您的当前位置：首页正文

古海洋生产力指标综述

来源：六九路网

Earth-ScienceReviews149(2015)23–52

ContentslistsavailableatScienceDirect

Earth-ScienceReviews

journalhomepage:www.elsevier.com/locate/earscirev

Totalorganiccarbon,organicphosphorus,andbiogenicbariumﬂuxesasproxiesforpaleomarineproductivity

ShaneD.Schoepfera,⁎,JunShenb,c,d,HengyeWeib,e,RichardV.Tysonf,ElleryIngallg,ThomasJ.Algeob,c,daDepartmentofEarthandSpaceSciences,UniversityofWashington,Seattle,WA98195,USADepartmentofGeology,UniversityofCincinnati,Cincinnati,OH45221-0013,USAcStateKeyLaboratoryofGeologicalProcessesandMineralResources,ChinaUniversityofGeosciences,Wuhan,Hubei430074,PRChinadStateKeyLaboratoryofBiogeologyandEnvironmentalGeology,ChinaUniversityofGeosciences,Wuhan,Hubei430074,PRChinaeDepartmentofEarthScience,EastChinaInstituteofTechnology,Nanchang,Jiangxi330013,PRChinafGETECH,KitsonHouse,ElmeteHall,ElmeteLane,LeedsLS82LJ,UnitedKingdomgSchoolofEarthandAtmosphericSciences,GeorgiaInstituteofTechnology,Atlanta,GA30332,USA

barticleinfoabstract

Althoughmarineproductivityisakeyparameterintheglobalcarboncycle,reliableestimationofproductivityinancientmarinesystemshasprovendifﬁcult.Inthisstudy,weevaluatetheaccumulationratesofthreecommonlyusedproxiesforproductivityfromasetofprimarilyQuaternarysedimentcoresat94marinesites,compiledfrom37publishedsources.Foreachcore,massaccumulationrateswerecalculatedfortotalorganiccarbon(TOC),or-ganicphosphorus(Porg),andbiogenicbarium(Babio).Calculatedmassaccumulationrateswerecomparedtotwoindependentestimatesofmodernregionalprimaryproductivityandexportproductivity,aswellastotwopoten-tialcontrollingvariables,bulkaccumulationrate(BAR)andredoxenvironment.BARwasfoundtoexerciseastrongcontrolonthepreservationoforganiccarbon.ThelinearregressionequationsrelatingpreservationfactortoBARcanbetransformedtoyieldequationsforprimaryandexportproductionasafunctionofTOCandBAR,twovariablesthatcanbereadilymeasuredorestimatedinpaleomarinesystems.Paleoproductivitycanalsobeestimatedfromempiricalrelationshipsbetweenelementalproxyﬂuxesandmodernproductivityrates.Althoughtheseequationsdonotattempttocorrectforpreservation,organiccarbonandphosphorus(butnotbarium)ac-cumulationsrateswerefoundtoexhibitasystematicrelationshiptoprimaryandexportproduction.Allofthepaleoproductivityequationsdevelopedherehavealargeassociateduncertaintyand,so,mustberegardedasyieldingorder-of-magnitudeestimates.

RelationshipsbetweenproxyﬂuxesandBARprovideinsightsregardingthedominantinﬂuencesoneachele-mentalproxy.IncreasingBARexerts(1)astrongpreservationaleffectonorganiccarbonthatissubstantiallylarg-erinoxicfaciesthaninsuboxic/anoxicfacies,(2)aweakclastic-dilutioneffectthatisobservablefororganicphosphorus(butnotfororganiccarbonorbiogenicbarium,owingtootherdominantinﬂuencesontheseprox-ies),and(3)alargenegativeeffectonbiogenicbariumthatisprobablyduetoreduceduptakeofbariumatthesediment–waterinterface.Theseeffectsbecameevidentthroughanalysisofourgloballyintegrateddataset;anal-ysisofindividualmarinesedimentaryunitsmostcommonlyrevealsautocorrelationsbetweenelementalproxyﬂuxesandBARasaresultofthelatterbeingafactorinthecalculationoftheformer.Weconcludethatorganiccarbonandphosphorusﬂuxeshaveconsiderablepotentialaswidelyusefulpaleoproductivityproxies,butthattheapplicabilityofbiogenicbariumﬂuxesmaybelimitedtospeciﬁcoceanicsettings.

Articlehistory:

Received1February2014Accepted27August2014

Availableonline7September2014Keywords:

MarineproductivityTotalorganiccarbonPhosphorusBarium

SedimentationrateSedimentmassﬂux

Contents1.2.

Introduction.........Paleoproductivityproxies...2.1.Generalconsiderations.2.2.Totalorganiccarbon(TOC)2.3.Organicphosphorus(Porg)

....

...............

.....................................................................................................................................................................................................................................................2424242526

⁎Correspondingauthor.

E-mailaddresses:shanedms@uw.edu(S.D.Schoepfer),thomas.algeo@uc.edu(T.J.Algeo).

24S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

2.4.Biogenicbarium(Babio)......................2.5.CovariationamongTOC,Porg,andBabio...............2.6.Inﬂuencesonpaleoproductivityestimates..............3.Methods................................

3.1.Compilationofmodernsedimentcoredatabase...........3.2.CalculationoforganicPandbiogenicBa...............3.3.Calculationofproductivityproxyﬂuxes...............3.4.Modernoceanicprimaryproductivityandexportproductivityestimates3.5.Calculationofpreservationfactors(PF)...............4.Results.................................

4.1.Robustnessofmodernproductivityestimates............4.2.Organiccarbonaccumulationrates(OCAR).............4.3.Organicphosphorusaccumulationrates(PAR)............4.4.Biogenicbariumaccumulationrates(BaAR).............5.Discussion...............................

5.1.RelationshipofproductivityproxyMARtoBAR...........5.2.Estimationofpaleoproductivityandestimateerrors.........5.3.PaleoproductivityestimatesbasedonOCAR.............5.4.PaleoproductivityestimatesbasedonPAR..............5.5.PaleoproductivityestimatesbasedonBaAR.............6.Conclusions...............................Acknowledgments..............................AppendixA.Supplementarydata......................References..................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................272929313132353636383838393940404546464748484848

1.Introduction

Organicproductivityisafundamentalparameterofallmarineecosys-tems,playingapivotalroleinecologicaldynamics,environmentalredoxconditions,andthecyclingofcarbon,nitrogen,phosphorus,andothernu-trientelements.Inthemodernopenocean,themainprimaryproducersaresingle-celledphytoplanktoninthesurfacemixedlayer(Levinton,2008).Somephytoplankton,e.g.,calcareouscoccolithophoresandsili-ceousdiatoms,producemineralizedtestswhoseﬂuxestothesedimentcanbeusedasproductivityproxies(e.g.,Kinkeletal.,2000;Rageneauetal.,2000).AlthoughbiogenicsedimentsarecommoninthePaleozoicandearlyMesozoicpelagicocean,mainlyasradiolarites(e.g.,Hori,1992;Algeoetal.,2010),mineralizedphytoplanktondidnotbecomecommonuntiltheTriassic,andphytoplanktontestsdidnotbecomeadominantcomponentofmarinesedimentsuntiltheCretaceous(Martin,1995;Ridgwell,2005).Eveninthemodernocean,manymarinealgaelackmineralizedtests(Tomas,1997)andcontributeonlyamorphousor-ganicmatter(AOM)tothesediment(Tayloretal.,1998).Inregionsdom-inatedbynon-mineralizedalgae,productivityhasbeenestimatedonthebasisofgeochemicalproxiessuchastotalorganiccarbon(TOC),organicphosphorus(Porg),andbiogenicbarium(Babio)(Tribovillardetal.,2006;CalvertandPedersen,2007).

TheutilityofTOC,Porg,andBabioaspaleomarineproductivityproxiesdependsonadominantlymarinesourceoforganicmatterandfavorableconditionsforpreservationinthesediment.Carbonandphosphorushavetheadvantagesofbeingmajorcomponentsofmarinealgalbiomassandhavingfewothersourcesinopen-oceansettings.Theonlyothersig-niﬁcantsourceofeithercomponenttomarinesedimentsisterrestrialor-ganicmatter,whichisprevalentmainlyincoastalareas(HedgesandParker,1976;ShowersandAngle,1986).Preservationfactors(PFs)fororganiccarbon(i.e.,thefractionofprimaryproductionpreservedinthesediment)canbeashighas30%inreducingfaciesbutarecommonlyfarlower(≤1%)inoxicfacies(Canﬁeld,1994;Tyson,2005).Ontheotherhand,burialefﬁciencies(BE;i.e.,thefractionoftheorganiccarbonsinkingﬂuxpreservedinthesediment)aretypicallyintherangeof10–50%(Canﬁeld,1994;Tyson,2005)and,thus,canbemorereliablyesti-matedforpaleomarinesystems(Algeoetal.,2013).Porgispreferentiallyrecycledbackintothewatercolumnunderreducingconditions(VanCappellenandIngall,1994)butcanbeeffectivelyretainedwithinthesedimentunderoxictosuboxicconditions(Föllmi,1996;Algeoand

Ingall,2007).TheutilityofBabioasaproductivityproxyisascribedtothecloserelationshipbetweentheproductionofauthigenicbariteandthedecayoforganicmatterincontactwithseawater,whichisthesourceofBabio(PaytanandGrifﬁth,2007).Anadvantageofthisproxyisthatthemineralbariteisrelativelyresistanttodissolutionunderoxictosuboxicconditions,providingameansofestimatingexportproductioninnon-reducingpaleomarinesystems.Becauseoftheincompletepreservationofallofthesecomponentsinmarinesediments,estimatesbasedonmeasuredconcentrationsrepresentminimumvaluesofbothprimaryproductivity(i.e.,theﬂuxofcarbonﬁxedfromtheatmosphereintothesurfaceocean)andexportpro-ductivity(i.e.,theﬂuxofcarbonfromthesurfacemixedlayertothethermoclineregionoftheocean).

Inthiscontribution,weundertakeananalysisofTOC,Porg,andBabioﬂuxesinmodernmarinesettingswiththegoalofevaluatingtheirutilityaspaleoproductivityproxies.Tothisend,we(1)calculatedtheﬂuxesofTOC,Porg,andBabioinarangeofmodernmarinesettings,(2)comparedthesedatawithestimatesofprimaryandexportproductivityforeachsetting,and(3)evaluatedtherelativeinﬂuencesofproductivityversuspreservation(whichiscloselyrelatedtosedimentbulkaccumulationrates)ontheaccumulationofTOC,Porg,andBabio.

InacompanionpaperinthisvolumebyShenetal.(inreview),theﬁndingsofthepresentstudyareappliedtoananalysisofproductivityvariationsduringthePermian–Triassictransition,themostseverebiodi-versitycrisisofthePhanerozoic(Erwinetal.,2002).Whilemarineanoxiaiswidelyagreedtohaveplayedamajorroleintheextinction(Isozaki,1997;WignallandNewton,2003),modelssuggestthatthiscouldnothaveoccurredwithoutasubstantialincreaseinmarineexportproduction(Hotinskietal.,2001;WinguthandWinguth,2012).2.Paleoproductivityproxies2.1.Generalconsiderations

Avarietyofgeochemicalproxieshavebeenusedtoreconstructpastchangesinbiologicalproductivity,includingmethodsbasedonCandNisotopes,organicbiomarkers,andtracemetal(Cu,Ni,Cd,Zn)abun-dances(seereviewsinTribovillardetal.,2006;CalvertandPedersen,2007).Eachproxyisaffectedbyahostofenvironmentalfactorssuchastemperature,redoxconditions,andoceancirculation,inadditionto

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5225

factorsthatinﬂuencethecompositionandstructureofmarineecosys-tems(Tribovillardetal.,2006).Sinceallpaleoproductivityproxiesaresubjecttosubstantialuncertaintiesandnosingleproxyisinherentlyre-liableunderallconditions,itisgenerallyadvisabletomakepaleoproductivityestimatesbasedonmultipleproxies(AverytandPaytan,2004).Inthisstudy,wefocusonTOC,Porg,andBabio,whichareamongthemostwidelyusedpaleoproductivityproxies.

Inferencesconcerningproductivityinpaleomarinesystemsarecom-monlybasedonﬂuxesratherthanconcentrationdata(e.g.,Algeoetal.,2011,2013).TheamountsofTOC,Porg,andBabioinasedimentarysucces-sionarenotgoodproxiesforproductivitybecauseminor-componentconcentrationsarestronglyinﬂuencedbysite-speciﬁcsedimentaccumulationrates,andchangesintheﬂuxofdiluents(e.g.,claymin-erals,orbiogeniccarbonateorsilica)canresultinvariationincomponentconcentrationsintheabsenceofanyactualchangesinproductivity.Themostrobustmethodofproductivityreconstructionthereforerequirestheconstructionofanage-depthmodelforeachstudysite,andthesub-sequentconversionofconcentrationdatatoﬂuxestimates.Age-depthmodelscanbeconstructedonthebasisofradiogenicisotopicdates(Winckleretal.,2005),astronomicaltuning(Algeoetal.,2011),or,atacoarsertemporalresolution,averagesedimentationratesforentiregeo-logicstagesorsubstages(Algeoetal.,2013).Althoughabsoluteproductivityestimatesareinherentlyuncertainowingtothemultitudeoffactorsthatinﬂuencepreservation,secularvariationinproxyﬂuxesatasinglelocaleoftencanprovidereliableinformationabouthowpro-ductivityhaschangedlocallythroughtime.

2.2.Totalorganiccarbon(TOC)

Organiccarbon,representingthesinglelargestconstituentoforgan-icmatter,providesthemostdirectproxyforproductivity(PedersenandCalvert,1990;Canﬁeld,1994;Tyson,2005;Zonneveldetal.,2010).Pri-maryproducersinthephoticzonetakeupCO2fromtheatmospheretoformorganicmatterviaphotosynthesis(Fig.1A).Asubstantialportionofcarbonﬁxedviaphotosynthesis(primaryproductivity)isrecycledwithintheocean-surfacemixedlayer(whichvariesspatiallyandtem-porally,butisgenerallyfromtenstoafewhundredsofmetersdeep),whiletheremaindersinksintothethermoclineanddeepoceanasnecromassandfecalpellets.Theﬂuxoforganiccarbonleavingtheocean-surfacelayer(knownas‘exportproduction’)canbeexpressedasafractionoftotalprimaryproductivity,i.e.,thepe(particleexport)ratiosensuDunneetal.(2005).Thef-ratio,referringtotheproportionofproductivityfueledbyexogenousnutrientinputs,isusedinter-changeablywith‘peratio’insomestudies(e.g.,Eagleetal.,2003),asthesetermsareexpectedtobeequalundersteady-stateconditions.Thisparametervarieswidelyinthemodernoceanasaresultofvaryingnutrientavailability,planktoncommunitystructure,andballastingbyinorganicsedimentcomponents,and,therefore,showslittlepredictablegeographicvariation(Dunneetal.,2005).

Mostoftheorganiccarboninsedimentscomesfromsinkingparti-clesoforganicmaterialproducedinthesurfaceocean,togetherwithacomponentofterrigenousorganiccarboninsomemarginalmarineset-tings(e.g.,Goñietal.,1997;OpsahlandBenner,1997).Inmostmarine

ABC

Fig.1.Schematicillustratingﬂuxesfromthesurfaceoceantothesedimentof(A)organiccarbon,(B)organicphosphorus,and(C)biogenicBa.Arrowwidthshowstherelativemagnitudeofﬂuxes.Cdetr=detritalorganiccarbon,Corg=primarymarineorganiccarbon;Pdetr=detritalP(insilicatesandotherminerals),Porg=organicP,PFe=Fe-boundP,Pau=authigenicP(e.g.,francolite),Pbio=biogenicP(ﬁshteeth,bones,scales,conodontelements);Badetr=detritalBa,Bacarb=Baincarbonatefacies,BaFe=Fe-boundBa(basedonFilippelliandDelaney,1996;Dymondetal.1992;PaytanandMcLaughlin,2007;Kraal,2010).

26S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

systems,thebulkoftheorganicmaterialexportedfromthesurfaceoceanisdecomposedbybacterialrespirationbeforereachingthesediment(Deuser,1971;OpsahlandBenner,1997).Arelativelysmallfractionofprimaryproduction(generally~0.1%to10%)survivestobedepositedatthesediment–waterinterface(MüllerandSuess,1979;Canﬁeld,1994;HedgesandKeil,1995),andasubstantialportionofthisorganicmatterissubsequentlylosttoanaerobicrespirationvia(i)denitriﬁcation,(ii)manganese,iron,orsulfatereduction,or(iii)methanogenesisafterburialinthesediments(e.g.,Fig.1A;Froelichetal.,1979).

Manystudiesofmodernandancientmarinesedimentshaveusedtotalorganiccarbon(TOC)toreconstructprimaryproductivity(e.g.,PedersenandCalvert,1990;Canﬁeld,1994;Kuypersetal.,2002;Algeoetal.,2013;Felix,2014).However,themeasuredTOCcontentofsedimentsisafunctionnotonlyofprimaryproductivitybutalso:1)peratio,2)preservationoforganiccarboninthewatercolumnandduringdiagenesis,and3)dilutionoforganiccarbonbylithogenicorbiogeniccomponentsinthesediment(Canﬁeld,1994;Tyson,1995,2001,2005).Sinceorganiccarbonistypicallyaminorcomponentofmarinesedi-mentsandonlyasmallfractionofprimaryproductionispreserved,smallchangesinthePFoforganiccarboncanhaveamajoreffectontheTOCcontentofthesediment(Tyson,2005).Aerobicrespirationoforganicmaterialbybacteriaisthemostefﬁcientformofcarbonremineralization(Fig.1A),andsedimentsunderlyinganoxicwatercol-umnmayhavealowTOCcontentevenifsurface-waterproductivityishigh(BernerandRaiswell,1983;Berner,1984).Organicmatterpreser-vationisenhancedbyminimizingtheexposureoforganicmaterialtoox-ygenthrough(1)morereducingbottom-waterconditions,and(2)morerapidsedimentation.Bothoftheseconditionsresultinmorerapidpas-sageoforganicmatterthroughthezoneofaerobicrespiration,inwhichdecayisrapid,andintothezoneofanaerobicrespiration,inwhichdecayisapproximatelyanorderofmagnitudeslower(Simonetal.,1994;Bastvikenetal.,2004).

Organiccarbonaccumulationratesandpreservationfactorsshowastrongpositivecorrelationwithsedimentationratesinthemodernocean,inwhichoxicbottom-waterconditionspredominate(Tyson,2005).Carbonaccumulationratesinsuboxicandanoxicregions,whichmaybeabetteranalogformanypaleomarinesystems,showlessofadependenceonsedimentationrates,probablybecauseoxygenexposureandpreservationratearelessofacontrolonsedimentTOCcontentundersuchconditions.Theconvergenceofthesetrendsat~103.1gm−2yr−1(Tyson,2005)suggeststhatlittletonoaerobicrespi-rationoccursbelowthesediment–waterinterfaceatbulkaccumulationrateshigherthanthisvalue(Algeoetal.,2013).HighsedimentationratesalsohavethepotentialtodilutetheOCﬂuxtothesediments,andinenvironmentswherethepredominantphytoplanktongroupshavemineralizedtests,‘autodilution’mayoccur,wherevariationsintheﬂuxesofOCandbiogenicdiluentsreﬂectthesameﬂuctuationsinprimaryproductivity.Althoughenhancedpreservationappearstobethepredominanteffectofhighsedimentationratesinthemodernocean,high-TOCsedimentsdepositedinreducingpaleomarinesystemsmayreﬂectperiodsofslowsedimentaccumulation(BettsandHolland,1991;Tyson,2001,2005;AlgeoandHeckel,2008).

Inadditiontowater-columnredoxconditionsandsedimentationrates,otherfactorsrelatedtosedimentpropertiesanddepositionalen-vironmentalconditionscanhavealargeinﬂuenceonthepreservationoforganicmatter(PedersenandCalvert,1990;Canﬁeld,1994).Biotur-bationofthesedimentsbymacrofaunaorphysicalmixingprocessescanincreasetheexposureoforganicmaterialtooxygenduringburial.Ontheotherhand,exposuretoH2Sinreducingporewaterenvironmentscancontributetotheformationofsulﬁdizedorganiccompoundsthatarerelativelymoreresistanttodegradation(Tribovillardetal.,2004;Zonneveldetal.,2010).Insomepaleomarinesystems,organiccontentalsoshowsastrongrelationshiptomineralsurfacearea(ametriclarge-lydependentonclay-mineral,especiallysmectite,content),suggestingthatadsorptiontoclaymineralsurfacesmaybeanimportantmecha-nismoforganiccarbonpreservation(Kennedyetal.,2002;Kennedy

andWagner,2011;Kennedyetal.,2014).Theseparametersarenotin-dependentandcaninteractinsynergisticwaystopromotepreservationordecompositionoforganicmaterial.Forexample,clay-richsedimentstendtohavelowpermeability,whichislikelytopromoteareducingporewaterenvironmentand,hence,sulﬁdization.

Theprocessesdiscussedabovereﬂectthechemical,physical,andecologicalconditionsinthesurfacelayer,watercolumn,andsedimentsduringformationanddepositionoforganicmatter.Latediageneticpro-cessescanalsoaffectTOCcontentthroughlossofcarbonduringthermalmaturation.Temperatureappearstobethesinglemostimportantfactorinthelossoforganicmatterfromsedimentsduringlateburialdiagene-sis(RaiswellandBerner,1987).Themaximumburialtemperaturecanbeestimatedforpaleomarinesystemsbyvitrinitereﬂectance,conodontalterationindex,oranumberofothermethods,andtheoriginalTOCcontentbeforethermalmaturationpotentiallycanbeback-calculated(DalyandEdman,1987;SkjervoyandSylta,1993;Petersetal.,2006;ModicaandLapierre,2012;Algeoetal.,2013).

Inordertogeneratepaleoproductivityestimatesfromtheburialﬂuxoforganiccarbon(Fig.1A),oneormoretransformfunction(s)arere-quiredtoaccountfororganiccarbonlosseswithinthewatercolumnandsediment(e.g.,Algeoetal.,2013).Organiccarbonlossesduetodia-geneticprocesseswithinthesedimentcanbeestimatedwithrelativeconﬁdence,yieldinganestimateofthesinkingﬂuxoforganiccarbon(i.e.,itsrateofdeliverytothesediment–waterinterface).Organiccarbonlosseswithinthethermoclineanddeeplayersoftheoceanaresomewhatlesspredictable,althoughrelationshipstowaterdepth(e.g.,Suess,1980;Sarntheinetal.,1988;Antiaetal.,2001)canbeusedtoestimateexportproductionfromtheorganiccarbonsinkingﬂux.However,burialefﬁciency,whichisdependentonsedimentationrate,appearstoaccountformorevariationinsedimentTOCthanorgan-iccarbonsinkingﬂux,whichisdependentonwaterdepth(Felix,2014).Ontheotherhand,thesubstantialandunpredictablevariabilityofpera-tiosmakesreliableestimationofprimaryproductivityfromtheorganiccarbonsinkingﬂuximpossibleinmostpaleomarinesystems(Dunneetal.,2005;Algeoetal.,2013).Fortunately,theexportproductionparameterisofconsiderableinterestinpaleomarinestudies,asitisdi-rectlyrelatedtobenthicecologyandredoxconditions.2.3.Organicphosphorus(Porg)

Phosphorusisanessentialnutrientformarinephytoplanktongrowth,beingastructuralandfunctionalcomponentofallorganisms(Redﬁeld,1958).Itispresentinseawaterinbothdissolvedandparticu-lateforms(PaytanandMcLaughlin,2007).Thedissolvedfractionin-cludesinorganicphosphorus(generallyinthesolubleorthophosphateform)andmacromolecularcolloidalphosphorus.ParticulatePincludesorganicPwithinlivinganddeadplanktonaswellasbioapatite(e.g.,ﬁshteethandscales,conodontelements),precipitatesofauthigenicphos-phorusminerals,andphosphorusadsorbedtoparticulates(e.g.,iron-boundP)(FilippelliandDelaney,1996;Schenauetal.,2005;PaytanandMcLaughlin,2007;Kraal,2010).

OrganicmatteristheultimatesourceofmostPinmarinesediments(Fig.1B;Ingalletal.,1993),whereasdetritalP(i.e.,interrigenoussili-catesandotherminerals)generallycomprisesb20%oftotalP(Fig.2;AlgeoandIngall,2007).Asaconsequence,totalPiscommonlyusedasaproxyfororganicallyderivedphosphorus(‘Porg’inthisstudy)(e.g.,AlgeoandIngall,2007).Bioapatiteisusuallyalsoaminorcompo-nentofmarinesedimentbutcanrepresentasubstantialportionoftheburialﬂuxofphosphorusﬂuxinsomehighlyproductiveupwellingset-tings(Schenauetal.,2005;Díaz-Ochoaetal.,2009).Intheoxygenatedmodernocean,mostorganicPisremineralizedafterburial,andaconsid-erablefractionoforganicandFe-boundPisreleasedtosedimentporewaters,mostofwhich(N90%)subsequentlydiffusesbackintotheoverlyingwatercolumn(Benitez-Nelson,2000).However,someofthereleasedPisretainedinthesedimentandultimatelyincorporatedintoauthigenicmineralphasessuchasfrancolite,acarbonateﬂuorapatite

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5227

Fig.2.DiageneticchangesinsedimentaryPphaseswithtimein(A)anoxicfaciesand(B)oxic–suboxicfacies.

AdaptedfromFilippelliandSouch,1999;Kraal,2010.

mineral(Fig.2;FilippelliandDelaney,1996;FilippelliandSouch,1999;AlgeoandIngall,2007).Asaconsequence,thedepositionalﬂuxofPadsorbedtoFe-oxyhydroxidescanbecomparabletothatassociatedwithorganicmatterincontinentalshelfdeposits(Filippelli,2001).

UnlikeTOC,whoseburialefﬁciencydependsprimarilyonpreserva-tionoftheoriginalsedimentarycomponent(organicmaterial),theburialefﬁciencyofPisafunctionofseveraldiageneticprocessesaffect-ingthefateofPinsedimentporewatersafterregeneration(IngallandVanCappellen,1990;IngallandJahnke,1997;Vinketal.,1997).Theef-ﬁcacyoftheseprocessesinretainingPwithinthesedimentdependsstronglyonbottom-wateroxygenconcentrations,withoxygenatedsitespromotingphosphorusretentionmorestronglythananoxicsites(Fig.1B;GächterandMüller,2003;Ingalletal.,2005).Underoxiccon-ditions,remineralizedorganicPisretainedthroughacombinationofprocesses,includingadsorptionontoandcomplexationwithmetaloxyhydroxides,aswellasbiologicalsequestrationinpolyphosphates(Filippelli,2001;Tribovillardetal.,2006;Díazetal.,2008).TheseP-trappingmechanismscanresultinporewaterPconcentrationsreachingsaturation,leadingtoprecipitationofauthigenicphosphatemineralsthat,onceformed,arestableintheburialenvironment(Fig.2;Kraal,2010).ThedominantmechanismforPretentionmaybeadsorptionontoFe-oxyhydroxides,withredoxcyclingofthelatterwithinthesedimentresultinginmultiphaseadsorptionandreleaseofP(Tribovillardetal.,2006).Underreducingconditions,ironislikelytobeintheformofdissolvedionsratherthanoxyhydroxideparticles,anda

lessactivebenthicmicrobiotaislesslikelytoformdissolvedortho-andpolyphosphateions(Díazetal.,2008).Asaresult,Preleasedtosedimentporewatersunderreducingconditionshasagreaterpotentialtodiffusebackintotheoverlyingwatercolumn(Ingalletal.,2005).However,somesulﬁde-oxidizingbacteriacanformapatiteandimmobilizeporewaterPunderanoxicconditions,whichmaybeimportanttothefor-mationofphosphoritesinupwellingzones(Goldhammeretal.,2010).

Phosphorusaccumulationrateshavebeenusedasaproductivityin-dicatorinbothmodern(SchenauandDeLange,2001)andancientsed-iments(Comptonetal.,1990;Schenauetal.,2005).However,caremustbetakentounderstandpotentialinteractionsbetweenproductivityandbottom-waterredoxconditions.HighPaccumulationratescoincidedwithperiodsofhighproductivityintheQuaternaryArabianSeadespiteaconcurrentdecreaseinPburialefﬁciency(Fig.3;Schenauetal.,2005).HighPaccumulationratesalsocoincidedwiththeonsetoforganicsed-imentationduringtheCenomanian–Turonianoceanicanoxicevent(OAE2),suggestingthatPaccumulationmirroredproductivitychangesuntilintensebottom-wateranoxiahinderedPretention(Mortetal.,2007).Asaresultofcomplexinteractionsbetweenproductivity,redoxconditions,andburialefﬁciency,Paccumulationratescannotbeas-sumedtohavealinearrelationshipwithprimaryorexportproductivity(Tribovillardetal.,2006).

Phosphoritesinthesedimentaryrecordtypicallycoincidewithhighorganiccarboncontentandhavebeenlinkedtohigh-productivityre-gionssuchasupwellingzones(Föllmi,1996;Goldhammeretal.,2010).ThePermianPhosphoriaFormationofthenorthwesternUnitedStateshasbeenlinkedtoaproductivecoastalupwellingsystemimping-ingonashallow,semi-restrictedshelfembayment(HiattandBudd,2003),andthefewregionsofthemodernoceanwherephosphoritefor-mationhasbeenobservedarecloselylinkedtocoastalupwelling(BremnerandRogers,1990;RaoandLamboy,1995).However,phos-phoritesalsocanbeproduced(orenhanced)throughphysicalsedimen-taryprocessessuchaswinnowing(Föllmi,1990).Amongalargesuiteofredoxandproductivityproxies,Brumsack(2006)foundthatPconcen-trationisoneofthefewmetricscapableofdistinguishingeffectivelybetweenorganic-richsedimentsformedinproductiveupwellingzonesversus‘stagnant’depositionalbasinssuchastheBlackSea,withPbeingconsistentlymoreenrichedinsedimentsoftheformer.Brumsack(2006)concludedthatPisoneofthemostrobustandwidelyapplicableproductivityproxies.2.4.Biogenicbarium(Babio)

Thechemicalbehaviorandcyclingofbariuminseawaterarenowfairlywellunderstood(Ganeshrametal.,2003;PaytanandGrifﬁth,2007;VanBeeketal.,2007;GrifﬁthandPaytan,2012).Bariumhasarel-ativelyshortresidencetimeinseawater(~10kyr;Dickensetal.,2003).Itsmainsourceisriverrunoff,anditsprimarysinkisburialofbarite(BaSO4)inmarinesediments.Alargepartofthebariteburialﬂuxisde-liveredtothesediment–waterinterfacebysinkingorganicmatter(Fig.1C),leadingtolocalBaenrichmentofthesediment.Thisfractionisreferredtoas‘biogenicbarium’(Babio),i.e.,Baassociatedwiththesinkingﬂuxoforganicmatter.OthersinksforseawaterBaincludeadsorptionontoaluminosilicate,carbonate,andferromanganeseoxyhydroxidesedimentaryphases(Dymondetal.,1992;Eagleetal.,2003;GonneeaandPaytan,2006).

TheoriginofBabiohasbeenamatterofsomecontroversy.Somephytoplankton(BertramandCowen,1997;Gonzalez-Muñozetal.,2012)andzooplankton(Riederetal.,1982;Bernsteinetal.,1992)as-similateBaintracellularly,whichcanyieldconcentrationssubstantiallyexceedingthatofseawater(Fisheretal.,1991).However,bariteprecip-itationbymarineorganisms(GoodayandNott,1982;BertramandCowen,1997)orreplacementofcelestite(SrSO4)inacantharians(Bernsteinetal.,1992,1998)areminorsourcescomparedwithauthigenicprecipitationofbaritewithindecayingorganicmatter(Fig.1B;Bishop,1988;VanBeeketal.,2007).Theoperationofthelatter

28S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

ABCDFig.3.Massaccumulationrates(MARs)forcoreODP117-722BfromtheOwenRidge,ArabianSea(16.62°N,59.80°E,waterdepth2022m).(A)Totalorganiccarbon(TOC),(B)phosphorus(P),(C)excessbarium(Baxs,aproxyforbiogenicbariumBabio),and(D)d18Opaleotemperatureproxydata.Noteglacial/interglacialcyclesinproductivityproxies;glacialintervalsareshaded.OxygenisotopesweremeasuredintheforaminiferGlobigerinoidessacculifer.Phosphorus,barium,andaluminumconcentrationdataarefromShimmieldandMowbray(1991)andTOCdatafromMurrayandPrell(1991).BabiowascalculatedusingaBa/Alratioof0.00475,basedonaBa-versus-Alcrossplot.MARwascalculatedbasedontheagemodelanddrybulkdensitymeasurementsofMurrayandPrell(1991).

DataaccessedthroughPANGAEAdataportalbhttp://www.pangaea.deN.

processisdemonstratedbyformationofbaritethroughouttheoceanicwatercolumnratherthaninthephoticzonealone(VanBeeketal.,2007).Bariteformationiscommonlyassociatedwiththedecayofcoccolithophoreanddiatombiomass;twophytoplanktongroupsthatarenotknowntoprecipitatebariteintracellularly(Ganeshrametal.,2003).Thedecayprocessresultsinthelocalizeddevelopmentofmicro-environmentsonorganicparticlesurfaces,whereBaisconcentratedviaadsorptionontoFe-oxyhydroxides(Sternbergetal.,2005),andwhere

2thesolubilityproductofbarium(Ba+2)andsulfate(SO−4)exceedstheequilibriumconstantofbarite(BaSO4),causingspontaneousnucle-ationofbaritecrystals(Dehairsetal.,1980;Bishop,1988;Dehairsetal.,1992).Thisprocessisassistedbyelevatedconcentrationsofsulfatewithinsinkingorganicparticlesduetooxidationoforganic-derivedsul-furcompoundssuchasaminoacids(DymondandCollier,1996)orre-oxidationofH2Sproducedthroughbacterialsulfatereduction.

Seawaterismoderatelyundersaturatedwithrespecttobaritethroughmostoftheglobalocean,includingmostsurfacewaters,althoughpartsofthedeepoceanaresupersaturated(Monninetal.,1999).Thispatternre-ﬂectstheexportofBafromtheocean-surfacelayerinassociationwithsinkingorganicparticles(Fig.1C).Despitewidespreadundersaturationofseawaterwithrespecttobarite,particulatebariteisnearlyubiquitousintheworld'soceans.Particulatebariteisgenerallysmall(meansize~1μm)andisfoundinassociationwithorganicaggregatesinsurfacewa-tersandasfreeparticlesindeepwaters(Bishop,1988;Dehairsetal.,1990;BertramandCowen,1997).However,formationofparticulatebar-iteisrelativelymorevigorousinthedeepocean,whereseawaterisclosertosaturationwithbarite,thanintheocean-surfacelayerandthermocline,wherelowersaturationlevelsarefound(Fig.1C).SpatialgradientsinBaconcentrationarefoundinthedeepoceanasaconsequenceofexportofparticulatebaritefromtheNorthAtlantictotheNorthPaciﬁcviatheglobaloceanicconveyorbelt(Broecker,1991).

BothBabioandbaritehavebeenusedasapaleoproductivityproxyinoceanographicstudiesofmarinesedimentsofvariousages,includingtheRecent(Dehairsetal.,1987;PrakashBabuetal.,2002;Weldeabetal.,2003),Quaternary(Paytanetal.,1996;Morenoetal.,2002;Klöckeretal.,2006;Jaccardetal.,2013),Neogene–Paleogene(LatimerandFilippelli,2002;AndersonandDelaney,2005;Paytanetal.,2007),Cretaceous(Zachosetal.,1989;Bąk,2007),andPaleozoic(Jewell,1994;Kastenetal.,2001).Asapaleoproductivityproxy,baritehastheadvan-tagesofbeingrelativelyrefractoryandhavingahighburialefﬁciency

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5229

underoxicconditions(~30–50%oftheBasinkingﬂuxversusb10%foror-ganiccarbonandphosphorus)(Dymondetal.,1992;PaytanandKastner,1996;BalakrishnanNairetal.,2005;PaytanandGrifﬁth,2007).Eveninreducingenvironmentsthatarenotconducivetobaritepreservation,thehighﬂuxofbiogenicBatothesediment–waterinterfacenonethelesscanmakeitausefulpaleoproductivityproxy(Falkneretal.,1993;Thomsonetal.,1995;VanSantvoortetal.,1996;Nijenhuisetal.,1999;Martinez-Ruizetal.,2000;Kastenetal.,2001;PrakashBabuetal.,2002;Martinez-Ruizetal.,2003;Scopellitietal.,2004;Bąk,2007).

Thefateofparticulatebariteinthesedimentiscriticaltoitspotentialutilityasapaleoproductivityproxy.Bariteisrelativelyinsolubleunderoxidizingconditionsand,hence,tendstobewellpreservedwherebottom-watersareoxygenated(Fig.1C;PaytanandKastner,1996;BalakrishnanNairetal.,2005;PaytanandGrifﬁth,2007),althoughbariteparticlesarepotentiallysubjecttowinnowingorcorrosion(Dymond,1981).However,bariteinsurﬁcialsedimentsmaydissolvewhenex-posedtoanoxicporewatersfromwhichthesulfatehasbeenremovedbymicrobialactivity(VanOsetal.,1991;VonBreymannetal.,1992;McManusetal.,1994;PaytanandKastner,1996;Torresetal.,1996;McManusetal.,1998,1999;Schenauetal.,2001),aprocesswhichlimitstheuseofBaasapaleoproductivityproxyinsomemoderncontinentalmarginsettings(Shimmieldetal.,1994;Ganeshrametal.,1999).Asre-ducingconditionsarewidelypresentatshallowdepthswithinmarinesediments,somelossofauthigenicBaduringburialmaybecommon(Berelsonetal.,1996).Comparisonofwater-columnsediment-trapwithsedimentdatasuggeststhat,onaverage,~30%oftheparticulateBaﬂuxtotheseaﬂoorispreservedintheopenPaciﬁc(Dymondetal.,1992).However,thisvaluevarieswidely(~10–70%)asafunctionofbulkaccumulationrate,withsuperiorbaritepreservationassociatedwithhighersedimentationrates(Dymondetal.,1992).Onceburied,bar-iteparticlestendtoremainstableowingtothegenerallysaturatedcon-ditionofsedimentporewaterswithrespecttobarite.

OneissueofimportancefortheapplicationofBabioasapaleo-productivityproxyiswhetherseawaterhasremainedsaturated(ornearlyso)withrespecttobaritethroughtime.Ifancientseawaterwaseverstronglyundersaturatedwithrespecttobarite,thenauthigenicbar-itewouldnothaveaccumulatedinthesedimentunderanyproductivityconditions.Seawaterundersaturationwithrespecttobaritecouldhaveresultedfromlarge-scaleremovalofsulfatetothesedimentaryreservoir.However,seawaterhascontainedsubstantialquantitiesofdissolvedsul-fatesincetheEarlyProterozoic,whenoxidativeweatheringofsulﬁdesintensiﬁedduetorisingatmosphericpO2(Canﬁeld,2005).Evenifsea-watersulfateconcentrationsfell,ashasoccurredepisodicallyduringthePhanerozoic(Lowensteinetal.,2001;Luoetal.,2010;WortmannandPaytan,2012;Songetal.,2014),itisnotcertainthatthiswouldhaveresultedinundersaturationwithrespecttobarite.BecausebariteisthedominantsinkforBainseawater,anydecreaseindissolvedsulfateconcentrationswouldhavebeencompensatedbyariseinBa2+concen-trations,maintaininganapproximatelyconstantsolubilityproductforbarite.Therefore,itseemslikelythatthesaturationlevelofbariteinsea-waterhasnotvariedgreatlysincepossibly~2Ga.2.5.CovariationamongTOC,Porg,andBabioBecausenosingleproxyisinherentlymorereliablethanothers,paleoproductivityassessmentsshouldgenerallybemadeonthebasisofmultipleproxies(AverytandPaytan,2004).Inmanymodernmarinesystems,theﬂuxesofTOC,Porg,andBabio(orauthigenicbarite)arecoupled(e.g.,Figs.3,4),suggestingthat—underconditionsfavoringtheirpreservation—allthreecomponentsmayserveasproxiesforex-portproductivity(Dymondetal.,1992;Françoisetal.,1995;DymondandCollier,1996;Jeandeletal.,2000;Weldeabetal.,2003;Fageletal.,2004;BalakrishnanNairetal.,2005;PaytanandGrifﬁth,2007).Furthermore,theseproxieshavebeenshowntoexhibitsigniﬁcantsec-ularcovariationatglacial–interglacialtimescales,e.g.,inLateQuaterna-rymarinesedimentsoftheArabianSea(Fig.3;MurrayandPrell,1991;

ShimmieldandMowbray,1991;Shimmield,1992;Rosteketal.,1997;Schultzetal.,1998)andtheeasterntropicalPaciﬁc(Fig.4;Lyleetal.,1992;Gardneretal.,1997;GaneshramandPedersen,1998;Murrayetal.,2000;Murrayetal.,2012).However,attentionmustbepaidtothepossibilitythatsuchcovariationislinkedtochangesinthesedimen-tationrateandprovenanceoflithogeniccomponentsthatmayinﬂu-enceBaorAlconcentrations(Weldeabetal.,2003).

Inthewell-ventilatedmodernocean,theﬂuxesofTOC,Porg,andBabiovarystronglywithwaterdepth.Thesepatternsreﬂectapreservationalcontrolonorganiccarbonandphosphorusassociatedwithdepth-dependentdecaytimeandredoxconditions.Remineralizationratesoforganiccarbonandphosphorusremainhighuntilreachingthesedi-ment–waterinterface,especiallyinwell-oxygenatedwatercolumns(Paceetal.,1987;LeMoigneetal.,2013).Ratesofdecayinanoxicwatercolumnsaresigniﬁcantlylower(Canﬁeld,1994;Tyson,2005).P-bearingorganiccompoundsarerelativelymorelabilethannon-P-bearingcompounds,asaconsequenceofwhichtheTOC/Pratiosofor-ganicmattercanchangewithincreasingwaterdepthanddecay(IngallandVanCappellen,1990).Inshallowdeltaandshelfenvironments,TOC/PratiosaregenerallyclosetothecanonicalRedﬁeldratio(106:1)owingtorapidsinkingandburialoforganicmatter(RuttenbergandGoñi,1997).Ontheuppercontinentalslope(i.e.,withintheoxygen-minimumzone,orOMZ),TOC/Pratiosincreasetoamaximumof~400owingtopreferentialremineralizationofP-bearingcompoundsandlackofretentionofPinthesedimentundersuboxicconditions(IngallandVanCappellen,1990;SchenauandDeLange,2001).Inthedeepocean,TOC/PratioscanfallbelowtheRedﬁeldratioowingtoretentionofPwithinthesedimentofoxicfacies(Fig.5A;AlgeoandIngall,2007).

ThedepthdependenceofBabioaccumulationisafunctionoftherel-ativesaturationofbariteinthewatercolumn.Thedeepoceanisclosertosaturationwithrespecttobaritethanthesurfacemixedlayerandthermocline,soparticulatebariteismorelikelytoprecipitatewithinandsurvivetransitthroughdeeperwaters(GingeleandDahmke,1994;VanBeeketal.,2007).TOC/Babiovariessystematicallywithwaterdepthinthemodernocean,especiallybelow1500m(Fig.5B;Dymondetal.,1992;Schenauetal.,2001;CalvertandPedersen,2007).TOC/Babioratiosof100–125aretypicalofsurfacewatersofthepelagicoceanbutdecreaseto≤25atabyssaldepths(Dymondetal.,1992).TOC/Babioratiosaresomewhathigherinmarginalmarineenvi-ronmentswithhighsedimentationrates,althoughthepatternofde-creasingratioswithincreasingwaterdepthstillprevails(Françoisetal.,1995;Dehairsetal.,2000;Schenauetal.,2001;Fageletal.,2004).Thispatternisaconsequenceofthesimultaneousdecayofor-ganicmatterandformationofauthigenicbaritewithinsinkingorganicaggregates.Forthisreason,sedimentsdepositedabovetheOMZonshallowcontinentalshelvesmayexhibitatightcouplingbetweenTOCandBabiothatreﬂectstheinﬂuenceofproductivitymorethanpreserva-tion(seeSection2.6).

TheratiosamongTOC,Porg,andBabiocanalsovaryalonglateraltran-sects,fromcontinentmarginstothedeepocean(Françoisetal.,1995;Dehairsetal.,2000).Asaresult,althoughorganiccarbon,phosphorus,andbiogenicbariumoftenshowcovariationthatislikelyrelateddirect-lytochangesinprimaryproductivity,theratioofTOCtoeitheroftheothertwoproxiescanvarysubstantially,makingitdifﬁculttoestimatecarbonexportfromPorgorBabiowithoutadditionalinformationthatoftencanonlybeestimatedinpaleomarinesystems.2.6.Inﬂuencesonpaleoproductivityestimates

Twoofthemostimportantcontrolsonorganicmatterpreservationand,hence,onpaleomarineproductivityestimatesare(1)bottom-waterredoxconditions(Canﬁeld,1994;Meyers,1997),and(2)sedi-mentbulkaccumulationrates(BAR)(Tyson,2005;Algeoetal.,2013).Redoxconditionsgenerallyexertastronginﬂuenceontheaccumula-tionofelementalproductivityproxiesinthesediment.Forexample,thePFoforganiccarbonincreasesrapidlywithdecreasingdissolved

30S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Fig.4.Massaccumulationrates(MARs)forcoreTT013-PC72fromthetropicaleasternPaciﬁc(0.11°N,139.40°W,waterdepth4298m).(A)Totalorganiccarbon(TOC),(B)phosphorus(P),(C)excessbarium(Baxs,aproxyforbiogenicbariumBabio),and(D)d18Opaleotemperatureproxydata.Noteglacial/interglacialcyclesinproductivityproxies;glacialintervalsareshaded.OxygenisotopesweremeasuredintheforaminiferCibicideswuellerstorﬁ.Phosphorus,barium,andaluminumconcentrationdataarefromMurrayetal.(2000)andTOCdatafromMurrayetal.(2012).BabiowascalculatedusingaBa/Alratioof0.0065,basedonaverageshale(PAAS,TaylorandMcLennan,1985).MARwascalculatedbasedontheagemodelofMurrayetal.(2000)withdrybulkdensityestimatedusingtheequationofCurryandLohmann(1986).DataaccessedthroughSedDBdataportalbhttp://www.earthchem.org/seddbN.

oxygenconcentration(Canﬁeld,1994),whereasthoseofPorgandBabiogenerallydecreaseunderthesameconditions(Fig.6).Incontrast,therefractorynatureofbariteinoxidizingenvironmentsmaymakeitamoreeffectiverecorderofproductivitythanTOCinoxicmarinesystemswithlowsedimentationrates.BARalsoexertsastronginﬂuenceontheaccumulationoftheseelementalproxies,thepreservationofwhichisenhancedathighersedimentationrates(Tyson,2005).WithincreasingBAR,theinﬂuenceofbenthicredoxconditionsontheaccumulationofTOC,Porg,andBabiobecomesmuchlesspronouncedasaconsequenceofrapidburialandreducedoxygenexposuretime(Canﬁeld,1994;Tyson,2005).

Therelativeimportanceofredoxversusproductivitycontrolsonor-ganicmatteraccumulationmaybedistinguishableonthebasisofpat-ternsofcovariationamongTOC,Porg,andBabio.Intheredox-dominantscenario,theaccumulationofthesecomponentsismorestronglyinﬂu-encedbythedifferingpreservationpatternsoftheseproxies,resultinginnegativecovariationofTOCwithPorgandBabio(Fig.6A).Thisscenarioappliesequallytooxidizingandreducingenvironments.Inreducingen-vironments,ahighsinkingﬂuxofTOCcommonlyintensiﬁesreducingconditionsthroughhighbiologicaloxygendemand,thusenhancingor-ganicmatterpreservationandincreasingtheburialﬂuxoforganiccar-bon.Incontrast,reducingconditionstendtodiminishtheburialﬂuxesofPorgandBabioowingtoreductivedissolutionofMn–Fe-oxyhydroxideparticles,thusreleasingadsorbedP,andbaritecrystals,releasingBa(Tyson,2005).Ontheotherhand,oxidizingenvironmentsgenerallyfa-cilitatethepreservationandretentionofPorgandBabiowithinthesedi-ment,whereastheburialefﬁciencyoforganiccarbonisreducedowingtogreateraerobicremineralization(Tyson,2005).Intheproductivity-

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5231

Fig.5.(A)Molar(C/P)orgratiosofsedimentaryorganicmatter;notedifferencesbetweenMississippiDeltaandGulfofMexicoshelfsediments(RuttenbergandGoñi,1997)versusArabianSeaslopesediments(SchenauandDeLange,2001).Amolar(C/P)orgof106:1istheRedﬁeldratioformarinephytoplankton(Redﬁeld,1958).(B)Corg/Babioratiosofsettlingparticles;notedifferencesbetweenopen-ocean(Dymondetal.,1992;Françoisetal.,1995)andshelf-marginsites(Françoisetal.,1995;Dehairsetal.,2000;Fageletal.,2004).

dominantscenario,theexportoforganicmatterfromthesurfaceoceanisthemainfactorinﬂuencingtheﬂuxesofTOC,Porg,andBabiotothesed-iment(Fig.6B).Althoughpreservationaleffectsmayaltertheratiosofthesecomponents,theirsensitivitytochangesinproductivityimposesastrongpositivecorrelationamongallproxiesacrossthespectrumofredoxconditions(Tyson,2005).

Examplesofbothredox-andproductivity-dominantenvironmentscanbefoundinHoloceneandoldermarinesystems,evenwithinasin-gletypeofenvironmentalsetting,e.g.,OMZsinproductiveupwellingregions.TheQuaternaryeasterntropicalPaciﬁcexempliﬁesaredox-dominantsystem,inwhichPorgandBabioshowstrongpositivecovaria-tion(r2=0.74)withlittlerelationshipofeitherproxytoorganiccarbonﬂuxonglacial–interglacialtimescales(Fig.4;Murrayetal.,2000,2012).AsecondexampleistheBenguelaupwellingsystem,inwhichTOCandBa/Alcovarynegatively(Robinsonetal.,2002;n.b.,neitherproxyshowsaconsistentrelationshiptoP).Ontheotherhand,theQuaternaryArabi-anSeaexempliﬁesaproductivity-dominantsystem,inwhichTOCex-hibitsstrongpositivecovariationwithbothPorgandBabio(Fig.3;Reichartetal.,1997;BalakrishnanNairetal.,2005)despitepossibledia-geneticlossesofBabio(MurrayandPrell,1991;ShimmieldandMowbray,1991).Asecondexampleofaproductivity-dominantsystemisthehighlyproductiveChileanmarginupwellingsystem,inwhichTOCcovariespositivelyBa/Al(Muñozetal.,2012;n.b.,Pwasnotmeasuredinthisstudy).

BARisasecondmajorfactorinﬂuencingtheaccumulationofTOC,Porg,andBabio.Organiccarbonaccumulationratesshowastrongcorre-lationwithBAR(MüllerandSuess,1979),althoughthenatureofthere-lationshipdiffersbetweenfullyoxicandfullyanoxicenvironments(Tyson,2005).Inoxicenvironments,higherBARminimizestheexpo-suretimeoforganicmattertoaerobicdecayintheshallowburialzone,increasingitsBE(Hartnettetal.,1998;IversenandPloug,2010).Whilethelabilefractionoforganiccarbonisdegradedefﬁcientlyevenunderanaerobicconditions(HenrichsandReeburgh,1987;Hultheetal.,1998;KristensenandHolmer,2001;Bastvikenetal.,2004),theresidencetimeoftheresidualorganicmatteratthesediment–waterin-terfaceorintheshallowburialzoneappearstobelessofacontrolonpreservationwherebottom-watersareanoxic(Tyson,2001,2005);

thismaybeduetointrinsicallyslowerdegradationofrefractorycom-poundsintheabsenceofoxygen(Benneretal.,1984;Colberg,1988;Schink,1988;DingandSun,2005),exclusionofbacteria-grazingproto-zoaandmacrofauna(Lee,1992),adsorptiontomineralsurfaces(HedgesandKeil,1995;Henrichs,1995)oradependenceofrefractorycompounddegradationontheoverallsedimentmetabolicrate(Canﬁeld,1994).

HigherBARalsoenhancestheBEofBabio,butforadifferentreason.Whenburialisrapid,biogenicbaritepassesquicklyintothezoneofre-ducingporewatersinwhichittendstodissolve,butarelativelycloseddiageneticsystempreventsdiffusionofBa2+outofthesedimentandfa-cilitatesitssubsequentreprecipitationasadiageneticbaritephase(Dymondetal.,1992).

IncontrasttoTOCandBabio,organicPdoesnotappeartoshowasim-plelinearrelationshiptoBAR.IngallandVanCappellen(1990)observedthatTOC:Porgratiosvaryinacomplexmannerasafunctionofsedimen-tationrate,withlowandhighrates(b2andN1000cmkyr−1)associatedwithTOC:Porgof~100–200andintermediaterates(~10–250cmkyr−1)associatedwithTOC:Porgof~500–800.Theunderlyingcontrolonthispatternisnotcertainbutpossiblyrelatedtotheinterplayofsedimenta-tionrateswithredoxconditions.HighTOC:Porgratiosareassociatedwithreducingenvironments(AlgeoandIngall,2007),whichtendtohavesed-imentationratesbetweenthoseofhighlyoxidizingenvironmentsinopen-oceansettings(lowBAR)andthoseincontinent–marginsystems(highBAR)(IngallandVanCappellen,1990;Trompetal.,1995).AlthoughTOC/Porgincreaseswithburialdepth,thislargelyreﬂectsatransferofPfromorganictoauthigenicphases,resultinginlittlechangeinCorg/Ptotalratios(Filippelli,2001;AlgeoandIngall,2007).3.Methods

3.1.Compilationofmodernsedimentcoredatabase

Ourprimarygoalinthisstudyistoevaluatetherobustnessofthesed-imentcomponentsTOC,Porg,andBabioasproxiesforprimaryandexportproductivityinpaleomarinesystems.Tothisend,wegeneratedadata-basecontaining(1)TOC,Porg,Babio,andAlconcentrationsfor5914

32S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Fig.6.Inferredcovariantrelationshipsamongﬂuxes(f)ofCorg,Porg,andBabioasafunctionofprimaryproductivity(abscissa)andbenthicredoxconditions(ordinate).Curvedlinesrepresentisoﬂuxes,andarrowsindicatedirectionofincreasingﬂuxes.Theoretically,Corgaccumulationisfavoredbyhighproductivityandreducingconditions,andPorgandBabioaccumulationbyhighproductivityandoxidizingconditions.Intheredox-dominantsce-nario(A),redoxvariationexertsastrongerinﬂuenceonproxyﬂuxesthanproductivityvariation(reﬂectedinsubhorizontalisoﬂuxcontours),resultinginnegativecovariationbetweenfCorgandfPorgorfBa-xs.Intheproductivity-dominantscenario(B),productivityvariationexertsthestrongerinﬂuence(reﬂectedinsubverticalisoﬂuxcontours),poten-tiallyallowingpositivecovariationamongallthreeproductivityproxies(e.g.,alongeitherofthetrendsshownbycoloredarrows).Seetextforexamples.

individualsamplesfrom94sedimentarycores,representingarangeofoceaniclocalitiesanddepositionalenvironments(Fig.7),and(2)modernprimaryproductivityestimatesforthesame94sitesbasedonseveraltechniques(seeSection3.4).ThecoresusedinthisstudyareallofCenozoic(b65Ma)andmostlyofQuaternary(b2.4Ma)age,allowingfordirectcomparisonsofsedimentproxieswithmodernprimaryproduc-tivitydata.Thesedimentgeochemicaldatawerecompiledfrom37refer-encesthataresummarizedinTable1,whichwereeitherdownloadedfromtheonlinePANGAEAdatabase(pangaea.de)orextractedfrompub-lishedsources.Toassesstheeffectsofredoxconditionsonproductivityproxies,eachsitewasassignedaredoxclassiﬁcationbasedonitslocation.SitesbelowthechemoclineoftheBlackSea,CariacoBasin,andSaanichInletwereclassiﬁedasanoxic(dissolvedO2=0mLL−1),thoseinupwell-ingzonesoftheArabianSeaandtheWestAfricancoastwereclassiﬁedassuboxic(dissolvedO2N0to2mLL−1),andtheremaining(mainlyopen-ocean)siteswereclassiﬁedasoxic(dissolvedO2N2mLL−1).3.2.CalculationoforganicPandbiogenicBa

Weusedorganicphosphorus(Porg)ratherthantotalphosphorus(PorPtotal)inourﬂuxcalculationsinordertomoreaccuratelyassessbiogenicPﬂuxes.WecalculatedPorgfromtotalPbysubtractinga

detritalphosphorusfraction(Pdetr)estimatedfromeachsample'sAlcontentasfollows:h

org¼½Ptotal󰀂–½Al󰀂ÂðP=AlÞdetr:

ð1Þ

WeassumedadetritalP/Alratioof0.0087basedontheaveragePandAlconcentrationsofuppercontinentalcrust(McLennan,2001).Be-causedetritalPcomprisedb5%oftotalPinallofthestudyunits,minoruncertaintiesregardingthechoiceofdetritalP/AlratiohavelittleeffectoncalculatedPorgﬂuxes.Indeed,thecalculatedproportionsofdetritalParesosmallinmostopen-oceansettingsastobeeffectivelynegligible.Thiscalculationassumesthatallnon-detritalPwasultimatelyderivedfrommarinephytoplanktonbiomass.ThisinferenceisgenerallytrueofPadsorbedontoFe-oxyhydroxidesorpreservedinauthigenicphos-phate(Kraal,2010),butitdoesnottakeintoaccountPcontributionsfrombiogenicﬂuorapatite,e.g.,ﬁshscalesandbones(PbioinFig.1B),whichmaybequantitativelyimportantinsomesamples(Fig.2).

BiogenicBawasdeterminedbycalculatingtheamountofbariuminexcess(Baxs)oftheexpecteddetritalBaconcentration(Badetr)asfol-lows:

½Baxs󰀂¼½Batotal󰀂–½Al󰀂ÂðBa=AlÞdetr:

ð2Þ

WeassumedthatBabioisequivalenttoBaxs,i.e.,thatallnon-detritalBaisbiogenicinorigin,althoughasmallfractionofnon-detritalBamayderivefromBaadsorbedontocarbonatesorferromanganeseoxyhydroxides.Theseadsorbedfractionsmaybequantitativelymoreimportantinsedimentsfromhigh-productivityzonessuchastheequa-torialPaciﬁc,whereBainbiogenicbariteaccountsforjust~70%ofBaxs(Eagleetal.,2003).

BecausedetritalBacanrepresentalargefractionoftotalBa(some-timesN50%),thechoiceofasuitabledetritalBa/AlratioisimportantforcorrectestimationofBabio.Althoughtheinﬂuenceof(Ba/Al)detron[Babio]isrelativelyminorintheabyssaloceanwheresedimentsarepri-marilybiogenic,calculatedvaluesof[Babio]arequitesensitivetovaria-tionin(Ba/Al)detrinsiliciclasticsedimentsfrommarginalmarinesettings(Dymondetal.,1992).Compilationsofcrustalcompositiondatayield(Ba/Al)detrratiosbetween0.005and0.010(e.g.,TaylorandMcLennan,1985),whichhavebeenusedtoestimateBabio(orBaxs)inmanystudies(Dymondetal.,1992;Nürnbergetal.,1997;Bonnetal.,1998;PrakashBabuetal.,2002).However,asubstantialfractionofBaindetritalsedimentsappearstobelostduringweatheringandtransportintheterrestrialenvironment,yieldingsiliciclastic(Ba/Al)detrratiosaround0.002–0.004upondepositioninmarinesystems(Rutschetal.,1995;Reitzetal.,2004).Thus,commonlyused(Ba/Al)detrratiosbasedonaverageuppercrustalcompositionsmayoverestimate(Ba/Al)detrandunderestimateBabio(orBaxs).

TheaccuracyofBabioestimatescanbeimprovedbyusingformation-speciﬁc(Ba/Al)detrratios.Inmodernmarinesystems,thiscanbedonebyﬁndingtheinterceptofanexponentialregressionlineonaBa/Alvs.waterdepthcrossplot(Klumpetal.,2000;Pfeiferetal.,2001)orbydetailedmodelingofvariousdetritalcontributionsintheareaofinterest(Pirrungetal.,2008).Asthesemethodsaredif-ﬁculttoapplytopaleomarinesystems,weusedanapproachbasedonestimating(Ba/Al)detrratiosfromAlvs.Bacrossplots,inwhichahighconcentrationofsamplesalongalinethatpassesthroughtheori-ginisassumedtorepresentthedetritalcomponentofBa(seeShenetal.,inreview,forexamples).Thismethodyielded(Ba/Al)detrratiosforspeciﬁcstudyunitsrangingfrom0.0032to0.0046,whichareconsis-tentwiththesiliciclasticBa/AlratiosreportedbyRutschetal.(1995)andReitzetal.(2004).Thismethodisconservative,sinceitassumesthatthesampleswiththelowestBa/Alratioscontainnobiogenicbari-um.However,onemustexercisecautioninmarinesystemsinwhichproductivityishighandstronglycorrelatedwithdetritalinput,andalargeproportionofbariumisintheformofbiogenicbarite.Insuch

Fig.7.Worldmapshowinglocationsofsitesusedinthisstudy,andtheirredoxcategory.SitenumberscorrespondtothoseusedinTable1.

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–523334S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Table1

Summaryofsitesusedinthisstudy.Then=columnindicatesthenumberofindividualsamplesfromeachsite.EstimatesofmodernproductivityatagivensitewerebasedontheoceanicprovincesofLonghurstetal.(1995;seeTable2).

Site123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

ODPLeg882(Hole)S-2andAC-2PC72

199-1218199-1218199-1218199-1218199-1218199-1218199-1219199-1218199-121828-26632-31062-4637-62A7-657-66A85-57289-5869-77GPC-3

CoreMD97-2140138-844B138-846Hole803DHole804B/CHole806BHole807A/CHole844

Hole846BHole851BHole34-319Aria-2Aria-5Aria-6Aria-8Aria-13

CoreRC11-210GS7202-35Site92-598W8709A-1BC

W8709A-8PCand8TCW8709A-13PCand13TCODPLeg169S(Hole1033B)ODPLeg169S(Hole1034B)ODPLeg169S(Hole1033E)112-679D112-681B112-688117-723117-7241756-51768-81772-82082-11754-11754-21756-51768-81575-11648-11821-6121-752AMD90940MD962073

Location

DetroitSeamountShatskyRise

AbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcSEIndianOceanridgeCentralPaciﬁcseamountCentralPaciﬁcseamountCarolineBasin

AbyssalEquatorialPaciﬁcAbyssalEquatorialPaciﬁcEastPaciﬁcRise

Ontong-JavaPlateauEastPaciﬁcRise

AbyssalCentralPaciﬁcCarolineBasinEastPaciﬁcRiseEastPaciﬁcRise

Ontong-JavaPlateauOntong-JavaPlateauOntong-JavaPlateauOntong-JavaPlateauEastPaciﬁcRise

EastPaciﬁcRiseEastPaciﬁcRise

BauerDeep,SEPaciﬁcEastPaciﬁcRiseEastPaciﬁcRiseEastPaciﬁcRiseEastPaciﬁcRiseEastPaciﬁcRise

AbyssalEquatorialPaciﬁcEastPaciﬁcRiseEastPaciﬁcRiseCaliforniamarginCaliforniamarginCaliforniamarginSaanichInletSaanichInletSaanichInlet

PerumarginPerumarginPerumarginOmanmarginOmanmargin

AbyssalSouthernOceanAntarcticmargin

AbyssalSouthernOceanAbyssalSouthernOceanMid-AtlanticRidgeMid-AtlanticRidgeMid-AtlanticRidgeMid-AtlanticRidgeAntarcticmarginAntarcticmarginAntarcticmarginBrokenRidgeMadingleyRiseSocotra

Latitude50.3533.360.118.898.898.898.898.898.897.808.898.89−56.4036.8721.351.874.352.381.44−0.500.4830.322.077.92−3.102.43

1.000.323.637.923.102.77−13.02−19.42−19.40−19.39−18.93−19.50

1.82−14.47−19.0041.5442.2642.1248.5948.6348.59−11.06−10.99−11.5418.0518.46−48.63−52.59−55.46−43.22−46.77−46.77−48.90−52.59−62.85−69.74−67.07−30.89

5.3410.94

Longitude167.58159.13−139.40−135.67−135.67−135.67−135.67−135.67−135.67−142.02−135.67−135.67110.11176.90174.67141.94176.99−166.12−113.84158.50−133.23−157.67142.27−90.48−90.82160.54161.59159.36156.63−90.48−90.82−110.57−101.52−119.83−119.88−119.81−116.84−114.96−140.05−113.50−124.68−131.96−127.68−125.75−123.50−123.50−123.50−78.27−77.96−78.9457.6157.796.714.481.1611.717.617.596.714.48−43.34−6.6937.4893.5861.4152.62

Waterdepth(m)3244310742984827.24827.24827.24827.24827.24827.25063.44827.24827.2416735162532259161305293389322074291570525473414.5329634103861252028063415329637604296365436803600335034354420304436993680311127122382032384621623829.880860038283299413746102519253438283299346125294027109738753142

n105613124517842199416135116376101291512810101692373222620159243568682164675174181502570512937203235448703371296855212388170465967

RedoxcategoryOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicAnoxicAnoxicAnoxicSuboxicSuboxicSuboxicSuboxicSuboxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxicOxic

Longhurstetal.(1995)provincePSAWNPSWPEQDPNECPNECPNECPNECPNECPNECPNECPNECPNECANTANPSENPTGWARMWARMWARMPEQDWARMPEQDNPSEWARMPNECPEQDWARMWARMWARMWARMPNECPNECPEQDSPSGSPSGSPSGSPSGSPSGSPSGPEQDSPSGSPSGOCALOCALCCALn/an/an/aCHIL*CHIL*CHILARABARABSANTSANTANTASSTCSANTSANTSANTSANTAPLRAPLRAPLRISSGMONSMONS

Reference

Jaccardetal.(2009)

AmoandMinagawa(2003)Murrayetal.(2000,2012)LyleandLyle(2005)LyleandLyle(2005)LyleandLyle(2005)LyleandLyle(2005)LyleandLyle(2005)LyleandLyle(2005)LyleandLyle(2005)LyleandLyle(2005)LyleandLyle(2005)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Moodyetal.(1988)Thevenonetal.(2004)

FilippelliandDelaney(1995)Emeisetal.(1995)

FilippelliandDelaney(1996),DelaneyandFilippelli(1994)FilippelliandDelaney(1996),DelaneyandFilippelli(1994)FilippelliandDelaney(1996),DelaneyandFilippelli(1994)FilippelliandDelaney(1996),DelaneyandFilippelli(1994)FilippelliandDelaney(1996),DelaneyandFilippelli(1994)FilippelliandDelaney(1996)FilippelliandDelaney(1996)Dymondetal.(1977)

LeinenandGraybeal(1986)LeinenandGraybeal(1986)LeinenandGraybeal(1986)LeinenandGraybeal(1986)LeinenandGraybeal(1986)Reaetal.(1991)Millsetal.(2010)

RuhlinandOwen(1986)Lyleetal.(1992)Lyleetal.(1992)Lyleetal.(1992)

RussellandMorford(2001),Filippelli(2001)Filippelli(2001)Filippelli(2001)

Lückgeetal.(1996)Lückgeetal.(1996)Lückgeetal.(1996)Lückgeetal.(1996)Lückgeetal.(1996)Nürnbergetal.(1997)Nürnbergetal.(1997)Nürnbergetal.(1997)Nürnbergetal.(1997)Franketal.(2000)Franketal.(2000)Franketal.(2000)Franketal.(2000)Bonnetal.(1998)Bonnetal.(1998)Bonnetal.(1998)

OwenandZimmerman(1991)DesCombesetal.(1999)DesCombesetal.(2005)

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Table1(continued)

Site67686970717273747576777879808182838485868788899091929394

115-711AODPLeg117(Hole722B)ODPLeg117(Hole724C)ODPLeg162(Site983)GeoB4301-1GeoB4242-4GeoB9526-5GeoB9527-5GeoB6518-1Hole1082Hole1084Hole1085Hole1087175-1085AODPSite1085ODPLet165(Hole1002)M40/4_SL87Core1430Core1432Core1436Core1437Core1440Core1443Core1452Core1462Core1470Core1472Core1484

Location

AbyssalNWIndianOceanOwenRidgeOwenRidgeReykjanesRidgeCanaryIslandsCanaryIslandsCapeVerdeCapeVerdeCongoFan

SWAfricanmarginSWAfricanmarginSWAfricanmarginSWAfricanmarginSWAfricanmarginSWAfricanmarginCariacoBasinBalearicBasinBlackSeaBlackSeaBlackSeaBlackSeaBlackSeaBlackSeaBlackSeaBlackSeaBlackSeaBlackSeaBlackSea

Latitude−2.7416.6218.4660.4029.1529.6812.4412.44−5.59−21.09−25.51−29.37−32.47−29.37−29.3710.7138.9941.4343.0143.4041.6942.2044.5942.7843.0542.0743.1544.70

Longitude

61.1659.8057.79−23.64−15.50−17.89−18.06−18.2211.2211.8213.0313.9915.3113.9913.99−65.174.0229.4334.0836.6036.4734.3631.9228.6033.0441.2739.9136.89

Waterdepth(m)443920285931983.736144292323136719621280.61992.21713.11371.61713.21713.1892.91897663224821589732071057728218610681588386

n22147155651178659144201820203373323513051518723242017151612

RedoxcategoryOxicSuboxicSuboxicOxicOxicOxicSuboxicSuboxicSuboxicSuboxicSuboxicSuboxicSuboxicSuboxicSuboxicAnoxicOxicAnoxicAnoxicAnoxicAnoxicAnoxicAnoxicAnoxicAnoxicAnoxicAnoxicAnoxic

Longhurstetal.(1995)provinceMONSARABARABARCTNASENASENATRNATRGUINBENGBENGBENGBENGBENGBENGGUIAMEDIMEDIMEDIMEDIMEDIMEDIMEDIMEDIMEDIMEDIMEDIMEDI

Reference

BostromandBackman(1990)ShimmieldandMowbray(1991)ShimmieldandMowbray(1991)Hyunetal.(1999)

Freudenthaletal.(2001)Freudenthaletal.(2001)

ZarriessandMackensen(2010)ZarriessandMackensen(2010)Weijersetal.(2009)Giraudeauetal.(2002)Giraudeauetal.(2002)Giraudeauetal.(2002)Giraudeauetal.(2002)Diester-Haassetal.(2001)Murrayetal.(2002)Yarinciketal.(2000)Weldeabetal.(2003)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)Hirst(1974)

*ThesesiteswereslightlyinshoreoftheCHILregionasdeﬁnedbycoordinates,andarereferredtoCHILforallpurposes.

cases,anarrayofsamplesexhibitingpositivecovariationofBawithAlmaynotreﬂectthedetritalBacomponent.Forexample,abyssalsamplesfromthemodernequatorialPaciﬁcyieldawell-deﬁnedarraywithaBa/Alratioof~0.32(Murrayetal.,2000),whichisnearlytwoordersofmagnitudehigherthantheaveragecrustal(Ba/Al)detrratioof0.0065(McLennan,2001).Inthiscase,theobservedstrongpositivecovariationofBaandAlprobablyreﬂectsaclosecouplingbetweenEoliandetritalinputandprimaryproductivity.3.3.Calculationofproductivityproxyﬂuxes

Thecalculationofmassaccumulationrates(MAR)fortheproductiv-ityproxiesusedinthisstudy(TOC,Porg,andBabio)isanimprovementovertheuseofraworAl-normalizedelementalconcentrationsoren-richmentfactors,whichcanbestronglyinﬂuencedbydilutioneffects(Tribovillardetal.,2006).Inthefollowinganalysis,wereportMARsforTOC,Porg,andBabiointhesameunitsofﬂuxperunitarea(i.e.,mgcm−2kyr−1)withwhichmarineproductivityisreportedinthisstudy.MARswerecalculatedbymultiplyingtheweightfractions[TOC],[Porg],and[Babio]bythebulkaccumulationrate(BAR)foragivensample:MARðXÞ¼½X󰀂ÂBAR

ð3Þ

Inordertocalculateproductivity-proxyMARs,itisnecessarytomeasureorestimatevaluesforρandLSRforeachsample.Forthemajor-ityofthe94sitesexaminedhere,ρwasreportedinthesourcepublica-tion.Fortheremainingsites,weusedanexponentialfunctiontoestimateρasafunctionofsedimentage:ρ¼0:0794ÂlnðxÞþ0:650

ð5Þ

whereXistheproductivityproxyofinterest(TOC,Porg,orBabio)andBARiscalculatedas:BAR¼ρÂLSR

ð4Þ

whereρissedimentdrybulkdensity(inunitsofgcm−3)andLSRislinearsedimentationrate(inunitsofcmkyr−1;Algeoetal.,2011,2013).Henceforth,wewillrefertotheMARsoforganicC,organicP,andbiogenicBaasOCAR,PAR,andBaAR,respectively.

whereρissedimentdrybulkdensity(inunitsofgcm−3)andxistheageofthesample(inunitsofkyr,orthousandsofyears).TherelationshipinEq.(5),whichyieldsastandarddeviationof0.08gcm−3andr2=0.90,wasderivedfromseveraldrybulkden-sitydatasets(MurrayandPrell,1991;Lyleetal.,1992;Freudenthaletal.,2001;AmoandMinagawa,2003;Gallego-Torresetal.,2007).Wefoundthatbulkdensityshowedabetterrelationshiptosedi-mentagethantoburialdepth.

Calculationoflinearsedimentationratesrequiresanage-depthmodelforeachsedimentcorethatistypicallybasedonaseriesofbiostratigraphicallyorradiometricallydatedtiepoints,betweeneachpairofwhichsedimentationratesareassumedtobelinear.Theaccura-cyofthisprocedureislimitedbythenumberofagetiepointsandtheageuncertaintyattachedtoeach.Ingeneral,theeffectofassuminglin-earsedimentationratesbetweentiepointsistoevenoutshort-termvariabilityinsedimentaccumulation.Withincreasingburialdepth,compactioncausesareductioninLSRs,butacorrespondingincreaseinthedrybulkdensityofthesedimentcompensatesforthiseffectandyieldsanunchangedmassaccumulationrate,assumingnolossorgainofmaterialinthesediment.

Anage-depthmodelforeachofthe94studyunitswasconstructedfromageanddepthinformationprovidedintheoriginalpublishedsource,fromwhichaLSRwasdeterminedforeachsample.Formanyofthestudyunits,highlydetailedage-depthmodelscouldbeconstruct-ed,e.g.,owingtotheavailabilityofoxygen–isotopecorrelationslinked

36S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

tothestandardmarineisotopestage(MIS)stratigraphy,forwhichcalibratedastrochronologiesarenowavailable(LiseckiandRaymo,2005).Inaddition,deep-oceansuccessionstendtoberelativelymorecompleteandtocontainfewertemporalgapsthanshallow-marineandcontinentalsuccessions(Sadler,1981),introducingfeweruncer-taintiesintotheage-depthmodelsforsuchunits.

3.4.Modernoceanicprimaryproductivityandexportproductivityestimates

WecomparedourcalculatedvaluesofOCAR,PAR,andBaARwithsite-speciﬁcestimatesofmodernmarineproductivityforthe94studysites.Inordertoassignaproductivityvaluetoeachsite,weusedestimatesderivedfromthreesources:(1)satellitemeasurementsofchlorophyll‘a’concentrationsfor57oceanicprovinces(Longhurstetal.,1995),(2)122ﬁeldmeasurementsofprimaryproductivitybasedonseveralmethods,ascompiledbyDunneetal.(2005),and(3)N3000ﬁeldmeasurementsofproductivitybasedonthe14Cuptakemethod,fromanOregonStateUniversitydatabase(www.science.oregonstate.edu/ocean.productivity/).ThesedatasetswillbereferredtosubsequentlyasL95,D05,andOS,respectively.Longhurstetal.(1995)andLonghurst(2010)undertookadetailedanalysisofspatialvariationinchlorophyll‘a’concentrationsinthemodernocean,fromwhichtheyidentiﬁed57irregularlyshapedoceanicprovinceswithin-ternallyconsistentproductivityranges.Sinceproductivitymeasure-mentsfromthethreesourcesabovewererarelyavailablefortheexactlocationofeachstudycoreinourdatabase,weemployedtheocean-provinceframeworkofL95toassignsite-speciﬁcestimatesofprimaryproductivityandexportproductivity(Table2;notethatour94studysitesarelocatedinjust25ofthe57oceanicprovincesofL95).InordertoaccuratelylocatethestudysitesrelativetotheL95oceanicprovinces,wemadeuseofGoogleEarthandashapeﬁledepictingtheprovinceboundariesthatisavailablefrombwww.ecomarres.comN.BecausenoneofthesourcesabovecontainedproductivityestimatesforSaanichInlet,wemadeuseofestimatesfromTimothyandSoon(2001)basedonthe14CuptakemethodandincludedthemintheOScompilation.Inthisstudy,‘primaryproductivity’isdeﬁnedastherateofﬁxationofatmosphericordissolvedinorganiccarbonasorganicmatterinthephoticzoneoftheocean,and‘exportproductivity’astherateatwhichsinkingmarineorganicmatteristransferredtotheoceanicthermocline(below~200mwaterdepth).Whilephytoplanktonbiomassisnotcom-prisedentirelyofcarbon,productivityistypicallyexpressedasthemassﬂuxoforganiccarbonperunitareainordertofacilitatecomparisonsacrossenvironmentswithdifferentprimaryproducers.TheproductivityestimatesinthisstudyareallreportedinunitsofmgCcm−2kyr−1.

Netprimaryproduction(NPP)isdeﬁnedastherateatwhichprima-ryproducers(phytoplanktoninmarinesystems)assimilatecarbonintotheirbodiesforpurposesoflong-termgrowth,ratherthancarbonﬁxa-tionthatisusedtosupportrespiration.Thisisincontrasttogrossprima-ryproduction(GPP),whichincludesallcarbonﬁxationbyprimaryproducers,includingcarbonthatisquicklyrespiredtosupportmetabol-icprocesses.The14Cuptakemethod,whichwasusedintheinsituOSproductivityestimatesandincalibratingtheL95satellite-basedesti-mates,hasbeenthesubjectofsomediscussionaboutwhethermeasure-mentsmorecloselyapproximateNPPorGPP.UnlikemeasurementsofO2inincubations,whereoxygencanbebothproducedandconsumed,14Cuptakemeasurementsarealwayspositive,indicatingthattheyfallsomewherebetweenNPPandGPP(seeMarra,2009).Sincemanyphy-toplanktonreassimilaterespiredcarbonbutnotrespiredoxygen(Marra,2008),14CuptakehasbeencontrastedwithestimatesofGPPderivedfromoxygenproductioninordertodetermineNPPasafractionofGPP(Hashimotoetal.,2005).Thus,itappearsthat14Cuptakemea-surementscloselyapproximatenetprimaryproduction(NPP)inmostnaturalsystemswhenincubatedoverafulldiurnal(24-hour)cycle(FaheyandKnapp,2007;Marra,2009).Sincetheproductivityestimatesusedinthisstudyareeitherdirectly(OS)orindirectly(L95)derived

from14Cuptake,theymostcloselyapproximatenetprimaryproduction(NPP).

Theexportratio,i.e.,thefractionofphotosyntheticallyﬁxedcar-bonthatisexportedfromthesurfacemixedlayer,canvaryoverawiderange(from4%to72%accordingtoDunneetal.(2005)).Duetothisextremevariability,sedimentaryproxiessuchasOCAR,PAR,andBaARcanbeexpectedtoexhibitastrongerrelationshiptoexportproductivitythantoprimaryproductivity.Weestimatedmodernex-portproductivityﬂuxesforour94studysitesfromtheD05dataset,whichcontainspairedmeasurementsofprimaryandexportproduc-tivity,allowingtheexportratiotobedeterminedforeachsiteinthatstudy.WeconvertedthesedatatotheL95ocean-provinceframe-workinordertominimizetheeffectsofseasonalandsmall-scalespatialvariabilityandtomaintainconsistencywithourprimarypro-ductivityestimates(Table2).WethencalculatedanaverageexportproductivityﬂuxandexportratioforeachL95province(notethatnoexportproductivitydatawereavailableinD05for12ofthe25oceanicprovincesconsideredinthisstudy).

Incomparingmodernproductivityestimateswithproductivity-proxyMARsinthesedimentrecord,itisimportanttokeepinminddif-ferencesinspatialandtemporalscalesbetweenthesedatasets.Oures-timatesofmodernproductivityﬂuxesarebasedonaveragesforlargeoceanicareas(i.e.,the57oceanicprovincesofL95),withineachofwhichsomespatialvariationexists,whereaseachofthe94studycoresrecordsproductivityatasingle,speciﬁcsite.Allmodernproduc-tivitymeasurementsareinstantaneousvaluesinageologicsense,evenwhenintegratedoveraseriesofyearsasfortheL95dataset.Incontrast,marinesedimentsrecordatime-averagedsignalinwhichasinglecentimetercanintegratehundredstothousandsofyearsofproductivityvariation,andsubstantialsmoothingoftheproductivitysignalcanresultfrompost-depositionalbioturbationofthesediment(e.g.,Wheatcroft,1990).Anotherconsiderationisthatallofthereport-edmodernproductivitymeasurementsweremadewithinthepasthalf-centuryand,thus,potentiallyrecordanthropogenicinﬂuencesonnutri-entcyclesandmarineecosystems(Gallowayetal.,2004;Behrenfeldetal.,2006).Ontheotherhand,someofthesedimentaryproductivityﬂuxestimatesrepresentconditionsduringQuaternaryglacialstagesorpriortotheonsetofNorthernHemisphereglaciationat~2.5Ma,i.e.,timeswhenoceancirculationintensitywassomewhatdifferentthanatpresent(Raymoetal.,1992).Asaresult,modernproductivityestimatesarenotnecessarilyrepresentativeofproductivityconditionsinagivenoceanicregionthousandsormillionsofyearsago.Suchchangesovertimearelikelytoaccountformuchofthedivergencebe-tweenthemodernin-situandancientsediment-basedproductivityes-timatesofthisstudy.

3.5.Calculationofpreservationfactors(PF)

Preservationfactors(PFs)representthefractionofeitherprimaryorexportproductivitythatisultimatelyburied(andthuspreserved)inthesedimentaryrecord(Trask,1953;BralowerandThierstein,1987).Itscomplement,1−PF,representsthecumulativelossofagivencompo-nentduetoremineralizationinthewatercolumnand/orthesediment.TherelationshipofPFtoproductivityisgivenbyEq.(6)fororganicCandEq.(7)fororganicP:PFZ¼OCAR=PRODZ

ð6Þ

PFZ¼PARÂðC=PÞorg=PRODZð7Þ

wherePRODZiseitherprimaryproductivity(PRODprim)orexportpro-ductivity(PRODexp),asquantiﬁedforaspeciﬁcstudysitefromtheL95,D05,orOSdataset.Becauseexportproductivityisalwaysafractionofprimaryproductivity,aﬁxedamountofagivencomponentinthe

Table2

ProductivityandexportproductionvaluescompiledwithintheregionalframeworkdeﬁnedbyLonghurstetal.(1995).ParticleexportratioswerecalculatedfromthemeanproductivityandexportdataintheD05dataset,whereprimaryandexportproductionwereseparatemeasurements.ExportproductionfortheL95andOSdatasetsisanestimatemadebymultiplyingthispe-ratiobythemeanprimaryproductivityineachprovince.SinceL95productivityestimatesaresatellitebasedandtimeintegrated,theyhavenostandarddeviation.SincethisprogramdoesnotdistinguishtheOCALandCCALprovinces,thesetwowerecombinedintheD05andOSdatasets.AseparateproductivityvaluewasusedforOCALintheL95dataset.Provincename

pe-ratio

Dunneetal.(2005)(D05)Primaryproduction(mgCcm−2kyr−1)Mean

AntarcticAustralPolar

NorthwestArabianUpwellingAtlanticArctic

BenguelaCurrentCoastalCaliforniaUpwellingCoastalChile–PeruCurrentCoastalGuianaCurrentCoastalGuineaCurrentCoastal

IndianOceanSouthSubtropicalGyreMediterraneanandBlackSeasIndianOceanMonsoonGyres

NorthAtlanticSubtropicalGyreEastNorthAtlanticTropicalGyre

NorthPaciﬁcSubtropicalGyreEastNorthPaciﬁcSubtropicalGyreWestNorthPaciﬁcTropicalGyreOffshoreCaliforniaCurrent⁎PaciﬁcEquatorialDivergence

NorthPaciﬁcEquatorialCountercurrentPaciﬁcSubarcticGyreWestSub-Antarctic

SouthPaciﬁcSubtropicalGyreSouthSubtropicalConvergenceWesternPaciﬁcWarmPool

38.45%45.12%20.48%71.15%–

31.68%32.09%––––

15.19%8.94%–––7.84%31.68%18.06%14.89%––9.01%–

13.59%

47,71822,41939,42022,564–

25,51586,783––––

23,23026,609––––

25,51557,97515,790––

16,796–

30,828

Std.deviation30,10813,64513,709––

10,7794463––––

10,2718033––––

10,77972,8381783––

11,053–6406

Exportproduction(mgCcm−2kyr−1)Mean17,98411,177764416,055–726627,672––––41072472––––726613,3442289––1652–4282

Std.deviation13,81078584705––32545526––––41721333––––325422,477675––1214–1773

11781–42––––22––––4123––5–2n=

Oregonstate(OS)Primaryproduction(mgCcm−2kyr−1)Mean3278–

42,80019,581–

37,94267,064–

54,473–

19,7022756–

22,401–142814,02337,94221,88010,542–84909297–9551

Std.deviation1638–

23,92618,618–

68,26151,273––

15,1079536–

20,814–2475719568,26116,4145526–42738036–7159

1260–876513,933–

12,01921,522––––419––––110012,01939511569––837–1298

11–223–937123–1–9741–16–235093713187–1530–40

16,50039,80045,40048,40032,30038,80026,90069,90049,500710021,60010,50012,20010,60011,10010,900590011,70011,30010,70026,40012,000870013,6008200

634417,959929834,438–

12,2918633––––15951091–––463370620401593––784–1114

Exportproduction(mgCcm−2kyr−1)

Longhurstetal.(1995)(L95)Primaryproduction(mgCcm−2kyr−1)

Exportproduction(mgCcm−2kyr−1)

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52⁎Softwareavailablethroughbhttp://www.lifewatch.be/Nwasusedtogrouplat-longcoordinatesintoLonghurstetal.(1995)provinces.

3738S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

sedimentwillyieldalargerPFfortheformer(exp)relativetothelatter(prim).

Basedoncalculatedprimaryandexportproductivityﬂuxes(Section3.4),wedeterminedpreservationfactors(PFs)forallstudysamplesforwhichTOCdatawereavailable.ForafewsamplesfromSaanichInlet,PFprimexceeded100%,andthesevalueswereconvertedto100%fordisplaypurposes.Thisresultisnotnecessarilyimpossibleasshort-termorganiccarbonaccumulationrates(asrecordedinthesedimentrecord)easilymightexceedtheaveragelong-termprimaryproductivityrateforagivensite.4.Results

4.1.Robustnessofmodernproductivityestimates

Weevaluatedthedegreeofagreementbetweenthetwoprincipalmethodsofmodernproductivityestimation(i.e.,chlorophyll‘a’and14Cuptake)foroceanicprovinceshavingproductivityestimatesbasedonbothmethods.Thetwodatasetsbasedon14Cuptakemeasurementsinpart(D05)orinfull(OS)showedsomeagreement,withanr2of0.37(n=11).TheOS14C-basedandL95satellite-basedestimates,whichwerebothbasedonamuchlargernumberofmeasurementsthantheD05dataset,showedthestrongestagreement,withanr2of0.39(n=17).However,therewasvirtuallynorelationshipbetweentheL95andD05datasets(r2b0.01,n=13),largelybecausetheD05datasetdetectedmuchgreaterproductivitythantheL95datasetincoastalup-wellingzones,theequatorialPaciﬁc,andtheSouthernOcean.Sincehighproductivitydrivesagreatdealofthescientiﬁcinterestinthesesystems,itispossiblethattheinsitupointmeasurementsusedintheD05datasetreﬂectaspatialortemporalselectionbias,whiletheL95datasetintegrateslargerareasoverlonger(multi-annual)timescales,thusyieldingloweraverageproductivityestimates.

Whereasalldatasetsexhibitasimilarrangeofproductivityvalues(approximately10,000to60,000mgCcm−2kyr−1),theOSdatasetin-cludesamuchlargernumberofindividualsamples(n=3170)thantheD05dataset,makingitmorelikelytoaverageoutseasonal,climatic,andspatialvariabilitywithineachoceanicprovince.Althoughtheoce-anicprovincesofL95arebasedoninternallyuniformproductivityvalues,itisworthnotingthatthe14C-uptakesamplepointstakentorepresentagivenregionsometimescoveronlyasmallfractionofaprovince'stotalareaand,hence,ofitsinternalvariationinproductivity.Basedontheseconsiderationsandthesuperioragreementbetweenthem(Fig.8),weadoptedtheL95andOSproductivitydatasetsasourprimarysourcesofestimatesofmodernprimaryproductivity,reservingthesmallerD05datasetforestimatesofexportproductivity.4.2.Organiccarbonaccumulationrates(OCAR)

Redoxconditionsstronglyinﬂuencethepreservationoforganiccar-bonforboththeL95andOSproductivitydatasets.BoththemeanandmaximumOCARincreasewithdecreasingdissolvedoxygenlevels(Fig.9A).MeanOCARvaluesincreasebyafactorof~30×(1.5logunits)betweenoxicandsuboxicfacies,andbyanotherfactorof~30×betweensuboxicandanoxicfacies,implyingamajorroleforwater-columnredoxconditionsinorganiccarbonpreservation.OCARgeneral-lyrepresentsaminutefractionofprimaryproductioninoxicandsuboxicfacies,inwhichPFprimaverages~0.1%and~0.6%,respectively(Table3).TheL95andOSdatasetsdivergesigniﬁcantlyforanoxicfacies,inwhichPFprimaverages~0.9%(max.6.3%)intheL95datasetbut18%(max.100%,seebelow)intheOSdataset.ThisdifferenceislargelyduetotheinclusionofSaanichInlet,ananoxicfjordlocatedonVancou-verIsland,intheOSdataset,forwhichcalculatedPFprimexceeded100%foranumberofsamples.ExcludingSaanichInlet,theaveragePFprimfortheOSdatasetisreducedto1.1%(max.6.9%),whichisingoodagree-mentwiththeL95dataset.ForboththeL95andOSdatasets,thediffer-encesbetweenredoxfaciesweresigniﬁcantatthep(α)b0.01level

Fig.8.ComparisonofestimatesofproductivityfromtheLonghurstetal.(1995)(L95)datasetbasedonsatellitechlorophyllmeasurements,andtheOregonState(OS)datasetbasedoninsitumeasurementsusingthe14CuptakemethodintegratedoverthedepthofthephoticzoneandaveragedbyL95oceanicprovince.Pointsarelabeledwithcodein-dicatingwhichprovinceisrepresented.Forprovincecodesandstandarddeviationsasso-ciatedwithOSregionalaverages,seeTable2.

usingatwo-tailedStudent'st-test.Althoughexportratiosforanoxicfa-ciesinourdatabasewerenotavailableinD05,meanPFexpwashigherinthesuboxicthanintheoxicfacies(Table3).ForboththeL95andOSexportproductionestimates,thedifferencebetweenoxicandsuboxicsettingswasstatisticallysigniﬁcantatthep(α)b0.05levelusingatwo-tailedStudent'st-test.

OCARshowsastrongcorrelationtoBAR(r2=0.86),witharegres-sionslope(m)of1.72forthefulldataset(Fig.10).Whenconsideredbyredoxfacies,asteeperslope(m=1.76)wasobtainedforoxicsitesrelativetosuboxic/anoxicsites(m=1.11).Thispatternisbroadlycon-sistentwiththatobservedinHolocenedatabyTyson(2005),althoughtheregressionslopeswecalculatedfromN5000datapointsaresteeper,indicatingastrongerinﬂuenceofBARonorganiccarbonpreservationthanpreviouslyrecognized(seeSection2.2).Thelinearequationinlog–logspaceforourfulldataset(Fig.10)is:log10ðOCARÞ¼1:72Âlog10ðBARÞþ0:09

ð8Þ

whereOCARisinunitsofmgcm−2kyr−1andBARisinunitsofgcm−2kyr−1.Thisisequivalenttothefollowingexponentialequationinlinearspace:OCAR¼10

0:09

ÂBAR

1:72

:ð9Þ

SinceOCARisequaltoTOC×BAR×1000,withthefactorof1000×accountingfortheconversionofgtomg,theequationcanbesimpliﬁedbyeliminatingtheBAR×1000termfrombothsides.Thisoperationcan-celstheunitsofmassﬂuxandyieldsadimensionlessterm(TOC)asafunctionofBAR,andrenderingtheequationintermsofarealrelation-shipbetweentwofullyindependentvariables:TOC¼10

−3:09

ÂBAR

0:72

:ð10Þ

Evaluatedasafunctionofredoxfacies,differentregressionrelation-shipsbetweenOCARandBARareexhibitedbysuboxic/anoxicversusoxicsites(Fig.10):

log10ðOCARÞ¼1:11Âlog10ðBARÞþ0:97

ðsuboxic=anoxicÞ

ð11Þlog10ðOCARÞ¼1:76Âlog10ðBARÞþ0:01ðoxicÞ:

ð12Þ

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5239

Fig.9.(A)Organiccarbonaccumulationrate(OCAR),(B)phosphorusaccumulationrate(PAR),and(C)biogenicbariumaccumulationrate(BaAR)byredoxfacies.Boxesrepresentmeansandstandarddeviationranges,whereasthewhiskersshowthefullrangeofvaluesforeachredoxcategory.

Theserelationshipscanbetransformedinthesamemannerasforthefulldataset(i.e.,viaEqs.(9)and(10))toyieldexpressionsrelatingtwoindependentvariables(i.e.,TOCandBAR)foreachredoxfacies.

WhenoneconsidersOCARasafractionofprimaryorexportproduc-tion(i.e.,asgivenbythepreservationfactor,orPF),strongcorrelationsareobservedwithBAR(inagreementwiththeﬁndingsofFelix,2014).ThisrelationshipisstrongerwhenPFiscalculatedasapercentageof

primaryproduction(Fig.11A)ratherthanofexportproduction(Fig.11B).PFprimcovariespositivelywithBARinbothproductivitydatasets(r2=0.87forOS;r2=0.77forL95),withPFexpshowingasim-ilarbutsomewhatweakerrelationship(r2=0.78forOS;r2=0.56forL95).Sincepreservationfactorsarecalculatedfromthreefullyindepen-dentvariables(i.e.,TOC,BAR,andproductivity),thisrelationshipallowsprimaryproductivitytobeisolatedasafunctionofTOCandBAR,twoparametersthatcanbemeasuredorestimatedinmostpaleomarinesystems(seeSection3.3).

OCARshowspositivecovariationwithestimatesofprimaryproduc-tivity,withr2valuesof0.26(L95)and0.33(OS)(Table4,Fig.12A).Comparablecorrelations(r2=0.30,L95;r2=0.20,OS)areseenbe-tweenOCARandestimatesofexportproduction(Table4,Fig.12B).WhileweakerthanthecorrelationsobservedbetweenBARandOCAR,thesecorrelationsarebetweenfullyindependentparametersandlikelyindicatearealrelationshipbetweenproductivityandorganiccarbonaccumulation,albeitmodiﬁedbypreservationalfactors.4.3.Organicphosphorusaccumulationrates(PAR)

Redoxconditionsinﬂuencesedimentaryphosphorusaccumulation.MeanPARincreasesundermorereducingconditions,withaveragevaluesmorethanafactorof10×higherinanoxicfaciesthaninoxicfa-cies(Fig.9B).AveragePorgconcentrationspeakinsuboxicsystems(1035ppm),althoughthedifferencesinmeanconcentrationsbetweenredoxfaciesarenotsigniﬁcant(Fig.13).WhilethereisconsiderableoverlapbetweenredoxfaciesatlowerPorgconcentrations,themaxi-mumobservedvaluesdecreaseundermorereducingconditions.Thus,oxicfaciesoccasionallyyieldPorgconcentrationstoN5500ppm,where-asPorgconcentrationsinanoxicfaciesareuniformlyb2500ppm(Fig.13).Thesepatternsreﬂectthepartialretentionofremineralizedor-ganicphosphorusinoxidizedsedimentscontainingFe-oxyhydroxides(FilippelliandDelaney,1996;Delaney,1998;AlgeoandIngall,2007).

PARshowspositivecovariationwithBAR,yieldingaregressionslope(m)of1.07(r2=0.92)forthefulldataset(Fig.14A).Bothoxicandsuboxic/anoxicsitesexhibitstrongpositiverelationshipswithBAR,withmof1.04(r2=0.84)and0.93(r2=0.89),respectively.Thesecal-culationsexcludethedatafromMillsetal.(2010),whicharefromanoceanichydrothermalplumeregionandrepresentobviousoutliersinourdataset.WhenPorgconcentrationsareplottedagainstBAR,thereislittlerelationshipbetweenthevariables(r2=0.05),withtheformerdeﬁningaroughlyhorizontaltrendcenteredonPorg=102.75(~560)ppm(Fig.14B).

PARalsoshowspositivecovariationwithbothprimaryandexportproduction(Fig.15).PARcorrelatesmorestronglywiththeL95satellite-basedprimaryproductivitymeasurements(r2=0.30)thanwiththe14C-basedOSdataset(r2=0.03),largelyduetolowerproduc-tivityestimatesfortheSouthernandIndianoceansinthelattersource(Fig.15A).However,PARdoesnotshowanysystematicincreaseaboveproductivityvaluesof~20×103(or~104.3)mgCcm−2kyr−1.Incontrast,PARincreasesnearlymonotonicallywithorganiccarbonex-port,althoughtherelationshipisweaker(Fig.15B).CorrelationswiththeL95productivityestimatesarestronger(r2=0.19)thanthosewiththeOSdataset(r2=0.05).

4.4.Biogenicbariumaccumulationrates(BaAR)

BaARexhibitsanunusualrelationshiptoredoxfacies.AlthoughthesuboxicfaciesexhibitsthehighestmeanBaAR(~4mgcm−2kyr−1),thehighestpeakvalues(N30mgcm−2kyr−1)arefoundintheoxicfacies(Fig.9C).However,therangeofBaARsfortheoxic,suboxic,andanoxicfaciesisnotsigniﬁcantlydifferent.

BaARisnotstronglyinﬂuencedbyBAR.Forthefulldataset,therela-tionshipbetweenBARandBabioaccumulationisstatisticallyinsigniﬁ-cant(m=0.16;r2=0.05;Fig.16).Individualredoxfaciesexhibitsomewhatstrongerrelationships,withoxicfaciesyieldingmof0.35

40S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Table3

Preservationfactor(PF)calculatedasaproportionofprimaryproductionandasaproportionexportproduction,fortheL95andOSproductivityestimates,andforeachofthethreeredoxcategoriesusedinthisstudy.

OxicL95

PreservationfactorPrimaryproductionMean

Std.deviationMinMaxn=

ExportproductionMean

Std.deviationMinMaxn=

SuboxicL95

AnoxicL95

0.07%0.14%b0.00%1.57%29790.38%0.86%b0.00%10.32%2788

0.12%0.46%b0.00%5.97%25800.69%3.14%b0.00%39.32%2389

0.56%0.67%0.03%5.91%7544.13%4.38%0.46%18.42%17

0.64%0.53%0.06%2.37%3062.96%2.91%0.19%9.70%17

0.94%0.89%b0.00%6.33%407––––0

18.17%(1.10%)27.30%(1.24%)b0.00%

100.00%(6.94%)258(172)––––0

(r2=0.20)andsuboxic/anoxicfaciesyieldingmof0.57(r2=0.12).AlthoughtheslopesoftheBaAR–BARrelationshipsfortheseredoxfa-ciesarenottoodifferent;adistinctoffsetofthedatadistributionsisev-ident:foragivenBAR,oxicfaciesyieldsigniﬁcantlyhigherBaARvaluesthansuboxic/anoxicfacies(Fig.16).Thispatternislikelytoberelatedtoredoxcontrolsonbiogenicbariteaccumulation,speciﬁcally,moreefﬁ-cientpreservationofbariteunderoxidizingconditions(seeSection2.4).

BaARshowslittlerelationshiptoeitherprimaryorexportproductiv-ity,yieldingr2b0.05forboththeL95andOSdatasets(Fig.17).ApplyingtheDymondetal.(1992)correctionforsedimentationrate-enhancedpreservation(seeSection5.4)toourBaARestimatesimprovedthecor-relationwithprimaryandexportproduction(althoughr2valuesremainedb0.15),largelybyincreasingtheestimatedBaARatlowBARsandsteepeningthenegativeslopeoftheBAR–BaARregressionline.TheseresultssuggestthattheDymondetal.(1992)equationmayovercorrectforlowsedimentationrates.BothBaARandproductiv-ityestimatesvaryovermorethantwoordersofmagnitude,sothelackofanyrelationshipappearstobringintoquestiontheutilityofbiogenicBaasageneralproductivityproxy.Eagleetal.(2003)foundregionaldifferencesintherelationshipbetweenBaARandprimaryproductivity,althoughtheyreportedastrongglobalcorrelationwithexportproduc-tivity.UnderstandingthereasonforthedifferentrelationshipsofBaARtoexportproductivitywillrequireadditionalinvestigation.

5.Discussion

5.1.RelationshipofproductivityproxyMARtoBAR

Instudiesinwhichsedimentﬂuxeshavebeencalculated,productiv-ityestimatesgenerallytrackBARclosely(e.g.,Sternbergetal.,2007;Murrayetal.,2012).Thismaybearealcorrelationinenvironmentsin

Fig.10.Bulkaccumulationrate(BAR)versusorganiccarbonaccumulationrate(OCAR)byredoxfacies,shownonalog–logscale.Thesuboxic/anoxicfaciesexhibitsalowerregressionslopem(1.11)thantheoxicfacies(1.76);mforthecombineddatasetis1.72(r2=0.86;n=4226).Fig.11.Bulkaccumulationrate(BAR)versusorganiccarbonpreservationfactor(PF)for(A)primaryproductionand(B)exportproduction,fortheL95andOSdatasets.L95:n=4140forprimaryproduction,n=2805forexportproduction.OS:n=3144forprimaryproduction,n=2406forexportproduction.

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Table4

Equationsandcorrelationcoefﬁcientsforlinearregressionequationsinlog–logspacere-latingmassﬂuxesoforganiccarbon,phosphorus,andexcessbariumtoprimaryandex-portproductionasestimatedintheL95andOSdatasets.

Regressionequation

r2OCAR

PrimaryproductionL95log10(OCAR)=2.53×log10(Prod.)−9.920.33OS

log10(OCAR)=2.07×log10(Prod.)−7.190.26ExportproductionL95log10(OCAR)=1.07×log10(Exp.)−3.420.20OSlog10(OCAR)=1.73×log10(Exp.)−5.70

0.30

PAR

PrimaryproductionL95log10(PAR)=1.39×log10(Prod.)−5.590.30OS

log10(PAR)=0.36×log10(Prod.)−1.330.03ExportproductionL95log10(PAR)=0.69×log10(Exp.)−2.250.19OSlog10(PAR)=0.32×log10(Exp.)−1.00

0.05

BaAR

PrimaryproductionL95log10(BaAR)=−0.25×log10(Prod.)+1.190.03OS

log10(BaAR)=−0.23×log10(Prod.)+1.140.04ExportproductionL95log10(BaAR)=−0.09×log10(Exp.)+0.440.02OS

log10(BaAR)=−0.09×log10(Exp.)+0.42

0.01

whichsedimentsaremostlybiogenicandsedimentationrateisafunc-tionofbiologicalproductivity,asintheequatorialPaciﬁc(Murrayetal.,2000,2012).Insuchcases,BARitself,aswellastheaccumulationrateofeachofthebiogeniccomponents,islikelytobeavalidproxyformarine

Fig.12.Primaryproduction(A)andexportproduction(B)versusorganiccarbonaccumu-lationrate(OCAR)fortheL95andOSdatasets.n=4226.

Fig.13.Phosphorusconcentrationsbyredoxfacies.Boxesrepresentmeansandstandarddeviationranges,whereasthewhiskersshowthefullrangeofvaluesforeachredoxcategory.

productivity.Suchconditionsarecommonintheopenpelagicocean,wherethebiogenicsedimentfractionisderivedmainlyfromprimaryproducers(e.g.,diatomsandcoccolithophores,Murrayetal.,2012)orlow-levelplanktonicconsumers(e.g.,radiolarians,Hori,1992;Algeo

Fig.14.(A)Bulkaccumulationrate(BAR)versusorganicphosphorusaccumulationrate(PAR),shownonalog–logscale.Slopesareapproximately1.0forbothoxic(m=1.04)andsuboxic/anoxic(m=0.93)units.DatafromMillsetal.(2010)appeartobeoutliersandarethereforeshownseparately.mforthecombineddataset,notincludingMillsetal.(2010),is1.07(r2=0.92;n=1935).(B)Bulkaccumulationrate(BAR)versusorganicphosphorusconcentration,shownonalog–logscale.Slopeisessentiallyﬂat(m=0.07forthefulldataset,r2=0.05).

42S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Fig.15.(A)Primaryproductionand(B)exportproductionversusphosphorusaccumula-tionrate(PAR)fortheL95andOSdatasets.n=1935.

etal.,2010).However,theyarelesscommoninshelforplatformenvi-ronments,wherethebiogenicsedimentfractioncontainsalargepor-tionofmaterialderivedfrombenthicmacrofauna,andwheresedimentscontainalargelithogeniccomponent,thereisnoapriori

Fig.16.Bulkaccumulationrate(BAR)versusorganiccarbonaccumulationrate(BaAR)byredoxfacies,shownonalog–logscale.Thesuboxic/anoxicfaciesexhibitsasteeperregres-sionslopem(0.57)thantheoxicfacies(0.35),whilethedownwardshiftininterceptsuggestsaredoxeffectonpreservation.Theslopemforthecombineddatasetis0.16(r2=0.05;n=2299).

Fig.17.Primaryproduction(A)andexportproduction(B)versusbiogenicbariumaccu-mulationrate(BaAR)fortheL95andOSdatasets.n=2299forallpanels.

reasonwhyproductivityshouldbepositivelycorrelatedwithBAR.OnecaveatisthatstudiesmakinguseofproductivityestimatesbasedonAl-orTi-normalizedproxyconcentrationsratherthanproxyMARsmaygenerateaspuriousnegativecorrelationwithBAR(e.g.,ShimmieldandMowbray,1991).

Theinﬂuenceofvariousfactorsonproductivity-proxyﬂuxes,in-cludingenhancedorganicmatterpreservation,siliciclasticdilution,andvariablebiogenicorganic:mineralratios,canbeevaluatedfromMAR/BARslopes(m).ThesepatternsareillustratedusingOCARasanexample(Fig.18),butsimilarrelationshipscouldbeinferredforPARand,possibly,BaAR(althoughthelatterproxyissubjecttoadditionalinﬂuencesasaresultofitsauthigenicorigin;seebelow).Inahypothet-icaldepositionalsystemaccumulatingonlybiogenicsedimentwithaﬁxedorganic:mineralratio(i.e.,aconstantweightpercentofTOC)andinwhichthereisnoeffectofBARonorganiccarbonpreservation,OCARincreasesindirectproportiontoBARwithmequalto1.0(or1:1,representingslopesinlog–logspace;Fig.18A).Thisisapureex-pressionoftheautocorrelationeffect,inwhichOCARexactlytracksBARasaresultofthelatterbeingafactorinthecalculationoftheformer(seeEq.(3)).Inthishypotheticalsystem,inputsofdetritalsiliciclasticmaterialdilutethebiogeniccomponentofthesediment,loweringmtob1.0(Fig.18A).Lowervaluesofmdevelopbecausetheadditionofnon-biogenicdiluentscausesBARtoincreasemorerapidlythanOCAR,whichincreasesindirectproportiontothebiogenicﬂux.Suchadiluenteffectshouldbemostevidentincontinentalshelfandepicratonicma-rinesettings,wheredetritalsiliciclasticﬂuxesarecomparativelylarge.

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5243

Fig.18.ConceptualmodelsofvariationinOCAR(A–C)andBaAR(D)asafunctionofbulkaccumulationrate(BAR)andvariouseffects.(A)‘Detritaldilutioneffect’:increasingdilutionlowerstheslopeoftheregressionline(m)from1:1towardzero.(B)‘Organic:mineralratioeffect’:increasingtheorganic:mineralratioofbiogenicmaterialshiftstheregressionlineup-wardwithoutchangingm.(C)‘Enhancedpreservationeffect’:mbecomesN1:1whenhigherBARscauseenhancedpreservationoforganiccarboninpurelybiogenicsediments.Thiseffectcanpotentiallybepartiallyoffsetornegatedbythedetritaldilutioneffect,yieldingmb1:1inmixedbiogenic–detritalsediments.oforganicmatteryieldsmN1:1.(D)BaAR,theaccumu-lationrateofauthigenicBa,ishypothesizedtodependonmultiplefactors.Thedashed1:1linerepresentsBauptakeindirectproportiontoOCARina100%biogenicsediment.Observedmvaluesaresigniﬁcantlylower(0.35foroxicfacies,and0.57forsuboxic/anoxicfacies;Fig.16)duetotwocontrols:(1)reduceduptakeofauthigenicBaatthesediment–waterinterfaceinhigh-BARsystems,and(2)reduceduptakeofBaperunitorganiccarbonforrefractoryrelativetolabileorganicmatter.Finally,Baissubjecttoaredoxpreservationeffect,characterizedbyasmallreductioninBaARforsuboxic/anoxicfaciesrelativetooxicfaciesasaresultofreductivedissolutionofbariteintheformer.

Inthisscenario,varyingtheorganic:mineralratioofthebiogenicﬂux(i.e.,theweightpercentofTOCinthesediment)wouldraiseorlowertheregressionline,changingthey-interceptwithoutchangingm(Fig.18B).TheforegoingscenariosassumenoeffectofBARonthepres-ervationoforganiccarbon,eventhoughitisgenerallyacceptedthathigherBARsenhanceorganicmatterpreservation.Indepositionalsys-temsdominatedbybiogenicinputs,theenhanced-preservationeffectwillcausemtoincreasetoN1.0(Fig.18C).SubtractingtheOCAR/BARautocorrelationslopeof1.0fromtheobservedmthusyieldsthemagni-tudeoftheenhanced-preservationeffect.However,thepositiveeffectofenhancedorganicpreservationonmmaybeoffsetbythenegativeef-fectofdetritaldilutioninsedimentarysystemswithlargenon-biogenicinputs,thusloweringmrelativetoitsvalueintheabsenceofdetritaldi-lution(Fig.18C).ObservedvaluesofmforOCAR/BARregressionsexhibitastrongredoxdependence.Tyson(2005)determinedmof0.84forsuboxic/an-oxicfaciesand1.38foroxicfaciesofHoloceneage.Ourdatasetshowsthesamepatternoflowermforsuboxic/anoxicfacies(1.11)relativetooxicfacies(1.76;Fig.10),butwithhigherabsolutevaluesofmthanthosereportedbyTyson(2005).Lowervaluesofminsuboxic/anoxicfa-ciesareduetothereducedimportanceofrapidburialfororganiccarbonpreservationwhentheoverlyingwatercolumnisdepletedofoxygen(Canﬁeld,1994).Asdiscussedabove,thedifferenceinmbetweenanob-servedsampleregressionandanautocorrelationline(m=1.0)isanin-dicationofthestrengthoftheenhanced-preservationeffect,andthedifferenceinmbetweenoxicandsuboxic/anoxicsamplesisanindica-tionoftherelationshipbetweenBARandtheenhanced-preservationeffect.Subtractingtheautocorrelationmof1.0yieldsanenhanced-

44S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

preservationeffectof0.76foroxicfaciesand0.11forsuboxic/anoxicfa-ciesinourdataset,and0.38foroxicfaciesand−0.16forsuboxic/anoxicfaciesintheTyson(2005)dataset.Ourdatasetthusshowstheenhanced-preservationeffectforoxicsedimentstobetwiceaslargeasfortheTyson(2005)dataset,implyingamuchstrongerinﬂuenceofBARonthepreservationofsedimentaryorganicmatterthanrecognizedheretofore.Further,ourdatasetdocumentsasmallpositiveeffectofBARonorganicmatterpreservationinsuboxic/anoxicsediments,versusanegativeeffectfortheTyson(2005)dataset.Evenunderanoxiccondi-tions,increasesinBARarelikelytoenhancethepreservationoforganicmattertoasmalldegree,whichisconsistentwiththeresultsofouranalysis.Thedifferenceinmbetweenoxicandsuboxic/anoxicfaciesreﬂectsthestrengthoftheinﬂuenceofredoxconditionsonorganicmatterpreservation.Thisdifferenceis0.65(i.e.,1.76–1.11)forourdatasetand0.54(i.e.,1.38–0.84)fortheTyson(2005)dataset.Theseslopesarequitesimilar,bothindicatingthatvariationinBARhasamuchlargerinﬂuenceonorganiccarbonpreservationinoxicfaciesthaninsuboxic/anoxicfacies.

Thepointofconvergenceoftheoxicandsuboxic/anoxictrendsonaBAROCARcrossplothasbeeninterpretedasrepresentingtheaccumula-tionrateabovewhichaerobicoxidationoforganicmatterinthesedimentbecomesinsigniﬁcant(Algeoetal.,2013).Inourdataset,thisconvergenceoccursataBARof~30(=101.5)gcm−2kyr−1(Fig.10),whichissubstantiallylowerthantheconvergencepointof~125(=102.1)gcm−2kyr−1showninFigure5ofTyson(2005,n.b.,thex-axisscaleinthatﬁgureshouldbegm−2yr−1,notgm−2kyr−1asshown).Inpart,thedifferenceinconvergencepointsisduetothelargerdatasetusedinthepresentstudy,whichmaybebetterabletodistinguishtheeffectsofBARonorganiccarbonaccumulation.Howev-er,itisalsoareﬂectionofthelargerm(1.11)forsuboxic/anoxicsitesinourdataset.Inourdataset,therelationshipsbetweenBARandOCARforoxicversussuboxic/anoxicfacies(Fig.10)appearlessastwoconvergingtrendsthanasacontinuum,wheretheregressionslopesﬂattenslightlyathigherBARsbutremainsabove1.0,indicatingthatrapidburialand,hence,post-depositionaldecompositionandlossofcar-bon,continuetoexertaninﬂuenceoncarbonpreservation.Sincesuboxicandanoxicenvironmentsaregenerallylocatedinrestrictedormarginal-marinesettings,theytendtohavehighBARs(Fig.10).Thus,thisﬂatteningoftheBAR–OCARrelationshipmayindicatethatrapidburialexercisesless,butstillsome,inﬂuenceonpreserva-tioninlow-oxygensettings.Itisalsopossiblethattheﬂatterregres-sionslopeathighBARsmayreﬂectincreasingsiliciclasticdilution(Fig.18A),assubstantialdetritalsedimentinputisnecessarytoachievethehighestBARs.

WhiletheOCAR–BARregressionslopesforourbulkdatasetaswellasitsoxicandsuboxic/anoxicfaciesareN1.0,individualstudyunitsgener-allyyieldmcloseto1.0,withameanof1.07andamedianof0.99(Fig.19;n=34).ThispatternimpliesthedominantinﬂuenceofautocorrelationbetweenOCARandBARforindividualstudyunits(cf.Fig.18A),whichisacharacteristicfeatureofsedimentaryunitsinwhichBARismorevariablethanTOC.Anadditionalfactorfavoringregressionslopesof~1.0maybethecompensatoryeffectsofBAR-enhancedpreservation(increasingm)andsiliciclasticdilution(decreas-ingm;Fig.18C).Theprobableinﬂuenceofsiliciclasticdilutiononm(Fig.18A)isparticularlyevidentforindividualstudyunitswithhighBARs(i.e.,N100gcm−2kyr−1),whichgenerallyconsistpredominantlyoflithogenicsedimentsandwhichyieldmof0.5to0.8(Fig.19).

PARexhibitsasimplerelationshiptoBARwithwell-deﬁnedmof1.04foroxicfaciesand0.93forsuboxic/anoxicfacies(Fig.14A).Theseslopesarebothclosetothevalueof1.0characteristicofautocorrelations(seeabove),indicatingthatBARisthedominantcontrolonPAR.ThestronginﬂuenceofBARonPARisexpectedgiventhattheformerisafactorinthecalculationofthelatter(Eq.(3)),andthatBARvariesoversixordersofmagnitudeversustwoforPorgconcentrations(Fig.14B).TherelativeinvarianceofPorgconcentrationsoverawiderangeofsedimentaccumulationratessuggeststheoperationofa

Fig.19.Bulkaccumulationrate(BAR)versustheorganiccarbonaccumulationrate(OCAR).Eachlineshownisthebestﬁtregressionlinesforanindividualsiteinthedataset.Notethat,althoughtheregressionslopes(m)ofindividualunitsare~1.0,mforthebulkdatasetis1.72.

negativefeedbackmechanismaffectingorganicPretentioninsedi-ments(Ingalletal.,1993;Murphyetal.,2000;seeSection5.4).PAR/BARslopesof~1.0alsoimplythatphosphorusaccumulationisnoten-hancedathighersedimentationrates,whichisunlikethestronglyen-hancedpreservationoforganiccarbon(seeabove).Theslightlylowermofsuboxic/anoxicfacies(0.93)relativetooxicfacies(1.04)isconsis-tentwiththeeffectsofclasticdilutionathigherBARs(Fig.18A).

BaAR,theaccumulationrateofauthigenicBa,exhibitsamorecomplexsetofcontrols.Threeobservationsmustbeaccountedfor:(1)BaAR/BARregressionsexhibitunusuallylowm(bb0.1),(2)aredoxeffectisevident,withoxicfacieshavingalowerm(0.35)thansuboxic/anoxicfacies(0.57),aswellasahigherBaARforagivenBARand(3)aBAReffectisevident,withbothredoxcategoriesshowingsep-aratetrendsofincreasingBaARwithincreasingBAR(Fig.16).ControlsonthemofBaAR/BARregressionscanbeinferredwithintheframeworkpreviouslydevelopedforOCAR/BAR(Fig.18A–C)butmodiﬁedtoallowfortheauthigenic(ratherthanbiogenic)originofBa.

IfBaaccumulatesindirectproportiontoOCAR(productivitycontrol),thenitshouldexhibitanmof1:1ina100%biogenicsedimentlackingpreservationeffects(Fig.18D).Thelowm(≪1.0)exhibitedbyBaAR/BARcannotbeduetodetritaldilution(Fig.18A;seeSection5.1),becausesuchaninﬂuencewouldhaveoperatedequallyontheOCAR/BARandPAR/BARrelationships,whichislargelynotob-served(Figs.10,14;notethataverysmallclasticdilutioneffectwasin-ferredaboveforPAR/BAR).WhileDymondetal.(1992)demonstratedthatoverallauthigenicbaritepreservationisenhancedathigherBARs,itispossiblethatnegativepreservationaleffectscontributetoanoverallBAR/BaARrelationshipwithasmallerpreservationaleffect(m≪0.1)thanthatseenfororganicmatter.

Thelabilityoftheorganicmatterand,hence,itsrateofdecaymayaf-fectratesofsulfatereduction,whichmayimpactbaritepreservationrate(Fig.18D).Asedimentcontainingmorerefractoryorganicmatterwillhavealowerreductantdemandthanasedimentcontainingmorelabileorganicmatter(WestrichandBerner,1984;Hultheetal.,1998).Rapidburial,whichminimizestheexposuretimeoforganicmaterialtooxygen,couldensurethatalargerpooloflabileorganicmaterialsur-vivestobeburiedinthezoneofsulfatereduction,whereitmayfacili-tatereductivebaritedissolution.Thismechanismmayoperateduringearlydiagenesisevenwheresedimentsunderlieanoxicwatercolumn,asanaerobicrespirationisresponsibleforthemajorityofcarbonoxida-tioneveninoxiccontinentalshelfsediments(Canﬁeld,1994).

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5245

Reductivedissolution,andlesscompleteorganicmatterdegradation,inthewatercolumnarepotentiallyresponsibleforgenerallylowerpreser-vationinsuboxic/anoxicenvironments(Fig.18D;seeSection4.4).5.2.Estimationofpaleoproductivityandestimateerrors

Paleoproductivityestimatesforanyancientmarinesedimentaryunitpotentiallycanbederivedfromproductivity-proxyﬂuxdata.Tolimittheinﬂuenceofautocorrelations(e.g.,betweenproxyMARsandBAR;seeSection5.1),wedevelopedpaleoproductivityalgorithmsbasedonproxyPFsandBAR.Althoughtheselattervariablesarenotfullyindepen-dent,thepotentialforautocorrelationsisreducedbecausePFsdependonlyindirectlyonBAR(seeEqs.(3)and(6)–(7)).Best-ﬁtlinearregres-sionsfororganiccarbonPFasafunctionofprimaryproduction(prim)versusBARwerenearlyidenticalfortheL95andOSproductivitydatasets(Fig.11).Agenericformofthisrelationship,withcoefﬁcientsintermedi-atebetweenthosecalculatedfortheL95andOSdatasets,is:log󰀂󰀃

10PFprim¼1:54Âlog10ðBARÞ–4:10ð13Þ

whichisequivalenttotheexponentialequation:PF10

prim¼10

−4:ÂBAR

1:54

ð14Þ

whereBARisinunitsofgcm−2kyr−1andPFisadimensionlessvariablebetween0and1.Asimilargenericequationcanbederivedfortheorgan-iccarbonPFasafunctionofexportproduction(exp),althoughthepoorinitialcorrelationofthesevariablesandtheweakeragreementbetweentheL95andOSproductivitydatasetsmeansthatitshouldbeusedwithcaution:PF:37

exp¼10

−3ÂBAR

1:47

:ð15Þ

CombiningEqs.(3)and(6)yieldsthefollowingexpressionsforPFprimandPFexp:

PFprim¼ðTOCÂBARÞ=PRODprim

ð16Þ

PFexp¼ðTOCÂBARÞ=PRODexp:ð17Þ

InsertingtheexponentsfromEqs.(14)and(15)andrearrangingtosolveforproductivityyieldsthefollowingequations:PROD󰀂104:10ÂTOC󰀃

=BAR

0:54

prim¼1000Âð18Þ

PRODexp¼1000Â10

3:37

ÂTOCÞ=BAR

0:47

ð19Þ

whereBARisinunitsofgcm−2kyr−1,TOCisadimensionlessweightratiobetween0and1,andthefactorof1000servestoexpressPRODZinourstandardunitsofmgcm−2kyr−1.ThesigniﬁcanceofEqs.(18)and(19)isthattheyallowestimationofprimaryandexportproductiv-ityasafunctionoftwocompletelyindependentvariables(TOCandBAR)thatarereadilydeterminableinmanypaleomarinesystems.

ThepresenceofBARinthedenominatoroftheseequationsmaybecounter-intuitive,ashighBARwaspreviouslyshowntocorrelatewithhighOCAR(Fig.10).However,therelationshipofBARtoTOCandpro-ductivitycanbeunderstoodinthecontextofitsrelationshiptoPFinEqs.(13)and(14).Speciﬁcally,they-interceptinEq.(13)(10−4.10)rep-resentstheaverageproportionofprimaryproductivitythatwillbepre-servedwhenBARiszero(i.e.,lessthanonepartintwelvethousand).Theinverseofthisvalue,seenasthecoefﬁcientinthenumeratorofEq.(18),representsthemaximumproductivityratethatcanbeinferredbasedonagivenTOCconcentrationbeforeaccountingforpreservationaleffects.

TheBARterminthedenominatorrepresentsthesepreservationaleffects.AsBARincreases,theorganiccarbonPFincreasesevenmorerap-idly(atalog–lograteof1.54;Eq.(14))sothatcalculatedprimarypro-ductivityratesforagivenTOCconcentrationmustdecline(Eq.(18)).Thesameconsiderationsapplytoexportproductivity,asgiveninEqs.(15)and(19).

ToassesstheutilityofEqs.(18)and(19)inreconstructingpaleoproductivity,weestimatedprimaryandexportproductivityratesforallsamplesinourdatabaseforwhichTOCdatawereavailableandthencomparedthemwithactualproductivityratemeasurementsfromtheL95andOSdatasets.Correlationcoefﬁcientsbetweentheesti-matedandmeasuredproductivityratesaregenerallylowalthoughsomewhatbetterfortheOSthanfortheL95productivitydataset:r2=0.24(OS)and0.06(L95)forprimaryproductivity,andr2=0.27(OS)and0.01(L95)forexportproductivity(Fig.20).Differencesines-timated(est)andmeasured(meas)productivityrateswerequantiﬁedasarelativeerror:

Error¼ðPRODZ‐est–PRODZ‐measÞ=PRODZ‐measÂ100%

ð20Þ

whereZrepresentseitherprimary(prim)orexport(exp)productivity,and‘measured’referstotheL95andOSproductivityestimates.Relativeerrorswerecalculatedasabsolutevaluesinordertoindicatethedevia-tionoftheproductivityestimatefromtheexpectedvalueineitherdirection.TherelativeerrorsthuscalculatedarecommonlylargealthoughrelativelysimilarfortheOSandL95productivitydatasets.Ab-solutemeanerrorvaluesforprimaryproductivityestimateswere152%(OS)and158%(L95),andthoseforexportproductivityestimateswere143%(OS)and178%(L95).Medianerrorswereconsiderablylower:58%(OS)and68%(L95)forprimaryproductivityestimates,and53%(OS)and76%(L95)forexportproductivityestimates.Thesmallermedian

Fig.20.(A)PrimaryproductionaspredictedfromTOCandBARbyEq.(18),versusprimaryproductionfromtheL95andOSdatasets.(B)ExportproductionaspredictedfromTOCandBARbyEq.(19),versusexportproductionfromtheL95andOSdatasets.n=4226foreachdataseries.

46S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

valuesindicatethaterrordistributionsarehighlyskewed,i.e.,withmanysmallerrorsandfewerlargeerrors(Fig.21).Thispatternisattrib-utableinparttotheupperlimitof100%oncalculatederrorswhenPRODZ-estbPRODZ-meas.

Overall,N70%ofproductivityestimatesforindividualsampleshadanabsoluterelativeerrorofb100%(Fig.21).Giventhelargenumberoffac-torsthatcancauseproductivityestimatesbasedonsedimentaryproxiestodeviatefrommoderninsituproductivityrates(seeSection3.4),thisisafairlyrobustresult.Becausetheproductivityestimatesinourdatasetrangeovermorethantwoordersofmagnitude,evena100%errorisanacceptablelevelofuncertaintyforproductivityestimatesinpaleomarinesystems.Thevalueofourapproachtopaleoproductivityestimationisen-hancedbythefactthatitisbasedontwoindependentandeasilydeter-minedsedimentologicalparameters(i.e.,TOCandBAR).ExamplesoftheapplicationofthismethodofpaleoproductivityanalysiswillbegivenincompanionpapersinthepresentvolumebyShenetal.(inreview)andWeietal.(2014).

5.3.PaleoproductivityestimatesbasedonOCAR

Amongtheelementalproxymassﬂuxesconsideredinthisstudy,the

strongestpositiverelationshipwithbothprimaryandexportproduc-tionisshownbyOCAR(Fig.12).ItissurprisingthatOCARshouldshowastrongerrelationshipwithprimarythanwithexportproduction,sinceitiscarbonexportedfromthesurfaceoceanthatisultimatelypre-servedinthesediment.Themarginallystrongerrelationshipwith

Fig.21.DistributionoferrorvaluescalculatedwithEq.(20)for(A)primaryproductiones-timatesfromEq.(18)versusactualprimaryproductionfromtheL95andOSdatasets,and(B)exportproductionestimatesfromEq.(20)versusactualexportproductionfromtheL95andOSdatasets.ErrorsN500%arenotshown,butbothpanelsshowN90%ofallerrorvalues.Errorvaluesbinnedinincrementsof5%.

primaryproductionmayreﬂecttheeffectsofhighproductivityonredoxconditionsortheassociationofhighproductivitywithelevatedsedimentationrates.TheregressionsrelatingOCARtoprimaryandex-portproduction(Table4)canberearrangedtopredictedprimary(prim)andexport(exp)productionasafunctionofOCAR,usingcoefﬁ-cientsintermediatebetweenthosedeterminedfortheL95andOSdatasets:

PROD󰀂8:55

󰀃0:43

prim¼10

ÂOCARð21Þ

PROD󰀂󰀃exp¼104:56

ÂOCAR

0:71ð22Þ

wherePRODZandOCARareinunitsofmgcm−2kyr−1.IfOCARisre-placedbyTOCandBAR(perEq.(3)),Eqs.(21)and(22)canberecasttoestimatepaleoproductivitybasedonthesameparametersasusedinEqs.(18)and(19).However,thisreformulationyieldsapositiverela-tionshipbetweenPRODZandBARthatisquitedifferentfromtheinverserelationshipsseeninEqs.(18)and(19).Unlikethoseequations,Eqs.(21)and(22)areempiricalderivationsofpaleoproductivityratesthatdonotaccountexplicitlyforthepreservationaleffectsofBAR.

PaleoproductivityestimatesbasedonEqs.(21)and(22)exhibitweaktomodestcorrelationstomodernproductivityratesandaresubjecttopotentiallylargeerrors.Estimatesofprimaryproductivity(Eq.(21))yieldanr2of0.44withL95productivityandanr2of0.06withOSproduc-tivity.Averageandmedianabsoluteerrors,ascalculatedperEq.(20),were64%and59%,respectively,whencalculatedagainstL95productiv-ity,and119%and68%,respectively,whencalculatedagainstOSproduc-tivity.Estimatesofexportproductivity(Eq.(22))yieldanr2of0.05withL95productivityandanr2of0.32withOSproductivity.Averageandme-dianabsoluteerrors,ascalculatedperEq.(20),were245%and83%,re-spectively,whencalculatedagainstL95productivity,and371%and83%,respectively,whencalculatedagainstOSproductivity.

Therelativeimportanceofproductivityratesandredoxconditionsonproductivity-proxyﬂuxeswasconsideredconceptuallyinFig.6,andtheproxyﬂuxdatasetgeneratedinthisstudyprovidesanopportu-nitytoassesstheserelationshipsquantitatively.AlthoughredoxfaciesisclearlyanimportantcontrolonOCAR(Fig.9A),itisworthnotingthatBARandredoxconditionsarenotfullyindependentvariablesinourdataset,anditisuncertaintowhatextentthisisreﬂectedinenhancedpreservationathighBARs.Infact,OCARappearstoshowtheinﬂuenceofbothredoxandproductivitycontrols,withthehighestorganiccarbonﬂuxesassociatedwithhigh-productivityanoxicsettings(i.e.,lowerrightofFig.22A).Becauseelevatedsurface-waterproductivitycantrig-gerbenthicanoxiathroughthebiologicaloxygendemandimposedbysinkingorganicmatter,theserelationshipsmightbeinterpretedasevi-denceofthegenerallydominantinﬂuenceofproductivityonorganicmatteraccumulationinmarinesediments.However,afeedbackinvolv-ingtherecyclingoforganicPinmarinesedimentsbackintothewatercolumnhelpstosustainhighsurface-waterproductivityinanoxicma-rinesystems(Ingalletal.,1993;Murphyetal.,2000).Thus,therelation-shipsdocumentedhere(Fig.22A)suggestthatelevatedproductivityratesarecommonlyimportantinestablishingmarineanoxiaandhighOCAR,butthattheinterplayofproductivityandredoxconditionsisin-tegraltosustainingsuchconditions(cf.PedersenandCalvert,1990;Tyson,2005).

5.4.PaleoproductivityestimatesbasedonPAR

ThemostnotablefeatureofthephosphorusdatasetistherelativeconsistencyinPconcentrationsoverawiderangeofredoxconditions(Fig.13)andBARs(Fig.14).AlthoughPARshowsonlyaweakrelation-shiptoprimaryproductivity,particularlyatlowerproductivityrates(Fig.15A),itbecomesnearlyinvariantatprimaryproductivityratesN20,000mgcm−2kyr−1,apointthatcorrespondstothetransition

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–5247

Fig.22.Massﬂuxesof(A)organiccarbon,(B)organicphosphorus,and(C)biogenicbar-iumforeachLonghurstetal.(1995)provinceconsideredinthisstudyasafunctionof

redoxsettingandprimaryproductivity.PositionofcirclesonhorizontalaxiscorrespondstoestimatesofprimaryproductionfromtheL95(blue)orOS(red)datasets,shownonalogarithmicscale.Positionofcirclesonverticalaxiscorrespondstoitscategorizationinonethethreediscreteredoxcategoriesusedinthisstudy.Diameterofcircleisproportion-altotheprovinceaveragemassﬂuxoftheproductivityproxyinquestion(inunitsofmgcm−2kyr−1).Wheretworedoxcategorieswererepresentedinthesameprovince,separateaverageswerecalculatedforeachredoxcategoryandseparatecircleswereusedtorepresenttheaverages.

betweenprimarilyoxicandprimarilysuboxic/anoxicfacies.Itsrelation-shipwithexportproductionismoremonotonicbutweaker.Itispossi-bletoestimateprimaryandexportproductionbasedonPARusingtheequationsinTable4(withcoefﬁcientsintermediatebetweenthosecal-culatedfortheL95andOSdatasets):PROD¼󰀂103:46

ÂPAR

󰀃1:14

prim

ð23Þ

PROD󰀂1:63

󰀃1:98

exp¼10ÂPAR

ð24Þ

whereproductivityandPARareinunitsofmgcm−2kyr−1.Correlationswereweakeranderrorshigherfortheseequationsthanforthosede-rivedfromOCAR,buttheywereofthesameorderofmagnitude.AswithOCAR,correlationswiththeL95productivityestimates(r2=

0.27forprimaryproduction,0.08forexportproduction)werestrongerthanthosewiththeOSestimates(r2=0.00forprimaryproduction,0.05forexportproduction).FortheL95dataset,meanandmedianabso-luteerrorsofprimaryproductionestimatesare229%and81%,respec-tively,andtheerrorsofexportproductionestimatesare763%and88%,respectively.FortheOSdataset,meanandmedianabsoluteerrorsforprimaryproductionestimatesare2996%and85%,respectively,andtheerrorsforexportproductionestimatesare1860%and93%,respectively.Forthelatterdataset,thelargedifferencesbetweenthemeanandmedi-anerrorsreﬂectinclusionofanomalouslyhighPARvaluesforSaanichInlet;whenthesedataareexcluded,themeanandmedianabsoluteerrorsforprimaryproductivityfallto314%and84%,respectively.

Porgappearstoaccumulatemostrapidlyinhighlyproductive,suboxic-to-anoxicsettings(Fig.22B),althoughthelackofanoxicsiteshavinglowproductivityratesmayinﬂuencethisinterpretation.Be-causeorganiccarbonandphosphorushavedifferentmodesofpreserva-tionasafunctionofredoxvariation(seeSection2.6),thesimilarityofthepatternsforOCARandPAR(Fig.22A,B)issigniﬁcant,suggestingthepotentiallydominantinﬂuenceofproductivityontheaccumulationoforganicmatterinmarinesediments(cf.Section5.2).ThenarrowingrangeofPorgconcentrationsobservedinincreasinglyanoxicenviron-ments,aneffectthatisseenmoreclearlyinmaximumthaninmeanvalues(Fig.14),suggeststhatanoxicconditionsfundamentallylimitPorgretentioninsediments.ThislimitationoperatesthroughanefﬁcientnegativefeedbackinvolvingthediffusionofremineralizedorganicPbackintothewatercolumnunderincreasinglymorereducingcondi-tions(Ingalletal.,1993;Murphyetal.,2000;seeSection4.3).5.5.PaleoproductivityestimatesbasedonBaAR

BaARdoesnotseemtoshowanyclearrelationshipwithproductivityonaglobalscale.Theweak(andnegative)correlationsbetweenesti-matesofproductivityandBaAR(Fig.17)donotinspireconﬁdenceintheuseofBabioasawidelyapplicableproductivityproxy.AlthoughEagleetal.(2003)foundthatbiogenicbariumvariednearlylinearlywithexportproductionafteruseofanappropriatelocalpe-ratio,thisanalysiswaslimitedtomarinesystemswithanestimatedexportproductionofb104mgCcm−2kyr−1.ApositiverelationshipcanbediscernedbetweenexportproductionandBaARintheOSdatasetoverthissameproductivityrange(Fig.17B),butitbreaksdownathigherlevelsofproductivity.Weincludethefollowingequationsforestimat-ingpaleoproductivityfromBaARforthesakeofcompleteness(i.e.,asanalogstoEqs.(21)–(22)forOCARandEqs.(23)–(24)forPAR),butweemphasizethattheyhavequestionablevalueforestimatingproduc-tivityinpaleomarinedepositionalsystems:PROD󰀂−1:17

󰀃−4:17

prim¼10

ÂBaARð25Þ

PROD¼󰀂10−0:43

󰀃−11:11

expÂBaAR

ð26Þ

whereBaARandproductionareinunitsofmgcm−2kyr−1.Thecorre-lationcoefﬁcientsareb0.01forallestimatesofprimaryandexportpro-ductivityversusmeasuredproductivityvalues,andaverageerrorsareordersofmagnitudelargerthanforproductivityestimatesbasedonOCARorPAR.

Thebariteproductivityproxywasdevelopedlargelyintheopenocean(Dymondetal.,1992;Françoisetal.,1995;PaytanandKastner,1996;Paytanetal.,1996;Eagleetal.,2003;Paytanetal.,2007),inenvi-ronmentsdominatedbypelagicbiogenicsediments(Murrayetal.,2000;PrakashBabuetal.,2002),anditsapplicabilitymaybelimitedtosuchsystems.Furtherworkwillbeneededtodetermineifbiogenicbariumhasvalueforestimatingpaleoproductivityinequivalentancientopen-oceanfacies,suchasradiolarites(e.g.,Algeoetal.,2010,2011).

48S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

6.Conclusions

Ouranalysisofthreewidelyusedelementalproductivityproxies(TOC,Porg,andBabio)providesinsightsregardingcontrolsontheiraccu-mulation,theirrobustnessaspaleoproductivityproxies,andtherangeofdepositionalenvironmentsinwhichtheymayusefullybeapplied.OrganiccarbonaccumulationratesweredeterminedtohaveastrongrelationshiptoBAR,withalargeslopem(1.72)indicatingstronglyen-hancedpreservationoforganiccarbonathighersedimentaccumulationrates.Organiccarbonpreservationfactors(PF)exhibitalinearrelation-shipwithBAR,indicatingthattheeffectsofrapidsedimentaccumula-tiononpreservationcanbecorrectedfor,andthatpaleoproductivitycanbeestimatedfromTOCandBAR.Theresultingequationscanyieldorder-of-magnitudeestimatesofprimaryandexportproductioninpaleomarinesystems.PhosphorusaccumulationratesarestronglycorrelatedtoBARwithanmof~1.0,implyingahighdegreeofautocor-relationand,thus,controlofPaccumulationbyBAR.ThisisconsistentwiththeobservedlimitedvariationofsedimentaryPconcentrations,whichislikelyduetotheoperationofhomeostaticfeedbacksrelatedtoporewaterredoxconditions.Ataglobalscale,theproductivity-dominantmodelappearstoaccountbetterforobservedpatternsofor-ganiccarbonandphosphorusaccumulationratesthantheredox-dominantmodel.BiogenicbariumexhibitsaweakrelationshiptoBAR,probablybecauseofreduceduptakeofBaatthesediment–waterinter-facewithincreasingsedimentationrates.Biogenicbariumﬂuxesshownosystematicrelationshiptoproductivityinmodernmarinedeposi-tionalsystemsgenerally,althoughpreviousstudieshaveidentiﬁedpos-itivecovariationwithproductivityinspeciﬁcenvironments,suchastheequatorialPaciﬁc.Weconcludethatorganiccarbonandphosphorusﬂuxeshaveconsiderablepotentialaswidelyusefulpaleoproductivityproxies,butthattheapplicabilityofbiogenicbariumﬂuxesmaybelim-itedtospeciﬁcoceanicsettings.Acknowledgments

WethankPeterWardforstimulatingdiscussionsandprofessionalmentoring.ResearchbySDSissupportedbytheSedimentaryGeologyandPaleobiologyprogramoftheU.S.NationalScienceFoundationandUniversityofWashingtonDepartmentEarthandSpaceSciences.Re-searchbyTJAissupportedbytheSedimentaryGeologyandPaleobiolo-gyprogramoftheU.S.NationalScienceFoundation,theNASAExobiologyprogram,theStateKeyLaboratoryofGeologicalProcessesandMineralResourcesattheChinaUniversityofGeosciences-Wuhan(program:GPMR201301),andtheNaturalScienceFoundationofChina(NSF-C).

AppendixA.Supplementarydata

Supplementarydatatothisarticlecanbefoundonlineathttp://dx.doi.org/10.1016/j.earscirev.2014.08.017.References

Algeo,T.J.,Heckel,P.H.,2008.TheLatePennsylvanianmidcontinentseaofNorthAmerica:

areview.Palaeogeogr.Palaeoclimatol.Palaeoecol.268,205–221.

Algeo,T.J.,Ingall,E.,2007.SedimentaryCorg:Pratios,Paleoceanventilation,andPhanero-zoicatmosphericpO2.Palaeogeogr.Palaeoclimatol.Palaeoecol.256,130–155.

Algeo,T.J.,Hinnov,L.,Moser,J.,Maynard,J.B.,Elswick,E.,Kuwahara,K.,Sano,H.,2010.

ChangesinproductivityandredoxconditionsinthePanthalassicOceanduringthelatestPermian.Geology38,187–190.

Algeo,T.J.,Kuwahara,K.,Sano,H.,Bates,S.,Lyons,T.,Elswick,E.,Hinnov,L.,Ellwood,B.B.,

Moser,J.,Maynard,J.B.,2011.Spatialvariationinsedimentﬂuxes,redoxconditions,andproductivityinthePermian–TriassicPanthalassicOcean.Palaeogeogr.Palaeoclimatol.Palaeoecol.308,65–83.

Algeo,T.J.,Henderson,C.M.,Tong,J.N.,Feng,Q.L.,Yin,H.F.,Tyson,R.V.,2013.Planktonand

productivityduringthePermian–Triassicboundarycrisis:ananalysisoforganiccarbonﬂuxes.Glob.Planet.Chang.105,52–67.

Amo,M.,Minagawa,M.,2003.Sedimentaryrecordofmarineandterrigenousorganic

matterdeliverytotheShatskyRise,westernNorthPaciﬁc,overthelast130kyr.Org.Geochem.34(9),1299–1312.

Anderson,L.D.,Delaney,M.L.,2005.UseofmultiproxyrecordsontheAgulhasRidge,

SouthernOcean(OceanDrillingProjectLeg177,Site1090)toinvestigatesub-AntarctichydrographyfromtheOligocenetotheearlyMiocene.Paleoceanography20,PA3011(16pp.).

Antia,A.N.,Koeve,W.,Fischer,G.,Blanz,T.,Schulz-Bull,D.,Scholten,J.,Neuer,S.,Kremling,

K.,Kuss,J.,Peinert,R.,Hebbeln,D.,Bathmann,U.,Conte,M.,Fehner,U.,Zeitzschel,B.,2001.Basin-wideparticulatecarbonﬂuxintheAtlanticOcean:regionalexportpat-ternsandpotentialforatmosphericCO2sequestration.Glob.Biogeochem.Cycles15(4),845–862.

Averyt,K.B.,Paytan,A.,2004.Acomparisonofmultipleproxiesforexportproductionin

theequatorialPaciﬁc.Paleoceanography19,PA4003(14pp.).

Bąk,K.,2007.Organic-richandmanganesesedimentationduringtheCenomanian–

TuronianboundaryeventintheOuterCarpathianbasins;anewrecordfromtheSkoleNappe,Poland.Palaeogeogr.Palaeoclimatol.Palaeoecol.256,21–46.

BalakrishnanNair,T.M.,Ittekkot,V.,Shankar,R.,Guptha,M.V.S.,2005.Settlingbarium

ﬂuxesintheArabianSea:criticalevaluationofrelationshipwithexportproduction.Deep-SeaRes.II52,1930–1946.

Bastviken,D.,Persson,L.,Odham,G.,Tranvik,L.,2004.Degradationofdissolvedorganic

matterinoxicandanoxiclakewater.Limnol.Oceanogr.49,109–116.

Behrenfeld,M.J.,O'Malley,R.T.,Siegel,D.A.,McClain,C.R.,Sarmiento,J.L.,Feldman,G.C.,

Milligan,A.J.,Falkowski,P.G.,Leterlier,R.M.,Boss,E.S.,2006.Climate-driventrendsincontemporaryoceanproductivity.Nature444(7120),752–755.

Benitez-Nelson,C.R.,2000.Thebiogeochemicalcyclingofphosphorusinmarinesystems.

EarthSci.Rev.51(1),109–135.

Benner,R.,Maccubbin,A.E.,Hodson,R.E.,1984.Anaerobicbiodegradationofthelignin

andpolysaccharidecomponentsoflignocelluloseandsyntheticligninbysedimentmicroﬂora.Appl.Environ.Microbiol.47,998–1004.

Berelson,W.M.,McManus,J.,Coale,K.H.,Johnson,K.S.,Kilgore,T.,Burdige,D.,Pilskaln,C.,

1996.Biogenicmatterdiagenesisontheseaﬂoor:acomparisonbetweentwoconti-nentalmargintransects.J.Mar.Res.54,731–762.

Berner,R.A.,1984.Sedimentarypyriteformation:anupdate.Geochim.Cosmochim.Acta48,

605–615.

Berner,R.A.,Raiswell,R.,1983.Burialoforganiccarbonandpyritesulfurinsediments

overPhanerozoictime:anewtheory.Geochim.Cosmochim.Acta47,855–862.

Bernstein,R.E.,Byrne,R.H.,Betzer,P.R.,Greco,A.M.,1992.Morphologiesandtransforma-tionsofcelestiteinseawater:theroleofacanthariansinstrontiumandbariumgeo-chemistry.Geochim.Cosmochim.Acta56,3273–3279.

Bernstein,R.E.,Byrne,R.H.,Schijf,J.,1998.Acantharians:amissinglinkintheoceanicbio-geochemistryofbarium.Deep-SeaRes.IOceanogr.Res.Pap.45(2–3),491–505.Bertram,M.A.,Cowen,J.P.,1997.Morphologicalandcompositionalevidenceforbiotic

precipitationofmarinebarite.J.Mar.Res.55,577–593.

Betts,J.N.,Holland,H.D.,1991.Theoxygencontentofoceanbottomwaters,theburialef-ﬁciencyoforganiccarbon,andtheregulationofatmosphericoxygen.Palaeogeogr.Palaeoclimatol.Palaeoecol.97(1),5–18.

Bishop,J.K.B.,1988.Thebarite–opal-organiccarbonassociationinoceanicparticulate

matter.Nature332,341–343.

Bonn,W.J.,Gingele,F.X.,Grobe,H.,Mackensen,A.,Fütterer,D.K.,1998.Palaeoproductivity

attheAntarcticcontinentalmargin:opalandbariumrecordsforthelast400ka.Palaeogeogr.Palaeoclimatol.Palaeoecol.139(3),195–211.

Bostrom,K.,Backman,J.,1990.GeochemistryandoriginofNeogenesedimentsinHole

711A.In:Duncan,R.A.,Backman,J.,Peterson,L.C.,etal.(Eds.),Proc.OceanDrill.Pro-gramSci.Results115,699–708.

Bralower,T.J.,Thierstein,H.R.,1987.OrganiccarbonandmetalaccumulationratesinHo-loceneandmid-Cretaceoussediments:palaeoceanographicsigniﬁcance.In:Brooks,J.,Fleet,A.J.(Eds.),MarinePetroleumSourceRocks.London,GeologicalSocietySpecialPublicationvol.26,pp.345–369.

Bremner,J.M.,Rogers,J.,1990.PhosphoritedepositsontheNamibiancontinentalshelf.

In:Burnett,W.C.,Riggs,S.R.(Eds.),PhosphateDepositsoftheWorldNeogenetoMod-ernPhosphoritesvol.3.CambridgeUniversityPress,Cambridge,pp.143–152.Broecker,W.S.,1991.Thegreatoceanconveyor.Oceanography4(2),79–89.

Brumsack,H.J.,2006.Thetracemetalcontentofrecentorganiccarbon-richsediments:

implicationsforCretaceousblackshaleformation.Palaeogeogr.Palaeoclimatol.Palaeoecol.232(2),344–361.

Calvert,S.E.,Pedersen,T.F.,2007.Elementalproxiesforpalaeoclimaticand

palaeoceanographicvariabilityinmarinesediments:interpretationandappli-cation.In:Hillaire-Marcel,C.,DeVernal,A.(Eds.),ProxiesinLateCenozoicPaleoceanographyDevelopmentsinMarineGeology1.Elsevier,pp.567–644.Canﬁeld,D.E.,1994.Factorsinﬂuencingorganiccarbonpreservationinmarinesediments.

Chem.Geol.114,315–329.

Canﬁeld,D.E.,2005.Theearlyhistoryofatmosphericoxygen:homagetoRobertM.

Garrels.Annu.Rev.EarthPlanet.Sci.33,1–36.

Colberg,P.J.,1988.Anaerobicmicrobialdegradationofcellulose,lignin,oliolignols,and

monoaromaticligninderivatives.In:Zehnder,A.J.B.(Ed.),BiologyofAnaerobicOr-ganisms.Wiley,pp.333–372.

Compton,J.S.,Snyder,S.W.,Hodell,D.A.,1990.Phosphogenesisandweatheringofshelf

sedimentsfromthesoutheasternUnitedStates:implicationsforMioceneδ13Cexcur-sionsandglobalcooling.Geology18,1227–1230.

Curry,W.B.,Lohmann,G.P.,1986.LateQuaternarycarbonatesedimentationattheSierra

Leonerise(easternequatorialAtlanticOcean).MarineGeology70(3),223–250.Daly,A.R.,Edman,J.D.,1987.Lossoforganiccarbonfromsourcerocksduringthermal

maturation.Am.Assoc.Pet.Geol.Bull.71(5),546.

Dehairs,F.,Chesselet,R.,Jedwab,J.,1980.Discretesuspendedparticlesofbariteandthe

bariumcycleintheopenocean.EarthPlanet.Sci.Lett.49,528–550.

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Dehairs,F.,Lambert,C.E.,Chesselet,R.,Risler,N.,1987.Thebiologicalproductionofma-rinesuspendedbariteandthebariumcycleintheWesternMediterraneanSea.Bio-geochemistry4,119–139.

Dehairs,F.,Goeyens,L.,Stroobants,N.,Bernard,P.,Goyet,C.,Poisson,A.,Chesselet,R.,

1990.OnsuspendedbariteandtheoxygenminimumintheSouthernOcean.Glob.Biogeochem.Cycles4,85–102.

Dehairs,F.,Baeyens,W.,Goeyens,L.,1992.Accumulationofsuspendedbariteatmesope-lagicdepthsandexportproductionintheSouthernOcean.Science258,1332–1335.Dehairs,F.,Fagel,N.,Antiam,A.N.,Peinert,R.,Elskens,M.,Goeyens,L.,2000.Exportpro-ductionintheGulfofBiscayasestimatedfrombarium–bariteinsettlingmaterial:acomparisonwithnewproduction.Deep-SeaRes.I47,583–601.

Delaney,M.L.,1998.Phosphorusaccumulationinmarinesedimentsandtheoceanic

phosphoruscycle.Glob.Biogeochem.Cycles12(4),563–572.

Delaney,M.L.,Filippelli,G.M.,1994.Anapparentcontradictionintheroleofphosphorus

inCenozoicchemicalmassbalancesfortheworldocean.Paleoceanography9(4),513–527.

DesCombes,J.,Caulet,J.P.,Tribovillard,N.P.,1999.Pelagicproductivitychangesinthe

equatorialareaofthenorthwestIndianOceanduringthelast400,000years.Mar.Geol.158(1),27–55.

DesCombes,J.,Caulet,J.P.,Tribovillard,N.,2005.MonitoringthevariationsoftheSocotra

upwellingsystemduringthelast250kyr:abiogenicandgeochemicalapproach.Palaeogeogr.Palaeoclimatol.Palaeoecol.223(3),243–259.

Deuser,W.G.,1971.Organic-carbonbudgetoftheBlackSea.Deep-SeaRes.18,995–1004.Díaz,J.,Ingall,E.,Benitez-Nelson,C.,Paterson,D.,deJonge,M.D.,McNulty,I.,Brandes,J.A.,

2008.Marinepolyphosphate:akeyplayeringeologicphosphorussequestration.Science320,652–655.

Díaz-Ochoa,J.A.,Lange,C.B.,Pantoja,S.,DeLange,G.J.,Gutiérrez,D.,Muñoz,P.,Salamanca,

M.,2009.FishscalesinsedimentsfromoffCallao,centralPeru.Deep-SeaRes.II56(16),1124–1135.

Dickens,G.,Fewless,T.,Thomas,E.,Bralower,T.J.,2003.Excessbariteaccumulationduring

thePaleocene–Eocenethermalmaximum:massiveinputofdissolvedbariumfromseaﬂoorgashydratereservoirs.In:Wing,S.L.,Gingerich,P.D.,Schmitz,B.,Thomas,E.(Eds.),CausesandconsequencesofgloballywarmclimatesintheearlyPaleogeneGeologicalSocietyofAmericaSpecialPaper369.GeologicalSocietyofAmerica,Boulder,Colorado,pp.11–23.

Diester-Haass,L.,Meyers,P.A.,Vidal,L.,Wefer,G.,2001.Datareport:sandfraction,car-bonateandorganiccarboncontentsofLateMiocenesedimentsfromsite1085,mid-dleCapeBasin.In:Berger,W.H.,Wefer,G.,Richter,C.(Eds.),Proc.OceanDrill.ProgramSci.Results175,1–23.

Ding,H.,Sun,M.-Y.,2005.Biochemicaldegradationofalgalfattyacidsinoxicandanoxic

sediment–seawaterinterfacesystems:effectsofstructuralassociationandrelativerolesofaerobicandanaerobicbacteria.Mar.Chem.93(1),1–19.

Dunne,J.P.,Armstrong,R.A.,Gnanadesikan,A.,Sarmiento,J.L.,2005.Empiricalandmech-anisticmodelsfortheparticleexportratio.Glob.Biogeochem.Cycles19,GB4026(16pp.).

Dymond,J.,1981.GeochemistryofNazcaplatesurfacesediments:anevaluationofhydro-thermal,biogenic,detrital,andhydrogenoussources.In:Kulm,L.D.,Dymond,J.,Dasch,E.J.,Hussong,D.M.(Eds.),NazcaPlate:CrustalFormationandAndeanConver-gence.GeologicalSocietyofAmericaMemoir154,pp.133–173.

Dymond,J.,Collier,R.,1996.Particulatebariumﬂuxesandtheirrelationshipstobiological

productivity.Deep-SeaRes.II43,1283–1308.

Dymond,J.,Corliss,J.B.,Heath,G.R.,1977.HistoryofmetalliferoussedimentationatDeep

SeaDrillingsite319intheSouthEasternPaciﬁc.Geochim.Cosmochim.Acta41(6),741–753.

Dymond,J.,Suess,E.,Lyle,M.,1992.Bariumindeep-seasediment:ageochemicalproxy

forpaleoproductivity.Paleoceanography7(2),163–181.

Eagle,M.,Paytan,A.,Arrigo,K.R.,vanDijken,G.,Murray,R.W.,2003.Acomparisonbe-tweenexcessbariumandbariteasindicatorsofcarbonexport.Paleoceanography18(1),1021(13pp.).

Emeis,K.C.,Doose,H.,Mix,A.C.,Schulz-Bull,D.,1995.Alkenonesea-surfacetemperatures

andcarbonburialatSite846(easternequatorialPaciﬁcOcean):thelast1.3My.In:Mayer,L.A.,Pisias,N.G.,Palmer-Julson,A.,vanAndel,T.H.(Eds.),Proc.OceanDrill.ProgramSci.Results138,605–614.

Erwin,D.H.,Bowring,S.A.,Jin,Y.G.,2002.End-Permianmassextinctions:areview.In:

Koeberl,C.,McLeod,K.G.(Eds.),CatastrophicEventsandMassExtinctions:ImpactsandBeyondGeologicalSocietyofAmericaSpecialPaper356.GeologicalSocietyofAmerica,Boulder,Colorado,pp.363–384.

Fagel,N.,Dehairs,F.,Peinert,R.,Antia,A.,André,L.,2004.Reconstructingexportproduc-tionattheNEAtlanticmargin:potentialandlimitsoftheBaproxy.Mar.Geol.204,11–25.

Fahey,T.J.,Knapp,A.K.(Eds.),2007.PrinciplesandStandardsforMeasuringPrimaryPro-duction.OxfordUniversityPress,NewYork,NewYork,USA288pp.

Falkner,K.K.,Klinkhammer,G.P.,Bowers,T.S.,Todd,J.F.,Lewis,B.L.,Landing,W.M.,

Edmond,J.M.,1993.Thebehaviorofbariuminanoxicmarinewaters.Geochim.Cosmochim.Acta57,537–554.

Felix,M.,2014.Acomparisonofequationscommonlyusedtocalculateorganiccarbon

contentandmarinepalaeoproductivityfromsedimentdata.Mar.Geol.347,1–11.Filippelli,G.M.,2001.CarbonandphosphoruscyclinginanoxicsedimentsoftheSaanich

Inlet,BritishColumbia.Mar.Geol.174,307–321.

Filippelli,G.M.,Delaney,M.L.,1995.Phosphorusgeochemistryandaccumulationratesin

theeasternequatorialPaciﬁc.In:Mayer,L.A.,Pisias,N.G.,Palmer-Julson,A.,vanAndel,T.H.(Eds.),Proc.OceanDrill.ProgramSci.Results138,757–767.

Filippelli,G.M.,Delaney,M.L.,1996.PhosphorusgeochemistryofequatorialPaciﬁcsedi-ments.Geochim.Cosmochim.Acta60,1479–1495.

Filippelli,G.M.,Souch,C.,1999.Effectsofclimateandlandscapedevelopmentonthe

terrestrialphosphoruscycle.Geology27,171–174.

Fisher,N.S.,Guillard,R.R.,Bankston,D.C.,1991.Theaccumulationofbariumbymarine

phytoplanktongrowninculture.J.Mar.Res.49(2),339–354.

Föllmi,K.B.,1990.Condensationandphosphogenesis:exampleoftheHelveticmid-Cretaceous(northernTethyanmargin).In:Notholt,A.J.G.,Jarvis,I.(Eds.),PhosphoriteResearchandDevelopment:GeologicalSocietyofLondonSpecialPublication52,pp.237–252.

Föllmi,K.B.,1996.Thephosphoruscycle,phosphogenesisandmarinephosphate-richde-posits.EarthSci.Rev.40,55–124.

François,R.,Honjo,S.,Manganini,S.J.,Ravizza,G.E.,1995.Biogenicbariumﬂuxestothe

deepsea:implicationsforpaleoproductivityreconstruction.Glob.Biogeochem.Cy-cles9(2),289–303.

Frank,M.,Gersonde,R.,Loeff,M.R.,Bohrmann,G.,Nürnberg,C.C.,Kubik,P.W.,Mangini,A.,

2000.SimilarglacialandinterglacialexportbioproductivityintheAtlanticsectoroftheSouthernOcean:multiproxyevidenceandimplicationsforglacialatmosphericCO2.Paleoceanography15(6),642–658.

Freudenthal,T.,Neuer,S.,Meggers,H.,Davenport,R.,Wefer,G.,2001.Inﬂuenceoflateral

particleadvectionandorganicmatterdegradationonsedimentaccumulationandstablenitrogenisotoperatiosalongaproductivitygradientintheCanaryIslandsre-gion.Mar.Geol.177(1),93–109.

Froelich,P.N.,Klinkhammer,G.P.,Bender,M.L.,Luedtke,G.R.,Heath,G.R.,Cullen,D.,

Dauphin,P.,Hammond,D.,Hartman,B.,Maynard,V.,1979.EarlyoxidationoforganicmatterinpelagicsedimentsoftheeasternequatorialAtlantic:suboxicdiagenesis.Geochim.Cosmochim.Acta43,1075–1090.

Gächter,R.,Müller,B.,2003.Whythephosphorusretentionoflakesdoesnotnecessarily

dependontheoxygensupplytotheirsedimentsurface.Limnol.Oceanogr.48,929–933.

Gallego-Torres,D.,Martínez-Ruiz,F.,Paytan,A.,Jiménez-Espejo,F.J.,Ortega-Huertas,M.,

2007.Pliocene–HoloceneevolutionofdepositionalconditionsintheeasternMediter-ranean:roleofanoxiavs.productivityattimeofsapropeldeposition.Palaeogeogr.Palaeoclimatol.Palaeoecol.246(2),424–439.

Galloway,J.N.,Dentener,F.J.,Capone,D.G.,Boyer,E.W.,Howarth,R.W.,Seitzinger,S.P.,

Asner,G.P.,Vöosmarty,C.J.,2004.Nitrogencycles:past,present,andfuture.Biogeo-chemistry70(2),153–226.

Ganeshram,R.S.,Pedersen,T.F.,1998.Glacial–interglacialvariabilityinupwelling

andbioproductivityoffNWMexico:implicationsforQuaternarypaleoclimate.Paleoceanography13(6),634–645.

Ganeshram,R.S.,Calvert,S.E.,Pedersen,T.F.,Cowie,G.L.,1999.Factorscontrollingthe

burialoforganiccarboninlaminatedandbioturbatedsedimentsoffNWMexico:im-plicationsforhydrocarbonpreservation.Geochim.Cosmochim.Acta63,1723–1734.Ganeshram,R.S.,François,R.,Commeau,J.,Brown-Leger,S.L.,2003.Anexperimentalin-vestigationofbariteformationinseawater.Geochim.Cosmochim.Acta67(14),2599–2605.

Gardner,J.V.,Dean,W.E.,Dartnell,P.,1997.BiogenicsedimentationbeneaththeCal-iforniaCurrentsystemforthepast30kyranditspaleoceanographicsigniﬁcance.Paleoceanography12(2),207–225.

Gingele,F.,Dahmke,A.,1994.Discretebariteparticlesandbariumastracersof

paleoproductivityinSouthAtlanticsediments.Paleoceanography9(1),151–168.Giraudeau,J.,Meyers,P.A.,Christensen,B.A.,2002.Accumulationoforganicandinorganic

carboninPliocene–PleistocenesedimentsalongtheSWAfricanmargin.Mar.Geol.180(1),49–69.

Goldhammer,T.,Brüchert,V.,Ferdelman,T.G.,Zabel,M.,2010.Microbialsequestrationof

phosphorusinanoxicupwellingsediments.Nat.Geosci.3,557–561.

Goñi,M.A.,Ruttenberg,K.C.,Eglinton,T.I.,1997.Sourcesandcontributionofterrigenous

organiccarbontosurfacesedimentsintheGulfofMexico.Nature389,275–278.Gonneea,M.E.,Paytan,A.,2006.Phaseassociationsofbariuminmarinesediments.Mar.

Chem.100,124–135.

Gonzalez-Muñoz,M.T.,Martinez-Ruiz,F.,Morcillo,F.,Martin-Ramos,J.D.,Paytan,A.,2012.

Precipitationofbaritebymarinebacteria:apossiblemechanismformarinebariteformation.Geology40(8),675–678.

Gooday,A.J.,Nott,J.A.,1982.Intracellularbaritecrystalsintwoxenophyophores,

AschemonellaramuliformisandGalatheamminasp.(Protozoa:Rhizopoda)withcom-mentsonthetaxonomyofA.ramuliformis.J.Mar.Biol.Assoc.U.K.62,595–605.Grifﬁth,E.M.,Paytan,A.,2012.Bariteintheocean–occurrence,geochemistryand

palaeoceanographicapplications.Sedimentology59(6),1817–1835.

Hartnett,H.E.,Keil,R.G.,Hedges,J.I.,Devol,A.H.,1998.Inﬂuenceofoxygenexposuretimeon

organiccarbonpreservationincontinentalmarginsediments.Nature391,572–574.Hashimoto,S.,Horimoto,N.,Yamaguchi,Y.,Ishimaru,T.,Saino,T.,2005.Relationshipbe-tweennetandgrossprimaryproductionintheSagamiBay,Japan.Limnol.Oceanogr.50(6),1830–1835.

Hedges,J.I.,Keil,R.G.,1995.Sedimentaryorganicmatterpreservation:anassessmentand

speculativesynthesis.Mar.Chem.49,81–115.

Hedges,J.I.,Parker,P.L.,1976.Land-derivedorganicmatterinsurfacesedimentsfromthe

GulfofMexico.Geochim.Cosmochim.Acta40,1019–1029.

Henrichs,S.M.,1995.Sedimentaryorganicmatterpreservation:anassessmentand

speculativesynthesis—acomment.Mar.Chem.49(2–3),127–136.

Henrichs,S.M.,Reeburgh,W.S.,1987.Anaerobicmineralizationofmarinesediment

organicmatter:ratesandtheroleofanaerobicprocessesintheoceaniccarbonecon-omy.GeomicrobiolJ.5,191–238.

Hiatt,E.E.,Budd,D.A.,2003.ExtremepaleoceanographicconditionsinaPaleozoicoceanic

upwellingsystem:organicproductivityandwidespreadphosphogenesisinthePermianPhosphoriaSea.In:Chan,M.A.,Archer,A.W.(Eds.),ExtremeDepositionalEnvironments:MegaEndMembersinGeologicTime.GeologicalSocietyofAmerica,SpecialPaper370,pp.245–264.

Hirst,D.M.,1974.GeochemistryofsedimentsfromelevenBlackSeacores.In:Degens,E.T.,

Ross,D.A.(Eds.),TheBlackSea—Geology,Chemistry,andBiology.AmericanAssocia-tionofPetroleumGeologistsMemoirsvol.20,pp.430–455.

50S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Hori,R.,1992.RadiolarianbiostratigraphyattheTriassic/Jurassicperiodboundaryinbed-dedchertsfromtheInuyamaArea,centralJapan.J.Geosci.OsakaCityUniv.35,53–65.Hotinski,R.M.,Bice,K.L.,Kump,L.R.,Najjar,R.G.,Arthur,M.A.,2001.Oceanstagnationand

end-Permiananoxia.Geology29,7–10.

Hulthe,G.,Hulth,S.,Hall,P.O.J.,1998.Effectofoxygenondegradationrateofrefractory

andlabileorganicmatterincontinentalmarginsediments.Geochim.Cosmochim.Acta62(8),1319–1328.

Hyun,S.,Ortiz,J.D.,Raymo,M.E.,Taira,A.,1999.Low-frequencyoscillationsinSite983

sediments:relationshipbetweencarbonateandproductivityproxies.In:Raymo,M.E.,Jansen,E.,Blum,P.,Herbert,T.(Eds.),Proc.OceanDrill.ProgramSci.Results162,197.

Ingall,E.D.,Jahnke,R.A.,1997.Inﬂuenceofwater-columnanoxiaontheelementalfrac-tionationofcarbonandphosphorusduringsedimentdiagenesis.Mar.Geol.139,219–229.

Ingall,E.D.,VanCappellen,P.,1990.Relationbetweensedimentationrateandburialofor-ganicphosphorusandorganiccarboninmarinesediments.Geochim.Cosmochim.Acta54,373–386.

Ingall,E.D.,Bustin,R.M.,VanCappellen,P.,1993.Inﬂuenceofwatercolumnanoxiaonthe

burialandpreservationofcarbonandphosphorusinmarineshales.Geochim.Cosmochim.Acta57,303–316.

Ingall,E.D.,Kolowith,L.,Lyons,T.,Hurtgen,M.,2005.Sedimentcarbon,nitrogenand

phosphoruscyclinginananoxicfjord,EfﬁnghamInlet,BritishColumbia.Am.J.Sci.305,240–258.

Isozaki,Y.,1997.Permo–Triassicboundarysuperanoxiaandstratiﬁedsuperocean:

recordsfromlostdeep-sea.Science276,235–238.

Iversen,M.H.,Ploug,H.,2010.Ballastmineralsandthesinkingcarbonﬂuxintheocean:

carbon-speciﬁcrespirationratesandsinkingvelocityofmarinesnowaggregates.Bio-geosciences7(9),2613–2624.

Jaccard,S.L.,Galbraith,E.D.,Sigman,D.M.,Haug,G.H.,Francois,R.,Pedersen,T.F.,Dulski,P.,

Thierstein,H.R.,2009.SubarcticPaciﬁcevidenceforaglacialdeepeningoftheoceanicrespiredcarbonpool.EarthPlanet.Sci.Lett.277(1),156–165.

Jaccard,S.L.,Hayes,C.T.,Martínez-García,A.,Hodell,D.A.,Anderson,R.F.,Sigman,D.M.,

Haug,G.H.,2013.TwomodesofchangeinSouthernOceanproductivityoverthepastmillionyears.Science229,1419–1422.

Jeandel,C.,Tachikawa,K.,Bory,A.,Dehairs,F.,2000.Biogenicbariuminsuspendedand

trappedmaterialasatracerofexportproductioninthetropicalN-EAtlantic(EUMELISites).Mar.Chem.71,125–142.

Jewell,P.W.,1994.PaleoredoxconditionsandtheoriginofbeddedbaritealongtheLate

DevonianNorthAmericancontinentalmargin.J.Geol.102,151–164.

Kasten,S.,Haese,R.R.,Zabel,M.,Ruhlemann,C.,Schulz,H.D.,2001.Bariumpeaksatglacial

terminationsinsedimentsoftheequatorialAtlanticOcean:relictsofdeglacialpro-ductivitypulses?Chem.Geol.175,635–651.

Kennedy,M.J.,Wagner,T.,2011.Claymineralcontinentalampliﬁerformarinecarbonse-questrationinagreenhouseocean.Proc.Natl.Acad.Sci.U.S.A.108(24),9776–9781.Kennedy,M.J.,Pevear,D.R.,Hill,R.J.,2002.Mineralsurfacecontroloforganiccarbonin

blackshale.Science295,657–660.

Kennedy,M.J.,Löhr,S.C.,Fraser,S.A.,Baruch,E.T.,2014.Directevidencefororganiccarbon

preservationasclay-organicnanocompositesinaDevonianblackshale;fromdeposi-tiontodiagenesis.EarthPlanet.Sci.Lett.388,59–70.

Kinkel,H.,Baumann,K.-H.,Čepek,M.,2000.CoccolithophoresintheequatorialAtlantic

Ocean:responsetoseasonalandLateQuaternarysurfacewatervariability.Mar.Micropaleontol.39,87–112.

Klöcker,R.,Ganssen,G.,Jung,S.J.A.,Kroon,D.,Henrich,R.,2006.LateQuaternary

millennial-scalevariabilityinpelagicaragonitepreservationoffSomalia.Mar.Micropaleontol.59,171–183.

Klump,J.,Hebbeln,D.,Wefer,G.,2000.Theimpactofsedimentprovenanceonbarium-basedproductivityestimates.Mar.Geol.169(3),259–271.

Kraal,P.,2010.Redox-dependentPhosphorusBurialinModernandAncientMarine

Sediments(Ph.D.dissertation)UniversiteitUtrecht,Netherlands(176pp.).

Kristensen,E.,Holmer,M.,2001.Decompositionofplantmaterialsinmarinesediment

exposedtodifferentelectronacceptors(O2,NO−3,andSO24−),withemphasisonsub-strateorigin,degradationkinetics,andtheroleofbioturbation.Geochim.Cosmochim.Acta65(3),419–433.

Kuypers,M.M.M.,Pancost,R.D.,Nijenhuis,I.A.,SinningheDamsté,J.S.,2002.Enhancedpro-ductivityledtoincreasedorganiccarbonburialintheeuxinicNorthAtlanticbasindur-ingthelateCenomanianoceanicanoxicevent.Paleoceanography17(4),1051(13pp.).Latimer,J.C.,Filippelli,G.M.,2002.EocenetoMioceneterrigenousimportsandexport

production:geochemicalevidencefromODPLeg177,Site1090.Palaeogeogr.Palaeoclimatol.Palaeoecol.182,151–164.

LeMoigne,F.A.C.,Henson,S.A.,Sanders,R.J.,Madsen,E.,2013.Globaldatabaseofsurface

oceanparticulateorganiccarbonexportﬂuxesdiagnosedfromthe234Thtechnique.EarthSyst.Sci.DataDiscuss.6,163–187.

Lee,C.B.,1992.Controlsonorganiccarbonpreservation:theuseofstratiﬁedwaterbodies

tocompareintrinsicratesofdecompositioninoxicandanoxicsystems.Geochim.Cosmochim.Acta56(8),3323–3335.

Leinen,M.,Graybeal,A.,1986.SedimentationinthevicinityofLeg-92drillsites:studiesof

sitesurveycores.In:Bailey,M.(Ed.),InitialRep.DeepSeaDrill.Proj.92,237–251.Levinton,J.S.,2008.MarineBiology:Function,Biodiversity,Ecology,3rded.OxfordUni-versityPress,(640pp.).

Lisecki,L.E.,Raymo,M.E.,2005.APliocene–Pleistocenestackof57globallydistributed

benthicd18Orecords.Paleoceanography20(1),PA1003.

Longhurst,A.R.,2010.EcologicalgeographyoftheseaAccessOnlineviaScienceDirect

http://www.sciencedirect.com/science/book/9780124555211.

Longhurst,A.,Sathyendranath,S.,Platt,T.,Caverhill,C.,1995.Anestimateofglobalprima-ryproductionintheoceanfromsatelliteradiometerdata.J.PlanktonRes.17(6),1245–1271.

Lowenstein,T.K.,Timofeeff,M.N.,Brennan,S.T.,Hardie,L.A.,Demicco,R.V.,2001.Oscilla-tionsinPhanerozoicseawaterchemistry:evidencefromﬂuidinclusions.Science294(5544),1086–1088.

Lückge,A.,Boussaﬁr,M.,Lallier-Vergès,E.,Littke,R.,1996.Comparativestudyoforganic

matterpreservationinimmaturesedimentsalongthecontinentalmarginsofPeruandOman.PartI:resultsofpetrographicalandbulkgeochemicaldata.Org.Geochem.24(4),437–451.

Luo,G.M.,Kump,L.R.,Wang,Y.,Tong,J.,Arthur,M.A.,Yang,H.,Huang,J.,Yin,H.,Xie,S.,

2010.Isotopeevidenceforananomalouslylowoceanicsulfateconcentrationfollow-ingend-Permianmassextinction.EarthPlanet.Sci.Lett.300,101–111.

Lyle,A.O.,Lyle,M.,2005.OrganiccarbonandbariuminEocenesediments:possiblecon-trolsonnutrientrecyclingintheEoceneequatorialPaciﬁcOcean.In:Wilson,P.A.,Lyle,M.W.,Firth,J.(Eds.),Proc.OceanDrill.ProgramSci.Results199,1–33.

Lyle,M.,Zahn,R.,Prahl,F.,Dymond,J.,Collier,R.,Pisias,N.,Suess,E.,1992.

PaleoproductivityandcarbonburialacrosstheCaliforniaCurrent:themultitracerstransect,42°N.Paleoceanography7(3),251–272.

Marra,J.,2008.Approachestothemeasurementofplanktonproduction.In:Williams,P.J.,

Le,B.,Thomas,D.N.,Reynolds,C.S.(Eds.),PhytoplanktonProductivity:CarbonAssim-ilationinMarineandFreshwaterEcology.Wiley-BlackwellPublishing,pp.78–108.Marra,J.,2009.Netandgrossproductivity:weighinginwith14C.Aquat.Microb.Ecol.56,

123–131.

Martin,R.E.,1995.Cyclicandsecularvariationinmicrofossilbiomineralization:cluesto

thebiogeochemicalevolutionofPhanerozoicoceans.Glob.Planet.Chang.11(1),1–23.

Martinez-Ruiz,F.,Kastner,M.,Paytan,A.,Ortega-Huertas,M.,Bernasconi,S.M.,2000.Geo-chemicalevidenceforenhancedproductivityduringS1sapropeldepositionintheeasternMediterranean.Paleoceanography15,200–209.

Martinez-Ruiz,F.,Paytan,A.,Kastner,M.,Gonzalez-Donoso,J.M.,Linares,D.,Bernasconi,S.

M.,Jimenez-Espejo,F.J.,2003.Acomparativestudyofthegeochemicalandmineral-ogicalcharacteristicsoftheS1sapropelinthewesternandeasternMediterranean.Palaeogeogr.Palaeoclimatol.Palaeoecol.190,23–37.

McLennan,S.M.,2001.Relationshipsbetweenthetraceelementcompositionofsedimen-taryrocksanduppercontinentalcrust.Geochem.Geophys.Geosyst.2(2000GC00109,24pp.).

McManus,J.,Berelson,W.M.,Klinkhammer,G.P.,Kilgore,T.M.,Hammond,D.E.,1994.Re-mobilizationofbariumincontinentalmarginsediments.Geochim.Cosmochim.Acta58(22),4899–4907.

McManus,J.,Berelson,W.M.,Klinkhammer,G.P.,Johnson,K.S.,Coale,K.H.,Anderson,R.F.,

Kumar,N.,Burdige,D.J.,Hammond,D.E.,Brumsack,H.J.,McCorckle,D.C.,Rushdi,A.,1998.Geochemistryofbariuminmarinesediments:implicationsforitsuseasapaleoproxy.Geochim.Cosmochim.Acta62,3453–3473.

McManus,J.,Berelson,W.M.,Hammond,D.E.,Klinkhammer,G.P.,1999.Bariumcyclingin

theNorthPaciﬁc:implicationsfortheutilityofBaasapaleoproductivityandpaleoalkalinityproxy.Paleoceanography14(1),53–61.

Meyers,P.A.,1997.Organicgeochemicalproxiesofpaleoceanographic,paleolimnologic,

andpaleoclimaticprocesses.Org.Geochem.27,213–250.

Mills,R.A.,Taylor,S.L.,Pälike,H.,Thomson,J.,2010.Hydrothermalsedimentsrecord

changesindeepwateroxygencontentintheSEPaciﬁc.Paleoceanography25,PA4226(16pp.).

Modica,C.J.,Lapierre,S.G.,2012.Estimationofkerogenporosityinsourcerocksasafunc-tionofthermaltransformation:examplefromtheMowryShaleinthePowderRiverBasinofWyoming.Am.Assoc.Pet.Geol.Bull.96(1),87–108.

Monnin,C.,Jeandel,C.,Cattaldo,T.,Dehairs,F.,1999.Themarinebaritesaturationstateof

theworld'soceans.Mar.Chem.65,253–261.

Moody,J.B.,Chaboudy,L.R.,Worsley,T.R.,1988.Paciﬁcpelagicphosphorusaccumulation

duringthelast10my.Paleoceanography3(1),113–136.

Moreno,A.,Nave,S.,Kuhlmann,H.,Canals,M.,Targarona,J.,Freudenthal,T.,Abrantes,F.,

2002.ProductivityresponseintheNorthCanaryBasintoclimatechangesduringthelast250,000yr:amulti-proxyapproach.EarthPlanet.Sci.Lett.196,147–159.

Mort,H.P.,Adatte,T.,Föllmi,K.B.,Keller,G.,Steinmann,P.,Matera,V.,Berner,Z.,Stuben,D.

,2007.Phosphorusandtherolesofproductivityandnutrientrecyclingduringocean-icanoxicevent2.Geology35,483–486.

Müller,P.J.,Suess,E.,1979.Productivity,sedimentationrate,andsedimentaryorganic

matterintheocean—I.Organiccarbonpreservation.Deep-SeaRes.26,1347–1362.Muñoz,P.,Dezileau,L.,Cardenas,L.,Sellanes,J.,Lange,C.B.,Inostroza,J.,Muratli,J.,

Salamanca,M.A.,2012.GeochemistryoftracemetalsinshelfsedimentsaffectedbyseasonalandpermanentlowoxygenconditionsoffcentralChile,SEPaciﬁc(~36°S).Cont.ShelfRes.33,51–68.

Murphy,A.E.,Sageman,B.B.,Hollander,D.J.,Lyons,T.W.,Brett,C.E.,2000.Blackshaledeposi-tionintheDevonianAppalachianBasin:siliciclasticstarvation,episodicwater-columnmixing,andefﬁcientrecyclingofbiolimitingnutrients.Paleoceanography15,280–291.Murray,D.W.,Prell,W.L.,1991.PliocenetoPleistocenevariationsincalciumcarbonate,or-ganiccarbon,andopalontheOwenRidge,northernArabianSea.In:Emeis,K.C.,Meyers,P.A.,Niitsuma,N.,Prell,W.L.(Eds.),Proc.OceanDrill.ProgramSci.Results117,343–363.

Murray,R.W.,Knowlton,C.W.,Leinen,M.,Mix,A.C.,Polsky,C.H.,2000.Exportproduction

andcarbonatedissolutioninthecentralequatorialPaciﬁcOceanoverthepast1Myr.Paleoceanography15(6),570–592.

Murray,R.W.,Christensen,B.A.,Kalbas,J.L.,Kryc,K.A.,2002.Plioceneexportproduction

andterrigenousprovenanceoftheSouthernCapeBasin,southwestAfricanmargin.Mar.Geol.180(1),133–150.

Murray,R.W.,Leinen,M.,Knowlton,C.W.,2012.Linksbetweenironinputandopaldepo-sitioninthePleistoceneequatorialPaciﬁcOcean.Nat.Geosci.5(4),270–274.

Nijenhuis,I.A.,Bosch,H.-J.,SinningheDamsté,J.S.,Brumsack,H.J.,DeLange,G.J.,1999.

Organicmatterandtraceelementrichsapropelsandblackshales:ageochemicalcomparison.EarthPlanet.Sci.Lett.169,277–290.

S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Nürnberg,C.C.,Bohrmann,G.,Schlüter,M.,Frank,M.,1997.Bariumaccumulationin

theAtlanticsectorofthesouthernocean:resultsfrom190,000-yearrecords.Paleoceanography12(4),594–603.

Opsahl,S.,Benner,R.,1997.Distributionandcyclingofterrigenousdissolvedorganic

matterintheocean.Nature386,480–482.

Owen,R.M.,Zimmerman,A.R.B.,1991.GeochemistryoftheCretaceous/Tertiaryboundary

atHole752B,BrokenRidge.In:Weissel,J.,Alt,J.,Peirce,J.,Taylor,E.(Eds.),Proc.OceanDrill.ProgramSci.Results121,423–433.

Pace,M.L.,Knauer,G.A.,Karl,D.M.,Martin,J.H.,1987.Primaryproduction,newproduction

andverticalﬂuxintheeasternPaciﬁcOcean.Nature325,803–804.

Paytan,A.,Grifﬁth,E.M.,2007.Marinebarite:recorderofvariationsinoceanexport

productivity.Deep-SeaRes.II54,687–705.

Paytan,A.,Kastner,M.,1996.BenthicBaﬂuxesinthecentralequatorialPaciﬁc,implica-tionsfortheoceanicBacycle.EarthPlanet.Sci.Lett.142,439–459.

Paytan,A.,McLaughlin,K.,2007.Theoceanicphosphoruscycle.Chem.Rev.107,563–576.Paytan,A.,Kastner,M.,Chavez,F.P.,1996.Glacialtointerglacialﬂuctuationsinproductiv-ityintheequatorialPaciﬁcasindicatedbymarinebarite.Science274,1355–1357.Paytan,A.,Averyt,K.,Faul,K.,Gray,E.,Thomas,E.,2007.Bariteaccumulation,oceanpro-ductivity,andSr/BainbariteacrossthePaleocene–Eocenethermalmaximum.Geol-ogy35,1139–1142.

Pedersen,T.F.,Calvert,S.E.,1990.Anoxiavs.productivity:whatcontrolstheformationof

organic-richsedimentsandsedimentaryrocks?Am.Assoc.Pet.Geol.Bull.74,454–466.Peters,K.E.,Magoon,L.B.,Bird,K.J.,Valin,Z.C.,Keller,M.A.,2006.Northslope,Alaska:

sourcerockdistribution,richness,thermalmaturity,andpetroleumcharge.Am.Assoc.Pet.Geol.Bull.90(2),261–292.

Pfeifer,K.,Kasten,S.,Hensen,C.,Schulz,H.D.,2001.Reconstructionofprimaryproductiv-ityfromthebariumcontentsinsurfacesedimentsoftheSouthAtlanticOcean.Mar.Geol.177(1),13–24.

Pirrung,M.,Illner,P.,Matthießen,J.,2008.Biogenicbariuminsurfacesedimentsofthe

EuropeanNordicSeas.Mar.Geol.250(1),89–103.

PrakashBabu,C.,Brumsack,H.-J.,Schnetger,B.,Böttcher,M.E.,2002.Bariumasaproduc-tivityproxyincontinentalmarginsediments:astudyfromtheeasternArabianSea.Mar.Geol.184,189–206.

Rageneau,O.,Tréguer,P.,Leynaert,A.,Anderson,R.F.,Brzezinski,M.A.,DeMaster,D.J.,

Dugdale,R.C.,Dymond,J.,Fischer,G.,François,R.,Heinze,C.,Maier-Reimer,E.,Martin-Jézéquel,V.,Nelson,D.M.,Quéguiner,B.,2000.AreviewoftheSicycleinthemodernocean:recentprogressandmissinggapsintheapplicationofbiogenicopalasapaleoproductivityproxy.Glob.Planet.Chang.26,317–365.

Raiswell,R.,Berner,R.A.,1987.Organiccarbonlossesduringburialandthermalmatura-tionofnormalmarineshales.Geology15(9),853–856.

Rao,V.P.,Lamboy,M.,1995.PhosphoritesfromtheOmanMargin,ODPLeg117.Oceanol.

Acta18(3),289–307.

Raymo,M.E.,Hodell,D.,Jansen,E.,1992.Responseofdeepoceancirculationtoinitiation

ofNorthernHemisphereglaciation(3–2Ma).Paleoceanography7(5),645–672.Rea,D.K.,Pisias,N.G.,Newberry,T.,1991.LatePleistocenepaleoclimatologyofthecentral

equatorialPaciﬁc:ﬂuxpatternsofbiogenicsediments.Paleoceanography6(2),227–244.

Redﬁeld,A.C.,1958.Thebiologicalcontrolofchemicalfactorsintheenvironment.Am.Sci.

46(3),205–221.

Reichart,G.J.,denDulk,M.,Visser,H.J.,vanderWeijden,C.H.,Zachariasse,W.J.,1997.A

225kyrrecordofdustsupply,paleoproductivityandtheoxygenminimumzonefromtheMurrayRidge(northernArabianSea).Palaeogeogr.Palaeoclimatol.Palaeoecol.134(1),149–169.

Reitz,A.,Pfeifer,K.,deLange,G.J.,Klump,J.,2004.BiogenicbariumandthedetritalBa/Al

ratio:acomparisonoftheirdirectandindirectdetermination.Mar.Geol.204,289–300.

Ridgwell,A.,2005.AMidMesozoicrevolutionintheregulationofoceanchemistry.Mar.

Geol.217(3),339–357.

Rieder,N.,Ott,H.A.,Pfundstein,P.,Schoch,R.,1982.X-raymicro-analysisofthemineral

contentsofsomeprotozoa.J.Protozool.29,15–18.

Robinson,R.S.,Meyers,P.A.,Murray,R.W.,2002.Geochemicalevidenceforvariationsin

deliveryanddepositionofsedimentinPleistocenelight–darkcolorcyclesundertheBenguelaCurrentUpwellingSystem.Mar.Geol.180(1),249–270.

Rostek,F.,Bard,E.,Beaufort,L.,Sonzogni,C.,Ganssen,G.,1997.Seasurfacetemperature

andproductivityrecordsforthepast240kyrintheArabianSea.Deep-SeaRes.II44,1461–1480.

Ruhlin,D.E.,Owen,R.M.,1986.Therareearthelementgeochemistryofhydrothermalsed-imentsfromtheEastPaciﬁcRise:examinationofaseawaterscavengingmechanism.Geochim.Cosmochim.Acta50(3),393–400.

Russell,A.D.,Morford,J.L.,2001.Thebehaviorofredox-sensitivemetalsacrossalaminat-ed–massive–laminatedtransitioninSaanichInlet,BritishColumbia.Mar.Geol.174(1),341–354.

Rutsch,H.-J.,Mangini,A.,Bonani,G.,Dittrich-Hannen,B.,Kubik,P.W.,Suter,M.,Segl,M.,

1995.10BeandBaconcentrationsinWestAfricansedimentstraceproductivityinthepast.EarthPlanet.Sci.Lett.133,129–143.

Ruttenberg,K.C.,Goñi,M.A.,1997.Phosphorusdistribution,C:N:Pratios,andδ13Cocinarc-tic,temperature,andtropicalcoastalsediments:toolsforcharacterizingbulksedi-mentaryorganicmatter.Mar.Geol.139,123–145.

Sadler,P.M.,1981.Sedimentaccumulationratesandthecompletenessofstratigraphic

sections.J.Geol.89,569–584.

Sarnthein,M.,Winn,K.,Duplessy,J.C.,Fontugne,M.R.,1988.Globalvariationsofsurface

oceanproductivityinlowandmidlatitudes:inﬂuenceonCO2reservoirsofthedeepoceanandatmosphereduringthelast21,000years.Paleoceanography3,361–399.

Schenau,S.J.,DeLange,G.J.,2001.Phosphorusregenerationvs.burialinsedimentsofthe

ArabianSea.Mar.Geol.75,201–217.

Schenau,S.J.,Prins,M.A.,DeLange,G.J.,Monnin,C.,2001.BariumaccumulationintheAra-bianSea:controlsonbaritepreservationinmarinesediments.Geochim.Cosmochim.Acta65(10),1545–1556.

Schenau,S.J.,Reichart,G.J.,DeLange,G.J.,2005.Phosphorusburialasafunctionof

paleoproductivityandredoxconditionsinArabianSeasediments.Geochim.Cosmochim.Acta69(4),919–931.

Schink,B.,1988.Principlesandlimitsofanaerobicdegradation:environmentalandtech-nologicalaspects.In:Zehnder,A.J.B.(Ed.),BiologyofAnaerobicOrganisms.Wiley,pp.771–846.

Schultz,H.,vonRad,U.,Erlenkeuser,H.,1998.CorrelationbetweenArabianSeaand

Greenlandclimateoscillationsofthepast110,000years.Nature393,54–57.

Scopelliti,G.,Bellanca,A.,Coccioni,R.,Luciani,V.,Neri,R.,Baudin,F.,Chiari,M.,

Marcucci,M.,2004.High-resolutiongeochemicalandbioticrecordsoftheTethy-an‘BonarelliLevel’(OAE2,latestCenomanian)fromtheCalabianca–Guidalocacompositesection,northwesternSicily,Italy.Palaeogeogr.Palaeoclimatol.Palaeoecol.208,293–317.

Shen,J.,Schoepfer,S.D.,Feng,Q.,Zhou,Lian,Yu,J.,Song,H.,Wei,H.,Algeo,T.J.,2014.Ma-rineproductivitychangesduringtheend-PermiancrisisandEarlyTriassicrecovery.EarthScienceReviews,(inreview).

Shimmield,G.B.,1992.Cansedimentgeochemistryrecordchangesincoastalupwelling

palaeoproductivity?EvidencefromnorthwestAfricaandtheArabianSea.In:Summerhayes,C.P.,Prell,W.L.,Emeis,K.C.(Eds.),UpwellingSystems:EvolutionSincetheMiocene.GeologicalSocietyofLondon,SpecialPublications64,pp.29–46.Shimmield,G.B.,Mowbray,S.R.,1991.Theinorganicgeochemicalrecordofthenorthwest

ArabianSea:ahistoryofproductivityvariationoverthelast400kyfromSites722and724.In:Emeis,K.C.,Meyers,P.A.,Niitsuma,N.,Prell,W.L.(Eds.),Proc.OceanDrill.ProgramSci.Results117,409–429.

Shimmield,G.,Derrick,S.,Mackensen,A.,Grobe,H.,Pudsey,C.,1994.Thehistoryofbari-um,biogenicsilicaandorganiccarbonaccumulationintheWeddellSeaandAntarcticOceanoverthelast150,000years.In:Zahn,R.,Kaminski,M.,Labeyrie,L.,Pedersen,T.(Eds.),CarbonCyclingintheGlacialOcean:ConstraintsontheOcean'sRoleinGlobalChangeNATOASISeries.Springer,Berlin,pp.555–574.

Showers,W.J.,Angle,D.G.,1986.Stableisotopiccharacterizationoforganiccarbonaccu-mulationontheAmazoncontinentalshelf.Cont.ShelfRes.6,227–244.

Simon,A.,Poulicek,M.,Velimirov,B.,MacKenzie,F.T.,1994.Comparisonofanaerobicand

aerobicbiodegradationofmineralizedskeletalstructuresinmarineandestuarineconditions.Biogeochemistry25,167–195.

Skjervoy,A.,Sylta,Ø.,1993.Modelingofexpulsionandsecondarymigrationalongthe

southwesternmarginoftheHordaplatform.In:Doré,A.G.,Augustson,J.H.,Stewart,D.J.,Sylta,Ø.(Eds.),BasinModeling:AdvancesandApplicationsProceedingsoftheNorwegianPetroleumSocietyConference,Stavanger,Norway,13–15March1991,NPFSpecialPublicationNo.3.Elsevier,pp.499–538.

Song,H.,Tong,J.,Algeo,T.J.,Song,H.,Qiu,H.,Zhu,Y.,Tian,L.,Bates,S.,Lyons,T.W.,Luo,G.,

Kump,L.,2014.EarlyTriassicseawatersulfatedrawdown.Geochim.Cosmochim.Acta128,95–113.

Sternberg,E.,Tang,D.,Ho,T.Y.,Jeandel,C.,Morel,F.M.M.,2005.Bariumuptakeandad-sorptionindiatoms.Geochim.Cosmochim.Acta69(11),2745–2752.

Sternberg,E.,Jeandel,C.,Miquel,J.C.,Gasser,B.,Souhaut,M.,Arraes-Mescoff,R.,Francois,

R.,2007.ParticulatebariumﬂuxesandexportproductioninthenorthwesternMed-iterranean.Mar.Chem.105(3),281–295.

Suess,E.,1980.Particulateorganiccarbonﬂuxintheoceans—surfaceproductivityand

oxygenutilization.Nature288,260–263.

Taylor,S.R.,McLennan,S.M.,1985.TheContinentalCrust:ItsCompositionandEvolution.

Blackwell,Oxford(312pp.).

Taylor,G.H.,Teichmüller,M.,Davis,C.(Eds.),1998.OrganicPetrology,16thed.Gebrüder

Borntraeger,Berlin704pp.

Thevenon,F.,Bard,E.,Williamson,D.,Beaufort,L.,2004.Abiomassburningrecordfrom

theWestEquatorialPaciﬁcoverthelast360ky:methodological,climaticandan-thropicimplications.Palaeogeogr.Palaeoclimatol.Palaeoecol.213(1),83–99.

Thomson,J.,Higgs,N.C.,Wilson,T.R.S.,Croudace,I.W.,DeLange,G.J.,Santvoort,P.J.M.,

1995.Redistributionandgeochemicalbehaviorofredox-sensitiveelementsaroundS1,themostrecenteasternMediterraneansapropel.Geochim.Cosmochim.Acta59,3487–3501.

Timothy,D.A.,Soon,M.,2001.Primaryproductionanddeep-wateroxygencontentoftwo

BritishColumbianfjords.Mar.Chem.73(1),37–51.

Tomas,C.R.(Ed.),1997.IdentifyingMarinePhytoplankton.AcademicPress,Elsevier,

Amsterdam(858pp.).

Torres,M.E.,Brumsack,H.J.,Bohrman,G.,Emeis,K.C.,1996.Baritefrontincontinental

marginsediments:anewlookatbariumremobilizationinthezoneofsulfatereduc-tionandformationofheavybaritesindiageneticfronts.Chem.Geol.127,125–139.Trask,P.D.,1953.ChemicalstudiesofsedimentsofthewesternGulfofMexico.Pap.Phys.

Oceanogr.Meteorol.12(4),49–120.

Tribovillard,N.,Riboulleau,A.,Lyons,T.,Baudin,F.,2004.Enhancedtrappingofmolybde-numbysulfurizedmarineorganicmatterofmarineorigininMesozoiclimestonesandshales.Chem.Geol.213,385–401.

Tribovillard,N.,Algeo,T.J.,Lyons,T.W.,Riboulleau,A.,2006.Tracemetalsaspaleoredox

andpaleoproductivityproxies:anupdate.Chem.Geol.232,12–32.

Tromp,T.K.,VanCappellen,P.,Key,R.M.,1995.Aglobalmodelfortheearlydiagenesisof

organiccarbonandorganicphosphorusinmarinesediments.Geochim.Cosmochim.Acta59,1259–1284.

Tyson,R.V.,1995.SedimentaryOrganicMatter:OrganicFaciesandPalynofacies.Chapman

andHall,London(615pp.).

Tyson,R.V.,2001.Sedimentationrate,dilution,preservation,andtotalorganiccarbon:

someresultsofamodelingstudy.Org.Geochem.32,333–339.

Tyson,R.V.,2005.The“productivityversuspreservation”controversy;cause,ﬂaws,and

resolution.In:Harris,N.B.(Ed.),DepositionofOrganic-carbon-richSediments:

52S.D.Schoepferetal./Earth-ScienceReviews149(2015)23–52

Models,Mechanisms,andConsequences.SocietyforSedimentaryGeology(SEPM-SSG)SpecialPublication82,pp.17–33.

VanBeek,P.,François,R.,Conte,M.,Reyss,J.L.,Souhaut,M.,Charette,M.,2007.228Ra/226Ra

and226Ra/Baratiostotrackbariteformationandtransportinthewatercolumn.Geochim.Cosmochim.Acta71(1),71–86.

VanCappellen,P.,Ingall,E.D.,1994.Benthicphosphorusregeneration,netprimarypro-duction,andoceananoxia:amodelofthecoupledmarinebiogeochemicalcyclesofcarbonandphosphorus.Paleoceanography9,677–692.

VanOs,B.J.H.,Middelburg,J.J.,DeLange,G.J.,1991.Possiblediageneticmobilizationof

bariuminsapropelicsedimentfromtheeasternMediterranean.Mar.Geol.100,125–136.

VanSantvoort,P.J.M.,DeLange,G.J.,Thomson,J.,Cussen,H.,Wilson,T.R.S.,Krom,M.D.,

Ströhle,K.,1996.Activepost-depositionaloxidationoftherecentsapropel(S1)insedimentsoftheeasternMediterraneanSea.Geochim.Cosmochim.Acta60,4007–4024.

Vink,S.,Chambers,R.M.,Smith,S.V.,1997.Distributionofphosphorusinsedimentsfrom

TomalesBay,California.Mar.Geol.139,157–179.

VonBreymann,M.,Emeis,K.-C.,Suess,E.,1992.Waterdepthanddiageneticconstraints

ontheuseofthebariumasapaleoproductivityindicator.In:Summerhayes,C.P.,Prell,W.L.,Emeis,K.C.(Eds.),UpwellingSystems:EvolutionSincetheEarlyMiocene.GeologicalSocietyofLondonSpecialPublication64,pp.273–284.

Wei,H.,Shen,J.,Schoepfer,S.D.,Krystyn,L.,Richoz,S.,Algeo,T.J.,2014.Environmental

controlsonmarineecosystemrecoveryfollowingmassextinctions,withanexamplefromtheEarlyTriassic.EarthScienceReviews,(inreview).

Weijers,J.W.,Schouten,S.,Schefuß,E.,Schneider,R.R.,SinningheDamsté,J.S.,2009.

Disentanglingmarine,soilandplantorganiccarboncontributionstocontinentalmar-ginsediments:amulti-proxyapproachina20,000yearsedimentrecordfromtheCongodeep-seafan.Geochim.Cosmochim.Acta73(1),119–132.

Weldeab,S.,Emeis,K.-C.,Hemleben,C.,Schmiedl,G.,Schulz,H.,2003.Spatialproductivity

variationsduringformationofsapropelsS5andS6intheMediterraneanSea:evidencefromBacontents.Palaeogeogr.Palaeoclimatol.Palaeoecol.191,169–190.

Westrich,J.T.,Berner,R.A.,1984.Theroleofsedimentaryorganicmatterinbacterial

sulfatereduction:theGmodeltested.Limnol.Oceanogr.29(2),236–249.

Wheatcroft,R.A.,1990.Preservationpotentialofsedimentaryeventlayers.Geology18,

843–845.

Wignall,P.B.,Newton,R.,2003.Contrastingdeep-waterrecordsfromtheUpperPermian

andLowerTriassicofSouthTibetandBritishColumbia:evidenceforadiachronousmassextinction.Palaios18,153–167.

Winckler,G.,Anderson,R.F.,Schlosser,P.,2005.EquatorialPaciﬁcproductivityanddust

ﬂuxduringthemid-Pleistoceneclimatetransition.Paleoceanography20(4)(10pp.).

Winguth,A.M.E.,Winguth,C.,2012.SimulatingPermian–Triassicoceanicanoxiadistribu-tion:Implicationsforspeciesextinctionandrecovery.Geology40,127–130.

Wortmann,U.G.,Paytan,A.,2012.Rapidvariabilityofseawaterchemistryoverthepast

130millionyears.Science337,334–336.

Yarincik,K.M.,Murray,R.W.,Lyons,T.W.,Peterson,L.C.,Haug,G.H.,2000.Oxygenationhis-toryofbottomwatersintheCariacoBasin,Venezuela,overthepast578,000years:resultsfromredox-sensitivemetals(Mo,V,Mn,andFe).Paleoceanography15(6),593–604.

Zachos,J.C.,Arthur,M.A.,Dean,W.E.,1989.Geochemicalevidenceforsuppressionof

pelagicmarineproductivityattheCretaceous/Tertiaryboundary.Nature337,61–64.Zarriess,M.,Mackensen,A.,2010.Thetropicalrainbeltandproductivitychangesoff

northwestAfrica:A31,000-yearhigh-resolutionrecord.Mar.Micropaleontol.76(3),76–91.

Zonneveld,K.A.F.,Versteegh,G.J.M.,Kasten,S.,Eglinton,T.I.,Emeis,K.C.,Huguet,C.,Koch,

B.P.,deLange,G.J.,Middelburg,J.J.,Mollenhauer,G.,Prahl,F.G.,Rethemeyer,J.,Wakeham,S.G.,2010.Selectivepreservationoforganicmatterinmarineenviron-ments;processesandimpactonthesedimentaryrecord.Biogeosciences7,483–511.

因篇幅问题不能全部显示，请点此查看更多更全内容

查看全文