Durham E-Theses

The Osmium Isotopic Composition of Seawater: Past

and Present

SPROSON, ADAM,DAVID

How to cite:

SPROSON, ADAM,DAVID (2017) The Osmium Isotopic Composition of Seawater: Past and Present,Durham theses, Durham University. Available at Durham E-Theses Online:http://etheses.dur.ac.uk/12262/

Use policy

The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission orcharge, for personal research or study, educational, or not-for-pro�t purposes provided that:

• a full bibliographic reference is made to the original source

• a link is made to the metadata record in Durham E-Theses

• the full-text is not changed in any way

The full-text must not be sold in any format or medium without the formal permission of the copyright holders.

Please consult the full Durham E-Theses policy for further details.

Academic Support O�ce, Durham University, University O�ce, Old Elvet, Durham DH1 3HPe-mail: e-theses.admin@dur.ac.uk Tel: +44 0191 334 6107

http://etheses.dur.ac.uk

2

1

TheOsmiumIsotopicCompositionofSeawater:PastandPresent

AdamD.Sproson

CollegeofSt.Hild&St.Bede

AthesissubmittedinpartialfulfilmentoftherequirementsforthedegreeofDoctorofPhilosophyatDurhamUniversity

DepartmentofEarthSciences,DurhamUniversity

June,2017

2

AbstractTheosmiumisotopiccompositionofseawater(187Os/188Os)reflectsabalancebetween

radiogeniccontinentalsourcesandunradiogenicmantleandextraterrestrialderived

sources.Reconstructionofthisvaluehasallowedustounlockvitalinformationabouta

seriesofEarthsystemprocesses,bothtodayandinEarth’sgeologicalpast.Thisbodyof

worklookstoreconstructthe187Os/188Osofseawaterforpastandpresentoceansusingthe

187Os/188Oscompositionofshalesandmacroalgae(seaweed)respectively.

The187Os/188OscompositionofIcelandic(0.16to0.99)andJapanese(0.16to1.09)

macroalgaearehighlyvariable,andreflectthemixingbetweenmultiplesources.The

187Os/188OsofIcelandiccoastalwatersisdominatedbyseawaterandlocalrivercatchments,

andhasbeenutilisedtotracetheinfluenceofbasalticweatheringontheglobalOscycle.

The187Os/188OsofJapanesecoastalwatersisdominatedbyseawaterandrivercatchments

drainingMiocene-Holocenecontinentalrocksoranthropogenicsources,andhasbeen

utilisedtotracemankind’simpactontheglobalOscycle.The187Os/188Osprofilesofshales

fromtheSilurianIreviken,Mulde,LauandKlonkbioventsaresimilartothosepreviously

recordedfortheHirnantianglaciation.ThisdatasuggeststheSilurianhasbeenpunctuated

byseveralglaciationsassociatedwithfluctuationsinglobaltemperatures,sea-levelandthe

carboncycle.WhencombinedwiththeLiisotopic(δ7Li)compositionofcarbonates,this

studysuggestsglacialprocessescausedlargechangesinoxidativeandsilicateweathering.

Thisstudyhassuccessfullyutilisedmacroalgaeasaproxyforthe187Os/188Osof

seawaterandprovenitcanbecomeapowerfultracerofEarthsystemprocessesand

humanactivity.ThisstudyhasalsoredefinedtheSilurianasanicehouse,andsuggeststhe

longtermdeclineinatmosphericCO2,duetoorogeny,land-plantdiversification,volcanic-

arcdegassingand/orpaleogeography,wasreversedbyperiodicglaciationswhichactedto

enhanceoxidativeweatheringwhilstsuppressingsilicateweathering.

3

Declaration

Ideclarethatthisthesis,whichIsubmitforthedegreeofDoctorofPhilosophyatDurham

University,ismyownworkandnotsubstantiallythesameasanywhichhaspreviously

beensubmittedatthisoranyotheruniversity.

AdamD.Sproson

CollegeofSt.Hild&St.Bede,DurhamUniversity

June2017

©Thecopyrightofthisthesisrestswiththeauthor.Noquotationfromitshould

bepublishedwithoutpriorwrittenconsentandinformationderivedfromit

shouldbeacknowledged.

4

Acknowledgements

IwouldliketodedicatethisPhDthesistomyparents,JulieandRichardSproson,for

withoutthem,noneofthiscouldbepossible.Theyhavesupportedmefinancially,

emotionallyandacademicallythroughoutmyentirelife.Theyhavealwayskeptanopen

mindandhaveneverpressuredmetotakemylifeinanyparticulardirection.Thishasled

metotheenvironmentalandpaleoclimaticsciencesIsomuchloveandtreasureandthe

startofacareerinsomethingIfeelintenselypassionateabout.ForthatIameternally

indebted.

IwouldliketothankmyPhDsupervisors,Prof.DavidSelby,Prof.KevinBurtonand

Prof.DavidHarperfortheirconstantadviceandsupportinthisendeavour.Inparticular,I

wouldliketothankProf.DavidSelbyforhisfinantialandacademicsupport.Withouthim

noneofthisresearchcouldhavebeenpossible.Healwaysallowedmetodevelopnew

ideasandcarryfurtherresearchifthemoodstruckme,withnosignoffinancialconstrait.

Thishasledtothedevelopmentofsomeuniqueresearchthathasgoneaboveandbeyond

theoriginalremitoftheproject.Inmyopinionheisanexceptionalsupervisorandthebest

supervisorIhavehad(sofar).

IwouldliketothankmyfriendsandcolleaguesinDurham.InparticularIwouldlike

tothankFienkeNanne,ThomasUnderwood,ElizabethAtar,EdwardInglisandBen

Maunder.Theyhavemadethelastfouryearssomeofthebestinmylife.Theyhavefilled

thedurationofmyPhDwithtrulyfunandenjoyableexperiencesandsupportedme

throughthehardtimes(ofwhichthereweremany).

IwouldliketothankJoannaHesselinkforhelpingmeinthelabduringthestartof

myPhDandteachingmethetricksofthetrade.IwouldalsoliketothankAntoniaHoffman

forherhelptowardstheendofmyPhDandmakingthingsrunsmoothly.ChrisOttleyand

GeoffNowellreceivemygratitudeforkeepingthemassspectrometersinoperation

5

thoughoutmyPhDandforhelpingwithanalyses.ThankyoutotherestoftheSelbyGroup

i.e.YangLi,JunjieLiuandZeyangLiuformakingthelabasaneplacetoworkandhelping

interpretmydata.

IwouldliketothankKatsuhikoSuzuki,RyokoSenda,MariaLuisaTejeda,Marc-

AlbanMilletandDieterKornfortheirassistancewithfieldworkduringmyPhD.Inparticular

IwouldliketothankDirector.KatzSuzukiforsupervisingmeduringmyresearchinJapan.I

wouldliketothankJindirchHladil,EmiliaJarochowska,JiriFryda,LadislavSlavikandDavid

Loydellfortheirtirelesseffortsinprovidingmewithenoughsamplestomakemystudyof

theSilurianpossible.TothatextentIwouldalsoliketothankPhilipPoggevonStrandmann

forhishelpandexpertiseinbothanalysingandinterpretinglithiumisotopedata.

Finally,IwouldliketothankTheJapaneseSocietyforthePromotionofScience,the

GeologicalSocietyofLondon,theCollegeofSt.Hild&St.Bede(JohnSimpsonGreenwell

memorialfund)andProf.DavidSelbyforhelpingtofundmyresearch.Withouttheir

money,noneofthisresearchwouldhavebeenpossible.

6

Contents

Abstract...................................................................................................................................2

Declaration..............................................................................................................................3

Acknowledgements.................................................................................................................4

Chapter1...............................................................................................................................10

1.1The187Os/188Osofcontemporaryseawater.................................................................13

1.1.1Osmiumisotopesinmacroalgae..........................................................................14

1.1.2OsmiumisotopesinIceland..................................................................................15

1.1.3OsmiumisotopesinJapan....................................................................................16

1.2The187Os/188OsofseawaterduringtheSilurian..........................................................18

1.2.1ClimaticchangeduringtheSilurian......................................................................18

1.2.2ApplicationoftheRe-OssystemtotheSilurian....................................................20

1.2.3ApplicationoftheLithiumisotopesystemtotheSilurian....................................21

1.3References...................................................................................................................22

Chapter2...............................................................................................................................28

2.1Introduction.................................................................................................................30

2.2Fieldandanalyticaltechniques...................................................................................33

2.2.1Samplingandstorage...........................................................................................34

2.2.2Macroalgaespeciesandhabitats.........................................................................35

2.2.3Re-Osanalysis.......................................................................................................36

2.2.3.1Macroalgae.......................................................................................................37

2.2.3.2Bedload.............................................................................................................37

2.2.3.3Seawater...........................................................................................................38

2.2.3.4MassSpectrometry............................................................................................38

2.2.4Statisticaltests.....................................................................................................39

2.3Results.........................................................................................................................40

2.3.1Macroalgae..........................................................................................................40

2.3.2Bedload................................................................................................................42

2.3.3Dissolvedload.......................................................................................................44

2.3.4Partitioncoefficient..............................................................................................46

2.4Discussion....................................................................................................................47

2.4.1BiologicalandenvironmentalcontrolsonReandOsuptakeinmacroalgae.......47

2.4.2Environmentalcontrolsonthe187Os/188Osofmacroalgae...................................52

7

2.4.2.1Influenceofestuarineconditionsonthe187Os/188Osofmacroalgae.................52

2.4.2.2Influenceofbasalticweatheringonthe187Os/188Osofmacroalgae..................54

2.4.3Biologicalandenvironmentalcontrolson187Re/188Osofmacroalgae..................57

2.5Implicationsandfutureoutlook..................................................................................60

2.6References...................................................................................................................62

Chapter3...............................................................................................................................66

3.1Introduction.................................................................................................................67

3.2Fieldandanalyticaltechniques...................................................................................71

3.2.1Samplingandstorage...........................................................................................71

3.2.2Macroalgaespeciesandhabitats.........................................................................73

3.2.3Re-Osanalysis.......................................................................................................73

3.2.3.1Macroalgae.......................................................................................................73

3.2.3.2MassSpectrometry............................................................................................74

3.3Results.........................................................................................................................74

3.3.1HokkaidoandNorthernHonshu...........................................................................76

3.3.2TokyoBay.............................................................................................................78

3.3.3OsakaBay.............................................................................................................79

3.3.4IseandMikawaBay.............................................................................................80

3.3.5IzuPeninsula.........................................................................................................82

3.3.6NotoPeninsula.....................................................................................................83

3.4Discussion....................................................................................................................85

3.4.1Biologicalandenvironmentalcontrolsonthe187Re/188Osofmacroalgae...........86

3.4.2Naturalsourcesofosmiumtomacroalgae..........................................................88

3.4.3Anthropogenicsourcesofosmiumtomacroalgae...............................................92

3.4.4Anthropogenicinfluenceontheglobalosmiumcycle..........................................99

3.4.4.1AnthropogenicimpactonJapanesecoastalwaters..........................................99

3.4.4.2AnthropogeniccontributionsofosmiumfromJapan......................................100

3.4.4.3Impactofanthropogenicosmiumonsurfacewaters......................................101

3.5Implicationsandfutureoutlook................................................................................102

3.6References.................................................................................................................104

Chapter4.............................................................................................................................111

4.1Introduction...............................................................................................................112

4.2Materialsandmethods.............................................................................................116

4.2.1Geologicalsetting...............................................................................................116

8

4.2.1.1Aizputecore.....................................................................................................117

4.2.1.2Lusklint&Lickershamn....................................................................................117

4.2.1.3Bartoszyce.......................................................................................................118

4.2.1.4Hunningecore.................................................................................................118

4.2.1.5Kosov...............................................................................................................118

4.2.1.6Klonkcore........................................................................................................119

4.2.2Samplepreparation............................................................................................119

4.2.3Osmiumisotopeanalysisofshales.....................................................................120

4.2.4Lithiumisotopeanalysisofbulkcarbonates.......................................................121

4.2.5Isotopemodeling................................................................................................123

4.2.5.1Osmiumisotopemodeling...............................................................................123

4.2.5.2Lithiumisotopemodeling................................................................................124

4.3Results.......................................................................................................................126

4.3.1Rhenium-osmiumisotopedata..........................................................................126

4.3.1.1Aizpute-41Core...............................................................................................129

4.3.1.2BartoszyceCore...............................................................................................130

4.3.1.3Kosovsection...................................................................................................131

4.3.1.4KlonkCore........................................................................................................133

4.3.2Lithiumisotopeandtracemetaldata................................................................134

4.3.2.1Lusklintsection................................................................................................135

4.3.2.2Lickershamnsection........................................................................................136

4.3.2.3Hunninge-1drillcore........................................................................................138

4.3.2.4Kosovsection...................................................................................................139

4.4Discussion..................................................................................................................139

4.4.1IsotopicconstraintsonSilurianseawaterchemistry..........................................141

4.4.1.1Climaticallyinducedchangesinoceancirculation..........................................141

4.4.1.2FloodBasaltVolcanism....................................................................................145

4.4.1.3Temperature-weatheringfeedbacks...............................................................146

4.4.1.4Hydrothermalactivity......................................................................................147

4.4.1.5Glaciation........................................................................................................149

4.4.2ScenariosfortriggeringSilurianGlaciations......................................................156

4.4.2.1Orogeny...........................................................................................................157

4.4.2.2Landplantdiversification................................................................................157

4.4.2.3Volcanicarcdegassing....................................................................................158

9

4.4.2.4Paleogeography..............................................................................................159

4.4.2.5Orbitalforcing.................................................................................................159

4.4.2.6Whydidwenotsee‘SnowballEarth’conditionsduringtheSilurian?.............160

4.4.3WeatheringfeedbackshelpregulateatmosphericCO2......................................161

4.5Implicationsandfutureoutlook................................................................................164

4.6References.................................................................................................................166

Chapter5.............................................................................................................................178

5.1Re-Osisotopeuptakeanddistributioninmacroalgae..............................................179

5.2Re-OsisotopesinmacroalgaeasatracerofEarthsystemprocesses.......................180

5.3Re-Osisotopesinmacroalgaeasatracerofanthropogenicprocesses....................181

5.4TheOsandLiisotopiccompositionofSilurianseawater..........................................182

5.6Futureoutlook...........................................................................................................183

5.7References.................................................................................................................185

10

Chapter1

Introduction

11

Temporalandspatialvariationsintheisotopiccompositionofseawaterreflecttheeffects

offluctuationinEarthsystemprocessesonoceanchemistry.Manyradiogenicisotope

systemsinseawateraresensitivetovariationsincontinentalweatheringanderosion,

makingthemapowerfularchiveforreconstructingresponsestoclimaticortectonic

change,withsilicateweatheringreceivingspecialattentiondueitsperceivedcontrolon

atmosphericCO2overgeologicaltimescales(Berneretal.,1983;Walkeretal.,1981).Of

theseradiogenicsystems,therubidium-strontium(87Rb-86Sr)radiogenicisotopesystemhas

beenthemostwidelyused,withvariationsinthemarine87Sr/86Srrecordreflecting

fluctuationsincontinentalinputscausedbyorogenesis(Raymoetal.,1988)andglaciations

(Armstrong,1971).However,asaconsequenceofthelongresidencetimeofSrinthe

oceans(2-4Myr)short-termfluctuationsininputsarehardtodetect(Richterand

Turekian,1993).

Theosmiumisotopiccompositionofseawater(187Os/188Os)reflectsabalance

betweenradiogeniccontinentalsourcesandunradiogenicmantleandextraterrestrial

derivedsources(Peucker-EhrenbrinkandRavizza,2000).Therefore,muchlikeSrisotopes,

Osisotopeshavebeenutilisedtoinferinformationaboutpastchangesincontinental

weathering(SeePeucker-EhrenbrinkandRavizza,2012).However,unlikeSr,theresidence

timeofOsintheocean(1-50kyr)issufficientlyshorttorespondtoshort-periodic

fluctuationsininput,whilststillbeinglongenoughtoattainaglobalsignal(Levasseuretal.,

1999;Oxburgh,2001;Rooneyetal.,2016).Therefore,the187Os/188Oscompositionof

seawaterofferstheabilitytodistinguishbetweenhigh-frequencyclimaticandlow

frequencytectonicforcing(Peucker-EhrenbrinkandRavizza,2000).Thishasallowedusto

unlockvitalinformationaboutaseriesofEarthsystemprocessesintheEarth’sgeological

pastsuchas:floodbasaltvolcanism(CohenandCoe,2002;DuVivieretal.,2014;Ravizza

andPeucker-Ehrenbrink,2003;TurgeonandCreaser,2008);paleoweathering(Finlayetal.,

2010;Ravizzaetal.,2001;Schmitzetal.,2004)basinconnectivity(PoirierandHillaire-

12

Marcel,2009);and,bolideimpacts(Paquayetal.,2008).However,despitethreedecades

ofwork,therestillremainsagreatdearthofdata187Os/188Osforpre-Cenozoictime(See

Peucker-EhrenbrinkandRavizza,2012).Inpart,thisstudylookstocorrectthisby

determiningthe187Os/188OsofseawaterfortheSilurian.

Inthemodernocean,the187Os/188Oscompositionofseawaterhasbeen

reasonablywellconstrainedthroughdirectanalysisusingultra-lowblanktechniques

capableofoxidisingallosmiumtoacommonoxidationstate(ChenandSharma,2009;

GannounandBurton,2014;Levasseuretal.,1998;Pauletal.,2009).Nevertheless,direct

analysisofseawaterremainsanalyticallychallengingduetothelowconcentrations

(Peucker-Ehrenbrinketal.,2013),andmeasurementsofrivers,estuariesandcoastal

watersarethereforesparse(Gannounetal.,2006;Huhetal.,2004;Sharmaetal.,2007;

SharmaandWasserburg,1997;Turekianetal.,2007).Thecompositionoftheglobal

riverineinputsthereforeremainspoorlyconstrained,raisingthepossibilitythatthe

osmiuminputintotheoceancouldbeunderestimatedbyafactorof~3(Oxburgh,2001).

Thismayhaveledtoadiscrepancybetweenoceanicosmiumresidencetimesestimated

frommassbalancecalculations(35-50kyr)andthoseinferredfromtheevolution(1-4

kyr)oftheosmiumisotoperecord(Levasseuretal.,1999;Oxburgh,2001;Rooneyetal.,

2016;Sharmaetal.,1999).

RecentworksuggestsmacroalgaeconcentratesOs(withabundancesthatvary

from12.6to78.5ppt),whilstmaintainingthe187Os/188Oscompositionoftheseawaterit

inhabits(Racionero-Gómezetal.,2016;Racionero-Gómezetal.,2017).Thissuggeststhat

macroalgaecouldactasaproxyforthe187Os/188Oscompositionoflocalwaterswhilst

removingsomeoftheanalyticalchallengesassociatedwithdirectanalysisofseawateri.e.

ultra-lowconcentrationsandmultipleoxidationstates.Macroalgaeexistingincoastal

13

waters,therefore,shouldrecordan187Os/188Ossignaturethatreflectsabalanceoflocal

inputs,includingriverineinput,localbedrock,anthropogenicactivityandseawater.

Inthisbodyofworkwewillapplythemacroalgae-Os-seawaterproxytoreal

worldsettingstotestitsabilitytorecordEarthsystemprocessesduringthepresent.In

Chapter2wewillutilisemacroalgaecollectedfromIcelandiccoastalwaterstotrace

fluctuationsinthe187Os/188OsoffreshwaterandseawateraroundIceland,anddetermine

theinfluenceofbasalticweatheringontheglobalosmiumbudget.InChapter3wewill

utilisemacroalgaefromJapanesecoastalwaterstohelpconstraintheanthropogenic

influenceontheglobalOscycle.Finally,inChapter4wewillutilisetheRe-Osisotope

systematicsofshalestodeterminefluctuationsinthe187Os/188Osofseawaterduringthe

Silurian,andshedlightonthemechanismsbehindabruptclimaticchangeduringthistime.

1.1The187Os/188Osofcontemporaryseawater

Osmiumisamongtheleastabundantelementsinseawater,thereforeearlyattemptsto

analysetheconcentrationandisotopiccompositionofseawaterdirectlywereplaguedwith

difficulties.Koideetal.(1996)useda25Lsampleofseawaterspikedwitha190Ostracer,

separatingOswithanion-exchangechromatographyandthempurifyingitusingdistillation

techniques.Sharmaetal.(1997)reducedseawaterandtracerOsbybubblingSO2(g)and

thenco-precipitatingOswithironoxyhydroxidein4-10Lsamples.Theproblemwiththese

techniquesisthattheyrequirehandlingalargevolumeofsample.Threesubsequent

techniquesattemptedtousesmallervolumesofsample(50gto1.5kg),andtriedto

equilibratetracerandwaterOsbyoxidisingittoacommonoxidationstate(OsO4).

However,ChenandSharma(2009)discoveredthateachofthesemethodsdidnotyield

identicalconcentrationstothoseofWoodhouseetal.(1999).Theyfoundhigher

temperatures(300°C)wererequiredtooxidiseallspeciesofOspresentinseawater.Witha

14

reliablechemicalseparationtechniqueinhandithasbecomepossibletomeasureOs,

althoughstillanalyticalchallenging,withaslittleas20gofseawater(Sharmaetal.,2012).

Despitethiscapability,problemsstillarisefromthenatureofOsinseawater.

ExtremelylowconcentrationsofOsinseawateri.e.90fgin100mlofseawater,meansyou

getsignificantinterferencefromproceduralblanksof~3.6fg(ChenandSharma,2009).This

iscompoundedbycontaminationfromtheuseofpolyethylenebottlesforseawater

storage(Sharmaetal,2012).WeproposetodevelopanewproxyfortheOsisotopic

compositionofseawaterbasedonOsmeasurementsofmacroalgae,whichdoesnotsuffer

fromtheseproblemsi.e.Osconcentrationsinseaweedare>50pg/gandthereforefar

higherthantheproceduralblank(~50fg),whilstthelongtermstorageofseaweeddoes

notsufferfromthestoragetechniquesused.

1.1.1Osmiumisotopesinmacroalgae

WorkconductedbyB.Racionero-GómezandmyselfattheUniversityofDurhamlooked

intothebiologicaluptakeofRe(Racionero-Gómezetal.,2016)andOs(RacioneroGomez

etal.,2017)intomacroalgae.Thesestudiesutilisedasinglemacroalgaespecies,Fucus

vesiculosus,andanalyseditsReandOsabundanceanduptake,aswellasassessingifit

couldrecordtheOsisotopecompositionoftheseawaterinwhichitlived.Itwas

demonstratedthatOsandRearenotlocatedinonespecificbiologicalstructure,butfound

throughouttheorganism.OsmiumuptakewasdeterminedbyculturingF.vesiculosuswith

differentconcentrationsofOswithaknown187Os/188Oscomposition(~0.16).Thecultured

samplestookontheisotopiccompositionofthecultureinwhichtheylived,whichis

significantlydifferenttothebackgroundcompositionofun-dopedseawater(~0.94)(Fig.1).

Thissuggestsmacroalgaecanattaintheisotopiccompositionofseawater,andtherefore

couldpotentiallyactasaproxyforunderstandingavarietyofEarthsystemprocesses.

15

Fig.1.Osmium(ppt)accumulation(circles)and187Os/188Oscompositions(squares)inF.

versiculosusunderdifferentculturemediaOsabundances.SeeRacionero-Gómezetal.

(2017)formoredetails.

1.1.2OsmiumisotopesinIceland

Icelandconsistsofanessentiallymonolithologicalbasalticterrainofvaryingages(historic

to12Ma),yieldingalargerangeinthe187Os/188Os(0.15to1.04)ofriverinedissolvedloads

(Gannounetal.,2006).Unradiogenic187Os/188Osvaluescanbeexplainedbycongruent

basaltweatheringand/orhydrothermalinput,withradiogenic187Os/188Osvaluesarising

fromtwodistinctprocesses:the187Os/188Osofglacier-fedriverscanbeexplainedbythe

entrainmentofseawateraerosolsintoprecipitationandsubsequentglacialmelting;while

the187Os/188Oscompositionofdirect-runoffandspring-fedriverscanbeexplainedbythe

incongruentweatheringofcertainprimarybasalticmineralsthatpossessexceptionallyhigh

16

187Re/188Os,whichovertimeevolvestoradiogenic187Os/188Osvalues(Gannounetal.,

2004).

Thechemistryofmacroalgaefromshallowcoastalwatersisdominatedbythe

brackishconditionsinwhichitlives.Macroalgaethereforecomeincontactwithbothfresh

waterandseawatersourcesthroughasingletidalcycle,andintheory,the187Os/188Os

compositionofmacroalgaeshouldrepresentamixingbetweenthe187Os/188Oscomposition

ofthesetwosources.InIceland,aspreviouslyexplained,the187Os/188Oscompositionof

riverineinputswillbehighlyvariable,rangingfromunradiogenicvaluesnearthecentralrift

zone,throughtohighlyradiogenicvaluesintheouterpartsofIceland.Thisdiverserangeof

187Os/188OscompositionsmakesIcelandauniqueenvironmentwithwhichtotesttheability

ofmacroalgaetorecordthe187Os/188Oscompositionoftheseawateritinhabits.

Chapter2presentsRe-Osabundanceandisotopedataformacroalgaeand

dissolvedandbedloadsfromcoastalwatersandriversdrainingbasalticwatershedsof

Iceland.Thisrepresentsthefirstexaminationoftheinfluenceofbothseawaterandriver

waterosmiumonthe187Os/188Oscompositionofmacroalgaeanddemonstratestheability

ofmacroalgaetotracefluctuationsinthe187Os/188Osoffreshwaterandseawateraround

Iceland.Chapter2utilisesthistodeterminetheinfluenceofbasalticweatheringonthe

globalosmiumbudget.

1.1.3OsmiumisotopesinJapan

Thepresent-dayopenoceanseawater187Os/188Osvalueof~1.06reflectsthebalance

betweenunradiogenicmantle-derivedOsandradiogeniccontinentalOs(Peucker-

Ehrenbrink,&Ravizza,2000).Dissolutionofbackgroundextraterrestrialmattercontributes

littletotheunradiogenicsourcesofOs.Ontheotherhand,Osreleasedbyanthropogenic

activitieshasbeendetectedincoastalsediments,lakesandestuariesfromsourcessuchas

sewagesludge,catalyticconvertors,anduseasastainingreagentinbiomedicalresearch

17

(EsserandTurekian,1993;Rauchetal.,2004;RavizzaandBothner,1996;Turekianetal.,

2007;Williamsetal.,1997).AtmosphericanthropogenicOsattributedtothesmeltingof

variousoresandcatalyticconvertors,hasalsobeendetectedinrainandsnow,whichis

impactingOsinoceanicsurfacewatersandthereforetheglobalOsbudget(Chenetal.,

2009).Theseanthropogenicsourcesaregenerallycharacterizedbyunradiogenic

187Os/188Oscompositions(~0.12)relatedtotheisotopiccompositionofPGEsrefinedfor

humanuse.

Japanoffersauniqueplaceinwhichtostudytheinfluenceofanthropogenic

processesonregionalvariationsinthemarineOscycle.Large,denselypopulated,

sprawlingmetropolitanareasoftenfallincloseproximitytocoastalwaters,inletseasor

bays.WatersintheseregionsarelikelytobedominatedbyunradiogenicOsproducedby

sewagetreatmentplants,hospitals,refineriesandvehicleexhaust.Moreover,these

metropolitanareasareoftenjuxtaposedtosparselypopulatedruraland/ormountainous

regions,withlittlehumanactivityandthereforeanthropogenicinfluence.The187Os/188Os

compositionofcoastalwatersintheseregionsisthereforelikelytobedominatedby

naturalsourcesofOs.Inparticular,Japaneserivercatchmentsaredominatedbythe

weatheringofMiocene-Holocenevolcanicandsedimentaryrocksofradiogenic187Os/188Os

compositions.Aspreviouslyexplained,the187Os/188Oscompositionofmacroalgaefrom

coastalwaterslikelyrepresentsamixingbetweenthe187Os/188Oscompositionofseawater

andlocalfreshwatersinputs.MacroalgaefromdenselypopulatedregionsofJapanwill

thereforelikelyshowastronginfluencefromhighlyunradiogenicPGE187Os/188Osvalues

derivedfromlocalhumanactivity.Meanwhile,the187Os/188Oscompositionofmacroalgae

frommoresparselypopulatedregionswillshowastrongerinfluencefromnatural

freshwatersources.

18

Chapter3presentsRe-Osabundanceandisotopedataformacroalgaefrom

TokyoBay,OsakaBay,IseBay,MikawaBay,IzuPeninsula,NotoPeninsula,Hokkaidoand

northernHonshu.TokyoBay,OsakaBayandIse/MikawaBaylieincloseproximitytothe

denselypopulatedKanto(Tokyo-Yokohama),Keihanshin(Osaka-Kobe)andChukyo

(Nagoya)metropolitanareas.The187Os/188Ossignaturefromtheseregionsismuchlower

thanexpectedfornaturalriverandoceanicsystems,butsimilartotheisotopecomposition

ofPGEores.ThissuggeststhathumanactivityhasinfluencedtheOsisotopiccomposition

ofmacroalgae,andthereforeseawater,throughtheburningofmunicipaland/orhospital

waste,processingofsewageandtheextensiveuseofautomobilesintheseareas.TheIzu

Peninsula,NotoPeninsula,HokkaidoandnorthernHonshuexhibit187Os/188Osvaluessimilar

toglobalriverwaterorPacificseawatermeasurements,suggestinglittleinfluencefrom

humanactivity.Theseresultsdemonstratetheabilityofmacroalgaetotracefluctuationsin

the187Os/188OsoffreshwaterandseawateraroundJapan,andutilisethistodeterminethe

influenceofhumanactivityontheglobalosmiumbudget.

1.2The187Os/188OsofseawaterduringtheSilurian

1.2.1ClimaticchangeduringtheSilurian

IncontrasttopreviousviewsthattheSilurianwasanenvironmentallystableperiod

betweentheOrdovicianicehouseandtheDevoniangreenhouse(Munneckeetal.,2010),it

hasmorerecentlybecomeapparentthattheSilurianischaracterisedbyahighlydynamic

climateattenuatedbymultipleshort-livedevents,strongeustaticsealevelchangeand

oceanicturnoverassociatedwithextinctioneventsafterarecoveryfromtheend-

Ordovicianglaciation(Calner,2008;Melchinetal.,2012).Thecarbonisotopiccomposition

ofcarbonate(δ13Ccarb)throughtheSilurianishighlyvariable,indicatingtheclimatesystem

andcarboncyclewereprobablymoreunstablethananyotherPhanerozoicperiodandcan

19

beregardedasoneofthemostvolatileperiodswhenconsideringtheocean-atmosphere

system(CramerandSaltzman,2005).

TheSilurianthroughtotheearlyDevonianismarkedbyfourlarge-amplitude

positivecarbonandoxygenisotopeexcursions,withtheδ13Ccarbexceeding+5‰duringthe

Ireviken,Mulde,LauandKlonkbioevents.TheLaucarbonisotopeexcursion(>+8‰)is

probablythelargestpostCambrianδ13CcarbexcursionoftheentirePhanerozoic.Such

excursionsarefarlargerthananythingintheMesozoicorCenozoic,andthereforeany

classicalinterpretationsconcerningproductivitychangesarenotviable(Bickertetal.,

1997).Theδ13Ccarbisotopeexcursionsareassociatedwithsignificantpositiveoxygen

isotope(δ18O)excursionsofapproximately1-3.5‰magnitude(Lehnertetal.,2010;

Munneckeetal.,2010;Žigaitėetal.,2010),whicharetoolargetobeexplainedbyeither

temperature,icevolumeorsalinityalone(Bickertetal.,1997).However,despitethese

discoveriesandovertwodecadesofresearch,thecauseoftheseclimateperturbationsis

stillnotunderstood.

Traditionalexplanationsfortheseeventshaveinvokedashiftbetweentwo

stableoceanic-climatestates,drivenbychangesinthelocationofdeep-waterformation

fromhightolowlatitudes(Jeppsson,1990),orglobalprecipitationratesandcontinental

runoff(Bickertetal.,1997).However,theseearlyattemptstoexplainSilurianclimatic

eventshavereceivedmuchcriticism(Johnson,2006;Kaljoetal.,2003;Loydell,1998).More

recentlyithasbeenpostulatedthatSilurianclimaticchangecouldhavebeendrivenby

glacialexpansionoverGondwana,inferredinpartfrompositiveoxygenisotopeshifts

(Trotteretal.,2016)coupledtosignificanteustaticsea-levelchange(Díaz-Martínezand

Grahn,2007;Lehnertetal.,2010),muchliketheLateOrdovicianthatprecededit(Algeoet

al.,2016;Harperetal.,2014).However,thelackofglacialsedimentsinthestratigraphic

20

recordformuchoftheSilurian(post-Wenlock)hashamperedthisnotion(Caputo,1998;

Díaz-MartínezandGrahn,2007;GrahnandCaputo,1992).

1.2.2ApplicationoftheRe-OssystemtotheSilurian

Toshedlightonthepossiblemechanismsbehindtheseabruptclimateperturbations,we

haveappliedtheRe-Osisotopesystemtoorganicrichshalesfromgeologicalformations

thatspantheIreviken,Mulde,LauandKlonkbioevents.The187Os/188Os

oforganicrichshalesmimicsthatofseawateratthetimeofdeposition,andreflectsa

balancebetweenradiogeniccontinentalsourcesandunradiogenicmantleand

extraterrestrialderivedsources(Peucker-EhrenbrinkandRavizza,2000).Therefore,much

likeSrisotopes,Osisotopeshavebeenutilisedtoinferinformationaboutpastchangesin

continentalweathering.However,unlikeSr,theresidencetimeofOsintheoceanis

sufficientlyshorttorespondtoshort-periodicfluctuationsininput,whilststillbeinglong

enoughtoattainaglobalsignal(Levasseuretal,1998;Oxburgh,2001).Therefore,theOs

isotopesystemofferstheabilitytodistinguishbetweenhigh-frequencyclimaticandlow

frequencytectonicforcing(Peucker-EhrenbrinkandRavizza,2000).

InChapter4osmiumisotopeprofilesfortheIreviken,Mulde,LauandKlonk

bioeventsarepresented.TheseprofilesaresimilartotheHirnantianglaciation(Finlayet

al.,2010),whichoccurredsome10MyrpriortotheIrevikenEvent.The187Os/188Osprofiles

arerelatedtotheweatheringoforganic-richsedimentaryrocksduringtheexpansionof

continentaliceoverGondwana,whichdeliversradiogenicOstotheocean.Thiswork

suggestsclimaticperturbationsduringtheSilurianarerelatedtoglaciationstriggeredby

decliningatmosphericCO2andtemperatures.Oncetheglaciationwastriggered,enhanced

oxidativeweatheringoforganicandsulphiderichsedimentaryrocksledtoareleaseinCO2

totheatmospherewhichhelpedreverseglobalcooling.

21

1.2.3ApplicationoftheLithiumisotopesystemtotheSilurian

Aspreviouslyexplained,variationsinOsisotopescanalsobecontrolledbyfactorsother

thancontinentalweathering,suchasfluctuationsinhydrothermalalterationofjuvenile

basalticcrustandextra-terrestrialinputs(Peucker-EhrenbrinkandRavizza,2000).Totest

thevalidityofourOsisotopecurvesastracersofcontinentalweathering,wealsoutilised

othercontinentalweatheringproxies.TheLiisotopiccompositionofseawater(δ7Li)

reflectsabalancebetweentheinputsofLifromrivers(weatheringofcontinentalsilicate

rocks)andmid-oceanridgespreadingcentres(weatheringofoceanicsilicaterocks)andthe

outputsofLifromtheincorporationintomarinesedimentsandalteredoceaniccrust

(MisraandFroelich,2012).Theδ7Liofcarbonatesisseentoreflecttheδ7Liofseawater

throughtime,andhasbeenusedtoreconstructcontinentalsilicateweatheringratesinthe

geologicalpast(MisraandFroelich,2012).Recently,PoggevonStrandmannetal.(in

review)utilisedLiisotopemeasurementsincarbonatesandshalesthatspantheHirnantian

glaciation.Apeakinδ7LioccursduringthetroughinOsisotopes(Finlayetal.,2010)

associatedwithglacialmaximum.Thisshowsastrongdeclineinsilicateweathering,

reducingtheeffectoftheEarth’sprimaryCO2removalmechanism,preventingfurther

coolingandallowingtemperaturestobuild-up,leadingtodeglaciation.

Totestthehypothesisinsection1.2.2wegeneratedLiisotoperecordsin

carbonatesfromsectionsspanningtheIreviken,Mulde,LauandKlonkbioevents(See

Chapter4).LithiumisotoperecordsshowasimilarprofiletotheHirnantianglaciation

suggestingsuppressionincontinentalsilicateweatheringunderenhancedcontinentalice

sheets.ThisreducesoneoftheEarth’smajorCO2withdrawalmechanisms,andwhen

combinedwithenhancedoxidativeweathering,leadstogreaterlevelsofatmosphericCO2.

ThisstudysuggeststhattheEarthhasastabilisingnegativefeedbackmechanisminwhicha

reductioninatmosphericCO2andglobaltemperatureswhichleadstoenhanced

22

continentaliceisultimatelyreversedbyweatheringprocesseswhichacttoincrease

atmosphericCO2andtemperatures,preventingarunawayicehouse.

1.3References

Algeo,T.J.,Marenco,P.J.,Saltzman,M.R.,2016.Co-evolutionofoceans,climate,andthe

biosphereduringthe‘OrdovicianRevolution’:Areview.Palaeogeography,

Palaeoclimatology,Palaeoecology458,1-11.

Armstrong,R.L.,1971.Glacialerosionandthevariableisotopiccompositionofstrontiumin

seawater.Nature230,132-133.

Berner,R.A.,Lasaga,A.C.,Garrels,R.M.,1983.Thecarbonate-silicategeochemicalcycleand

itseffectonatmosphericcarbondioxideoverthepast100millionyears.AmJSci283,641-

683.

Bickert,T.,Pätzold,J.,Samtleben,C.,Munnecke,A.,1997.Paleoenvironmentalchangesin

theSilurianindicatedbystableisotopesinbrachiopodshellsfromGotland,Sweden.

GeochimicaetCosmochimicaActa61,2717-2730.

Calner,M.,2008.Silurianglobalevents–atthetippingpointofclimatechange,Mass

Extinction.SpringerBerlinHeidelberg,pp.21-57.

Caputo,M.V.,1998.Ordovician–Silurianglaciationsandglobalsea-levelchanges.Silurian

cycles,15-25.

Chen,C.,Sedwick,P.N.,Sharma,M.,2009.Anthropogenicosmiuminrainandsnowreveals

global-scaleatmosphericcontamination.ProceedingsoftheNationalAcademyofSciences

oftheUnitedStatesofAmerica106,7724-7728.

Chen,C.,Sharma,M.,2009.Highprecisionandhighsensitivitymeasurementsofosmiumin

seawater.Analyticalchemistry81,5400-5406.

Cohen,A.S.,Coe,A.L.,2002.Newgeochemicalevidencefortheonsetofvolcanisminthe

CentralAtlanticmagmaticprovinceandenvironmentalchangeattheTriassic-Jurassic

boundary.Geology30,267.

Cramer,B.D.,Saltzman,M.R.,2005.Sequestrationof12Cinthedeepoceanduringthe

earlyWenlock(Silurian)positivecarbonisotopeexcursion.Palaeogeography,

Palaeoclimatology,Palaeoecology219,333-349.

23

Díaz-Martínez,E.,Grahn,Y.,2007.EarlySilurianglaciationalongthewesternmarginof

Gondwana(Peru,BoliviaandnorthernArgentina):Palaeogeographicandgeodynamic

setting.Palaeogeography,Palaeoclimatology,Palaeoecology245,62-81.

DuVivier,A.D.C.,Selby,D.,Sageman,B.B.,Jarvis,I.,Gröcke,D.R.,Voigt,S.,2014.Marine

187Os/188Osisotopestratigraphyrevealstheinteractionofvolcanismandocean

circulationduringOceanicAnoxicEvent2.EarthandPlanetaryScienceLetters389,23-33.

Esser,B.K.,Turekian,K.K.,1993.Anthropogenicosmiumincoastaldeposits.Environmental

science&technology27,2719-2724.

Finlay,A.J.,Selby,D.,Gröcke,D.R.,2010.TrackingtheHirnantianglaciationusingOs

isotopes.EarthandPlanetaryScienceLetters293,339-348.

Gannoun,A.,Burton,K.,2014.Highprecisionosmiumelementalandisotope

measurementsofNorthAtlanticseawater.JAAS.

Gannoun,A.,Burton,K.W.,Thomas,L.E.,Parkinson,I.J.,vanCalsteren,P.,Schiano,P.,2004.

Osmiumisotopeheterogeneityintheconstituentphasesofmid-oceanridgebasalts.

Science303,70-72.

Gannoun,A.,Burton,K.W.,Vigier,N.,Gíslason,S.R.,Rogers,N.,Mokadem,F.,Sigfússon,B.,

2006.Theinfluenceofweatheringprocessonriverineosmiumisotopesinabasaltic

terrain.EarthandPlanetaryScienceLetters243,732-748.

Grahn,Y.,Caputo,M.V.,1992.EarlySilurianglaciationsinBrazil.Palaeogeography,

Palaeoclimatology,Palaeoecology99,9-15.

Harper,D.A.T.,Hammarlund,E.U.,Rasmussen,C.M.Ø.,2014.EndOrdovicianextinctions:A

coincidenceofcauses.GondwanaResearch25,1294-1307.

Huh,Y.,Birck,J.-L.,Allègre,C.J.,2004.OsmiumisotopegeochemistryintheMackenzie

Riverbasin.EarthandPlanetaryScienceLetters222,115-129.

Jeppsson,L.,1990.Anoceanicmodelforlithologicalandfaunalchangestestedonthe

Silurianrecord.JournaloftheGeologicalSociety147,663-674.

Johnson,M.E.,2006.RelationshipofSiluriansea-levelfluctuationstooceanicepisodesand

events.GFF128,115-121.

24

Kaljo,D.,Martma,T.,Männik,P.,Viira,V.,2003.ImplicationsofGondwanaglaciationsin

theBalticlateOrdovicianandSilurianandacarbonisotopictestofenvironmentalcyclicity.

BulletindelaSociétégéologiquedeFrance174,59-66.

Koide,M.,Goldberg,E.D.,Walker,R.,1996.Theanalysisofseawaterosmium.DeepSea

ResearchPartII:TopicalStudiesinOceanography43,53-55.

Lehnert,O.,Männik,P.,Joachimski,M.M.,Calner,M.,Frýda,J.,2010.Palaeoclimate

perturbationsbeforetheSheinwoodianglaciation:Atriggerforextinctionsduringthe

‘IrevikenEvent’.Palaeogeography,Palaeoclimatology,Palaeoecology296,320-331.

Levasseur,S.,Birck,J.-L.,Allegre,C.,1999.Theosmiumriverinefluxandtheoceanicmass

balanceofosmium.EarthandPlanetaryScienceLetters174,7-23.

Levasseur,S.,Birck,J.-L.,Allègre,C.J.,1998.Directmeasurementoffemtomolesofosmium

andthe187Os/186Osratioinseawater.Science282,272-274.

Loydell,D.K.,1998.EarlySiluriansea-levelchanges.GeologicalMagazine135,447-471.

Melchin,M.J.,Sadler,P.M.,Cramer,B.D.,Cooper,R.A.,Gradstein,F.M.,Hammer,O.,2012.

Chapter21-TheSilurianPeriod,in:Gradstein,F.M.,Ogg,J.G.,Schmitz,M.D.,Ogg,G.M.

(Eds.),TheGeologicTimeScale.Elsevier,Boston,pp.525-558.

Misra,S.,Froelich,P.N.,2012.LithiumisotopehistoryofCenozoicseawater:changesin

silicateweatheringandreverseweathering.Science335,818-823.

Munnecke,A.,Calner,M.,Harper,D.A.T.,Servais,T.,2010.OrdovicianandSiluriansea–

waterchemistry,sealevel,andclimate:Asynopsis.Palaeogeography,Palaeoclimatology,

Palaeoecology296,389-413.

Oxburgh,R.,2001.Residencetimeofosmiumintheoceans.Geochemistry,Geophysics,

Geosystems2,1018.

Paquay,F.S.,Ravizza,G.E.,Dalai,T.K.,Peucker-Ehrenbrink,B.,2008.Determiningchondritic

impactorsizefromthemarineosmiumisotoperecord.Science320,214-218.

Paul,M.,Reisberg,L.,Vigier,N.,2009.Anewmethodforanalysisofosmiumisotopesand

concentrationsinsurfaceandsubsurfacewatersamples.ChemicalGeology258,136-144.

Peucker-Ehrenbrink,B.,Ravizza,G.,2000.Themarineosmiumisotoperecord.TerraNova

12,205-219.

25

Peucker-Ehrenbrink,B.,Ravizza,G.,2012.OsmiumIsotopeStratigraphy.145-166.

Peucker-Ehrenbrink,B.,Sharma,M.,Reisberg,L.,2013.Recommendationsforanalysisof

dissolvedosmiuminseawater.Eos,TransactionsAmericanGeophysicalUnion94,73-73.

PoggevonStrandmann,P.A.E.,Desrochers,A.,Murphy,M.J.,Finlay,A.J.,Selby,D.,Lenton,

T.M.,inreview.GlobalclimatestabilisationbychemicalweatheringduringtheHirnantian

glaciation.GeochemicalPerspectivesLetters.

Poirier,A.,Hillaire-Marcel,C.,2009.Os-isotopeinsightsintomajorenvironmentalchanges

oftheArcticOceanduringtheCenozoic.GeophysicalResearchLetters36.

Racionero-Gómez,B.,Sproson,A.,Selby,D.,Gröcke,D.,Redden,H.,Greenwell,H.,2016.

Rheniumuptakeanddistributioninphaeophyceaemacroalgae,Fucusvesiculosus.Royal

SocietyOpenScience3,160161.

Racionero-Gómez,B.,Sproson,A.D.,Selby,D.,Gannoun,A.,Gröcke,D.R.,Greenwell,H.C.,

Burton,K.W.,2017.Osmiumuptake,distribution,and187Os/188Osand187Re/188Os

compositionsinPhaeophyceaemacroalgae,Fucusvesiculosus:Implicationsfordetermining

the187Os/188Oscompositionofseawater.GeochimicaetCosmochimicaActa199,48-57.

Rauch,S.,Hemond,H.F.,Peucker-Ehrenbrink,B.,2004.Recentchangesinplatinumgroup

elementconcentrationsandosmiumisotopiccompositioninsedimentsfromanurbanlake.

Environmentalscience&technology38,396-402.

Ravizza,G.,Bothner,M.H.,1996.Osmiumisotopesandsilverastracersofanthropogenic

metalsinsedimentsfromMassachusettsandCapeCodbays.GeochimicaetCosmochimica

Acta60,2753-2763.

Ravizza,G.,Norris,R.N.,Blusztajn,J.,Aubry,M.P.,2001.Anosmiumisotopeexcursion

associatedwiththeLatePaleocenethermalmaximum:Evidenceofintensifiedchemical

weathering.Paleoceanography16,155-163.

Ravizza,G.,Peucker-Ehrenbrink,B.,2003.ChemostratigraphicevidenceofDeccan

volcanismfromthemarineosmiumisotoperecord.Science302,1392-1395.

Raymo,M.E.,Ruddiman,W.F.,Froelich,P.N.,1988.InfluenceoflateCenozoicmountain

buildingonoceangeochemicalcycles.Geology16,649-653.

26

Richter,F.M.,Turekian,K.K.,1993.Simplemodelsforthegeochemicalresponseofthe

oceantoclimaticandtectonicforcing.EarthandPlanetaryScienceLetters119,121-131.

Rooney,A.D.,Selby,D.,Lloyd,J.M.,Roberts,D.H.,Lückge,A.,Sageman,B.B.,Prouty,N.G.,

2016.Trackingmillennial-scaleHoloceneglacialadvanceandretreatusingosmium

isotopes:InsightsfromtheGreenlandicesheet.QuaternaryScienceReviews138,49-61.

Schmitz,B.,Peucker-Ehrenbrink,B.,Heilmann-Clausen,C.,Åberg,G.,Asaro,F.,Lee,C.-T.A.,

2004.Basalticexplosivevolcanism,butnocometimpact,atthePaleocene–Eocene

boundary:high-resolutionchemicalandisotopicrecordsfromEgypt,SpainandDenmark.

EarthandPlanetaryScienceLetters225,1-17.

Sharma,M.,Balakrishna,K.,Hofmann,A.W.,Shankar,R.,2007.ThetransportofOsmium

andStrontiumisotopesthroughatropicalestuary.GeochimicaetCosmochimicaActa71,

4856-4867.

Sharma,M.,Chen,C.,Blazina,T.,2012.Osmiumcontaminationofseawatersamplesstored

inpolyethylenebottles.LimnologyandOceanography:Methods10,618-630.

Sharma,M.,Papanastassiou,D.,Wasserburg,G.,1997.Theconcentrationandisotopic

compositionofosmiumintheoceans.GeochimicaetCosmochimicaActa61,3287-3299.

Sharma,M.,Wasserburg,G.,1997.Osmiumintherivers.Geochimicaetcosmochimicaacta

61,5411-5416.

Sharma,M.,Wasserburg,G.,Hofmann,A.,Chakrapani,G.,1999.Himalayanupliftand

osmiumisotopesinoceansandrivers.GeochimicaetCosmochimicaActa63,4005-4012.

Trotter,J.A.,Williams,I.S.,Barnes,C.R.,Männik,P.,Simpson,A.,2016.Newconodontδ18O

recordsofSilurianclimatechange:Implicationsforenvironmentalandbiologicalevents.

Palaeogeography,Palaeoclimatology,Palaeoecology443,34-48.

Turekian,K.K.,Sharma,M.,Gordon,G.W.,2007.Thebehaviorofnaturalandanthropogenic

osmiumintheHudsonRiver–LongIslandSoundestuarinesystem.Geochimicaet

CosmochimicaActa71,4135-4140.

Turgeon,S.C.,Creaser,R.A.,2008.Cretaceousoceanicanoxicevent2triggeredbya

massivemagmaticepisode.Nature454,323-326.

27

Walker,J.C.G.,Hays,P.B.,Kasting,J.F.,1981.Anegativefeedbackmechanismforthelong-

termstabilizationofEarth'ssurfacetemperature.JournalofGeophysicalResearch86,

9776.

Williams,G.,Marcantonio,F.,Turekian,K.K.,1997.Thebehaviorofnaturaland

anthropogenicosmiuminLongIslandSound,anurbanestuaryintheeasternUS.Earthand

planetaryscienceletters148,341-347.

Woodhouse,O.,Ravizza,G.,Falkner,K.K.,Statham,P.,Peucker-Ehrenbrink,B.,1999.

Osmiuminseawater:verticalprofilesofconcentrationandisotopiccompositioninthe

easternPacificOcean.EarthandPlanetaryScienceLetters173,223-233.

Žigaitė,Ž.,Joachimski,M.M.,Lehnert,O.,Brazauskas,A.,2010.δ18Ocompositionof

conodontapatiteindicatesclimaticcoolingduringtheMiddlePridoli.Palaeogeography,

Palaeoclimatology,Palaeoecology294,242-247.

28

Chapter2

Tracingtheinfluenceofweatheringprocessesoncoastalwaterssurroundinga

basalticterrainusingosmiumisotopesinmacroalgae*

*AversionofthischapterwillbesubmittedtoGlobalBiogeochemicalCycles,co-authored

withDavidSelby,AbdelmouchineGannoun,KevinW.BurtonandJeremyM.Lloyd.

29

Thisstudypresentsrhenium(Re)andosmium(Os)abundanceandisotopedatafor

macroalgae,dissolvedloadandbedloadfromtheIcelandiccoastalwaters,anenvironment

adjacenttopredominantlybasalticterrains,ranginginagefromhistorictoca.12Ma.Both

theReandOsabundanceinmacroalgaeareshowntobeprimarilycontrolledbyuptake

fromthedissolvedloadoflocalseawater.Inahabitatwithvaryingsalinitythroughthetidal

cycle,macroalgaeReandOsabundancesvary,dependingontherelativeinfluenceoflocal

freshwaterinputs.IncorporationofReandOsintomacroalgaeiscomplicatedbyadditional

Reuptakefromsuspendedparticulatesand/orbedload,whichisnotobservedforOs,

suggestingdifferentuptakepathwaysforbothReandOs.The187Os/188Os(0.16to0.99)and

187Re/188Os(~65to40,320)compositionsofmacroalgaearehighlyvariable,andcanbe

explainedintermsofanunradiogenic187Os/188Oscontributionwithlow187Re/188Osfrom

riversdrainingyoungercatchmentsthathaveundergonecongruentbasaltweathering

(and/orhydrothermalinput),andaradiogenic187Os/188Oscontributionfromtwodistinct

sources:riversdrainingoldercatchmentsthathaveundergoneincongruentweatheringof

primarybasalticmineralsthatpossessexceptionallyhigh187Re/188Osratiosthathave

evolvedtoradiogenic187Os/188Osratios;and,NorthAtlanticseawater.Macroalgaecan

attaina187Re/188OsfarhigherthanthatrecordedfortheIcelandicgeochemicalreservoirs

duetothepreferentialuptakeofReoverOsathighReconcentrationsinthedissolved

load,thereforemacroalgaecannotbeusedtodeterminethe187Re/188Osofseawater.This

studyconfirmstheutilityofmacroalgaeasaproxyfortheOsisotopiccompositionof

seawater,whichholdsthepotentialtoelucidatearangeofEarthsystemprocesses.

However,itisnotyetpossibletodirectlyrelatethemacroalgaeOsconcentrationtothatof

thewaterinwhichtheylive.Finally,theseresultssuggestthatmacroalgaeisnota

substantialsinkforeitherReorOs.Therefore,globalmacroalgaebiomass,todayorduring

theEarth’sgeologicalpast,doesnotplayasignificantroleinthemarineOsandRecycles.

30

2.1Introduction

Temperatureandatmosphericcarbondioxide(CO2)havefluctuatedwidelythroughoutthe

Phanerozoic,fromwarmgreenhousetocoldicehouseconditions.Nevertheless,

throughoutthistime,temperatureandCO2havealwaysremainedwithinthenarrowlimits

thatallowlifetoexistandevolvethroughtheinteractionbetweentheatmosphere,

hydrosphere,biosphereandlithosphere(Berneretal.,1983).Overgeologicaltimescales

(Myr)temperaturehaspartlybeencontrolledbyinteractionsbetweenatmosphericCO2

andcontinentalweathering.Risingtemperaturesstimulateincreasedchemicalweathering

ofsilicaterocksdrawingdownCO2fromtheatmosphere,leadingtoadeclinein

temperature(Berneretal.,1983;Walkeretal.,1981).Manyradiogenicisotopesystemsin

seawateraresensitivetovariationsincontinentalweatheringanderosion,makingocean

chemistryapowerfularchiveforreconstructingresponsestoclimaticortectonicchange.

Ofthese,therubidium-strontium(87Rb-86Sr)radiogenicisotopesystemhasbeen

themostwidelyused,withvariationsinthemarine87Sr/86Srrecordreflectingfluctuations

incontinentalinputscausedbyorogenesis(Raymoetal.,1988)andglaciations(Armstrong,

1971).However,asaconsequenceofthelongresidencetimeofSrintheoceans(2-4Myr)

short-termfluctuationsininputsarehardtodetect(RichterandTurekian,1993).The

osmiumisotopiccompositionofseawater(187Os/188Os)reflectsabalancebetween

radiogeniccontinentalsourcesandunradiogenicmantleandextraterrestrialderived

sources(Peucker-EhrenbrinkandRavizza,2000).Therefore,muchlikeSrisotopes,Os

isotopeshavebeenutilisedtoinferinformationaboutpastchangesincontinental

weathering(SeePeucker-EhrenbrinkandRavizza,2012andreferencestherein).However,

unlikeSr,theresidencetimeofOsintheocean(1-50kyr)issufficientlyshorttorespondto

short-periodicfluctuationsininput,whilststillbeinglongenoughtoattainaglobalsignal

(Levasseuretal.,1998;Oxburgh,2001;Rooneyetal.,2016).Therefore,the187Os/188Os

31

compositionofferstheabilitytodistinguishbetweenhigh-frequencyclimaticandlow

frequencytectonicforcing(Peucker-EhrenbrinkandRavizza,2000).

Inthemodernocean,the187Os/188Oscompositionofseawaterhasbeen

reasonablywellconstrainedthroughdirectanalysisusingultra-lowblanktechniques

capableofoxidisingallosmiumtoacommonoxidationstate(ChenandSharma,2009;

GannounandBurton,2014;Levasseuretal.,1998;Pauletal.,2009).Nevertheless,direct

analysisofseawaterremainsanalyticallychallengingbecauseofthelowconcentrations

andmultipleoxidationstates(Peucker-Ehrenbrinketal.,2013),andmeasurementsof

rivers,estuariesandcoastalwatersarethereforesparse(Gannounetal.,2006;Huhetal.,

2004;Sharmaetal.,2007;SharmaandWasserburg,1997;Turekianetal.,2007).Therefore

thecompositionoftheglobalriverineinputremainspoorlyconstrained,raisingthe

possibilitythattheosmiuminputtotheoceanisunderestimated,therebyaccountingfor

thediscrepancybetweenoceanicosmiumresidencetimesestimatedfrommassbalance

calculations(35-50kyr)andthoseinferredfromtheevolutionoftheosmiumisotope

record(1-4kyr)(Levasseuretal.,1999;Oxburgh,2001;Rooneyetal.,2016;Sharmaetal.,

1999).Althoughitisalsoproposedthatthedifferenceinresidencetimeestimatesrelates

tothelocalremovalofOs(Sharmaetal.,2007).

Recentworkindicatesthatmacroalgae(seaweed)concentratesOs(with

abundancesthatvaryfrom12.6to78.5ppt),whilstmaintainingthe187Os/188Os

compositionoftheseawateritinhabits(Racionero-Gómezetal.,2017;Rooneyetal.,

2016).Thissuggeststhatmacroalgaecouldactasaproxyforthe187Os/188Oscompositionof

localwaterswhilstremovingsomeoftheanalyticalchallengesassociatedwithdirect

analysisofseawateri.e.ultra-lowconcentrationsandmultipleoxidationstates.Macroalgae

existingincoastalwaters,therefore,shouldrecordan187Os/188Ossignaturethatreflectsa

balanceoflocalinputs,includingriverineinput,localbedrockandseawater.

32

Icelandconsistsofanessentiallymonolithologicalbasalticterrainofvaryingages

(historicto12Ma),yieldingalargerangeinthe187Os/188Os(0.15to1.04)ofriverine

dissolvedloads(Gannounetal.,2006).Unradiogenic187Os/188Osvaluescanbeexplainedby

congruentbasaltweatheringand/orhydrothermalinput,withradiogenic187Os/188Osvalues

arisingfromtwodistinctprocesses.The187Os/188Osofglacier-fedriverscanbeexplainedby

theentrainmentofseawateraerosolsintoprecipitationandsubsequentglacialmelting.

The187Os/188Oscompositionofdirect-runoffandspring-fedriverscanbeexplainedbythe

incongruentweatheringofcertainprimarybasalticmineralsthatpossessexceptionallyhigh

187Re/188Os,whichovertimeevolvestoradiogenic187Os/188Osvalues(Gannounetal.,2004).

Icelandthereforeprovidesauniqueenvironmentwithrespectto187Os/188Osinwhichto

testtheabilityofmacroalgaetorecordthe187Os/188Oscompositionoftheseawaterit

inhabits.

ThisstudypresentsRe-Osabundanceandisotopedataformacroalgaeand

dissolvedandbedloadsfromcoastalwatersandriversdrainingbasalticwatershedsof

Iceland.GiventhatwatershedsofIcelandareessentiallymonolithological,their187Os/188Os

compositionisdeterminedbytheageofbasalt,thepreferentialweatheringofconstituent

basaltmineralsandtheentrainmentofrain,seaandhydrothermalwaters(Gannounetal.,

2006).Thisstudyrepresentsthefirstexaminationoftheinfluenceofbothseawaterand

riverwaterosmiumonthe187Os/188Oscompositionofmacroalgae.Theseresults

demonstratetheabilityofmacroalgaetotracefluctuationsinthe187Os/188Osoffreshwater

andseawateraroundIceland,andutilisethistodeterminetheinfluenceofbasaltic

weatheringontheglobalosmiumbudget.

33

2.2Fieldandanalyticaltechniques

34

Fig.1.GeologicalmapofIcelandmodifiedafterJóhannesson(2014):1=Holocene

sediments;2=basicandintermediatelavas(postglacial,historic,youngerthanAD871);3=

basicandintermediatelavas(postglacial,prehistoric,olderthanAD871);4=acidlavas

(postglacial,historic,youngerthanAD871);5=acidlavas(postglacial,prehistoric,older

thanAD871);6=acidextrusives(Miocene,PlioceneandPleistocene,olderthan11,000

yrs);7=basicandintermediatehyaloclastite,pillowlavaandassociatedsediments(Upper

Pleistocene,youngerthan0.8myr);8=basicandintermediateinterglacialandsupraglacial

lavaswithintercalatedsediments(UpperPleistocene,youngerthan0.8myr);9=basicand

intermediateextrusiverockswithintercalatedsediments(UpperPlioceneandLower

Pleistocene,0.8-3.3myr);10=basicandintermediateextrusiverockswithintercalated

sediments(MioceneandLowerPliocene,olderthan3.3myr);11=basicandintermediate

intrusions,gabbro,doleriteanddiorite;12=acidintrusions,rhyolite,granophyreand

granite.Sampletypeandlocalitiesareindicatedinthelegend.Dashedlinesrepresentan

outlineofareasshowninFigures2aand2b.

2.2.1Samplingandstorage

Macroalgae,bedloadsandsandwatersfromtheIcelandiccoastlineweresampledat

eighteenlocationsduringlateAugustof2014(Fig.1;samples7-23and27).Afurthernine

locationsweresampledbetweenlateJulyandearlyAugustof2015(Fig.1;samples1-6and

24-26).Intotaltwenty-sevenmacroalgae,elevenbedloadandeightwatersampleswere

collected(Fig.1;Table1-2).Macroalgaeandbedloadwerewashedusingdeionised(Milli-

Q™)watertoremoveanyattachedsedimentandsalt.Theywerethendriedfor12hat60

°Candstoredinplasticzip-lockbags.Macroalgaeandbedloadwerelatercrushedusingan

agatepestleandmortarpriortoanalysis.Watersampleswerefilteredthrough0.2μm

celluloseacetatefiltersusingapressurizedSartorius®Teflonunit.Filtratealiquotswere

storedinpre-cleanedSavillex®TeflonbottlestopreventOscontamination(Sharmaetal.,

2012).SalinitywasmeasuredusingaHanna®HI98192conductivitymeter.

35

2.2.2Macroalgaespeciesandhabitats

Specificmacroalgaespecieswerenottargetedduringthisstudy,andsampleswere

selectedbasedontheiravailabilityatsamplesites.Therewashoweverapreferenceto

brownmacroalgaeovergreenandredmacroalgaeduetotheirrelativelyhighabundancein

Re(Masetal.,2005;Proutyetal.,2014;Racionero-Gómezetal.,2016;Yang,1991)andOs

(Racionero-Gómezetal.,2017;Rooneyetal.,2016).Fourspeciesofbrownmacroalgae

(Fucusvesiculosus,Fucusspiralis,FucusdistichusandAscophyllumnodosum)were

analysed.

Fig.2.GeologicalmapsoftheReykjanesPeninsula(a)andEasternFjords(b).Keyisthe

sameasinFigure1.

a

36

2.2.3Re-Osanalysis

TheRe-OsanalysisformacroalgaeandbedloadwerecarriedoutintheDurham

GeochemistryCentre(LaboratoryforSulfideandSourceRockGeochronologyand

Geochemistry).TheseawaterOsanalyseswereconductedatLaboratoireMagmaset

37

VolcansattheCampusUniversitairedesCézeaux,withtheRefractionprocessedatthe

DurhamGeochemistryCentre.

2.2.3.1Macroalgae

ThetechniqueforchemicalseparationofReandOsfrommacroalgaeisreportedby

Racionero-Gómezetal.(2017).Inbrief,approximately200mgofpowderedmacroalgae

wasintroducedintoaCariustubetogetherwith11NHCl(3mL),15.5NHNO3(6mL)anda

knownamountof185Re+190Ostracersolutionandheatedto220°Cinanovenfor24h.The

OswasisolatedfromtheacidmediumusingCHCl3solventextractionandthenback

extractedintoHBr.TheOswasfurtherpurifiedusingaCrO3-H2SO4–HBrmicro-distillation

(Bircketal.,1997;CohenandWaters,1996).TheremainingRe-bearingacidmediumwas

evaporatedtodrynessat80°C,withtheReisolatedandpurifiedusingbothNaOH-acetone

solventextractionandHNO3-HClanionchromatography(Cummingetal.,2013).

2.2.3.2Bedload

ThedetailedanalyticalprocedureforsilicateshasbeenadaptedfromIshikawaetal.(2014).

Approximately1gofbedloadwascrushedusinganagatemortar.Thepowderisdissolved

withHCl+HF(4mL:2mL)ina22ml-savillex®vialat100°C.Theacid-samplemediumwas

evaporatedtodrynessat80°Cbefore11NHCl(1mL)wasaddedandsubsequently

evaporatedtwicetoremoveremainingHF.Theresultingacidmediumwasintroducedinto

aCariustubetogetherwith11NHCl(3mL),15.5NHNO3(6mL)andaknownamountof

the185Re+190Ostracersolutionandheatedat220°Cfor48h.Theextractionand

purificationoftheOsandRefractionsfollowsthesameanalyticalprotocoloutlinedin

section2.3.1.

38

2.2.3.3Seawater

ThetechniqueforchemicalseparationofReandOsfromseawaterisreportedby

Racionero-Gómezetal.(2017).Briefly,~60gofwatersample,plusaknownamountof

mixed(190Os+185Re)tracersolution,togetherwith2mLofBr2,2mLofCrO3-H2SO4solution

and1.5mLof98%H2SO4weresealedintoa120mLsavillexvialandheatedto100˚Cinan

ovenfor72htoequilibratesampleandspike(GannounandBurton,2014).TheOswas

extractedfromthesampleintoliquidBr2followedbyasecondextractionofOsusing1mL

ofBr2.The1mLofliquidBr2wasaddedtothesamplesolution,reactedfor1h,andthen

removed.TheextractedBr2ismixedwith1mLof9NHBrandevaporatedtodryness,and

furtherpurifiedusingaCrO3-H2SO4–HBrmicro-distillation.TheRewaspurifiedasoutlined

forthemacroalgaesamples(Cummingetal.,2013).

2.2.3.4MassSpectrometry

ThepurifiedReandOsfractionswereloadedontoNiandPtfilamentsrespectively,and

measuredusingNTIMS(Creaseretal.,1991;Völkeningetal.,1991)onaThermoScientific

TRITONmassspectrometerusingFaradaycollectorsinstaticmode,andanelectron

multiplierindynamicmoderespectively.TheReandOsabundancesandisotope

compositionsarepresentedwith2s.e.(standarderror)absoluteuncertaintieswhich

includefullerrorpropagationofuncertaintiesinthemassspectrometermeasurements,

blank,spikecalibrations,andsampleandspikeweights.Fullanalyticalblankvaluesforthe

macroalgaeanalysisare10.9±5.9pgforRe,0.13±0.13pgforOs,witha187Os/188Os

compositionof0.61±0.34(1SD,n=4).Forthebedloadanalysisthefullanalyticalblank

valuesare15.9±0.23pgforRe,2.12±0.01pgforOs,witha187Os/188Oscompositionof

0.27±0.001(2s.e.,n=1).Fortheseawateranalysisthefullanalyticalblankvaluesare10±

39

1.3pgforRe,0.043±0.002pgforOs,witha187Os/188Oscompositionof0.72±0.02(1SD,n

=4).

Tomonitorthelong-termreproducibilityofmassspectrometermeasurements

ReandOs(DROsS,DTM)referencesolutionswereanalysed.The125pgResolutionyields

anaverage185Re/187Reratioof0.5987±0.0023(2SD.,n=8),whichisinagreementwith

publishedvalues(e.g.,Cummingetal.,2013andreferencestherein).A50pgDROsS

solutiongavean187Os/188Osratioof0.16111±0.0008(2SD.,n=8),whichisinagreement

withreportedvaluefortheDROsSreferencesolution(Nowelletal.,2008).Forthe

seawaterOsanalysisattheLaboratoireMagmasetVolcansmonitorinstrument

reproducibilityusinga1pgDTMOssolution,whichyieldsa187Os/188Osvalueof0.1740±

0.0002(2SD,n=4),whichisinagreementwithpublishedvalues(ChenandSharma,2009;

GannounandBurton,2014).

2.2.4Statisticaltests

StatisticaltestswerecarriedoutusingMATLAB®R2017a(MATLAB9.2,TheMathWorks

Inc.,Natick,MA,2017).Aone-wayanalysisofvariance(ANOVA)wasutilisedtodetermine

whetherseveralgroupsofafactorhaveacommonmean.ANOVAtestsfordifferences

betweengroupmeansbypartitioningvariationinthedataintotwocomponents:variation

ofindividualobservationsfromtheirgroupmean;and,variationofthegroupmeansfrom

theoverallmean(WuandHamada,2000;Neteretal,1996).IfduringtheANOVAtest,the

p-valueoftheF-statistic(ratioofthemeansquares)issmallerthanthesignificancelevel

(0.05),thetestrejectsthenullhypothesis,thatallgroupmeansareequal,andoneofthe

groupmeansisthereforedifferentfromtheothers.WheretheANOVAtestisusedthe

meansofeachgroup(avg),theF-statistic(F)andthep-valuewillbecited.Intheeventthe

nullhypothesisisrejectedamultiplecomparisontestiscarriedout.Thiswilldetermine

whichpairsofmeansaresignificantlydifferentfromeachother.

40

2.3Results

2.3.1Macroalgae

Table1

RheniumandosmiumabundanceandisotopedataforIcelandicmacroalgaesamples

TheReandOsabundanceandisotopedataformacroalgaearegiveninTable1.Rhenium

andOsabundancesshowalargerangefrom0.1to88.4ppbandfrom3.3to254.5ppt

respectively.Individualspecies,suchasFucusvesiculosus,Fucusspiralis,Fucusdistichus

andAscophyllumnodosumshowvariableReabundancesfrom3.6to71.9ppb,14.5to28.8

ppb,0.10to1.6ppband3.2to88.4ppb,respectively(Fig.3a).Individualspecies,suchas

Fucusvesiculosus,Fucusspiralis,FucusdistichusandAscophyllumnodosumalsoshowa

largerangeinOsabundancesfrom5.0to254.5ppt,3.3to14.5ppt,4.7to7.6pptand9.5

to53.0ppt,respectively(Fig.3a).The187Os/188Oscompositionsofthemacroalgaerange

1 Fucusvesiculosus 44.32 2.53 27.85 0.10 7879.7 452.6 0.337 0.0032 Fucusspiralis 14.52 0.57 8.81 0.07 8534.2 342.4 0.700 0.0203 Fucusvesiculosus 38.04 1.50 5.03 0.04 39405.6 1646.4 0.750 0.0304 Fucusvesiculosus 53.37 2.10 23.42 0.10 11864.8 471.8 0.750 0.0105 Fucusvesiculosus 40.52 2.32 44.92 0.16 4483.7 257.4 0.368 0.0036 Fucusvesiculosus 23.56 0.93 106.42 0.42 1087.5 43.5 0.270 0.0107 Ascophyllumnodosum 3.17 0.01 53.04 0.41 307.1 4.3 0.634 0.0128 Fucusvesiculosus 71.92 0.23 10.14 0.04 37751.4 211.2 0.926 0.0069 Fucusvesiculosus 15.39 0.05 6.93 0.06 11692.8 180.0 0.835 0.01810 Fucusvesiculosus 61.68 0.20 13.71 0.05 23395.3 120.6 0.736 0.00411 Fucusvesiculosus 25.66 0.54 7.87 0.05 17023.4 381.1 0.765 0.00912 Fucusvesiculosus 3.58 0.01 13.73 0.36 1323.9 77.5 0.535 0.04413 Fucusvesiculosus 8.86 0.19 10.59 0.12 4358.6 130.2 0.750 0.02214 Fucusdistichus 0.11 0.03 4.69 0.10 123.5 34.6 0.970 0.05815 Fucusvesiculosus 14.65 0.05 9.21 0.09 8533.3 130.4 0.995 0.02116 Fucusvesiculosus 50.36 0.13 8.34 0.04 32096.9 237.1 0.913 0.00917 Fucusvesiculosus 43.93 0.11 7.72 0.04 30419.2 226.2 0.967 0.00918 Fucusdistichus 1.55 0.01 7.41 0.05 1107.6 10.8 0.879 0.01119 Fucusspiralis 28.83 0.08 14.45 0.08 10523.5 80.0 0.854 0.00920 Fucusdistichus 0.10 0.03 7.56 0.15 64.6 20.5 0.612 0.03621 Fucusspiralis 24.96 0.08 3.30 0.03 40320.4 683.7 0.938 0.02322 Ascophyllumnodosum 4.23 0.01 21.98 0.44 1022.7 41.8 0.925 0.05323 Fucusvesiculosus 60.27 0.15 8.24 0.06 38953.6 311.5 0.930 0.01424 Ascophyllumnodosum 88.43 0.29 13.83 0.28 34155.5 1422.1 0.962 0.05625 Fucusspiralis 18.15 0.06 6.45 0.04 14725.3 135.5 0.787 0.01026 Ascophyllumnodosum 63.96 3.65 9.45 0.04 35167.1 2017.9 0.727 0.00627 Fucusvesiculosus 9.58 0.55 254.47 1.69 182.3 10.7 0.161 0.003

2s.e. 2s.e. 2s.e.187Re/188Os 187Os/188Os2s.e.[Os](ppt)

[Re](ppb)MacroalgaeSpecies

SampleLocation

41

from0.16to0.99(Fig.3b).Thereason(s)forthevariabilityisdiscussedbelow.The

187Re/188Osratiosofthemacroalgaearehighlyvariable(~65to40,320;Fig.3b).

Fig.3.Rhenium(opendiamonds)andosmium(filledsquares)abundance(a)andisotopic

composition(b)forF.vesiculosus(green),F.spiralis(orange),F.distichus(purple)andA.

nodosum(blue).

Aone-wayANOVAtestwasconductedtodetermineiftherewasanystatistically

significantvariationinReabundance,Osabundance,187Re/188Osand187Os/188Osbetween

eachmacroalgaespecies.WhencomparingFucusvesiculosus,Fucusspiralis,Fucusdistichus

andAscophyllumnodosum,theANOVAtestsforReabundance(avg=35.4,21.6,0.6,39.9;

42

F=2.29;p-value=0.1054),Osabundance(avg=34.9,8.3,6.5,24.6;F=0.45;p-value=

0.7178),187Re/188Os(avg=16904,18525,431,17663;F=1.18;p-value=0.3385)and

187Os/188Os(avg=0.68,0.82,0.82,0.82;F=0.68;p-value=0.5744)allhaveap-values

whichfallabovethesignificancelevel(p-value=0.05).Wecanthereforeacceptthenull

hypothesisthateachspecieshasacommonmeanforeachparameterstudied.

2.3.2Bedload

Table2

RheniumandosmiumabundanceandisotopedataforIcelandicbedloads

SampleLoacation

[Re](ppb) 2s.e.

[Os](ppt) 2s.e. 187Re/188Os 2s.e. 187Os/188Os 2s.e.

7 0.73 0.02 41.21 0.18 85.6 2.1 0.157 0.002

8 0.91 0.03 7.23 0.04 646.3 23.0 0.591 0.007

9 0.89 0.02 15.60 0.07 277.8 6.6 0.231 0.003

10 1.11 0.04 9.13 0.04 593.03 20.98 0.228 0.003

11 0.89 0.03 8.81 0.04 494.3 17.6 0.250 0.003

12 0.55 0.02 24.99 0.12 108.02 3.97 0.337 0.004

15 0.69 0.02 23.71 0.11 143.2 5.2 0.292 0.004

18 0.51 0.02 10.42 0.05 238.6 8.8 0.224 0.003

19 1.03 0.04 12.52 0.06 405.5 14.4 0.271 0.003

20 0.61 0.02 11.41 0.05 259.96 9.45 0.220 0.003

27 0.63 0.02 69.05 0.31 44.2 1.6 0.204 0.003

TheReandOsabundanceandisotopicdataofbedloadsarepresentedinTable2.Rhenium

andOsabundancesrangefrom0.5to1.1ppband7.3to69.1pptrespectively.This

compareswelltoReandOsabundancespreviouslyrecordedinIcelandicbasalts,which

rangefrom0.05to1.8ppband3.7to1954.9pptrespectively(Debailleetal.,2009;

Gannounetal.,2006).MacroalgaeandbedloadretainsimilarlevelsofOswiththe

exceptionofthemacroalgaesamplefromlocation27(Fig.1),whichisheavilyenrichedin

Os(~254ppt;Fig.4a).MacroalgaeisenrichedinRebyseveralordersofmagnitudewhen

comparedtobedload.Asignificantcorrelation(R2=0.6061)inReabundanceisobserved

43

betweenmacroalgaeandthecorrespondingbedload(Fig.4b).The187Os/188Oscomposition

ofthebedloadrangesfrom0.16to0.59,withanaverageof0.27,whichisclosertothe

unradiogenicend-member.Thebedload187Os/188Osvaluesaremoreradiogenicthanthose

previouslyrecordedforIcelandicbasalts(Gannounetal.,2006),althoughdatainthisstudy

isfromtheolderextremitiesofIceland(Figs.1-2),wheremoreradiogenicingrowthof

187Oswillhaveoccurred.Inalmostallcases,the187Os/188Osratiosofmacroalgaeare

significantlymoreradiogenicthanthecorrespondingbedload(Fig.5a).The187Re/188Os

ratioofthebedloadrangesfrom44.2to646.3,whichfallclosetothelowerendofthe

rangereportedforIcelandicbasalts(45-1698;Gannounetal.,2006).Althoughthe

macroalgaepossessmuchgreater187Re/188Osthanthecorrespondingbedload,astrong

correlation(R2=0.7962)betweenthemacroalgaeandbedload187Re/188Osisobserved(Fig.

5b).

Fig.4.Bedloadosmium(a)andrhenium(b)abundanceanddissolvedloadosmium(c)and

rhenium(d)abundanceversusthecorrespondingabundanceinmacroalgae.Seetextfor

discussion.

44

2.3.3Dissolvedload

Table3

RheniumandosmiumabundanceandisotopedataforIcelandicdissolvedloads

SampleLoacation

[Re](ppt) 2s.e.

[Os](ppq) 2s.e. 187Re/188Os 2s.e. 187Os/188Os 2s.e.

Salinity(ppt) ±

7 1.9 0.2 69.2 0.3 130.99 11.07 0.160 0.002 0.140 0.001

8 10.0 0.7 9.7 0.1 5467.3 375.4 0.876 0.023 33.580 0.336

9 8.3 0.6 18.9 0.2 2229.6 152.8 0.569 0.017 29.680 0.297

11 4.5 0.3 11.2 0.1 2097.98 147.97 0.693 0.014 30.080 0.301

15 2.1 0.2 8.5 0.1 1184.03 105.84 0.230 0.009 0.220 0.002

19 19.4 1.3 750.7 3.0 125.4 8.4 0.168 0.002 23.780 0.238

20 0.6 0.1 8.1 0.1 365.7 60.8 0.198 0.007 0.000 0.001

27 3.8 0.3 30.1 0.2 613.3 43.9 0.237 0.006 9.170 0.092

TheReandOsabundanceandisotopicdataoffilteredwatersamplesarereportedinTable

3.RheniumandOsabundancesrangefrom0.6to19.4pptand8.1to750.7ppq

respectively.ThesevaluesarehighlyvariablewhencomparedtooceanicRe(~8.2ppt)

(Anbaretal.,1992;Colodneretal.,1995;Colodneretal.,1993b)andOs(~10ppq)

(GannounandBurton,2014;Levasseuretal.,1998;Sharmaetal.,1997;Woodhouseetal.,

1999)concentration.However,theycomparewellwithReandOsconcentrationsfoundin

globalriverestimateswhichrangefrom24ppqto2.3ppband4.6to52.1ppqrespectively

(Colodneretal.,1993a;Levasseuretal.,1999;Milleretal.,2011;SharmaandWasserburg,

1997),withtheexceptionofsample19(750.7ppq),whichhasanOsconcentrationseveral

ordersofmagnitudehigherthanforanywatereverrecorded.Withtheexceptionof

sample19,OsconcentrationsaresimilartopreviouslyrecordedIcelandicrivers(1.0-20.5

ppq;Gannounetal.,2006).MacroalgaearehighlyenrichedinReandOswhencompared

tothedissolvedloadoftheriversmeasuredhere(Figs.4c,d),inkeepingwithprevious

observationswherebymacroalgaeisseentotakeupandconcentratetheseelementsfrom

seawater(Racionero-Gómezetal.,2016;Racionero-Gómezetal.,2017).

45

Fig.5.Bedload187Os/188Os(a)and187Re/188Os(b)anddissolvedload187Os/188Os(c)and187Re/188Os(d)versusthecorrespondingratioinmacroalgae.Seetextfordiscussion.

The187Os/188Osratioofthedissolvedloadrangesfrom0.16to0.88.Thesevalues

arelessradiogenicthanopenocean187Os/188Osvalues(~1.06)(GannounandBurton,2014;

Levasseuretal.,1998;Sharmaetal.,1997;Woodhouseetal.,1999)fallingclosetothe

rangeofIcelandicrivers(0.15-1.04)(Gannounetal.,2006).Inalmostallcasesthe

187Os/188Osratiosofmacroalgaearesignificantlymoreradiogenicthanthecorresponding

dissolvedload(Fig.5c).Themostradiogenicvalueswereobtainedformacroalgaecloseto

theoldestbasaltintheeasternandnorth-westernpartsofIceland,whilethemost

unradiogenicisotoperatioswerefoundeitherclosetothecentralactivezoneortherivers

drainingit(Table1;Fig.1).The187Re/188Osratioofthedissolvedloadrangesfrom125.4to

5467.3.Theloweranduppervaluesofthisrangearesimilartoaverageglobalriverine(227)

andseawater(4270)187Re/188Osratios,respectively(Peucker-EhrenbrinkandRavizza,2000)

suggestingtheinfluenceofbothriverwaterandseawateratthesesamplinglocations.This

46

isconfirmedbysalinitydata,whichshowsaweakcorrelation(R2=0.469)with187Re/188Os

andgenerallyhigherratiosathighersalinityandviceversa(Table3).Themacroalgae

187Re/188Osaregenerallyseveraltimesgreaterthanthecorrespondingdissolvedload(Fig.

5d).Astrongcorrelation(R2=0.9088)in187Re/188Osvaluesisobservedbetween

macroalgaeandthedissolvedload(Fig.5d).

2.3.4Partitioncoefficient

Thepartitioncoefficient(Kd)forOsandReis:

Kd=X s

[X]l

where[X]sisthetotalelementalabundanceformacroalgae,[X]listheelemental

abundanceinthedissolvedloadofseawater,andXreferstotheelementinquestion.The

Kdvaluesforsamplelocationswherebothmacroalgaeandseawatermeasurementsare

recordedintable4.

Table4

PartitioncoefficientsbetweenmacroalgaeandseawaterforReandOsabundance.

SampleLocation ReKd OsKd

7 1693 7678 7200 10499 1860 36611 5644 70215 7113 108419 1484 1920 156 93227 2536 8448

47

2.4Discussion

2.4.1BiologicalandenvironmentalcontrolsonReandOsuptakeinmacroalgae

RheniumandOsabundanceinmacroalgaeareconsistentwithpreviousstudies,showing

variationsbothbetweenspeciesandwithineachspeciesitself(Fig.3).Rhenium

abundancesinFucusvesiculosusfromsamplelocation3,8,10and24(Fig.1)comparewell

withvaluesof51.0to103.4ppbintheUK(Racionero-Gómezetal.,2016)and60.8to84.9

ppbinNorway(Masetal.,2005).However,mostFucusvesiculosusinthisstudyhavelower

concentrationsthanthosefromtheliterature,rangingfrom3.6to50.4ppb.Likewise,

FucusdisichusandAscophyllumnodosumshowalowerabundanceofRethanrecorded

valuesforCalifornia(Yang,1991)andGreenland(Rooneyetal.,2016),withtheexception

ofsample21.PreviousOsabundancedeterminationsforFucusvesiculosusfromtheUK

(33.8ppt;Racionero-Gómezetal.,2017)andAscophyllumnodosumfromGreenland(12.6

ppt;Rooneyetal.,2016)fallwithintherangesfoundinthisstudy.However,Osabundance

forFucusdistichusrecordedinGreenland(14.0ppt;Rooneyetal.,2016)isfarhigherthan

therangeforIcelandicFucusdistichusofthisstudy.

Ithaspreviouslybeensuggestedthattheageofthemacroalgae-duetoitsrapid

accumulationrates(Yang,1991)-andgeographicaldistribution-duetorelatively

ubiquitousoceanicReconcentrations(Masetal.,2005)-donotplayapartincontrolling

theReconcentrationinmacroalgae.However,seasonalvariations,thechemicalspeciesof

theReperrhenatecompound,fertilityandgrowth-mediaReconcentrationhaveallbeen

suggestedaspossiblecausesofRevariationinmacroalgae(Masetal.,2005;Racionero-

Gómezetal.,2016).Likewise,duetofastOsuptake(Racionero-Gómezetal.,2017)and

relativelyubiquitousoceanicOsconcentrations(Peucker-EhrenbrinkandRavizza,2000),it

isconceivablethatseasonalvariationsandgrowth-mediaOsconcentrationcouldcausethe

variationsinOsseeninmacroalgae(Racionero-Gómezetal.,2017).

48

Allsampleswerecollectedduringthesameseason(July-August)suggesting

seasonalitydoesplayanimportantroleinthevariationsinReandOsabundanceobserved

inthisstudy.AthighconcentrationsithasbeenfoundthatthespeciesofResaltdoesnot

greatlyaffectthelevelsofReincorporatedintomacroalgaeandbyhomogenisingthe

sample,aswasdoneinthisstudy,theinfluenceofhighRenon-fertiletipscanbemitigated

(Racionero-Gómezetal.,2016).However,culturingofmacroalgaeunderincreasing

seawaterReandOsconcentrationshasbeenshowntocauseanincreaseinuptakeinRe

(Racionero-Gómezetal.,2016)andOs(Racionero-Gómezetal.,2017)inmacroalgae.

AlthoughoceanicRe(8.2ppt)andOs(~10ppq)concentrationisseentobefairly

ubiquitous(Anbaretal.,1992;Colodneretal.,1993b;Levasseuretal.,1998;Sharmaetal.,

1997;Woodhouseetal.,1999),riverineRe(3.1ppt)andOs(9.1ppq)abundanceis

generallymuchlower,butalsohighlyvariable(Colodneretal.,1993a;Levasseuretal.,

1999;Milleretal.,2011;SharmaandWasserburg,1997).Therefore,thegeographical

distributioncouldplayanimportantroleifmacroalgaelivingonthecontinentalshelfinan

estuarinehabitatareincloseproximitytoafreshwatersource.Underestuarineconditions,

amixtureoffreshwaterandseawatercouldcausehighlyvariableReandOsabundances

throughatidalcycle,leadingtovariableabundancesinmacroalgae.

MacroalgaeinthisstudywerecollectedfromcoastalwaterssurroundingIceland,

generallyclosetothemouthsofmajorrivers(Figs.1-2).Previously,ithasbeenshownthat

ReconcentrationactsconservativelyintheAmazonestuarywithlowconcentrations(0.21

ppt)atlowsalinityandhighconcentrations(8.6ppt)athighsalinity(Colodneretal.,

1993a).Osmiumontheotherhand,hasbeenshowntobehavenon-conservativelyin

estuaries,experiencingremovalfromthewatercolumnatlowsalinitiesintemperateand

arcticestuaries(Levasseuretal.,2000;Turekianetal.,2007)andathighsalinitiesin

tropicalestuaries(Martinetal.,2001;Sharmaetal.,2007),suggestingpotentialOsremoval

atlowsalinitiesinIceland.Giventhatthemacroalgaestudiedheretendstoliveinbrackish

49

water,itwillinteractwithseawaterofvaryingsalinitythroughatidalcycle.Samples

locatedclosertoafreshwatersourcecouldthereforereceivelessReandOsthanadeeper

watersample,whichcanbeseeninsamples16,17and18(Table1;Fig.2b),whichshow

progressivelyhigherReandOsabundancesasyoumoveseaward.Thiscouldexplainsome

oftheRevariabilityinF.vesiculosus,whereassamples(7,9,11,12,13,14,15,18,20and

27;Table1;Fig.1)closetoriverineinputsgenerallyhavelowerReabundances(8.1ppb)

thanthosethatarenotspatiallynearriverineinputs(43ppb).However,Osdoesnotfollow

thistrendsuggestingthatsomeothermechanismscontrolmacroalgaeOsuptakeand

concentration.

TheOsabundanceofthedissolvedloadofwatermeasuredinthisstudy(8.1to

750.7ppq)compareswellwithglobalriverestimates(4.6-52.1ppq)(Levasseuretal.,

1999;SharmaandWasserburg,1997)withtheexceptionofsample19(750.7ppq),which

hasanOsabundanceseveralordersofmagnitudehigherthananyrecordedwater.This

highvariationinOsabundancecanbeattributedtotheinfluenceofIcelandicfreshwater

sources(1.0-20.5ppq),withlowerconcentrationsfoundinriversdrainingolder,more

radiogenic187Os/188Osbearingbasalticcatchments(1.9ppq)whencomparedtoyounger,

lessradiogenic187Os/188Osbasalticcatchments(10.5ppq)(Gannounetal.,2006).Thiscould

alsoexplainsomeofthevariationseeninthemacroalgaeOsabundance,wherebysamples

(13,14,15,16,17,18and20;Table1;Fig.1)foundclosetothemouthsofriversdraining

oldercatchmentshavelowerconcentrations(8.0ppt)thanthosedrainingyounger

catchments(153.8ppt;Sample:7and27;Table1;Fig.1).However,thisisfurther

complicatedbythepotentialinfluenceofgeothermalwatersthatpossessrelativelyhighOs

concentrations(19.7ppq;Gannounetal.,2006),whichcouldexplainthehigherOs

concentration(36.1ppt)foundinmacroalgaearoundtheReykjanespeninsula(Samples1-

6;Table1;Fig.2a).Thisissupportedbyexperimentalstudies,whichshownincreasingRe

50

andOsabundancesinculturedF.vesiculosuswithincreasingReandOsabundancesinthe

culturesmedia(Racionero-Gómezetal.,2016;Racionero-Gómezetal.,2017).

Fig.6.Partitionscoefficients(Kd)forOs(opensquares)andRe(closedcircles)forvarying

salinities.

RheniumandOspartitioncoefficientsbetweenmacroalgaeandlocalseawater

(Table4)showalargevariationwithsalinity(Fig.6).OsmiumKdremainsrelativelyhighat

bothhighandlowsalinities,butdecreasestowardsintermediatesalinities(Fig.6).Rhenium

showsvariableKdatlowsalinitiesandhighKdathighsalinity,butdecreasestowards

intermediatesalinities(Fig.6).ThissuggestssalinitycontrolsonReandOsuptakeinto

macroalgaewithgreateruptakeatfreshwater(low)andseawater(high)salinity.Uptakeof

bothelementsisreducedatintertidalzone(intermediate)salinities.Thiscouldarisefor

severalreasonsinmacroalgaesuchas:growthrates;salinitystressorthe

conservative/unconservativebehaviourofReandOsinanestuarinehabitat.Growthrates

areknowntoincreaseatintermediatesalinities(15to20psu)insomemacroalgaespecies

(Martinsetal.,1999).However,ifReandOsuptakeislinkedtometabolicactivity,we

wouldexpecthigherabundancesinmacroalgaeatintermediatesalinitiesasmoreReand

Osareconfusedforelementsofasimilarionicradius(Racionero-Gómezetal.,2016;

0

1000

2000

3000

4000

5000

6000

7000

8000

0

200

400

600

800

1000

1200

0 5 10 15 20 25 30 35

ReK

d

OsK

d

Salinity(ppt)

51

Racionero-Gómezetal.,2017).Salinitystresscanbecausedbyfluctuatingsalinityinthe

fluidmediumwherebythemacroalgaetransportssalts,andthereforepotentiallyReand

Os,tomaintaintheircellularosmolality.Inmarineandfreshwatersystems,salinitywill

remainrelativelyconstantandthereforemacroalgaewillnotundergosalinitystress.

However,intheintertidalzone,salinitywillvarythroughatidalcycleandmacroalgaewill

thereforebeaffectedbysalinitystress.Thiscouldpotentiallypreventmacroalgaefrom

maintainingconstantReandOsuptakeandthereforeReandOsabundance(Fig.6).Aswill

bediscussedfurtherinsection2.4.2.1,Rehasbeenseentobehaveconservativelyinan

estuarinehabitatwhileOshasbeenobservedtobehavebothconservativelyor

unconservativelyinestuarinehabitats(Levasseuretal.,2000;Martinetal.,2001;Sharma

etal.,2007;Turekianetal.,2007;Milleretal.,2011).ThiscouldleadtohigherReuptake

highRe,highsalinity,seawater(Fig.6).Rheniumuptakedecreasesasyoumovelandwards

asmixingwithfreshwaterdrivestheconcentrationofRedowntowardsfreshwatervalues

(Fig.6).Osmiumontheotherhandhasbeenseentoactunconservativelyinestuarieswith

Osremovalatintermediatesalinitiesassedimentdispersalduringmixingcauses

scavengingontoparticulatematerial(Martinetal.,2001).ThiswouldacttolowerOs

abundanceinintertidalwatersandthereforelowerOsuptakeinmacroalgae(Fig.6).

However,Osabundanceinseawatermeasuredinthisstudyisnotseentovarywithsalinity

(Table3).

PreviousstudieshaveidentifiedthatReaccumulationinF.vesiculosusisvariable

acrossstructuralcomponents,indicatingsomecellsaremorespecialisedforReuptake

(Racionero-Gómezetal.,2016),whichisnotthecaseforOs(Racionero-Gómezetal.,

2017).AlthoughthebiologicalmechanismbywhichmacroalgaeextractReandOsfrom

theirmediaisnotknown,culturingstudieshaveshownthatmacroalgaedirectlytakeup

bothReandOsfromthedissolvedloadofseawaterthroughsyn-life

bioadsorption/bioaccumulation(Racionero-Gómezetal.,2016;Racionero-Gómezetal.,

52

2017).However,thesestudiesdidnotlookintoReandOsuptakeinmacroalgaefromthe

particulateorbedloadofseawater.Astrongcorrelationbetweenmacroalgaeand

dissolvedloadOsabundanceisclearfrompreviousstudies(Racionero-Gómezetal.,2017),

buttheabsenceofanycorrelationbetweenmacroalgaeandbedloadOsabundancefound

inthisstudy(Fig.4a)suggeststhatOsuptakeispurelyfromthedissolvedloadwithlittle

uptakedirectlyfromthebedload.However,macroalgaeReshowsacorrelationwithboth

thedissolved(Racionero-Gómezetal.,2016)andbedload(Fig4b).Thissuggestsa

potentialadditionalsourceofRetomacroalgaefromthebedload.However,ithasbeen

shownthattheReabundanceoftheholdfastinF.vesiculosusisrelativelylowwhen

comparedtotheotherbiologicalstructuresandthereforeReisprobablynottakenupfrom

thebedloadviathispathway(Racionero-Gómezetal.,2016).

Althoughnotmeasuredinthisstudy,ithasbeenshownthatReabundanceinthe

particulateloadofIcelandicriversiswellcorrelatedwiththerespectivebedload(Gannoun

etal.,2006).ThesimilarityofthebedloadReabundancemeasuredinthisstudy(0.5-1.1

ppb)toIcelandicrivers(0.3-1.8ppb;Gannounetal.,2006)couldsuggestapossible

correlationwithReintheparticulateloadatthesampledlocations.MacroalgaeRe

abundancecouldpotentiallybecorrelatedwiththeparticulateloadsuggestingdifferent

biologicalpathwaysfortheuptakeofOsandReinmacroalgae.

2.4.2Environmentalcontrolsonthe187Os/188Osofmacroalgae

2.4.2.1Influenceofestuarineconditionsonthe187Os/188Osofmacroalgae

Ithasbeenshownthatthe187Os/188Osoffloatingmacroalgae(SargassumfluitansandS.

natans)fromtheGulfofMexico(Rooneyetal.,2016)areindistinguishablefromthatofthe

presentdayoceanic187Os/188Osvalueof1.06(Peucker-EhrenbrinkandRavizza,2000).In

contrast,macroalgaefromwatersoffthewestcoastofGreenlanddeviatefromthisvalue

53

(0.9-1.9),insteadrecordinglocalcontinentalOsfluxintothecoastalregion(Rooneyetal.,

2016).Moreover,FucusvesiculosusfromanestuaryontheeastcoastoftheUKrecordsthe

187Os/188Osoftheseawaterinwhichitlives(0.94)andwhenculturedinseawaterdoped

withOsofaknown187Os/188Oscomposition(0.16),takesonthecompositionofthenew

source(Racionero-Gómezetal.,2017).Ithasthereforebeenpostulatedthatthe

187Os/188Osofmacroalgaecanactasaproxyforthe187Os/188Osoftheseawaterinwhichit

lives.

The187Os/188Osofmacroalgaefromthisstudy(0.16-1.0;Fig.3b)showsasimilar

rangetothe187Os/188OsofdissolvedOsinIcelandicrivers(0.15-1.04;Gannounetal.,

2006).However,whenwecomparethe187Os/188Osofmacroalgaewiththatofthe

dissolvedloadfromthesamelocationweseenorelationship(Fig.5c).Thisisprobablydue

totheentrainmentofseawaterinanestuarinehabitatfromwhichthesemacroalgaeand

watersamplesweretaken.Underestuarineconditionsmacroalgaewillcomeintocontact

withvaryingamountsoffreshwaterandseawaterthroughatidalcycle.Itscompositionwill

thereforerepresentamixingbetweenthe187Os/188Osofalocalfreshwatersourceandthe

187Os/188OsofNorthAtlanticseawater(1.02;GannounandBurton,2014).Macroalgaefrom

deeperwaters,withlittlefreshwaterinfluence,aremorelikelytotakeonan187Os/188Os

compositionsimilartoseawater(Rooneyetal.,2016;thisstudy),unliketheirshallowwater

counterparts(Racionero-Gómezetal.,2017;thisstudy).

Thisisapparentwhenthe187Os/188Osofthedissolvedloadfromthisstudy(0.16-

0.88)isplottedagainstsalinity(Fig.7).Atlowsalinity,the187Os/188Osofthefreshwater

sourceisdominantandvaluesareclosertothatexpectedforIcelandicrivers(Avg.=0.4;

Gannounetal,2006),whileathighsalinitythe187Os/188OsisclosertoaNorthAtlantic

source(Avg.=1.02;GannounandBurton,2014;GreensquareinFig.7).Foreachlocation,

macroalgaeinhabitwaterswithanintermediatesalinitybetweenthesetwoend-members

54

(blacksquaresinFig.7),andthereforethe187Os/188Osrepresentsamixtureofboth

freshwaterandseawatersources.Thisissupportedbyarctic,temperateandtropical

estuaries,whichshowanincreasingseawater187Os/188Osinfluenceasyoumoveoceanward

tohighersalinities(Levasseuretal.,2000;Martinetal.,2001;Sharmaetal.,2007;Turekian

etal.,2007).

Fig.7.Osmiumisotopic(187Os/188Os)compositionversussalinity.Blueandgreenopen

squaresrepresentvaluesforthedissolvedload(thisstudy)andNorthAtlanticseawater

(GannounandBurton,2014)respectively.Salinityhasbeeninterpolatedbasedon187Os/188Osmeasurementsformacroalgae(openblacksquares).Macroalgaegenerallyplots

atintermediatesalinitiesalongamixinglinebetweenthecorrespondingdissolvedloadfor

eachlocationandestimatesfortheNorthAtlantic.Seetextfordiscussion.

2.4.2.2Influenceofbasalticweatheringonthe187Os/188Osofmacroalgae

Duringmantlemeltingandbasaltgenesis,bothReandOsbecomefractionated.Osmium

behavesasacompatibleelementduringmeltingandisretainedinthemantle,unlikeRe,

whichismoderatelyincompatibleandentersthemelt.Mantlederivedbasaltsthushave

veryhighRe/OsvaluesandtheirprimarymineralphasescrystalliseinahighRe/Os

environment(Burtonetal.,2002;Gannounetal.,2004).Primarymineralssuchasolivine,

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 10 20 30 40

187 O

s/18

8 Os

Salinity(ppt)

55

pyroxeneandplagioclasepossesshigher187Re/188Osthanbulk-rockorglass,andovershort-

timescales(<106yrs)canproducemeasurableshiftsinradiogenicosmium(187Osr)(SeeFig.

9inGannounetal.,2006).

InIceland,riversdrainingoldercatchments(>106yrs)areundersaturatedwith

respecttothesehigh187Re/188Osbearingminerals(olivine,pyroxeneandplagioclase)and

preferentialweatheringofthesephasesfromthehostbasalt,combinedwiththeirage,

whichallowsforradiogenicingrowthof187Osfromthedecayof187Re,causeselevated

radiogenic187Os/188Osinthedissolvedload(Gannounetal.,2006).Incontrast,rivers

drainingyoungercatchments(<106yrs)areapproachingsaturationwithrespecttothese

sameminerals,whichinthisinstancehavenothadtimetoattainsignificantradiogenic

ingrowthof187Os,andweatheringtendstowardscongruencycausingthedissolvedloadto

approachthatofbulk-rock(Gannounetal.,2006).

Macroalgae187Os/188Osvaluesinthisstudy(0.16–0.99)aregenerallymuch

lowerthanthe187Os/188Osratiosof0.81to1.89previouslyrecorded(Rooneyetal.,2016;

Racionero-Gómezetal.,2017).However,macroalgaeinthisstudyreachsimilarvaluesto

culturesdopedwiththeOsstandardsolution,DROsS(Racionero-Gómezetal.,2017),

suggestingthatanunradiogenic187Os/188OssourceoftheOstakenupbythemacroalgae.

Whenthe187Os/188Osofmacroalgaeisplottedagainstthereciprocaloftheconcentration,

allthedatafallwithinafielddelimitedbythreepotentialend-members(Fig.8a):Seawater

(radiogenic187Os/188Os,intermediate[Os]);riverwaterdraininganoldbasalticcatchment

(lessradiogenic187Os/188Os,low[Os]);and,geothermalwaterandriverwaterdraininga

youngbasalticcatchment(unradiogenic187Os/188Os,high[Os]).Variationsinthe187Os/188Os

ofmacroalgaecanthereforebeexplainedbythemixingbetweenradiogenicseawaterand:

relativelyless-radiogenic187Os/188Osbearingrivers(BluesquaresinFig.8a)drainingan

oldercatchmentwhichhasundergoneincongruentweatheringofhigh187Re/188Os,and

56

thereforeradiogenic187Os/188Os,minerals;orunradiogenic187Os/188Osbearingrivers(Pink

circlesinFig.8a)drainingyoungercatchmentswhichhaveundergonecongruent

weatheringofmineralswithunradiogenic187Os/188Os.Samplesclosetoglaciallyfedrivers

(Samples9and10;Fig.1)arelikelytoattainaradiogenic187Os/188Ossignal(Orange

diamondsinFig.8a)duetotheentrainmentofseawaterintoglacialice(Gannounetal.,

2006).

Fig.8.Osmiumisotopic(187Os/188Os)compositionagainstthereciprocaloftheOs

abundancefor(a)macroalgaesampledcloseto:riversdrainingayoungbasalticterrainor

geothermalwaters(pinkcircles);riversdraininganoldbasalticterrain(bluesquares);and,

Holocenesediments(orangediamonds)and(b)F.vesiculosuswheretheOsabundanceis

convertedtothatofseawater(blacksquares)usingtherelationshipfoundinRacionero-

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

187 O

s/18

8 Os

1/[Os](ppt)

Seawater

River(old)

River(young)

b

57

Gómezetal.(2017).Athree-pointend-membermixingdiagrambasedonextremeNorth

Atlanticseawater(Green;GannounandBurton,2014)andriversdrainingayoungbasaltic

catchment(pink;Gannounetal,2006)andoldbasalticcatchments(blue,Gannounetal,

2006).Seetextfordiscussion.

ThisrelationshipbecomesmoreapparentwhentheOsabundanceinF.

vesiculosusisconvertedtotheOsabundanceinseawaterusingtherelationshipfoundin

Racionero-Gómezetal.(2017;y=0.0004x1.6607,R²=0.9898),withitsreciprocalplotted

againstthe187Os/188OsofF.vesiculosus,andthencomparedtoanidealised3-pointend-

membermixingmodel(Fig.8b).Datafalluponeither:amixinglinebetweenseawater

(187Os/188Os=1.02,[Os]=10ppq)anddirect-runoffriversdrainingoldcatchments

(187Os/188Os=0.6,[Os]=3.65ppq);amixinglinebetweenseawateranddirect-runoffrivers

drainingyoungriversorgeothermalwaters(187Os/188Os=0.14,[Os]=22.7ppq);or,closeto

thevalueforseawateritself(Fig.8b).However,thedataisoffsetinconcentrationspace

fromanidealisedmixingmodel.Thisismostlikelycausedbytheconversionrateused,

whichwasdevelopedbydopingmacroalgaeunderexceptionallyhighseawaterOs

concentrations(Racionero-Gómezetal.,2017).Moreworkisneededtoascertainhow

macroalgaeculturesbehaveatlowerOsconcentrations,moreakintothosefoundin

nature,forF.vesiculosusandothercommonmacroalgaespecies.

2.4.3Biologicalandenvironmentalcontrolson187Re/188Osofmacroalgae

The187Re/188Osratiosofmacroalgaeshowawiderangefrom~64to40,320(Fig.3b)similar

topreviousstudies(6.8to31,983;Rooneyetal.,2016,Racionero-Gómezetal.,2017),but

farexceedingthatfoundinglobalseawater(avg.187Re/188Os=4270;Peucker-Ehrenbrink

andRavizza,2000),globalriverwater(avg.187Re/188Os=227;Peucker-Ehrenbrinkand

Ravizza,2000)andthedissolvedloadofwatersmeasuredinthisstudy(125-5467).As

58

previouslydiscussedthismaybeduetothedifferentsourcesofReandOsinmacroalgae.

Sampleslocatedclosetoayoungbasalticterrain(PinkcirclesinFig.9a)arecharacterised

byrelativelyunradiogenic187Os/188Osandlow187Re/188Oscompositions,astheirisotopic

signatureiscontrolledbygeothermalandriverwatersdominatedbycongruentweathering

ofbasalticbed-rockwithrelativelyunradiogenic187Os/188Os(0.12-0.18)andlow

187Re/188Os(23.8-1310)compositions.Sampleslocatedclosetotheolderbasalticterrains

(BluesquaresinFig.8a)exhibitradiogenic187Os/188Osandhigh187Re/188Osratios,astheir

isotopicsignatureisdominatedbyriverwatersthathavebeeninfluencedbythe

preferentialweatheringofoldprimarybasalticmineralssuchasolivine,clinopyroxeneand

plagioclase,whichhaverelativelyradiogenic187Os/188Os(0.13-0.25whichhasundergone

radiogenicingrowth)andhigh187Re/188Os(288-7164)ratios(Burtonetal.,2002;Gannoun

etal.,2004;Gannounetal.,2006).Theinfluenceofseawater,witharelativelyradiogenic

187Os/188Os(1.02)andintermediate187Re/188Os(4270;Peucker-EhrenbrinkandRavizza,

2000),willpullthemacroalgaevaluestowardsaseawaterend-member(SeeFig.9).

Macroalgaethereforeholdthepotentialtonotonlyrecordthe187Os/188Osoftheseawater

inwhichtheylive,butalsotherelative187Re/188Osratiooftheseawatersources.Thisis

confirmedbythestrongrelationshipbetweenthe187Re/188Osofmacroalgaeandthe

dissolvedloadmeasuredinthisstudy(Fig5d).

Despitethisrelationship,macroalgaecanstillexhibit187Re/188Osratiosgreater

thanexpectedforIcelandicgeochemicalreservoirs.Ithasbeenproposedthatmacroalgae

uptakeReandOsthroughthesamemechanismleadingtocompetitionbetweenthese

elementsandthereforelowerReconcentrationinmacroalgaeunderhigherOsseawater

concentrationandviceversa(Racionero-Gómezetal.,2017).Thisisillustratedwhenthe

187Re/188OsratioofmacroalgaeisplottedagainstthereciprocaloftheReconcentration

(Fig.9b).AtlowmacroalgaeReconcentration(<10ppb),thereisnocompetitionbetween

ReandOsforuptakeintomacroalgae,and187Re/188Osratiosremainconsistentlylow

59

(<5,000).AstheReconcentrationrisesabove~10ppb,ReisfavouredoverOsasthetwo

elementsbegintocompeteforuptakeintomacroalgaeleadingtoanincreasein187Re/188Os

ratios(5,000-15,000).AtexceptionallyhighReconcentration(>25ppb),Recontinuestobe

takenupandthemacroalgaebecomesenrichedinRe,leadingtoexceptionallyhigh

187Re/188Osratios(15,000-40,000).

Fig.9.(a)187Re/188Osversus187Os/188Osand(b)187Re/188OsagainstthereciprocalofRe

abundanceformacroalgaesampledcloseto:riversdrainingayoungbasalticterrainor

geothermalwaters(pinkcircles);riversdraininganoldbasalticterrain(bluesquares);and,

Holocenesediments(orangediamonds).Seetextfordiscussion.

R²=0.74086

051015202530354045

0 2 4 6 8 10 12

187 Re/

188 O

sx103

1/[Re](ppb)

b

05

1015202530354045

0.0 0.2 0.4 0.6 0.8 1.0 1.2

187 Re/

188 O

sx103

187Os/188Os

a

60

ThiscouldbefurtherexacerbatedbyadditionaluptakeofRefromthebedload

and/orparticulateloadleadingtoevenhigherReconcentrationrelativetoOsin

macroalgae(Fig.4b),ortheadditionofanother187Re/188Ossignalimpartedfromthe

bedload(Fig.5b).ThesebiologicalcontrolsonReandOsuptakecouldpossiblybethe

causeofthe187Re/188Osratiosinmacroalgaeclosetotheoutflowofolderrivercatchments

beingfarhigherthanthoseobservedforrecordedprimarybasalticminerals(Fig.9a).

2.5Implicationsandfutureoutlook

TheRe-Osdataformacroalgaepresentedherehavebeensuccessfullyusedtotracethe

influenceofbasalticweatheringonthe187Os/188Osand187Re/188OsofIcelandiccoastal

watersandsubsequentmixingwithmoreradiogenicseawater.Geothermalwaterand

riversdrainingyoungbasalticcatchmentsimpartanunradiogenic187Os/188Oscomposition

andalsoarelativelylow187Re/188Ossignaltocoastalwaters.Riversdrainingolderbasaltic

catchments,wherepreferentialweatheringofprimarybasalticmineralshighinRe/Os

ratios,andwhichhavehadsufficienttimeforthedecayof187Retoform187Osr,impartsa

moreradiogenic187Os/188Oscompositionandrelativelyhigh187Re/188Ossignaltocoastal

waters.Thisprovidesfurtherevidencetosupporttheuseofthe187Os/188Osofmacroalgae

asaproxyforthe187Os/188Osofseawater,overcomingthedifficultiesassociatedwithdirect

seawateranalysisandpotentiallybecomingausefultoolfortracingavarietyofEarth

systemprocesses.

Ifsamplesobtainedatthemouthofariverrepresentstheentiredrainagearea

(Milleretal.,2011),thenitispossibletocalculatetheentirequantityandisotopic

compositionofdissolvedOsthatissuppliedtotheNorthAtlanticfromIcelandicrivers

usingmacroalgae.TheriverscontainingF.vesiculosusstudiedhaveanannualdischargeof

2.3km3/yr,accountingfor1.3%ofthetotaldischargeofIcelandicrivers(175km3/yr).Given

61

theOsabundanceofF.vesiculosus,whichhasbeenconvertedtothatofseawaterandthen

offsettobemorerepresentativeofIcelandicgeochemicalreservoirs(SeeFig.8b),we

estimateadissolvedOsfluxof2.35kg/yr.Thesevaluesarefarhigherthanpreviously

recordedestimatesforIcelandicriversof0.98kg/yr(Gannounetal.,2006).Someofthe

discrepancybetweenthesetwovaluesislikelyduetotheunderrepresentationofIcelandic

riversinthisstudy(1.3%)whencomparedtopreviousstudies(21%)(Gannounetal.,2006),

butalsosuggeststhatalthoughthisnewlydevelopedproxyoffersthepotentialtobetter

constrainthe187Os/188Oscompositionofglobalriverineinputs,moreworkisneededto

understandtheuptakerateofOsbymacroalgaeatnaturallevelsinordertoabetter

estimatetheglobalriverineabundance.Ifthiscanbedone,macroalgaecouldholdthe

potentialtoyieldabetterunderstandingoftheglobalOscycleandoceanicresidence

times.

ItisalsopossibletoestimatetheentirequantityofReandOstakenupby

macroalgaefromIceland.Thefourmacroalgaespeciesstudiedhereaccountfor46%ofthe

totalbiomassofcommonmacroalgaespeciesfoundaroundIceland(Munda,1987).Given

theaverageReandOsabundanceforeachspecies,wecalculate4.3to14.1kgofReand

3.3to11.8gofOsareabsorbedbythesemacroalgaeovertheirlifetime.Thisisrather

insignificantwhencomparedtotheIcelandicriverinputofOstotheoceanasdescribed

previously.However,ithasbeenshownthatupondeathmacroalgaedoesnotappreciably

looseRe(Racionero-Gómezetal.,2016).IfthisisthesameforOs,and6to14%ofglobal

macroalgae(6084Tg/yr)(Krause-JensenandDuarte,2016)survivesdecompositionandis

sequesteredtosediment,thenmacroalgaecouldactasasinkof0.03to89kg/yrofReand

0.001to0.2kg/yrforOs,whenbasedontherangeofcompositionsofallmacroalgae

speciesstudiedtodate(Masetal.,2005;Racionero-Gómezetal.,2016;Racionero-Gómez

etal.,2017;Rooneyetal.,2016;Yang,1991).Thisisinsignificantwhencomparedto

estimatesforglobalsedimentsinksof18to29x103kg/yrand260to1350kg/yrforReand

62

Os,respectively(MorfordandEmerson,1999;Oxburgh,2001),suggestingmacroalgae

alonedoesnothaveasignificantcontrolintheglobalmarineReandOscycle,bothtoday

andinEarth’sgeologicalpast.

2.6References

Anbar,A.D.,Creaser,R.A.,Papanastassiou,D.A.,Wasserburg,G.J.,1992.Rheniuminseawater:Confirmationofgenerallyconservativebehavior.GeochimicaetCosmochimicaActa56,4099-4103.

Armstrong,R.L.,1971.Glacialerosionandthevariableisotopiccompositionofstrontiuminseawater.Nature230,132-133.

Berner,R.A.,Lasaga,A.C.,Garrels,R.M.,1983.Thecarbonate-silicategeochemicalcycleanditseffectonatmosphericcarbondioxideoverthepast100millionyears.AmJSci283,641-683.

Birck,J.L.,Barman,M.R.,Capmas,F.,1997.Re-Osisotopicmeasurementsatthefemtomolelevelinnaturalsamples.Geostandardsnewsletter21,19-27.

Burton,K.W.,Gannoun,A.,Birck,J.-L.,Allègre,C.J.,Schiano,P.,Clocchiatti,R.,Alard,O.,2002.Thecompatibilityofrheniumandosmiuminnaturalolivineandtheirbehaviourduringmantlemeltingandbasaltgenesis.EarthandPlanetaryScienceLetters198,63-76.

Chen,C.,Sharma,M.,2009.Highprecisionandhighsensitivitymeasurementsofosmiuminseawater.Analyticalchemistry81,5400-5406.

Cohen,A.S.,Waters,F.G.,1996.Separationofosmiumfromgeologicalmaterialsbysolventextractionforanalysisbythermalionisationmassspectrometry.AnalyticaChimicaActa332,269-275.

Colodner,D.,Edmond,J.,Boyle,E.,1995.RheniumintheBlackSea:comparisonwithmolybdenumanduranium.EarthandPlanetaryScienceLetters131,1-15.

Colodner,D.,Sachs,J.,Ravizza,G.,Turekian,K.,Edmond,J.,Boyle,E.,1993a.Thegeochemicalcycleofrhenium:areconnaissance.EarthandPlanetaryScienceLetters117,205-221.

Colodner,D.C.,Boyle,E.A.,Edmond,J.M.,1993b.Determinationofrheniumandplatinuminnaturalwatersandsediments,andiridiuminsedimentsbyflowinjectionisotopedilutioninductivelycoupledplasmamassspectrometry.AnalyticalChemistry65,1419-1425.

Creaser,R.A.,Papanastassiou,D.A.,Wasserburg,G.J.,1991.Negativethermalionmassspectrometryofosmium,rheniumandiridium.GeochimicaetCosmochimicaActa55,397-401.

63

Cumming,V.M.,Poulton,S.W.,Rooney,A.D.,Selby,D.,2013.AnoxiaintheterrestrialenvironmentduringthelateMesoproterozoic.Geology41,583-586.

Debaille,V.,Trønnes,R.G.,Brandon,A.D.,Waight,T.E.,Graham,D.W.,Lee,C.-T.A.,2009.Primitiveoff-riftbasaltsfromIcelandandJanMayen:Os-isotopicevidenceforamantlesourcecontainingenrichedsubcontinentallithosphere.GeochimicaetCosmochimicaActa73,3423-3449.

Gannoun,A.,Burton,K.,2014.HighprecisionosmiumelementalandisotopemeasurementsofNorthAtlanticseawater.JAAS.

Gannoun,A.,Burton,K.W.,Thomas,L.E.,Parkinson,I.J.,vanCalsteren,P.,Schiano,P.,2004.Osmiumisotopeheterogeneityintheconstituentphasesofmid-oceanridgebasalts.Science303,70-72.

Gannoun,A.,Burton,K.W.,Vigier,N.,Gíslason,S.R.,Rogers,N.,Mokadem,F.,Sigfússon,B.,2006.Theinfluenceofweatheringprocessonriverineosmiumisotopesinabasalticterrain.EarthandPlanetaryScienceLetters243,732-748.

Huh,Y.,Birck,J.-L.,Allègre,C.J.,2004.OsmiumisotopegeochemistryintheMackenzieRiverbasin.EarthandPlanetaryScienceLetters222,115-129.

Ishikawa,A.,Senda,R.,Suzuki,K.,Dale,C.W.,Meisel,T.,2014.Re-evaluatingdigestionmethodsforhighlysiderophileelementand187Osisotopeanalysis:Evidencefromgeologicalreferencematerials.ChemicalGeology384,27-46.

Jóhannesson,H.,2014.GeologicalMapofIceland.1:600000.BedrockGeology.IcelandicInstituteofNaturalHistory,Reykjavík(2ndedition).

Krause-Jensen,D.,Duarte,C.M.,2016.Substantialroleofmacroalgaeinmarinecarbonsequestration.NatureGeosci9,737-742.

Levasseur,S.,Birck,J.-L.,Allegre,C.,1999.Theosmiumriverinefluxandtheoceanicmassbalanceofosmium.EarthandPlanetaryScienceLetters174,7-23.

Levasseur,S.,Birck,J.-L.,Allègre,C.J.,1998.Directmeasurementoffemtomolesofosmiumandthe187Os/186Osratioinseawater.Science282,272-274.

Levasseur,S.,Rachold,V.,Birck,J.-L.,Allegre,C.,2000.Osmiumbehaviorinestuaries:theLenaRiverexample.EarthandPlanetaryScienceLetters177,227-235.

Martin,C.E.,Peucker-Ehrenbrink,B.,Brunskill,G.,Szymczak,R.,2001.Osmiumisotopegeochemistryofatropicalestuary.GeochimicaetCosmochimicaActa65,3193-3200.

Mas,J.,Tagami,K.,Uchida,S.,2005.RheniummeasurementsonNorthAtlanticseaweedsamplesbyID-ICP-MS:anobservationontheReconcentrationfactors.JournalofRadioanalyticalandNuclearChemistry265,361-365.

64

Miller,C.A.,Peucker-Ehrenbrink,B.,Walker,B.D.,Marcantonio,F.,2011.Re-assessingthesurfacecyclingofmolybdenumandrhenium.GeochimicaetCosmochimicaActa75,7146-7179.

Morford,J.L.,Emerson,S.,1999.Thegeochemistryofredoxsensitivetracemetalsinsediments.GeochimicaetCosmochimicaActa63,1735-1750.

Munda,I.M.,1987.DistributionanduseofsomeeconomicallyimportantseaweedsinIceland,TwelfthInternationalSeaweedSymposium.Springer,pp.257-260.

Nowell,G.,Luguet,A.,Pearson,D.,Horstwood,M.,2008.Preciseandaccurate186Os/188Osand187Os/188Osmeasurementsbymulti-collectorplasmaionisationmassspectrometry(MC-ICP-MS)partI:Solutionanalyses.ChemicalGeology248,363-393.

Oxburgh,R.,2001.Residencetimeofosmiumintheoceans.Geochemistry,Geophysics,Geosystems2,1018.

Paul,M.,Reisberg,L.,Vigier,N.,2009.Anewmethodforanalysisofosmiumisotopesandconcentrationsinsurfaceandsubsurfacewatersamples.ChemicalGeology258,136-144.

Peucker-Ehrenbrink,B.,Ravizza,G.,2000.Themarineosmiumisotoperecord.TerraNova12,205-219.

Peucker-Ehrenbrink,B.,Ravizza,G.,2012.Osmiumisotopestratigraphy.TheGeologicTimeScale2012,145-166.

Peucker-Ehrenbrink,B.,Sharma,M.,Reisberg,L.,2013.Recommendationsforanalysisofdissolvedosmiuminseawater.Eos,TransactionsAmericanGeophysicalUnion94,73-73.

Prouty,N.G.,Roark,E.B.,Koenig,A.E.,Demopoulos,A.W.,Batista,F.C.,Kocar,B.D.,Selby,D.,McCarthy,M.D.,Mienis,F.,Ross,S.W.,2014.Deep-seacoralrecordofhumanimpactonwatershedqualityintheMississippiRiverBasin.GlobalBiogeochemicalCycles28,29-43.

Racionero-Gómez,B.,Sproson,A.,Selby,D.,Gröcke,D.,Redden,H.,Greenwell,H.,2016.Rheniumuptakeanddistributioninphaeophyceaemacroalgae,Fucusvesiculosus.RoyalSocietyOpenScience3,160161.

Racionero-Gómez,B.,Sproson,A.D.,Selby,D.,Gannoun,A.,Gröcke,D.R.,Greenwell,H.C.,Burton,K.W.,2017.Osmiumuptake,distribution,and187Os/188Osand187Re/188OscompositionsinPhaeophyceaemacroalgae,Fucusvesiculosus:Implicationsfordeterminingthe187Os/188Oscompositionofseawater.GeochimicaetCosmochimicaActa199,48-57.

Raymo,M.E.,Ruddiman,W.F.,Froelich,P.N.,1988.InfluenceoflateCenozoicmountainbuildingonoceangeochemicalcycles.Geology16,649-653.

Richter,F.M.,Turekian,K.K.,1993.Simplemodelsforthegeochemicalresponseoftheoceantoclimaticandtectonicforcing.EarthandPlanetaryScienceLetters119,121-131.

65

Rooney,A.D.,Selby,D.,Lloyd,J.M.,Roberts,D.H.,Lückge,A.,Sageman,B.B.,Prouty,N.G.,2016.Trackingmillennial-scaleHoloceneglacialadvanceandretreatusingosmiumisotopes:InsightsfromtheGreenlandicesheet.QuaternaryScienceReviews138,49-61.

Sharma,M.,Balakrishna,K.,Hofmann,A.W.,Shankar,R.,2007.ThetransportofOsmiumandStrontiumisotopesthroughatropicalestuary.GeochimicaetCosmochimicaActa71,4856-4867.

Sharma,M.,Chen,C.,Blazina,T.,2012.Osmiumcontaminationofseawatersamplesstoredinpolyethylenebottles.LimnologyandOceanography:Methods10,618-630.

Sharma,M.,Papanastassiou,D.,Wasserburg,G.,1997.Theconcentrationandisotopiccompositionofosmiumintheoceans.GeochimicaetCosmochimicaActa61,3287-3299.

Sharma,M.,Wasserburg,G.,1997.Osmiumintherivers.Geochimicaetcosmochimicaacta61,5411-5416.

Sharma,M.,Wasserburg,G.,Hofmann,A.,Chakrapani,G.,1999.Himalayanupliftandosmiumisotopesinoceansandrivers.GeochimicaetCosmochimicaActa63,4005-4012.

Turekian,K.K.,Sharma,M.,Gordon,G.W.,2007.ThebehaviorofnaturalandanthropogenicosmiumintheHudsonRiver–LongIslandSoundestuarinesystem.GeochimicaetCosmochimicaActa71,4135-4140.

Völkening,J.,Walczyk,T.,Heumann,K.G.,1991.Osmiumisotoperatiodeterminationsbynegativethermalionizationmassspectrometry.InternationalJournalofMassSpectrometryandIonProcesses105,147-159.

Walker,J.C.G.,Hays,P.B.,Kasting,J.F.,1981.Anegativefeedbackmechanismforthelong-termstabilizationofEarth'ssurfacetemperature.JournalofGeophysicalResearch86,9776.

Woodhouse,O.,Ravizza,G.,Falkner,K.K.,Statham,P.,Peucker-Ehrenbrink,B.,1999.Osmiuminseawater:verticalprofilesofconcentrationandisotopiccompositionintheeasternPacificOcean.EarthandPlanetaryScienceLetters173,223-233.

Yang,J.S.,1991.Highrheniumenrichmentinbrownalgae:abiologicalsinkofrheniuminthesea?Hydrobiologia211,165-170.

66

Chapter3

TracinganthropogenicosmiumaroundJapanusingtheosmiumisotopic

compositionofmacroalgae*

*AversionofthischapterwillbesubmittedtoEnvironmentalScienceandTechnology,co-

authoredwithDavidSelbyandKatsuhikoSuzuki.

67

Thisstudypresentsrhenium(Re)andosmium(Os)abundanceandisotopedatafor

macroalgaefromJapanesecoastalwatersadjacenttodenselypopulatedmegacities

(Tokyo,OsakaandNagoya)andsparselypopulatedregionsofJapan(TheIzupeninsula,The

Notopeninsula,northernHonshuandHokkaido).The187Os/188Oscompositionof

macroalgaeishighlyvariable(0.16to1.09)andcanbeexplainedintermsofmixing

betweenseawaterand:riversdrainingMiocene-Holocenecontinentalrockswitha

radiogenic187Os/188Oscomposition;and/or,anthropogenicsourcesofOswithan

unradiogenic187Os/188Oscomposition.TheOsemittedfromcatalyticconvertorusein

vehiclespredominantlycontrolstheOsisotopebudgetofcoastalwatersindensely

populatedregionsofJapan.However,Osderivedfrommedicalresearch,municipalsolid

wasteincineratorsandsewageoutflowcangeneratelocalisedpointsourcesof

anthropogenicOs.AnthropogenicOsattributedtocoastalcities,whichistransferredvia

surfacewaterstoworld’soceans,representsasubstantialsourceofunradiogenicOswhich

isdrivingsurfaceseawatertolower187Os/188Osratios(~0.95)thandeeperwaters(~1.05).

ThissuggeststhedemandinPGEs,theirrefiningandeventualuseincatalyticconvertorsis

impactingtheglobalOscycle.Finally,thisstudysuggeststhatosmiumisotopesin

macroalgaecouldbecomeapowerfulenvironmentalindicatorandtracerofcontinental

inputstothemarineOscycle.

3.1Introduction

Theosmiumisotopiccomposition(187Os/188Os)ofseawaterreflectsabalancebetween

radiogeniccontinentalsourcesandunradiogenicmantleandextraterrestrialderived

sources,andhastraditionallybeenusedtotraceawiderangeofEarthsystemprocesses,

bothtodayandintheEarth’sgeologicalpast(Peucker-EhrenbrinkandRavizza,2000,2012).

However,humanactivityisperturbingthemarineOscycle(Chenetal.,2009)and

anthropogenicOscontaminationhasbeendetectedinestuaries(Turekianetal.,2007;

68

Williamsetal.,1997),coastalsediments(EsserandTurekian,1993;RavizzaandBothner,

1996),lakes(Rauchetal.,2004)andprecipitation(Chenetal.,2009)(SeeTable1).

PotentialsourcesofanthropogenicOstotheatmosphereandthemarineenvironment

include:(i)thecombustionoffossilfuels;(ii)smeltingofplatinumgroupelement(PGE:Os,

Ir,Pt,Pd,Ru,Rh)sulphideores;(iii)smeltingofbase-metal(Pb,Ni,Cu,andZn)sulphide

ores;(iv)smeltingofaluminiumore(v)smeltingofchromiumore;(vi)exhaustfrom

automobilecatalyticconvertors;(vii)emissionsfromincinerators;and,(viii)sewage

outflow(SeeTable1).

Table1

ComparisonofnaturalandanthropogenicsourcesofOsandRe.Datareferencesare:

seawater(Peucker-EhrenbrinkandRavizza,2000);riverwater(Peucker-Ehrenbrinkand

Ravizza,2000);precipitation(Chenetal.,2009);loess(Peucker-EhrenbrinkandJahn,2001);

cosmicdust(Peucker-EhrenbrinkandRavizza,2000);HThydrothermal(Peucker-Ehrenbrink

andRavizza,2000);LThydrothermal(Peucker-EhrenbrinkandRavizza,2000);fossilfuels

(SelbyandCreaser,2005;Selbyetal.,2007);base-metalsulphideores(Lambertetal.,

1998;Morganetal.,2002;Walkeretal.,1994);PGEores(McCandlessandRuiz,1991);

chromites(Walkeretal.,2002);aluminiumsmelter(Gogotetal.,2015);airborneparticles

(Rauchetal.,2005);urbansediments(Rauchetal.,2004);carcatalysts(Poirierand

Gariépy,2005);sewagesludge(EsserandTurekian,1993);MSWIashes(Funarietal.,2016).

[Os](fg/g) 187Os/188Os [Re](pg/g) 187Re/188Os

Natural

Seawater 10 1.06 8.2 4270

RiverWater 9.1 1.4 428 227

Precipitation 0.3-16.9 0.16-0.98

Loess 3.1x104 1.05 3x105 50

CosmicDust 5x108 0.13 3.7x107 0.4

HThydrothermal 2.8-38 0.13-0.39 LThydrothermal 98 0.11 Anthropogenic

FossilFuels 0.01-295x103 0.9-6 0.3-40.7x103 400-1300

Base-metalSulphideOre 0.02-89x106 0.13-24 0.6-286x103 0.8-5232

PGEOre

0.15-0.19

69

Chromites 2-1320x106 0.12 0.01-9.3x103 0.01-2.2

AluminiumSmelter 30x103 2.5

AirborneParticles 0.001-0.82pg/m3 0.3-2.8

UrbanSediments 56-395x103 0.36-2.2

CarCatalysts 6-228x103 0.16-0.19

SewageSludge 10-100x103 0.6-1.7

MSWIAshes 26-1650x103 0.24-0.7

Highlyradiogenicanthropogenic187Os/188Ossourcesincludethesmeltingof

bauxiteore(~2.5)andfossilfuels(1.1to13.7)(Finlayetal.,2011;Gogotetal.,2015;Selby

andCreaser,2005;Selbyetal.,2007).Incontrast,the187Os/188OsvaluesofPGEsulphide

oresfromtheBushveldComplex,SouthAfrica(0.15to0.2),chromites(0.12to0.14),

municipalsolidwasteincinerator(MSWI)ash(0.24to0.7)andsewagesludgearegenerally

unradiogenic(EsserandTurekian,1993;Funarietal.,2016;McCandlessandRuiz,1991;

RavizzaandBothner,1996;Turekianetal.,2007;Walkeretal.,1994;Williamsetal.,1997).

The187Os/188Osofbase-metalsulphideoresarehighlyvariable,andinsomecasescanbe

relativelyradiogenic(e.g.Sudbury,Canada)duetocontaminationfromcontinental

sources,whilstinothercasestheycanrecordunradiogenic187Os/188Osvaluessimilarto

mantlederivedsources(e.g.Noril’sk,RussiaandYilgarncraton,Australia)(Lambertetal.,

1998;Morganetal.,2002;Walkeretal.,2002).Roughly95%(Jones,1999)ofallPGEsare

derivedfromunradiogenicsources(0.15-0.2),whichisreflectedinthe187Os/188Os

measurementsofthecatalyticconvertorsfromwhichtheyaremade(0.1-0.2)(Poirierand

Gariépy,2005),suggestingthe187Os/188Osofautomobileexhaustwillbeofsimilarvalues

(Chenetal.,2009).

Smelters(Gogotetal.,2015;Rodushkinetal.,2007b),incinerators(Funarietal.,

2016;Turekianetal.,2007;WilliamsandTurekian,2002)andsewageoutflow(Esserand

Turekian,1993;RavizzaandBothner,1996;Turekianetal.,2007;Williamsetal.,1997)

representpointsourcesinanthropogenicOs,whereascatalyticconvertorsrepresenta

70

regionalinfluenceindenselypopulatedareas(PoirierandGariépy,2005).Onaglobalscale,

therefiningofPGEores,andtoalesserextentautomobileexhaust,hasdrivenglobal

precipitationtowardstheisotopiccompositionofores(~0.2)(Chenetal.,2009).Moreover,

present-daysurfacewatershavelower187Os/188Osvalues(~0.95)thandeepwaters(~1.05),

suggestinghumanshaveimpactedtheglobalOscycleandtheworld’soceans(Chenetal.,

2009).Osmiumisotopesthereforeholdthepotentialtobecomeapowerfultracerof

hydrologic,oceanographicandanthropogenicprocesses‘similartoPbfromleadedgasoline

usagebefore1978ortritiumfromatmosphericatomicbombtestingintheearly1960s’

(Chenetal.,2009).

Despitethispotential,andtheestablishmentofultra-lowblankanalytical

techniquescapableofoxidisingallosmiumtoacommonoxidationstate(ChenandSharma,

2009;GannounandBurton,2014;Levasseuretal.,1998;Pauletal.,2009),thedirect

analysisofOsinseawaterstillremainsanalyticallychallengingduetoultra-lowOs

concentrations(Peucker-Ehrenbrinketal.,2013).Therefore,themeasurementsof

anthropogenicOsinrivers,estuariesandcoastalandoceanicwatersremainsparse.Recent

workindicatesthatmacroalgae(seaweed)concentratesOs(withabundancesthatvary

from3.3to254.5ppt),whilstmaintainingthe187Os/188Oscompositionoftheseawaterit

inhabits(Racionero-Gómezetal.,2017;Rooneyetal.,2016;Chapter2).Thissuggeststhat

macroalgaecouldactasaproxyforthe187Os/188Oscompositionoflocalwaterswhilst

removingsomeoftheanalyticalchallengesassociatedwithdirectanalysisofseawateri.e.

ultra-lowconcentrationsandmultipleoxidationstates.Macroalgaeexistingincoastal

waters,therefore,shouldrecordan187Os/188Ossignaturethatreflectsabalanceoflocal

inputs,includingriverineinput,localbedrockandseawaterandhasbeenutilisedto

understandnaturalprocessesinGreenland(Rooneyetal.,2016)andIceland(Chapter2).

Japanoffersauniqueplaceinwhichtostudytheinfluenceofanthropogenic

processesonregionalvariationsinthemarineOscycle.Large,denselypopulated,

71

sprawlingmetropolitanareasoftenfallincloseproximitytocoastalwaters,inletseasor

bays.Thesemetropolitanareasareoftenjuxtaposedtosparselypopulatedruraland/or

mountainousregions,withlittlehumanactivityandthereforeanthropogenicinfluence.

ThisstudypresentsRe-OsabundanceandisotopedataformacroalgaefromTokyoBay,

OsakaBay,IseBay,MikawaBay,IzuPeninsula,NotoPeninsula,Hokkaidoandnorthern

Honshu.

TokyoBay,OsakaBayandIse/MikawaBaylieincloseproximitytotheKanto

(Tokyo-Yokohama),Keihanshin(Osaka-Kobe)andChukyo(Nagoya)metropolitanareas.The

187Os/188Ossignaturefromtheseregionsismuchlowerthanexpectedfornaturalriverand

oceanicsystems,butsimilartotheisotopecompositionofPGEores.Thissuggeststhat

humanactivityhasinfluencedthe187Os/188Osofmacroalgae,andthereforeseawater,

throughtheburningofmunicipaland/orhospitalwaste,processingofsewageandthe

extensiveuseofautomobilesintheseareas.TheIzuPeninsula,theNotoPeninsula,

HokkaidoandnorthernHonshuexhibit187Os/188Osvaluessimilartoglobalriverwateror

Pacificseawatermeasurements,suggestinglittleinfluencefromhumanactivity.These

resultsdemonstratetheabilityofmacroalgaetotracefluctuationsinthe187Os/188Os

compositionsoffreshwaterandseawateraroundJapan,andutilisethistodeterminethe

influenceofhumanactivityontheglobalosmiumbudget.

3.2Fieldandanalyticaltechniques

3.2.1Samplingandstorage

MacroalgaefromOsakaBay(SampleLocation16-22)weresampledat7locationsduring

September2014.MacroalgaefromTokyoBay(SampleLocation1-15),theIzupeninsula

(SampleLocation31-36)andtheNotopeninsula(SamplesLocation37-43)weresampledat

28locationsbetweenJulyandSeptember2015.Afurther8locationsintheIse(Sample

72

Location23-24)andMikawa(SampleLocation25-30)BaysweresampledduringDecember

2015.MacroalgaefromnorthernHonshu(SampleLocation44-48)weresampledat5

locationsduringSeptember2016.SamplesfromHokkaidowerepurchasedattsukijifish

marketinTokyo.Intotalsixty-fourmacroalgaesampleswerecollected(Fig.1;Table1).

Macroalgaewerewashedusingdeionised(Milli-Q™)watertoremoveanyattached

sedimentandsalt.Theywerethendriedfor12hat60°Candstoredinplasticzip-lock

bags.Macroalgaewerelatercrushedusinganagatepestleandmortarpriortoanalysis.

Fig.1.MapofJapan.MacroalgaesamplelocationsfornorthernHonshuandHokkaidoare

displayed(Circles;Table2).DashedlinesrepresentanoutlineofareasshowninFigures2

(TokyoBay),3(OsakaBay),4(IseandMikawaBay),5(TheIzupeninsula)and6(Kanazawa

andtheNotopeninsula).

73

3.2.2Macroalgaespeciesandhabitats

Specificmacroalgaespecieswerenottargetedduringthisstudy,andsampleswere

selectedbasedontheiravailabilityatsamplesites.Therewashoweverapreferenceto

brownmacroalgaeovergreenandredmacroalgaeduetotheirrelativelyhighabundancein

Re(Masetal.,2005;Proutyetal.,2014;Racionero-Gómezetal.,2016;Yang,1991)andOs

(Racionero-Gómezetal.,2017;Chapter2;Rooneyetal.,2016).Twentyspeciesof

macroalgaewereanalysed:Monostromanitidum;Urosporapenicilliformis;Hizikia

fusiformis;Undariapinnatifidia;DictyopterisundulataHolmes;Gloiopeltiscomplanata;

Spatoglossumcrissum;Gloiopeltisfurcate;Sargassumfusiforme;Sargassummuticum;

GelidumelegansKutzing;Gracilariavermiculophylla(Ohmi)Papenfuss;Sargassumhorneri;

Grateloupiacarnosa;Schizymeniadubyi;Grateloupialanceolate;PorphyrateneraKjellman;

Pachydictyoncoriaceum(Holmes)Okamura;Gracilariabursa-pastoris;and,Laminaria

japonica.

3.2.3Re-Osanalysis

TheRe-OsanalysisofmacroalgaewascarriedoutintheDurhamGeochemistryCentre

(LaboratoryforSulfideandSourceRockGeochronologyandGeochemistry).

3.2.3.1Macroalgae

ThetechniqueforchemicalseparationofReandOsfrommacroalgaeisreportedby

Racionero-Gómezetal.(2017).Inbrief,approximately200mgofpowderedmacroalgae

wasintroducedintoaCariustubetogetherwith11NHCl(3mL),15.5NHNO3(6mL)anda

knownamountof185Re+190Ostracersolutionandheatedto220°Cinanovenfor24h.The

OswasisolatedfromtheacidmediumusingCHCl3solventextractionandthenback

extractedintoHBr.TheOswasfurtherpurifiedusingaCrO3-H2SO4–HBrmicro-distillation

74

(Bircketal.,1997;CohenandWaters,1996).TheremainingRe-bearingacidmediumwas

evaporatedtodrynessat80°C,withtheReisolatedandpurifiedusingbothNaOH-acetone

solventextractionandHNO3-HClanionchromatography(Cummingetal.,2013).

3.2.3.2MassSpectrometry

ThepurifiedReandOsfractionswereloadedontoNiandPtfilamentsrespectively,and

measuredusingNTIMS(Creaseretal.,1991;Völkeningetal.,1991)onaThermoScientific

TRITONmassspectrometerusingFaradaycollectorsinstaticmode,andanelectron

multiplierindynamicmoderespectively.TheReandOsabundancesandisotope

compositionsarepresentedwith2s.e.(standarderror)absoluteuncertaintieswhich

includefullerrorpropagationofuncertaintiesinthemassspectrometermeasurements,

blank,spikecalibrations,andsampleandspikeweights.Fullanalyticalblankvaluesforthe

macroalgaeanalysisare10.9±5.9pgforRe,0.13±0.13pgforOs,witha187Os/188Os

compositionof0.61±0.34(1SD,n=4).

Tomonitorthelong-termreproducibilityofmassspectrometermeasurements

ReandOs(DROsS,DTM)referencesolutionswereanalysed.The125pgResolutionyields

anaverage185Re/187Reratioof0.5987±0.0023(2SD.,n=8),whichisinagreementwith

publishedvalues(e.g.,Cummingetal.,2013andreferencestherein).A50pgDROsS

solutiongavean187Os/188Osratioof0.16111±0.0008(2SD.,n=8),whichisinagreement

withreportedvaluefortheDROsSreferencesolution(Nowelletal.,2008).

3.3Results

TheReandOsabundanceandisotopedataformacroalgaearepresentedinTable2.The

ReandOsabundanceshowalargerangefrom0.02to21.43ppband1.96to122.76ppt

respectively.The187Os/188Osand187Re/188Oscompositionsarehighlyvariableandrange

75

from0.16to1.09and8.7to11234.5respectively.Thereason(s)forthevariationis

discussedissection4.Insection3.3.2to3.3.6,thepositionofsamplelocationsare

comparedtoseveralindicatorsofhumanactivity.Populationdensitydata(ESRI,2014),Os

pollutionsources(GoogleEarth,2016)andCO2emissiondatafromEAGrid2010(Kannariet

al.,2007)canbefoundinpanelb,candd(Figs.2to6)respectively.

Table2

RheniumandosmiumabundanceandisotopedataforJapanesemacroalgaesamples

SampleLocation MacroalgaeSpecies

[Re](ppb) 2s.e.

[Os](ppt) 2s.e. 187Re/188Os 2s.e. 187Os/188Os 2s.e.

TokyoBay1 Monostromanitidum 0.02 0.03 3.90 0.08 28.7 38.9 0.519 0.0322 Monostromanitidum 3.56 0.20 11.27 0.22 1603.9 113.4 0.541 0.0313 Monostromanitidum 2.88 0.05 3.33 0.07 4462.4 215.1 0.674 0.0414 Monostromanitidum 0.07 0.03 7.67 0.08 42.2 19.4 0.363 0.011

5Urospora

penicilliformis 3.72 0.04 24.03 0.11 789.5 9.3 0.582 0.0065 Monostromanitidum 0.305 0.001 12.18 0.05 124.7 1.0 0.397 0.0046 Monostromanitidum 3.65 0.04 29.19 0.56 639.5 26.7 0.592 0.034

7Urospora

penicilliformis 0.52 0.03 29.86 0.31 87.9 5.4 0.463 0.014

7Urospora

penicilliformis 0.938 0.009 30.06 0.31 156.3 3.6 0.428 0.012

8Urospora

penicilliformis 0.233 0.003 26.57 0.50 43.9 1.9 0.436 0.025

rptUrospora

penicilliformis 0.37 0.03 18.14 0.34 101.6 9.3 0.455 0.0269 Monostromanitidum 0.127 0.001 17.21 0.13 37.0 0.6 0.418 0.009 9 Monostromanitidum 0.269 0.001 17.60 0.08 76.7 0.6 0.450 0.005 10 Monostromanitidum 0.05 0.03 3.77 0.08 71.0 40.5 0.551 0.03411 Monostromanitidum 0.06 0.03 3.44 0.07 85.7 45.2 0.669 0.04112 Hizikiafusiformis 1.63 0.09 10.00 0.07 861.6 50.3 0.867 0.01213 Undariapinnatifidia 2.14 0.02 5.85 0.04 1948.2 27.2 0.950 0.015

14Dictyopterisundulata

Holmes 0.10 0.03 7.40 0.15 68.5 21.7 0.910 0.053

15Gloiopeltiscomplanata 7.65 0.14 17.10 0.20 2362.2 66.3 0.857 0.026

OsakaBay

16Spatoglossum

crassum 1.33 0.07 13.25 0.08 502.6 26.7 0.445 0.00716 Gloiopeltisfurcata 0.58 0.03 21.31 0.22 137.3 8.5 0.497 0.01417 Sargassumfusiforme 3.64 0.17 122.76 0.54 143.6 6.9 0.165 0.00217 Undariapinnatifidia 4.21 0.15 23.91 0.45 876.0 47.7 0.380 0.022rpt Undariapinnatifidia 3.49 0.17 24.87 0.39 697.4 40.6 0.364 0.01718 Undariapinnatifidia 7.05 0.25 22.78 0.25 1589.2 65.5 0.628 0.01818 Undariapinnatifidia 4.72 0.22 14.34 0.20 1683.0 91.7 0.597 0.02319 Undariapinnatifidia 4.21 0.15 18.81 0.36 1144.2 62.5 0.591 0.03419 Gloiopeltisfurcata 0.55 0.04 11.36 0.12 246.6 18.5 0.521 0.01520 Sargassumfusiforme 2.58 0.10 25.89 0.16 518.8 20.0 0.758 0.01021 Sargassummuticum 0.38 0.03 16.59 0.18 114.5 10.3 0.508 0.01522 Undariapinnatifidia 3.18 0.12 19.34 0.37 830.8 45.6 0.497 0.029

IseBay23 Gelidumelegans 0.16 0.03 88.80 0.84 8.7 1.7 0.158 0.005

76

Kutzing

23

Gracilariavermiculophylla(Ohmi)Papenfuss 0.08 0.03 27.06 0.49 15.0 5.4 0.185 0.011

24 Undariapinnatifida 1.37 0.07 22.01 0.13 318.5 16.7 0.602 0.008MikawaBay

25 Sargassumhorneri 0.36 0.03 10.62 0.21 176.3 18.5 0.742 0.04326 Sargassummuticum 0.24 0.03 7.31 0.15 167.3 23.8 0.689 0.04127 Grateloupiacarnosa 3.82 0.18 5.30 0.09 3661.5 217.3 0.544 0.02728 Sargassummuticum 21.43 1.00 14.78 0.08 7523.2 354.7 0.716 0.00829 Schizymeniadubyi 0.19 0.03 3.00 0.06 327.6 55.1 0.513 0.032

30Grateloupialanceolata 0.27 0.03 3.51 0.04 390.0 48.1 0.537 0.017

IzuPeninsula31 Hizikiafusiformis 4.83 0.17 16.92 0.19 1504.3 62.5 0.842 0.025

32Porphyratenera

Kjellman 0.61 0.04 7.78 0.09 407.4 26.4 0.704 0.022

33

Pachydictyoncoriaceum(Holmes)

Okamura 4.22 0.15 10.25 0.14 2129.3 97.2 0.692 0.027

34Dictyopterisundulata

Holmes 4.79 0.17 15.21 0.14 1628.8 64.8 0.697 0.01735 Hizikiafusiformis 0.65 0.04 28.10 0.32 122.0 7.6 0.905 0.02636 Undariapinnatifidia 0.89 0.04 8.69 0.06 537.4 26.9 0.776 0.010

KanazawaandtheNotoPeninsula37 Monostromanitidum 0.16 0.03 5.18 0.10 158.6 30.7 0.459 0.028

38Gracilariabursa-

pastoris 0.12 0.03 6.13 0.12 100.8 25.5 0.476 0.02838 Monostromanitidum 0.03 0.03 2.29 0.05 70.2 66.1 0.483 0.031

39Gracilariabursa-

pastoris 0.61 0.04 19.24 0.38 166.0 13.2 0.848 0.049

40Dictyopterisundulata

Holmes 3.18 0.15 27.16 0.54 623.1 39.0 0.937 0.05441 Sargassummuticum 1.73 0.09 32.80 0.21 279.1 14.1 0.888 0.011

42Dictyopterisundulata

Holmes 2.56 0.11 12.20 0.07 1101.0 49.0 0.822 0.00943 Sargassummuticum 1.60 0.07 25.26 0.17 341.0 16.1 1.017 0.013

NorthernHonshu44 Laminariajaponica 0.82 0.04 3.98 0.05 1078.3 54.7 0.763 0.02445 Laminariajaponica 2.12 0.06 6.40 0.04 1727.7 54.1 0.765 0.00846 Laminariajaponica 1.44 0.05 16.16 0.14 470.1 17.3 0.884 0.01847 Laminariajaponica 0.60 0.03 12.10 0.14 263.2 16.0 0.921 0.02748 Laminariajaponica 3.92 0.11 11.10 0.19 1877.8 83.3 0.923 0.045

Hokkaido49 Laminariajaponica 7.71 0.02 3.69 0.05 11234.5 257.2 1.016 0.03150 Laminariajaponica 2.637 0.009 4.02 0.08 3517.9 154.1 0.984 0.05951 Laminariajaponica 1.27 0.07 1.96 0.05 3485.2 269.9 1.015 0.06852 Laminariajaponica 2.878 0.009 5.06 0.11 3078.9 132.4 1.075 0.06453 Laminariajaponica 11.87 0.04 11.32 0.10 5689.1 82.2 1.093 0.022

3.3.1HokkaidoandNorthernHonshu

TheReandOsabundanceinmacroalgaefromNorthernHonshuandHokkaido(Fig.1)

variesfrom0.6to11.87ppband1.96to16.16pptrespectively.The187Os/188Osand

187Re/188Oscompositionsrangefrom0.76to1.09and263.2to11234.52respectively.No

discerniblepatterncanbefoundbetweenReandOsabundanceandgeographicalposition.

77

a

c

b

d

78

Fig.2.Samplelocationsand187Os/188Osmacroalgaevalues(a),populationdensity(b),potentialOspointsourcelocations(c)andannualvehicleCO2emissionsint-CO2km-2yr-1(d)fortheTokyoBayarea.

The187Os/188Osand187Re/188Oscompositionofmacroalgaeisgenerallyhigherin

macroalgaefromHokkaido(avg.187Os/188Os=~1.03;avg.187Re/188Os=~5401)than

northernHonshu(avg.187Os/188Os=~0.85;avg.187Re/188Os=~1083).

3.3.2TokyoBay

TheReandOsabundanceinmacroalgaefromTokyoBay(Fig.2a)variesfrom0.02to7.65

ppband3.33to30.06pptrespectively.The187Os/188Osand187Re/188Oscompositionsrange

from0.36to0.95and28.7to4462.4respectively.Nodiscerniblepatterncanbefound

betweenReabundanceand187Re/188Oscompositionandgeographicalposition.Osmium

abundanceisgenerallyhigherinmacroalgaewithproximitytocentralTokyo(Samples5to

9).The187Os/188OscompositionofmacroalgaefromTokyoBayarehighlyvariable.

Relativelyunradiogenicvalues(~0.36to0.67)arefoundinmacroalgaefromthenorthern

andnorth-easternpartsofthebay,withconsistentlylow(~0.4)187Os/188Osvaluesnear

centralTokyo(Samples7to9).MacroalgaemoreproximaltothePacificOcean,possess

187Os/188Osvaluesthatbecomeprogressivelymoreradiogenic(~0.7to0.95).

ThedenselypopulatedcitiesofYokohama,TokyoandChibaoccupywestern,

northernandnorth-easternpartsofTokyoBayrespectively(Fig.2b).Alargenumberof

hospitalsoccupythewesternandnorthernpartsofthebay,whilemunicipalsolidwaste

incinerators(MSWIs)andsewagetreatmentplantsdominatepartsofthebaycloseto

centralTokyoandChiba(Fig.2c).MajorhighwaysconnectingthecitiesofYokohama,Tokyo

andChiba,generatehighannualvehicleCO2emissionsalongtheedgeofthenorthernpart

oftheBay(Fig.2d).Majorhighwaysandtrafficcongestionleadtoexceptionallyhigh

vehicleemissionsincentralTokyo(Fig.2d).

79

3.3.3OsakaBay

TheReandOsabundanceinmacroalgaefromOsakaBay(Fig.3a)variesfrom0.38to7.05

ppband11.36to122.76pptrespectively.The187Os/188Osand187Re/188Oscompositions

rangefrom0.17to0.76and114.5to1683respectively.Nodiscerniblepatterncanbe

foundbetweenReabundance,Osabundanceand187Re/188Oscompositionsand

geographicalposition.The187Os/188OscompositionofmacroalgaefromOsakaBayishighly

variable.SamplesalongtheeasterncoastofAwajiIslandvaryfromrelativelyunradiogenic

187Os/188Osvalues(0.16to0.45)towardsthesouth(Samplesite18)andthecentre(Sample

site17)torelativelyradiogenic187Os/188Osvalues(0.52to0.63)inthenorth(Samplesite18

and19).Themostradiogenic187Os/188Osvalues(~0.76)canbefoundclosetocentralOsaka

(Samplesite20),withrelativelyhomogenous187Os/188Osvaluesof~0.5occurringsouthof

Osaka(Samplesite21and22).

c

b

d

a

80

Fig.3.Samplelocationsand187Os/188Osmacroalgaevalues(a),populationdensity(b),

potentialOspointsourcelocations(c)andannualvehicleCO2emissionsint-CO2km-2yr-1

(d)fortheOsakaBayarea.SeeFig.2forpopulationdensityinformation.

ThedenselypopulatedmegacityofOsakaliestothenorth-eastofthebayand

thelargecityofKobeoccupiesthenortherncoast(Fig3b).Thesecitiesgenerallyrepresent

morethan4000peoplepersquarekm,whileAwajiIslandtothesouth-westhasa

populationlowerthan400peoplepersquarekm(Fig.3b).Alargenumberofhospitals,

MSWI,sewagetreatmentplantsandsteelmillsoccupytheeasterncoastlineofOsakaBay

(Fig.3c).Asmallsectionoftheeasternbayisoccupiedbyseverallargeoilrefineries(Black

circlesinFig.3c).ThreehospitalsareclusterednearcentralAwajiIslandwithafurthertwo

atthenortherntipoftheisland(Fig.3c).MajorhighwaysrunningtoOsakaandKobe

generatehighannualvehicleCO2emissionsalongthenorthernedgeoftheBay(Fig.3d).

ExceptionallyhighvehicleemissionscanbefoundincentralOsaka(Fig.3d).

3.3.4IseandMikawaBay

TheReandOsabundanceinmacroalgaefromIseandMikawaBay(Fig.4a)varyfrom0.16

to21.43ppband3to88.8pptrespectively.The187Os/188Osand187Re/188Oscompositions

rangefrom0.16to0.74and8.7to7521.2respectively.Nodiscerniblepatterncanbefound

betweenReabundance,Osabundance,and187Re/188Osandgeographicalposition.The

187Os/188OscompositionofmacroalgaefromIseandMikawaBayarehighlyvariable.

Moderatelyradiogenic187Os/188Osvalues(~0.51to0.54)arefoundalongthenorthern

coastlineofMikawaBayclosetothecitiesofGamagori(Sample29)andNishio(Sample

30).The187Os/188OsvaluesalongthesoutherncoastlineofMikawaBayaregenerallymore

radiogenic(~0.69to0.74)withtheexceptionofsample27(~0.54).IntheIseBay,

187Os/188Osvaluesarerelativelyunradiogenic(~0.16)closetothecityofMatsusaka(Sample

81

23),withmoreradiogenic187Os/188Osvalues(~0.6)closetothemouthofthebayandthe

PhilippineSea(Sample24).

b

cd

a

82

Fig.4.Samplelocationsand187Os/188Osmacroalgaevalues(a),populationdensity(b),

potentialOspointsourcelocations(c)andannualvehicleCO2emissionsint-CO2km-2yr-1

(d)fortheIseandMikawaBayareas.SeeFig.2forpopulationdensityandpointsource

information.

ThedenselypopulatedcitiesofGamagoriandNishiooccupythenortherncoast

ofMikawaBay,withToyohashitotheeast(Fig.4b).Thedenselypopulatedcityof

MatsuakaoccupiesthewesternsectionofIseBay(Fig.4b),withthemegacity,Nagoya,to

thenorthofthemap.SeveralhospitalsarespreadacrossthewesterncoastlineofIseBay

withamoredenselygroupinginToyohashi(Fig.4c).Severalsewagetreatmentplants

occupytheeasternandnorthwestregionsofMikawaBay(Fig.4c).VehicleCO2emissions

arerelativelyhighinToyohashiandalongthewesterncoastlineoftheIseBay(Fig.4d).

3.3.5IzuPeninsula

TheReandOsabundanceinmacroalgaefromtheIzuPeninsula(Fig.5a)varyfrom0.61to

4.83ppband7.78to28.1pptrespectively.The187Os/188Osand187Re/188Oscompositions

rangefrom0.69to0.91and122to2129.3respectively.Nodiscerniblepatternisobserved

betweenReabundance,Osabundance,and187Re/188Osandgeographicalposition.The

187Os/188OscompositionofmacroalgaefromtheIzuPeninsulaarevariable,butallrelatively

radiogenic(~0.7to0.91).Alongthesoutherntipofthepeninsula(SampleLocation32to

34)the187Os/188Osvaluesare~0.7,withslightlymoreradiogenicvaluesbeingfoundfurther

northonthewesterncoast(187Os/188Os=~0.84)andeasterncoast(187Os/188Os=0.78to

0.91).TheIzuPeninsulaisgenerallysparselypopulated,withthelargestcity(Numazu)to

thenorthwestofthepeninsula(Fig.5b).SeveralhospitalsandMSWIslinetheeasterncoast

ofthepeninsula(Fig.5c).VehicleCO2emissionsarelowacrosstheentirepeninsula,with

theexceptionofthecityofNumazutothenorth(Fig.5d).

83

Fig.5.Samplelocationsand187Os/188Osmacroalgaevalues(a),populationdensity(b),potentialOspointsourcelocations(c)andannualvehicleCO2emissionsint-CO2km-2yr-1(d)fortheIzuPeninsula.SeeFig.2forpopulationdensityinformation.

3.3.6NotoPeninsula

TheReandOsabundanceinmacroalgaefromtheNotoPeninsula(Fig.6a)varyfrom0.03

to3.18ppband2.29to32.8pptrespectively.The187Os/188Osand187Re/188Oscompositions

rangefrom0.46to1.02and70.3to1101respectively.TheReabundanceofmacroalgae

fromKanazawaandthewesterncoastoftheNotoPeninsula(SampleLocation37to39)is

b

c d

a

84

Fig.6.Samplelocationsand187Os/188Osmacroalgaevalues(a),populationdensity(b),

potentialOspointsourcelocations(c)andannualvehicleCO2emissionsint-CO2km-2yr-1

(d)fortheNotoPeninsula.SeeFig.2forpopulationdensityinformation

lower(0.03to0.61ppb)thanformacroalgaeofthenorthernandeasterncoastoftheNoto

Peninsula(1.6to3.18ppb).TheOsabundanceinmacroalgaeislower(2.29to6.13ppt)

ba

c d

85

nearthecityofKanazawa(SampleLocation37and38)thanformacroalgaeoftheNoto

Peninsula(12.2to32.8ppt).The187Re/188Oscompositionofmacroalgaefromthewestern

coastoftheNotoPeninsula(SampleLocation37to39)arelower(70.2to166)thanforthe

northernandeasterncoastoftheNotoPeninsula(279.1to1101).The187Os/188Os

compositionofmacroalgaefromtheIzuPeninsulaishighlyvariable.Relatively

unradiogenicvalues(0.46to0.48)arefoundclosetothemajorcityofKanazawa(Sample

Location37and38),whereassamplessurroundingtheNotoPeninsulaarerelatively

radiogenic(0.82to1.02).

TheNotoPeninsulaisgenerallysparselypopulated,withthelargestcity

(Kanazawa)tothesouthwest,andthecitiesofTakaoka,ImizuandToyamatothesouth

eastofthepeninsula(Fig.6b).Hospitals,MSWIsandsewagetreatmentplantsaregenerally

clusteredinmajorcities,withseveralhospitalsspreadaroundtheNotoPeninsula(Fig.6c).

VehicleCO2emissionsarelowacrosstheentirepeninsula,withtheexceptionofKanazawa

tothesouthwestandTakaokaandToyamatothesoutheast(Fig.6d).

3.4Discussion

Thelargerangeinthe187Os/188Oscompositionofmacroalgae(0.16to1.09)impliesthat

bothhighlyradiogenicandhighlyunradiogenicsourcesofOscontributetothebulkOsin

macroalgae.NaturalsourcesofradiogenicOsincludeseawater,chemicalweatheringof

radiogeniccontinentalcrustandthedepositionofAeoliandust,whileanthropogenic

sourcesincludethesmeltingofradiogenicbase-metalsulphideoresandcontamination

fromfossilfuels.NaturalsourcesofunradiogenicOsincludehydrothermalalterationof

oceaniccrust,volcanism,cosmicdustandthechemicalweatheringofjuvenilebasaltic

crust,whereasanthropogenicsourcesinvolveexhaustfromautomobilesandthe

processingofPGEores,chromitesandunradiogenicbase-metalsulphideores.Inthe

followingsectionswewillfirstdiscusstheenvironmentalandbiologicalinfluenceonReand

86

Osuptakeinmacroalgaebeforediscussingeachofthesourcespresentedabovetoinfer

thattheOsproducedbyhumanactivitydominatestheisotopiccompositionofcoastal

watersinproximitytomajorJapanesecities.

3.4.1Biologicalandenvironmentalcontrolsonthe187Re/188Osofmacroalgae

The187Re/188Osofmacroalgaeshowawiderange8.7to11234.5(Table2)similarto

previousstudies(6.8to40,320)(Racionero-Gómezetal.,2017;Chapter2;Rooneyetal.,

2016).ThismaybeduetothedifferentsourcesofReandOstomacroalgae.Whenthe

187Re/188Osofmacroalgaeisplottedagainstthe187Os/188Osofmacroalgae(Fig.7a),most

Japaneseseaweedsshowalargevariationin187Os/188Osforasmallrangein187Re/188Os

whencomparedtoIcelandicmacroalgae(greycirclesinFig.7a).Thismaysuggestthat

JapanlacksasourcehighinReandthereforehigh187Re/188Osvalues,suchasoldprimary

basalticminerals(SeeChapter2;Gannounetal.,2006).Insteadthemacroalgaeappearto

bedominatedbysourceslowinResuchasriverwater(avg.187Re/188Os=227;Peucker-

EhrenbrinkandRavizza,2000).The187Re/188Osratiosincreaseasriverinesourcesmixwith

seawaterwithhigher187Re/188Os(avg.187Re/188Os=4270;Peucker-EhrenbrinkandRavizza,

2000).ThisisparticularlydominantinHokkaido(blacksquaresinFig.7a)wheremacroalgae

fromdeeperwatershaveastrongerinfluencefromPacificOceanseawater,andthus

higher187Re/188Osvalues(Fig.7a).

Whenthe187Re/188OsofmacroalgaeisplottedagainstthereciprocaloftheRe

concentration(Fig.7b)weseeasimilarrelationshiptoIcelandicmacroalgaefromChapter2

(greycirclesinFig.7b).ThissuggestssimilarmechanismscouldcontrolReandOsuptakein

Japanesemacroalgae.IfReandOsaretakenupviathesamepathwayitwillleadto

competitionbetweentheseelementsandthereforelowerReconcentrationinmacroalgae

underhigherOsseawaterconcentrationandviceversa(Chapter2;Racionero-Gómezetal.,

2017).AtlowReconcentration,thereisnocompetitionbetweenReandOsforuptakeinto

87

macroalgae,andthe187Re/188Osratiosremainconsistentlylow.AstheReconcentration

rises,ReisfavouredoverOsasthetwoelementsbegintocompeteforuptakeinto

macroalgae,leadingtoanincreaseinthe187Re/188Osratios.AtexceptionallyhighRe

concentrations,RecontinuestobetakenupandthemacroalgaebecomesenrichedinRe,

leadingtoexceptionallyhigh187Re/188Osratiosinmacroalgae(Fig.7b).

y=445.66x-0.917R²=0.82191

0.00

0.01

0.10

1.00

10.00

100.00

0.01 0.10 1.00 10.00 100.00

187 Re/

188 Os

x103

1/Re(ppb)

0

5

10

15

20

25

30

35

40

45

0.0 0.2 0.4 0.6 0.8 1.0 1.2

187 Re/

188 Os

x103

187Os/188Os

TokyoBayBosoPeninsulaOsakaBayIseBayMikawaBayIzuPeninsulaKanazawaNotoPeninsulaKanazawaHokkaidoIceland

a

b

88

Fig.7.a)The187Re/188Osagainstthe187Os/188OsofmacroalgaefromIceland(Chapter2)andJapan(Chapter3).b)187Re/188OsratiosagainstthereciprocaloftheReconcentrationforJapanese(blacksquares)andIcelandic(greycircles)macroalgae.

3.4.2Naturalsourcesofosmiumtomacroalgae

Alargenumberofmacroalgaeinhabitbrackishwaters,andtheir187Re/188Osand187Os/188Os

compositionwillthereforerepresentamixingbetweenfreshwaterriverineinputsand

seawater(SeeChapter2).The187Os/188OscompositionandOsabundanceofadepthprofile

fromtheEastPacificOceanhasbeenconstrainedat~1.04and10ppqrespectively(Chen

andSharma,2009;GannounandBurton,2014;Woodhouseetal.,1999).Further,the

hydrogenous187Os/188Oscompositionofpresentdaymarinesedimentssuggestthatthe

SeaofJapanhasasimilar187Os/188Oscompositionof~1.03(Dalaietal.,2005).

Fig.8.Osmiumisotopecomposition(187Os/188Os)ofmacroalgaefromHokkaidosamplelocations(SeeFig.1).BlacklineshowsdatafromdirectPacificOceanseawatermeasurements(SeeGannounandBurton,2014).

These187Os/188Osvaluesarereflectedinthe187Os/188Oscompositions(Table2)of

macroalgaefarmedfromdeeperpristinewaters>2milesofftheeastcoastofHokkaido

89

(~1.02),whichisindistinguishable,withinuncertainty,fromOsisotopemeasurements

obtaineddirectlyfromseawater(Fig.8).The187Os/188Oscompositionofmacroalgaefrom

shallowercoastalwatersaroundJapanwillrepresentthemixingbetweenPacificseawater

values(~1.04)andlocalcontinentalriverineinputswithadiverserangeof187Os/188Os

compositionsandOsabundancesdependentontheunderlyinglithologybeingweathered.

ThelongchainofJapaneseislandsalongthenorthwestmarginofthePacific

Oceanconsistsoftwotrench-arcsystems:theEastJapanIslandArcwhichmarksthe

boundarybetweentheEurasian,PacificandPhilippineSeaplates;andtheWestJapan

IslandArcwhichrepresentstheboundarybetweentheEurasianandPhilippineSeaplates

(Hashimoto,1991).Thesefeaturesgenerateacomplexgeologythatcanbebrokendown

intofivemajorelements:poorlylithifiedNeogene-QuaternarysedimentsandPaleogene

sedimentaryrocks(~40%);accretionarycomplexesconsistingmainlyofmelange,

mudstoneandsandstone(~17%);non-alkalineNeogene-Quaternaryvolcanicrocks(23%);

graniticrocksintrudedduringtheCretaceous(~10%);and,metamorphicrocks(~4%).The

underlyingingeologyofJapanthereforerepresentsalargerangeofagesandsourcesi.e.

mantletouppercrust.

Parent-daughterfractionationoftheRe-Ossystemduringmantlemeltingleads

toRebecomingpreferentiallypartitionedintothemeltoverOs.Asaproductofmantle

differentiation,continentalcrustischaracterisedbyhigherRe/Osratiosrelativetothe

mantle.Withanaveragecontinentalcrustageof~2.2Gyr,insitudecayof187Rehasledto

theaccumulationofappreciable187Osandaradiogenic187Os/188Oscompositionof~1.4

(Peucker-EhrenbrinkandJahn,2001).Theerosionandsubsequentweatheringofcomplex

underlyinglithologieshasledtohighlyvariable187Os/188Osvalues(0.64to2.94)andOs

abundances(4.6to52.1ppq)inglobalrivers(Levasseuretal.,1999a).Themajorcities

studiedhere,andmorethan70%oftheJapanesepopulation,resideontheJapanese

‘plains’whicharelargelyunderlinedbyQuaternarysediments(Hashimoto,1991).Itis

90

thereforeexpectedthatthe187Os/188Oscompositionofriversdrainingtheseregionstobe

similartotheglobalriverineaverageof~1.54(Levasseuretal.,1999a).

Pleistocene-Holocenesedimentsandsedimentaryrocksoccupytheentire

drainagebasinoftheKantoplains(Ohtaetal.,2011),Osakaplains(Ohtaetal.,2005;Ohta

etal.,2007),Kanazawaplains(Ohtaetal.,2004)andIseplains(Ohtaetal.,2005;Ohtaet

al.,2007),andthereforerepresentthedominantcontrolonthechemicalcompositionof

riversdrainingtherelevantwatershed(Ohtaetal.,2005).Wewouldthereforeexpectthe

187Os/188OsofwatersinTokyoBay(Fig.2)torepresentamixturebetweenradiogenicrivers

(~1.54)drainingthecoastalplainsandPacificseawater(~1.04)entrainedintothebayfrom

thePacificOceanviaestuarinegravitationalcirculation.However,the187Os/188Osof

macroalgaefromTokyoBay(Fig.2a),OsakaBay(Fig.3a),Kanazawa(Fig.6a)andIseBay

(Fig.4a)arerelativelyunradiogenicandrangefrom0.36to0.95,0.17to0.76,0.16to0.6

and0.46to0.48(Table2).ThissuggeststhatOsfromnaturalsourceshasverylittle

influenceontheisotopiccompositionofTokyoBay.

Althoughmostregionsstudiedareunderlainbysedimentaryplains,exceptions

tothisincludetheRyokebelt,IzupeninsulaandNotoPeninula.TheRyokebeltis

characterisedbylowP/Ttypemetamorphismandextensivefelsicigneousactivitythat

coversa~800kmstretchofinnerSouthwestJapan(Nakajima,1994).Graniticrocksare

dominantovermetamorphicrocksthroughouttheRyokebelt,andwerederivedfrom

magmasproducedinthelowercrustand/oruppermantle;withpotentialassimilationof

metamorphicrocksorPrecambriancrust(Yuharaetal.,2000).Weatheringofgranitic

materialdominatescentralAwajiIsland,KobecityalongthenorthcoastofOsakaBayand

theYahagiRiver,whichflowsintonorthernMikawaBay(Ohtaetal.,2005;Ohtaetal.,

2007).TheweatheringofhighpressuremetamorphicrocksdominatestheToyoRiver

whichflowsintotheeasternMikawaBay(Ohtaetal.,2005;Ohtaetal.,2007).Moststudies

suggesttheweatheringofold(Precambrian)graniticterraindelivershighlyradiogenicOs

91

(187Os/188Os=≥2.5)toriversdrainingthesecatchments(Chenetal.,2006;Ehrenbrinkand

Ravizza,1996;Huhetal.,2004).However,theRyokeBeltismadeupofmorerecent

(Cretaceous)I-typegraniteswithsomesedimentarycomponentsoftheMinoTerrain

(IshiharaandWu,2001).Althoughhardtoestimatewithoutdirectmeasurements,thelow

OsconcentrationcombinedwithhighRe/Osratiosingranites(Johnsonetal.,1996),

combinedwithpossiblecrustalcontaminationofradiogenicOswouldleadtohigh

187Os/188OsvaluesinsilicicmaterialsincetheCretaceous(Alvesetal.,1999;Alvesetal.,

2002;Hartetal.,2003;Hartetal.,2002).Wethereforesuggestthatrelativelyunradiogenic

187Os/188OsvaluesofmacroalgaefromOsakaBay(0.17to0.76)andMikawaBay(0.51to

0.74)arenotrelatedtotheweatheringoflocalgranites.

The187Os/188Oscompositionofmacroalgaefromthesouthandeastcoastofthe

IzuPeninsula(SampleLocation31to36;Fig.5a)andthecoastoftheNotoPeninsula

(SampleLocation39to43;Fig.6a)rangefrom0.69to0.91andfrom0.82to1.02

respectively.Thesevaluesareconsistentlylowerthanthe187Os/188Oscompositionof

seawater(~1.04),suggestingtheinfluenceofanunradiogenicendmember.Lowhuman

activity(Fig.5bandFig.6b)andtheabsenceofsubstantialanthropogenicsourcesofOsin

theregionsstudied(Fig.5c,dandFig.6c,d)suggeststhatthedominantsourceof

unradiogenicOsislikelytobenatural.FormationsexposedontheIzuPeninsulaarealmost

entirelycomposedofsubmarineandterrestrialvolcanicseruptedsincetheearlyMiocene,

andtheirreworkeddeposits(KoyamaandUmino,1991).Thenorth-easttipofthe

peninsulaisdominatedbyrecent(0-0.6Ma)HigashiizuandShiofukibasaltictoandesitic

lavasandpyroclastics.ThiscontrastswiththesouthernIzuPeninsula,whichisdominated

byolder(MiddleMiocenetoEarlyPleistocene)basalticandandesiticvolcanicsfromthe

YugashimaandShirahamagroup.Mioceneandesiticlavaandsedimentaryrocksfromthe

IwaineandKurosendaidominatethelithologyoftheNotopeninsula(Japan,1992).

92

Weatheringofthesevolcanicsystemsdominatesstreamandcoastalsediments

emanatingfromthem(Ohtaetal.,2005;Ohtaetal.,2004;Ohtaetal.,2007).Wewould

thereforeexpectthe187Os/188Oscompositionofriversintheseregionstohaveamore

mantlesignature(~0.12)duetotheweatheringofbasalticmaterialdominatinglocal

catchmentareas(Gannounetal.,2006).However,thedecayofabundant187Reintheolder

YugashimaandShirahamagroupsandtheNotoPeninsulamayleadtoradiogenicingrowth

of187Osandthereforehigher187Os/188Oscompositionsintheseregions(Gannounetal.,

2004;Gannounetal.,2006).The187Os/188OscompositionsofmacroalgaefromtheIzu

coastalwatersthereforeconstitutesamixingbetweenradiogenicPacificseawater(~1.04)

andunradiogenicriversdrainingabasalticterrain(>0.12).Thisissupportedbythe

187Os/188OscompositionofmarinesedimentsfromtheYasakaestuaryinnorthernKyushu,

whichrangefrom0.67to0.73,andrepresentamixtureoflocallyweatheredMiocene

volcanicsandseawater(Zhengetal.,2014).

3.4.3Anthropogenicsourcesofosmiumtomacroalgae

AnthropogenicOshasbeendetectedinestuaries(Turekianetal.,2007;Williamsetal.,

1997),lakes(Rauchetal.,2004),coastalsediments(EsserandTurekian,1993;Ravizzaand

Bothner,1996),biologicalorganisms(Rodushkinetal.,2007a,b),airborneparticles(Rauch

etal.,2005)andprecipitation(Chenetal.,2009).SourcesofanthropogenicOsinclude

hospitals(EsserandTurekian,1993;Turekianetal.,2007;Williamsetal.,1997),MSWIs

(Funarietal.,2016),vehicleexhaust(PoirierandGariépy,2005)andsmelters(Chenetal.,

2009;Rodushkinetal.,2007b),andcanbeeitherdirectlyintroducedintoriversandcoastal

watersviasewageoutflow(EsserandTurekian,1993;RavizzaandBothner,1996;Turekian

etal.,2007;Williamsetal.,1997)ordirectlytotheatmosphereasvolatileOsO4during

high-temperatureprocesses(Smith,1974).The187Os/188OscompositionofOsfromthese

sourcesdependsontheoriginalmaterialusedintheircreation.The187Os/188Os

93

compositionoffossilfuelsarehighlyradiogenic,rangingfrom1to13.7(Finlayetal.,2011;

SelbyandCreaser,2005).ThePGEsulphideoresandchromiteshaveunradiogenic

187Os/188Osvaluesof0.15to0.2(McCandlessandRuiz,1991)and0.12to0.14(Walkeret

al.,2002)respectively.The187Os/188Oscompositionofbase-metalsulphidedepositsare

highlyvariable:rangingfromvaluessimilartoPGEores(Lambertetal.,1998;Walkeretal.,

1994),throughtohighlyradiogenicvaluescausedbycontaminationfromcontinentalcrust

duringtheircreation(Morganetal.,2002).

Osmiumhasmajoruses–asatissuestainforelectronmicroscopyandasa

catalystinsteroidsynthesis–inmedicalresearch(Smith,1974).AnthropogenicOsfrom

biologicalandmedicalresearchlaboratorieshasbeendetectedintheHudsonRiver-Long

IslandSoundestuarinesystemandChesapeakeBaywithanestimated187Os/188Os

compositionof~0.13(EsserandTurekian,1993;Helzetal.,2000;Turekianetal.,2007;

Williamsetal.,1997).Thedominantmodeoftransportisbelievedtobeseweroutflow

fromnearbyhospitalsandatmospherictransportfromhospitalincinerators,theinfluence

ofwhichcanbedetectedupto70kmfromthesource(EsserandTurekian,1993;Ravizza

andBothner,1996;Williamsetal.,1997).Althoughnotmeasuredhere,sewagefrom

outflowsinNewHaven(EsserandTurekian,1993),NewYorkCity(Williamsetal.,1997)

andBoston(RavizzaandBothner,1996)have187Os/188OscompositionsandOsabundances

ofbetween0.15to0.3and0.57to4.01ppbrespectively.

DenselypopulatedregionsofJapan(SeepanelbinFigs.2to6)aregenerally

servedbyalargenumberofmajorhospitals(SeeredcirclesinpanelcofFigs.2to6).We

wouldthereforeexpectasignificantinfluxofunradiogenicOsfromhospitalssituatedin

centralTokyoandYokohama(Fig.2c),centralOsaka(Fig.3c),westernIseBayandeastern

MikawaBay(Fig.4c)andKanazawa(Fig.6c).The187Os/188Oscompositionofmacroalgae

fromtheseregionsaregenerallyhighlyunradiogenicrangingfrom0.39to0.55nearcentral

TokyoandYokohama,0.5to0.76nearOsaka,0.16to0.19inwesternIseBay,0.51to0.54

94

innorthernMikawaBayand0.46to4.48nearKanazawa(Table2).Additionally,

macroalgaesamplesfromcentralAwajiIsland(SampleLocation17)showhighly

unradiogenic187Os/188Osvaluesof0.17to0.38,closetothepreviouslyrecorded187Os/188Os

ofsewage(0.15to0.3).Thisismostlikelyduethehighnumberofhospitalsinthenearby

cityofSumototothesouthorthesewagetreatmentplanttothenorth(Fig.3c).This

suggeststhatthemacroalgaehaveincorporatedOsrelatedtomedicalresearchfrompoint

sourcessuchashospitalsandsewagetreatmentplantseitherviasewageoutflowor

incinerationofmedicalwaste(Fig.9).However,despitetheubiquitousnatureofhospitals

inmajorcitiese.g.TokyoandOsaka,the187Os/188Oscompositionofmacroalgaefromthese

regionsarenotuniformlylow.Forexample,macroalgaeinTokyoBayshowanincreaseto

moreradiogenicvalueswestwardsfromcentralTokyo,despitetheubiquityofhospitalsin

thisregion.Meanwhile,themacroalgaeclosesttocentralOsaka(SampleLocation20)has

thehighest187Os/188Osvalues(0.76)forOsakaBay.Thissuggeststhatalthoughmedical

facilitiescanofferapointsourceforanthropogenicOse.g.AwajiIsland,other

anthropogenicsourcescandominateinotherregions.

Fig.9.MacroalgaeOsconcentration(filledcircles)andisotopiccomposition(opencircles)

withdistancefromhospitalsincentralAwajiIsland(redcircles)oranoilrefineryincentral

Osaka(green).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0

10

20

30

40

50

60

70

80

0 5 10 15 20 25 30 35 40

187 Os/

188 Os

[Os](p

pt)

Distancefromsource(km)

95

MSWIsareconsideredoneofthebestoptionsformunicipalsolidwaste

managementinindustrialisedcountries(Chandleretal.,1997)andarewidelyusedin

majorJapanesecities.AnalysisofbottomandflyashfromItalianMSWIsshowawiderange

of187Os/188Oscompositions(0.24to0.7)andconcentrations(0.026to1.65ppb),andithas

beenpredictedthattheOsreleasedfromMSWIsmokestackswillrangefrom16to38ng

Os/m2/yr(Funarietal.,2016).ThisisthereforeanimportantsourceofOsnearcentral

TokyoandOsaka(SeelightgreencirclesinFig.2cand3c).Theseadditionalinputsof

potentiallyunradiogenicOscouldcausetherelativelyunradiogenic187Os/188Osvaluesin

regionswithrelativelylowvehicleemissionsi.e.Yokosuka(SampleLocation10),Chiba

(SampleLocation1,2and5),Kanazawa(SampleLocation38)andthesouthcoastofOsaka

Bay(SampleLocation21and22).

Automobilecatalyticconvertorscouldpotentiallyprovidealarger,regionally

dispersed,sourceofanthropogenicOs.Directmeasurementsofcatalyticconvertorsyield

unradiogenic187Os/188OsvaluesandOsabundancesof0.1to0.2and6to228ppt

respectively(PoirierandGariépy,2005).Moreover,newcatalyticconvertorscouldbe

responsibleforasmuchas120pgOs/m2inthefirstyearoftheirlife(PoirierandGariépy,

2005),andanthropogenicOssourcedfromvehicleexhausthasbeendetectedinairborne

particles(Rauchetal.,2005)andglobalprecipitation(Chenetal.,2009).Asthemajorityof

PGEsaresourcedfromtheBushveldComplex(SouthAfrica)andNoril’sk(Russia),we

expectcatalyticconvertorsinJapantohaveunradiogenicvaluesof0.15to0.2(Jones,

1999).HerewehaveutilisedtotalvehicleCO2emissionsfor2015fromtheEastAsianAir

PollutantEmissionGrid(EAGrid2011)database(Kannarietal.,2007)totryanddetermine

theinfluenceofcatalyticconvertorsonregionalvariationsinanthropogenicOs(Seepanel

dinFigs.2to6).

Wecangenerallyseethatinregionsofhighvehicleemissions,wegetthemost

unradiogenic187Os/188Osvaluesinmacroalgae.The187Os/188Oscompositionofmacroalgae

96

incentralTokyorangefrom0.42to0.46.Thesevaluesareindistinguishablefromthe

hydrogenousloadofpreviouslyrecordedmarinesedimentsfromtheTamaRiver(~0.3to

0.4)(Zhengetal.,2014),coincidenttowherethehighestvehicleemissionsinJapanare

found(Fig.2d).The187Os/188Oscompositionofmacroalgaebecomesmoreradiogenic(0.55

to0.67)inregionsofintermediatevehicleemissionssuchasYokosukaandChiba,withthe

exceptionofSampleLocation4whichishighlyunradiogenic(0.36).The187Os/188Os

compositionofmacroalgaebeginstoreachmoreradiogenic187Os/188Osvalues(0.85to

0.95),closetoPacificOceanestimates(~1.04),inregionsoflowvehicleemissionsonthe

BosoPeninsula(Fig.2d).InOsakaBay,macroalgae187Os/188Osvaluesarerelatively

unradiogenic(0.5to0.59)inregionsofintermediatevehicleemissions(Fig.3d).However,

macroalgaefromSampleLocation20and17wherevehicleemissionsarehighestand

lowestrespectively,recordthemostradiogenic187Os/188Osvalues(0.76)andunradiogenic

values(0.17to0.38)respectively.Inthecaseofsamplelocation17,aspreviously

explained,localinfluencefromhospitalandsewageoutflowwillcontributeOswithan

unradiogenic187Os/188Oscompositiontothisregion.Meanwhile,samplelocation20isclose

toanoilrefinery(Fig.3c),whichcouldactasapointsourceofanthropogenicOs.Osmium

inoilisgenerallyhighlyradiogenic(187Os/188Os=1to13.7)(Finlayetal.,2011;Selbyand

Creaser,2005;Selbyetal.,2007)andanyoilspillsoratmosphericemissionwilltherefore

carrysimilarcompositions.Thisisthereforethemostlikelycauseofradiogenic187Os/188Os

values(0.76)inthisarea(Fig.9).MikawaandIseBayshowsimilartrendswithrelatively

unradiogenicvaluesclosetotheregionsofhighestvehicleemissions(Fig.4d).

Finally,otherpotentialpointsourcesofanthropogenicOsincludetheprocessing

ofchromites,PGEoresandbase-metalsulphideores.AlthoughPGEoreandchromite

smeltersarenotknowntoexistintheregionsstudiedhere,awidenumberofsteelmills

arelocatedintheJapaneseindustrialcentressuchasTokyoandOsaka(Fig.2cand3c).The

187Os/188Oscompositionofbase-metalsulphidedepositsarehighlyvariable:rangingfrom

97

unradiogenicmantlevalueslessthan0.3(Lambertetal.,1998;Walkeretal.,1994),

throughtohighlyradiogenicvaluesgreaterthan0.9(Morganetal.,2002).Notknowingthe

sourceoforeusedatJapanesesteelmills,andduetothelackofstudiesofemissions,itis

hardtoestimatetheinfluenceofthesesourcesonregionalfluctuationsatmosphericOs.

However,itshouldbenotedthatsteelmillscouldpotentiallyactasasignificantsourceof

anthropogenicOswithpotentiallyhighlyvariableisotopiccomposition.

00.10.20.30.40.50.60.70.80.91

0 2000 4000 6000 8000 10000

187 O

s/18

8 Os

PopulationDensity(people/km2)

y=-0.097ln(x)+1.3351R²=0.84271

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 2000 4000 6000 8000 10000

187 O

s/18

8 Os

PopulationDensity(people/km2)

a

b

98

Fig.10.187Os/188Osagainstpopulationdensityfora400km2areasurroundingeach

samplefor(a)populatedbaysstudieshere(Mikawa,Ise,OsakaandTokyoBay)and(b)

TokyoBay.Blacksymbolsrepresentmacroalgaeaffectedbypointsourcessuchashospitals

andoilrefineries(SeeFig.9).

Inconclusion,duringthisstudywehavefoundawiderangeofnaturaland

anthropogenicsourcestoJapanesewaters.InregionssuchasHokkaido,northernHonshu,

theIzuPeninsula,andNotoPeninsulawherehumanactivityislow,theosmiumisotope

budgetofcoastalwatersisdominatedbynaturalOsfromPacificOceanseawaterand

riverinesourcesdominatedbytheweatheringoflocalrocktypessuchasvolcanic,granitic

andsedimentaryrocks.IndenselypopulatedregionssuchasTokyo,Yokohama,Chiba,

Osaka,Kanazawa,MatsuakaandToyohashi,anthropogenicsourcesofOsdominate.

OsmiumfromcatalyticconvertorsdominatestheOsisotopebudgetincoastalwaterin

theseregions.However,pointsourcescanbecomemoreinfluentialinsampleslocated

nearhospitals,sewagetreatmentplantsandMSWIs.Thisisbestshownwhenthe

187Os/188Oscompositionofmacroalgaeisplottedagainstpopulationdensity(Fig.10).In

areasoflowpopulationdensity,the187Os/188Osofmacroalgaerangesfrom0.6to1(Fig.

10a).However,aspopulationdensityincreases,theisotopiccompositionofmacroalgae

becomesmoreunradiogenic(0.6to0.4).Pointsourcessuchashospitalsandoilrefineries

pulltheisotopiccompositionofmacroalgaetowardstheirrespectivesourcecompositions

awayfromthistrend(BlackcirclesinFig.10a).Thisrelationshipbecomesmoreapparent

whenthe187Os/188OsofmacroalgaeisplottedagainstpopulationdensityforTokyoBay

(Fig.10b).MacroalgaeinthelesspopulatedregionoftheBosoPeninsulahaveanisotopic

compositionof~0.9.However,asyoumoveintothebay,andtowardsthemorepopulated

regionstotheNorthoftheBay,theisotopiccompositionbecomesmoreunradiogenic

(0.4).

99

3.4.4Anthropogenicinfluenceontheglobalosmiumcycle

3.4.4.1AnthropogenicimpactonJapanesecoastalwaters

Thisstudyhasshownthatthe187Os/188Oscompositionofmacroalgaecansuccessfullytrace

thefluctuationsinnaturalandanthropogenicprocessesaroundJapan.Whenthe

187Os/188Oscompositionofmacroalgaeisplottedagainstthereciprocalofthe

concentration,allofthedatafallwithinfielddelimitedbyfourpotentialend-members(Fig.

9):seawater(radiogenic187Os/188Os,intermediate[Os]);riverwaterdrainingQuaternary

sedimentarymaterial(radiogenic187Os/188Os,low[Os]);riverwaterdrainingMiocene

volcanicrocks(intermediate187Os/188Os,low[Os]);and,anthropogenicOswithaPGE

source(unradiogenic187Os/188Os,high[Os]).Variationsinthe187Os/188Osofmacroalgaecan

thereforebeexplainedbythemixingbetweenthesedistinctsources.

Fig.11.187Os/188OsratiosagainstthereciprocaloftheOsconcentrationforJapanese(Chapter3)andIcelandic(Chapter2)macroalgae.SeeFig.7bforkey.

HighlypopulatedregionssuchasTokyoBay(redsquares),OsakaBay(dark

orangesquares),IseBay(lightorangesquares)andKanazawa(yellowsquares)fallona

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0 0.1 0.2 0.3 0.4 0.5 0.6

187 Os/

188 Os

1/[Os](ppt)

100

mixinglinebetweenradiogenicseawaterand/orriverwaterandanthropogenicsources

(Fig.11).Meanwhile,sparselypopulatedregionssuchasHokkaido(blacksquares),the

BosoPeninsula(lightbluesquares),theIzuPeninsula(darkgreensquares)andnorthern

Honshu(lightbluesquares)fallonamixinglinebetweenradiogenicseawaterand

radiogenicorrelativelyunradiogenicriverwater(Fig.11).Thissuggeststhathumanactivity

hasaddedanadditionalsourceofOstotheJapanesemarineOsbudgetandisdrivingthe

systemtowardsunradiogenicvalues.Thishascausedthe187Os/188Osprofileofarelatively

oldislandarclikeJapan,totakeonthe187Os/188Osprofileofayoungmid-oceanriftsystem

likeIceland(greycirclesinFig.11).

3.4.4.2AnthropogeniccontributionsofosmiumfromJapan

Osmiumoxidebecomesvolatileatacatalyticconvertorsoperatingtemperature(>400°C),

releasing75-95%ofitsOstotheatmospherewithinthefirstyearofuse(Poirierand

Gariépy,2005).Ithasbeenestimatedthata1kgmonolithiccatalystcontainsbetween6

and228pptOswhichwillbereleasedduringitslifetime(PoirierandGariépy,2005).Given

apopulationof127millionpeopleinJapan,androughly5.5millionnewvehiclesonthe

roadinJapanin2015,weestimateroughly23peoplepercar.Givenapopulationof~15.5

M,~4.2Mand~9.3MpeoplefortheTokyo-Kawasaki-Yokohama-chiba,Osaka-Kobeand

Mie-Aichiimetropolitanregionsrespectively,thisamountsto~675,000,~182,000and

~400,000vehiclesrespectively.Ifweassumeeachcarhasa1kgmonolithiccatalyst,this

amountsto3to105,1.5to54and0.2to8.4pgOs/m2/yrrespectively.

ThisiscomparabletopreviousOsemissionestimatesforNewYorkCityof3to

126pgOs/m2/yr(PoirierandGariépy,2005).ThissuggeststhatOsemissionfromcatalytic

convertorsaresignificantinurbanareaswhencomparedtonaturalOsinputsof~810pg

Os/m2/yr(PoirierandGariépy,2005)fromcontinentalerosionandatmosphericdustinputs

of1pgOs/m2/yr(WilliamsandTurekian,2002).However,pointsourceOsemissions,such

101

asMSWIs(16to38ngOs/m2/yr),canbecomefarmoresignificantthenvehicleemissionsin

denselypopulatedregions(Funarietal.,2016).

3.4.4.3Impactofanthropogenicosmiumonsurfacewaters

Contemporarysurfaceseawaterhasalower187Os/188Oscomposition(~0.95)thandeep

waters,whichhasbeenattributedtohumanactivities(Chenetal.,2009;Levasseuretal.,

1999b).Ifweassumethatsurfacewatersatthemouthofabay(M)representsthemixing

betweenananthropogenicsourceofOs(A)comingfromthebayandanaturalsourceofOs

comingfromseawater(SW),wecandescribetheisotopiccompositionandconcentration

atthemouthusingthefollowingequation:

187Os/188OsM Os SW+ Os A = 187Os/188OsSW[Os]SW+187Os/188OsA[Os]A

Wecanthenrearrangetheaboveequationtodeterminetheconcentrationof

anthropogenicosmium:

Os A =Os SW(187Os/188OsSW-187Os/188OsM)

187Os/188OsM − 187Os/188OsA

Ifweassumethattheisotopiccompositionofmacroalgaefromthemouthofthe

bayrepresents187Os/188OsM(187Os/188OsTokyoBay=0.86;187Os/188OsOsakaBay=0.47;

187Os/188OsIseBay=0.6),andgivena187Os/188OsSWand[Os]SWforthePacificOcean

(187Os/188OsPacific=1.04;[Os]Pacific=8ppq)andan187Os/188OsArepresentativeofPGEores

(0.15)weestimatetheconcentrationofanthropogenicOsinsurfacewatersatthemouth

ofTokyo,OsakaandIseBaystobe2ppq,14ppqand8ppqrespectively.Givenaflowrate

of8x109m3/yr,8x109m3/yrand37x109m3/yrforsurfacewatersleavingTokyo,Osaka

102

andIseBayrespectively(SmithandYanagi,1997),weestimateatotalofapproximately4.2

kgOs/yrofanthropogenicOsisdeliveredtothePacificOceanfromthesethreebays.

Usinganaveragevalueof1.4kgOs/yrdeliveredbyeachJapanesemegacityand

consideringtherearecurrently16megacitiesgloballysituatedincoastalregions,this

equatesto22.6kgOs/yrdeliveredfromglobalcoastalcities.However,coastalmegacities

onlyrepresent242millionpeopleor~8.2%ofpeoplelivingwithinthecoastalregion

(PellingandBlackburn,2014).Therefore,apotential276kgofanthropogenicOscouldbe

deliveredviasurfacewaterstotheworld’soceanseveryyearifweextrapolatetotheentire

globalcoastalpopulation.Chenetal.(2009)calculated2,391kg/yrarerequiredtodrive

theOsisotopiccompositionofocean’ssurfacewaters(depth=200m)from1.05to0.95.

TheysuggestedthatthemajorityofthisanthropogenicOsisdeliveredtotheatmosphere

duringtherefiningofPGEores.However,thisstudysuggests~12%ofanthropogenicOsin

surfacewaterscanbeattributedmainlytotheuseofPGEsincatalyticconvertors,and

confirmshumanactivityisimpactingtheglobalOsbudget.

3.5Implicationsandfutureoutlook

TheRe-Osdatapresentedheresuggestsmacroalgaecansuccessfullytracetheimpactof

environmentalandhumanactivityontheglobalosmiumbudget.Itparticular,the

187Os/188OscompositionofmacroalgaehastracedfluctuationsinnaturalOsfromthe

weatheringofvolcanicsandsedimentarymaterial,anditsmixingwithseawaterfromthe

PacificOceanandtheSeaofJapaninbrackishwatersaroundHokkaido,northernHonshu,

theIzuPeninsulaandNotoPeninsula.Moreover,the187Os/188Oscompositionof

macroalgaehasalsosuccessfullytracedtheimpactofanthropogenicOsemissionsfromthe

useofcatalyticconvertors,MSWIs,hospitalsandrefineriesonwatersinTokyoBay,Osaka

BayandIse/MikawaBayneardenselypopulatedregionsofJapan.Thissupportsprevious

workthatindicatesOsisotopescanactasapowerfultracerofearthsurfaceprocessessuch

103

asbasalticweatheringinIceland(Chapter2)andcontinentalweatheringinGreenland

(Rooneyetal.,2016).

Thewidespreaduseofcatalyticconvertorshasledtotheremovalofarangeof

pollutantsfromtheenvironment.Ironically,thishasledtoglobalOspollutionattheEarth’s

surfacefromtheproductionofplatinumforuseincatalyticconvertors.Unlikeprevious

studies,wesuggestthattherefineryofPGEsdoesnothaveasignificantinfluenceonthe

JapanesemarineOsbudget.Instead,thewidespreaduseofOsinmedicalresearchandthe

lackoftreatmentinsewageandmunicipalwastehasledtosignificantcontaminationin

regionsdirectlyadjacenttohospitals,MSWIsandsewagetreatmentplants.Indensely

populatedregions,OsemissionsfromcatalyticconvertorsdominateregionalOsbudgets,

andrepresentasignificantsourceofOstotheatmosphere.TransferofanthropogenicOs

fromthesesources,viasurfacewaters,totheworld’soceansrepresents~12%of

anthropogenicOsinputtothesurfaceocean,whichactstodrivethe187Os/188Os

compositionofseawaterfrom1.05to0.95.ThisstudyechoesChenetal.(2009)who

suggestedthat‘Osisotopescouldbeavaluabletracerforthehydrologicalcycle,similarto

Pbfromleadedgasolineusagebefore1978ortritiumfromatmosphericatomicbomb

testingintheearly1960s.’

FurtherworkisneededtounderstandthespecificuptakerateofOsfrom

seawaterbymacroalgaeinasimilarmannertoRacionero-Gómezetal.(2017).Suchdata

forJapanesemacroalgaespecieswillprovideabetterestimateoftheOsabundancein

Japanesecoastalwaters.ThiswillhelpyieldabetterunderstandingoftheglobalOscycle

andoceanicresidencetimes.Itwouldbewisetoutiliseotherisotopesystemsin

conjunctionwiththeOsisotopesystemtodistinguishbetweenthesourcesof

anthropogenicOsfrompopulatedregions.Forinstance,AlandorganicChaveusedin

conjunctionwithosmiumisotopestotracetheflowofanthropogenicOsrelatedtosewage

104

outflow.ThiswillallowthepossibilitytodistinguishOsattributedtocatalyticconvertors

fromothersourcesincomplexregionssuchasTokyo.

3.6References

Alves,S.,Schiano,P.,Allègre,C.J.,1999.Rhenium–osmiumisotopicinvestigationofJava

subductionzonelavas.EarthandPlanetaryScienceLetters168,65-77.

Alves,S.,Schiano,P.,Capmas,F.,Allègre,C.J.,2002.Osmiumisotopebinarymixingarrays

inarcvolcanism.EarthandPlanetaryScienceLetters198,355-369.

Birck,J.L.,Barman,M.R.,Capmas,F.,1997.Re‐Osisotopicmeasurementsatthe

femtomolelevelinnaturalsamples.Geostandardsnewsletter21,19-27.

Chandler,A.J.,Eighmy,T.,Hjelmar,O.,Kosson,D.,Sawell,S.,Vehlow,J.,VanderSloot,H.,

Hartlén,J.,1997.Municipalsolidwasteincineratorresidues.Elsevier.

Chen,C.,Sedwick,P.N.,Sharma,M.,2009.Anthropogenicosmiuminrainandsnowreveals

global-scaleatmosphericcontamination.ProceedingsoftheNationalAcademyofSciences

oftheUnitedStatesofAmerica106,7724-7728.

Chen,C.,Sharma,M.,2009.Highprecisionandhighsensitivitymeasurementsofosmiumin

seawater.Analyticalchemistry81,5400-5406.

Chen,C.,Sharma,M.,Bostick,B.C.,2006.LithologiccontrolsonosmiumisotopesintheRio

Orinoco.EarthandPlanetaryScienceLetters252,138-151.

Cohen,A.S.,Waters,F.G.,1996.Separationofosmiumfromgeologicalmaterialsbysolvent

extractionforanalysisbythermalionisationmassspectrometry.AnalyticaChimicaActa

332,269-275.

Creaser,R.A.,Papanastassiou,D.A.,Wasserburg,G.J.,1991.Negativethermalionmass

spectrometryofosmium,rheniumandiridium.GeochimicaetCosmochimicaActa55,397-

401.

Cumming,V.M.,Poulton,S.W.,Rooney,A.D.,Selby,D.,2013.Anoxiaintheterrestrial

environmentduringthelateMesoproterozoic.Geology41,583-586.

105

Dalai,T.K.,Suzuki,K.,Minagawa,M.,Nozaki,Y.,2005.Variationsinseawaterosmium

isotopecompositionsincethelastglacialmaximum:AcasestudyfromtheJapanSea.

ChemicalGeology220,303-314.

Earth,G.,2016.GoogleEarth7.1.2.2041.Hospital,sewagetreatmentplant,steelmilland

oilrefienerylocations.Google,SKtelecomandZenrindataproviders.In:

https://www.google.com/earth/index.html(assessed30thMay2017).

Ehrenbrink,B.P.,Ravizza,G.,1996.ContinentalrunoffofosmiumintotheBalticSea.

Geology24,327-330.

ESRI,2014.ArcGISDesktop:Release10.3.Redlands,CA:EnvironmentalSystemsResearch

Institute.

Esser,B.K.,Turekian,K.K.,1993.Anthropogenicosmiumincoastaldeposits.Environmental

science&technology27,2719-2724.

Finlay,A.J.,Selby,D.,Osborne,M.J.,2011.Re-Osgeochronologyandfingerprintingof

UnitedKingdomAtlanticmarginoil:Temporalimplicationsforregionalpetroleumsystems.

Geology39,475-478.

Funari,V.,Meisel,T.,Braga,R.,2016.Thepotentialimpactofmunicipalsolidwaste

incineratorsashesontheanthropogenicosmiumbudget.TheScienceofthetotal

environment541,1549-1555.

Gannoun,A.,Burton,K.,2014.Highprecisionosmiumelementalandisotope

measurementsofNorthAtlanticseawater.JAAS.

Gannoun,A.,Burton,K.W.,Thomas,L.E.,Parkinson,I.J.,vanCalsteren,P.,Schiano,P.,2004.

Osmiumisotopeheterogeneityintheconstituentphasesofmid-oceanridgebasalts.

Science303,70-72.

Gannoun,A.,Burton,K.W.,Vigier,N.,Gíslason,S.R.,Rogers,N.,Mokadem,F.,Sigfússon,B.,

2006.Theinfluenceofweatheringprocessonriverineosmiumisotopesinabasaltic

terrain.EarthandPlanetaryScienceLetters243,732-748.

Gogot,J.,Poirier,A.,Boullemant,A.,2015.Osmiumisotopictracingofatmospheric

emissionsfromanaluminumsmelter.ComptesRendusGeoscience347,277-283.

106

Hart,G.L.,Johnson,C.M.,Hildreth,W.,Shirey,S.B.,2003.Newosmiumisotopeevidencefor

intracrustalrecyclingofcrustaldomainswithdiscreteages.Geology31,427-430.

Hart,G.L.,Johnson,C.M.,Shirey,S.B.,Clynne,M.A.,2002.Osmiumisotopeconstraintson

lowercrustalrecyclingandplutonpreservationatLassenVolcanicCenter,CA.Earthand

PlanetaryScienceLetters199,269-285.

Hashimoto,M.,1991.GeologyofJapan.SpringerScience&BusinessMedia.

Helz,G.R.,Adelson,J.M.,Miller,C.V.,Cornwell,J.C.,Hill,J.M.,Horan,M.,Walker,R.J.,2000.

OsmiumisotopesdemonstratedistaltransportofcontaminatedsedimentsinChesapeake

Bay.Environmentalscience&technology34,2528-2534.

Huh,Y.,Birck,J.-L.,Allègre,C.J.,2004.OsmiumisotopegeochemistryintheMackenzie

Riverbasin.EarthandPlanetaryScienceLetters222,115-129.

Ishihara,S.,Wu,C.,2001.GenesisofLateCretaceous-PaleogeneGranitioidswith

ContrastingChemicalTrendsintheChubuDistrict,CentralJapan.BULLETIN-GEOLOGICAL

SURVEYJAPAN52,471-491.

Japan,G.S.o.,1992.GeologicalMapofJapan,1:1,000,000.3rded.GeologicalSurveyof

Japan,Tsukuba.

Johnson,C.M.,Shirey,S.B.,Barovich,K.M.,1996.Newapproachestocrustalevolution

studiesandtheoriginofgraniticrocks:whatcantheLu–HfandRe–Osisotopesystemstell

us?GeologicalSocietyofAmericaSpecialPapers315,339-352.

Jones,R.,1999.PlatinumsmeltinginSouthAfrica.SouthAfricanJSci95,525–534.

Kannari,A.,Tonooka,Y.,Baba,T.,Murano,K.,2007.Developmentofmultiple-species<

span><imgheight=.AtmosphericEnvironment41,3428-3439.

Koyama,M.,Umino,S.,1991.WhyDoestheHigashi-IzuMonogeneticVolcanoGroupExist

intheIzuPeninsula?JournalofPhysicsoftheEarth39,391-420.

Lambert,D.,Foster,J.,Frick,L.,Hoatson,D.,Purvis,A.,1998.ApplicationoftheRe‐Os

isotopicsystemtothestudyofPrecambrianmagmaticsulfidedepositsofWestern

Australia.AustralianJournalofEarthSciences45,265-284.

107

Levasseur,S.,Birck,J.-L.,Allegre,C.,1999a.Theosmiumriverinefluxandtheoceanicmass

balanceofosmium.EarthandPlanetaryScienceLetters174,7-23.

Levasseur,S.,Birck,J.-L.,Allègre,C.J.,1998.Directmeasurementoffemtomolesofosmium

andthe187Os/186Osratioinseawater.Science282,272-274.

Levasseur,S.,Birck,J.-L.,Allègre,C.J.,1999b.Osmiumisotopiccompositionofcorals:

Evidencesformultiplesources.Geochimicaetcosmochimicaacta63,1335-1343.

Mas,J.,Tagami,K.,Uchida,S.,2005.RheniummeasurementsonNorthAtlanticseaweed

samplesbyID-ICP-MS:anobservationontheReconcentrationfactors.Journalof

RadioanalyticalandNuclearChemistry265,361-365.

McCandless,T.,Ruiz,J.,1991.Osmiumisotopesandcrustalsourcesforplatinum-group

mineralizationintheBushveldComplex,SouthAfrica.Geology19,1225-1228.

Morgan,J.W.,Walker,R.J.,Horan,M.F.,Beary,E.S.,Naldrett,A.J.,2002.190Pt-186Osand

187Re-187OssystematicsoftheSudburyigneouscomplex,Ontario.Geochimicaet

CosmochimicaActa66,273-290.

Nakajima,T.,1994.TheRyokeplutonometamorphicbelt:crustalsectionoftheCretaceous

Eurasiancontinentalmargin.Lithos33,51-66.

Nowell,G.,Luguet,A.,Pearson,D.,Horstwood,M.,2008.Preciseandaccurate186Os/188

Osand187Os/188Osmeasurementsbymulti-collectorplasmaionisationmass

spectrometry(MC-ICP-MS)partI:Solutionanalyses.ChemicalGeology248,363-393.

Ohta,A.,Imai,N.,Terashima,S.,Tachibana,Y.,2005.Applicationofmulti-element

statisticalanalysisforregionalgeochemicalmappinginCentralJapan.Applied

Geochemistry20,1017-1037.

Ohta,A.,Imai,N.,Terashima,S.,Tachibana,Y.,2011.Regionalgeochemicalmappingin

easternJapanincludingthenation'scapital,Tokyo.Geochemistry:Exploration,

Environment,Analysis11,211-223.

Ohta,A.,Imai,N.,Terashima,S.,Tachibana,Y.,Ikehara,K.,Nakajima,T.,2004.Geochemical

mappinginHokuriku,Japan:influenceofsurfacegeology,mineraloccurrencesandmass

movementfromterrestrialtomarineenvironments.AppliedGeochemistry19,1453-1469.

108

Ohta,A.,Imai,N.,Terashima,S.,Tachibana,Y.,Ikehara,K.,Okai,T.,Ujiie-Mikoshiba,M.,

Kubota,R.,2007.Elementaldistributionofcoastalseaandstreamsedimentsintheisland-

arcregionofJapanandmasstransferprocessesfromterrestrialtomarineenvironments.

AppliedGeochemistry22,2872-2891.

Paul,M.,Reisberg,L.,Vigier,N.,2009.Anewmethodforanalysisofosmiumisotopesand

concentrationsinsurfaceandsubsurfacewatersamples.ChemicalGeology258,136-144.

Pelling,M.,Blackburn,S.,2014.Megacitiesandthecoast:risk,resilienceand

transformation.Routledge.

Peucker-Ehrenbrink,B.,Ravizza,G.,2000.Themarineosmiumisotoperecord.TerraNova

12,205-219.

Peucker-Ehrenbrink,B.,Ravizza,G.,2012.OsmiumIsotopeStratigraphy.145-166.

Peucker‐Ehrenbrink,B.,Jahn,B.m.,2001.Rhenium‐osmiumisotopesystematicsand

platinumgroupelementconcentrations:Loessandtheuppercontinentalcrust.

Geochemistry,Geophysics,Geosystems2.

Peucker‐Ehrenbrink,B.,Sharma,M.,Reisberg,L.,2013.Recommendationsforanalysisof

dissolvedosmiuminseawater.Eos,TransactionsAmericanGeophysicalUnion94,73-73.

Poirier,A.,Gariépy,C.,2005.Isotopicsignatureandimpactofcarcatalystsonthe

anthropogenicosmiumbudget.Environmentalscience&technology39,4431-4434.

Prouty,N.G.,Roark,E.B.,Koenig,A.E.,Demopoulos,A.W.,Batista,F.C.,Kocar,B.D.,Selby,

D.,McCarthy,M.D.,Mienis,F.,Ross,S.W.,2014.Deep‐seacoralrecordofhumanimpact

onwatershedqualityintheMississippiRiverBasin.GlobalBiogeochemicalCycles28,29-43.

Racionero-Gómez,B.,Sproson,A.,Selby,D.,Gröcke,D.,Redden,H.,Greenwell,H.,2016.

Rheniumuptakeanddistributioninphaeophyceaemacroalgae,Fucusvesiculosus.Royal

SocietyOpenScience3,160161.

Racionero-Gómez,B.,Sproson,A.D.,Selby,D.,Gannoun,A.,Gröcke,D.R.,Greenwell,H.C.,

Burton,K.W.,2017.Osmiumuptake,distribution,and187Os/188Osand187Re/188Os

compositionsinPhaeophyceaemacroalgae,Fucusvesiculosus:Implicationsfordetermining

the187Os/188Oscompositionofseawater.GeochimicaetCosmochimicaActa199,48-57.

109

Rauch,S.,Hemond,H.F.,Peucker-Ehrenbrink,B.,2004.Recentchangesinplatinumgroup

elementconcentrationsandosmiumisotopiccompositioninsedimentsfromanurbanlake.

Environmentalscience&technology38,396-402.

Rauch,S.,Hemond,H.F.,Peucker-Ehrenbrink,B.,Ek,K.H.,Morrison,G.M.,2005.Platinum

groupelementconcentrationsandosmiumisotopiccompositioninurbanairborneparticles

fromBoston,Massachusetts.Environmentalscience&technology39,9464-9470.

Ravizza,G.,Bothner,M.H.,1996.Osmiumisotopesandsilverastracersofanthropogenic

metalsinsedimentsfromMassachusettsandCapeCodbays.GeochimicaetCosmochimica

Acta60,2753-2763.

Rodushkin,I.,Engstrom,E.,Sorlin,D.,Ponter,C.,Baxter,D.C.,2007a.Osmiumin

environmentalsamplesfromNortheastSweden.PartI.Evaluationofbackgroundstatus.

TheScienceofthetotalenvironment386,145-158.

Rodushkin,I.,Engstrom,E.,Sorlin,D.,Ponter,C.,Baxter,D.C.,2007b.Osmiumin

environmentalsamplesfromNortheastSweden.PartII.Identificationofanthropogenic

sources.TheScienceofthetotalenvironment386,159-168.

Rooney,A.D.,Selby,D.,Lloyd,J.M.,Roberts,D.H.,Lückge,A.,Sageman,B.B.,Prouty,N.G.,

2016.Trackingmillennial-scaleHoloceneglacialadvanceandretreatusingosmium

isotopes:InsightsfromtheGreenlandicesheet.QuaternaryScienceReviews138,49-61.

Selby,D.,Creaser,R.A.,2005.Directradiometricdatingofhydrocarbondepositsusing

rhenium-osmiumisotopes.Science308,1293-1295.

Selby,D.,Creaser,R.A.,Fowler,M.G.,2007.Re–Oselementalandisotopicsystematicsin

crudeoils.GeochimicaetCosmochimicaActa71,378-386.

Smith,I.,1974.Osmium:Anappraisalofenvironmentalexposure.EnvironHealthPerspect

8,201–213.

Smith,S.,Yanagi,T.,1997.NPbudgetsforthreeJapaneseBays.LOICZ-Biogeochemical

ModellingNode.

Turekian,K.K.,Sharma,M.,Gordon,G.W.,2007.Thebehaviorofnaturalandanthropogenic

osmiumintheHudsonRiver–LongIslandSoundestuarinesystem.Geochimicaet

CosmochimicaActa71,4135-4140.

110

Völkening,J.,Walczyk,T.,Heumann,K.G.,1991.Osmiumisotoperatiodeterminationsby

negativethermalionizationmassspectrometry.InternationalJournalofMass

SpectrometryandIonProcesses105,147-159.

Walker,R.,Morgan,J.,Horan,M.,Czamanske,G.,Krogstad,E.,Fedorenko,V.,Kunilov,V.,

1994.Re-Osisotopicevidenceforanenriched-mantlesourcefortheNoril'sk-type,ore-

bearingintrusions,Siberia.GeochimicaetCosmochimicaActa58,4179-4197.

Walker,R.J.,Prichard,H.M.,Ishiwatari,A.,Pimentel,M.,2002.Theosmiumisotopic

compositionofconvectinguppermantlededucedfromophiolitechromites.Geochimicaet

CosmochimicaActa66,329-345.

Williams,G.,Marcantonio,F.,Turekian,K.K.,1997.Thebehaviorofnaturaland

anthropogenicosmiuminLongIslandSound,anurbanestuaryintheeasternUS.Earthand

planetaryscienceletters148,341-347.

Williams,G.A.,Turekian,K.K.,2002.Atmosphericsupplyofosmiumtotheoceans.

Geochimicaetcosmochimicaacta66,3789-3791.

Woodhouse,O.,Ravizza,G.,Falkner,K.K.,Statham,P.,Peucker-Ehrenbrink,B.,1999.

Osmiuminseawater:verticalprofilesofconcentrationandisotopiccompositioninthe

easternPacificOcean.EarthandPlanetaryScienceLetters173,223-233.

Yang,J.S.,1991.Highrheniumenrichmentinbrownalgae:abiologicalsinkofrheniumin

thesea?Hydrobiologia211,165-170.

Yuhara,M.,Kagami,H.,Nagao,K.,2000.Geochronologicalcharacterizationand

petrogenesisofgranitoidsintheRyokebelt,SouthwestJapanArc:constraintsfromK–Ar,

Rb–SrandSm‐Ndsystematics.IslandArc9,64-80.

Zheng,G.,Suzuki,K.,Kuno,A.,Mastuo,M.,Takano,B.,Shimizu,H.,2014.Osmium

geochemistryofmodernestuarinesedimentsfromtheTamaandYasakariversinJapan.

AppliedGeochemistry40,82-88.

111

Chapter4

Osmiumandlithiumisotopeevidenceforfluctuatingoxidativeandsilicate

weatheringduringperiodicSilurianglaciations*

*AversionofthischapterwillbesubmittedtoNature,co-authoredwithPhilipPoggevon

Strandmann,TimothyM.Lenton,DavidSelby,EmiliaJarochowska,JiriFryda,JindrichHladil,

DavidLoydell,LadislavSlavik,MikaelCalnerandAxelMunnecke

112

TheSilurian(419.2to443.8Ma)wasoneofthemostclimaticallyunstableperiodsofthe

Phanerozoic,punctuatedbyfourlargepositivecarbonisotope(δ13C)excursions,associated

withrapidturnoverinmarinebiota.Despiteovertwodecadesofresearch,thecauseof

theseclimaticfluctuationsisstillunclear.Hereweuseosmium(187Os/188Os)andlithium

(δ7Li)isotopemeasurementsofshalesandcarbonatesspanningfourofthemost

prominentSilurianδ13Cexcursions,toassesstheroleofcontinentalweatheringduringthis

time.Wefindtwopeaksofradiogenic187Os/188Oscompositionseithersideofthelightest

δ7Livaluesduringeachδ13Cexcursion.Geochemicalmodellingattributesthistoperiodic

continentalglaciations,whichacttoenhancetheoxidativeweatheringoforganic-and

sulphide-richlithologies,whilstsuppressingglobalsilicateweathering.Theproductionof

atmosphericCO2fromoxidativeweathering,coupledtoareductioninsilicateweathering–

andthereforeatmosphericCO2removal–actedtoreversethelong-termdeclinein

atmosphericCO2andglobaltemperaturesdrivenbyorogenesis,landplantdiversification,

reducedvolcanicarcdegassingand/orchangesinpaleogeographyduringtheSilurian.

4.1Introduction

OverthepasttwodecadesithasbecomeapparentthattheSilurianisthemostclimatically

unstableperiodofthePhanerozoic,punctuatedbyfourlargeamplitude(>5‰)positive

carbonisotope(δ13C)excursions(Fig.1)associatedwithfluctuationsinthecarboncycle,

seawatertemperaturesandfaunalextinctionrates(Calner,2008;Lehnertetal.,2010;

Melchinetal.,2012;Munneckeetal.,2010;Nobleetal.,2005).Traditionalexplanationsfor

theseeventshaveinvokedashiftbetweentwostableoceanic-climatestates,drivenby

changesinthelocationofdeep-waterformationfromhightolowlatitudes(Jeppsson,

1990),orglobalprecipitationratesandcontinentalrunoff(Bickertetal.,1997).However,

113

theseearlyattemptstoexplainSilurianclimaticeventshavereceivedmuchcriticism

(Johnson,2006;Kaljoetal.,2003;Loydell,1998).

Figure1.Osmium(187Os/188Os,greensquares),lithium(δ7Li,reddiamonds),oxygen(δ18O,

bluecircles)andcarbon(δ13C,blackcircles)isotoperatiosforshaleandcarbonatesections

measured.a,Klonk,b,Kosov(Os),c,BartoszyceIG-1,d,Aizpute-41,e,Kosov(Li),f,

Hunningeandg,Lusklint&Lickershamn.Datafromthisstudyiscomparedtotheglobal

conodont(C)andgraptolite(G)bio-events(Melchinetal.,2012),δ13C(Saltzmanand

Thomas,2012;Crameretal.,2011a),δ18O(SeeTrotteretal.,2016),glacialtillites(thick

dashedblackline)(Caputo,1998;Díaz-MartínezandGrahn,2007;GrahnandCaputo,1992)

andsea-levelreconstructions(HaqandSchutter,2008;Johnson,2006,2010;Loydell,

1998).

114

MorerecentlyithasbeenpostulatedthatSilurianclimaticchangecouldhave

beendrivenbyglacialexpansionoverGondwana,inferredinpartfrompositiveoxygen

isotope(δ18O)shifts(Fig.1)(Azmyetal.,1998;Brandetal.,2006;ErikssonandCalner,

2007;Kaljoetal.,2003;Lehnertetal.,2010;Trotteretal.,2016;Žigaitėetal.,2010)

coupledtosignificanteustaticsea-levelchange(HaqandSchutter,2008;Johnson,2006,

2010;Loydell,1998),muchliketheLateOrdovicianthatprecededit(Algeoetal.,2016;

Harperetal.,2014).However,thelackofglacialsedimentsinthestratigraphicrecordfor

muchoftheSilurian(post-Wenlock)hashamperedthisnotion(Caputo,1998;Díaz-

MartínezandGrahn,2007;GrahnandCaputo,1992).

Intheabsenceofstratigraphicevidenceforglacialsediments,seawater

chemistrycanbecomeapowerfularchiveforreconstructingearthsystemresponsesto

climaticortectonicchange,andseveralisotopesystemshavebeenutilisedtoreconstruct

continentalweatheringanderosion.Traditionally,therubidium-strontium(87Rb-86Sr)

radiogenicisotopesystemhasbecomethemostwidelyused,andvariationsinmarine

87Sr/86Srareseentoreflectfluctuationsincontinentalinputscausedbyorogenesis(Raymo

etal.,1988)andglaciation(Armstrong,1971).However,thelongresidencetimeof

strontiumintheoceans(2-4Myr)andmultiplecontinentalsourcesfrombothcarbonate

andsilicateweathering(Jacobsonetal.,2002;PalmerandEdmond,1992)meansthat

short-periodicfluctuationsinunambiguousinputsarehardtodetect(Hodelletal.,1990;

RichterandTurekian,1993).Unlikestrontium,OsandLiisotopesystemscanovercome

someofthesedifficulties.

Theosmiumisotopiccomposition(187Os/188Os)ofseawaterreflectsabalance

betweentheweatheringofradiogenicOs-richsedimentaryrocksorsilicatemineralsand

unradiogenicmantleandextraterrestrialderivedsources(Georgetal.,2013;Peucker-

EhrenbrinkandRavizza,2000).Likewise,thelithiumisotopiccomposition(δ7Li)ofseawater

115

reflectsthebalancebetweenchemicalweatheringofcontinentalsilicateminerals,high

temperatureweatheringofoceanicsilicatemineralsatmid-oceanridgespreadingcentres

(Chanetal.,1993;ElderfieldandSchultz,1996;Huhetal.,1998),andincorporationinto

alteredoceaniccrustandauthigenicclays(Chanetal.,1992;Chanetal.,2006;Jamesetal.,

1999;MackenzieandGarrels,1966;MisraandFroelich,2012;Seyfriedetal.,1998;Verney-

Carronetal.,2011;Vigieretal.,2008).Whencombinedwithrelativelyshortresidence

timesinseawater(~1-50kyrforOsand~1-1.5myrforLi)(Huhetal.,1998;Oxburgh,

2001;Rooneyetal.,2016;StoffynegliandMackenzie,1984),theseisotopesystemshave

permittedtheabilitytounlockvitalinformationaboutaseriesofEarthsystemprocesses

suchasfloodbasaltvolcanism(CohenandCoe,2002;DuVivieretal.,2014;Ravizzaand

Peucker-Ehrenbrink,2003;TurgeonandCreaser,2008),paleoweathering(Hathorneand

James,2006;Lechleretal.,2015;MisraandFroelich,2012;PoggevonStrandmannetal.,

2013;Ravizzaetal.,2001;Schmitzetal.,2004),basinconnectivity(PoirierandHillaire-

Marcel,2009),bolideimpacts(Paquayetal.,2008)andmantleandperidotiteformation

(PoggevonStrandmannetal.,2011).Intandem,thesesystemshaverecentlybeenutilised

todeterminehowsilicateweatheringhasbeeninfluencedbycontinentalicevolume,

atmosphericCO2andglobaltemperaturesduringtheLateOrdovicianHirnantianmass

extinction(Finlayetal.,2010;PoggevonStrandmannetal.,inreview).

Here,shalesweremeasuredforosmiumisotopes(187Os/188Os),andbulk

carbonatesweremeasuredforlithiumisotopes(δ7Li)andtracemetals,instratigraphic

sectionsthatspanthelateTelychian-earlySheinwoodian,mid-Homerian,mid-Ludfordian

andlatePridoli-earlyLochkovianpositivecarbonisotopeexcursions(SeeFig.1).Thisstudy

representsthefirstapplicationofOsandLiisotopestostratigraphicsectionsfromthe

Silurian.TheresultsdemonstratetheabilityofOsandLiisotopestotraceglobal

fluctuationsinEarthsystemprocesses,andwhencombinedwithdynamicisotopemodels,

elucidatepotentialcauses,suchasfluctuationsinoceancirculation,floodbasaltvolcanism,

116

globaltemperature,hydrothermalactivityand/orcontinentalicevolume.Finally,thisstudy

looksatgeologicalandclimatictrendsinatmosphericCO2duringtheSilurian,andrelates

themtonegativefeedbackmechanismsthathelpmaintainahabitableplanet.

4.2Materialsandmethods

4.2.1Geologicalsetting

Inthisstudy,fourshalesectionswereanalysedforosmiumisotopesandfourbulk

carbonatesectionswereanalysedforlithiumisotopes(SeeFig.3to9).Combined,these

sectionscoveredfourperiodsofSiluriantime:thelateTelychiantoEarlySheinwoodian;

midHomerian;lateLudfordian;andtheSilurian-Devonianboundary.Shalesfromthe

Aizpute-41core(Latvia)andcarbonatesfromtheLusklint&Lickershamnsections(Gotland,

Sweden)coverthelatestTelychiantoearliestSilurian.ShalesfromtheBartoszycecore

(Poland)andbulkcarbonatesfromtheHunninge-1core(Sweden)coverthemidHomerian.

ShalesandbulkcarbonatesfromtheKosov(CzechRepublic)sectioncovertheLate

Ludfordian.ShalesfromtheKlonkcore(CzechRepublic)covertheSilurian-Devonian

Boundary.ThelocationofthesesectionscanbefoundinFigure2.Sectionswerechosen

becausetheyhaveallhadextensivecarbonisotopeandbiologicalstratigraphycarriedout.

Meanwhiletheyrepresentthemostdistalcoastalenvironemntsawayfromrestricted

basins,andthereforearethemostlikelysectionstorecordoceansignatures.Thefollowing

willdetailthegeology,samplingstrategyandpaleoenvironmentofeachsectionstudied.

117

Figure2.Silurian(425Ma)paleogeographicmap(adaptedfromMelchinetal.,2012).

SamplelocationsfromGotland(Hunninge-1),Aizpute-41,BartoszyceandtheBarrandian

(KosovandKlonk)arehighlighted.Darkandlightbrownrepresentthepositionof

continentsandcontinentalshelvesrespectively,duringtheSilurian.Whitelinesrepresent

theconfigurationofmoderncontinents.

4.2.1.1Aizputecore

ThelateTelychian-earlySheinwoodianAizpute-41coreislocatedinthetownofAizpute,

situatedinwesternLatvia,inthedeepershelfpartoftheEasternBaltoscandianBasin.The

latestLlandoverybedsconsistofgreenishandbrownishgreymarlstones,whiletheearliest

Wenlockconsistsofgreen,greyandbrownmarlstoneswithcalcareousmarlstones(Loydell

etal.,2003).Samplingstrategyofgraptolite-rich,relativelyhightotalorganiccarbon(TOC)

shaleswasfollowedaccordingtoLoydelletal.(2003).Theδ13Ccarbdataforthedrillcore

materialcanbefoundinCrameretal.(2010).

4.2.1.2Lusklint&Lickershamn

ThelateTelychian-earlySheinwoodianLowerVisbyformationandtheearliest

SheinwoodianUpperVisbyformationcanbefoundoutcroppingatLusklintand

Lickershamnalongthenorth-westerncoastofGotland,Sweden.TheLowerVisby

formationconsistsofupto12mofregularalternationsof2-5cmthick,wavybeddedto

118

nodularargillaceouslimestonesandapproximately10cmthickmarlswhichwere

depositedinadistalshelfenvironment,belowstormwavebaseandthephoticzone

(Maier,2010;Calneretal.,2004).However,thebeddingintheUpperVisbyformationis

notasregular,withtheratioofmarltolimestoneincreasingintheuppermostpartofthe

formationwheredetriticlimestonesbecomemoredominant(Maier,2010).Samplesfrom

theLowerVisbyformationweresampledfromLusklintwhilesamplesfromtheUpperVisby

formationweresampledfromLusklint.Theδ13Ccarbdataforthesamplescanbefoundin

Maier(2010).

4.2.1.3Bartoszyce

Themid-HomerianBartoszyceIG1boreholeislocatedintheeasternpartofthePeribaltic

SyneclizeofthePolishpartoftheEastEuropeanplatform.Thecoreconsistsofsparsely

bioturbated,light-greylaminatedandcalcareousmudstones(Porębskaetal.,2004).

SamplingstrategyofrelativelyhighTOCshales,δ13Ccarbandδ18Ocarbdatacanbefoundin

Porębskaetal.(2004).

4.2.1.4Hunningecore

Themid-Homerianhunninge-1coreislocatedintheHunningequarryofwesternGotland,

Sweden.TheGannarvememberconsistsofalternatingbedsofbrownish,argillaceous,fine

dolostonewithsiltydolomarlstoneandalumshales(Calneretal.,2006).Theoverlyingbara

oolitememberandbrickclaymemberconsistofcoatedgrainsandargillaceous,nodular

limestonealternatingwithshalerespectively(Calneretal.,2006).Samplingstrategyfor

carbonatesandδ13CcarbdatacanbefoundinCalneretal.(2006).

4.2.1.5Kosov

Themid-LudfordianKosovsectionislocatedintheBarrandianregionoftheCzechRepublic.

Thekozlowskiibiozoneconsistsofalternatingbedsofgreyfinelylaminatedshaleandlight-

greypackstonesandgrainstones,whiletheoverlyingPristiograptusdubiuspostfrequens

119

biozoneconsistsofalternatingbedsoflight-greypackstones/grainstonesandmudstonesor

greycoarselylaminatedcalcareousshales.Samplingstrategyofcarbonatesandrelatively

highTOCshalesandδ13CcarbdatacanbefoundinFrýdaandManda(2013).

4.2.1.6Klonkcore

ThePridoli-Lochkovian(Silurian-Devonian)GSSPislocatedintheCzechRepublic35km

southwestofPrague.ThelatestPridolianandearliestLochkovianbedsconsistofgrayish-

black,platy,mostlyfine-grainedbituminouslimestonesalternatingwithcalcareousshale

interbedswithoccasionalstringersofcrinoidallimestones(Beckeretal.,2012).Sampling

strategyofrelativelyhightotalorganiccarbon(TOC)drillcorematerialwasfollowed

accordingtoCricketal.(2001).δ13Ccarbandδ18Ocarbdataforthedrillcorematerialcanbe

foundinBuggischandMann(2004).

4.2.2Samplepreparation

Priortocrushing,20-80gofshalesamplewaspolishedtoeliminatecontaminationfrom

cuttinganddrillingmarksandsampleswithanysignsofveiningorweatheringwere

avoided.Theshalesampleswerethendriedat60°Cfor~12hbeforebeingbrokeninto

chipswithnometalcontact.Bulkcarbonatesandshaleswerecrushedtoafinepowder

(~30μm)inaZirconiaceramicdishusingashatterbox.Bulkcarbonateswereleachedusing

asequentialextractionmethod(PoggevonStrandmannetal.,2013;Tessieretal.,1979),

whereby~0.1gofcarbonatewasleachedfor5hatroomtemperatureusingNaacetate

bufferedtopH5byaceticacid.Leachingofinterstitialsilicateswasmonitoredusing

elementalratiossuchasAl/CaandMn/Ca.TheAl/Caratiomustbegreaterthan>0.8

mmol/molbeforesilicate-derivedLiwillperturbthed7Limeasuredincarbonates(Pogge

vonStrandmannetal.,2013).ThesamplepreparationandRe-Osisotopeandtracemetal

analysiswascarriedoutattheDurhamGeochemistryCentre(LaboratoryforSulfideand

120

SourceRockGeochronologyandGeochemistry)atDurhamUniversity.TheLiisotope

analysiswascarriedoutatthestableisotopelabtheUniversityofOxford.

4.2.3Osmiumisotopeanalysisofshales

Rheniumandosmiumabundancesandisotopiccompositionsweredeterminedusing

isotopedilutionnegativethermalionisationmassspectrometryusingCrVI–H2SO4digestion

andsolventextraction(CHCl3),micro-distillationandanionchromatographymethods

(Creaseretal.,1991;Cummingetal.,2013;SelbyandCreaser,2003).TheCrVI–H2SO4

digestionemployedhereprincipallydissolvestheorganicfractionofashale,thusliberating

thehydrogenousRe-Osloadofthesediment,andthereforeavoidingdetrital

contamination(Kendalletal.,2004;SelbyandCreaser,2003).

Forallshalesamplesbetween0.5and1gofpowderwasaddedtoacarius-tube

withaknownamountof185Re-190Ostracersolutionand8mlof0.25g/gCrVIO3-4NH2SO4at

220°Cfor48h.OsmiumwasisolatedfromtheacidmediumusingCHCl3solventextraction,

withbackextractioninHBr,andthenfurtherpurifiedusingamicro-distillationtechnique.

RheniumwasisolatedusingNaOH-C3H6Osolventextractionandpurifiedusinganion

chromatography.TheisolatedReandOsfractionswereloadedontoNiandPtfilaments

respectively,andtheirisotopiccompositionwasdeterminedusingaThermoScientific

TRITONmassspectrometerusingFaradaycollectorsandthesecondaryelectronmultiplier,

respectively.

TotalproceduralblanksforReandOsare1.1and0.1pgrespectively,withan

average187Os/188Osof1.3(n=6).RawReandOsoxidevalueswerecorrectedforoxygen

contributionandmassfractionation.Calculateduncertaintiesincludethoseassociatedwith

massspectrometermeasurements,blankabundanceandisotopiccomposition,spike

121

calibration,andsampleandspikeweights.In-housestandardsolutionsofReandOs

(DROsS)yieldanaverage185Re/187Revalueof0.59872±0.00135(1SD,n=24),and

187Os/188Osof0.16101±0.000401(1SD,n=41),respectively,whichisidentical,within

uncertaintytothepreviouslypublishedvalues(Nowelletal.,2008).

Initial187Os/188Os(187Os/188Osi)valuesinthisstudyweredeterminedfromRe-Os

dataandthe187Redecayconstant(1.666e−11a−1)(Smoliaretal.,1996)andinterpolated

graptolitebiozoneages(Melchinetal.,2012).Analyticaluncertaintyforindividual

calculatedOsiis<0.05.Thereproducibilityofcalculated187Os/188Osiwasbasedon15

analysesoftheUSGSrockreferencematerialSBC-1(BushCreekShale)andhasavalueof

~0.65±0.1(2SD).Thisuncertaintywasusedtoaccountforthemaximumuncertaintyin

thesampleset.Calculated187Os/188Osiratiosassumeclosedsystembehaviorafter

depositionwithrespecttobothrheniumandosmiumandthatthe187Os/188Osratiosreflect

theisotopecompositionofthelocalseawateratthetimeofsedimentdeposition,andare

unaffectedbymineraldetritus.

4.2.4Lithiumisotopeanalysisofbulkcarbonates

AsplitofeachsamplesolutionwasretainedforcationanalysisusinganElanQuadrupole

inductivelycoupledplasmamassspectrometer.Sampleswerematrixmatchedto10μg/g

Caandcalibratedagainstasetofsyntheticstandardsmadeupfromsingleelement

solutions.TheAl/CaratioincarbonateswasmonitoredtodetecttheinfluenceofLileached

fromsilicateclays.Previousworksuggestscarbonatesmustbe>0.8mmol/molbefore

carbonateLiisotoperatiosbecomemeasurablyperturbedbyLileachedfromclays(See

PoggevonStrandmannetal.,2013).Accuracyandprecisionwereassessedbyrepeated

analysesofseawater,theinternationalreferencematerialJLs-1and,inordertoassess

reproducibilityofbothanalysesandcarbonateleaching,repeateddissolutionsandanalysis

122

ofasampleofthePlenusMarlfromEastbournewereundertaken.Samplereproducibility

ofLi/CaandAl/Cawas~7%(2SD,n=6).

Thelargerpartofeachsample(typicallycontaining5–10ngLi)waspurifiedby

passingitthroughatwo-stagecation-exchangeprocedure(PoggevonStrandmannetal.,

2011).GiventhatLiisotopesfractionateduringcationchromatography,itiscriticaltohave

columnyieldscloseto100%(Tomascak,2004).Toassesstheefficacyofthisprocess,splits

ofthesolutionwerecollectedbeforeandafterthecollectedbracketforLi,andwere

analysedforLicontent.Resultsshowedthat<0.1%ofLiwaspresentinthesesplits.

ThetotalproceduralblankforLiisotopeanalysisis~0.02ngLi,whichis

insignificantcomparedtothemassofsampleused.AnalyseswereperformedonaNu

PlasmaHRmulti-collectorICP-MS,usingasample-standardbracketingsystemrelativeto

theLSVECstandard(Fleschetal.,1973).Eachsamplewasmeasuredthreeseparatetimes

duringananalyticalsession,repeatmeasurementsbeingseparatedbyseveralhours,but

duringthesameanalyticalsession.Eachindividualmeasurementconsistedof10ratios(10

stotalintegrationtime),givingatotalintegrationtimeof300s/sample.Atanuptakerate

of75μl/min,thesensitivityfora20ng/mlsolutionis~18pAof7Li.Background

instrumentalLiintensity,typically~0.01pA,wassubtractedfromeachmeasurement.

Accuracyandexternalreproducibility,asassessedfromseawater,is31.1±0.6‰(2SD,n=

16,chemistry=16).Precisionwasalsoassessedfromrepeatedanalyses(includingleaching

andchemistry)ofanin-housemarlstandard,whichalsogivesareproducibilityof±0.6‰

(n=7).

123

4.2.5Isotopemodeling

DynamicboxmodelsforseawaterOsandLicycleshavebeenutilisedtoexplorecausesof

thevariationsseeninthedata.ThemodelswereadaptedfromthosepresentedinPogge

vonStrandmannetal.(2013).

4.2.5.1Osmiumisotopemodeling

TheOsmodelwasconstructedfromthefollowingmassbalanceequation:

+,Os

+-= 𝐹sil + 𝐹ORLW + 𝐹lth + 𝐹hth + 𝐹dust − 𝐹out Eq.1

whereNistheseawaterOsreservoir,andFxrepresentstheinputandoutput

fluxes:sil=riverinputrelatedtosilicateweathering;ORLW=riverinputrelatedtothe

weatheringoforganic-sulphide-richlithologies;lht=low-temperaturehydrothermal;hth=

high-temperaturehydrothermal;anddust=aeoliandust.Theisotopebalanceequationis

thengivenby:

𝑁Os+1SW

+-= 𝐹sil 𝑅sil − 𝑅SW + 𝐹ORLW 𝑅ORLW − 𝑅SW + 𝐹lht 𝑅lth − 𝑅SW +

𝐹hth 𝑅hth − 𝑅SW + 𝐹dust 𝑅dust − 𝑅SW − 𝐹out(𝑅out − 𝑅SW)

Eq.2

whereRxistheisotoperatioofthevariousfluxes.Finally,thecalculationofthe

sinkofOsfromseawaterisbasedontheassumptionthatpartitioningintothesinkisdue

toaconstantpartitioncoefficientk,where:

𝐹out = 𝑘×𝑁 Eq.3

Theseequationsdifferfromtheoriginalmodel(PoggevonStrandmannetal.,

2013)astheriverineinputhasbeenpartitionedintotwocomponentsderivedfromsilicate

weatheringandorganic-sulphide-richlithologyweathering(ORLW)(SeeGeorgetal.,2013).

DuringtheCenozoic,riverineOsfluxesarelargelycontrolledbytheoxidativeweatheringof

124

ORLandrepresents70%ofriverineOsbudget,withsilicateweatheringrepresentingthe

other30%(Georgetal.,2013;Lietal.,2009).However,duringtheSilurian,theabsenceof

extensiveterrestrialvegetationwilllowertheavailabilityoforganic-richsedimentsfor

weathering,andsilicateweatheringwillthereforerepresentamoredominantcontrolon

theriverineOsbudget(60%).Cosmicdustisnotincludedinthemodelasitisunlikelyits

fluxchangedduringthistime.

4.2.5.2Lithiumisotopemodeling

TheLimodelwasconstructedfromthefollowingmassbalanceequation:

+,Li

+-= 𝐹riv + 𝐹hth − 𝐹sed Eq.4

whereNistheseawaterLireservoir,andFxrepresentstheinputandoutput

fluxes:riv=river;hth=hydrothermal;andsed=sediment(combineduptakeintomarine

sediments,andalterationoftheoceaniccrust).Theisotopebalanceequationisthengiven

by:

𝑁Li+1SW

+-= 𝐹riv 𝑅riv − 𝑅SW + 𝐹hth 𝑅hth − 𝑅SW − 𝐹sed(𝑅sed − 𝑅SW) Eq.5

whereRxistheisotoperatioofthevariousfluxeswhereSW=seawater.Rsinkis

givenby∆sink=Rsink-Rsw,where∆7Lisink=15-16‰(Chanetal.,1993;Huhetal.,1998;Misra

andFroelich,2012).Finally,thecalculationofthesinkofLifromseawaterisbasedonthe

assumptionthatpartitioningintothesinkisduetoaconstantpartitioncoefficientk,

where:

𝐹out = 𝑘×𝑁 Eq.6

ThefractionationofLiuptakeintocarbonatesisaccountedforbythemodel.

Assumingamodernoceanicsink(Hazenetal.,2013)andhydrothermalδ7Li,an

125

unfractionatedriverineδ7Li(~0‰)similartocontinentalcrust(Sauzéatetal.,2015)is

requiredtoproducetheδ7LiofseawaterrecordedfortheHirnantian(Poggevon

Strandmannetal.,inreview).Suchanisotopicallylightriverinefluxwasmostlikelydueto

thedominationofillites,whichcauselittlefractionation(MillotandGirard,2007),priorto

theadventofterrestrialplants(PoggevonStrandmannetal.,inreview).Liwasmodelled

over10kyrsteps,whileOswasmodeledover5kyrsteps.Table1liststhevaluesusedin

themodelsforeachsystem.

Table1

Modelinputparameters.StartingparametersarebasedonPoggevonStrandmannetal.,

2013andPoggevonStrandmannetal.,2017.However,Osriverinefluxhasbeenmodified

torepresentcontributionsfromtheweatheringofbothsilicatesandORL.Seetextfor

detailsonmodelperturbations.

StartingParameters Li Os

Friver(Gmol/yr) 20 Silicate(mol/yr) 600

ORLW(mol/yr) 400

Fhydro(Gmol/yr) 9.3 low-Thydro(mol/yr) 95.4

high-Thydro(mol/yr) 366

Fdust(mol/yr)

150

Rriver 0 Silicate 0.6

ORLW 1.34

Rhydro 7 low-Thydro 0.878

high-Thydro 0.115

Rdust 1.05

126

4.3Results

4.3.1Rhenium-osmiumisotopedata

RheniumandosmiumisotopecompositionsandabundancedatafortheAizpute-41

(Latvia),Bartoszyce(Poland),Kosov(CzechRepublic)andKlonk(CzechRepublic)samples

arepresentedinTable2.

Table2

RheniumandosmiumabundanceandisotopedatafortheAizpute-41,Bartoszyce,Kosov

andKlonkshalesamples.Initial187Os/188Os(187Os/188Osi)werecalculatedusinggraptolite

biozoneagesfrom(Melchinetal.(2012).

Depth Re 2s.e. Os 2s.e. 187Re/188Os 2s.e. 187Os/188Os 2s.e. 187Os/188Osi 2s.e.

(m) (ppb) (ppt)

Aizpute-41Core,Latvia

910.06 2.22 0.01 65.0 0.7 206.9 2.9 2.10 0.04 0.602 0.015

910.19 2.31 0.01 67.2 0.7 208.8 3.0 2.11 0.04 0.595 0.014

910.60 2.42 0.01 74.0 0.8 196.9 2.8 2.06 0.04 0.629 0.015

910.90 2.91 0.05 73.8 0.6 244.6 4.9 2.33 0.03 0.558 0.013

911.33 3.23 0.03 87.2 0.7 226.6 3.0 2.18 0.03 0.540 0.010

912.00 2.79 0.05 77.3 0.6 217.4 4.3 2.04 0.03 0.466 0.011

912.90 2.17 0.04 98.9 0.5 121.7 2.3 1.30 0.01 0.414 0.008

914.01 2.02 0.15 84.9 0.5 136.6 10.1 1.59 0.01 0.602 0.045

914.74 5.60 0.01 130.7 1.1 271.3 2.5 2.53 0.03 0.564 0.009

914.80 5.53 0.02 135.4 1.1 250.0 2.3 2.20 0.03 0.391 0.006

914.95 3.26 0.01 115.1 0.7 166.7 1.2 1.83 0.02 0.620 0.007

915.90 7.88 0.25 128.0 0.9 433.6 14.0 3.66 0.03 0.521 0.017

916.50 10.52 0.19 152.8 1.0 502.4 9.2 4.06 0.03 0.422 0.008

917.70 10.20 0.33 144.1 1.1 525.5 17.0 4.28 0.03 0.468 0.016

919.96 9.59 0.17 144.5 1.1 481.5 8.9 4.00 0.03 0.515 0.010

924.70 1.88 0.01 81.2 0.5 131.7 1.1 1.51 0.01 0.557 0.007

924.98 1.52 0.01 78.6 0.5 108.2 0.8 1.38 0.01 0.593 0.007

925.21 3.77 0.04 98.3 0.5 237.6 2.6 2.32 0.01 0.597 0.007

925.65 9.64 0.03 152.7 1.3 460.6 3.5 4.07 0.04 0.733 0.009

928.01 16.74 0.04 207.6 1.8 644.7 4.5 5.18 0.05 0.511 0.006

930.38 17.95 0.32 230.0 1.8 606.8 11.2 4.83 0.04 0.433 0.009

932.35 27.46 0.07 274.0 2.2 929.8 5.2 7.23 0.05 0.489 0.004

934.20 15.86 0.04 249.7 1.9 439.6 2.9 3.47 0.03 0.287 0.003

935.82 5.98 0.02 143.5 0.9 253.5 1.7 2.13 0.02 0.296 0.003

127

BartoszyceCore,Poland

1674.20 12.46 0.22 162.3 1.3 597.0 11.1 4.83 0.04 0.551 0.011

1672.90 12.55 0.40 196.3 1.5 454.8 14.8 3.78 0.03 0.517 0.017

1670.00 8.71 0.02 131.5 1.0 482.9 3.0 4.06 0.03 0.595 0.006

1666.65 9.46 0.17 147.8 1.0 461.5 8.5 3.93 0.03 0.618 0.012

1665.00 14.96 0.27 238.4 1.6 449.5 8.3 3.86 0.03 0.638 0.013

1663.75 3.01 0.05 69.8 0.6 275.5 5.5 2.65 0.03 0.670 0.016

1662.30 1.71 0.06 42.7 0.5 253.2 8.9 2.53 0.05 0.711 0.029

1661.80 1.10 0.02 27.9 0.4 244.2 6.7 2.35 0.07 0.603 0.024

1661.70 1.29 0.01 30.0 0.4 268.5 5.7 2.44 0.07 0.513 0.018

1661.40 1.17 0.02 34.6 0.5 203.6 5.6 2.04 0.06 0.580 0.023

1660.72 0.74 0.02 26.2 0.6 166.2 8.7 1.76 0.10 0.567 0.044

1660.70 0.51 0.01 34.2 0.7 81.1 3.4 1.18 0.07 0.594 0.042

1660.00 4.09 0.04 87.4 2.1 298.5 12.4 2.61 0.15 0.471 0.033

1659.75 3.80 0.07 106.3 0.9 221.5 4.4 2.32 0.03 0.727 0.017

1659.65 0.53 0.01 39.8 0.2 72.9 1.1 1.16 0.01 0.639 0.012

1657.95 3.81 0.01 126.9 1.0 177.5 1.5 1.87 0.02 0.601 0.009

1656.72 4.38 0.25 126.4 0.9 210.1 12.1 2.12 0.02 0.611 0.036

1652.36 8.61 0.49 192.2 1.2 283.8 16.3 2.54 0.02 0.500 0.029

1648.20 7.93 0.02 184.6 1.3 272.9 2.0 2.57 0.02 0.615 0.007

1647.20 9.32 0.02 204.1 1.2 290.9 1.5 2.60 0.02 0.515 0.004

Kosov,CzechRepublic

-7.60 5.60 0.39 79.7 1.0 499.9 35.7 3.78 0.07 0.235 0.017

-6.5 3.28 0.01 55.9 0.6 401.1 4.9 3.33 0.05 0.486 0.010

-6 4.26 0.01 60.9 0.7 518.7 6.1 4.25 0.07 0.570 0.011

-5.00 4.45 0.08 100.9 0.8 285.9 5.8 2.79 0.03 0.761 0.017

-3.65 9.34 0.53 134.0 1.0 511.4 29.4 4.13 0.03 0.509 0.029

-2.20 2.07 0.04 22.3 0.3 802.9 19.5 6.25 0.13 0.559 0.018

-1.00 8.35 0.16 102.7 1.1 623.6 12.8 4.66 0.06 0.243 0.006

-0.15 15.28 0.87 145.7 1.4 949.4 54.7 6.86 0.06 0.129 0.008

0.15 1.65 0.01 25.1 0.2 473.3 5.9 3.89 0.05 0.533 0.009

1.75 0.33 0.03 12.9 0.1 146.1 11.7 1.48 0.03 0.443 0.036

2.10 0.84 0.02 30.0 0.5 160.7 5.4 1.61 0.06 0.472 0.024

4.25 0.22 0.00 10.4 0.1 124.8 3.1 1.63 0.04 0.740 0.024

9.45 0.64 0.04 27.0 0.6 140.7 10.0 1.93 0.11 0.930 0.085

14.60 1.16 0.00 21.9 0.3 355.3 7.6 3.12 0.09 0.605 0.022

16.60 3.02 0.01 35.4 0.4 721.7 7.2 5.90 0.08 0.788 0.013

KlonkCore,CzechRepublic

16.37 1.93 0.07 66.3 0.5 175.0 6.4 2.02 0.03 0.789 0.031

17.53 2.77 0.01 81.6 1.1 206.7 4.3 2.13 0.06 0.682 0.024

18.67 7.30 0.26 176.3 1.5 259.9 9.5 2.44 0.03 0.619 0.024

19.82 4.45 0.16 152.2 1.2 172.9 6.3 1.87 0.02 0.662 0.026

20.73 2.60 0.01 137.6 1.4 110.7 1.6 1.78 0.03 1.006 0.024

21.22 1.98 0.01 104.8 1.1 110.6 1.6 1.76 0.03 0.984 0.024

22.25 2.37 0.01 111.1 1.1 126.8 1.8 1.93 0.04 1.044 0.025

128

23.27 1.81 0.01 64.0 0.7 176.5 2.7 2.39 0.05 1.149 0.029

23.45 3.50 0.01 76.4 1.1 304.0 5.8 3.03 0.08 0.898 0.030

23.65 6.46 0.02 148.2 1.9 282.7 4.7 2.77 0.06 0.792 0.022

23.95 2.35 0.01 59.6 0.7 248.2 4.3 2.48 0.06 0.745 0.021

24.25 4.61 0.02 142.2 1.7 198.0 3.3 2.17 0.05 0.787 0.022

24.4 6.00 0.02 158.1 1.9 234.3 4.0 2.29 0.05 0.643 0.019

24.54 8.84 0.03 182.4 1.6 321.0 3.2 3.00 0.04 0.750 0.012

24.65 3.52 0.02 124.9 1.5 167.0 3.0 1.88 0.04 0.710 0.021

24.72 3.76 0.02 125.2 1.5 181.9 3.2 2.11 0.05 0.832 0.024

24.77 4.07 0.02 161.2 1.8 146.6 2.5 1.70 0.04 0.671 0.019

24.87 5.77 0.02 150.2 1.8 235.5 4.1 2.22 0.05 0.567 0.016

24.87 6.04 0.02 159.7 1.3 232.2 2.1 2.22 0.03 0.590 0.009

25.01 3.12 0.02 84.5 1.1 229.3 4.4 2.33 0.06 0.721 0.022

25.21 5.06 0.02 123.4 1.6 258.1 4.6 2.47 0.06 0.665 0.020

25.35 6.71 0.02 136.2 1.8 325.3 5.7 2.97 0.07 0.691 0.020

25.42 2.54 0.01 111.1 1.2 137.3 2.0 2.00 0.04 1.039 0.025

25.60 6.34 0.02 116.8 1.4 370.3 5.3 3.32 0.07 0.725 0.018

25.98 3.75 0.01 118.8 1.3 190.5 2.7 2.07 0.04 0.732 0.018

26.22 2.13 0.01 99.4 1.0 123.5 1.8 1.62 0.03 0.754 0.018

26.45 5.13 0.02 152.4 1.3 215.9 2.0 2.66 0.03 1.148 0.017

26.78 10.42 0.04 175.5 2.4 411.3 6.9 3.48 0.08 0.602 0.017

27.25 6.47 0.02 208.0 2.4 186.2 3.1 1.98 0.04 0.673 0.019

27.8 17.46 0.06 200.9 1.4 729.0 3.9 5.81 0.03 0.697 0.005

28.34 4.55 0.02 105.6 0.9 274.3 2.6 2.60 0.03 0.678 0.010

29.28 6.86 0.24 185.3 1.0 229.0 8.2 2.30 0.01 0.690 0.025

30.15 9.95 0.35 224.8 1.3 283.7 10.1 2.65 0.02 0.665 0.024

31.45 6.94 0.25 206.0 1.1 204.8 7.3 2.14 0.01 0.701 0.025

129

4.3.1.1Aizpute-41Core

Figure3.Osmium(187Os/188Os,greensquares),oxygen(δ18Ocarb,bluecircles)andcarbon

(δ13Ccarb,blackcircles)isotoperatiosforshalesandcarbonatesfromtheLlandovery-

WenlockAizputecore.Biozone,lithologyandcarbonandoxygendatahavebeenadapted

fromCrameretal.2010.Seetextfordetails.

TheReandOsabundancesand187Re/188Osand187Os/188Osratiosarevariablethroughout

theAizpute-41section([Re]=1.52to27.46ppb;[Os]=65to274ppt;187Re/188Os=108to

930;187Os/188Os=1.3to7.2;Table2;Fig.3).Initial187Os/188Osvaluesrangefrom0.29to

0.73(Table2d;Fig.2).From934.82to965.28m,withinthelatestlapworthiandearliest

murchisonibiozones,187Os/188Osiincreasesfrom~0.29to~0.73.From925.65to916.5m,

187Os/188Osidecreasesfrom~0.73to~0.42.The187Os/188Osithenfluctuatesbetween~0.39

and~0.62duringthelatterpartofthemurchisonibiozone.From912.9to910.6m,spanning

130

thefirmusbiozone,the187Os/188Osiincreasesfrom0.41to0.63.Ahiatusinthe

riccartonensisbiozonepreventsfurtheranalysisintotheWenlock.

4.3.1.2BartoszyceCore

Figure4.Osmium(187Os/188Os,greensquares),oxygen(δ18Ocarb,bluecircles)andcarbon

(δ13Ccarb,blackcircles)isotoperatiosforshalesandcarbonatesfromthemid-Homerian

Bartoszycesection.Biozone,lithologyandcarbonandoxygendatahavebeenadapted

fromPorębskaetal.(2004).Seetextfordetails.

TheReandOsabundancesand187Re/188Osand187Os/188Osratiosarevariablethroughout

theBartoszyceIG-1core([Re]=0.5to15ppb;[Os]=26.2to238.4ppt;187Re/188Os=73to

131

597;187Os/188Os=1.2to4.8;Table2;Fig.4).Initial187Os/188Osvaluesrangefrom0.47to

0.73(Table2;Fig.2c).From1672.9to1662.3m,fromthelatestlundgrenitothebaseof

thenassabiozone,the187Os/188Osiincreasesfrom~0.52to~0.71.Immediatelyafterwards,

the187Os/188Osidecreasesfrom~0.71to~0.51whereitremainsrelativelylow(~0.5)

between1661.7to1660m.From1660to1659.7m,the187Os/188Osisharplyincreasesfrom

~0.47to~0.73.The187Os/188Osithenproceedstodecreasethroughouttherestofthenassa

biozone.

4.3.1.3Kosovsection

TheReandOsabundancesand187Re/188Osand187Os/188Osratiosarevariablethroughout

theKosovsection([Re]=0.2to15.3ppb;[Os]=10.4to145.7ppt;187Re/188Os=124.8to

949.4;187Os/188Os=1.5to6.9;Table2;Fig.5).Initial187Os/188Osvaluesrangefrom0.13to

0.93(Table2;Fig.2b).From7.6to5m,the187Os/188Osiincreasesfrom~0.23to~0.76.Prior

tothebaseofthedubiuspostfrequensbiozone,the187Os/188Osidecreasesfrom~0.76to

~0.13between5and0.13m.The187Os/188Osithenincreasesfrom~0.13to~0.93between

0.13and-9.45mbeforedecreasingagaintowardstheendofthedubiuspostfrequens

biozone

132

.

Figure5.Osmium(187Os/188Os,greensquares),Lithium(δ7Li,redsquares),oxygen(δ18Ocarb,

bluecircles)andcarbon(δ13Ccarb,blackcircles)isotoperatiosforcarbonatesandshales

fromtheLudfordianKosovsection.Biozone,lithologyandcarbonandoxygendatahave

beenadaptedfromFrýdaandManda(2013).Seetextfordetails.

133

4.3.1.4KlonkCore

Figure6.Osmium(187Os/188Os,greensquares),oxygen(δ18Ocarb,bluecircles)andcarbon

(δ13Ccarb,blackcircles)isotoperatiosforcarbonatesandshalesfromtheSilurian-Devonian

GSSPatKlonk.BiozoneinformationadaptedfromCricketal.(2001).Carbonandoxygen

isotopedatahavebeenadaptedfromBuggischandMann(2004).Seetextfordetails.

TheReandOsabundancesand187Re/188Osand187Os/188Osratiosarevariablethroughout

theKlonkcore([Re]=1.8to17.4ppb;[Os]=59.6to224.8ppt;187Re/188Os=110to729;

187Os/188Os=1.6to5.8;Table2;Fig.6).Initial187Os/188Osvaluesrangefrom0.57to1.15

134

(Table2;Fig.2a).From31.45to26.78m,the187Os/188Osiaremoderatelyradiogenic,

rangingfrom~0.6to~0.7.Between26.78and25.35m,the187Os/188Osifluctuatesbetween

~0.6and~1.15.AcrosstheSilurian-Devonianboundary,the187Os/188Osiisalsomoderately

radiogenic,withvaluesbetween~0.57and~0.83(25.35-23.95m).From23.95to23.27m,

the187Os/188Osiincreasesfrom0.74to1.15beforesubsequentlydecreasingfrom1.15to

0.62duringtheearliestDevonian.

4.3.2Lithiumisotopeandtracemetaldata

LithiumisotopemeasurementsandtracemetaldatafortheHunninge-1(Sweden)and

Kosov(CzechRepublic)samplesarepresentedinTable3.

Table3

LithiumisotopeandtracemetaldatafortheHunninge-1andKosovcarbonatesamples.

Depth δ7Li 2s.d. Mg/Ca Al/Ca Mn/Ca Sr/Ca(m) (‰) (μmol/mol) (μmol/mol) (μmol/mol) (μmol/mol)

Lusklint&Lickerdhamn,Sweden38 11.0 0.2 6.50 0.27 0.51 0.1730 12.3 0.2 4.81 0.17 0.49 0.0921 11.1 0.3 5.11 0.15 0.45 0.1014 11.6 0.6 4.39 0.12 0.38 0.124 11.6 0.5 5.53 0.18 0.50 0.181 11.1 0.8 4.45 0.17 0.47 0.09

-0.19 13.5 0.5 4.00 0.15 0.41 0.09-0.66 12.2 0.1 6.63 0.22 0.38 0.20-2.25 16.1 0.2 5.36 0.14 0.29 0.22-3.27 17.6 0.5 4.18 0.21 0.22 0.15-4.2 16.1 0.5 4.15 0.08 0.26 0.12-5.8 14.3 0.1 5.15 0.20 0.32 0.17

Hunninge-1Drillcore,Sweden -4.4 11.2 0.7 39.32 0.536 2.74 0.75

-2.8 11.4 0.3 28.51 0.290 3.64 0.74-1.5 11.4 0.5 40.49 1.006 2.46 0.66-1.1 10.6 0.5 32.11 0.521 1.76 0.29-0.7 9.9 0.1 20.51 0.123 1.97 0.39-0.4 12.2 0.2 41.29 0.827 1.89 0.60-0.2 12.5 0.2 40.12 0.536 1.47 0.680.2 12.7 0.3 48.25 1.002 1.07 0.96

135

0.7 12.6 0.6 24.21 0.108 0.67 0.831 13.4 0.3 15.61 0.007 0.50 0.701.5 13.8 0.2 19.00 0.007 0.53 0.631.8 14.7 0.6 24.94 0.114 0.49 1.274.8 13.0 0.7 22.07 -0.005 0.61 1.35

Kosov,CzechRepublic -4.77 9.6 0.6 13.64 -0.018 1.23 0.56-2.65 6.2 0.5 12.69 0.066 0.78 0.55-2.2 7.8 0.4 14.60 0.014 0.74 0.51-1.25 5.8 0.6 14.48 0.108 0.85 0.44-1 4.6 0.1 12.34 0.200 0.85 0.79-0.7 7.8 0.6

-0.45 6.2 0.4 -0.05 15.4 0.2 10.93 0.052 0.47 0.97

0.5 6.2 0.7 0.9 9.0 0.3 13.24 0.129 0.35 0.95

1.15 5.3 0.5 1.5 6.3 0.6 2.7 6.2 0.2 3.3 5.6 0.2 4.05 6.1 0.1 13.67 0.072 0.41 0.99

5.28 12.8 0.6 13.29 0.065 0.31 0.98

4.3.2.1Lusklintsection

TheMg/Ca,Al/Ca,Mn/CaandSr/CaratiosarevariablethroughouttheHunninge-1drillcore

(Mg/Ca=4.38to6.5μmol/mol;Al/Ca=0.12to0.27μmol/mol;Mn/Ca=0.38to0.51

μmol/mol;Sr/Ca=0.09to0.18μmol/mol;Table3;Fig.7).TheAl/Caratiosremainbelow

the~0.8mmol/molthreshold,suggestinglittleinfluencefromLileachedfromclays.The

δ7Livaluesrangefrom11to12.3‰(Table3;Fig.7).Theδ7Livaluesremainrelatively

constantthroughoutthebicornisandprocerusbiozones.

136

Figure7.Lithium(δ7Li,redsquares)andcarbon(δ13Ccarb,blackcircles)isotoperatiosfor

carbonatesfromtheLlandovery-WenlockLusklintsection.Biozone,lithologyandcarbon

datahavebeenadaptedfromMaier(2010).Seetextfordetails.

4.3.2.2Lickershamnsection

TheMg/Ca,Al/Ca,Mn/CaandSr/CaratiosarevariablethroughouttheHunninge-1drillcore

(Mg/Ca=4to6.63μmol/mol;Al/Ca=0.08to0.22μmol/mol;Mn/Ca=0.22to0.41

μmol/mol;Sr/Ca=0.09to0.22μmol/mol;Table3;Fig.7).TheAl/Caratiosremainbelow

the~0.8mmol/molthreshold,suggestinglittleinfluencefromLileachedfromclays.The

δ7Livaluesrangefrom12.2to17.6‰(Table3;Fig.8).Theδ7Livaluesrelativelylow(~13

‰)atthebaseoftheupperprocerusbiozonesbeforerisingto17.6‰bytheendofthe

upperbiozone,decreasingagainto~14.3‰duringthelowerranuliformiszone.

137

Figure8.Lithium(δ7Li,redsquares)andcarbon(δ13Ccarb,blackcircles)isotoperatiosfor

carbonatesfromtheLlandovery-WenlockLickershamnsection.Biozone,lithologyand

carbondatahavebeenadaptedfromMaier(2010).Seetextfordetails.

138

4.3.2.3Hunninge-1drillcore

Figure9.Lithium(δ7Li,redsquares),oxygen(δ18Ocarb,bluecircles)andcarbon(δ13Ccarb,

blackcircles)isotoperatiosforcarbonatesfromthemid-HomerianHunningecore.Biozone,

lithologyandcarbonandoxygendatahavebeenadaptedfromCalneretal.(2006).Seetext

fordetails.

139

TheMg/Ca,Al/Ca,Mn/CaandSr/CaratiosarevariablethroughouttheHunninge-1drillcore

(Mg/Ca=15.6to48.3μmol/mol;Al/Ca=-0.005to1.006μmol/mol;Mn/Ca=0.49to3.64

μmol/mol;Sr/Ca=0.29to1.35μmol/mol;Table3).ThemajorityofAl/Caratiosremain

belowthe~0.8mmol/molthreshold,suggestinglittleinfluencefromLileachedfromclays.

However,samplesfromadepthof-1.5and0.2mhaveanAl/Caof~1suggestingpossible

influencefromLileachedfromclays.Theδ7Livaluesrangefrom9.9to14.7‰(Table3;Fig.

9).From4.4to1.5m,justpriortothebaseofthenassabiozone,δ7Liremainsrelatively

constant(~11.3‰).Justafterthebaseofthenassabiozone,δ7Lidropsto9.9‰before

rapidlyincreasingto14.7‰between0.7and-1.8m.Theδ7Lithendecreasesto13‰.

4.3.2.4Kosovsection

TheMg/Ca,Al/Ca,Mn/CaandSr/CaratiosarevariablethroughouttheKosovsection

(Mg/Ca=10.93to14.6μmol/mol;Al/Ca=-0.02to0.2μmol/mol;Mn/Ca=0.31to1.23

μmol/mol;Sr/Ca=0.44to0.99μmol/mol;Table3).Al/Caremainsbelowthe0.8mmol/mol

threshold,suggestinglittleinfluencefromLileachedfromclays.Theδ7Livaluesrangefrom

4.6to15.4‰(Table3;Fig.2e).From4.77to1m,priortothebaseofthedubius

postfrequensbiozone,δ7Lidecreasesfrom9.6to4.6‰.Between1and0.05m,theδ7Li

increasesfrom4.6to15.4‰beforedecreasingagainto5.3‰by-1.15m.Theδ7Livalues

remainrelativelyconstantat~6‰beforesubsequentlyincreasingto12.8‰between-

4.05and-5.28m.

4.4Discussion

TheOsandLiisotoperecords(Fig.1andFigs.3to9)showsimilarprofilesforeachtime

periodstudied,butwithdifferingmagnitudesofchange.Priortotheδ13Cexcursionthereis

140

changeof0.19to0.56inthe187Os/188Oscompositiontomoreradiogenicvalues.This

changeinthe187Os/188Oscompositionisoftenassociatedwithanincreaseinδ18Oof

between0.55and1.74‰(Fig.1andFigs.3to9).Thisisfollowedbyadeclineinthe

187Os/188Oscompositiontopreviousvalues.Thisintervalcontinuestobecharacterizedby

highδ18Ovalues.Duringtheδ13Cexcursionofbetween0.9and8.29‰,the187Os/188Os

compositiongenerallyremainslow(unradiogenic).Incontrast,theδ7Liincreasesby4.8to

9.2‰.Duringtherelativelyplateauedδ13Cinterval,theδ18Oandδ7Livaluesbeginto

returntopre-excursionvalues.Duringthisdecline,the187Os/188Osvaluesincreaseby0.26

to0.8beforereturningtothemoreunradiogenicpre-excursionvalueseitherintimewith

thedescendinglimboftheδ13Crecordorpriortoit,withexceptionoftheTelychian-

Sheinwoodianboundary,whichhasahiatusaftertheascendinglimboftheδ13Cexcursion.

Processesthatcouldcausethesevariationsincludecontaminationduringsample

processing,diagenesis,oraprimaryseawatersignaldrivenbylocalorglobalchangesin

Earthsystemprocesses.ContaminationofReandOsfromthedetritalfractionofanalysed

shaleswasavoidedusingbytheCrO3–H2SO4digestionmethod,whilecationexchangeor

leachingofclays,whichcouldimpartanisotopicallylightδ7Lisignal,wasmonitoredby

analysingcation/Caratiosofthecarbonatesamples(Seesection4.2).Diagenesiscanbe

discountedbecausesimilartrendsandabsolutevaluesarerepeatedinsectionsfrom

temporallyseparatedsections(Thisstudy;Finlayetal.,2010;PoggevonStrandmannetal.,

inreview).Furthermore,carbonandoxygenisotopesinstudiedprofilesshowsimilarvalues

toothercorrelatedsectionsthatspanthesameintervals(Fig.1;SaltzmanandThomas,

2012;Trotteretal.,2016,Crameretal.,2011).Itisthereforesuggestedthattheisotopic

shiftsinthecarbonateandshalesectionsmustrepresentprimaryseawatersignatures.

InthefollowingsectionwewilldiscusshowchangesinavarietyofEarthsystem

processescaninfluenceboththe187Os/188Osandδ7Liofseawater,anddetermineifthese

141

processescouldhavecausedthevariationsinisotopedataseeninthisstudyusingdynamic

OsandLicyclemodels.Initiallywewillexploretraditionaloceancirculationmodels(Bickert

etal.,1997;Jeppsson,1990)usedtoexplainlithologicalandcarbonandoxygenisotope

variationsduringtheSilurian.Wewillthendiscussthepotentialofrapidclimaticcooling

(e.g.Lechleretal.,2015;PoggevonStrandmannetal.,2013;Ravizzaetal.,2001),flood

basaltvolcanism(e.g.DuVivieretal.,2014;Lechleretal.,2015;PoggevonStrandmannet

al.,2013)andhydrothermalactivityaspossibleexplanations.FinallywewilldiscussSilurian

glacialexpansionsoverGondwanaasadriverofglobalenvironmentalchange(Azmyetal.,

1998;Brandetal.,2006;Kaljoetal.,2003;Trotteretal.,2016).

4.4.1IsotopicconstraintsonSilurianseawaterchemistry

4.4.1.1Climaticallyinducedchangesinoceancirculation

Severalocean-climatemodelsweredevelopedduringthe1990stotryandexplainthe

observedchangesinfaunalturnover,lithologyandcarbonandoxygenisotopes.Thefirst

modelwasestablishedbyJeppsson(1990)toexplainobservedchangesinthelithologyand

conodontfaunasofGotland.Themodelproposesthatmajorbio-eventsoccurredduring

thetransitionbetweentwostableoceanicstates,knownasPrimoandSecundoEpisodes,

drivenbyachangeinthelatitudinalpositionofdeep-waterformation.Primoepisodes

werecharacterisedbylowatmosphericCO2concentration,coolerclimatesandrelatively

lowglobalsealevel.Downwellingofcold-densehighlatitudesurfacewatersventilatedthe

deepocean,causingupwellingofoxic,nutrient-richwatersatlowlatitudes,whichdrovean

increaseinprimaryproductivityandshelffaunadiversity.Lowlatitudeclimatewashumid,

andhighrainfallwouldhaveintensifiedcontinentalweathering,deliveringabundantclay

andnutrientstoshelfseas.TheSecundoepisodeswerecharacterisedbyhigher

atmosphericCO2,warmerclimatesandthereforethermalexpansionoftheocean.Cold-

142

densewatersnolongerformedathighlatitudes,andsalinitydrivendeep-waterformation

occurredatintermediatelatitudesinstead.Thiscreatedastratifiedoceanwithlower

primaryproductivityandshelffaunadiversity.Thelowlatitudeclimatewasdry,leadingto

lowerrun-offandadeclineincontinentalweatheringintensity.TheIreviken,Muldeand

Laueventsallsupposedlyculminatedfromaswitchbetweenthesetwooceanicstates

(Jeppsson,1998).

Aspreviouslyexplained,theswitchbetweenPrimoandSecundoepisodeswould

havehadaprofoundinfluenceonlowlatitudecontinentalweatheringintensityand

thereforetheriverinefluxandisotopiccompositionofosmiumandlithiumdeliverytothe

ocean.DuringthePrimoepisodes,warmertemperatures,ahumidclimateandhigher

precipitationrateswouldbeassociatedwithenhancedchemicalweatheringratesof

radiogeniccontinentalcrustresultinginanincreaseinthe187Os/188Oscompositionof

seawater(Peucker-EhrenbrinkandRavizza,2000;Peucker-EhrenbrinkandRavizza,2012;

Ravizzaetal.,2001).Highintensityweatheringandextensiveclayformation,asproposed

bytheJeppsson(1990)model,wouldlead7Li-depletedclaysandothersecondaryminerals

toremainintheweatheringzoneforalongtime,leadingtodissolutionofthesephasesand

relativelylowδ7Liinriverwaterandthereforeadecreaseintheδ7Liofseawater(Bouchez

etal.,2013;Dellingeretal.,2015).DuringtheSecundoepisodes,lowertemperatures,a

morearidclimateandlowerprecipitationrateswouldbeassociatedwithreducedchemical

weatheringratesandthereforeresultinaloweringthe187Os/188Oscompositionof

seawater(Peucker-EhrenbrinkandRavizza,2000;Peucker-EhrenbrinkandRavizza,2012).A

lowerweatheringintensityandreducedclayformationwouldcause6Litoberetainedin

precipitatedsecondaryminerals,leadingtoisotopefractionationandrelativelyhighδ7Liin

thedissolvedloadofriversandanincreaseintheδ7Liofseawater(Bouchezetal.,2013;

Dellingeretal.,2015).

143

Accordingtotheabovescenarios,aswitchbetweenthesetwo(Primo-Secundo)

statesduringtheIreviken,MuldeandLaueventswouldbeassociatedwithpermanent

changefromrelativelyradiogenic187Os/188Osandlowδ7Litorelativelyunradiogenic

187Os/188Osandhighδ7Li.However,theobservedvariationsin187Os/188Osandδ7Li,

associatedwithcontinentalinputs,fromthisstudy(Fig.1)donotfollowthistrend.

Therefore,ourdatadoesnotsupportaswitchbetweenPrimoandSecundoepisodesasa

driverofSilurianclimateperturbations,atleastintermsofcontinentalprecipitation

changes.

ThemodelbyJeppsson(1990)waslaterfurtherdevelopedbyBickertetal.

(1997)usinggeochemicaldata.TheynoticedthattheIreviken,MuldeandLaueventswere

allassociatedwithanomaliesintheδ13Candδ18OrecordfromGotland.Bickertetal.(1997)

arguedthatthepositiveshiftsinδ18Oduringtheseeventsweredrivenbysalinity

fluctuationsassociatedwithchangesinglobalevaporation-precipitationratesand

continentalrun-off.AsaresulttheoceancirculationmodelofJeppsson(1990)wasadapted

accordinglytothestableisotopefluctuationsintoshiftsbetweenhumidperiods(H-

periods)withhighcontinentalinputandupwellingincoastalwaters,andaridperiods(A-

periods)withlittlecontinentalrunoffanddownwellingincoastalwaters.

Generally,themodelsuggeststhattheSilurianclimateisinahumidstate

punctuatedbyshortaridepisodes.Afullhumid-aridcycle(A-Hmodes)duringtheIreviken,

MuldeandLaueventswouldbeassociatedwithadeclineinlow-latitudecontinental

weatheringintensityduringA-periods.TotesttheA-Hmodesaseriesofdynamicmodels

wereruntoconstraintheinfluenceofagradualdecreaseinglobalcontinentalweathering

ratesby~50%,over250kyr,ontheOsandLiisotopesystems(Fig.10).Duringaswitch

fromanH-toA-period,adeclineinriverinefluxesrelatedtosilicateand/ororganic-and

sulphide-richlithologyweathering(Fig.10b)causesashifttolessradiogenic187Os/188Os

144

values(Fig.10a),duetoareductioninthefluxofradiogenicOsintotheocean.The

187Os/188Osvaluesproceedtoplateauatrelativelyunradiogenicvaluesfor~750kyr(Fig.

10a),beforerisingtomoreradiogenicvaluesastheriverineinputofmoreradiogenicOs

resumesastheclimatereturnstoahumidmode(Fig.10b).ForLi,theδ7Ligradually

increasestomorepositivevalues,peakingtowardstheendofthearidperiod(Fig.10c).

DuringtheswitchbacktoanH-period,thefluxofcontinentalderivedLiincreases(Fig.

10d),drivingtheδ7Libacktopreviousvaluesoverthenext5Myr(Fig.10d).

Figure10.DynamicmodelofOs(a)andLi(c)isotopesforchangesincontinental

precipitationratesaccordingtoBickertetal.(1997).Themodelshownwasgeneratedby

assuminga50%dropincontinentalweatheringandthereforeriverineflux(b,d)duringa

switchfromahumid-period(H)toanarid-period(A).InthecaseofLi,theweathering

congruencyismodelledbyvaryingtheisotoperatiooftheriverineendmember(d).See

textfordetails.

Thisissupportedbythedatafromthisstudy(Fig.1),whichshowslessradiogenic

187Os/188Oscompositionsatthesametimeasthepeakinδ7Livaluesofsimilarmagnitudes

tothemodel(Fig.10)duringtheA-periodsassociatedwiththeMuldeandLauevents.We

145

alsoseelessradiogenic187Os/188OscompositionsduringtheIrevikenandKlonkevents(Fig.

1).However,wewouldexpectthecorrespondingH-periodstobeassociatedwithrelatively

radiogenic187Os/188Oscompositions(Fig.10a),whichwouldpersistformuchoftheSilurian.

Incontrast,short-livedpeakscharacterizedbyradiogenic187Os/188Oscompositionsare

observed,withbackgroundlevelsduringpredictedH-periodssimilartothedurationofthe

A-period(Fig.1).Inaddition,theascendinglimbtowardsradiogenic187Os/188Os

compositionsisassociatedwithanincreaseinδ18Oinmostinstances(Fig.1).Accordingto

theBickertetal.(1997)model,thisshouldbetheconverse,withadeclinetounradiogenic

187Os/188Oscompositionsduringanincreaseinδ18Oasalowercontinentalfluxisassociated

withmoresalinecoastalwaters.Thisiscompoundedbypreviouscriticismsthatarguethe

conceptofgloballyincreasingaridityorhumidityisnotsupportedbymodellingorexisting

data,andisthereforedifficulttoargueasacauseforglobalevents(Johnson,2006;Kaljoet

al.,2003;Loydell,1998;Munneckeetal.,2010).

4.4.1.2FloodBasaltVolcanism

Periodsofintensesubmarineorcontinentalvolcanismandtheemplacementoflarge

igneousprovinces(LIPs)haveaprofoundinfluenceonglobalclimateandare

penecontemporaneouswiththeonsetofoceananoxicevents(OAE).Additionally,newly

formedbasalticterrainscanhavealargeimpactonsecularvariationsintheosmium

(Bottinietal.,2012;CohenandCoe,2002;DuVivieretal.,2014;RavizzaandPeucker-

Ehrenbrink,2003;TurgeonandCreaser,2008)andlithium(Lechleretal.,2015;Poggevon

Strandmannetal.,2013)isotoperecords.Submarinefumaroleandhydrothermalalteration

/weatheringofjuvenilebasaltsdeliversafluxofunradiogenicOstotheoceans,drivingthe

187Os/188Oscompositionofseawatertowardsmantlevalues.Further,duringOAEs,abrupt

globalwarmingassociatedwithrisingatmosphericCO2causedenhancedweatheringof

146

maficsilicatematerialwhichdeliveredlowδ7Li(high[Li])inputstotheocean,drivingthe

δ7Liofseawatertolowervalues.Therefore,anyigneousactivityduringtheSilurian

intervalsstudiedherewouldbemetbyarapiddecreaseinboththe187Os/188Osandδ7Liof

seawater,whichisnotobserved(Fig.1).Furthermore,thisiscompoundedbythelackof

evidenceforLIPsduringtheSilurian.

4.4.1.3Temperature-weatheringfeedbacks

Overgeologicaltimescales(Myr)temperaturehaspartlybeencontrolledbyinteractions

betweenatmosphericCO2andcontinentalweathering(Berneretal.,1983;Walkeretal.,

1981).Risingtemperaturesstimulateincreasedchemicalweatheringofsilicaterocks

drawingdownCO2fromtheatmosphere,leadingtoadeclineintemperatureandviceversa

(Berneretal.,1983;Walkeretal.,1981).Rapidtemperaturefluctuationsinthegeological

pastcouldinfluencecontinentalweatheringandthereforesecularvariationsinbothOsand

Liisotoperecords.ThisissupportedbyOsandLiisotopedatafromthePaleocene-Eocene

ThermalMaximum(PETM)andOAEsrespectively,whichareinterpretedtoreflectabrupt

releaseofgreenhousegaseswhichresultedinanincreaseinglobaltemperatures,

stimulatingcontinentalsilicateweatheringanddeliveringradiogenicOsandisotopically

heavyLitotheocean(Lechleretal.,2015;PoggevonStrandmannetal.,2013;Ravizzaet

al.,2001).

Silurianoxygenisotoperecordsofphosphates(Fig.1)(Trotteretal.,2016)show

globallyrecognisedpositiveexcursions,indicativeofcoolingduringtheIreviken,Muldeand

Lauevents.Likewise,oxygenisotopedatafromsectionsstudiedhereshowanincreasein

δ18Ointimewith,orslightlypreceding,theOsandLiisotopeexcursions(Fig.1).According

totheabovetheory,arapidcoolingwouldbeassociatedwithadeclineinglobal

continentalsilicateweatheringandthereforedecreasethefluxoftheradiogenicOsand

147

isotopicallyheavyLitotheocean,drivingthe187Os/188Osandδ7Licompositionofseawater

tolowerandhighervaluesrespectively.However,thisstudyshowsanincreaseinthe

187Os/188Oscompositionofthehydrogenouscomponentoftheorganic-richsedimentary

units,andthusbyinferencethatofthecontemporaneousseawaterduring,orslightly

precedingtheincreaseintheδ18Ovalues;withtheexcursiontomorepositiveδ7Livalues

occuringsometimeaftertheinitialriseintheδ18Ovalues(Fig.1).Thissuggeststhat

temperaturechangealoneisnotthedriverofOsandLiisotopevariationsseeninthis

study.

4.4.1.4Hydrothermalactivity

Inthemodernoceans,hydrothermalactivityatmid-oceanridgesaccountsforasignificant

inputofOsandLitotheocean,witha187Os/188Oscompositionandδ7Livalueof~0.12and

~8‰respectively(HathorneandJames,2006;MisraandFroelich,2012;Peucker-

EhrenbrinkandRavizza,2000).Theisotopiccompositionofthesefluxesisseentobe

relativelyconsistentthroughtime,howeverachangeintheirfluxcouldcausevariationsin

theisotopiccompositionofseawater.Althoughhardtoconstrain,areductioninpluton

emplacement(Hardie,1996)andanincreaseintheMg/Caratioofseawater(Stanleyand

Hardie,1998)suggestsareductioninsea-floorspreadingduringthelate-Wenlocktoearly-

Ludlow(Crameretal.,2011b).However,thelowresolutionofavailabledatapreventsan

estimationofvariationsofhydrothermalactivityovertherelativelyshorttimescales

studiedhere.

148

Severaldynamicmodelswereruntoconstraintheinfluenceofshortperiodic

fluctuationinthefluxofhightemperaturehydrothermalactivityontheOsandLiisotope

systems(Fig.11).Twoscenariosweremodelled:1.Apulseddecreaseinthehydrothermal

fluxby~50%(Fig.11b);and,2.Asingle,extendedperiodofreduced(~50%)hydrothermal

flux(Fig.11d).Duringthepulsedscenarioweseeasimultaneousincreasetomore

radiogenic187Os/188Oscompositionsandmorepositiveδ7Livalueswitheachdecreasein

hydrothermalflux(Fig.11a).Althoughthe187Os/188Oscompositionsbecomemore

unradiogenicintimewithanincreaseinhydrothermalflux,theδ7Livaluesshowadelayed

response,decliningovertheproceeding250kyr(Fig.11a).Thiscompareswellwiththe

187Os/188Osdatafromthisstudy(Fig.1),whichshowtwouniformpeaksinradiogenic

187Os/188Oscompositions.However,themagnitudeofchangepredictedbythemodel(0.07)

ismuchsmallerthanmeasuredisotoperatios.Likewise,theδ7Livaluesmeasuredhere(Fig.

1)showonesingularpeakinδ7Litowardsthesecondpeakin187Os/188Oscompositions,

whichisnotseeninthemodel(Fig.11a).

Figure11.DynamicmodelofOsandLiisotopesforapulsedreductioninhydrothermalflux

(a)andanextendedperiodofreducedhydrothermalflux(c).Themodelshownwas

149

generatedbyassuminga50%dropinhydrothermalfluxovertworelativelyshort250kyr

periods(b)oroverasingle,moreextended750kyrperiod(d).Seetextfordetails.

Duringanextendedperiodofreducedhydrothermalflux(Fig.11d),the

187Os/188Oscompositionsandδ7Livaluesincreaseoverasimilarperiodoftime(Fig.11c).

ThiscompareswellwithLiisotopedatafromthisstudy(Fig.1),whichshowsasingular

peakinδ7Li.However,themagnitudeofchangepredictedbythemodel(<0.5‰)isfartoo

smallwhencomparedtotheexcursionobserved(>4‰).Likewise,the187Os/188Os

compositiondeterminedhere(Fig.1)showtwopeaksinradiogenic187Os/188Osvalues,

whichisnotseeninthemodel(Fig.11c).Inbothmodeloutputs(Fig.11aandc)theδ7Li

decreasesby>0.3‰belowbackgroundlevelsaftertheexcursionduetoadependenceof

thepartitioncoefficientonseawaterconcentration,whichremainslowevenaftertheinput

hasincreased.Thesemodellingstudies(Fig.11)suggestthatfluctuationsinthe

hydrothermalfluxcannotcausethevariationsinOsandLiisotopedataseenhere(Fig.1).

4.4.1.5Glaciation

OsmiumandLiisotopevariationsfoundinthisstudybearastrikingresemblancetothose

measuredfortheHirnantianglaciation(Finlayetal.,2010;PoggevonStrandmannetal.,in

review),some12MyrsearlierthantheTelychian-Sheinwoodianboundary(Fig.1).Authors

postulatedthesevariationsweredrivenbyfluctuationsinchemicalweatheringrates

relatedtoenhancedcontinentalicevolumeoverGondwana.DuetothesimilaritiesinOs

andLiisotopetrendsdeterminedhere,combinedwithreoccurringglobaltrendsinoxygen

isotopes(Azmyetal.,1998;Calner,2008;Lehnertetal.,2010;Munneckeetal.,2010;

Nobleetal.,2005;Trotteretal.,2016),carbonisotopes(Crameretal.,2011a;Crameret

al.,2010;Munneckeetal.,2010;SaltzmanandThomas,2012),sea-levelreconstructions

(HaqandSchutter,2008;Johnson,2006,2010;Loydell,1998),marinefaunalturnover

150

(Calner,2008;Cooperetal.,2013;Melchinetal.,2012),andinthecaseoftheearly

Sheinwoodian,thepresenceofglacialsediments(Díaz-MartínezandGrahn,2007),we

suggestchangesincontinentalweatheringcausedbyglaciationeventscouldhaveoccurred

duringeachofthefourSiluriantimeintervalsstudied.

Duringglacialadvance,physicalerosionoftheunderlyingbedrockprovides

abundantfreshlycomminutedrockflourtoice-sheetmarginsandtheproglacialzone

(Tranter,1982).Subglacialandproglacialconditionspromotesulphideoxidationandthe

chemicalweatheringofcarbonatesinthesenewlyformedreactivemineralsurfaces

regardlessofunderlyinglithology(Andersonetal.,2000;Cooperetal.,2002;Fairchildet

al.,1999;Tranteretal.,2002)whilststimulatingthemicrobiallymediatedoxidationof

ancientorganicmattersuppliedbycomminutedshales(Petschetal.,2001a;Wadhamet

al.,2004).SedimentarysulphidesandorganicmatterareoftenassociatedwithhighOs

concentrationsandhighlyradiogenic187Os/188Oscompositions(Jaffeetal.,2002;Peucker-

EhrenbrinkandHannigan,2000;Peucker-EhrenbrinkandRavizza,2000;Pierson-Wickmann

etal.,2002;RavizzaandTurekian,1989),andtheirerosionandsubsequentoxidationby

advancingice-sheetswouldimpactontheglobalriverineOsend-member,drivingseawater

tomoreradiogenicvalues.Theinitialpeakin187Os/188Osvaluesseeninthisstudy(Fig.1)

andduringtheHirnantianglaciation(Finlayetal.,2010)canthereforebeattributedto

glacialadvance.

However,lowtemperaturesinnewlyglaciatedregionswouldacttosuppress

chemicalsilicateweatheringtolowerratesthanpreviouslynon-glaciatedregions

(Anderson,2005,2007;Andersonetal.,2000;Andersonetal.,1997;Gislasonetal.,2009;

MaherandChamberlain,2014).Ifphysicalerosionratesincrease,whilesilicatechemical

weatheringremainsconstantordecreases,weatheringintensitywilldeclineandtherefore

increasethedissolvedδ7Livaluesofrivers(Bouchezetal.,2013;Dellingeretal.,2015;Li

151

andWest,2014;PoggevonStrandmannetal.,2017).Likewise,regionsoflowchemical

weatheringintensitywilldrivetheδ7Liofglacialriverstomorepositivevalues,as6Liis

retainedbysecondaryminerals(PoggevonStrandmannetal.,2006),andtheformationof

Fe-oxyhydroxidesduringsulphideoxidationunderice-sheetswillalsopreferentiallyuptake

6Liontomineralsurfaces,contributingtotheriseintheδ7Liofglacialwaters(Wimpennyet

al.,2010).Anincreaseintheδ7Livaluesoftheriverinefluxcombinedwithadecreaseinthe

overallglobalriverinefluxofLi,duetothegradualcoveringofGondwanabycontinental

ice-sheetspreventingweatheringoftheunderlyingsilicateminerals(KumpandAlley,1994;

PoggevonStrandmannetal.,inreview),willdrivetheδ7Livalueofseawatertomore

positivevalues(Fig.1).

Asglacialexpansionbegantoslowandglacialmaximumwasestablished,the

previouslyhighdenudationrateswouldhavediminishedalongwiththeoxidationof

proglacialandsubglacialsulphidesandorganicmatter,reducingtheinputofOswitha

radiogenic187Os/188Oscompositiontotheoceans.Whendiminutivedenudationiscoupled

withenhancedlowlatitudecontinentalcoverbyicesheetsandgenerallycolder,drier

conditions,overallglobalchemicalweatheringrateswouldhaveremainedlowunder

glacialmaximum(KumpandAlley,1994),maintaininghighδ7Livaluesandlow187Os/188Os

compositionsinseawater(Fig.1)(Finlayetal.,2010;PoggevonStrandmannetal.,in

review).FortheHirnantianglaciation,authorshavesuggestedthatthistransientdeclinein

silicateweathering,andthereforeadeclineinoneoftheEarth’smajoratmosphericCO2

withdrawalmechanisms,wouldhavecausedanincreaseinatmosphericCO2thatultimately

terminatedtheglaciation(Kumpetal.,1999;PoggevonStrandmannetal.,inreview).The

similaritybetweenLiisotoperecordsfortheHirnantian(PoggevonStrandmannetal.,in

review)andthisstudy(Fig.1)suggestasimilarmechanismfordeglaciationcouldhave

occurredthroughouttheSilurian.

152

Whateverthecauseofdeglaciation,theextensiveavailabilityoffreshmaterial,

increasedmeltwaterandgenerallywetterconditionsduringthedemiseofcontinentalice

sheetswouldenhancetheoxidationand/orweatheringofsulphides,shalesandsilicates

leftbehind(BluthandKump,1994;DreverandZobrist,1992;Gaillardetetal.,1999;Huh

andEdmond,1999;Meybeck,1987;Petschetal.,2001b;Vanceetal.,2009;Wildmanetal.,

2004).ThiswouldcauseashiftintheglobalriverineOsend-member,andtherefore

seawater,tomoreradiogenicvalues(Finlayetal.,2010;Peucker-EhrenbrinkandBlum,

1998)creatingthesecondpeakinthe187Os/188Oscompositionrecord(Fig.1).Greater

weatheringintensityofsilicateswilldecreaseLiisotopefractionationbetweenthe

dissolvedandsuspendedloadsofglacialrivers,asless6Liisretainedinsecondaryminerals,

drivingδ7Litolowervalues(Huhetal.,1998;PoggevonStrandmannetal.,2006).Asthese

newlyformedriversystemsbecomemoremature,continuedweatheringofsuspended

materialwillincreasethesaturationstateofthedissolvedloadwithrespecttosecondary

minerals,loweringδ7Livaluesfurther(Dellingeretal.,2015;PoggevonStrandmannetal.,

2006).Adecreaseintheδ7LivalueandLiconcentrationofriverineend-memberswould

leadtoadecreaseinδ7Livalueofseawater,eventuallyrestoringthesystemtopre-

excursionvalues(Fig.1).

Severaldynamicmodelswereruntoconstraintheinfluenceofglacialprocesses

ontheOsandLiisotopesystems(Fig.12).Biozoneboundaryages(Melchinetal.,2012)

wereusedtoconstrainthetimingsofglaciation(~250kyr;lightblueboxdenotedbyGin

Fig.12),glacialmaximum(~500kyr;darkblueboxdenotedbyGMinFig.12)and

deglaciation(~250kyr;lightblueboxdenotedbyDinFig.12).Duringglaciation,the

relativecontinentalfluxrelatedtosilicateweathering(Fig.12band12e)gradually

decreasesby50%,whiletheweatheringoforganic-sulphide-richlithologiesincreasesby

~80%(Fig.12c).InthecaseoftheOsisotopesystem,thiscausesaswitchfromariverine

budgetdominatedbysilicateweathering(60%oftotalFriv),toariverinebudgetdominated

153

byORLW(70%oftotalFriv).ThehighlyradiogenicnatureofORL(187Os/188Os=1.34;Table1)

whencomparedtosilicateminerals(187Os/188Os=0.6;Table1)causestotalRrivtobecome

moreradiogenic,whilstFrivremainsconstant,drivingthe187Os/188Oscompositionof

seawatertomoreradiogenicvalues(Fig.12a).Meanwhile,areductioninriverineinputs

relatedtosilicateweathering(Fig.12e)drivesδ7Litomorepositivevalues(Fig.12d).

Figure12.DynamicmodelofOs(a)andLi(d)isotopesduringachangeinhigh-latitude

continentalicevolume.Themodelshownwasgeneratedbyassuminga50%decreasein

continentalsilicateweathering(b,e)duringglaciation(G),whichremainslowduringglacial

maximum(GM),beforeincreasingduringdeglaciation(D).InthecaseofOs,twoperiodsof

enhancedORLWduringglaciationanddeglaciationaregeneratedbyassuminga70%

increaseintheriverinefluxrelatedtoORLW(c).InthecaseofLi,theweathering

congruencyismodelledbyvaryingtheisotoperatiooftheriverineendmember(f).See

textfordetails.

Duringglacialmaximum,icesheetmovementstops,causingtheimmediate

cessationofenhancedORLWandtherelativefluxofFORLWdropstobelowpre-excursion

values(Fig.12c).Thiscausesadecreaseinthe187Os/188Oscompositionofseawaterasthe

inputofradiogenicOstoseawaterisremoved(Fig.12a).Theδ7Liofseawatercontinuesto

increase,reachingpeakvaluestowardstheendoftheglacialmaximum(Fig.12d).During

154

deglaciation,theriverinefluxrelatedtosilicateweatheringgraduallyincreasesbacktopre-

excursionlevels(Fig.12band12e),whiletheriverineinputrelatedtoORLWshowsa

secondinstantaneousincreasebeforegraduallydecliningtopre-excursionlevels(Fig.12c).

Thiscausesanincreaseinthe187Os/188Oscompositionofseawaterbeforethe187Os/188Os

compositiondecreasestopre-excursionvalues(Fig.12a).Theδ7Livalueofseawater

graduallydecreasestopre-excursionvaluesoverthenext~2Myr(Fig.12d).

ThiscompareswelltoOsisotopedatafortheIreviken,Mulde,LauandKlonk

events(Fig.1),whichallshowasimilartwo-peakprofiletomodeloutputs(Fig.12a).

However,themagnitudeofchangefrommodeloutputs(~0.12)islowerthandescribedin

thedata(0.19-0.8).Possiblecausesofthesediscrepanciescouldbe:amisrepresentationof

modelinputparameters;or,astrongerinfluencefromlocalcontinentalsourcesofOsin

coastalshelfsamplinglocalities,whichwilldrivelargervariationsinrecorded187Os/188Os

whencomparedtoawell-mixedocean.Lithiumisotoperecords(Fig.1)showasimilar

singularpeakinδ7Livaluestothemodeloutputs(Fig.12d).Ascenarioinwhichriverflux

alonevariedthroughouttheglaciationeventcausesaδ7Liexcursionof~3‰(Fig.12d),

whichcompareswelltorecordeddatafortheMuldeandLauevents(<4.4‰;Fig.1).Peak

δ7Livaluesoccurduringdeglaciationandthesecondpeakofradiogenic187Os/188Osvalues

(Fig.12aandd),consistentwithdatafortheMuldeevent(Fig.1).Ascenarioinwhichthe

riverfluxdecreased,whileriverineδ7Liratiosincreasedcausedavariationinδ7Li(~6‰)

greaterthanobservedinthedata,suggestingweatheringcongruencyremainedconstant

throughouttheglaciation.

ThefailureofSilurianoceancirculationmodels(Seesection4.1.1),floodbasalt

volcanism(Seesection4.1.2),rapidcooling(Seesection4.1.3)andvariationsin

hydrothermalactivity(Seesection4.1.4)indescribingvariationsinOsandLiisotopesfound

here(Fig.2),leadsustotheconclusionthatperiodicglaciationsarethemostlikelycauseof

155

Silurianclimateevents,basedonthecurrentunderstandingoftheOsandLiisotope

systems.Thedataandmodelsareconsistentwithglaciationeventsduringthelate

TelychiantoearlySheinwoodian,mid-Homerian,mid-LudfordianandtheSilurian-Devonian

boundary.Duringglaciation,expandingcontinentalicesheetsenhancephysicalerosion

whilststimulatingchemicalweatheringoforganicandsulphiderichrocksasevidencedby

anincreaseinthe187Os/188Oscompositiontomoreradiogenicvalues(Fig.1;Fig.12a).

Meanwhile,glacialcover,subglacialprocessesanddecliningglobaltemperaturesreduce

silicateweathering,decreasingtheriverinefluxofLitotheocean,leadingtoanincreasein

theδ7Livalueofseawater(Fig.1;Fig.12e).Decliningtemperaturescoupledtoenhanced

continentalicevolumedrivetheδ18Oofseawatertomorepositivevalues(Azmyetal.,

1998;Calner,2008;Lehnertetal.,2010;Munneckeetal.,2010;Nobleetal.,2005;Trotter

etal.,2016).Adropineustaticsea-level(HaqandSchutter,2008;Johnson,2006,2010;

Loydell,1998)exposescarbonateshelvestoweathering,drivingtheδ13Cofseawaterto

morepositivevalues(Crameretal.,2011a;Kumpetal.,1999;SaltzmanandThomas,2012).

Astheglacialmaximumisreached,icesheetexpansionabruptlyterminates,reducingthe

availabilityoffreshlycomminutedrockfortheoxidationofancientorganicmatterand

sulphiderichlithologiesandthereforereducingthefluxofOstotheoceanbearinga

radiogenic187Os/188Oscomposition(Fig.1;Fig.12a).Extensivestablecontinentalice-sheets

maintainlowlevelsofsilicateweathering(Fig.1;Fig.12e)andsea-level(Johnson2006,

2010;HaqandSchutter,2008;Loydell,1998)whilstmaintainingahigherδ18Ovalueof

seawater(Trotteretal.,2016).Duringdeglaciation,risingtemperaturesandtheincreased

availabilityofmeltwaterandfreshlycomminutedand/orscouredbedrock,enhances

chemicalsilicateweatheringandtheoxidationoforganicandsulphiderichrocks,drivingan

increaseinthe187Os/188Oscompositionofseawater,andadecreaseintheδ7Livalueof

seawater(Fig.1;Fig.12aand12e).Anincreaseinglobaltemperaturesandareductionin

continentalice-volume,drivesadecreaseintheδ18Ovalueofseawater(Trotteretal.,

156

2016)andanincreaseinglobaleustaticsea-level(HaqandSchutter,2008;Johnson,2006,

2010;Loydell,1998).

Acoupleddropineustaticsealevelandglobaltemperatureswouldhavehada

profoundinfluenceonmarinebiota.TheIreviken,Mulde,LauandKlonkbio-eventsare

definedbyglobalextinctionsofconodonts,graptolites,acritarchs,andotherbenthos

(Aldridgeetal.,1993;Calner,2008;Cooperetal.,2013;Jeppsson,1990,1997;Jeppsson,

1998;JeppssonandAldridge,2000;Loydell,2007;Melchinetal.,1998;Štorch,1995).

Cooperetal.(2013)utilisedgraptoloidevolutionaryratestotrackglobalclimaticchange.

RelativelycalmOrdovicianextinctionandoriginationratesgavewaytohighlyvolatilerates

duringthelateKatianthroughtotheEarlyDevonian,withsharpextinctionepisodes

triggeredbyenvironmentalcrisis(Cooperetal.,2013).Thissupportstheideapresentedin

thisstudy,inwhichaswitchfromrelativelystablegreenhouseconditionsduringtheEarly

andMiddleOrdovician,torelativelyunstableicehouseconditionsduringtheHirnantian

andSilurian,createdhighlyvolatileconditionsformarinebiota.Abruptglaciations

presentedherecoincidewithhighgraptloidextinctionratesduringtheearlySheinwoodian,

midHomerian,lateLudfordianandlatestPridolian,withsubsequenthighlevelsof

originationsduringrecoveryphases(Cooperetal.,2013).

4.4.2ScenariosfortriggeringSilurianGlaciations

Severalmechanisms,usuallyinvolvingadropbelowthresholdatmosphericCO2levels(~6

PAL;Pohletal.,2014),havebeenputforthastriggersfortheHirnantianglaciation,and

includechangesin:orogeny(Kumpetal.,1999);landplantdiversification(Lentonetal.,

2012;Poradaetal.,2016);volcanicarcdegassing(McKenzieetal.,2016;Poggevon

Strandmannetal.,inreview);and,paleogeography(Nardinetal.,2011).Likethe

Ordovicianthatprecededit,manyoftheseprocesseswereactiveduringtheSilurian,

157

providingseveralpossiblecandidatesforthemechanismthattriggeredtheglaciations

proposedinthisstudy.

4.4.2.1Orogeny

IthasbeenarguedthatenhancedweatherabilityofsilicaterocksduringtheTaconic

orogeny,ledtoalong-termdeclineinatmosphericCO2whicheventuallydroppedbelow

thethresholdlevelsthatwouldallowcontinentalicetoexpand(Kumpetal.,1999).During

theSilurian,AvaloniaandBalticacollided,withsubsequentcollisionofthesecombined

landmasseswithLaurentiatoformtheScandianorogeny(CocksandTorsvik,2002;Torsvik

andCocks,2013).Therapidrisetomoreradiogenic87Sr/86Srisotoperatiosduringthe

Silurianhasbeenattributedtoenhancedweatheringofoldsialiccrustexposedduringthe

aforementionedformationoftheCaledonian-Appalachianorogenicbelt(Crameretal.,

2011b).Moreover,steepinflectionpointsintheSrisotoperecordoccurpriortoand/or

duringthelate-Telychian,mid-Homerian,mid-LudfordianandtheSilurian-Devonian

boundary(Crameretal.,2011b;Frýdaetal.,2002).Theserepeatingperiodsofenhanced

weatherabilityofexposedsialiccrustwouldhaveledtothelong-termdeclinein

atmosphericCO2andglobaltemperatureswhich,ifdroppedbelowthresholdlevels(Kump

etal.,1999),wouldinitiatetheformationofcontinentalicesheets.

4.4.2.2Landplantdiversification

Ithasbeenhypothesized(Lentonetal.,2012)thattheexpansionofnon-vascularland-

plantsduringtheOrdovicianacceleratedglobalchemicalweatheringbyuptothreetimes

modernweatheringfluxes(Poradaetal.,2016),leadingtothedrawdownofenough

atmosphericCO2(<6PAL)totriggerthegrowthoficesheets.Moreover,thedevelopment

158

ofvascularlandplantsduringtheDevonianisthoughttohavecausedadramatic

drawdowninatmosphericCO2thatledtoglobalcoolingandpolarglaciationsduringthe

LateDevonian,ultimatelyleadingtoCarboniferousicehouseconditions(Algeoand

Scheckler,2010;AlgeoandScheckler,1998;Berner,1997).DuringtheSilurian,non-

vascularandvascularlandplantsexpandedgeographicallytoinhabitnewcontinentsand

landmasses(Steemansetal.,2009).Meanwhile,thedevelopmentofevolutionarytraitsin

vascularland-plants,suchasmoreextensiverootingstructuresand‘theseedhabit’,

originallyassignedtotheDevonian(AlgeoandScheckler,2010;Algeoetal.,1995;Algeo

andScheckler,1998),arenowthoughttohavedevelopedearlier,duringtheSilurian

(Edwardsetal.,2014;Gensel,2008;Kenricketal.,2012).Geographicalexpansionand/or

thedevelopmentofmoreextensiverootsystemswouldacttoenhancetheweathering

ratesofsilicates,leadingtoareductioninatmosphericCO2,globalcoolingandperiodic

glaciations.

4.4.2.3Volcanicarcdegassing

Recently,variationsinvolcanicarcactivityhavebeenshowntohaveadirectrelationship

withclimaticshiftsbetweenicehouseandgreenhouseconditions(McKenzieetal.,2016).

ContinentalcollisionsduringtheassemblyofGondwanaledtoareductionincontinental

arcvolcanism,andthereforeatmosphericCO2emissions,culminatinginglobalcoolingand

theHirnantianglaciation(McKenzieetal.,2016;PoggevonStrandmannetal.,inreview).In

theMid-LateSilurian,theIapetusOceanclosedduringtheformationofLaurussia.This

newlyformedcontinentrapidlytravelledsouthward,eventuallycollidingwithGondwana,

causingtheclosureoftheRheicOceanbytheCarboniferous(Cocksandtorsvik,2002).This

wouldhaveledtoareductionincontinentalarcvolcanism,asevidencedbycumulativeage

159

distributionsofSilurianZircons(McKenzieetal.,2016),leadingtoglobalcoolingand

intermittentSilurianglaciations.

4.4.2.4Paleogeography

AnotherexplanationfortheupperOrdovicianthroughtolateLlandoveryglacialperiod

involveschangesinpaleogeographythatcausedthemovementoffreshlyexposedvolcanic

rocks(Lefebvreetal.,2010;Youngetal.,2009)throughtheintertropicalconvergencezone

(ITCZ),stimulatingcontinentalweatheringandadeclineinatmosphericCO2belowa

thresholdthatwouldallowcontinentalglaciationstooccur(Nardinetal.,2011).In

particular,theprogressiveamalgamationofBaltica,AvaloniaandLaurentiaattropical

latitudesledtohighrunoffandthereforeweathering,maintainingthelowatmosphericCO2

requiredtoinitiatetheearlySilurianglaciations(Nardinetal.,2011).Althoughthismodel

couldexplainthemechanismbywhichaglaciationjustaftertheTelychian-Sheinwoodian

boundary(Fig.1),itprecludestheexistenceoflateSilurianglaciationsduetodecreasesin

continentalrunoffascontinentallandmassesmoveoutoftheITCZ(Nardinetal.,2011).

However,thismaybeduetothelowtemporalresolutionofthemodel.

4.4.2.5Orbitalforcing

AlthoughtherehavebeenmanyPaleozoic-centricmechanismsputforthasdriversfor

Paleozoicglaciations(Seesection4.4.2.1to4.4.2.4)ithasrecentlybecomeapparentthata

more‘Cenozoic-style’scenariomayhavecausedtheend-Ordovicianglaciations(Ghienneet

al.,2014).TheHirnantianglaciationisgenerallyviewedasasingularevent,withmarine

extinctionstiedtoglaciationanddeglaciation.However,highresolutionstratigraphic

recordshaveledtothediscoveryofseveralhigh-orderglacialcycles,suggestingamulti-

orderclimatesignalsimilartotheCenozoic(Sutcliffeetal.,2000;Ghienneetal.,2014).

Bothtwo(Sutcliffeetal.,2000)andthree(Ghienneetal.,2014)glacialcyclesofice-sheet

160

growthhavebeenrecordedfortheend-Ordovician,controlledbya0.1Myreccentricity

periodicity(Sutcliffeetal.,2000)or1.2Myrobliquitymodulation(Ghienneetal.,2014)

respectively.

MuchlikepreviousviewsfortheOrdovician,theSilurianclimaticeventsare

generallyseentohaveasingularpeak,lackinginhighorderglacialcycles.However,the

Silurianisseverelylackinginhighresolutionstratigraphicandisotopicrecords,precluding

thediscoveryoforbitallyforced‘Cenozoic-style’glacialcycles.Moreover,OsandLiisotopic

recordsfromthisstudy(SeeFig.1)areofsuchlow-resolutionthattheypreventrecognition

ofsuchevents.However,Silurianoxygenisotope(Trotteretal.,2016)andsea-level((Haq

andSchutter,2008;Johnson,2006,2010;Loydell,1998)recordsshowhigh-order

fluctuationspossiblerelatedtoglacialeustasyonglacial-interglacialtimescales(SeeFig.1).

Furthermore,theHirnantian,Sedgwicki,Ireviken,Mulde,LauandKlonkeventsappearto

bespacedover4to5Myrintervals(SeeFig.1)suggestingapacingtoSilurianglaciations

whichcouldbesetbytheinteractionbetweeneccentricity,obliquityandprecessioncycles,

ultimatelycontrolledthroughchangesininsolation.Moreworkwillneedtobegenerate

high-resolutionstratigraphicandisotopicrecordsforSilurianclimaticeventsinasimilar

mannertotheend-Ordovician(SeeSutcliffeetal.,2000andGhienneetal.,2014).

4.4.2.6Whydidwenotsee‘SnowballEarth’conditionsduringtheSilurian?

Largeglaciationsencompassingtheentireglobe,otherwiseknownas‘SnowballEarths’,are

believedtohaveoccurredduringtheNeoproterozoic(1000–542Ma),asevidencedfrom

theexistenceoflow-latitudeglacialdepositsatsea-level(SeeHoffman&Schrag,2002and

referencestherein).Theseeventscouldhavepotentiallybeentriggeredbyhighobliquity

causedbythelunar-formingimpact(SeeWilliams,2008)orpositiveice-albedofeedbacks

whichledtoglacialsthatlastedtensofmillionsofyearsbeforetheslowaccumulationof

atmosphericCO2duetovolcanicdegassingbroughtthemtoanend(SeeHoffman&Schrag,

161

2002).However,unliketheNeoproterozoicsnowballEarths,Paleozoicglaciationssuchas

thosefoundintheCambrian,LateOrdovician,Silurian,LateDevonianandCarboniferous

aregenerallynotmetbylow-latitudeglacialdepositsanddeglacialcap-carbonatesand

negatived13Cexcursions(>10‰).Thissuggeststherewaseithermore/lessfeedback

mechanismsactiveduringtheNeoproterozoicwhencomparedtothePaleozoic,that

allowedforrunawayicehouseconditionstooccur.

Phanerozoicpaleogeographyreconstructions(CocksandTorsvik,2002;Torsvik

andCocks,2013)oftensituatescontinentalmassesathigh-latitudesoveroneorbothof

thepoles,eitherasasupercontinente.g.Pangeaorasmallercontinente.g.Antarctica.

However,unlikethePhanerozoic,NeoproterozoicsnowballEarthsaredominatedbya

severelackofhigh-latitudecontinentalmasses.Thiswouldhavecreatedanabsenceof

continentalice-weatheringfeedbacksproposedinthisstudyfortheSilurian(Seesection

4.4.1.5).Therefore,oncepolarglaciationisinitiated,sea-iceformationwouldproceedto

lowlatitudesviapositiveice-albedofeedbackwithoutthesubsequentreductioninsilicate

weatheringandenhancementofoxidativeweatheringassociatedwithcontinentalice,

whichwouldotherwiseacttoincreaseatmosphericCO2leadingtoglobalwarming.This

suggestsweatheringfeedbackshelpregulateatmosphericCO2,preventingrunaway

icehouseconditionsandsnowballEarths(Seesection4.4.3).

4.4.3WeatheringfeedbackshelpregulateatmosphericCO2

WeatheringregulatesatmosphericCO2overmultimillionyeartimescalesandcanbe

summarisedbyseveralchemicalreactions(BernerandBerner,2012;Berner,2006;Berner

etal.,1983;Torresetal.,2014;Walkeretal.,1981).ThedissolutionofMg-andCa-bearing

silicatemineralsbycarbonicacid,andthesubsequenttransportofsolutestotheocean

whereinorganiccarbonisburiedasmarinecarbonates,sequestersatmosphericCO2:

CO2+(Ca,Mg)SiO3→(Ca,Mg)CO3+SiO2 Eq.7

162

Ontheotherhand,theoxidativeweatheringofancientsedimentaryorganicmatter

releasesCO2backtotheatmosphere:

CH2O+O2→CO2+H2O Eq.8

OthersourcesofatmosphericCO2includetheoxidativedissolutionofpyriteandthe

subsequentdissolutionofcarbonatesbysulfuricacid:

15O2+4FeS2+14H2O→4Fe(OH)3+8H2SO4 Eq.9

CaCO3+H2SO4→CO2+H2O+Ca2++SO42- Eq.10

and

2CaCO3+H2SO4→2Ca2++2HCO3-+SO4

2-↔CaCO3+CO2+SO42-+Ca2+

Eq.11

Reactions7to11showastrongdependenceonclimaticparameterssuchastemperature,

runoffand/orphysicalweathering(Georgetal.,2013;Torresetal.,2014).Duringthe

Cenozoic,concurrentincreasesinthe87Sr/86Sr,187Os/188Osandδ7Lirecords(Klemmetal.,

2005;McArthuretal.,2001;MisraandFroelich,2012)suggestenhancedphysicaland

chemicaldenudationduringextensiveupliftofmountainrangesmayhavestimulatedCO2

consumptionviasilicateweathering(Raymoetal.,1988).However,intheabsenceof

enhancedCO2productionfromothersources,thiswouldhavestrippedtheatmosphereof

allitsCO2withinafewmillionyears(BernerandCaldeira,1997).Thisdeclinein

atmosphericCO2mayhavebeenoffsetbyareleaseofCO2fromasimultaneousincreases

inoxidativeweatheringofancientorganiccarbon(Lietal.,2009)and/orcarbonate–sulfuric

acidweathering(Torresetal.,2014).

DuringQuaternaryglacial-interglacialcycles,enhancedphysicalweatheringleft

behindfine-grainedmaterialforchemicalweatheringatglacialterminals(BellandLaine,

163

1985;GoodbredandKuehl,2000;Hinderer,2001;ThomasandThorp,1995).Thiscaused

deglacialpulsesinsilicateweatheringthatcouldhavetheoreticallyloweredatmospheric

CO2by10-20ppm(Vanceetal.,2009).However,icecoresrecordariseinCO2during

deglaciations(Petitetal.,1999)duetoaconcurrentreleaseofCO2fromtheenhanced

weatheringoforganic-andsulphide-richsedimentaryrocks,whichexceededCO2

consumptionbysilicateweathering(Georgetal.,2013).

Here,weproposesimilarprocessesoccurredduringtheSilurian.Thelong-term

declineinatmosphericCO2causedbyorogeny,land-plantdiversification,volcanic

degassingand/orpaleogeographicchanges,inducedglobalcooling.Eventuallyatmospheric

CO2droppedbelowthresholdlevels(~6PAL)triggeringanexpansionofsouthern

hemisphereicesheetsoverGondwana,whichproceededtoexpandviapositiveice-albedo

feedback.Duringglaciation,enhancedphysicalweatheringandglacialprocessesactedto

increasetheoxidationofancientorganiccarbonandsedimentarysulphides,which

subsequentlydissolvedcarbonates,releasingatmosphericCO2totheatmosphere.

Meanwhile,silicateweatheringdeclined,partiallysuppressingoneoftheEarth’smajorCO2

removalmechanisms.Thecombinedinfluenceofthesetwoeffectsbegantoreversethe

globalcoolingtrend.Duringglacialmaximum,relativelylowsilicateweatheringrates

allowedCO2tobuildupintheatmosphereviaotherprocessese.g.volcanism,leadingto

rapidwarmingandeventuallydeglaciation.Thisdeglaciationexposedscouredbedrockto

theatmosphereandgeneratedfreshlycomminutedglacialtill,actingasafertilesubstrate

formeltwatertochemicallyattack.AdeclineinatmosphericCO2duetoenhancedsilicate

weatheringatglacialterminalswouldhavebeenlargelyoffsetbyenhancedoxidationof

organic-andsulphide-richsedimentaryrocks.ThisworklendsitselftotheideaoftheEarth

havingaself-regulatingclimatethatallowslifetoremainwithinhabitableconditions

(BernerandCaldeira,1997;Berneretal.,1983;Garrelsetal.,1976;Walkeretal.,1981).

164

4.5Implicationsandfutureoutlook

OsmiumandLiisotopedatapresentedhereareinterpretedtotracefluctuationsin

continentalweatheringthroughSilurianclimaticperturbations.DuringtheSilurian,

orogeny,thediversificationandglobalexpansionoflandplants,changesinpaleogeography

and/orareductioninvolcanicarcdegassing,ledtoalong-termdeclineinatmosphericCO2

andglobalcooling.DuringthelateTelychian-earlySheinwoodian,mid-Homerian,mid-

LudfordianandacrosstheSilurian-Devonianboundary,atmosphericCO2droppedbelow

thethresholdlevels(~6PAL)thatwouldhaveallowedcontinentaliceoverGondwanato

expandthroughice-albedofeedbacks.Oncetriggered,theexpansionofcontinentalice

enhancedthedenudationofunderlyingbedrockandtheproductionoffine-grained

materialforchemicalattack.Subglacialandproglacialprocessesfavourtheoxidationof

ancientorganiccarbonandsulphides,whilstsuppressingsilicateweathering,causinganet

releaseofCO2totheatmosphere.Underglacialmaximum,theproductionoffine-grained

materialceased,andoxidativeweatheringratesdiminished.Areductioninsilicate

weathering,andthereforeoneoftheEarth’smajorCO2removalmechanisms,allowedCO2

tobuildupintheatmosphereviaotherprocesses,leadingtoglobalwarmingandrapid

deglaciation.Duringdeglaciation,retreatingice-sheetswouldhaveenhancedchemical

weatheringthroughtheprovisionoffreshmaterial,increasedmeltwaterandgenerally

wetterconditions.Theoxidativeweatheringoforganicandsulphide-richlithologieswould

havelargelyoffsetthereductioninatmosphericCO2fromenhancedsilicateweathering

duringdeglaciation.

FluctuationsintheSilurianδ18Orecordweredrivenbychangesincontinentalice

volumeandglobaltemperatures,withmorepositiveδ18Ovaluesduringglacialperiods.The

associateddropineustaticsealevel,exposedcarbonateshelvestoweatheringanddrove

positiveshiftsintheδ13Crecord,attainingthehighestδ13Cvaluesduringglacialmaximum.

Acoupleddropineustaticsealevelandglobaltemperatureswouldhavehadaprofound

165

influenceonmarinebiota.TheIreviken,Mulde,LauandKlonkbio-eventsaredefinedby

globalextinctionsofconodonts,graptolites,acritarchs,andotherbenthos,withhighlevels

oforiginationsduringpost-glacialrecovery.

IncontrasttopreviousviewsoftheSilurianasagreenhouseworldpunctuated

bylargecarboncycleperturbationsassociatedwithchangesinoceancirculationand

precipitationrates,thisstudypresentsanewoutlookontheSilurian.Wesuggestthe

SilurianisanicehouseworldlikethelateOrdovicianthatprecededit,punctuatedby

glaciationsassociatedwithabruptclimaticandbiologicalchange.Thelong-termdeclinein

atmosphericCO2duringtheSilurianwasperiodicallyreversedbynegativefeedback

mechanismsassociatedwithsaidglaciations,andpreventeda‘runaway’icehouse,helping

tomaintainahabitableplanet.Futureworkwillneedtofocusongeneratingsimilar

osmiumandlithiumisotopecurvesfordifferentpaleogeographiclocationstovalidatea

globalsignalandhelpascertainthetimingsofglaciation,glacialmaximumanddeglaciation.

Moreworkisnecessarytohelpdeterminebetweenorogeny,landplantdiversification,

paleogeographicchangesorvolcanicarcdegassingasthedriverofatmosphericCO2decline

andglobalcooling.

MuchliketheCenozoic,generatingOsandLiisotopecurvesfortheentire

Siluriancouldhelptrackthetimingandextentoforogenicevents.Moredetailedsporeand

macrofossilrecordsforSilurianland-plantscouldhelptrackevolutionaryandgeographical

changesandhelpascertaintheirpotentialinfluenceonglobalweatheringandatmospheric

evolutionthroughcomputermodelling.Modellingcouldalsohelppredicttheinfluenceof

enhancedvolcanicweatheringonatmosphericCO2duringLaurussia’spassagethroughthe

ITCZ.Finally,thehuntforglacialsedimentsinSouthAmericaandAfricashouldberesumed

toprovideunequivocalevidenceforSilurianglaciationsandtheirtiming.

166

4.6References

Aldridge,R.,Jeppsson,L.,Dorning,K.,1993.EarlySilurianoceanicepisodesandevents.JournaloftheGeologicalSociety150,501-513.

Algeo,T.,Scheckler,S.,2010.LandplantevolutionandweatheringratechangesintheDevonian.J.EarthSci.21,75-78.

Algeo,T.J.,Berner,R.A.,Maynard,J.B.,Scheckler,S.E.,1995.LateDevonianoceanicanoxiceventsandbioticcrises:“rooted”intheevolutionofvascularlandplants.GSAtoday5,45.

Algeo,T.J.,Marenco,P.J.,Saltzman,M.R.,2016.Co-evolutionofoceans,climate,andthebiosphereduringthe‘OrdovicianRevolution’:Areview.Palaeogeography,Palaeoclimatology,Palaeoecology458,1-11.

Algeo,T.J.,Scheckler,S.E.,1998.Terrestrial-marineteleconnectionsintheDevonian:linksbetweentheevolutionoflandplants,weatheringprocesses,andmarineanoxicevents.PhilosophicalTransactionsoftheRoyalSocietyB:BiologicalSciences353,113-130.

Anderson,S.P.,2005.Glaciersshowdirectlinkagebetweenerosionrateandchemicalweatheringfluxes.Geomorphology67,147-157.

Anderson,S.P.,2007.BiogeochemistryofGlacialLandscapeSystems.AnnualReviewofEarthandPlanetarySciences35,375-399.

Anderson,S.P.,Drever,J.I.,Frost,C.D.,Holden,P.,2000.Chemicalweatheringintheforelandofaretreatingglacier.GeochimicaetCosmochimicaActa64,1173-1189.

Anderson,S.P.,Drever,J.I.,Humphrey,N.F.,1997.Chemicalweatheringinglacialenvironments.Geology25,399-402.

Armstrong,R.L.,1971.Glacialerosionandthevariableisotopiccompositionofstrontiuminseawater.Nature230,132-133.

Azmy,K.,Veizer,J.,Bassett,M.G.,Copper,P.,1998.OxygenandcarbonisotopiccompositionofSilurianbrachiopods:implicationsforcoevalseawaterandglaciations.GeologicalSocietyofAmericaBulletin110,1499-1512.

Becker,R.T.,Gradstein,F.M.,Hammer,O.,2012.Chapter22-TheDevonianPeriod,in:Gradstein,F.M.,Ogg,J.G.,Schmitz,M.D.,Ogg,G.M.(Eds.),TheGeologicTimeScale.Elsevier,Boston,pp.559-601.

Bell,M.,Laine,E.,1985.ErosionoftheLaurentideregionofNorthAmericabyglacialandglaciofluvialprocesses.QuaternaryResearch23,154-174.

Berner,E.K.,Berner,R.A.,2012.Globalenvironment:water,air,andgeochemicalcycles.PrincetonUniversityPress.

Berner,R.A.,1997.TheriseofplantsandtheireffectonweatheringandatmosphericCO2.Science276,544-546.

167

Berner,R.A.,2006.GEOCARBSULF:AcombinedmodelforPhanerozoicatmosphericO2andCO2.GeochimicaetCosmochimicaActa70,5653-5664.

Berner,R.A.,Caldeira,K.,1997.Theneedformassbalanceandfeedbackinthegeochemicalcarboncycle.Geology25,955-956.

Berner,R.A.,Lasaga,A.C.,Garrels,R.M.,1983.Thecarbonate-silicategeochemicalcycleanditseffectonatmosphericcarbondioxideoverthepast100millionyears.AmJSci283,641-683.

Bickert,T.,Pätzold,J.,Samtleben,C.,Munnecke,A.,1997.PaleoenvironmentalchangesintheSilurianindicatedbystableisotopesinbrachiopodshellsfromGotland,Sweden.GeochimicaetCosmochimicaActa61,2717-2730.

Bluth,G.J.,Kump,L.R.,1994.Lithologicandclimatologiccontrolsofriverchemistry.GeochimicaetCosmochimicaActa58,2341-2359.

Bottini,C.,Cohen,A.S.,Erba,E.,Jenkyns,H.C.,Coe,A.L.,2012.Osmium-isotopeevidenceforvolcanism,weathering,andoceanmixingduringtheearlyAptianOAE1a.Geology40,583-586.

Bouchez,J.,vonBlanckenburg,F.,Schuessler,J.A.,2013.Modelingnovelstableisotoperatiosintheweatheringzone.AmericanJournalofScience313,267-308.

Brand,U.,Azmy,K.,Veizer,J.,2006.EvaluationoftheSalinicItectonic,CancañiriglacialandIrevikenbioticevents:BiochemostratigraphyoftheLowerSiluriansuccessionintheNiagaraGorgearea,CanadaandU.S.A.Palaeogeography,Palaeoclimatology,Palaeoecology241,192-213.

Buggisch,W.,Mann,U.,2004.CarbonisotopestratigraphyofLochkoviantoEifelianlimestonesfromtheDevonianofcentralandsouthernEurope.InternationalJournalofEarthSciences93,521-541.

Calner,M.,2008.Silurianglobalevents–atthetippingpointofclimatechange,MassExtinction.SpringerBerlinHeidelberg,pp.21-57.

Calner,M.,Kozłowska,A.,Masiak,M.,Schmitz,B.,2006.AshorelinetodeepbasincorrelationchartforthemiddleSiluriancoupledextinction-stableisotopicevent.GFF128,79-84.

Caputo,M.V.,1998.Ordovician–Silurianglaciationsandglobalsea-levelchanges.Siluriancycles,15-25.

Chan,L.-H.,Edmond,J.M.,Thompson,G.,1993.AlithiumisotopestudyofhotspringsandmetabasaltsfromMid-OceanRidgeHydrothermalSystems.JournalofGeophysicalResearch98,9653.

168

Chan,L.,Edmond,J.,Thompson,G.,Gillis,K.,1992.Lithiumisotopiccompositionofsubmarinebasalts:implicationsforthelithiumcycleintheoceans.EarthandPlanetaryScienceLetters108,151-160.

Chan,L.H.,Leeman,W.P.,Plank,T.,2006.Lithiumisotopiccompositionofmarinesediments.Geochemistry,Geophysics,Geosystems7.

Cocks,L.,Torsvik,T.,2002.Earthgeographyfrom500to400millionyearsago:afaunalandpalaeomagneticreview.JournaloftheGeologicalSociety159,631-644.

Cohen,A.S.,Coe,A.L.,2002.NewgeochemicalevidencefortheonsetofvolcanismintheCentralAtlanticmagmaticprovinceandenvironmentalchangeattheTriassic-Jurassicboundary.Geology30,267.

Cooper,R.,Wadham,J.,Tranter,M.,Hodgkins,R.,Peters,N.,2002.Groundwaterhydrochemistryintheactivelayeroftheproglacialzone,Finsterwalderbreen,Svalbard.JournalofHydrology269,208-223.

Cooper,R.A.,Sadler,P.M.,Munnecke,A.,Crampton,J.S.,2013.GraptoloidevolutionaryratestrackOrdovician–Silurianglobalclimatechange.GeologicalMagazine151,349-364.

Cramer,B.D.,Brett,C.E.,Melchin,M.J.,Maennik,P.,Kleffner,M.A.,McLaughlin,P.I.,Loydell,D.K.,Munnecke,A.,Jeppsson,L.,Corradini,C.,2011a.RevisedcorrelationofSilurianProvincialSeriesofNorthAmericawithglobalandregionalchronostratigraphicunitsandδ13Ccarbchemostratigraphy.Lethaia44,185-202.

Cramer,B.D.,Loydell,D.K.,Samtleben,C.,Munnecke,A.,Kaljo,D.,Mannik,P.,Martma,T.,Jeppsson,L.,Kleffner,M.A.,Barrick,J.E.,Johnson,C.A.,Emsbo,P.,Joachimski,M.M.,Bickert,T.,Saltzman,M.R.,2010.TestingthelimitsofPaleozoicchronostratigraphiccorrelationviahigh-resolution(<500k.y.)integratedconodont,graptolite,andcarbonisotope(13Ccarb)biochemostratigraphyacrosstheLlandovery-Wenlock(Silurian)boundary:IsaunifiedPhanerozoictimescaleachievable?GeologicalSocietyofAmericaBulletin122,1700-1716.

Cramer,B.D.,Munnecke,A.,Schofield,D.I.,Haase,K.M.,Haase-Schramm,A.,2011b.ARevised87Sr/86SrCurvefortheSilurian:ImplicationsforGlobalOceanChemistryandtheSilurianTimescale.TheJournalofGeology119,335-349.

Creaser,R.A.,Papanastassiou,D.A.,Wasserburg,G.J.,1991.Negativethermalionmassspectrometryofosmium,rheniumandiridium.GeochimicaetCosmochimicaActa55,397-401.

Crick,R.,Ellwood,B.,Hladil,J.,ElHassani,A.,Hrouda,F.,Chlupac,I.,2001.MagnetostratigraphysusceptibilityofthePřıd́olian–Lochkovian(Silurian–Devonian)GSSP(Klonk,CzechRepublic)andacoevalsequenceinAnti-AtlasMorocco.Palaeogeography,Palaeoclimatology,Palaeoecology167,73-100.

Cumming,V.M.,Poulton,S.W.,Rooney,A.D.,Selby,D.,2013.AnoxiaintheterrestrialenvironmentduringthelateMesoproterozoic.Geology41,583-586.

169

Dellinger,M.,Gaillardet,J.,Bouchez,J.,Calmels,D.,Louvat,P.,Dosseto,A.,Gorge,C.,Alanoca,L.,Maurice,L.,2015.RiverineLiisotopefractionationintheAmazonRiverbasincontrolledbytheweatheringregimes.GeochimicaetCosmochimicaActa164,71-93.

Díaz-Martínez,E.,Grahn,Y.,2007.EarlySilurianglaciationalongthewesternmarginofGondwana(Peru,BoliviaandnorthernArgentina):Palaeogeographicandgeodynamicsetting.Palaeogeography,Palaeoclimatology,Palaeoecology245,62-81.

Drever,J.I.,Zobrist,J.,1992.ChemicalweatheringofsilicaterocksasafunctionofelevationinthesouthernSwissAlps.GeochimicaetCosmochimicaActa56,3209-3216.

DuVivier,A.D.C.,Selby,D.,Sageman,B.B.,Jarvis,I.,Gröcke,D.R.,Voigt,S.,2014.Marine187Os/188OsisotopestratigraphyrevealstheinteractionofvolcanismandoceancirculationduringOceanicAnoxicEvent2.EarthandPlanetaryScienceLetters389,23-33.

Edwards,D.,Morris,J.L.,Richardson,J.B.,Kenrick,P.,2014.Cryptosporesandcryptophytesrevealhiddendiversityinearlylandfloras.Newphytologist202,50-78.

Elderfield,H.,Schultz,A.,1996.Mid-oceanridgehydrothermalfluxesandthechemicalcompositionoftheocean.AnnualReviewofEarthandPlanetarySciences24,191-224.

Eriksson,M.J.,Calner,M.,2007.AsequencestratigraphicalmodelfortheLateLudfordian(Silurian)ofGotland,Sweden:implicationsfortimingbetweenchangesinsealevel,palaeoecology,andtheglobalcarboncycle.Facies54,253-276.

Fairchild,I.J.,Killawee,J.A.,Sharp,M.J.,Spiro,B.,Hubbard,B.,Lorrain,R.D.,Tison,J.L.,1999.Solutegenerationandtransferfromachemicallyreactivealpineglacial–proglacialsystem.EarthSurfaceProcessesandLandforms24,1189-1211.

Finlay,A.J.,Selby,D.,Gröcke,D.R.,2010.TrackingtheHirnantianglaciationusingOsisotopes.EarthandPlanetaryScienceLetters293,339-348.

Flesch,G.,Anderson,A.,Svec,H.,1973.Asecondaryisotopicstandardfor6Li/7Lideterminations.InternationalJournalofMassSpectrometryandIonPhysics12,265-272.

Frýda,J.,Hladil,J.,Vokurka,K.,2002.SeawaterstrontiumisotopecurveattheSilurian/Devonianboundary:astudyoftheglobalSilurian/Devonianboundarystratotype.Geobios35,21-28.

Frýda,J.,Manda,Š.,2013.Along-lastingsteadyperiodofisotopicallyheavycarboninthelateSilurianocean:evolutionoftheδ13Crecordanditssignificanceforanintegratedδ13C,graptoliteandconodontstratigraphy.BulletinofGeosciences88.

Gaillardet,J.,Dupré,B.,Louvat,P.,Allegre,C.,1999.GlobalsilicateweatheringandCO2consumptionratesdeducedfromthechemistryoflargerivers.Chemicalgeology159,3-30.

Garrels,R.M.,Lerman,A.,Mackenzie,F.T.,1976.ControlsofatmosphericO2andCO2:past,present,andfuture:geochemicalmodelsoftheearth'ssurfaceenvironment,focusing

170

onO2andCO2cycles,suggestthatadynamicsteady-statesystemexists,maintainedovertimebyeffectivefeedbackmechanisms.AmericanScientist,306-315.

Gensel,P.G.,2008.TheEarliestLandPlants.AnnualReviewofEcology,Evolution,andSystematics39,459-477.

Georg,R.B.,West,A.J.,Vance,D.,Newman,K.,Halliday,A.N.,2013.IsthemarineosmiumisotoperecordaprobeforCO2releasefromsedimentaryrocks?EarthandPlanetaryScienceLetters367,28-38.

Ghienne,J-F.,etal.,2014.ACenozoicstylescenariofortheend-Ordovicianglaciation.Naturecommunications,5:4485.

Gislason,S.R.,Oelkers,E.H.,Eiriksdottir,E.S.,Kardjilov,M.I.,Gisladottir,G.,Sigfusson,B.,Snorrason,A.,Elefsen,S.,Hardardottir,J.,Torssander,P.,Oskarsson,N.,2009.Directevidenceofthefeedbackbetweenclimateandweathering.EarthandPlanetaryScienceLetters277,213-222.

Goodbred,S.,Kuehl,S.A.,2000.Thesignificanceoflargesedimentsupply,activetectonism,andeustasyonmarginsequencedevelopment:LateQuaternarystratigraphyandevolutionoftheGanges–Brahmaputradelta.SedimentaryGeology133,227-248.

Grahn,Y.,Caputo,M.V.,1992.EarlySilurianglaciationsinBrazil.Palaeogeography,Palaeoclimatology,Palaeoecology99,9-15.

Haq,B.U.,Schutter,S.R.,2008.AchronologyofPaleozoicsea-levelchanges.Science322,64-68.

Hardie,L.A.,1996.Secularvariationinseawaterchemistry:Anexplanationforthecoupledsecularvariationinthemineralogiesofmarinelimestonesandpotashevaporitesoverthepast600my.Geology24,279-283.

Harper,D.A.T.,Hammarlund,E.U.,Rasmussen,C.M.Ø.,2014.EndOrdovicianextinctions:Acoincidenceofcauses.GondwanaResearch25,1294-1307.

Hathorne,E.C.,James,R.H.,2006.Temporalrecordoflithiuminseawater:Atracerforsilicateweathering?EarthandPlanetaryScienceLetters246,393-406.

Hazen,R.M.,Sverjensky,D.A.,Azzolini,D.,Bish,D.L.,Elmore,S.C.,Hinnov,L.,Milliken,R.E.,2013.Claymineralevolution.AmericanMineralogist98,2007-2029.

Hinderer,M.,2001.LateQuaternarydenudationoftheAlps,valleyandlakefillingsandmodernriverloads.GeodinamicaActa14,231-263.

Hodell,D.A.,Mead,G.A.,Mueller,P.A.,1990.Variationinthestrontiumisotopiccompositionofseawater(8Matopresent):Implicationsforchemicalweatheringratesanddissolvedfluxestotheoceans.ChemicalGeology:IsotopeGeosciencesection80,291-307.

Hoffman,P.F.&SchragD.P.2002.ThesnowballEarthhypothesis:testingthelimitsofglobalchange.TerraNova14,129-155.

171

Huh,Y.,Chan,L.-H.,Zhang,L.,Edmond,J.M.,1998.Lithiumanditsisotopesinmajorworldrivers:implicationsforweatheringandtheoceanicbudget.GeochimicaetCosmochimicaActa62,2039-2051.

Huh,Y.,Edmond,J.M.,1999.ThefluvialgeochemistryoftheriversofEasternSiberia:III.TributariesoftheLenaandAnabardrainingthebasementterrainoftheSiberianCratonandtheTrans-BaikalHighlands.GeochimicaetCosmochimicaActa63,967-987.

Jacobson,A.D.,Blum,J.D.,Chamberlain,C.P.,Poage,M.A.,Sloan,V.F.,2002.Ca/SrandSrisotopesystematicsofaHimalayanglacialchronosequence:carbonateversussilicateweatheringratesasafunctionoflandscapesurfaceage.GeochimicaetCosmochimicaActa66,13-27.

Jaffe,L.A.,Peucker-Ehrenbrink,B.,Petsch,S.T.,2002.Mobilityofrhenium,platinumgroupelementsandorganiccarbonduringblackshaleweathering.EarthandPlanetaryScienceLetters198,339-353.

James,R.H.,Rudnicki,M.D.,Palmer,M.R.,1999.ThealkalielementandborongeochemistryoftheEscanabaTroughsediment-hostedhydrothermalsystem.EarthandPlanetaryScienceLetters171,157-169.

Jeppsson,L.,1990.AnoceanicmodelforlithologicalandfaunalchangestestedontheSilurianrecord.JournaloftheGeologicalSociety147,663-674.

Jeppsson,L.,1997.Theanatomyofthemid-earlySilurianIrevikenEventandascenarioforPSevents.Paleontologicalevents:stratigraphic,ecological,andevolutionaryimplications,451-492.

Jeppsson,L.,1998.Silurianoceanicevents:summaryofgeneralcharacteristics.Siluriancycles:Linkagesofdynamicstratigraphywithatmospheric,oceanicandtectonicchanges,239-257.

Jeppsson,L.,Aldridge,R.J.,2000.Ludlow(lateSilurian)oceanicepisodesandevents.JournaloftheGeologicalSociety157,1137-1148.

Johnson,M.E.,2006.RelationshipofSiluriansea-levelfluctuationstooceanicepisodesandevents.GFF128,115-121.

Johnson,M.E.,2010.TrackingSilurianeustasy:Alignmentofempiricalevidenceorpursuitofdeductivereasoning?Palaeogeography,Palaeoclimatology,Palaeoecology296,276-284.

Kaljo,D.,Martma,T.,Männik,P.,Viira,V.,2003.ImplicationsofGondwanaglaciationsintheBalticlateOrdovicianandSilurianandacarbonisotopictestofenvironmentalcyclicity.BulletindelaSociétégéologiquedeFrance174,59-66.

Kendall,B.S.,Creaser,R.A.,Ross,G.M.,Selby,D.,2004.ConstraintsonthetimingofMarinoan“SnowballEarth”glaciationby187Re–187OsdatingofaNeoproterozoic,post-glacialblackshaleinWesternCanada.EarthandPlanetaryScienceLetters222,729-740.

172

Kenrick,P.,Wellman,C.H.,Schneider,H.,Edgecombe,G.D.,2012.Atimelineforterrestrialization:consequencesforthecarboncycleinthePalaeozoic.Phil.Trans.R.Soc.B367,519-536.

Klemm,V.,Levasseur,S.,Frank,M.,Hein,J.R.,Halliday,A.N.,2005.Osmiumisotopestratigraphyofamarineferromanganesecrust.EarthandPlanetaryScienceLetters238,42-48.

Kump,L.,Alley,R.B.,1994.Globalchemicalweatheringonglacialtimescales.MaterialFluxesontheSurfaceoftheEarth,46-60.

Kump,L.,Arthur,M.,Patzkowsky,M.,Gibbs,M.,Pinkus,D.,Sheehan,P.,1999.AweatheringhypothesisforglaciationathighatmosphericpCO2duringtheLateOrdovician.Palaeogeography,Palaeoclimatology,Palaeoecology152,173-187.

Lechler,M.,PoggevonStrandmann,P.A.E.,Jenkyns,H.C.,Prosser,G.,Parente,M.,2015.Lithium-isotopeevidenceforenhancedsilicateweatheringduringOAE1a(EarlyAptianSellievent).EarthandPlanetaryScienceLetters432,210-222.

Lefebvre,V.,Servais,T.,François,L.,Averbuch,O.,2010.DidaKatianlargeigneousprovincetriggertheLateOrdovicianglaciation?:Ahypothesistestedwithacarboncyclemodel.Palaeogeography,Palaeoclimatology,Palaeoecology296,310-319.

Lehnert,O.,Männik,P.,Joachimski,M.M.,Calner,M.,Frýda,J.,2010.PalaeoclimateperturbationsbeforetheSheinwoodianglaciation:Atriggerforextinctionsduringthe‘IrevikenEvent’.Palaeogeography,Palaeoclimatology,Palaeoecology296,320-331.

Lenton,T.M.,Crouch,M.,Johnson,M.,Pires,N.,Dolan,L.,2012.FirstplantscooledtheOrdovician.NatureGeoscience5,86-89.

Li,G.,Ji,J.,Chen,J.,Kemp,D.B.,2009.EvolutionoftheCenozoiccarboncycle:TherolesoftectonicsandCO2fertilization.GlobalBiogeochemicalCycles23,n/a-n/a.

Li,G.,West,A.J.,2014.EvolutionofCenozoicseawaterlithiumisotopes:Couplingofglobaldenudationregimeandshiftingseawatersinks.EarthandPlanetaryScienceLetters401,284-293.

Loydell,D.,Männik,P.,Nestor,V.,2003.IntegratedbiostratigraphyofthelowerSilurianoftheAizpute-41core,Latvia.GeologicalMagazine140,205-229.

Loydell,D.K.,1998.EarlySiluriansea-levelchanges.GeologicalMagazine135,447-471.

Loydell,D.K.,2007.EarlySilurianpositiveδ13Cexcursionsandtheirrelationshiptoglaciations,sea-levelchangesandextinctionevents.GeologicalJournal42,531-546.

Mackenzie,F.T.,Garrels,R.M.,1966.Chemicalmassbalancebetweenriversandoceans.AmericanJournalofScience264,507-525.

Maher,K.,Chamberlain,C.,2014.Hydrologicregulationofchemicalweatheringandthegeologiccarboncycle.Science343,1502-1504.

173

Maier,G.2010.Developmentofδ13Ccarbandδ13CorgisotopevaluesfromtheLowerSilurianofGotland(Sweden).Diplomathesis,GeoZentrumNordbayern.

McArthur,J.,Howarth,R.,Bailey,T.,2001.Strontiumisotopestratigraphy:LOWESSversion3:bestfittothemarineSr-isotopecurvefor0–509Maandaccompanyinglook-uptableforderivingnumericalage.TheJournalofGeology109,155-170.

McKenzie,N.R.,Horton,B.K.,Loomis,S.E.,Stockli,D.F.,Planavsky,N.J.,Lee,C.-T.A.,2016.Continentalarcvolcanismastheprincipaldriveroficehouse-greenhousevariability.Science352,444-447.

Melchin,M.,Koren,T.,Štorch,P.,1998.GlobaldiversityandsurvivorshippatternsofSiluriangraptoloids.NewYorkStateMuseumBulletin491,165-182.

Melchin,M.J.,Sadler,P.M.,Cramer,B.D.,Cooper,R.A.,Gradstein,F.M.,Hammer,O.,2012.Chapter21-TheSilurianPeriod,in:Gradstein,F.M.,Ogg,J.G.,Schmitz,M.D.,Ogg,G.M.(Eds.),TheGeologicTimeScale.Elsevier,Boston,pp.525-558.

Meybeck,M.,1987.Globalchemicalweatheringofsurficialrocksestimatedfromriverdissolvedloads.Americanjournalofscience287,401-428.

Millot,R.,Girard,J.,2007.Lithiumisotopefractionationduringadsorptionontomineralsurfaces,InternationalMeetingonClaysinNatural&EngineeredBarriersforRadioactiveWasteConfinement,Lille,France.

Misra,S.,Froelich,P.N.,2012.LithiumisotopehistoryofCenozoicseawater:changesinsilicateweatheringandreverseweathering.Science335,818-823.

Munnecke,A.,Calner,M.,Harper,D.A.T.,Servais,T.,2010.OrdovicianandSiluriansea–waterchemistry,sealevel,andclimate:Asynopsis.Palaeogeography,Palaeoclimatology,Palaeoecology296,389-413.

Nardin,E.,Godderis,Y.,Donnadieu,Y.,Hir,G.L.,Blakey,R.C.,Puceat,E.,Aretz,M.,2011.ModelingtheearlyPaleozoiclong-termclimatictrend.GeologicalSocietyofAmericaBulletin123,1181-1192.

Noble,P.J.,Zimmerman,M.K.,Holmden,C.,Lenz,A.C.,2005.EarlySilurian(Wenlockian)δ13CprofilesfromtheCapePhillipsFormation,ArcticCanadaandtheirrelationtobioticevents.CanadianJournalofEarthSciences42,1419-1430.

Nowell,G.,Luguet,A.,Pearson,D.,Horstwood,M.,2008.Preciseandaccurate186Os/188Osand187Os/188Osmeasurementsbymulti-collectorplasmaionisationmassspectrometry(MC-ICP-MS)partI:Solutionanalyses.ChemicalGeology248,363-393.

Oxburgh,R.,2001.Residencetimeofosmiumintheoceans.Geochemistry,Geophysics,Geosystems2,1018.

Palmer,M.,Edmond,J.,1992.Controlsoverthestrontiumisotopecompositionofriverwater.GeochimicaetCosmochimicaActa56,2099-2111.

174

Paquay,F.S.,Ravizza,G.E.,Dalai,T.K.,Peucker-Ehrenbrink,B.,2008.Determiningchondriticimpactorsizefromthemarineosmiumisotoperecord.Science320,214-218.

Petit,J.-R.,Jouzel,J.,Raynaud,D.,Barkov,N.I.,Barnola,J.-M.,Basile,I.,Bender,M.,Chappellaz,J.,Davis,M.,Delaygue,G.,1999.Climateandatmospherichistoryofthepast420,000yearsfromtheVostokicecore,Antarctica.Nature399,429-436.

Petsch,S.,Eglinton,T.,Edwards,K.,2001a.14C-deadlivingbiomass:evidenceformicrobialassimilationofancientorganiccarbonduringshaleweathering.Science292,1127-1131.

Petsch,S.T.,Smernik,R.,Eglinton,T.,Oades,J.,2001b.Asolidstate13C-NMRstudyofkerogendegradationduringblackshaleweathering.GeochimicaetCosmochimicaActa65,1867-1882.

Peucker-Ehrenbrink,B.,Blum,J.D.,1998.Re-OsisotopesystematicsandweatheringofPrecambriancrustalrocks:Implicationsforthemarineosmiumisotoperecord.GeochimicaetCosmochimicaActa62,3193-3203.

Peucker-Ehrenbrink,B.,Hannigan,R.E.,2000.Effectsofblackshaleweatheringonthemobilityofrheniumandplatinumgroupelements.Geology28,475-478.

Peucker-Ehrenbrink,B.,Ravizza,G.,2000.Themarineosmiumisotoperecord.TerraNova12,205-219.

Peucker-Ehrenbrink,B.,Ravizza,G.,2012.Osmiumisotopestratigraphy.TheGeologicTimeScale2012,145-166.

Pierson-Wickmann,A.-C.,Reisberg,L.,France-Lanord,C.,2002.BehaviorofReandOsduringlow-temperaturealteration:resultsfromHimalayansoilsandalteredblackshales.GeochimicaetCosmochimicaActa66,1539-1548.

PoggevonStrandmann,P.A.E.,Burton,K.W.,James,R.H.,vanCalsteren,P.,Gíslason,S.R.,Mokadem,F.,2006.Riverinebehaviourofuraniumandlithiumisotopesinanactivelyglaciatedbasalticterrain.EarthandPlanetaryScienceLetters251,134-147.

PoggevonStrandmann,P.A.E.,Desrochers,A.,Murphy,M.J.,Finlay,A.J.,Selby,D.,Lenton,T.M.,inreview.GlobalclimatestabilisationbychemicalweatheringduringtheHirnantianglaciation.GeochemicalPerspectivesLetters.

PoggevonStrandmann,P.A.E.,Elliott,T.,Marschall,H.R.,Coath,C.,Lai,Y.-J.,Jeffcoate,A.B.,Ionov,D.A.,2011.VariationsofLiandMgisotoperatiosinbulkchondritesandmantlexenoliths.GeochimicaetCosmochimicaActa75,5247-5268.

PoggevonStrandmann,P.A.E.,Frings,P.J.,Murphy,M.J.,2017.LithiumisotopebehaviourduringweatheringintheGangesAlluvialPlain.GeochimicaetCosmochimicaActa198,17-31.

175

PoggevonStrandmann,P.A.E.,Jenkyns,H.C.,Woodfine,R.G.,2013.LithiumisotopeevidenceforenhancedweatheringduringOceanicAnoxicEvent2.NatureGeoscience6,668-672.

Pohl,A.,Donnadieu,Y.,LeHir,G.,Buoncristiani,J.F.,Vennin,E.,2014.EffectoftheOrdovicianpaleogeographyonthe(in)stabilityoftheclimate.ClimPast10,2053-2066.

Poirier,A.,Hillaire-Marcel,C.,2009.Os-isotopeinsightsintomajorenvironmentalchangesoftheArcticOceanduringtheCenozoic.GeophysicalResearchLetters36.

Porada,P.,Lenton,T.M.,Pohl,A.,Weber,B.,Mander,L.,Donnadieu,Y.,Beer,C.,Poschl,U.,Kleidon,A.,2016.Highpotentialforweatheringandclimateeffectsofnon-vascularvegetationintheLateOrdovician.Naturecommunications7,12113.

Porębska,E.,Kozłowska-Dawidziuk,A.,Masiak,M.,2004.ThelundgrenieventintheSilurianoftheEastEuropeanPlatform,Poland.Palaeogeography,Palaeoclimatology,Palaeoecology213,271-294.

Ravizza,G.,Norris,R.N.,Blusztajn,J.,Aubry,M.P.,2001.AnosmiumisotopeexcursionassociatedwiththeLatePaleocenethermalmaximum:Evidenceofintensifiedchemicalweathering.Paleoceanography16,155-163.

Ravizza,G.,Peucker-Ehrenbrink,B.,2003.ChemostratigraphicevidenceofDeccanvolcanismfromthemarineosmiumisotoperecord.Science302,1392-1395.

Ravizza,G.,Turekian,K.K.,1989.Applicationofthe187Re-187Ossystemtoblackshalegeochronometry.GeochimicaetCosmochimicaActa53,3257-3262.

Raymo,M.E.,Ruddiman,W.F.,Froelich,P.N.,1988.InfluenceoflateCenozoicmountainbuildingonoceangeochemicalcycles.Geology16,649-653.

Richter,F.M.,Turekian,K.K.,1993.Simplemodelsforthegeochemicalresponseoftheoceantoclimaticandtectonicforcing.EarthandPlanetaryScienceLetters119,121-131.

Rooney,A.D.,Selby,D.,Lloyd,J.M.,Roberts,D.H.,Lückge,A.,Sageman,B.B.,Prouty,N.G.,2016.Trackingmillennial-scaleHoloceneglacialadvanceandretreatusingosmiumisotopes:InsightsfromtheGreenlandicesheet.QuaternaryScienceReviews138,49-61.

Saltzman,M.R.,Thomas,E.,2012.Chapter11-CarbonIsotopeStratigraphy,TheGeologicTimeScale.Elsevier,Boston,pp.207-232.

Sauzéat,L.,Rudnick,R.L.,Chauvel,C.,Garçon,M.,Tang,M.,2015.NewperspectivesontheLiisotopiccompositionoftheuppercontinentalcrustanditsweatheringsignature.EarthandPlanetaryScienceLetters428,181-192.

Schmitz,B.,Peucker-Ehrenbrink,B.,Heilmann-Clausen,C.,Åberg,G.,Asaro,F.,Lee,C.-T.A.,2004.Basalticexplosivevolcanism,butnocometimpact,atthePaleocene–Eoceneboundary:high-resolutionchemicalandisotopicrecordsfromEgypt,SpainandDenmark.EarthandPlanetaryScienceLetters225,1-17.

176

Selby,D.,Creaser,R.A.,2003.Re–Osgeochronologyoforganicrichsediments:anevaluationoforganicmatteranalysismethods.ChemicalGeology200,225-240.

Seyfried,W.,Chen,X.,Chan,L.-H.,1998.Traceelementmobilityandlithiumisotopeexchangeduringhydrothermalalterationofseafloorweatheredbasalt:anexperimentalstudyat350C,500bars.GeochimicaetCosmochimicaActa62,949-960.

Smoliar,M.I.,Walker,R.J.,Morgan,J.W.,1996.Re-OsAgesofGroupIIA,IIIA,IVA,andIVBIronMeteorites.Science271,1099-1102.

Stanley,S.M.,Hardie,L.A.,1998.Secularoscillationsinthecarbonatemineralogyofreef-buildingandsediment-producingorganismsdrivenbytectonicallyforcedshiftsinseawaterchemistry.Palaeogeography,Palaeoclimatology,Palaeoecology144,3-19.

Steemans,P.,Herisse,A.L.,Melvin,J.,Miller,M.A.,Paris,F.,Verniers,J.,Wellman,C.H.,2009.Originandradiationoftheearliestvascularlandplants.Science324,353.

Stoffynegli,P.,Mackenzie,F.T.,1984.Massbalanceofdissolvedlithiumintheoceans.GeochimicaetCosmochimicaActa48,859-872.

Štorch,P.,1995.Bioticcrisesandpost-crisisrecoveriesrecordedbySilurianplanktonicgraptolitefaunasoftheBarrandianarea(CzechRepublic).Geolines3,59-70.

Sutcliffe,O.E.,2000.CalibratingtheLateOrdovicianglaciationandmassextinctionbytheeccentricitycyclesofEarth'sorbit.Geology28(11),967-970.

Tessier,A.,Campbell,P.G.,Bisson,M.,1979.Sequentialextractionprocedureforthespeciationofparticulatetracemetals.Analyticalchemistry51,844-851.

Thomas,M.F.,Thorp,M.B.,1995.GeomorphicresponsetorapidclimaticandhydrologicchangeduringthelatePleistoceneandearlyHoloceneinthehumidandsub-humidtropics.QuaternaryScienceReviews14,193-207.

Tomascak,P.B.,2004.Developmentsintheunderstandingandapplicationoflithiumisotopesintheearthandplanetarysciences.ReviewsinMineralogyandGeochemistry55,153-195.

Torres,M.A.,West,A.J.,Li,G.,2014.SulphideoxidationandcarbonatedissolutionasasourceofCO2overgeologicaltimescales.Nature507,346-349.

Torsvik,T.H.,Cocks,L.R.M.,2013.NewglobalpalaeogeographicalreconstructionsfortheEarlyPalaeozoicandtheirgeneration.GeologicalSociety,London,Memoirs38,5-24.

Tranter,M.,1982.Controlsonthechemicalcompositionofalpineglacialmeltwaters.UniversityofEastAnglia.

Tranter,M.,Sharp,M.J.,Lamb,H.R.,Brown,G.H.,Hubbard,B.P.,Willis,I.C.,2002.GeochemicalweatheringatthebedofHautGlacierd'Arolla,Switzerland?anewmodel.HydrologicalProcesses16,959-993.

177

Trotter,J.A.,Williams,I.S.,Barnes,C.R.,Männik,P.,Simpson,A.,2016.Newconodontδ18OrecordsofSilurianclimatechange:Implicationsforenvironmentalandbiologicalevents.Palaeogeography,Palaeoclimatology,Palaeoecology443,34-48.

Turgeon,S.C.,Creaser,R.A.,2008.Cretaceousoceanicanoxicevent2triggeredbyamassivemagmaticepisode.Nature454,323-326.

Vance,D.,Teagle,D.A.,Foster,G.L.,2009.VariableQuaternarychemicalweatheringfluxesandimbalancesinmarinegeochemicalbudgets.Nature458,493-496.

Verney-Carron,A.,Vigier,N.,Millot,R.,2011.ExperimentaldeterminationoftheroleofdiffusiononLiisotopefractionationduringbasalticglassweathering.GeochimicaetCosmochimicaActa75,3452-3468.

Vigier,N.,Decarreau,A.,Millot,R.,Carignan,J.,Petit,S.,France-Lanord,C.,2008.QuantifyingLiisotopefractionationduringsmectiteformationandimplicationsfortheLicycle.GeochimicaetCosmochimicaActa72,780-792.

Wadham,J.L.,Bottrell,S.,Tranter,M.,Raiswell,R.,2004.StableisotopeevidenceformicrobialsulphatereductionatthebedofapolythermalhighArcticglacier.EarthandPlanetaryScienceLetters219,341-355.

Walker,J.C.G.,Hays,P.B.,Kasting,J.F.,1981.Anegativefeedbackmechanismforthelong-termstabilizationofEarth'ssurfacetemperature.JournalofGeophysicalResearch86,9776.

Wildman,R.A.,Berner,R.A.,Petsch,S.T.,Bolton,E.W.,Eckert,J.O.,Mok,U.,Evans,J.B.,2004.TheweatheringofsedimentaryorganicmatterasacontrolonatmosphericO2:I.Analysisofablackshale.AmericanJournalofScience304,234-249.

Wimpenny,J.,James,R.H.,Burton,K.W.,Gannoun,A.,Mokadem,F.,Gíslason,S.R.,2010.Glacialeffectsonweatheringprocesses:NewinsightsfromtheelementalandlithiumisotopiccompositionofWestGreenlandrivers.EarthandPlanetaryScienceLetters290,427-437.

Young,S.A.,Saltzman,M.R.,Foland,K.A.,Linder,J.S.,Kump,L.R.,2009.Amajordropinseawater87Sr/86SrduringtheMiddleOrdovician(Darriwilian):Linkstovolcanismandclimate?Geology37,951-954.

Žigaitė,Ž.,Joachimski,M.M.,Lehnert,O.,Brazauskas,A.,2010.δ18OcompositionofconodontapatiteindicatesclimaticcoolingduringtheMiddlePridoli.Palaeogeography,Palaeoclimatology,Palaeoecology294,242-247.

178

Chapter5

Conclusion

179

Thisbodyofworkhasdeterminedtheosmiumisotopiccompositionseawaterforpastand

presentoceansbyapplyingRe-Osisotopemeasurementstoorganic-richshalesand

macroalgaerespectively.Thishasledtoabetterunderstandingofbasalticweatheringin

Iceland,mankind’sinfluenceontheOsisotopecycleandthediscoveryofseveralSilurian

glaciations.Thisfinalchapterbrieflydrawstogetherkeyaspectsofeachchapterandthen

suggestspossibledirectionsforfuturework.

5.1Re-Osisotopeuptakeanddistributioninmacroalgae

ResearchconductedbyB.Racionero-GómezattheUniversityofDurham,inpartunder

supervisionbymyself,lookedatReandOsuptakeanddistributioninthecommon

macroalgaespecies,Fucusvesiculosus(Racionero-Gómezetal.,2016;Racionero-Gómezet

al.,2017).ThesestudiesdemonstratedReandOsvarieswithinthemacroalgaeandthatRe

andOsarenotlocatedwithinonespecificstructure.Rhenium,andthereforemostlikely

Os,isnotheldwithinthechloroplastsorcytoplasmicproteins.RheniumandOsabundance

inculturedmacroalgaeincreasedwithincreasingculturemediaabundance.Moreover,

culturedmacroalgaedopedinseawaterwithaknownOsisotopiccomposition,tookonthe

isotopicsignatureofthefluid,withnosignsoffractionation(Racionero-Gómezetal.,

2017).Rheniumdidnotaccumulateindeadmacroalgae,suggestingsyn-life

bioadsorption/bioaccumulation(Racionero-Gómezetal.,2016).

ThisworksuggestedReandOsinmacroalgaearetakenupdirectlyfromthewater

columninwhichtheylive,wheretheyaccumulatetofarhigherconcentrationsthan

seawater.Meanwhile,theuptakeofOsbymacroalgaedoesnotfractionateOsisotopes

andmacroalgaeretainstheisotopiccompositionofseawater.Thisovercomesseveral

problemsassociatedwithdirectOsisotopeanalysisinseawater,suchasultra-low

concentrationsandmultipleoxidationstates(Peucker-Ehrenbrinketal.,2013).Itis

180

thereforepossibleforOsisotopesinmacroalgaetobecomeaproxyfortheOsisotopic

compositionofseawater,apowerfultracerofEarthsystemprocesses.Moreworkwillneed

tobedonetounderstandtheuptakemechanismandstoragelocationforReandOswithin

themacroalgae.

5.2Re-OsisotopesinmacroalgaeasatracerofEarthsystemprocesses

Chapter2utilisedReandOsabundanceandisotopedatafromIcelandicmacroalgae,

dissolvedloadandbedload.ThisworkestablishedthatReandOsabundanceandisotopic

compositionreflectedthatofthebrackish,estuarinehabitatsinwhichtheylive.Inthe

instanceofIceland,estuarineconditionsrepresentthemixingbetweenriversdraining

eitheryoungerbasalticcatchmentsthathaveundergonecongruentweathering,or,older

basalticcatchmentsthathaveundergoneincongruentweatheringofprimarybasaltic

mineralswithNorthAtlanticseawater.However,althoughthe187Os/188Oscompositionof

macroalgaeiscontrolledbythatofseawater,the187Re/188Oscompositionofseawaterisfar

higherthanIcelandicgeochemicalreservoirs.ThissuggeststhepreferentialuptakeofRe

overOsathighambientseawaterReconcentrations,anotionsupportedbyfurtherdata

presentedinChapter3.

Thisstudybuiltonpreviouswork(Seesection5.1)byconfirmingtheuseofOs

isotopesinmacroalgaeasaproxyforthe187Os/188Oscompositionofseawaterinareal-

worldsetting.However,italsobecameapparentthatmacroalgaecouldnotbeusedto

tracefluctuationintheOsabundanceand187Re/188Oscompositionofseawater.Morework

willneedtobedonetounderstandOsuptakeratesinarangeofcommonmacroalgae

species.ThiswillallowformoreaccurateestimatesoftheOsabundanceofseawater,

whichinturncanbeutilisedalongside187Re/188Osratiostohelpconstraintheresidence

timeofOsintheocean.Finally,macroalgaeisnotasubstantialsinkofReandOsandglobal

181

macroalgaebiomassdoesnotplayasignificantroleinthemarineOsandRecycle,both

todayandinEarth’sgeologicalpast.

5.3Re-Osisotopesinmacroalgaeasatracerofanthropogenicprocesses

Chapter3utilisedReandOsabundanceandisotopedatainJapanesemacroalgaetotrace

notonlynaturalprocesses,butalsoanthropogeniccontaminationassociatedwithhuman

activityindenselypopulatedregionsofJapan.The187Os/188Osofmacroalgaereflects

mixingbetweenseawaterand:riversdrainingMiocene-Holocenecontinentalrocks;and/or,

anthropogenicsources.Inparticular,theuseofPGEoresincatalyticconvertorshasledto

thewidespreadreleaseofappreciableunradiogenicOstotheatmosphereandmarine

environmentsurroundingmajorJapanesecities.Thisiscompoundedbyfurtherpoint

sourcereleaseofPGEderivedOsassociatedwithmedicalresearchandmunicipalsolid

wasteprocessing.

Thisstudybuiltonpreviouswork(Seesection5.1and5.2)byconfirmingthatthe

187Os/188OsofmacroalgaecouldtraceenvironmentalfluctuationsinanthropogenicOs

relatedtothewidespreaduseofPGEores.Osmiumisotopesinmacroalgaecantherefore

becomepowerfultracersofpollution.However,moreworkwillneedtobedonetofind

waystodistinguishbetweenthevariousanthropogenicsourcesofOs.Onewaycouldbeto

developtheuseofotherisotopeandelementalsystemswithspecificassociationtovehicle

use,municipalsolidwasteorrelevantmedicalresearchsources.Asanexample,Aland

organiccarbonconcentrationsinsedimentshavebeenusedinconjunctionwithOs

isotopestotraceanthropogenicOsassociatedwithsewage.Finally,Chapter3suggested

coastalcitiesareasignificantsourceofanthropogenicOstotheocean,andtheuseofPGE

oresincatalyticconvertorshasledtolower187Os/188Osratiosinglobaloceanicsurface

waters.

182

5.4TheOsandLiisotopiccompositionofSilurianseawater

InChapter4,Re-OsandLiisotopesystematicsweremeasuredinorganic-richshalesand

carbonates,fromsectionsthatspannedtheSilurianIreviken,Mulde,LauandKlonk

bioevents.OsmiumandLiisotopeprofilesweresimilartothosepreviouslyrecordedforthe

Hirnantianglaciation(Finlayetal.,2010;PoggevonStrandmannetal.,inreview)andare

associatedwithfluctuationintheweatheringoforganic-sulphide-richshalesandsilicates

respectively.ThissuggeststhatmuchliketheOrdovician,theSilurianispunctuatedby

severalglaciationeventsassociatedwithenhancedcontinentalicevolumeoverGondwana,

andissupportedbycarbonandoxygenisotopedataandsea-levelreconstructions.

Ascontinentaliceexpandssubglacialandproglacialprocessesacttoenhancethe

weatheringoforganicandsulphiderichshales,deliveringanincreasedfluxofradiogenicOs

totheoceans.Meanwhile,subglacialprocessesacttosuppresssilicateweatheringcausing

areductioninthedeliveryofisotopicallylightLitotheocean.Underglacialmaximumthe

supplyoffreshlycomminutedshalesceasesalongwithweatheringandthesupplyof

radiogenicOstotheocean.However,theenhancedicecovermaintainsrelativelylow

levelsofsilicateweathering.Duringdeglaciation,subglacialandproglacialprocesses

enhancetheweatheringoforganicandsulphiderichshalesandsilicates,delivering

radiogenicOsandisotopicallylightLitotheocean.

Thisstudyhasseveralmajorimplications:

1. TheSilurianwastraditionallythoughtofasagreenhouseliketheDevonianthat

proceedsit.However,thisworksuggeststheSilurianisanicehousemuchlike

theOrdovicianthatprecededit.

183

2. TheSilurianispunctuatedbyseveralglaciationsassociatedwithminormarine

extinctionsandareductioninglobaltemperaturesandsea-level.

3. OsmiumandLiisotopescanbeusedintandemtoreconstructfluctuationin

continentaloxidativeandsilicateweatheringrespectively.Ifusedproperly,

theseproxiescouldhelpusunderstandhowweatheringinfluences

atmosphericCO2throughgeologicaltime.

4. Subglacialandproglacialprocessesleadtoenhancedoxidativeweathering

whilstsuppressingsilicateweatheringduringglaciations.Thisactstoreverse

globalcoolingbyreleasingnetcarbondioxidetotheatmosphere.

5. Along-termdeclineinatmosphericCO2associatedwithorogeny,land-plant

diversification,reducedvolcanicarcdegassingand/orpaleogeography,is

ultimatelyreversedbyweatheringprocessesassociatedwithenhanced

continentalicevolume,preventingrunawayicehouseconditions.

6. ThisstudysupportstheideaoftheEarthhavingaself-regulatingclimatethat

maintainsahabitableplanet.

5.6Futureoutlook

Osmiumisotopesinmacroalgaeasaproxyforthe187Os/188Oscompositionofseawater

remainsinitsinfancydespitethegreatstridesmadeinthisbodyofwork.Onemajor

limitationistheinabilitytodeterminetheOsabundanceinseawaterusingmacroalgae.

MoreworkwillneedtobedonetodetermineOsuptakerelationshipsinasimilarmanner

topreviouswork(Racionero-Gómezetal.,2017)butatnaturalOslevels.Thiswillthen

havetoberepeatedforthemostcommongloballydistributedmacroalgaespecies.This

studyhasshownthe187Os/188Osofmacroalgaerepresentsthatoftheseawaterinwhichit

lives,andifitcouldbecombinedwithareconstructionof[Os],wouldallowthe

determinationofcontinentalinputsofOstotheocean.Thiswouldallowforbetter

constraintsonthemarineOscycleandOceanicresidencetimes.

184

Thisworkhasshownthe187Os/188Osofmacroalgaecantraceanthropogenic

sourcesofOs.However,itisdifficulttodistinguishbetweenthevarioussourcesduetothe

similarnatureofisotopicsignatures.Otherisotopeorelementalsystemswithaspecific

affinityforoneofthesesourceswillneedtobedevelopedinmacroalgaetodistinguish

betweenthesesources.Thiscouldhelpmacroalgaebecomeapowerfultracerofhuman

activityandausefulenvironmentalindicator.Inparticular,itcanbeusedtotracethe

influenceofwidespreadcatalyticconvertoruse,‘similartoPbfromleadedgasolineusage

before1978ortritiumfromatmosphericatomicbombtestingintheearly1960s’(Chenet

al.,2009).

ThisbodyofworkprovidedfurthergeochemicalevidenceforperiodicSilurian

glaciationsbyusingOsandLiisotopes.However,physicalevidenceformid-lateSilurian

glaciationsstillremainselusive.Moreworkwillneedtobedonetofindglacialsediments

forthetimeperiodsstudied.ThiswillprovideunequivocalevidenceforaSilurianicehouse

world.Chapter4suggestedthatOsandLiisotopeprofilesrepresentchangesinoxidative

andsilicaterespectively.However,likemostisotopessystems,thesesignalsremain

ambiguous.Datacollectedinthisstudycouldreflectchangesinweatheringofvariable

continentalrocks(notjustoxidativeweathering)orachangeinweatheringcongruency.

Moreweatheringproxiesneedtobeutilisedtoseparateoutthesesignals.Moreover,

sampleswereonlyanalysedinoneregionallocation.Inordertodetermineaglobalsignal

samplesfromgloballydistributessitesneedtobemeasured.

Finally,thecauseofSilurianglaciationsremainselusive.Mostproposed

mechanismsinthisstudysuggestalong-termdeclineinatmosphericCO2andglobal

temperaturesresultingfromchangesinorogeny,land-plantdiversification,volcanic-arc

degassingorpaleogeography.Moreworkwillneedtobedonetodeterminebetweenthese

185

driversofclimaticchange.Inparticular,complexcarboncyclemodellingfortheSilurian

mayprovideunparalleledinformation.

5.7References

Chen,C.,Sedwick,P.N.,Sharma,M.,2009.Anthropogenicosmiuminrainandsnowreveals

global-scaleatmosphericcontamination.ProceedingsoftheNationalAcademyofSciences

oftheUnitedStatesofAmerica106,7724-7728.

Finlay,A.J.,Selby,D.,Gröcke,D.R.,2010.TrackingtheHirnantianglaciationusingOs

isotopes.EarthandPlanetaryScienceLetters293,339-348.

Peucker-Ehrenbrink,B.,Sharma,M.,Reisberg,L.,2013.Recommendationsforanalysisof

dissolvedosmiuminseawater.Eos,TransactionsAmericanGeophysicalUnion94,73-73.

PoggevonStrandmann,P.A.E.,Desrochers,A.,Murphy,M.J.,Finlay,A.J.,Selby,D.,Lenton,

T.M.,inreview.GlobalclimatestabilisationbychemicalweatheringduringtheHirnantian

glaciation.GeochemicalPerspectivesLetters.

Racionero-Gómez,B.,Sproson,A.,Selby,D.,Gröcke,D.,Redden,H.,Greenwell,H.,2016.

Rheniumuptakeanddistributioninphaeophyceaemacroalgae,Fucusvesiculosus.Royal

SocietyOpenScience3,160161.

Racionero-Gómez,B.,Sproson,A.D.,Selby,D.,Gannoun,A.,Gröcke,D.R.,Greenwell,H.C.,

Burton,K.W.,2017.Osmiumuptake,distribution,and187Os/188Osand187Re/188Os

compositionsinPhaeophyceaemacroalgae,Fucusvesiculosus:Implicationsfordetermining

the187Os/188Oscompositionofseawater.GeochimicaetCosmochimicaActa199,48-57.