Geobiology. 2018;16:35–48. wileyonlinelibrary.com/journal/gbi  | 35© 2017 John Wiley & Sons Ltd

Received:25February2017  |  Accepted:01September2017DOI: 10.1111/gbi.12262

O R I G I N A L A R T I C L E

The late- stage “ferruginization” of the Ediacara Member (Rawnsley Quartzite, South Australia): Insights from uranium isotopes

L. G. Tarhan1 | N. J. Planavsky1 | X. Wang1 | E. J. Bellefroid1 | M. L. Droser2 |  J. G. Gehling3

1DepartmentofGeologyandGeophysics,YaleUniversity,NewHaven,CT,USA2DepartmentofEarthSciences,UniversityofCalifornia,Riverside,Riverside,CA,USA3SouthAustralianMuseumandUniversity ofAdelaide,Adelaide,SA,Australia

CorrespondenceL.G.Tarhan,DepartmentofGeologyandGeophysics,YaleUniversity,NewHaven,CT,USA.Email:lidya.tarhan@yale.edu

Funding informationNASAExobiologyprogram;NSF-ELTprogram;AlternativeEarthsNASAAstrobiologyInstitute;NSFEarthSciencesPostdoctoralFellowship;NationalGeographicSociety;AustralianResearchCouncilDiscoveryprogram;AmericanPhilosophicalSocietyLewisandClarkFundforExplorationandFieldResearchinAstrobiology

AbstractThepaleoenvironmentalsettinginwhichtheEdiacaraBiotalived,died,andwaspre-served in the eponymous Ediacara Member of the Rawnsley Quartzite of SouthAustralia isanissueof long-standinginterestandrecentdebate.Overthepastfewdecades,interpretationshaverangedfromdeepmarinetoshallowmarinetoterres-trial.Oneofthekeyfeaturesinvokedbyadherentsoftheterrestrialpaleoenvironmenthypothesisisthepresenceofironoxidecoatings,inferredtorepresenttheupperho-rizonsofpaleosols,alongfossiliferoussandstonebedsoftheEdiacaraMember.Wefindthatthesesurficialoxidesarecharacterizedby(234U/238U)valueswhicharenotinsecularequilibrium,indicatingextensivefluid-richalterationofthesesurfaceswithinthepastapproximately2millionyears.Specifically,theoxidecoatingsarecharacter-izedby (234U/238U)values>1, indicating interactionwithhigh-(234U/238U)fluidsde-rivedfromalpha-recoildischarge.Theseoxidesarealsocharacterizedbylight“stable”δ238/235Uvalues,consistentwithagroundwaterUsource.TheseUisotopedatathuscorroboratesedimentologicalobservationsthatferricoxidesalongfossiliferoussur-facesoftheEdiacaraMemberconsistofsurficial,non-bedform-parallelstaining,andsharply irregularpatches,stronglyreflectingpost-depositional, late-stageprocesses.Therefore, both sedimentological and geochemical evidence indicate that Ediacaraironoxidesdonotreflectsynsedimentaryferruginizationandthatthepresenceofironoxidescannotbeusedtoeitherinvokeaterrestrialpaleoenvironmentalsettingfororreconstruct the taphonomic pathways responsible for preservation of the EdiacaraBiota.Thesefindingsdemonstratethatcarefulassessmentofpaleoenvironmentalpa-rametersisessentialtothereconstructionofthehabitatoftheEdiacaraBiotaandthefactorsthatledtothefossilizationoftheseearlycomplexecosystems.

1  | INTRODUCTION

The Ediacara Biota—Earth’s earliest fossilized communities of com-plex,macroscopic,multicellularorganisms—isrecordedworldwide inupper Ediacaran strata. Although the systematic affinities of manyEdiacarataxaremainenigmatic,themorphologyandpaleoecologyoftheseexceptionallypreservedassemblagesofsoft-bodiedorganisms

provideawindowintotheevolutionofcomplexseafloorecosystems,foremosttheestablishmentofnovel,metazoan-typeecologicaladap-tations that set theseearlybenthiccommunitiesapart fromsimplerprecursorecosystems,aswellasthehabitatsinwhichtheyflourished.

TheupperEdiacaranEdiacaraMemberoftheRawnsleyQuartziteofSouthAustraliacontainsthetaxonomicallyrichestandmostpaleo-ecologicallydiverse fossil assemblagesof theEdiacaraBiota.At the

36  |     TARHAN eT Al.

NationalHeritagesiteNilpena,onthewesternmarginoftheFlindersRanges,inparticular,sequentialexcavationandreconstructionofhun-dredsofsquaremetersoffossiliferousbeddingplaneshave,overthepastdecade,facilitatedextensiveandhighlyresolvedbedding-planeand stratigraphic assessment of the paleoecological, paleoenviron-mental, and taphonomic character of Ediacara fossil assemblages.Coupled in situ paleontological and sedimentological analyses haveplayedanespecially important role inexplicating theenvironmentaldistribution of Ediacara taxa, and disentangling primary (reflectingdisparity inhabitat) fromsecondary (reflectingdisparity inpostmor-temfossilization)environmentalsignals(e.g.,Gehling&Droser,2013;Tarhan,Hood,Droser,Gehling,&Briggs,2016).

PaleoenvironmentalmodelsforthefossiliferousEdiacaraMemberand,byextension,reconstructionsofthehabitatoftheEdiacaraBiotahave, historically, covered a range of marginal and shallowmarinesettings.Jenkins, Ford, andGehling (1983), for instance, suggesteddepositioninanintertidal,lagoonal,andbarrier-barsetting.However,thisinterpretationwaspremisedlargelyonthepresenceoffeaturesoriginallydiagnosedasmudcracks (Jenkinsetal.,1983)butsubse-quentlyreinterpretedassyneresiscracksoccurring inatempestite-dominatedsuccessionentirelydevoidofmudstone (Gehling,2000).Onthebasisofmorerecentfieldmappingandsedimentologicalandpaleontologicalanalyses,variousauthorshaveobservedthatEdiacarafossils occur in sandstone-dominated successions—rich and com-monly current-perturbed in situ Ediacara fossil assemblages occuralong symmetrically rippled and tool-marked sandstone bedding-plane surfaces.These observations have been employed to recon-struct the depositional environment of the fossiliferous EdiacaraMember as shallowmarine and storm-reworked, ranging from be-tweenfair-weatherandstormwavebasetoprodeltaanddelta-fronttoincisedvalleys(e.g.,Gehling&Droser,2013;Gehling,2000;Tarhan,Droser,&Gehling,2015;Tarhan,Droser,Gehling,&Dzaugis,2017).Conversely,EdiacarafossilsareabsentfromtheunderlyingBonneySandstoneandChaceQuartziteMemberoftheRawnsleyQuartziteandtheoverlyingunnamedmemberof theRawnsleyQuartzite—allofwhichhavebeeninterpreted, inthefaciesmodelofGehlingandDroser (2013), to representmarginalmarine, floodplain, and fluvialenvironments.

In contrast, recent studies (Retallack, 2013) have alternativelyproposed a terrestrial (paleosol) depositional environment for thefossiliferous Ediacara Member. Despite major issues raised andstrong debate surrounding the paleosolmodel (e.g.,Tarhan,Droser,&Gehling,2015;Xiao&Knauth,2013;Xiaoetal.,2013),ithascon-tinued to play an active role in the dialogue of the geological andbiological research communities (e.g., Algeo, Marenco, & Saltzman,2016; Beraldi-Campesi, 2013; Beraldi-Campesi & Retallack, 2016;Kump,2014;Pandey&Sharma,2017;Retallack,2012,2013,2014a,b,2015,2016a,b).OneofthefundamentaltenetsonwhichthepaleosolmodelfortheEdiacaraMemberhasrestedistheargumentthattheironoxidecoatingspresentonmanyfossiliferoushorizonsrepresent“synsedimentaryferruginization”associatedwithsoilhorizondevelop-ment,andthus,aterrestrialdepositionalenvironmentfortheEdiacaraMember(Retallack,2013).

Although it has been suggested, on the basis of outcrop-scalesedimentologicalobservation,thattheseoxidesarelikelylate-stagealteration features (Tarhan, Droser, & Gehling, 2015), geochemi-cal support for this interpretationhaspreviouslybeen lacking.Wehereinemployuraniumisotopestodirectlytestwhetherthesefossil-associatedoxidesarelate-stageinorigin.Thisapproachalsoallowsus to explore the diagenetic history of oxide-rich surfaces that donotcontaindirectandobvioussedimentarysignaturesoflate-stageoxideformation.Ouranalysesindicatethattheironoxide-richlayersonwhichRetallack (2013)has foundedhisEdiacarapaleosolargu-ment record open-system behavior and surface- or groundwater-mediatedalterationandoxide formationwithin thepast~2millionyears.Therefore,ourUisotopeworkprovidesunequivocalevidencethat the presence and distribution of these iron oxides cannot beusedtoinferenvironmental,ecological,orpreservationalconditionssynchronouswiththedepositionandearlydiagenesisoftheEdiacaraMember.

2  | GEOLOGIC SETTING AND SAMPLE DESCRIPTION

The Flinders Ranges region of South Australia contains one of themost complete and best-exposed Neoproterozoic–Cambrian suc-cessionsintheworldandincludesthetypesectionfortheEdiacaranPeriod.TheEdiacaraMemberoftheRawnsleyQuartziteistheyoung-est Ediacaran unit in theNeoproterozoic–middle Cambrian succes-sionof theAdelaideFoldBeltofSouthAustralia (Figure1;Gehling,2000). The Ediacara Member, which is 10–300m in thickness, isdefinedatitsbasebyanerosionalunconformitywiththeunderlyingChaceQuartziteMemberoftheRawnsleyQuartziteandtheBonneySandstone(Countsetal.,2016).AttheNationalHeritagesite,Nilpena,onthewesternmarginoftheFlindersRanges,33discreteandfossilif-erousbeddingplanes,representingapproximately300squaremetersofEdiacaraseafloor,have,todate,beenexcavatedandreassembled.Sequential(bed-by-bed)excavationacrossmultiplefossiliferousfacieshaspermittedbothbroad-scaleandhighly resolved (e.g.,mm-scale)spatialandstratigraphicreconstructionoftheenvironmentalhabitat,ecologicalstructure,andsuccessionofEdiacaracommunities,aswellasthetaphonomicprocessesresponsiblefortheirpreservationinthefossilrecord.

The Ediacara Member consists of a siliciclastic sequence, tra-ditionally interpreted as shallow marine and deltaic in origin, re-peated in up to five parasequences across theAdelaide Fold Belt(Figure1).AtNilpena, fossils of theEdiacaraBiotaoccurprimarilyinfourfacies.InthefaciesmodelproposedbyGehlingandDroser(2013) and Tarhan, Droser, etal. (2017), these facies have beentermed the Flat-Laminated to Linguoid-Rippled Sandstone Facies,theOscillation-RippledSandstoneFacies,thePlanar-LaminatedandRip-UpSandstoneFacies,andtheChannelizedSandstoneandSand-BrecciaFacies(Gehling&Droser,2013;Tarhan,Droser,etal.,2017).Atthebaseofeachparasequence,theFlat-LaminatedtoLinguoid-Rippled Sandstone Facies (formerly described as the Delta-Front

     |  37TARHAN eT Al.

SandstoneFacies;cf.Gehling&Droser,2013)consistsoffining-andthinning-upwardsetsofpoorlysorted,laminated,redsiltysandstonebeds,alongwhichlinguoidripplesandtoolmarksoccur,withfossilsoccurringprimarilyinhyporeliefasexternalandentrainedcompos-itemolds,andoccasionally inepirelief (Droseretal.,2014;Tarhan,Droser,Gehling,&Dzaugis,2015).TheOscillation-RippledSandstoneFacies (formerlydescribedas theWave-BaseSandstoneFacies;cf.Gehling&Droser,2013),inwhichmanyofthemosticonicEdiacarafossilsoftheWhiteSeaAssemblage(cf.Waggoner,2003)occurashyporelief external and internalmolds and, less commonly, epire-lief casts, consists of thin-bedded, rippled, fine- to coarse-grainedfeldspathic quartz arenites. The Planar-Laminated and Rip-UpSandstoneFacies(formerlydescribedastheSheet-FlowSandstoneFacies;cf.Gehling&Droser,2013)consistsoflaterallycontinuous,planar-laminated fine-grained sandstone beds with erosive bases(some ofwhich bear crisply preserved toolmarks) and commonlycharacterizedbybed-topandintrabedsandstoneintraclasts.Fossilassemblagesinthisfaciesarepreservedashyporeliefexternalandinternalmolds.TheChannelizedSandstoneandSand-BrecciaFacies

(formerlydescribedastheMass-FlowSandstoneFacies;cf.Gehling&Droser,2013) ischaracterizedbylenticular,discontinuousmedi-um- to coarse-grained sandstone bodies ofmeter-scale thickness,inwhichball-and-pillowbedsgradeupintomassivesandstonebedswith scoured bases and bearing dish-and-pillar dewatering struc-tures. Fossils are preserved threedimensionallywithin these bedsandare commonlybiostratinomicallydeformed (Gehling&Droser,2013;Tarhan,Droser,etal.,2017).Thisfaciesmodelcontrastsstrik-inglywiththatofRetallack(Retallack,2012,2013),inwhichallfos-siliferousfaciesoftheEdiacaraMemberareinterpretedasterrestrialin origin.

Likemuchof theAustraliancontinent, theFlindersRanges re-gionhas,overthepastcouplehundredmillionsofyears,beenrel-ativelytectonicallystableand,sincethePermian,unglaciated,andhas experienced long intervals of emergence and subaerial expo-sure(e.g.,Anand,2005;Biermanetal.,2002;Bird&Chivas,1988;Milnes, Bourman, & Northcote, 1985; Pillans, 2005, 2007). Dueto this protracted exposure, the SouthAustralian surface and theFlindersRangesregioninparticularhave,overtheCenozoic,been

F IGURE  1 Geographicdistribution(a)andfacies-scalestratigraphiccharacterization(b)ofthefossiliferousEdiacaraMemberintheFlindersRangesofSouthAustralia.Gray(ina)denotesexposureofthePoundSubgroup.SamplesforthisstudywerecollectedfromtheNationalHeritageEdiacarafossilsiteNilpena(notedinred).ModifiedfromTarhan,Droser,etal.(2017)[Colourfigurecanbeviewedatwileyonlinelibrary.com]

35°S

30°S

135°E 140°E

31°00′S

139°00′E138°00′E

31°00′S

32°00′S

138°00′E

0 20 km

0 300 km

Adelaide

Quorn

Hawker

Mt. Havelock

Moralana

Bunyeroo Gorge

Brachina Gorge

Bathtub Gorge

Parachilna Gorge

Blinman

Nilpena

Ediacara Range

Leigh CreekMt. Scott

Range

Mt. Mantell Range

Tooth Nob

Wilpena Pound

Chace Range

Elde

r Ran

geH

eyse

n R

ange

Lake Torrens

(a) (b)

180

100

140

160

120

80

0

20

40

60

m

Cha

ce Q

uartz

ite M

brE

diac

ara

Mbr

Raw

nsle

y Q

uartz

ite

Cross-Bedded Sandstone Facies: white-colored, medium- to coarse-grained cross-stratified quartz-feldspathic arenite

Oscillation-Rippled Sandstone Facies: gray-white-colored, fine- to coarse-grained wave-rippled, thin- to medium-bedded sandstone

Channelized Sandstone and Sand-Breccia Facies: white- to brown-colored, medium- to coarse-grained massive, slumped feldspathic sandstone

Flat-Laminated to Linguoid-Rippled Sandstone Facies: khaki-colored, red-weathering, fine- to coarse-grained, silty, laminated to irregularly bedded, feldspathic and poorly sorted sandstone

Chace Quartzite Member: white-colored, fine- to coarse-grained, petee-bedded, rippled feldspathic sandstone

Bonney Sandstone: red- to brown-colored, well-bedded, poorly sorted sandstone

Legend: Nilpena succession

Bonney Sandstone

38  |     TARHAN eT Al.

subjectedtoaseriesofintensiveweatheringepisodesand,overtheNeogeneandQuaternary,upliftanderosion,whichhaveresultedinageomorphologicallycomplexandvariablelandscape.Geochemicalandgeochronologicalarchives(e.g.,δ18OsignaturesofregolithclaysandpaleomagneticdatingofsecondaryFeandMnoxides)indicatethatduringthePaleocene–EoceneandMiocene,thisregionexperi-encedepisodesofparticularlyintensiveweathering,includingdeepoxidation, regolithdevelopment, and iron remobilization, followedbyaPleistoceneshifttomorearidconditions(Anand,2005;Bird&Chivas,1988;Pillans,2005,2007).

The Flinders Ranges are a topographically rugged, relativelyhigh-relief(upto600–1,000m)uplandssystemcomprisedofridgesandvalleys and surrounded by low-elevation, low-relief areaswithinternallydrainingplaya lakesystems (Quigley,Sandiford,Fifield,&Alimanovic,2007a,b;Twidale,2000).TheFlindersregionisoneofthemostseismicallyactivezonesinAustralia,andthemorphologyoftheFlinderslandscapehasbeenshapedlargelybyrecent(lateMiocene–present) tectonic activity (Quigley, Sandiford, Fifield,&Alimanovic,2007b).AlthoughtheFlindersmay,aslow-reliefhills,havehadsur-faceexpressionasfarbackastheMesozoic (Callen,1977;Twidale,2000),sedimentary,cosmogenic,andseismicdataindicatethatupliftalongrange-front reverse faults initiated in the lateMiocene, likelydue toeast–west compressional stresses (the intraplatemanifesta-tion of far-field plate-margin stresses; Anand, 2005; Callen, 1977;Quigley, Sandiford, Fifield, & Alimanovic, 2007a,b). Data from insitu-produced cosmogenic nuclides, particularly 10Be, indicate thattheFlinders region ischaracterizedbyhighlyvariable ratesofbed-rockandalluviumdenudation,rangingfromaverageerosionratesof3.0m/myrforWilpenaPoundand2.7m/myrfortheBrachinapied-monts (Bierman etal., 1998, 2002) to 10m/myr in the Parachilnaareato14.2–22.8m/myrforthenorthernFlindersand5–79m/myrforcatchments inthecentralFlinders(Quigley,Sandiford,Fifield,&Alimanovic, 2007a,b and references therein). Hill-slope and valleyfloor erosion rates and stream incision rates are commonly higherthan summit bedrockdenudation rates; this disparity is, in concertwithhighupliftrates,responsibleforthehigh-reliefandtopographicruggedness of the Flinders. Areas of adjacent ridge-front alluvialand playa basins are also undergoing active subsidence (Quigley,Sandiford,&Cupper,2007).Thevariability inerosion ratescharac-teristicoftheFlindersregionindicatesthat,evenwherelocalerosionrates are slow (decimeters to severalmeters of erosionpermillionyears), weathering and erosion are ongoing and dynamic driversof landscapemorphology (Bierman&Caffee, 2002; Bierman etal.,2002;Quigley,Sandiford&Cupper2007;Quigley,Sandiford,Fifield,&Alimanovic,2007a,b).

Wecollectedfossiliferoussamplesforgeochemicalanalysesfromthree facies (Figure1) inwhich ironoxidesoccurprominently alongexcavatedfossiliferousbedding-planesurfaces(Figure2).Specifically,wesampledfrom[followingthefaciesmodelofTarhan,Droser,etal.(2017)]theFlat-LaminatedtoLinguoid-RippledSandstoneFacies,theOscillation-RippledSandstoneFacies,and thePlanar-LaminatedandRip-UpSandstoneFacies.Ineachofthesefacies,epireliefandhypore-liefbedding-planesurfacesofmature,silica-cementedquartz-feldspar

arenites are characterized by ruddy-hued iron oxide-rich coatings(Figure3).Theseoxides,whichrepresenttheupper,“depletedzone”ofRetallack’sputativepaleosols,areconfinedtobedding-planesurfacesand do not permeate the sandstone beds (Figure3), nor do oxidescontributetocementphases,whichareessentiallyexclusivelyquartz(Figure4;Tarhan,Droser,&Gehling,2015;Tarhanetal.,2016).Ouranalyticaleffortswerethereforefocusedoncharacterizationoftheseoxidecoatings.

3  | URANIUM ISOTOPE SYSTEMATICS

Uraniumisotopes—particularlytheshort-livedisotope234U—provideausefulproxytoconstrainweatheringprocesses. Ithas longbeenobserved that authigenic minerals record the U isotopic (e.g.,234U/238U) composition of the fluids from which they precipitate(e.g.,Broecker&Thurber,1965).Further, theactivity ratioof234U to 238U (denoted as (234U/238U)) of secondarymineral phases andcontinentalfluidshasbeenextensivelyemployedasameanstotrackweatheringprocesses(Chabaux,Bourdon,&Riotte,2008;DePaolo,Lee, Christensen, & Maher, 2012; DePaolo, Maher, Christensen,&McManus, 2006;Dosseto, Bourdon, &Turner, 2008;Hansen&Stout,1968;Maher,DePaolo,&Christensen,2006;Mathieu,Bernat,&Nahon,1995).Aspartoftheuranium–thoriumradiogenicdecaychain, 238U (t1/2=4.46gyr) transforms to

234U (t1/2=245kyr) bymeans of decay through two short-lived intermediate nuclides:alphadecayto234Th(t1/2=24.1days)andbetadecayto

234Pa(t1/2 =1.2min) (Bourdon,Turner,Henderson,&Lundstrom,2003).Duetothedisparatedecayconstantsof234U and 238U,therelativeratesof234Uadditiontoandlossfromthesystemwillrapidlyequilibrateand (234U/238U) will, under closed-system conditions and regard-lessofinitialactivityratiovalue,reachwhatisreferredtoassecularequilibrium((234U/238U)=1;Bourdonetal.,2003). Inaclosedsys-tem, secular equilibriumwill be reachedwithin approximately fivehalf-livesof234U—that is, inapproximately2myr (Figure5).Underoxidizing conditions, interaction ofU-bearingmineral phaseswithexogenousfluidswill result inpreferential leachingof234U,duetoalpha-recoil damage incurred to the mineral lattice during alphadecay(DePaoloetal.,2012).ThiswillleadtoU-bearingfluidphaseswith(234U/238U)>1andresidualUsolidphaseswith(234U/238U)<1. Uraniumwill,underoxidizingconditions, take the formofhexava-lenturanium(U(VI))andbesignificantlymoresolublethanreducedtetravalenturanium(U(IV)) (Langmuir,1978).SolubleU(VI)canbe-come incorporated into iron oxide-rich sediments through copre-cipitationwith or sorption to ferric oxides (Brennecka,Wasylenki,Bargar,Weyer,&Anbar,2011).

Therefore,intheEdiacaraMember,eithercoprecipitationofUwithFeoxides, as suggestedbyoutcrop-scale evidence for reac-tion fronts (Tarhan,Droser,&Gehling, 2015), or adsorption ofUto previously precipitated Fe oxidesmust have occurred throughmobilizationofUbyoxidizedfluids.IfUwasrecently(<2Ma)mo-bilized(i.e.,leachedfrom,precipitatedwithorsorbedtoFeoxides),oxide-boundUshouldbecharacterizedby(234U/238U)valueswhich

     |  39TARHAN eT Al.

are not at secular equilibrium (i.e., (234U/238U)<1 or >1, not=1).Therefore, a (234U/238U) signature for secular disequilibrium canbe used as a direct indication of open-system behavior and a

late-stage (i.e., influenced by interaction with oxidized surfacewaters or groundwaters within the past 2myr) origin for theEdiacaraMemberironoxides.

F IGURE  2  IronoxidecoatingsofthefossiliferousEdiacaraMember.(a)Excavatedbedding-planesurfacecharacterizedbyprominentbutpatchyFeoxidecoatings.Examplesofsharp,irregularjunctions,suggestiveofprecipitationalongreactionfronts,denotedbyblack-rimmedwhitearrows.Blackarrowsdenoteexamplesofthebiogenicstructure“mop,”characterizedbystrongunidirectionalalignmentandinterpretedtohaveformedbydraggingandremovaloffrondoseorganismsfromtheorganically-boundsubstrate(Tarhan,Droser,&Gehling,2010).Oxidizedandnon-oxidizedportionsofthebedaresedimentologicallyindistinguishable;preservationoffossilsandmopsdoesnotcovarywithextentofoxidation.Hyporelief.Eachincrementofscalebar=1cm;chalkmarksdenote1m×1mgrids.(b,c)FieldphotographsoftheholdfastformgenusAspidella,associatedwithirregularFeoxidepatches(notedbyblack-rimmedwhitearrows).EvenwhereFeoxidespatchilycoverportionsofindividualAspidella,fossilsarecharacterizedbynodiscernibledifferenceinpreservationalfidelityorsedimentology.Hyporelief.Scalebars=5cm[Colourfigurecanbeviewedatwileyonlinelibrary.com]

40  |     TARHAN eT Al.

F IGURE  3 PhotographsofrepresentativefossilsamplesusedforUisotopeanalyses.Bedding-plane(hyporelief)(a,c,e,g)andcross-sectional(b,d,f,h)viewsareshownforeachspecimen;cross-sectionalviewsarestratigraphicallyoriented(fossiliferoussurfacealongbase).(a,b)Dickinsonia,associatedwiththetexturedorganicsurface“micropucker”(seeblack-rimmedwhitearrow),collectedfromthePlanar-LaminatedandRip-UpSandstoneFaciesofTarhan,Droser,etal.(2017).(c,d)ThetubularfossilFunisia(black-rimmedwhitearrow),collectedfromthePlanar-LaminatedandRip-UpSandstoneFaciesofTarhan,Droser,etal.(2017).(e,f)Dickinsonia,collectedfromtheOscillation-RippledSandstoneFaciesofTarhan,Droser,etal.(2017).(g,h)Dickinsonia,associatedwiththetexturedorganicsurface“weave”(seeblack-rimmedwhitearrows),collectedfromtheOscillation-RippledSandstoneFaciesofTarhan,Droser,etal.(2017).Notethat,inallcases,asshownincross-section(b,d,f,h),Feoxides(blackarrows)areconfinedtosampleexteriorandlate-stagefractures,whereastheinteriorsofsandstonesamplesconsistofquartz-cementedquartz-feldsparareniteandaredepauperateinFe-richdetritalorauthigenicphases.Scalebars=1cm[Colourfigurecanbeviewedatwileyonlinelibrary.com]

     |  41TARHAN eT Al.

F IGURE  4 PetrographyandelementalcompositionofEdiacaraMemberfossils.(a)CathodoluminescencebrightnessmapofAspidella holdfastfromthePlanar-LaminatedandRip-UpSandstoneFacies,consistingofauthigenicsilica-cementeddetritalquartzandpotassiumfeldspar(ksp)grains.Areaslabeled“r”areresin-impregnatedholes.(b–e)Energydispersivespectroscopyelementalmapsofsameareaasdepictedin(a),whichindicateEdiacarafossilsareassociatedwithenrichmentsinSiandO,andapaucityofFeorC(areasofbrightnessinCmapindicateresin-impregnatedvoidsandfracturesinfeldspargrains).Relativeelementalabundanceisdenotedbydistributionandbrightnessofcolor.(a,bandd)modifiedfromTarhanetal.(2016).Scalebars=100μm[Colourfigurecanbeviewedatwileyonlinelibrary.com]

Si Fe

O C

(b)

(c) (e)

(d)(a)ksp

kspksp

ksp

ksp + r

r

F IGURE  5 Theoreticalclosed-systemradiogenicdecaycurvescalculatedfromarangeofinitial234U/238Uactivityratios(234U/238U).Positionofx-axis(at(234U/238U)=1)denotessecularequilibrium.(234U/238U)valueswerecalculatedusingtheformula(234U∕238U)= ((234U∕238U)i−1)×e(−2.82×10

−6)t+1(Faure,1986),where(234U/238U)idenotesinitialactivityratiovalue,tdenotesdurationoftime(inyears),and2.82×10−6isthedecayconstantof234U(Chengetal.,2013).Forallinitial(234U/238U)values,rangingfrom1.15to10,secularequilibriumwasreachedin<5myr.Insetshowstherangeof(234U/238U)spaceoccupiedbyEdiacaraFeoxide-richsamples(tealbox)anddemonstratesthat,underallmodeledconditions,EdiacaraFeoxide-richsamplesareinseculardisequilibrium.Solidblackline:initial(234U/238U)=1.15;dashedblackline:initial(234U/238U)=2;solidgrayline:initial(234U/238U)=5;dashedgrayline:initial(234U/238U) = 10 [Colourfigurecanbeviewedatwileyonlinelibrary.com]

4.00 0.5 1.0 1.5 3.53.02.52.01

2

3

4

5

6

7

89

10

Time (myr)

(234 U

/238 U

)

1.701.651.601.551.501.451.401.351.301.251.201.101.701.051.00

0 0.5 1.0 1.5 2.0 2.5 3.0Time (myr)

(234 U

/238 U

)

42  |     TARHAN eT Al.

4  | METHODS

4.1 | Sample preparation and purification

EdiacaraFeoxidesampleswereisolatedfromfossiliferoushandsam-ples(representingallthreeoftheEdiacarafossiliferousfaciescharac-terizedbyabundantFeoxidecoatings),usingaDremelmicrodrillwithatungstencarbidebit.SampleswerecompletelydigestedinaHNO3-HClmixtureat130°Cfor24h.AsubsequentHNO3-HCl-HFthree-stepdi-gestwascarriedoutontheresiduetoconfirmquantitative(>99%)Umobilization.Uraniumandotherelementalconcentrationsweremeas-uredonaThermoFinniganElementXRICP-MS.Accuracyandpreci-sionweremonitoredwith analysis of USGS standard BHVO-2, anderrorwaslessthan5%forUandThconcentrationmeasurements.

For U isotope analysis, we followed the procedure of Wang,Johnson, andLundstrom (2015),modified fromWeyer etal. (2008).Inbrief,weusedthe233U-236Udoublespiketechniquetocorrectforisotope fractionationsduring samplepreparationandmeasurement.BasedonUconcentrations,appropriateamountsofsamplealiquotscontaining20–50ngUweredopedwiththe233U-236Udoublespikesoastoyield238U/236Uratiosof~30.Followingspiking,sampleswereslowlyevaporated,thenredissolvedin3NHNO3toachievesamplespikeequilibration.UraniumwasisolatedusingUTEVAionexchangeresin,basedontheprocedureofWeyeretal.(2008).

4.2 | Uranium isotope measurement

The U isotope composition of Ediacara Fe oxide-rich samples wasmeasuredonaThermoNeptunePlusMC-ICP-MShousedattheYaleMetalGeochemistryCenter.Theinstrumentwasoperatedat1,100Wplasmapowerandwassetatlow-massresolution.Sample(Ar-carried)gasandN2gasweretunedtomaximizetheirsensitivityandstability.PurifiedUsampleswerequantitativelydissolvedin0.75NHNO3todilutesample[U]to50ppb.Sampleswereintroducedintotheplasmawitha50μl/minPFAμFlownebulizer(ElementalScientific)connectedtoanApexIRsampleintroductionsystemwithoutmembranedesol-vation.WeusedN2tostabilizethesignal,jetsample,andHskimmercones.Inthisconfiguration,≥30volt238Uwasroutinelyachievedona1011 Ωresistorwitha50ppbUsolution.

Collector configuration settings followed those of Wang etal.(2015). Amplifier gain factors were measured before each session.Giventhelargecontrastin238U and 236Usignals,toavoidinterferenceof238U on 236Umeasurement,weusedatwo-zerobackgroundsub-tractingmethod,wherebyon-peakbackgroundsignalsweremeasuredfor 30s each at halfmasses below and above the nominalmasses.Resultingaveragevaluesweresubtractedfromthosemeasuredatthenominalmasses.Onemeasurementcomprisesfiveblocksof10cycles,witheachcyclelasting4.19s.EverythreesampleswerebracketedbyspikedinternationalstandardCRM112a(NewBrunswickLaboratory,U.S. DOE) solution at the same concentration. Sample valueswerenormalized to the bracketing standards, with errors propagated.Reportedδ238Uvalues represent thedeviationof sample 238U/235U fromCRM112a,inunitsofpartsperthousand:

4.3 | Data quality

The blank U for the total procedure (digestion, column chemistry)was always <100pg, which was negligible compared to 50ng sam-ples. Therefore blank correction was not performed. To monitoraccuracy and precision, we repeatedly processed USGS BHVO-2through the digestion and column chemistry (δ238U=−0.24±0.06‰ 2σ, n=6; (234U/238U)=1.001±0.005 2σ, n=6), and we ana-lyzed unprocessed secondary standard solutions of CRM129a(New Brunswick Laboratory, U.S. DOE; δ238U=−1.71±0.04‰ 2σ,n=17; (234U/238U)=0.973±0.013 2σ, n=17) and IRMM REIMEP18a (JRC, Brussels, Belgium; δ238U=−0.16±0.05 ‰ 2σ, n = 14; (234U/238U)=1.027±0.0082σ,n=14).Theresultsofthesestandardsagree,withinerror,withpreviousstudies(e.g.,Shieletal.,2013;Tissot&Dauphas,2015).Theexternalprecisionsforδ238Uand(234U/238U)meas-urements(below)are0.06‰and0.013,respectively,estimatedbythelargesttwo-standarddeviationsofvariousmeasuredstandards.Samplemeasurementuncertaintyisreportedaseitherthetwo-standarddevia-tionofBHVO-2orasthetwo-standarddeviationof50measurementcyclesforeachindividualsample(withthelargeruncertaintyreported).AsnotedinTable1,threesamples(withlimitedmaterial)havelargerer-rors(δ238U 2σ=0.2‰and(234U/238U) 2σ = 0.045).

5  | RESULTS

EdiacaraFeoxide-richlayersarecharacterizedbyU/Thratiosnotablyhigher than crustal values (mean U/Th=22.78, 1σ = 13.60; n = 16; Figure6,Table1).Essentially, allU (>99%)was removed in thehotHNO3-HCldigest,withnegligibleUextractedduringtheHF-HNO3-HCldigest.Therefore,eachHNO3-HCldigestwasregardedas,essen-tially,atotaldigest.Theδ238Uvaluesrangefrom−1.14‰to−0.27‰ (mean δ238U=−0.62‰, 1σ=0.29‰, n=12; Figures6 and 7,Table1).Thesevaluesaredepletedrelativetotheaveragecontinentalcrust,surfacewaters,andthetopsofmodernsoilprofiles(cf.Tissot&Dauphas,2015;Weyeretal.,2008).The(234U/238U)valuesallfallabove the secular equilibrium line; mean sample (234U/238U)=1.19(1σ=0.12,n=14)(Figures6and7,Table1),indicatingthatEdiacaraoxide-boundUisatseculardisequilibrium.

6  | DISCUSSION

Uraniumisotopesignaturesinironoxide-richlayersalongfossilifer-ous sandstones of the EdiacaraMember are tied to authigenicUenrichments. Ediacara fossils are preserved as casts andmolds inmaturequartz-feldspararenitesdevoidoforganicmatterand,apartfromtheanalyzedsurficialcoatings(whichweresampledbymicro-drilling),depauperateinoxides(e.g.,Figures3and4;Tarhanetal.,

δ238U(‰) =

(

238U∕235Usample

238U∕235UCRM112a

− 1

)

× 1,000.

     |  43TARHAN eT Al.

2016).AlloftheexaminedsamplescontainU/Thratioswellabovecrustal averages, aswell as substantiallyhigher than the rangeofU/Thratiosrecordedbycontinentaltopsoildatabases(cf.Rudnick&Gao,2014;Cole,Zhang,&Planavsky,2017),consistentwithlim-itedU in the detrital fraction. Further, the examined samples arecharacterized by non-crustal δ238U values (cf. Tissot & Dauphas,2015;Figures6 and7), alsonecessitating that themeasuredU ispredominatelyauthigenic (non-detrital),givenevidencefor limited

δ238Uvariabilityinigneousrocks(e.g.,Tissot&Dauphas,2015)andNorthAmericansoils,whicharerepresentativeofdetritalmaterials(DeCorte,2016).

Ediacara (234U/238U) values are consistently higher than secularequilibrium ((234U/238U)>1) and fallwithin the range typicalofmod-ern aqueous fluids, including groundwaters [natural groundwater(234U/238U)range:0.40–15.38,mean:2.35±2.191σbasedon>490datapoints(Abdul-Hadi,Alhassanieh,&Ghafar,2001;Andrews&Kay,

TABLE  1 UraniumdatafromanalyzedEdiacaraFeoxide-richsamples.Reported2σvaluesdenoteexternalprecisiononUisotopemeasurements

Sample U (ppm) Th (ppm) U/Th δ238U (‰) 2σ (‰) (234U/238U) 2σ

N-E-MM5-3A 0.61 0.04 15.92 – – – –

N-E-MM3-1A 0.65 0.03 19.25 −0.40 0.07 1.04 0.013

N-E-TSSS-3 0.15 0.01 20.05 – – 1.18 0.019

N-E-1T-D-2 0.18 0.01 22.74 −1.14 0.23 1.15 0.013

RQ-EM-H-8A-xB 17.99 0.30 60.84 −0.40 0.06 1.53 0.013

N-E-TSSS-4 0.53 0.06 9.42 −0.56 0.08 1.21 0.013

N-E-TSSS-5 2.11 0.09 23.72 −0.73 0.11 1.22 0.013

N-E-TSSS-1 0.61 0.06 10.97 −0.45 0.06 1.24 0.013

N-E-MM5-2 0.70 0.04 13.06 −0.27 0.06 1.10 0.013

RQ-EM-H-8A-T 1.20 0.11 11.04 −0.33 0.06 1.16 0.013

N-E-MM3-2T 1.00 0.06 16.99 – – 1.29 0.045

RQ-EM-H-8A-xA 3.20 0.07 44.13 – – – –

N-E-MAB-F-2 1.70 0.07 24.29 −1.03 0.13 1.18 0.013

N-E-MM3-2B 3.70 0.10 35.60 −0.98 0.23 1.09 0.013

N-E-R-AspA 1.20 0.06 20.77 −0.53 0.07 1.13 0.013

N-E-S-2 1.60 0.10 15.76 −0.68 0.11 1.16 0.013

F IGURE  6 δ238Uvalues(a)and234U/238Uactivityratios(234U/238U)(b)versusU/ThratiosforanalyzedEdiacaraFeoxide-richsamples.Horizontaldashedgrayline(a)denotesaveragecrustalδ238U(cf.Tissot&Dauphas,2015).Verticaldashedblueline(a,b)denotesaveragecrustalU/ThvalueasapproximatedbytopsoilmeanU/Thvalueof0.285,with2σ(±0.369)denotedbyblueshading(Coleetal.,2017);thismeanvalue,whichwascalculatedfromUSGSsoildataforthecontinentalUSA(Coleetal.,2017),closelymatchesotherestimatesforaveragecrustalU/Th valuederivedfromsedimentary,loess,androckdata(cf.Rudnick&Gao,2014).Horizontaldashedblackline(b)denotessecularequilibrium.Wherenotvisible,errorbarssmallerthansymbols[Colourfigurecanbeviewedatwileyonlinelibrary.com]

U/Th (ppm/ppm)

1.6

1.5

1.4

1.3

1.2

1.1

1.0

0.9

0.80 10 20 30 40 50 60 70

(234 U

/238 U

)

secular equilibrium

0 10 20 30 40 50 60 70–1.4

–1.2

–1.0

–0.8

–0.6

–0.4

–0.2

0.0

δ238 U

(‰)

n = 12 n = 14

(a) (b)

44  |     TARHAN eT Al.

1982;Asikainen, 1981; Avisar & Kronfeld, 2009; Basu etal., 2015;Chkir, Guendouz, Zouari, Ammar, & Moulla, 2009; Cowart, 1980;Grabowski & Bem, 2012; Hakam, Choukri, Reyss, & Lferde, 2001;Hussain&Krishnaswami,1980;Ioannidou,Samaropoulos,Efstathiou,&Pashalidis,2011;Kronfeld,1974;Kronfeld&Adams,1974;Kronfeld,Godfrey-Smith,Johannessen,&Zentilli,2004;Kronfeld,Gradsztajn,&Yaniv, 1979;Kronfeld,Vogel,&Talma,1994; Lee,Choi,Cho, Lee,&Shin,2001;Reyes&Marques,2008;Tripathietal.,2013;Vogel,Talma,Heaton,&Kronfeld,1999)](Figure7).Theseculardisequilibriumchar-acteristic of oxide-associated U in Ediacara samples suggests thattheseFeoxideslikelyprecipitatedfromfluidswithinthepast~2mil-lionyears.Wecannotusetheseactivityratiostocalculatetheprecisetimingofalteration,givenuncertaintyregardingtheinitial(234U/238U) value of the precipitating fluid. However, even systems character-izedby relativelyhigh initial (234U/238U)values (e.g., (234U/238U)≥10)will,underclosed-systemconditions,reachsecularequilibriumwithin<5millionyears (Figure5).Therefore,evenconsideringawiderangeof potential initial (234U/238U) values (Figure5), the (234U/238U) dis-equilibriumcharacteristicoftheEdiacaraFeoxidesindicatesthatthesampledUwasmobilewithinthepastfewmillionyears,andrulesoutthepossibilitythattheoxidesrecordFecyclingandredoxconditionssynchronouswiththeca.560–550MadepositionandearlydiagenesisoftheEdiacaraMember.TheassociationofUwithFeoxidesmayre-flectcoprecipitationofFeoxidesandUphases,sorptionofUontopre-existingFeoxides,ortheoperationofbothmechanisms.Forinstance,the(234U/238U)signatureofpreexistingoxidescouldhavebeenresetthroughrecentalterationbygroundwater.However,thiswouldlikelyrequiredissolutionandreprecipitation,asmineralalterationinanopensystemtypically results inpreferential loss, rather thangainof 234U. Regardless,theEdiacara(234U/238U)datastillnecessitatethepresenceofextensivefluid-richoxicalteration,indicatingtheoxidescannotbeused for reconstructionof thepaleoenvironmentorearlydiagenetic

historyoftheEdiacaraMember.Therefore,themodern-daypresenceofFeoxidecoatingsonfossiliferousEdiacarasandstonebedscannotbeinvokedasevidenceforaterrestrialdepositionalenvironmentforand“synsedimentaryferruginization”oftheEdiacaraMember.There-cent(234U/238U)signatureofEdiacaraFeoxidesfurthercorroboratesfield-andoutcrop-scalesedimentologicalevidenceindicatingthatthepresent-daydistributionofFeoxideson fossiliferoussurfaces in theEdiacaraMember reflects recent, late-stage and likely groundwater-mediatedalterationandprecipitation.Bed-surfaceoxidesare irregu-larlyshapedandpatchilydistributed;oxidefrontscommonlysharplycross-cut fossil specimens and sedimentary features without beingassociatedwithanychange inpreservationormatrix sedimentology(Figures2and3).Thesesharpoxidationboundaries likelyreflectthepresenceofreactionfrontsassociatedwithrecentfluidflowin(post-deposition and post-exhumation) continental settings. Further, sed-imentological, geochemical, and geochronological data indicate thatsurfaceandsubsurfaceoxidationandironmobilizationwerecommon-placeintheFlindersregionthroughmuchoftheCenozoic(e.g.,Pillans,2005,2007).Moreover,evidenceforrecentironmobilization,linkedtofluctuatinggroundwaterconditions,alongintra-sequencehorizonsofCenozoicsedimentarysuccessionsinsouthernSouthAustralia(Milnesetal.,1985),suggeststhatgroundwater-mediatedprocessesarelike-wiseplausibledriversofchemicalalterationintheFlindersregion.TheTorrens Basin and thewestern piedmont of the Flinders—the areassurroundingNilpena—arecharacterizedbynotonlyacomplexanddy-namicinterplayofupliftandsubsidence(Quigley,Sandiford,&Cupper,2007;Quigley,Sandiford,Fifield,&Alimanovic,2007b)butalsoarel-ativelydeepandvariablewatertable (Williams&Polach,1971). It istherefore plausible that the (234U/238U) disequilibrium characteristicofEdiacaraFeoxidesnotonlyindicatesgroundwater-mediatedalter-ationbutalsothatthis(234U/238U)signaturerecordsmerelythemostrecentepisodeofaseriesofoxidationevents—reflectingthelonganddynamichistoryofpost-depositionalweatheringcharacteristicoftheSouthAustralianlandscape.

Falsification, onboth sedimentological andgeochemical grounds,ofthehypothesisthatsynsedimentaryferruginization indicatesapa-leosol setting for the Ediacara Biota provides additional support forindependent paleontological and sedimentological evidence for asubaqueous,marinehabitatfortheEdiacaraBiota.EdiacaraBiotafos-sil assemblagescontaingloballydistributed taxapreserved inawiderangeof facies (Tarhan,Hood,Droser,Gehling,&Briggs,2017;Xiaoetal., 2013). Moreover, morphological characterization of Ediacarataxa indicatesthatthey lackedanyobviousphysiologicaladaptationsforlifeonland(Antcliffe&Hancy,2013).ThefossiliferousfaciesoftheEdiacaraMemberarecharacterizedbystrongevidenceofstorm-andcurrent-mediatedtransportanddeposition.Fossiliferousbedsarecom-posedoftexturallyandcompositionallymature,thin(sub-mm-tocm-scale) and discretely packaged quartz-feldspar arenites, alongwhichfeatures such as symmetrical ripples and toolmarks are common toabundant (Gehling&Droser,2013;Tarhan,Droser,&Gehling,2015;Tarhan, Droser, etal., 2017). Moreover, Ediacara fossil assemblagesare also directly characterized by evidence for organism–current in-teractions, in the formofcurrent-aligned tetheredstalksand fronds,

F IGURE  7 δ238Uvaluesversus234U/238Uactivityratios(234U/238U)foranalyzedEdiacaraFeoxide-richsamples.Horizontaldashedblacklinedenotessecularequilibrium;verticaldashedgraylinedenotesaveragecrustalδ238Uvalue(cf.Tissot&Dauphas,2015).Wherenotvisible,errorbarssmallerthansymbols

1.6

1.5

1.4

1.3

1.2

1.1

1.0

0.9

0.8

(234 U

/238 U

)

secular equilibrium

–1.4 –1.2 –1.0 –0.8 –0.6 –0.4 –0.2 0.0δ238U (‰)

n = 12

     |  45TARHAN eT Al.

aligneddeformation featuresalongthemarginsof insituDickinsonia (Dickinsonia “lifts”), alignedholdfast removal anddragmarks (“mop”),and aligned matground textures (Evans, Droser, & Gehling, 2015;Tarhan,Droser,&Gehling,2014,2015;Tarhanetal.,2010).Therefore,thebalanceofevidenceisgreatlyonthesideofamarinedepositionalenvironment,withastrikingpaucityofdatawhichuniquelyindicateaterrestrialpaleoenvironmentfortheEdiacaraMember.

More fundamentally, however, a signature for secular disequi-librium indicates that Ediacara Fe oxides unequivocally recordopen-systembehavior.Therefore, as theseoxidesdonot constitutegeochemicallyclosedsystems,neitherthepresence,distributionnorchemicalsignatureof ironoxidescanbereliablyusedasapaleoen-vironmentalproxyfortheEdiacaraMember(contraRetallack,2013).Theunreliability of present-dayFeoxidedistribution as ameaning-fulproxyforEdiacaranprocessesalsoextendstothetaphonomyofEdiacarafossilassemblages;notonlythelikelyveryrecentorigin,butalso, more importantly, the open-system signature characteristic oftheseoxidesindicatethattheyshouldnotbeusedtomakeinferencesregardingtheearlydiageneticprocessesresponsible for fossilizationoftheEdiacaraBiota(contraGehling,1999).

Thestronglynegativeδ238Uvaluescharacteristicoftheexaminedsamplesareconsistentwithrecent,groundwater-mediatedoxidefor-mation intheEdiacaraMember.Asdiscussedabove,thesenegativevaluescouldbelinkedtoeitherUsorptiontoorcoprecipitationwithiron oxides. The fractionation during U sorption to oxides typicallyrangesfrom−0.2‰to−0.15‰(e.g.,Brenneckaetal.,2011;Jemison,Johnson, Shiel, & Lundstrom, 2016). Under average river waterδ238Uvalues (−0.34‰—equivalenttothecrustalaverageof−0.3‰;Andersenetal.,2016;Tissot&Dauphas,2015;Weyeretal.,2008),thefractionationduringUsorptiontooxideswouldresultinaδ238U valueofapproximately–0.5‰.Thereareδ238Uvaluesinironoxide-richlayersintheEdiacaraMemberthataremuchmorenegativethan−0.5‰.However,groundwatersoftenhavemorenegativeδ238U val-uesthanaveragesurfacewatersduetoremovalofisotopicallyheavyU(VI)viareductionduringgroundwatertransport(Basuetal.,2015).Theanomalously lightvaluesofEdiacara samplesmay therefore re-flectUincorporationintoironoxidesfromgroundwatercharacterizedby an evolved δ238Usignature.Recentexperimentalwork(Styloetal.,2015)suggeststhatU(VI)reductionbyferrousironcanresult intheformationofrelativelyinsolubleandisotopicallylightU(IV),incontrastto the fractionation expected during microbial U reduction. Givenevidence for iron oxidation reaction fronts (see above), isotopicallylightδ238Uvaluescould,alternatively,reflectferrousiron-mediatedUreductionnearredoxboundariesinthegroundwatersystem.Insum,thereareseveralpossibleinterpretationsoftheδ238Uvalues,buttheyareallconsistentwithUbeingderivedfromoxicgroundwatersduringalterationoftheEdiacaraMemberoverthepastfewmillionyears.

7  | CONCLUDING REMARKS

Thepaleoenvironmental settingof theEdiacaraMember (RawnsleyQuartzite,SouthAustralia)—theunit containing the richestEdiacara

Biota fossil assemblages (Earth’s oldest complex ecosystems)—hasbeenthesubjectofrecentdebate.Inspiteofstrongsedimentologicalandpaleoecologicalevidenceforamarinedepositionalenvironment,thepresenceofFeoxidecoatingsalongfossiliferoushorizonsoftheEdiacaraMemberhasbeenmootedasevidenceforterrestrial,“syn-sedimentaryferruginization,”andsyn-depositionalpaleosoldevelop-ment.Totestthehypothesisthattheseoxidesarelatediageneticinorigin,wemeasuredtheUisotope((234U/238U) and δ238U)signatureoffossil-associatedFeoxides.Wefoundthatoxide-associatedUphasesareatseculardisequilibrium((234U/238U)>1),indicatingEdiacarasurfi-cialoxidesprecipitatedfromgroundwatersveryrecently,likelywithinthepast~2myr.Thesegeochemicaldata corroborateoutcrop- andhand sample-scale sedimentological evidence for a late-stage dia-geneticoriginfortheseoxides.Moreover,anunequivocalsignal foropen-systembehaviorinEdiacaraMemberoxidesindicatesthatthemodern-daypresenceanddistributionofFeoxidescannotbeusedtoinferanythingconcerningthedepositionalorearlydiagenetichistoryoftheEdiacaraMember.Therefore,Feoxidedistributionlacksmerit,inthecaseoftheEdiacaraMember,asapaleoenvironmentalproxyand paleoenvironmental interpretations premised upon this featureshouldbediscounted.

Although sedimentological andpaleontological linesof evidenceforthepaleoenvironmentalsettingoftheEdiacaraMemberhaveal-readybeenpresentedandscrutinizedindetail,Uisotopesand,spe-cifically, (234U/238U) values provide a useful and powerful tool tointerrogatewhatremainsatopicofconsiderabledebate.ThestrengthoftheapplicationofUisotopesystematicstoEdiacaraironoxidesliesin the testability of the hypotheses it can be employed to address,andthequantitativeframeworkinwhichanalyticalresultscanbecon-sidered.Ourdiscovery that thepresenceofFeoxides, likemanyoftheotherputativepaleoenvironmentalproxiesofRetallack(2013),isapoorandenvironmentallynon-specificproxyfortheEdiacaraMemberhighlights the pitfalls of indiscriminately employing proxieswithoutpetrographicandgeochemicalverificationthatthephasesinquestionarebothauthigenicandsyn-Ediacaran inorigin.Besidesdirectly re-futingthefindingsofRetallack(2013),thisstudyservesasacaution-ary taleof theneedforcomprehensivescrutinyofsedimentologicalcontextinallpaleoenvironmentalwork.

ACKNOWLEDGMENTS

This researchwas supportedbyaNASAExobiologygrant (M.L.D.andL.G.T.), aNSF-ELTgrant (N.J.P.), theAlternativeEarthsNASAAstrobiology Institute (N.J.P. and X.W.), a NSF Earth SciencesPostdoctoral Fellowship (L.G.T.), a National Geographic grant(M.L.D.), an Australian Research Council Discovery grant (J.G.G.),and theAmericanPhilosophical Society Lewis andClark Fund forExploration andFieldResearch inAstrobiology (L.G.T.).We thankRossandJaneFargherforaccesstotheNationalHeritageEdiacarafossil siteNilpenaon their property, acknowledging that this landlieswithintheAdnyamathanhaTraditionalLands,andM.P.Dzaugis,M.E. Dzaugis, S. Evans, C. Hall, and D. Rice for assistance withfieldwork.

46  |     TARHAN eT Al.

REFERENCES

Abdul-Hadi, A., Alhassanieh, O., & Ghafar, M. (2001). Disequilibrium ofuranium isotopes in someSyriangroundwater.Applied Radiation and Isotopes,55,109–113.

Algeo, T. J., Marenco, P. J., & Saltzman, M. R. (2016). Co-evolution ofoceans,climate,andthebiosphereduringthe‘OrdovicianRevolution’:Areview.Palaeogeography, Palaeoclimatology, Palaeoecology,458,1–11.

Anand,R.R.(2005).Weatheringhistory,landscapeevolutionandimplica-tionsforexploration. InRRAnand&P.Broekert (Eds.),Regolith land-scape evolution across Australia(pp.2–40).Perth:CooperativeResearchCentreforLandscapeEnvironmentsandMineralExploration.

Andersen,M.B.,Vance,D.,Morford,J.L.,Bura-Nakic,E.,Breitenbach,S.F.M.,&Och,L.(2016).ClosinginonthemarineU-238/U-235budget.Chemical Geology,420,11–22.

Andrews,J.N.,&Kay,R.L.F.(1982).U-234/U-238activityratiosofdis-solveduranium ingroundwaters fromaJurassic limestoneaquifer inEngland.Earth and Planetary Science Letters,57,139–151.

Antcliffe,J.B.,&Hancy,A.(2013).Criticalquestionsaboutearlycharacteracquisition-CommentonRetallack2012:“SomeEdiacaranfossilslivedonland”.Evolution & Development,15,225–227.

Asikainen,M.(1981).Stateofdisequilibriumbetween238U,234U,226Ra and 222Rningroundwaterfrombedrock.Geochimica et Cosmochimica Acta,45,201–206.

Avisar,D.,&Kronfeld,J.(2009).UraniumdisequilibriumasahydrologicalaidinstudyingthesalinizationprocessesintheSoutheasternCoastalPlainofIsrael.Applied Radiation and Isotopes,67,220–226.

Basu,A., Brown, S. T., Christensen, J. N., DePaolo, D. J., Reimus, P.W.,Heikoop, J. M., …Maher, K. (2015). Isotopic and geochemical trac-ersforU(VI) reductionandUmobilityatan insiturecoveryUmine.Environmental Science and Technology,49,5939–5947.

Beraldi-Campesi,H.(2013).Earlylifeonlandandthefirstterrestrialeco-systems.Ecological Processes,2,1–17.

Beraldi-Campesi,H.,&Retallack,G.J.(2016).TerrestrialecosystemsinthePrecambrian. In B.Weber, B. Büdel& J. Belnap (Eds.),Biological soil crusts: An organizing principle in drylands(Vol.226,pp.37–54).EcologicalStudies(AnalysisandSynthesis).Cham,Switzerland:Springer.

Bierman,P.R.,Albrecht,A.,Bothner,M.,Brown,E.,Bullen,T.,Gray,L.,&Turpin,L.(1998).Weathering,erosionandsedimentation.InC.Kendall&J.J.Mcdonnel(Eds.),Isotope tracers in catchment hydrology(pp.647–678).Amsterdam:Elsevier.

Bierman, P. R., & Caffee,M. (2002). Cosmogenic exposure and erosionhistoryofAustralianbedrock landforms.Geological Society of America Bulletin,114,787–803.

Bierman,P.R.,Caffee,M.W.,Davis,P.T.,Marsella,K.,Pavich,M.,Colgan,P., … Larsen, J. (2002). Rates and timing of earth surface processesfrom in situ-produced cosmogenic Be-10.Reviews in Mineralogy and Geochemistry,50,147–205.

Bird,M.I.,&Chivas,A.R.(1988).OxygenisotopedatingoftheAustralianregolith.Nature,331,513–516.

Bourdon, B., Turner, S., Henderson, G. M., & Lundstrom, C. C. (2003).Introduction to U-series geochemistry. Reviews in Mineralogy and Geochemistry,52,1–21.

Brennecka, G. A.,Wasylenki, L. E., Bargar, J. R.,Weyer, S., &Anbar, A.D. (2011). Uranium isotope fractionation during adsorption to Mn-oxyhydroxides.Environmental Science and Technology,45,1370–1375.

Broecker,W.S.,&Thurber,D.L. (1965).Uranium-seriesdatingofcoralsandoolites fromBahamanandFloridaKey limestones.Science,149,58–60.

Callen,R.A.(1977).LateCainozoiclandscapeofpartofnortheasternSouthAustralia.Journal of the Geological Society of Australia,24,151–169.

Chabaux, F., Bourdon, B., & Riotte, J. (2008). U-series geochemistry inweatheringprofiles,riverwatersandlakes.InS.Krishnaswami&J.K.Cochran(Eds.),Radioactivity in the environment(pp.49–104).NewYork:Elsevier.

Cheng,H.,Edwards,R.L.,Shen,C.-C.,Polyak,V.J.,Asmerom,Y.,Woodhead,J., … Alexander, E. C. (2013). Improvements in 230Th dating, 230Thand 234Uhalf-lifevalues, andU-Th isotopicmeasurements bymulti-collector inductively coupled plasma mass spectrometry. Earth and Planetary Science Letters,371,82–91.

Chkir,N.,Guendouz,A.,Zouari,K.,Ammar,F.H.,&Moulla,A.S. (2009).Uranium isotopes in groundwater from the continental intercalaireaquifer in Algerian Tunisian Sahara (Northern Africa). Journal of Environmental Radioactivity,100,649–656.

Cole,D.B.,Zhang,S.,&Planavsky,N.J.(2017).Anewestimateofdetritalredox-sensitivemetalconcentrationsandvariabilityinfluxestomarinesediments.Geochimica et Cosmochimica Acta,215,337–353.

Counts,J.W.,Rarity,F.,Ainsworth,R.B.,Amos,K.J.,Lane,T.,Moron,S.,…Nanson,R.(2016).SedimentologicalinterpretationofanEdiacarandelta: Bonney Sandstone, SouthAustralia.Australian Journal of Earth Sciences,63,257–273.

Cowart, J. B. (1980).The relationship of uranium isotopes to oxidation-reduction in the Edwards carbonate aquifer of Texas. Earth and Planetary Science Letters,48,277–283.

DeCorte,B.P. (2016).Uranium isotope ratios in modern and Precambrian soils.Appleton,WI:LawrenceUniversity.

DePaolo,D.J.,Lee,V.E.,Christensen,J.N.,&Maher,K.(2012).Uraniumcomminution ages: Sediment transport and deposition time scales.Comptes Rendus Geoscience,344,678–687.

DePaolo, D. J., Maher, K., Christensen, J. N., & McManus, J. (2006).Sediment transport time measured with U-series isotopes: ResultsfromODPNorthAtlantic drift site 984. Earth and Planetary Science Letters,248,394–410.

Dosseto,A.,Bourdon,B.,&Turner,S.P. (2008).Uranium-series isotopesinrivermaterials:Insightsintothetimescalesoferosionandsedimenttransport.Earth and Planetary Science Letters,265,1–17.

Droser,M.L.,Gehling,J.G.,Dzaugis,M.E.,Kennedy,M.J.,Rice,D.,&Allen,M.F.(2014).AnewEdiacaranfossilwithanovelsedimentdisplacivelifehabit.Journal of Paleontology,88,145–151.

Evans, S. D., Droser, M. L., & Gehling, J. G. (2015). Dickinsonia lift-off: Evidence of current derived morphologies. Palaeogeography, Palaeoclimatology, Palaeoecology,434,28–33.

Faure,G.(1986).Principles of isotope geology.NewYork:Wiley.Gehling,J.G. (1999).Microbialmats in terminalProterozoicsiliciclastics:

Ediacarandeathmasks.Palaios,14,40–57.Gehling,J.G.(2000).Environmentalinterpretationandasequencestrati-

graphic framework for the terminal Proterozoic Ediacara MemberwithintheRawnsleyQuartzite,SouthAustralia.Precambrian Research,100,65–95.

Gehling,J.G.,&Droser,M.L.(2013).HowwelldofossilassemblagesoftheEdiacaraBiotatelltime?Geology,41,447–450.

Grabowski,P.,&Bem,H.(2012).Uraniumisotopesasatracerofgroundwa-tertransportstudies.Journal of Radioanalytical and Nuclear Chemistry,292,1043–1048.

Hakam,O.K.,Choukri,A.,Reyss,J.L.,&Lferde,M.(2001).DeterminationandcomparisonofuraniumandradiumisotopesactivitiesandactivityratiosinsamplesfromsomenaturalwatersourcesinMorocco.Journal of Environmental Radioactivity,57,175–189.

Hansen,R.O.,&Stout,P.R.(1968).Isotopicdistributionsofuraniumandthoriuminsoils.Soil Science,105,44–50.

Hussain,N.,&Krishnaswami,S.(1980).U-238seriesradioactivedisequi-librium in groundwaters: Implications to the origin of excess U-234andfateofreactivepollutants.Geochimica et Cosmochimica Acta,44,1287–1291.

Ioannidou, A., Samaropoulos, I., Efstathiou, M., & Pashalidis, I. (2011).Uranium in ground water samples of Northern Greece. Journal of Radioanalytical and Nuclear Chemistry,289,551–555.

Jemison,N.E.,Johnson,T.M.,Shiel,A.E.,&Lundstrom,C.C.(2016).UraniumisotopicfractionationinducedbyU(VI)adsorptionontocommonaquiferminerals.Environmental Science and Technology,50,12232–12240.

     |  47TARHAN eT Al.

Jenkins,R.J.F.,Ford,C.H.,&Gehling,J.G.(1983).TheEdiacaraMemberof theRawnsleyQuartzite - the contextof theEdiacara assemblage(latePrecambrian,FlindersRanges).Journal of the Geological Society of Australia,30,101–119.

Kronfeld,J.(1974).UraniumdepositionandTh-234alpha-recoil:Anexpla-nationforextremeU-234/U-238fractionationwithintheTrinityaqui-fer.Earth and Planetary Science Letters,21,327–330.

Kronfeld, J.,Godfrey-Smith,D. I., Johannessen,D.,&Zentilli,M. (2004).Uranium series isotopes in theAvonValley, Nova Scotia. Journal of Environmental Radioactivity,73,335–352.

Kronfeld,J.,Gradsztajn,E.,&Yaniv,A.(1979).AflowpatterndeducedfromuraniumdisequilibriumstudiesfortheCenomaniancarbonateaquiferoftheBeershevaregion,Israel.Journal of Hydrology,44,305–310.

Kronfeld,J.,&Adams,JAS(1974).Hydrologicinvestigationsoftheground-watersofcentralTexasusingU-234/U-238disequilibrium.Journal of Hydrology,22,77–88.

Kronfeld,J.,Vogel,J.C.,&Talma,A.S. (1994).Anewexplanationforex-treme234U/238Udisequilibriainadolomiticaquifer.Earth and Planetary Science Letters,123,81–93.

Kump,L.R.(2014).HypothesizedlinkbetweenNeoproterozoicgreeningofthelandsurfaceandtheestablishmentofanoxygen-richatmosphere.Proceedings of the National Academy of Sciences of the United States of America,111,14062–14065.

Langmuir,D. (1978).Uraniumsolution-mineralequilibriaat lowtempera-tures with applications to sedimentary ore deposits. Geochimica et Cosmochimica Acta,42,547–569.

Lee, M. H., Choi, G. S., Cho, Y. H., Lee, C. W., & Shin, H. S. (2001).Concentrationsandactivityratiosofuraniumisotopesinthegroundwa-teroftheOkchunBeltinKorea.Journal of Environmental Radioactivity,57,105–116.

Maher, K., DePaolo, D. J., & Christensen, J. N. (2006). U-Sr isotopicspeedometer: Fluid flow and chemical weathering rates in aquifers.Geochimica et Cosmochimica Acta,70,4417–4435.

Mathieu,D.,Bernat,M.,&Nahon,D.(1995).Short-livedUandThisotopedistribution in a tropical laterite derived from granite (Pitinga riverbasin,Amazonia,Brazil):Applicationtoassessmentofweatheringrate.Earth and Planetary Science Letters,136,703–714.

Milnes,A. R., Bourman, R. P., &Northcote, K.H. (1985). Field relation-shipsofferricretesandweatheredzonesinsouthernSouthAustralia:Acontributionto‘laterite’studiesinAustralia.Australian Journal of Soil Research,23,441–465.

Pandey,S.K.,&Sharma,M.(2017).EnigmaticEdiacaranmegascopicbed-dingplanestructuresontheSoniaSandstone,JodhpurGroup,MarwarSupergroup, India: Seaweed or problematica?Geological Journal, 52,784–807.

Pillans, B. (2005). Geochronology of the Australian regolith. Regolith-Landscape Evolution Across Australia,pp.41–61.

Pillans, B. (2007). Pre-Quaternary landscape inheritance in Australia.Journal of Quaternary Science,22,439–447.

Quigley, M. C., Sandiford, M., & Cupper, M. L. (2007). Distinguishingtectonic from climatic controls on range-front sedimentation. Basin Research,19,491–505.

Quigley,M.,Sandiford,M.,Fifield,K.,&Alimanovic,A. (2007a).BedrockerosionandreliefproductioninthenorthernFlindersRanges,Australia.Earth Surface Processes and Landforms,32,929–944.

Quigley,M.,Sandiford,M.,Fifield,L.K.,&Alimanovic,A.(2007b).Landscaperesponsestointraplatetectonism:Quantitativeconstraintsfrom10Benuclideabundances.Earth and Planetary Science Letters,261,120–133.

Retallack, G. J. (2012). Were Ediacaran siliciclastics of South Australiacoastalordeepmarine?Sedimentology,59,1208–1236.

Retallack,G.J.(2013).Ediacaranlifeonland.Nature,493,89–92.Retallack,G.J.(2014a).AffirminglifeaquaticfortheEdiacarabiotainChina

andAustraliaComment.Geology,42,E325.Retallack,G.J. (2014b).Howwell do fossil assemblagesof theEdiacara

Biotatelltime?Geology,42,E332.

Retallack,G.J.(2015).Commenton:“Dickinsonialiftoff:Evidenceofcurrentderivedmorphologies”byS.D.Evans,M.L.Droser,andJ.G.Gehling.Palaeogeography, Palaeoclimatology, Palaeoecology,434,28–33.

Retallack, G. J. (2016a). Ediacaran fossils in thin-section.Alcheringa, 40,583–600.

Retallack, G. J. (2016b). Field and laboratory tests for recognition ofEdiacaranpaleosols.Gondwana Research,36,107–123.

Reyes, E., & Marques, L. S. (2008). Uranium series disequilibria inground waters from a fractured bedrock aquifer (MorungabaGranitoids-Southern Brazil): Implications to the hydrochemicalbehaviorof dissolvedUandRa.Applied Radiation and Isotopes,66,1531–1542.

Rudnick,R.L.,&Gao,S.(2014).Compositionofthecontinentalcrust.InH.D.Holland&K.K.Turekian(Eds.),Treatise on geochemistry(pp.1–51).Oxford:Elsevier.

Shiel,A.E., Laubach,P.G.,Johnson,T.M., Lundstrom,C.C., Long,P.E.,&Williams, K. H. (2013). Nomeasurable changes in 238U/235U due to desorption-adsorption of U(VI) from groundwater at the Rifle,Colorado, Integrated Field Research Challenge site. Environmental Science and Technology,47,2535–2541.

Stylo,M.,Neubert,N.,Wang,Y.H.,Monga,N.,Romaniello,S.J.,Weyer,S., &Bernier-Latmani, R. (2015).Uranium isotopes fingerprint bioticreduction.Proceedings of the National Academy of Sciences of the United States of America,112,5619–5624.

Tarhan,L.G.,Droser,M.L.,&Gehling,J.G.(2010).TaphonomiccontrolsonEdiacarandiversity:Uncoveringtheholdfastoriginofmorphologicallyvariableenigmaticstructures.Palaios,25,823–830.

Tarhan,L.G.,Droser,M.L.,&Gehling,J.G.(2014).Puckered,wovenandgrooved:TheimportanceofsubstrateforEdiacarapaleoecology,paleo-environmentandtaphonomy.In:Annual Meeting of the Palaeontological Association,Leeds,U.K.,pp.48.

Tarhan,L.G.,Droser,M.L.,&Gehling,J.G.(2015).Depositionalandpres-ervationalenvironmentsoftheEdiacaraMember,RawnsleyQuartzite(South Australia): Assessment of paleoenvironmental proxies andthe timing of ‘ferruginization’. Palaeogeography, Palaeoclimatology, Palaeoecology,434,4–13.

Tarhan, L. G., Droser, M. L., Gehling, J. G., & Dzaugis, M. P. (2015).Taphonomy and morphology of the Ediacara form genus Aspidella. Precambrian Research,257,124–136.

Tarhan,L.G.,Droser,M.L.,Gehling,J.G.,&Dzaugis,M.P.(2017).Microbialmat sandwiches and other anactualistic sedimentary features of theEdiacaraMember (RawnsleyQuartzite, SouthAustralia): Implicationsfor interpretation of the Ediacaran sedimentary record. Palaios, 32,181–194.

Tarhan,L.G.,Hood,A.V.S.,Droser,M.L.,Gehling,J.G.,&Briggs,D.E.G.(2016).Exceptionalpreservationofsoft-bodiedEdiacaraBiotapro-motedbysilica-richoceans.Geology,44,951–954.

Tarhan,L.G.,Hood,A.V.,Droser,M.L.,Gehling,J.G.,&Briggs,D.E.G.(2017). Exceptional preservation of soft-bodied Ediacara Biotapromotedbysilica-richoceans—Reply.Geology,45,E408.

Tissot,F.L.H.,&Dauphas,N.(2015).Uraniumisotopiccompositionsofthecrustandocean:Agecorrections,Ubudgetandglobalextentofmod-ernanoxia.Geochimica et Cosmochimica Acta,167,113–143.

Tripathi,R.M.,Sahoo,S.K.,Mohapatra,S.,Lenka,P.,Dubey,J.S.,&Puranik,V.D.(2013).Studyofuraniumisotopiccompositioningroundwateranddeviationfromsecularequilibriumcondition.Journal of Radioanalytical and Nuclear Chemistry,295,1195–1200.

Twidale, C. R. (2000). EarlyMesozoic (?Triassic) landscapes inAustralia:Evidence, argument, and implications. The Journal of Geology, 108,537–552.

Vogel,J.C.,Talma,A.S.,Heaton,T.H.E.,&Kronfeld,J.(1999).EvaluatingtherateofmigrationofanuraniumdepositionfrontwithintheUitenhageAquifer.Journal of Geochemical Exploration,66,269–276.

Waggoner,B. (2003).TheEdiacaranbiotas inspaceandtime. Integrative and Comparative Biology,43,104–113.

48  |     TARHAN eT Al.

Wang, X., Johnson, T. M., & Lundstrom, C. C. (2015). Isotope fraction-ation during oxidation of tetravalent uranium by dissolved oxygen.Geochimica et Cosmochimica Acta,150,160–170.

Weyer,S.,Anbar,A.D.,Gerdes,A.,Gordon,G.W.,Algeo,T.J.,&Boyle,E.A.(2008).Naturalfractionationof238U/235U. Geochimica et Cosmochimica Acta,72,345–359.

Williams,G. E.,&Polach,H.A. (1971). Radiocarbondating of arid-zonecalcareous paleosols. Geological Society of America Bulletin, 82,3069–3086.

Xiao,S.H.,Droser,M.,Gehling,J.G.,Hughes, I.V.,Wan,B.,Chen,Z.,&Yuan,X.L.(2013).AffirminglifeaquaticfortheEdiacarabiotainChinaandAustralia.Geology,41,1095–1098.

Xiao,S.H.,&Knauth, L.P. (2013).Fossils come in to land.Nature,493,28–29.

How to cite this article:TarhanLG,PlanavskyNJ,WangX,BellefroidEJ,DroserML,GehlingJG.Thelate-stage“ferruginization”oftheEdiacaraMember(RawnsleyQuartzite,SouthAustralia):Insightsfromuraniumisotopes.Geobiology. 2018;16:35–48. https://doi.org/10.1111/gbi.12262