dynamic redox conditions control late ediacaran metazoan ... · of the nama group, the northern...

De

RRAa

b

c

d

e

f

a

ARR3AA

KONBMEE

h0

Precambrian Research 261 (2015) 252–271

Contents lists available at ScienceDirect

Precambrian Research

jo ur nal home p ag e: www.elsev ier .com/ locate /precamres

ynamic redox conditions control late Ediacaran metazoancosystems in the Nama Group, Namibia

.A. Wooda,∗, S.W. Poultonb, A.R. Pravec, K.-H. Hoffmannd, M.O. Clarksona,1,

. Guilbaudb, J.W. Lynea, R. Tostevine, F. Bowyera, A.M. Pennya,

. Curtisa, S.A. Kasemannf

School of Geosciences, University of Edinburgh, West Mains Road, Edinburgh EH9 3JW, UKSchool of Earth and Environment, University of Leeds, Leeds LS2 9JT, UKDepartment of Earth Sciences, University of St Andrews, St Andrews KY16 9AL, UKGeological Survey of Namibia, Private Bag 13297, Windhoek, NamibiaDepartment of Earth Sciences, University College London, Gower Street, London WC1E 6BT, UKDepartment of Geosciences, University of Bremen, P.O. Box 330 440, 28334 Bremen, Germany

r t i c l e i n f o

rticle history:eceived 21 July 2014eceived in revised form0 December 2014ccepted 4 February 2015vailable online 12 February 2015

eywords:xygenationeoproterozoiciomineralisationetazoans

diacarancosystems

a b s t r a c t

The first appearance of skeletal metazoans in the late Ediacaran (∼550 million years ago; Ma) has beenlinked to the widespread development of oxygenated oceanic conditions, but a precise spatial and tem-poral reconstruction of their evolution has not been resolved. Here we consider the evolution of oceanchemistry from ∼550 to ∼541 Ma across shelf-to-basin transects in the Zaris and Witputs Sub-Basinsof the Nama Group, Namibia. New carbon isotope data capture the final stages of the Shuram/Wonokadeep negative C-isotope excursion, and these are complemented with a reconstruction of water columnredox dynamics utilising Fe–S–C systematics and the distribution of skeletal and soft-bodied metazoans.Combined, these inter-basinal datasets provide insight into the potential role of ocean redox chemistryduring this pivotal interval of major biological innovation.

The strongly negative �13C values in the lower parts of the sections reflect both a secular, global changein the C-isotopic composition of Ediacaran seawater, as well as the influence of ‘local’ basinal effectsas shown by the most negative �13C values occurring in the transition from distal to proximal rampsettings. Critical, though, is that the transition to positive �13C values postdates the appearance of calcifiedmetazoans, indicating that the onset of biomineralization did not occur under post-excursion conditions.

Significantly, we find that anoxic and ferruginous deeper water column conditions were prevalentduring and after the transition to positive �13C that marks the end of the Shuram/Wonoka excursion.Thus, if the C isotope trend reflects the transition to global-scale oxygenation in the aftermath of theoxidation of a large-scale, isotopically light organic carbon pool, it was not sufficient to fully oxygenatethe deep ocean.

Both sub-basins reveal highly dynamic redox structures, where shallow, inner ramp settings expe-rienced transient oxygenation. Anoxic conditions were caused either by episodic upwelling of deeperanoxic waters or higher rates of productivity. These settings supported short-lived and monospecificskeletal metazoan communities. By contrast, microbial (thrombolite) reefs, found in deeper inner- andmid-ramp settings, supported more biodiverse communities with complex ecologies and large skeletalmetazoans. These long-lived reef communities, as well as Ediacaran soft-bodied biotas, are found par-
ticularly within transgressive systems, where oxygenation was persistent. We suggest that a mid-rampposition enabled physical ventilation mechanisms for shallow water column oxygenation to operate dur-ing flooding and transgressive sea-level rise. Our data support a prominent role for oxygen, and for stable oxygenated conditions in particular, in controlling both the distribution and ecology of Ediacaran skeletal metazoan communities.
∗ Corresponding author. Tel.: +44 131 650 6014; fax: +44 131 668 3184.E-mail address: [email protected] (R.A. Wood).

1 Current address: Department of Chemistry, University of Otago, Union Street, PO Box

ttp://dx.doi.org/10.1016/j.precamres.2015.02.004301-9268/Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.

Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.

56, Dunedin 9016, New Zealand.

dx.doi.org/10.1016/j.precamres.2015.02.004

http://www.sciencedirect.com/science/journal/03019268

http://www.elsevier.com/locate/precamres

http://crossmark.crossref.org/dialog/?doi=10.1016/j.precamres.2015.02.004&domain=pdf

mailto:[email protected]

dx.doi.org/10.1016/j.precamres.2015.02.004

an Res

1

stenCyfio(zrPiwsbie

gSMotfl2Be(lrbJi(setu2st2g(dfswe2dwe2o

maisw

R.A. Wood et al. / Precambri

. Introduction

The Ediacaran (635–541 Ma) witnessed a profound biologicalhift from a world with minimal multicellular diversity and evolu-ionary stasis, to one of new body plans, skeletal types and novelcologies, culminating in the appearance of modern-style commu-ities by the early Cambrian (Butterfield, 2007; Erwin et al., 2011).omplex, multicellular, body fossils appeared about 575 millionears ago (Ma), represented by the soft-bodied Ediacaran biotaound initially in deep waters and later in shallow-marine sett-ngs (Martin et al., 2000; Narbonne and Gehling, 2003). Embryosf possible metazoans are known from at least ∼600–580 MaXiao et al., 1998), while the first unequivocal calcified meta-oans were present by ∼550 Ma (Germs, 1972). The subsequentapid diversification of metazoans with hard parts around therecambrian–Cambrian boundary (541 Ma) marks a step changen biodiversity, the complexity of marine ecosystems, and in the

orkings of the global carbon cycle. Metazoans demand oxygen toupport aerobic metabolisms and skeletal hard-parts, and so it haseen presumed that a rise in oxygen, perhaps incrementally, facil-

tated the evolution of this complexity (Fike et al., 2006; Canfieldt al., 2007, 2008; McFadden et al., 2008; Scott et al., 2008).

Chemical tracers reveal a profound change in major bio-eochemical cycles during the Ediacaran, such as the globalhuram/Wonoka deep negative C-isotope excursion (Burns andatter, 1993). This has variously been interpreted as being due to

xidation of a substantial reservoir of organic carbon dissolved inhe deep ocean (Rothman et al., 2003; Fike et al., 2006), to a largeux of methane released from clathrates (Bjerrum and Canfield,011), or to diagenetic phenomena (Derry, 2010). The models ofristow and Kennedy (2008), however, suggest that there were notnough oxidants available for the model proposed by Fike et al.2006), and thus that the Shuram could not have represented aarge scale oxidation event. Indeed, the global response of oceanedox chemistry to rising oxygen levels through this period haseen shown to be complex (Fike et al., 2006; Canfield et al., 2008;ohnston et al., 2010, 2012b, 2013; Sperling et al., 2013a), includ-ng in South China the presence of metastable zones of euxinicanoxic and sulfidic) waters impinging on the continental shelf andandwiched within ferruginous [Fe(II)-enriched] deep waters (Lit al., 2010). Detailed reconstructions of ocean chemistry suggesthat a globally anoxic and ferruginous deep ocean state existedntil at least ∼580 Ma, and beyond in certain areas (Canfield et al.,008; Planavsky et al., 2011; Poulton and Canfield, 2011), whereasurface-water oxygenation is thought to be a near-continuous fea-ure throughout the latter half of the Ediacaran (Canfield et al.,008). Indeed some have argued that pervasive and persistent oxy-enation of the deep ocean did not occur until the later Palaeozoice.g. Canfield et al., 2008). There is also evidence to suggest that mid-epth euxinia (free-sulphide in the water column) may have been aeature (but possibly temporally restricted) along some continentalhelves at certain times in the Neoproterozoic, but these conditionsere sparse compared to the preceding Mesoproterozoic (Canfield

t al., 2008; Johnston et al., 2010; Li et al., 2010; Sperling et al.,013a). The evolution of large metazoans and skeletal hardpartsuring the Ediacaran period was therefore set within the frame-ork of major perturbations to the C, Fe and S cycles (Des Marais

t al., 1992; Logan et al., 1995; Rothman et al., 2003; Fike et al.,006; Canfield et al., 2007) which are all potentially linked to risingxygen.

Oxygen requirements in metazoans vary widely, and are deter-ined by size, metabolism, mobility, and the presence or absence of

circulatory system (Vaquer-Sunyer and Duarte, 2008). As a resultt has been proposed that metazoans may have been limited tomall size (<3 mm), thin body plans, and low diversity communitiesith simple foodwebs by the relatively low levels of oxygen of the

earch 261 (2015) 252–271 253

Proterozoic, with the explosion of larger and ecologically diverseorganisms in the late Ediacaran and Cambrian related, in part, toincreasing oxygen levels (e.g. Cloud, 1968; Runnegar, 1982). Exper-imental work has also shown that the oxygen levels necessary tosupport small, primitive metazoans such as sponges (Porifera) arefar lower (Mills et al., 2014) than those required for large, active,and ecologically important animals such as carnivorous predators(Sperling et al., 2013b; Knoll and Sperling, 2014).

In modern marine environments benthic diversity and biomassdecreases with bottom-water oxygen levels, as does individual sizeand abundance until the skeletal macrobenthos is excluded (e.g.Rhoads and Morse, 1971). Hypoxia is a major factor in structur-ing benthic communities: pelagic-benthic coupling is reduced asare other measures of ecological complexity such as communitysuccession. As oxygen levels decrease, large individuals and long-lived equilibrium species are eliminated, and populations shifttowards younger individuals, and smaller and short-lived speciesthat possess opportunistic life histories (Diaz and Rosenberg, 1995).The loss of skeletal biota occurs when oxygen drops below ∼0.10present atmospheric levels, a threshold therefore postulated to fuelthe Cambrian radiation (Rhoads and Morse, 1971). In addition, fluc-tuating and unpredictable redox conditions are deleterious to somemetazoans, so establishing stable oxygenated conditions (Johnstonet al., 2012), even if pO2 remained relatively low, may have beenjust as important as a rise in absolute pO2.

Some suggest, however, that the Ediacaran oxygen transitionwas a consequence, not a cause, of metazoan diversification, asthe pumping activity of poriferans and cnidarians could have aug-mented the removal of DOC and smaller phytoplankton fromthe water column, so substantially enhancing its oxygenation(Butterfield, 2009; Lenton et al., 2014). The evolution of Eumeta-zoa would have shifted oxygen demand to shelf sea sediments anddeeper waters, in turn reducing total phosphorus recycling fromsediments so reinforcing through positive feedback the shift to amore oxygenated ocean state (Lenton et al., 2014). These authorsargue that this could have facilitated the rise of more mobile andpredatory animals.

Such unresolved controversies persist as to the precise role ofoxygen in driving metazoan evolution because current records ofthis interval are limited both spatially and temporally, and withoutsuch, a detailed causal understanding of the relationships betweenanoxia, evolution and carbon cycle processes is not possible.

We focus on the first appearance of calcified metazoans and theirecology, as found in the ∼550 to ∼541 Ma Nama Group of southernNamibia, one of the best archives of late Ediacaran Earth history(Germs, 1972; Grotzinger and Miller, 2008). Here, soft-bodied Edi-acaran biotas and horizontal burrow systems are found preservedin marine siliciclastic rocks, and abundant calcified metazoansoccur in shallow to mid-ramp carbonate settings. This distinc-tive early calcified biota of stem group Eumetazoa, Cnidaria andBilateria, or Cnidaria (Wood, 2011) terminates globally at the Pre-cambrian/Cambrian boundary at ∼541 Ma (Amthor et al., 2003). Agreater diversity of body plans and skeletal organisations, includ-ing recognisable stem-group members of extant bilaterian phyla,is subsequently recorded in the first 10 Myr of the Cambrian (Knoll,2003).

To explore relationships between ocean chemistry, the firstappearance of calcified metazoans and their ecology, we presentnew carbon isotope data across two basin transects that capturethe final stages of the Shuram/Wonoka deep negative C-isotopeexcursion (∼550 to ∼548 Ma). We then consider Fe–S–C systemat-ics and the distribution of skeletal metazoans across these basins
to provide detailed insight into the potential role of ocean redoxchemistry during this pivotal interval of major biological innova-tion. We focus on a suite of clastic and carbonate strata whichrecord coeval sedimentation across a mixed clastic-carbonate ramp

254 R.A. Wood et al. / Precambrian Research 261 (2015) 252–271

Fig. 1. (A) Simplified map showing the geological setting, subdivisions, and sub-basins of the late Ediacaran to early Cambrian Nama Group of southern Namibia (modifiedf m); 2,A e pala

D

sbWrtupisssao

2

flMabts

bezMtuIa(Gt(∼s

atbf

rom Grotzinger and Miller, 2008), and sampled sites: 1, Zwartmodder (Omkyk Farrasab; 7, Grens; 8, Pinnacle Reefs (Swartkloofberg Farm); 9, Swartpunt. (B) Relativ

epositional interpretations from Saylor et al. (1995, 1998).

ystem (Grotzinger et al., 1995). Our samples are from the twoasins of the Nama Group, the northern Zaris Basin and southernitputs Basin, encompassing shallow, inner ramp, through mid-

amp, to deeper, outer-ramp settings (Fig. 1). The succession spanshe Kuibis Subgroup (∼550 to ∼547 Ma) in both basins, and theppermost Schwarzrand Subgroup (∼544 to ∼541 Ma) in the Wit-uts Basin, and records the first appearance of calcified metazoans

n the fossil record (Germs, 1972), as well as typical late Ediacaranoft-bodied biota. Using these data, in combination with associatededimentological and palaeobiological observations from our nineample sites, we have built a spatially resolved redox history thatllows exploration of metazoan ecological response to changes incean oxygenation.

. Geological setting

The Nama Group consists of a variety of fluvial to marine plat-orm carbonate and siliciclastic rocks ranging from upper shore-ine/tidal flats to below-wave-base lower shoreface (Grotzinger and

iller, 2008; Saylor et al., 1995, 1998; Germs, 1995) that forms 3000 m thick foreland basin fill (Gresse and Germs, 1993). Theasin is divided by a palaeohigh, the Osis Arch, separating it intohe Zaris Sub-Basin in the north and the Witputs Sub-Basin in theouth (Fig. 1).

The age of the base of the Nama Group is poorly constrained,ut is probably between 550 and 553 Ma (Ries et al., 2010). How-ver, ash beds in the overlying units (Fig. 2) provide robust U–Pbircon age constraints ranging from 548.8 ± 1 Ma in the Hooglandember of the Kuibis Subgroup (Grotzinger et al., 1995; revised

o 547.32 ± 0.31 Ma by Schmitz, 2012) to 538.18 ± 1.24 Ma in thepper part of the Schwarzrand Subgroup (Grotzinger et al., 1995).

n the Witputs Sub-Basin the Ediacaran–Cambrian (E-C) bound-ry is marked by a regionally extensive erosional unconformityGerms, 1983; Saylor and Grotzinger, 1996; Narbonne et al., 1997;rotzinger et al., 1995) overlain by an incised-valley fill sequence

hat contains the earliest Cambrian trace fossil Treptichnus pedumWilson et al., 2012). Therefore, the Nama Group section spans10 Myr and extends to beyond the Proterozoic–Cambrian tran-

ition (Saylor et al., 1998; Ries et al., 2010).Each Sub-Basin has its own distinct stratigraphic framework,

lthough key sequence boundaries and systems tracts can be iden-ified in both, with Transgressive Systems Tracts (TSTs) generallyeing dominated by siliciclastic rocks whereas various carbonateacies distinguish Highstand Systems Tracts (HSTs; Saylor et al.,

Omkyk (Omkyk Farm); 3, Zebra River; 4, Dreidornvlagte; 5, Brak (Ababis Farm); 6,eodepth position of sites within the northern Zaris and southern Witputs Basins.

1995, 1998;Fig. 2). In general, the Kuibis Subgroup thins towardsand completely disappears over the Osis Ridge (Germs, 1983, 1995).The basal unit in both sub-basins is the siliciclastic Kanies Mem-ber that is overlain by variably developed mixed siliciclastic- tocarbonate-dominated units (Grotzinger and Miller, 2008; Sayloret al., 1995, 1998; Germs, 1995). In the Zaris Sub-Basin, carbonateunits forming 10–20 m-scale upward-coarsening, mid-inner rampto shoreface cycles typify the Lower and Upper Omkyk Members,defining units OS1 and OS2 respectively (Fig. 2). The overlyingHoogland Members primarily of heterolithic interbeds that recordtransgressive–regressive depositional successions in mostly mid-inner ramp to middle shoreface settings; these rocks have beenplaced in a sequence stratigraphic framework based on the recog-nition of two unconformity-bounded sequences, K1 and K2 (Sayloret al., 1995, 1998;Fig. 2).

In the Witputs Sub-Basin, above the Kanies Member is thecarbonate-dominated HST of the Mara Member itself overlain bythe TST of the Kliphoek Member, marked by upper-shoreface totide-dominated shoreline sandstones (Saylor et al., 1995). Theoverlying Mooifontein Member is a thin-bedded shallow-marinelimestone that exhibits a relatively uniform constant thickness(30–40 m) across the Witputs Sub-Basin but does pinch out nearthe Osis Ridge. Its top is marked by an unconformity (S1) havingdeep canyon-like erosional relief that is filled with conglomerate(Saylor et al., 1998).

The Schwarzrand Group in the Witputs Sub-Basin reaches amaximum thickness of 1000 m. Its lower part consists of the mixedmostly shallow marine clastic and carbonate rocks of the Nudaus(which contains Sequence Boundary S2) through Nasep Member ofthe Urusis Formation (overlain by Sequence Boundary S3) whereasthe Huns (overlain by Sequence Boundary S4), Feldschuhhorn andSpitzkopf Members (overlain by Sequence Boundary S5) compris-ing its middle part are carbonate-dominated, including spectacularpinnacle reefs that initiated on the flooding surface at the top ofthe Huns Member and then became enveloped within siltstonesof the Feldschuhhorn Member. The top of the Schwarzrand Groupis defined by an upper conglomeratic unit, the Nomtsas Forma-tion, that infills incised valleys bracketed by ash beds dated at542.68 ± 1.245 Ma and 540.61 ± 0.67 Ma (U–Pb zircon; Grotzingeret al., 1995) (Fig. 2).

The Nama Group yields various metazoan assemblages. Large(up to 30 cm) horizontal burrow systems with spriete suggestingformation by bilaterian organisms have been found in the siltstonesof the Omkyk Member in the Zaris Sub-Basin, 100 m below an ash

R.A. Wood et al. / Precambrian Research 261 (2015) 252–271 255

WITP UTS SUB-BASIN

EDIA

CARA

N

CAM

BRIA

N

NOMTS AS FM

Spitzkop Member

Feldschuhh orn Member

Huns Member

Nasep Membe r

NUDAUS FM

Mooifontein Membe r

Kliph oek Member

Mara Member

Kanies Membe r U

RUSI

S FO

RMAT

ION

538.18 ± 1.24

540.61 ± 0.67 542.68 ± 1.245

ZFM

S5

S4

S3

S2

S1

K2

V

V

V

V

V

V

V

V

Ash Beds (U/ Pb dat es Ma)

KUIB

IS S

UBG

ROU

P

S

CHW

ARZR

AND

SU

BGRO

UP

ZARI

S FO

RMAT

ION

V Kanies Membe r

Omkyk Membe r

Hoo gland Membe r

V OS2

V 547.32 ± 0.65

ZARIS SUB-BASIN

V OS1

First Cloud ina First Ed iacaran biota Sequence Boundar ies

Thrombolite bu ild-up s

ZFM - Zaris Forma�on DA

BIS

FORM

ATIO

N

DF - Da bis Forma�on

Urikos Membe r

V

V

KUIB

IS S

UBG

ROU

P

SCH

WAR

ZRAN

D SU

BGRO

UP

DF

K1

beds oM

baaB(S((affetuBf(mptt

3

tmlsb

Fig. 2. Stratigraphy, sequence boundaries, and dated ash

odified from Saylor et al. (1995, 1998), Grotzinger et al. (2005), Hall et al. (2013).

ed dated at 547.32 ± 0.7 Ma (Macdonald et al., 2014). Burrowsre also known from the Huns, Nomtsas and Nasep Members andppear in abundance in the Spitzkopf Member in the Witputs Sub-asin some 60 m below the Ediacaran–Cambrian unconformityJensen et al., 2000). Ediacaran fossils such as Ernietta, Pteridinium,wartpuntia, and Rangea (e.g. Narbonne et al., 1997), vendotaenidsGerms et al., 1986), and the calcified metazoans Cloudina spp.Germs, 1983; Grant, 1990), Namacalathus (Grotzinger et al., 2000),nd locally Namapoikia (Wood et al., 2002), have been describedrom both the Zaris and Witputs Basins. The oldest Ediacara-typeossils occur in the Mara Member in the Witpuits Sub-Basin (Saylort al., 1995), and the youngest, including Swartpuntia, are found inhe Spitzkop Member some 60 m below the Ediacaran–Cambriannconformity in the more southerly portion of the Witputs Sub-asin (Narbonne et al., 1997). The oldest reported Cloudina is

rom the carbonate-dominated Mara Member in the Witputs BasinGerms, 1983). All calcified metazoans occur in either shallow

arine carbonate (limestone and dolomite) facies, and may beresent along bedding planes of mudstones or dolostones, in thino thickly bedded packstones and grainstones, or associated withhrombolite reefs.

. Methods

Nine sections have been studied (see Supplementary Informa-ion Table S1 for co-ordinates), from shallow, inner ramp, through

id-ramp, to deeper, outer-ramp settings (Fig. 1). Sections wereogged for metazoan occurrence and sampled within the sequencetratigraphic framework of Saylor et al. (1995, 1998) as establishedy Grotzinger et al. (2005) and Hall et al. (2013) (Fig. 2) which allows

f the Zaris and Witpus Sub-Basins Nama Group, Namibia.

for consideration of redox dynamics with changes in relative sea-level through time within shelf to basin transects. Samples werechosen that were devoid of mineralised veins or other visible alter-ation. All weathered surfaces were removed, and the whole rockwas either pulverised to a fine powder or sub-sampled via manualexcavation for geochemical analysis.

3.1. Carbon and oxygen isotope analyses

The validity of using carbonate rocks as proxy archives forthe �13C composition of ancient seawater requires an assessmentthat geochemical data are representative of primary ocean chem-istry. While the Nama Group carbonates have undergone pervasiverecrystallization, as well as localised early dolomitization, samplesgenerally show high quality preservation. For example, analysedsamples from Zebra River show low Mn/Sr ratios from partialleaches (97% of samples <1), and high [Sr] (294–5802 ppm; with anaverage of 1509 ± 986 ppm; 40% of samples >1500 ppm) (Fig. 3A),and relatively heavy �18O values (89% of samples >−10‰) (Fig. 3B).There is also no co-variance between FeT vs �18O and Mn/Sr data(Fig. 3). For these reasons the isotopic records presented in thisstudy are considered to track the secular trends of contemporaryseawater.

Powdered carbonate samples were analysed at 75 ◦C using a KielCarbonate III preparation device. The resulting CO2 was then ana-
lysed on a Thermo Electron Delta+ Advantage stable isotope ratiomass spectrometer. The standard deviation (n = 61) of a powderedcoral laboratory standard (COR1D, �13C = −0.648, �18O = −4.923)run on the same days as the study samples were ±0.04‰ for �13C


F ance.

fi

aV

3

tai1

3

ratPeemapFttc(((ubSucis1occtomo

ig. 3. Cross plots of A, FeT vs �18O and, B, FeT vs of Mn/Sr data showing no co-varigure legend, the reader is referred to the web version of the article.)

nd ±0.07‰ for �18O. All isotopic values are quoted relative toPDB.

.2. Total organic C (TOC)

Total organic C (TOC) was measured on an EA mass spectrome-er after carbonate removal (two? 10% HCl washes for 48 h). Excesscid was evaporated and samples dried at 70 ◦C. The reproducibil-ty of 13 internal standards run on the same day as samples was.1%.

.3. Fe-speciation

Fe-speciation is a widely utilised proxy for identifying theegional nature of water column redox conditions. The techniquellows identification of oxic and anoxic water column condi-ions (Raiswell and Canfield, 1998; Poulton and Raiswell, 2002;oulton and Canfield, 2011), where anoxia may be separated intouxinic (H2S-bearing) and ferruginous (Fe(II)-containing) (Poultont al., 2004a). However, the technique lacks the sensitivity toore precisely identify changes in the extent of ocean oxygen-

tion (i.e. where the water column was fully oxic or dysoxic). Theroxy gives a measure of highly reactive Fe to total Fe (FeHR/FeT).eHR refers to Fe minerals that are considered highly reactiveowards biological and abiological reduction under anoxic condi-ions (Canfield et al., 1992; Poulton et al., 2004b), and includesarbonate-associated Fe (Fecarb; e.g., ankerite and siderite), ferricoxyhydr)oxides (Feox; e.g., goethite and hematite), magnetite FeFemag) and Fe sulfide minerals (Fepy; e.g., makinawite and pyrite)Poulton and Canfield, 2005). Sediments may be enriched in FeHRnder anoxic marine conditions due to either the export of remo-ilized Fe(aq) from the oxic shelf (Anderson and Raiswell, 2004;evermann et al., 2008) or under more widespread anoxia, due topwelling of deep water Fe(II) (Poulton and Canfield, 2011). Pre-ipitation of this mobilised water column Fe is then potentiallynduced through a variety of processes, for example through Feulfide (pyrite) formation under euxinic conditions (Canfield et al.,996a,b; Raiswell and Canfield, 1998), or through precipitationf oxidised or partially oxidised Fe minerals under ferruginousonditions (e.g., Zegeye et al., 2012). These processes have theonsequence that FeHR/FeT ratios provide a particularly sensi-
ive means to determine whether a depositional setting was oxicr anoxic. Calibration in modern and ancient marine environ-ents suggests that FeHR/FeT < 0.22 provides a robust indication
f oxic conditions, while FeHR/FeT > 0.38 suggests deposition from

Blue, limestone; red, dolomite. (For interpretation of the references to color in this

an anoxic water column (Poulton and Canfield, 2011; Raiswell andCanfield, 1998; Poulton and Raiswell, 2002). Values between 0.22and 0.38, however, are somewhat equivocal, and care needs to betaken to determine if such values are a consequence of maskingof the additional anoxic water column flux of FeHR, either due torapid sedimentation (Raiswell and Canfield, 1998; Poulton et al.,2004a) or due to post-depositional transformation of unsulfidizedFeHR minerals to less reactive sheet silicate minerals (Cumminget al., 2013; Poulton and Raiswell, 2002; Poulton et al., 2010). Addi-tional examination of the Fepy/FeHR ratio has the unique advantagein that it allows the separation of anoxic settings into euxinic (sul-fidic) environments (Fepy/FeHR > 0.7–0.8) and non-sulfidic (Fe-rich;ferruginous) environments (Fepy/FeHR < 0.7) (Poulton and Canfield,2011; Poulton et al., 2004a).

Fe-speciation extractions were performed according to cal-ibrated extraction procedures (Poulton and Canfield, 2005),whereby FeCarb was extracted with Na-acetate at pH 4.5 and 50 ◦Cfor 48 h, FeOx was extracted via Na-dithionite at pH 4.8 for 2 h,and FeMag was extracted with ammonium oxalate for 6 h. Total Fe(FeT) extractions were performed on ashed samples (8 h at 550 ◦C)using HNO3–HF–HClO4. Boric acid was used to prevent the forma-tion of Al complexes, allowing total Al to be determined on thesame extractions. All Fe concentrations were measured via atomicabsorption spectrometry and replicate extractions gave a RSD of<5% for all steps. Pyrite Fe was determined stoichiometrically byweight from precipitated Ag2S after chromous chloride distillation(Canfield et al., 1986).

In turbidite settings, high sedimentation rates can cause a dilu-tion of FeHR (Raiswell and Canfield, 1998; Poulton et al., 2004b) andthus decrease the FeHR/FeT ratio of sediments deposited under ananoxic water column, giving a falsely oxic signal. However, noneof the clastic samples measured for FeHR are from strata that showevidence of turbiditic deposition. Some of the sandstone units inthe Spitzkof Member may have been rapidly emplaced, but ouranalyses are restricted to shales and siltstones in these sections. AtOmkyk, the lower part of the succession is dominated by carbon-ate turbidites, slumps and storm-beds, but all of these units yieldan unequivocal anoxic signal. Conversely, preferential trapping ofFeHR in inner shore or shallow marine environments, such as flood-plains, salt marshes, deltas and lagoons, can create local FeHR/FeTenrichments (Poulton and Raiswell, 2002). Although the shallow
ramp settings of the Nama Group were influenced by fluvial inputand experienced intermittent restriction, such FeHR enrichment isnot expected in the normal fully marine settings of the samplesanalysed.


Fig. 4. Distribution of all Fe-speciation data from the Nama Group. Yellow, clastic; blue, limestone; red, dolomite. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of the article.)

F ions ii to the

3

ta

ig. 5. Fepy/FeHR against FeHR/FeT to identify ferruginous vs euxinic anoxic conditnterpretation of the references to color in this figure legend, the reader is referred

.4. Fe-speciation in carbonates

To date, the Fe-speciation technique has been widely appliedo shales and mudrocks, for which it was originally calibrated,nd the technique has also been successfully applied to some

n samples with FeT > 0.5 wt%. Yellow, clastic; blue, limestone; red, dolomite. (For web version of the article.)

carbonate-rich sediments (e.g., März et al., 2008; Kendall et al.,
2010). Indeed, Clarkson et al. (2014) recently demonstrated theutility of Fe-speciation for application to appropriate carbonate-rich strata. The primary concerns for the use of Fe-speciation incarbonate-rich sediments are related to the decreased detrital


Fig. 6. Zwartmodder section: Stratigraphy, lithology, metazoan distribution, �13C, FeT, TOC, Fe speciation and Fe/Al data. Where FeT > 0.5 wt%, FeHR/FeT < 0.22 provides arobust indication of oxic conditions and FeHR/FeT > 0.38 suggests deposition from an anoxic water column. Fepy/FeHR ratio separates anoxic settings into euxinic (sulfidic)e ments (Fepy/FeHR < 0.7). Fe/Al is used to help interpret FeHR/FeT values between 0.22 and0 d <0.5 wt% TOC may indicate oxic depositional conditions, but this requires independentp

csadseiFieFtfpand2

ns<papocbwa

Table 1Summary of threshold values for the use of Fe speciation and Fe/Al as palaeo-redoxproxies in carbonate-rich sediments that have not experienced Fe addition duringdeep burial dolomitization.

Water column redox FeT (wt%) TOC (wt%) FeHR/FeT Fe/Al

Oxic >0.5 N/A <0.22 0.55 ± 0.11Probably oxica <0.5 <0.5 N/A N/AAnoxic >0.5 NA >0.38 >0.64

nvironments (Fepy/FeHR > 0.7–0.8) and non-sulfidic (Fe-rich; ferruginous) environ.38; when Fe/Al is greater than 0.66 samples are likely to be anoxic. <0.5 wt% FeT analaeontological evidence.

omponents, and hence low FeHR and FeT. This can increase theensitivity of the proxy to FeHR inputs that are unrelated to thenoxic water column formation of FeHR minerals, as might occururing early diagenesis. The impact of this process has been demon-trated to potentially be significant when FeT is <0.5 wt% (Clarksont al., 2014). At FeT >0.5 wt%, carbonate-rich sediments may behaven a manner which is consistent with shales, generally allowinge-speciation and Fe/Al to be used as water column redox prox-es for carbonate-rich sediments that have not been subject to FeHRnrichments during deep burial dolomitization. Thus, samples witheT < 0.5 wt% have not been used to identify water column redox inhe present study. Furthermore, there is no petrographic evidenceor deep burial dolomitization in our limestone or dolostone sam-les, nor has this been documented in other studies (Grotzingernd Miller, 2008; Ries et al., 2010). Early stage dolomitization doesot generally cause an increase in FeT due to the lack of Fe-richolomitizing fluids in shallow burial environments (Clarkson et al.,014).

These constraints for the application of Fe speciation to carbo-ates are summarised in Table 1. Clarkson et al. (2014) additionallyuggest that carbonate-rich rocks containing <0.5 wt% FeT and0.5 wt% TOC may indicate oxic depositional conditions (Table 1),articularly when both parameters are very low. We apply thispproach with caution in the present study, but also, whereverossible, we support this approach by considering the presencef bioturbation to provide independent evidence for oxic water
olumn deposition. It is important to note that wherever possi-le we sampled shales, siltstones, limestones and dolomites, oftenithin single depositional high-frequency cycles. Therefore, we are
ble to consider data from siliciclastic horizons alongside data for

Modified from Clarkson et al. (2014).TOC, total organic carbon; NA, not applicable (i.e., no threshold value required).

a In the absence of late stage dolomitzation or rapid rates of sedimentation.

carbonates. All FeT > 0.5 wt% carbonates behave consistently withshales and siltstones (Figs. 4 and 5), except where shales have suf-fered from late stage transformation of unsulfidized highly reactiveFe phases to poorly reactive Fe-rich clays, such that Fe-speciationdata can be applied to accurately identify anoxic depositional sett-ings in carbonates providing additional strong support for thevalidity of our approach. Thus, the carbonate analyses accuratelycapture the redox chemistry of the ocean during deposition of thesesediments.

3.5. Oxidative weathering

Outcrop samples in Namibia have potentially been affected byoxidative weathering. Many shale samples are olive grey with no
evidence of Fe oxide staining, suggesting minimal oxidation, butsome shale, siltstone and anoxic limestone samples show a per-vasive red colouration that might suggest oxidative weathering.Such a colouration might, however, be expected for limestones


Fig. 7. Omkyk section: stratigraphy, lithology, metazoan distribution, �13C, FeT, TOC, Fe speciation and Fe/Al data. Where FeT > 0.5 wt%, FeHR/FeT < 0.22 provides a robust indi-cation of oxic conditions and FeHR/FeT > 0.38 suggests deposition from an anoxic water column. Fepy/FeHR ratio separates anoxic settings into euxinic (sulfidic) environments( R < 0.7i ay ine

dtnihrdtn(enFanibu

3

g2(enib

Fepy/FeHR > 0.7–0.8) and non-sulfidic (Fe-rich; ferruginous) environments (Fepy/FeH

s greater than 0.66 samples are likely to be anoxic. <0.5 wt% FeT and <0.5 wt% TOC mvidence.

eposited under ferruginous conditions. We nevertheless considerhe possibility for oxidation in more detail. Firstly, it should beoted that pyrite and Fecarb minerals, such as siderite, are the dom-

nant FeHR minerals that may undergo oxidation. In both cases,owever, oxidation will not significantly impact FeHR/FeT or Fe/Alatios, and hence any interpretation of anoxia remains robust. Oxi-ation of siderite would transfer Fecarb to the FeOx pool, and hencehis would not affect identification of a ferruginous or euxinic sig-al. A robust euxinic signature is identified by Fepy/FeHR ratios >0.8Anderson and Raiswell, 2004; Poulton and Canfield, 2011; Poultont al., 2004a,b). If we consider an extreme and highly unlikely sce-ario, whereby there was no FeOx in the primary sample, and allepy has been oxidised to FeOx, with no oxidation of Fecarb to FeOx,pproximately 10% of the anoxic samples would give a euxinic sig-al. However, significant Fecarb (i.e. >20% of the FeHR fraction) occurs

n ∼57% of the anoxic samples, indicating that the rocks have noteen completely weathered, and hence this extreme scenario isnlikely.

.6. Fe/Al ratios

Calibration studies suggest that when Fe/Al in siliciclastics isreater than the Phanerozoic average of 0.53 ± 0.11 (Raiswell et al.,008) these samples are likely to be anoxic. Similarly, Clarkson et al.2014) derived a normal oxic Fe/Al ratio of 0.55 ± 0.11 for mod-
rn sediments, independent of carbonate content. The Fe/Al ratioormalises the FeT abundance in order to consider Fe-enrichments
ndependent of background (clastic) supply, and the ratio cane used in tandem with Fe-speciation to better understand Fe

). Fe/Al is used to help interpret FeHR/FeT values between 0.22 and 0.38; when Fe/Aldicate oxic depositional conditions, but this requires independent palaeontological

systematics. Thus, Fe/Al ratios above ∼0.66 provide a robust indi-cation of anoxic water column conditions during deposition.

4. Results

4.1. Kuibis Subgroup: Zaris Basin

Saylor et al. (1995, 1998) have shown that these units weredeposited in settings that ranged from a shallow, inner rampenvironments through to distal, deep ramp environments thateverywhere shallowed-up through time.

4.1.1. Inner rampThe sampled section at Zwartmodder (Fig. 6; Table S2) consists

of ∼18 m Kanies Member, followed by 22 m Lower Omkyk Mem-ber, 75 m Upper Omkyk Member and ∼30 m Hoogland Member. TheLower Omkyk is dominated by dolostones at the base, shallowing tomainly limestone laminates, packstones and grainstones towardsthe top. There are layers of abundant calcified metazoans (Cloud-ina hartmannae and Namacalathus) in the Upper Omkyk and LowerHoogland Members (HST through TST), including minor thrombo-lites in the former.

Carbon isotopes show an increasing trend from −2.93 to −0.07‰in the first ∼33 m of the section; for the remaining ∼88 m valuesoscillate on a high frequency from 5.66 to 2.34 ‰ (Fig. 6).

Both FeT and TOC are variable, from 0 to ∼2 wt% and 0.04 to5.66 wt%, respectively (Fig. 6). Intervals where FeT and TOC areboth <0.5 wt% are intermittent in the Upper Omkyk Member TSTand HST, and Lower Hoogland Member TST, and are interpreted to


Fig. 8. Zebra River section: stratigraphy, lithology, metazoan distribution, �13C, FeT, TOC, and Fe speciation. Where FeT > 0.5 wt%, FeHR/FeT < 0.22 provides a robust indicationo colum( R < 0.7i ay ine

rsabfO

∼rliastem

sf

iwarsataai

f oxic conditions and FeHR/FeT > 0.38 suggests deposition from an anoxic water


s greater than 0.66 samples are likely to be anoxic. <0.5 wt% FeT and <0.5 wt% TOC mvidence.

epresent probable oxic horizons. Fe/Al is above 0.66 in 6 of the 9amples where FeT > 0.5 wt%. Fe speciation in shales in the Kaniest the base of the section indicates an oxic signature. Limited car-onate samples with FeT > 0.5 wt% indicate intermittently anoxicerruginous conditions in the Lower Omkyk Member TST and Uppermkyk Member TST and HST, and Lower Hoogland Member TST.

The Omkyk Site consists of ∼73 m Lower Omkyk Member and30 m Upper Omkyk Member. At Omkyk (Fig. 7; Table S3) mid-

amp re-sedimented carbonate beds and storm-beds typify theower part of the succession and then shallows upward to tidallynfluenced limestone grainstones with limited thrombolites andbundant C. hartmannae and Namacalathus in the upper part. Somehoaling cycles contain evidence for deposition in supra- to inter-idal conditions, which may have been subjected to exposure andvaporitic conditions. There are horizons of abundant calcifiedetazoans in the Upper Omkyk Member late TST only.Carbon isotopes show a general increasing trend throughout the

ection, from as negative as −4.22, up to −0.20‰ for first 22 m, thenor the remainder of the section climbing to 4.80‰ (Fig. 7).

FeT is highly variable, from 0.02 to 2.66 wt%. Fe/Al is above 0.66n 15 of the 19 samples where FeT > 0.5 wt%. Carbonate intervals

here FeT and TOC are both <0.5 wt% are intermittent in the Lowernd Upper Omkyk Member TST and HST, and are interpreted toepresent probable oxic horizons. Fe speciation in carbonates andhales throughout the lower part of the Lower Omkyk Member,nd the Upper Omkyk Member indicates anoxic ferruginous condi-
ions. One carbonate sample in the late TST Upper Omkyk Memberssociated with bioclasts gives an oxic signature from Fe speciation,nd another gives an equivocal oxic signature. Anoxic horizons arenterbedded with bioclast-bearing horizons.
n. Fepy/FeHR ratio separates anoxic settings into euxinic (sulfidic) environments). Fe/Al is used to help interpret FeHR/FeT values between 0.22 and 0.38; when Fe/Aldicate oxic depositional conditions, but this requires independent palaeontological

4.1.2. Mid rampZebra River consists of ∼65 m Lower Omkyk Member, fol-

lowed by ∼100 m Upper Omkyk Member, ∼60 m Lower HooglandMember and ∼75 m Upper Hoogland Member (Fig. 8; Table S4).The Kanies Member is a m-scale package of coarse grained,cross-bedded sandstones. The Lower Omkyk Member is dom-inated by grainstones whereas, through the TST of the UpperOmkyk Member, thrombolite-stromatolite reefs nucleate and formlaterally continuous biostrome layers updip (Grotzinger et al.,2005). Cloudina riemkeae, C. hartmannae, and Namacalathus arefound within thrombolite heads and lag beds in inter-reef shales.Towards the top of the Upper Omkyk Member the section shal-lows into grainstone-dominated facies with subordinate shalehorizons, containing thinner, discontinuous, microbial biostromes.Rip-up clasts and hummocky cross-bedding indicate storm-dominated conditions throughout. The Hoogland Member containsstorm-dominated laminites and heterolithics, shallowing towardsgrainstone-dominated facies.

Carbon isotopes show a general increasing trend throughout thesection, from −1.75‰ at the base, reaching values as positive as6.3‰ at 140 m, then for the remainder of the section are stable ataround 3‰ (Fig. 8).

FeT in carbonate rocks is highly variable, from 0.03 to 1.28 wt%,and follows sequence stratigraphic changes with elevated FeT atthe base of sequences during early TSTs. Carbonate intervals whereFeT and TOC are both <0.5 wt% are present throughout the Lower
Omkyk Member, the late TST of the Upper Omkyk Member, thelate TST of the Lower Hoogland Member, and the entire UpperHoogland Member, and are interpreted to represent probable oxichorizons. Fe speciation on shales and carbonate rocks indicates oxic


F h, 199o ce.

ctMTMoarrshP2

MUaaOargatHetNaN(

ig. 9. Driedornvlagte section: stratigraphy, lithology and �13C (modified from Smitxic depositional conditions, but this requires independent palaeontological eviden

onditions in the HST of the Upper Omkyk Member, the TST ofhe Upper Omkyk Member, and the TST of the Lower Hoogland

ember, with anoxic ferruginous signatures limited to the earlyST of the Upper Omkyk Member, the HST of the Upper Omkykember and the TST of the Lower Hoogland Members. At the base

f the Upper Omkyk Member, laterally continuous thromboliticnd other carbonate rocks show a ferruginous signature, but inter-eef and post-reef shales give an oxic signature. However, FeT/Alatios in siliciclastics give an anoxic signal, suggesting that thehales have suffered from late stage transformation of unsulfidizedighly reactive Fe phases to poorly reactive Fe-rich clays (e.g.,oulton and Raiswell, 2002; Poulton et al., 2010; Cumming et al.,013).

At Driedoornvlagte, the succession covers ∼47 m of Kaniesember, followed by ∼60 m Lower Omkyk Member and over 380 mpper Omkyk Member. The Driedoornvlagte reef (Fig. 9; Table S5) is

large pinnacle reef complex, 7 km long and over 250 m high, thatccumulated in a mid-ramp setting during the TST of the Uppermkyk, with markedly increased accommodation space associ-ted with a deeper-water setting compared to Zebra River. Theeef is predominantly composed of thrombolitic limestones andrainstones, with minor dolomitisation. Calcified metazoan fossilsre found throughout the complex, but are particularly abundantowards the top of the section (Unit 3M of Adams et al., 2004).ere, reef-building C. hartmannae (up to 8 mm diameter; Pennyt al., 2014) and smaller C. riemkeae are found, as well as cryp-ic C. riemkeae within thrombolitic reefs (Wood and Curtis, 2015).
amacalathus occurs intergrown with C. riemkeae, as monospecificggregations, and also grew in crypts (Wood and Curtis, 2015).amapoikia is present occupying fissures within thrombolite reefs
Wood et al., 2002).

8), metazoan distribution, FeT, and TOC. <0.5 wt% FeT and <0.5 wt% TOC may indicate

Carbon isotope values are given in Smith (1998), and show ini-tial negative values of ∼−3.5‰ at the base of the Kanies Member,transitioning to stable positive values in the Upper Omkyk TST of∼2‰ through the main reef unit (Fig. 9).

TOC is very low, between 0.03 and 0.07 wt%. FeT is also very low,between 0.005 and 0.383 wt%. Both TOC and FeT measurementsgive results that are consistently far less than 0.5 wt% throughoutthe sampled section, interpreted to indicate probable persistentlyoxic conditions. No bioturbation has been noted, however, althoughpreservation may not be common in such energetic settings.

4.1.3. Outer rampAt Brak, the succession covers ∼95 m of Kanies Member, fol-

lowed by ∼70 m Lower Omkyk Member. The Brak section (Fig. 10;Table S6) is dominated by fine-grained dolostones inter-beddedwith thin sandstones and shales indicative of deeper, outer rampconditions, with shallow subtidal grainstones (limestones) presentonly towards the very top of the section. There are no metazoansor thrombolites present, but limited stromatolite development atapproximately 30 m in the Kanies Member.

Carbon isotopes are negative in the Kanies Member (between−7.44‰ and ∼2.0‰), and increase through the Lower Omkyk Mem-ber to 2.45‰ (Fig. 10).

FeT is variable, from 0.05 to ∼1.89 wt% in the Kanies Member,and from 0.08 to 0.132 in the limestones of the Lower OmkykMember. Fe/Al is above 0.66 in all samples where FeT > 0.5 wt%. AllFe speciation values in carbonate rocks and shales in the Kanies
Member indicate an anoxic ferruginous signature. No unequivocalredox proxy data are available for the Lower Omkyk Member asall FeT < 0.5 wt%, but TOC is notably >0.5 wt%, ranging from 1.22 to3.39 wt%.


Fig. 10. Brak section: stratigraphy, lithology, �13C, FeT, TOC, Fe speciation and Fe/Al data. Where FeT > 0.5 wt%, FeHR/FeT < 0.22 provides a robust indication of oxic conditionsand FeHR/FeT > 0.38 suggests deposition from an anoxic water column. Fepy/FeHR ratio separates anoxic settings into euxinic (sulfidic) environments (Fepy/FeHR > 0.7–0.8)a to heF are lic

4

nZamsdasBb

4

p(rtSct(

4

MK

nd non-sulfidic (Fe-rich; ferruginous) environments (Fepy/FeHR < 0.7). Fe/Al is usedeHR/FeT values between 0.22 and 0.38; when Fe/Al is greater than 0.66 samplesonditions, but this requires independent palaeontological evidence.

.1.4. Fe/Al trends across the Zaris BasinFe/Al was run for the end-points in the Zaris Basin transect,

amely Zwartmodder (inner ramp), Omkyk (deeper inner ramp),ebra River, and Brak (outer ramp) sections to explore behaviourcross shallow to deep marine settings. These data show enrich-ents in FeT/Al in the deeper, Brak Section compared to the

hallowest section, Zwartmodder (Fig. 11). The significance of thisifference was tested using a Kruskal–Wallis test followed by

Dunn’s post-test for multiple comparisons. The Dunn’s resulthowed a significant difference (p < 0.05) for Zwartmodder andrak, and Zebra River and Brak, but no significant differenceetween other comparisons.

.2. Kuibis Subgroup: Witputs Basin

Two sections were the focus of study and sampling in the Wit-uts Basin, Arasab (Kanies to Moofontein Members) and GrensKanies to Aar Members), both of which record inner rampestricted settings. Also examined were the Pinnacle Reefs at theop of the Huns Member as well as the Spitzkopf Member atwartpunt that record a shallowing trend from deeper marineoarsening-upward TST siliciclastic beds to limestones interpretedo have formed in a subtidal ramp positions below storm wave baseGrotzinger and Miller, 2008).

.2.1. Inner rampArasab consists of ∼2 m Kanies Member, followed by ∼30 m

ara Member, ∼28 m Lower Kliphoek Member, ∼50 m Upperliphoek (Aar) Member, and ∼40 m Moofontein Member (Fig. 12;

lp interpret FeHR/FeT values between 0.22 and 0.38; Fe/Al is used to help interpretkely to be anoxic. <0.5 wt% FeT and <0.5 wt% TOC may indicate oxic depositional

Table S7A and B). The Mara Member consists of limestone,commonly with evaporitic fabrics, interbedded with thin shale lay-ers. The Lower Kliphoek is dominated by sandstone, followed bythe interbedded shales and limestones of the upper Kliphoek (Aar)Member. The Mooifontein Member, a thick limestone containingdistinct oolite bands, defines the top of the section. In situ Erniettaand transported Pteridinium and Rangea are abundant in nearbylocalities of the Lower Kliphoek sandstone and Aar Member units(Hall et al., 2013) but none have been observed in the Arasab sec-tion.

�13C values are negative in the lower 22 m of the Mara Member,from −6.22‰ to −0.22‰, and then rise to a maximum of +4.45‰ inthe Aar Member before steadily converging on ∼3‰ in the upperMooifontein Member (Fig. 12).

FeT data from shales range from 0.51 to 0.77 wt% and Fespeciation data indicate fully anoxic ferruginous conditions. Car-bonate rocks in the Mara, Kliphoek, and Aar Members hadFeT > 0.5 wt%, likewise indicating ferruginous conditions. In sam-ples with FeT < 0.5 wt%, TOC is from 0.05 to 0.13 wt%; these samplesare from rocks interbedded with anoxic layers in the Mara and AarMembers, and continuously through the Mooifontein Member, andare taken to indicate sporadic short-lived intervals of probable oxicwater column conditions although no bioturbation has been noted.

Stratigraphic coverage of the Grens section (Fig. 13; TableS8) covers ∼25 m of the Kanies Member, followed by ∼110 m
Mara Member, ∼30 m Lower Kliphoek Member, and ∼30 m UpperKliphoek (Aar) Member. Conglomerates, sandstones and shales ofthe Kanies Member are overlain by interbedded limestone andshale, arranged in m-scale shallowing cycles commonly exhibiting

R.A. Wood et al. / Precambrian Res

Fig. 11. Fe/Al trends across the Zaris Basin, in a column scatter plot for four studiedZaris Basin Sites, showing frequency distribution of the data. Data are offset hori-zontally to avoid overlap, and horizontal line marks median value. All 4 sites haveaufv

eawMasp

FFnp

non-Gaussian but similar shaped distribution allowing comparison of medianssing a Kruskal–Wallis test. The plotted lines are the median values of Fe/Al = 1.520or Brak, 1.110 for Omkyk, 0.610 for Zebra River, and 0.762 for Zwartmodder. palue ≤ 0.0001.

vaporitic fabrics of the Mara Member, which is markedly thickert Grens than at Arasab implying increasing accommodation spaceith the deepening ramp succession to the south. The Kliphoek
ember is composed of limestone with thin layers of sandstone
nd is overlain by the shale, limestone, and dolo-cemented sand-tones of the Aar Member. At 180 m (Fig. 13), in situ dense beddinglane assemblages of the Ediacaran fossil Nemiana are abundant.

ig. 12. Arasab section: stratigraphy, lithology, �13C, FeT, TOC, and Fe speciation. WhereeHR/FeT > 0.38 suggests deposition from an anoxic water column. Fepy/FeHR ratio separon-sulfidic (Fe-rich; ferruginous) environments (Fepy/FeHR < 0.7). <0.5 wt% FeT and <0.5

alaeontological evidence.

earch 261 (2015) 252–271 263

C-isotope data increase gradually from as low as −7.5‰ in thelower Mara to +2.5‰ in the lower Aar Member, where the firstpositive values occur (Fig. 13).

FeT data was >0.5 wt% for only 17 of the 50 samples analysed,with FeHR/FeT for these 17 samples indicating anoxic ferruginousconditions. Where FeT is <0.5 wt%, TOC is very low, from 0.04 to0.2 wt%. This is taken as an indication of probable oxic conditionsalthough no bioturbation has been noted, and these horizons arefound interbedded with anoxic horizons in the upper Middle andKliphoek Members.

4.3. Schwarzrand Group: Witputs Basin

4.3.1. Inner to outer rampAt Swartpunt (Fig. 14; Table S10), 500 m of the Spitzkopf

Member is exposed terminating at the Ediacaran–Cambrian (E-C)unconformity, and consists of mainly inner ramp HST carbonaterocks and mid- to outer ramp TST siliciclastics. The lower carbonateunits define m-scale cycles that formed in an outer-ramp posi-tion and contain in situ Namacalathus and other bioclastic materialin some beds. Overlying these rocks is a outer ramp TST succes-sion consisting of coarsening-upward beds typically comprised ofa basal green mudstone upward to ripple-laminated and thick-bedded planar-laminated or cross-bedded sandstones containingburrows and soft-bodied Ediacaran fossils including Swartpuntiaand a bed with Pteridinium (Narbonne et al., 1997). Carbonate rocks
above this consist of thin-bedded limestones and dolomites withdm-scale microbialites (thrombolites), some with a ‘leopardskin’texture (patchy dark grey micrite on a light grey background),and thick and thin-bedded limestones with rare flat-pebble and
FeT > 0.5 wt%, FeHR/FeT < 0.22 provides a robust indication of oxic conditions andates anoxic settings into euxinic (sulfidic) environments (Fepy/FeHR > 0.7–0.8) andwt% TOC may indicate oxic depositional conditions, but this requires independent


F OC, ano colum( R < 0.7r

islat

0tT

<fctctt5wtaH

4

aihgh

ig. 13. Grens section: stratigraphy, lithology, metazoan distribution, �13C, FeT, Tf oxic conditions and FeHR/FeT > 0.38 suggests deposition from an anoxic water


equires independent palaeontological evidence.

ntraclastic breccias deposited in shallow to deeper subtidal rampettings below storm wave base. These in turn give way to a HST ofaminated, flaggy limestones containing small (<5 mm) C. riemkeaend thrombolites. Horizontal burrows are first noted ∼60 m belowhe E-C unconformity.

Carbon isotope compositions show steady values of around.5‰, rising to 2‰ at the level of ash beds 1 and 2, then decliningo a minimum of almost 0‰ at the base of the TST shales (Fig. 14).his value increases to ∼1.5‰ at the top of the section.

All but two of the FeT values from Swartpunt carbonates are0.5 wt%, and these also show consistently low TOC throughout,rom 0.04 to 0.07 wt% (Fig. 14). This is taken to indicate oxiconditions, and is confirmed by the presence of bioturbation inhe upper 60 m of the section. TST clastic samples consistentlyontain FeT > 0.5 wt%. FeHR/FeT values vary significantly throughhe section, but all clastics show an oxic signature. By contrast,wo HST carbonate samples with FeT > 0.5 wt% occurring some5 m and 20 m below the E-C unconformity show FeHR/FeT > 0.38,ith low Fepy/FeHR giving an anoxic ferruginous signature. Taken

ogether, these data suggests predominantly oxic conditionst Swartpunt, with fleeting ferruginous episodes in the finalST.

.3.2. Mid rampThe Pinnacle Reefs at Swartkloofberg Farm (Fig. 15; Table S9A

nd B), initiate on the top of the Huns Member (characterised by
nterfingering carbonate rocks, siltstones and shales) and form bio-erms with a synoptic relief of up to 40 m. After termination of reefrowth, the reefs were enveloped by ∼70 m shales of the Feldschuh-orn Member (Saylor et al., 1995) and then ∼20 m of shales the
d Fe speciation. Where FeT > 0.5 wt%, FeHR/FeT < 0.22 provides a robust indicationn. Fepy/FeHR ratio separates anoxic settings into euxinic (sulfidic) environments). <0.5 wt% FeT and <0.5 wt% TOC may indicate oxic depositional conditions, but this

Spitskopf Member. Above this is the E-C unconformity and incisedvalley-fill conglomerate of the Nomtsas Formation.

C-isotope data from the Pinnacle reefs show values from ∼0.6to 1.7‰ with no discernible trends (Fig. 15).

Carbonate rock samples were taken from one of the pinna-cle reefs, and the Feldschuhhorn Member and Spitskopf Membershales were sampled nearby. The sampled reef was composedpredominantly of microbialite, with some laminar micritic layers.Microbialite textures are predominantly stromatolitic at the base(approximately 0–5 m), with thrombolitic textures appearing atapproximately 5 m height, and aggregations of Namacalathus (upto 12 mm diameter) becoming more prevalent towards the top ofthe reef.

FeT through the reef carbonates is low, with values of0.042–0.33 wt%, and TOC is very low, from 0.02 to 0.04 wt% (Fig. 15).These are taken to indicate probable continuous oxic conditions. Nobioturbation has been noted, however, although preservation maynot be common in such energetic settings.

The shales of the Feldschuhhorn and Spitskopf Members give FeTmeasurements ranging from 3.45 to 4.93 wt%. FeHR/FeT have valuesof 0.05–0.21, indicating a persistent oxic depositional environment.One sample, close to the base of the section, gives an equivocalFeHR/FeT value of 0.296.

5. Discussion

5.1. Carbon isotopes

The smooth and consistent variation in �13C values throughoutthe sampled sections parallels �13C variations of contemporaneous


Fig. 14. Swartpunt section: stratigraphy, lithology, metazoan distribution, �13C, FeT, TOC, and Fe speciation. Where FeT > 0.5 wt%, FeHR/FeT < 0.22 provides a robust indicationo colum( R < 0.7r

t(2Nad

ftavTnwsdSu

tSeoasbdpwc

f oxic conditions and FeHR/FeT > 0.38 suggests deposition from an anoxic water


equires independent palaeontological evidence.

erminal Neoproterozoic sections (Ries et al., 2010) from ChinaCondon et al., 2005; Jiang et al., 2003), Australia (Calver, 1995,000) and Oman (Fike et al., 2006), and suggests that the sampledama Group sections are a near-primary record of oceanic �13Cnd that the Nama Sub-Basins were connected to the global oceanuring this interval (Saylor et al., 1998; Grotzinger et al., 1995).

The �13C isotope curves show a striking transition in both basins,rom negative values in the Lowermost Kuibis Subgroup climbingo positive values by the Upper Omkyk Member in the Zaris Basin,nd the coeval Aar Member in the Witputs Basins with persistentalues of ∼+4 to +6‰ in the Kuibis Subgroup thereafter (Fig. 16A).he oldest sections, which are in the Witputs Basin, show values asegative as −7.2‰. This climbing trend coincides in the Zaris Basinith a change of �13C from −2 to −8‰ the distal, deeper ramp

etting of the Kanies Member to values of �13C from −4.5 to 4‰uring the Lower Omkyk Member shallow settings. In the upperchwarzrand Subgroup localities values range from +0.5 to ∼2‰p to the E-C unconformity.

The strongly negative �13C values in the lower parts of the sec-ions are interpreted as recording the recovery from the globalhuram/Wonoka excursion. In the Nama Group, the completexcursion is not preserved owing to the time-transgressive naturef the basal unconformity (Workman et al., 2002). This reflects both

secular, global change in the C-isotopic composition of Ediacaraneawater, as well as the influence of ‘local’, basinal effects, as showny the most negative �13C values occurring in the transition from
istal to proximal ramp settings (up to −3‰) and, likewise, the mostositive values in the most proximal settings (e.g. Zwartmodderith values up to 6.5‰ as compared to the progressively more distal
oeval portions of the Omkyk and Brak sections with values of 5.5‰

n. Fepy/FeHR ratio separates anoxic settings into euxinic (sulfidic) environments). <0.5 wt% FeT and <0.5 wt% TOC may indicate oxic depositional conditions, but this

and 2.5‰, respectively). Critical, though, is that while the appear-ance of calcified metazoans at these sites in both basins postdatesthe end the positive (rising) �13C excursion (Fig. 16A), calcifiedmetazoans have been noted in older sediments (Mara Formation)in other locations (Germs, 1983). This indicates that the onset ofbiomineralization did not occur under post-excursion conditions.

5.2. Evolution of redox in the Nama Group

Redox data reveal a highly dynamic record for the Kuibisand upper Schwarzrand Subgroups, with ferruginous anoxia andoxia varying both spatially and temporally (Fig. 16B). There is noevidence for euxinic (anoxic and sulfidic) waters. In the KuibisSubgroup in both the Zaris and Witputs Basins, proximal, innerramp sites show intermittent and laterally discontinuous oxia.Ferruginous water column conditions were a prevalent featurethroughout deposition in deep and proximal ramp settings, withpersistent oxia forming only in mid-ramp settings. Support for thiscomes from depth trends in Fe/Al (Fig. 11), which suggests thatdeeper basin enrichments are a consequence of progressive watercolumn precipitation and deposition of Fe during upwelling intoshallower waters (hence gradually removing Fe(II) from the watercolumn, leading to lower enrichments in the shallower settings).Conversely, during oxygenated periods, Fe remained in the deepbasin, reflected by the low FeT values of many of the shallow water
oxic samples.
In the upper Schwarzrand Subgroup, oxia becomes more per-sistent, but again is notably only stable in mid-ramp settings,particularly within TSTs. Indeed, proximal, shallow sites still


F , FeT,

b

ef

deCiotd

ranopslisewipoBd

ssKBs

ig. 15. Pinnacle Reefs section: stratigraphy, lithology, metazoan distribution, �13Cut this requires independent palaeontological evidence.

xperienced episodic anoxia in late HSTs close to the E-C uncon-ormity (Figs. 14 and 16B).

Significantly, anoxic and ferruginous deeper water column con-itions persist after the transition to positive �13C that marks thend of the Shuram/Wonoka excursion (Fig. 16A and B). Thus, if the

isotope trend reflects the transition to global-scale oxygenationn the aftermath of the oxidation of a large-scale, isotopically lightrganic carbon pool (Rothman et al., 2003; Johnston et al., 2012),hen this oxidation event was not sufficient to fully oxygenate theeep ocean.

The redox-stratified water column does not simply track depth-elated stratigraphy, as shallow water facies such as grainstonesnd shallow-marine mudstones may show either oxic or anoxic sig-atures. Taken together, these observations suggest that upwellingf deeper anoxic, ferruginous waters into shallow waters was arevalent feature throughout the succession, resulting in progres-ive precipitation of water column Fe (II) across the shelf and henceower Fe/Al ratios in the shallowest water sediments. Interestings that mid-ramp reef settings, particularly those growing in TSTs,how persistent oxia. Such reefs grew elevated into shallow, highlynergetic waters away from proximal influence. So although theater depth of carbonate production may have been similar to

nner ramp settings, this suggests that a mid-ramp position enabledhysical ventilation mechanisms for water column oxygenation toperate during flooding and transgressive sea-level rise (Fig. 17A).y contrast shallow settings during HSTs with limited accommo-ation space experienced only episodic oxygenation (Fig. 17B).

In sum, while placing minimum constraints on global atmo-pheric pO2 levels from local iron speciation data of marine
ediments is problematic, these data show that throughout theuibis Subgroup, from ∼550 to ∼547 Ma, the Zaris and Witputsasins were often only intermittently oxic in inner ramp shallowettings, persistently oxic in shallow waters in mid-ramp settings,
and TOC. <0.5 wt% FeT and <0.5 wt% TOC may indicate oxic depositional conditions,

but persistently anoxic in deep settings. There are some data tosuggest that from the latest Schwartzrund Group up to the E-C unconformity (∼543 to ∼541 Ma), redox conditions may havebecome more persistently oxygenated in shallow marine settings,but still experienced episodic anoxia in late HSTs.

5.3. Distribution of metazoan ecology

In the Kuibis Subgroup in both the Zaris and Witputs Basins,inner ramp sites show intermittent and laterally discontinousoxia. Oxic horizons often coincide with in situ metazoans, includ-ing small representatives of the skeletal forms Cloudina riemkeae,C. hartmannae and Namacalathus (Fig. 18G) at Zwartmodder andOmkyk, and the Ediacaran soft-bodied Nemiana at Grens. By con-trast, mid-ramp settings show longer intervals of constant oxia,where carbonates are expressed as thrombolitic reefs, particu-larly during the TSTs. These reefs record relatively biodiversecommunities of skeletal metazoans, including large reef-buildingCloudina hartmanne (Penny et al., 2014;Fig. 18E), large Namaca-lathus (Fig. 18C) and large, cryptic Namapoikia (Fig. 18D). The deepwater, outer ramp setting at Brak (Fig. 10) records persistent anoxia.

The Pinnacle reefs (Fig. 18A) contain abundant in situ Namaca-lathus, and these show persistent oxia throughout both the reeflimestones and the enclosing clastics of the Feldshuhhorn Forma-tion. Oxic conditions persist into the overlying Spitzkopf Member.The stratigraphically youngest site, Swartpunt (Fig. 14) within theSpitzkopf Member shows mainly oxia, but two episodes of imper-sistent anoxia appear high in the section within the late HSTs.Soft-bodied Ediacaran biota and bioturbation is found in the clastic
TST, and aggregations of small Namacalathus are found at intervalsthroughout the succession with small C. riemkeae associated withmodest thrombolite growth (Fig. 18H), as well as burrows, in theoverlying HST.


Fig. 16. (A) Carbon isotopes, stratigraphy and inferred general water depth settings for 9 sites that transect the Zaris and Witputs Basin, Nama Group, Namibia. (B) Redoxdynamics based on FeHR/FeT, Fe/Al, and FeP/FeHR profiles, and calcified metazoan occurrence and ecology.

268 R.A. Wood et al. / Precambrian Res

Fig. 17. Inferred general redox structure for the Nama Group, Namibia, based on Fe-speciation and Fe/Al. (A) Transgressive Systems Tract; (B) Highstand Systems Tract.Facies distribution based on the Kuibis Formation, after Saylor et al., 1995.

Fig. 18. Metazoan response to persistence of oxygenation. Persistent oxygenation: (A) R(20–40 mm); (D) Namapoikia (cryptic); (E) Reef-building Cloudina hartmanne (5–8 mm). Ihorizon (Namacalathus) showing lower laminated lithology, with surficial upper bioclastishowing in situ monospecific assemblage of Namacalathus (<5 mm) with evidence of distoto low relief thrombolite. (I) In situ aggregating Nemiana.

earch 261 (2015) 252–271

Noteworthy is that in shallow, proximal settings in both theKuibis and Schwarzrand Subgroups, horizons stratigraphicallybelow and between those bearing metazoans show evidence foranoxic ferruginous deposition. These samples often occur withinflooding events at the base of m-scale cycles that shallow upwardsinto sediments that contain small metazoans. Metazoan horizonsgenerally have either oxic FeHR/FeT ratios (<0.38), or very low FeT,suggesting minimal Fe enrichment together with very low TOC.Some horizons (e.g. in the Upper Omkyk), however, record ananoxic water column signal (Figs. 7 and 18F). Here, metazoan bio-clasts are present only within the upper few mm of individualbeds, and are particularly abundant on bedding planes, whereasthe lower part of the bed is expressed as dark, fine-scale laminae(laminite) devoid of any bioclastic material (Fig. 18F). However, allmicrosamples from the lower, dark, laminite to the upper bioclast-bearing region of a single bioclastic horizon have high Fe contents(1.17–2.27 wt%), and anoxic Fe speciation signatures (Fig. 18F).

The distribution of calcified metazoans, as well as their ecology,appears to track redox states, with distinctive ecologies formingin proximal shallow settings with highly transient oxia, comparedto deeper mid-ramp, settings with persistent oxia. This highlylocalised distribution of bioclasts within individual beds (Fig. 18F)emphasises the transient nature of oxygenation. This fine-scale
intra-bed variation may be the explanation for the presence of awater column signal which is, on average, anoxic when consid-ering bulk Fe-speciation analyses that incorporate a significantportion of time of single-bed samples, as well as the microsamples,
eefs; (B) Thrombolites (microbialites); (C) Thrombolite-associated Namacalathusmpersistent oxygenation: (F) Vertical polished section through individual bioclasticc horizon, and attendant FeT and Fe speciation data. (G) Aggregating, bedding planertion due to close-packing during growth (arrowed); (H) Cloudina (<3 mm) attached

an Res

euhltimmli2wteNgi

otctuwmaEtsat

eseomtlrEptsbegsm

6

spbrttb

dmtac


ven where thin bioclastic horizons are present. This is partic-larly noteworthy as the upper bedding plane of some anoxicorizons (Fig. 18F) show clear evidence of small, in situ Namaca-

athus (Fig. 18G). The presence of in situ metazoans does not supporthe contention that all bioclasts were reworked from oxic conditionnto anoxic settings. These data do, however, suggest that these

etazoans were either adapted to intermediate redox settings or,ore likely, that oxic conditions persisted only on very short (eco-

ogical) timescales. A similar sedimentary relationship has beenllustrated in Cloudina-rich shell beds from Paraguay (Warren et al.,013), suggesting that this highly localised distribution of bioclastsithin individual horizons may be typical for Ediacaran skele-

al biota that grew in proximal settings, and indeed may be anxpression of the transient nature of oxygenation. In these settings,amacalathus and Cloudina occur in short-lived but densely aggre-ating monospecific communities (Fig. 18G and H) where the size ofndividuals is often small, <3–5 mm diameter (Figs. 16B, 18G and H).

By contrast, all thrombolitic reefs (Fig. 18A and B), independentf their size and thickness, show very low FeT and TOC, sugges-ive of persistent oxic conditions. These reefs are often complexommunities that include large and relatively diverse in situ skele-al metazoans, including large Cloudina (Fig. 18E), Namacalathusp to 40 mm diameter (Fig. 18C), and locally Namapoikia (Fig. 18D),hich can reach up to 1 m. Noteworthy is that these reefs are foundainly in TSTs in deeper proximal and mid-ramp settings where

ccommodation space was high, and likewise in situ soft-bodieddiacaran biota, such as Nemiana (Fig. 18I) are also found in clas-ic TSTs. By contrast TSTs have variable oxygenation in inner rampettings, e.g. the Aar Member at Grens is intermittently oxic, butnoxic at Arasab. The Ediacaran-bearing mid- to outer ramp TST ofhe Spitzkof Member at Swartpunt is persistently oxic.

In summary, we find a similar relationship to modern marinenvironments where benthic diversity, as well as individualize and abundance, decreases with bottom-water oxygen lev-ls (Rhoads and Morse, 1971). While monospecific communitiesf small, short-lived taxa grew in fleetingly oxic conditions, theost diverse communities of the Nama Group are found in persis-

ently oxic mid-ramp settings (particularly in TSTs) and includearge skeletal forms such as the reef crypt-dweller Namapoikia,eef-building C. hartmanne, and large Namacalathus. Soft-bodieddiacaran metazoans are also restricted to clastic TSTs which recordersistent oxygenation in the sampled sections. Of note also ishat large (up to 30 cm) horizontal burrow systems with sprieteuggesting formation by mobile bilaterian organisms have alsoeen found in the clastic TST in the Omkyk Member (Macdonaldt al., 2014). These observations suggest that the persistent oxy-enation present in these particular settings promoted large bodyize, ecological complex and biodiverse communities, and moreetabolically costly feeding strategies.

. Conclusions

The strongly negative �13C values in the lower parts of theections reflect both a secular, global change in the C-isotopic com-osition of Ediacaran seawater, as well as the influence of ‘local’asinal effects, as shown by the most negative �13C values occur-ing in the transition from distal to proximal ramp settings. Critical,hough, is that the transition to positive �13C values postdateshe appearance of calcified metazoans, indicating that the onset ofiomineralization did not occur under post-excursion conditions.

Significantly, anoxic and ferruginous deeper water column con-itions persist during and after the positive �13C excursion that
arks the end of the Shuram/Wonoka excursion. Thus, if the C iso-
ope trend reflects the transition to global-scale oxygenation in theftermath of the oxidation of a large-scale, isotopically light organicarbon pool, it was not sufficient to fully oxygenate the deep ocean.

earch 261 (2015) 252–271 269

Previous studies have inferred a direct link between the devel-opment of stable oxygenated water column conditions and earlyMetazoan evolution (Canfield et al., 2007; Johnston et al., 2012a,b).Our data build upon this and show that the oxygenation stateof shallow waters during the late Ediacaran was, in fact, highlyvariable. These data reveal the independent redox histories of thetwo Nama Group basins, demonstrating that oxygenic ventilationof oceans in the late Ediacaran was probably dependent on localbasin hydrodynamics. Persistent oxygenation is found preferen-tially in mid ramp and deeper inner ramp settings, particularlyduring transgressive system tracts, suggesting that ventilation wasprovided by high energy flooding events.

This distinctive redox distribution controlled the distributionand ecology of skeletal metazoan communities of the late Edi-acaran. Persistent oxygenation – notably present in mid-rampsettings during TSTs – promoted large body size, ecological complexand relatively biodiverse communities, and more metabolicallycostly feeding strategies, while shallow settings with impersistentoxygenation supported small, short-lived, monospecific communi-ties. Our data support a prominent role for oxygen, and for stableoxygenated conditions in particular, in the distribution and struc-turing of early skeletal metazoan ecosystems.

Acknowledgments

RAW, SAK, and ARP acknowledge support from the Royal Soci-ety, and the Namibian Geological Survey. MOC acknowledgessupport from Edinburgh University Principal’s Scholarship Fund.MOC and ARP acknowledge support from the International Cen-tre for Carbonate Reservoirs (ICCR). SWP, RG, RT, FB, ARP, andRAW acknowledge support from NERC through the ‘Co-evolutionof Life and the Planet’ scheme (NE/1005978/1). FB acknowledgessupport from the Laidlaw Hall fund. We are grateful for permis-sion to access localities on many farms, and we thank A. Horn ofOmkyk, U. Schulze Neuhoff of Ababis, L. and G. Fourie of Zebra River,C. Husselman of Driedornvlagte, and L. G’ Evereet of Arasab andSwartpunt.

Appendix A. Supplementary data

Supplementary data associated with this article can be found,in the online version, at http://dx.doi.org/10.1016/j.precamres.2015.02.004.

References

Adams, E.W., Schroder, S., Grotzinger, J.P., McCormick, D.S., 2004. Digital reconstruc-tion and stratigraphic evolution of a microbial-dominated, isolated carbonateplatform (terminal Proterozoic, Nama Group, Namibia). J. Sediment. Res. 74,479–497.

Amthor, J.E., Grotzinger, J.P., Schroder, S., Bowring, S.A., Ramezani, J., Martin,M.W., Matter, A., 2003. Extinction of Cloudina and Namacalathus at thePrecambrian–Cambrian boundary in Oman. Geology 31, 431–434.

Anderson, T.F., Raiswell, R., 2004. Sources and mechanisms for the enrichment ofhighly reactive iron in euxinic Black Sea sediments. Am. J. Sci. 304, 203–233.

Bjerrum, C., Canfield, D.E., 2011. Towards a quantitative understanding of the lateNeoproterozoic carbon cycle. Proc. Natl. Acad. Sci. U.S.A. 109, 1542–1547.

Bristow, T.F., Kennedy, M.J., 2008. Carbon isotope excursions and the oxidant budgetof the Ediacaran atmosphere and ocean. Geology 36 (11), 863–866.

Burns, S.J., Matter, A., 1993. Carbon isotopic record of the latest Proterozoic fromOman. Eclogae Geol. Helv. 86, 595–607.

Butterfield, N.J., 2007. Macroevolution and macroecology through deep time.Palaeontology 50, 41–55.

Butterfield, N.J., 2009. Oxygen, animals and oceanic ventilation: an alternative view.Geobiology 7, 1–7.

Calver, C.R., (Ph.D. dissertation) 1995. Ediacaran Isotope Stratigraphy of Australia.Macquarie University.

Calver, C.R., 2000. Isotope stratigraphy of the Ediacaran (Neoproterozoic III) of theAdelaide Rift Complex, Australia, and the overprint of water column stratifica-tion. Precambrian Res. 100 (1), 121–150.

Canfield, D.E., Lyons, T.W., Raiswell, R., 1996a. A model for iron deposition to euxinicBlack Sea sediments. Am. J. Sci. 296, 818–834.

http://dx.doi.org/10.1016/j.precamres.2015.02.004

http://dx.doi.org/10.1016/j.precamres.2015.02.004

http://refhub.elsevier.com/S0301-9268(15)00045-5/sbref0005


































































































































































































2 an Res

C

C

C

C

C

C

C

C

C

D

D

D

E

F

G

G

G

G

G

G

G

G

G

G

J

J

J

J

J

J

J

H

70 R.A. Wood et al. / Precambri

anfield, D.E., Raiswell, R., Westrich, J., Reaves, C., Berner, R.A., 1986. The use ofchromium reduction in the analysis of inorganic sulphur in sediments andshales. Chem. Geol. 54, 149–155.

anfield, D.E., Raiswell, R., Bottrell, S., 1992. The reactivity of sedimentary iron min-erals toward sulfide. Am. J. Sci. 292, 818–834.

anfield, D.E., Lyons, T.W., Raiswell, R., 1996b. A model for iron deposition to euxinicBlack Sea sediments. Am. J. Sci. 292, 818–834.

anfield, D.E., Poulton, S.W., Narbonne, G.M., 2007. Late-Neoproterozoic deep-oceanoxygenation and the rise of animal life. Science 315, 92–95.

anfield, D., Poulton, S.W., Knoll, A.H., Narbonne, G.M., Ross, G., Goldberg, T., Strauss,H., 2008. Ferruginous conditions dominated later Neoproterozoic deep-waterchemistry. Science 321, 949–953.

larkson, M.O., Poulton, S.W., Guilbaud, R., Wood, R., 2014. Assessing the utility ofFe/Al and Fe-speciation to record water column redox conditions in carbonate-rich sediments. Chem. Geol. 382, 111–122.

loud Jr., P.E., 1968. Atmospheric and hydrospheric evolution on the primitive Earth.Science 160, 729–736.

ondon, D., Zhu, M., Bowring, S., Wang, W., Yang, A., Jin, Y., 2005. U–Pb ages fromthe Neoproterozoic Doushantuo Formation, China. Science 308, 95–98.

umming, V.M., Poulton, S.W., Rooney, A.D., Selby, D., 2013. Anoxia in the terrestrialenvironment during the late Mesoproterozoic. Geology 41, 583–586.

erry, L.A., 2010. On the significance of d13C correlations in ancient sediments. EarthPlanet. Sci. Lett. 296, 497–501.

es Marais, D.J., Strauss, H., Summons, R.E., Hayes, J.M., 1992. Carbon isotope evi-dence for the stepwise oxidation of the Proterozoic environment. Nature 359,605–609.

iaz, R.J., Rosenberg, R., 1995. Marine benthic hypoxia: a review of its ecologicaleffects and the behavioural responses of benthic macrofauna. Oceanogr. Mar.Biol.: Annu. Rev. 33, 245–303.

rwin, D.H., Laflamme, M., Tweedt, S.M., Sperling, E.A., Pisani, D., Peterson, K.J., 2011.The Cambrian Conundrum: early divergence and later ecological success in theearly history of animals. Science 334, 1091–1097.

ike, D.A., Grotzinger, J.P., Pratt, L.M., Summons, R.E., 2006. Oxidation of the Edi-acaran Ocean. Nature 444, 744–747.

erms, G.J., 1972. New shelly fossils from the Nama Group, South-West Africa. Am.J. Sci. 272, 752–761.

erms, G.J.B., 1983. Implications of a sedimentary facies and depositional environ-mental analysis of the Nama Group in South West Africa/Namibia. Evol. DamaraOrogen: Geol. Soc. S. Afr., Spec. Publ. 11, 89–114.

erms, G.J.B., 1995. The Neoproterozoic of Southwestern Africa, with emphasis onplatform stratigraphy and paleontology. Precambrian Res. 73 (1-4), 137–151.

erms, G.B.J., Knoll, A.H., Vidal, G., 1986. Latest Proterozoic fossils from the NamaGroup, Namibia. Precam. Res. 32, 45–62.

rant, S.W.F., 1990. Shell structure and distribution of Cloudina, a potential indexfossil for the terminal Proterozoic. Am. Jour. Sci. 290-A, 261–294.

resse, P.G., Germs, G.J.B., 1993. The Nama Foreland Basin: sedimentation, majorunconformity Bounded Sequences and Multisided Active Margin Advance. Pre-cambrian Res. 63, 247–272.

rotzinger, J., Adams, E.W., Schroeder, S., 2005. Microbial–Metazoan Reefs of the Ter-minal Proterozoic Nama Group (c. 550–543 Ma), vol. 142. Geological Magazine,Namibia, pp. 499–517.

rotzinger, J.P., Bowring, S.A., Saylor, B.Z., Kaufman, A.J., 1995. Biostratigraphicand geochronological constraints on early animal evolution. Science 13,229–272.

rotzinger, J.P., Miller, R., 2008. The Nama Group. In: Miller, R. (Ed.), The Geol-ogy of Namibia, Vol. 2: Geological Survey of Namibia Special Publication, pp.13229–13272.

rotzinger, J.P., Watters, W.A., Knoll, A.H., 2000. Calcified Metazoans in Thrombolite-Stromatolite Reefs of the Terminal Proterozoic Nama Group. Paleobiology,Namibia, pp. 334–359.

ensen, S., Saylor, B.Z., Gehling, J.G., Germs, G.J.B., 2000. Complex tracefossils fromthe terminal Proterozoic of Namibia. Geology 28, 143–146.

iang, G., Sohl, L.E., Christie-Blick, N., 2003. Neoproterozoic stratigraphic comparisonof the Lesser Himalaya (India) and Yangtze block (south China): Paleogeographicimplication. Geology 31, 917–920.

ohnston, D.T., Poulton, S.W., Dehler, C., Porter, S., Husson, J., Canfield, D.E., Knoll, A.H.,2010. An emerging picture of Neoproterozoic ocean chemistry: insight from theChuar Group, Grand Canyon, USA. Earth Planet. Sci. Lett. 290, 64–73.

ohnston, D.T., Poulton, S.W., Goldberg, T., Sergeev, V.N., Podkovyrov, V., Vorob’eva,N., Bekker, A., 2012. Late Ediacaran redox stability and metazoan evolution. EarthPlanet. Sci. Letts. 335–336, 25–35.

ohnston, D.T., Macdonald, F.A., Gill, B.C., Hoffman, P.F., Schrag, D.P., 2012a. Uncov-ering the Neoproterozoic Carbon cycle. Nature 483, 320–323.

ohnston, D.T., Poulton, S.W., Goldberg, T., Sergeev, V.N., Podkovyrov, V., Vorob’eva,N.G., Bekker, A., Knoll, A.H., 2012b. Late Ediacaran redox stability and metazoanevolution. Earth Planet. Sci. Lett. 335, 25–35.

ohnston, D.T., Poulton, S.W., Tosca, N.J., O’Brien, T., Halverson, G.P., Schrag, D.P.,Macdonald, F.A., 2013. Searching for an oxygenation event in the fossiliferousEdiacaran of northwestern Canada. Chem. Geol. 362, 273–286.

all, M., Kaufman, A.J., Vickers-Richs, P., Ivantsov, A., Trusler, P., Linnemann, U.,Hofmann, M., Elliott, D., Cui, H., Fedonkin, M., Hoffmann, K.-H., Wilson, S.A.,
Schneider, G., Smith, J., 2013. Stratigraphy, palaeontology and geochemistry ofthe late Neoproterozoic Aar Member, southwest Namibia: Reflecting environ-mental controls on Ediacara fossil preservation during the terminal Proterozoicin African Gondwana. Precambrian Res. 238, 214–232.
earch 261 (2015) 252–271

Kendall, B., Reinhard, C.T., Lyons, T.W., Kaufman, A.J., Poulton, S.W., Anbar, A.D.,2010. Pervasive oxygenation along late Archaean ocean margins:. Nat. Geosci.3, 647–652.

Knoll, A.H., 2003. Vestiges of a beginning? Paleontological and geochemical con-straints on early animal evolution. Ann. Paleontol. 89, 205–221.

Knoll, A.H., Sperling, E.A., 2014. Oxygen and animals in Earth history. Proc. Natl.Acad. Sci. U.S.A. 111, 3907–3908.

Lenton, T.M., Boyle, R.A., Poulton, S.W., Shields-Zhou, G.A., Butterfield, N.J., 2014. Co-evolution of eukaryotes and ocean oxygenation in the Neoproterozoic era. Nat.Geosci. 7, 257–265.

Li, C., Love, G.D., Lyons, T.W., Fike, D.A., Sessions, A.L., Chu, X., 2010. A stratified redoxmodel for the Ediacaran Ocean. Science 328, 80–83.

Logan, G.A., Hayes, J.M., Hieshima, G.B., Summons, R.E., 1995. Terminal Proterozoicreorganization of biogeochemical cycles. Nature 376, 53–56.

Macdonald, F.A., Pruss, S.B., Strauss, J., 2014. Trace fossils with Spreiten from the LateEdiacaran Nama Group, Namibia: complex feeding patterns five million yearsbefore the precambrian–cambrian boundary. J. Paleontol. 88, 299–308.

Martin, M.W., Grazhdankin, D.V., Bowring, S.A., Evans, D.A., Fedonkin, D., Kirschvink,M.A.J.A., 2000. Age of Neoproterozoic bilaterian body and trace Fossils, WhiteSea, Russia: implications for metazoan evolution. Science 288, 841–845.

März, C., Poulton, S.W., Beckmann, B., Kuster, K., Wagner, T., Kasten, S., 2008. Redoxsensitivity of P cycling during marine black shale formation: dynamics of sul-fidic and anoxic non-sulfidic bottom waters. Geochim. Cosmochim. Acta 72,3703–3717.

McFadden, K.A., Huang, J., Chu, X., Jiang, G., Kaufman, A.J., Zhou, C., Yuan, X., Xiao,S., 2008. Pulsed oxidation and biological evolution in the Ediacaran Doushantuoformation. Proc. Natl. Acad. Sci. USA 105, 3197–3202.

Mills, D.B., Ward, L.M., Jones, C., Sweeten, B., Forth, M., Treusch, A.H., Canfield, D.E.,2014. Oxygen requirements of the earliest animals. Proc. Natl. Acad. Sci. U.S.A.111, 4168–4172.

Narbonne, G.M., Gehling, J.G., 2003. Life after Snowball: the oldest complex Edi-acaran fossils. Geology 31, 27–30.

Narbonne, G.M., Saylor, B.Z., Grotzinger, J.P., 1997. The youngest Ediacaran fossilsfrom Southern Africa. J. Paleontol. 71 (6), 953–967.

Penny, A.M., Wood, R., Curtis, A., Bowyer, F., Tostevin, R., Hoffman, K.-H., 2014. Edi-acaran metazoan reefs from the Nama group, Namibia. Science 344, 1504–1506.

Planavsky, N.J., McGoldrick, P., Scott, C.T., Li, C., Reinhard, C.T., Kelly, A.E., Chu, X.,Bekker, A., Love, G.D., Lyons, T.W., 2011. Widespread iron-rich conditions in themid-Proterozoic ocean. Nature 477, 448–451.

Poulton, S.W., Raiswell, R., 2002. The low-temperature geochemical cycle ofiron: from continental fluxes to marine sediment deposition. Am. J. Sci. 302,774–805.

Poulton, S.W., Canfield, D.E., 2005. Development of a sequential extraction procedurefor iron: implications for iron partitioning in continentally derived particulates:.Chem. Geol. 214, 209–221.

Poulton, S.W., Fralick, P.W., Canfield, D.E., 2004a. The transition to a sulphidic ocean∼1.84 billion years ago. Nature 431, 173–177.

Poulton, S.W., Krom, M.D., Raiswell, R., 2004b. A revised scheme for the reactivity ofiron (oxyhydr)oxide minerals towards dissolved sulfide. Geochim. Cosmochim.Acta 68, 3703–3715.

Poulton, S.W., Fralick, P.W., Canfield, D.E., 2010. Spatial variability in oceanic redoxstructure 1.8 billion years ago. Nat. Geosci. 3, 486–490.

Poulton, S.W., Canfield, D.E., 2011. Ferruginous conditions: a dominant feature ofthe ocean through Earth’s history. Elements 7, 107–112.

Raiswell, R., Canfield, D.E., 1998. Sources of iron for pyrite formation in marinesediments. Am. J. Sci. 298, 219–245.

Raiswell, R., et al., 2008. Turbidite depositional influences on the diagenesis ofBeecher’s Trilobite Bed and the Hunsruck Slate; sites of soft tissue pyritization.Am. J. Sci. 308 (2), 105–129.

Rhoads, D.C., Morse, J.W., 1971. Evolutionary and ecologic significance of oxygen-deficient basins. Lethaia 4, 413–428.

Ries, J.B., Fike, D.A., Pratt, L.M., Lyons, T.W., Grotzinger, J.P., 2010. Super-heavypyrite (�34Spyr >�34SCAS) in the terminal Proterozoic Nama Group, SouthernNamibia: a consequence of low seawater sulfate at the dawn of animal life.Geology 37, 743–746.

Rothman, D.H., Hayes, J.M., Summons, R.E., 2003. Dynamics of the Neoproterozoiccarbon cycle. Proc. Natl. Acad. Sci. U.S.A. 100, 8124–8129.

Runnegar, B., 1982. Oxygen requirements, biology and phylogenetic significance ofthe late Precambrian worm Dickinsonia, and the evolution of the burrowinghabit. Alcheringa 6, 223–239.

Saylor, B.Z., Grotzinger, J.P., 1996. Reconstruction of important Proterozoic–Cambrian boundary exposures through the recognition of thrust deformationin the Nama Group of southern Namibia. Commun. Geol. Surv. Namib. 11, 1–12.

Saylor, B.Z., Grotzinger, J.P., Germs, G.J.B., 1995. Sequence stratigraphy and sedimen-tology of the Neoproterozoic Kuibis and Schwarzrand Subgroups (Nama Group),southwestern Namibia:. Precambrian Res. 73, 153–171.

Saylor, B.Z., Kaufman, A.J., Grotzinger, J.P., Urban, F., 1998. A composite referencesection for terminal Proterozoic strata of southern Namibia. J. Sediment. Res. 68(6), 1223–1235.

Schmitz, M.D., 2012. Radiogenic isotope geochronology. In: Gradstein, F.M., Ogg, J.G.,Schmidt, M.D., Ogg, G.M. (Eds.), Geologic Time Scale 2012. Elsevier, pp. 115–
126.
Scott, C., Lyons, T.W., Bekker, A., Shen, Y., Poulton, S.W., Chu, X., Anbar, A.D.,2008. Tracing the stepwise oxygenation of the Proterozoic ocean. Nature 452,456–459.

































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































an Res

S

S

S

S

V

W


evermann, S., Lyons, T.W., Anbar, A., McManus, J., Gordon, G., 2008. Modern iron iso-tope perspective on the benthic iron shuttle and the redox evolution of ancientoceans. Geology 36, 487–490.

mith, O.A., (unpublished M.Sc. thesis) 1998. Terminal Proterozoic Carbonate Plat-form Development: Stratigraphy and Sedimentology of the Kuibis Subgroup(ca. 550–548 Ma), Northern Nama Basin, Namibia. Massachusetts Institute ofTechnology, Cambridge, MA, 132 p.

perling, E.M., Halverson, G.P., Knoll, A.H., Macdonald, F.A., Johnston, D.T., 2013a.A basin redox transect at the dawn of animal life. Earth Planet. Sci. Lett. 371,143–155.

perling, E.A., Freider, C.A., Raman, A.V., Mirguis, P.R., Levin, L.A., Knoll, A.H., 2013b.Oxygen, ecology and the Cambrian radiation of animals. Proc. Natl. Acad. Sci.U.S.A. 110, 13446–13451.

aquer-Sunyer, R., Duarte, C.M., 2008. Thresholds of hypoxia for marine biodiversity.Proc. Natl. Acad. Sci. U.S.A. 105, 15452–15457.

arren, L.V., Simões, M.G., Fairchild, T.R., Riccomini, C., Gaucher, C., Anellli, L.E.,Freitas, B.T., Boggiani, P.C., Quaglio, F., 2013. Origin and impact of the oldestmetazoan bioclastic sediments. Geology 41, 507–510.

earch 261 (2015) 252–271 271

Wilson, J.P., et al., 2012. Deep-water incised valley deposits at the Ediacaran–Cambrian boundary in Southern Namibia contain abundant Treptichnus pedum.Palaios 27, 252–273.

Wood, R., Curtis, A., 2015. Extensive metazoan reefs from the Ediacaran Nama Group,Namibia: the rise of benthic suspension feeding. Geobiology 13, 112–122.

Wood, R.A., Grotzinger, J.P., Dickson, J.A.D., 2002. Proterozoic modular biomineral-ized metazoan from the Nama Group. Science 296, 2383–2386.

Wood, R.A., 2011. Paleoecology of the earliest skeletal metazoans: implications forearly biomineralization. Earth Sci. Rev. 106, 184–190.

Workman, R.K., Grotzinger, J.P., Hart, S.R., 2002. Constraints on Neoproterozoicocean chemistry from 13C and 11B analyses of carbonates from the Witvlei andNama Groups, Namibia. Geochim. Cosmochim. Acta 66, 847.

Xiao, S., Zhang, Y., Knoll, A.H., 1998. Three-dimensional preservation of algae and
animal embryos in a Neoproterozoic phosphorite. Nature 391, 553–558.
Zegeye, A., Bonneville, S., Benning, L.G., Sturm, A., Fowle, D.A., Jones, C., Canfield, D.E.,Ruby, C., MacLean, L.C., Nomosatryo, S., Crowe, S.A., Poulton, S.W., 2012. Greenrust formation controls nutrient availability in a ferruginous water column.Geology 40, 599–602.






















































































































































































































































dynamic redox conditions control late ediacaran metazoan ... · of the nama group, the northern...

Documents