IXE

SGa

b

c

d

e

f

g

a

ARRAA

KESDCRE

1

Cue22E

tf

0d

Precambrian Research 192– 195 (2012) 125– 141

Contents lists available at SciVerse ScienceDirect

Precambrian Research

journa l h omepa g e: www.elsev ier .com/ locate /precamres

ntegrated chemostratigraphy of the Doushantuo Formation at the northerniaofenghe section (Yangtze Gorges, South China) and its implication fordiacaran stratigraphic correlation and ocean redox models

huhai Xiaoa,e,∗, Kathleen A. McFaddenb, Sara Peekc, Alan J. Kaufmand, Chuanming Zhoue,anqing Jiangf, Jie Hug

Department of Geosciences, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USAConocophillips, Houston, TX 77079, USADepartment of Geology, University of Maryland, College Park, MD 20742, USADepartment of Geology and Earth System Science Interdisciplinary Center, University of Maryland, College Park, MD 20742, USAState Key Laboratory of Paleobiology and Stratigraphy, Nanjing Institute of Geology and Paleontology, Chinese Academy of Sciences, Nanjing 210008, ChinaDepartment of Geoscience, University of Nevada, Las Vegas, NV 89154, USAPetroChina Tarim Oilfield Company, Korla, Xinjiang 841000, China

r t i c l e i n f o

rticle history:eceived 12 July 2011eceived in revised form 22 October 2011ccepted 28 October 2011vailable online 4 November 2011

eywords:diacaranouth Chinaoushantuo Formationarbon, sulfur, strontium isotopesedox

a b s t r a c t

The Ediacaran Doushantuo Formation in the Yangtze Gorges of South China plays an important role inour understanding of biological evolution, global correlation, and ocean redox conditions, because of theavailability of high-resolution paleontological and geochemical data and numerous radiometric dates.However, integrated study has been focused on the Jiulongwan section that was largely deposited belowwave base in a restricted shelf lagoon (Jiang et al., 2011; Zhu et al., 2011). Studies of shallower watersuccessions are lacking, and this presents a challenge to test Ediacaran stratigraphic correlation andocean redox models. To fill this knowledge gap, we conducted a high-resolution integrated study of theDoushantuo Formation at the northern Xiaofenghe (NXF) section approximately 35 km to the northeastand paleogeographically updip of the Jiulongwan section. With the exception of the basal 20 m, NXFsediments were deposited above normal wave base. Integrated biostratigraphic and chemostratigraphicdata indicate that the 140 m thick NXF section correlates with the lower Doushantuo Formation (Member

uxinia I and much of Member II; i.e., the lower ca. 70 m of the formation) at Jiulongwan. Geochemical data fromNXF and other Doushantuo sections indicate that euxinic conditions may have been limited to a shelflagoon (represented by the Jiulongwan section) that was restricted between the proximal inner shelf anda distal shelf margin shoal complex, at least during early Doushantuo time following the deposition of theDoushantuo cap dolostone. Further integrated studies are necessary to test whether euxinic conditionsexisted in open marine shelves in South China and elsewhere during the Ediacaran Period.

. Introduction

The Doushantuo Formation (ca. 635–551 Ma) on the Yangtzeraton of South China has played an important role in ournderstanding of Ediacaran stratigraphic correlation, biologicalvolution, and ocean redox history (Xiao et al., 1998; Jiang et al.,

007; Zhou et al., 2007; Zhu et al., 2007, 2011; McFadden et al.,008, 2009; Li et al., 2010; Liu et al., 2011). Current models ofdiacaran stratigraphic correlation in South China are based upon

∗ Corresponding author at: Department of Geosciences, Virginia Polytechnic Insti-ute and State University, Blacksburg, VA 24061, USA. Tel.: +1 540 231 1366;ax: +1 540 231 3386.

E-mail address: xiao@vt.edu (S. Xiao).

301-9268/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.precamres.2011.10.021

© 2011 Elsevier B.V. All rights reserved.

biostratigraphic, chemostratigraphic, and geochronological datafrom the Yangtze Gorges area, particularly the Jiulongwan andother sections on the southern limb of the Huangling anticline(Condon et al., 2005; McFadden et al., 2008; Yin et al., 2011). Inthe area around Jiulongwan, the Doushantuo Formation has beensampled at high resolution by several research groups, in both out-crop and drill core, and has been analyzed using a combinationof micropaleontological and geochemical tools (Jiang et al., 2007;McFadden et al., 2008, 2009; Sawaki et al., 2010; Liu et al., 2011;Yin et al., 2011; Zhu et al., 2011). However, the Doushantuo Forma-tion is characterized by significant facies and thickness variations

on the shelf (McFadden et al., 2009; Jiang et al., 2011), and it is com-plicated by the presence of allochthonous olistostromes in upperslope environments (Vernhet et al., 2006; Vernhet and Reijmer,2010). Furthermore, Doushantuo fossil occurrences are subject to

126 S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141

Fig. 1. Geological maps, localities, and stratigraphic correlations. (A) Map showing location of the Yangtze, Cathaysia, North China, and Tarim blocks in China. (B) A generalizedpaleogeographic map of the Yangtze Craton during the early Ediacaran Period (Jiang et al., 2011). Rectangle at Yichang marks the Huangling anticline. Localities (Baokang, Lan-tian, Longe, Minle, Weng’an, and Zhongling) mentioned in text are marked. (C) Geological map of the Huangling anticline. Localities (Jiulongwan = JLW, Tianjiayuanzi = TJYZ,northern Xiaofenghe = NXF, southern Xiaofenghe = SXF, and Zhangcunping) are marked. Note that the location of Xiaofenghe and Zhangcunping were incorrectly marked in pre-vious publications, including Zhou et al. (2007), Zhu et al. (2007), McFadden et al. (2009), Jiang et al. (2011), and Yin et al. (2011). (D–E) Possible correlation between DoushantuoFormation in the Yangtze Gorges area (e.g., Jiulongwan section) and in shallower facies (e.g., Weng’an section), highlighting the implications for the biostratigraphic division ofacanthomorph assemblages. (D) Correlation proposed by Zhu et al. (2007, 2011), in which Tianzhushania spinosa is restricted to the lower assemblage. (E) Correlation proposedby Zhou et al. (2007), in which Tianzhushania spinosa extends to the upper assemblage. (F) Alternative correlation, in which potentially three assemblages can be recognizedin the Doushantuo Formation. �13Ccarb curves, radiometric ages, and exposure surfaces are omitted in (E)–(F) for clarity. NT: Nantuo Formation; DY: Dengying Formation.

esearc

paDfttmmrmapccmwdfLtphs

sed(tgdtJtpast

2s

2

so∼grtogoVnJtPwFFr

C

S. Xiao et al. / Precambrian R

aleoenvironmental and taphonomic biases (Zhou et al., 2007). As result, fine-scale litho- and biostratigraphic correlation of theoushantuo Formation at the regional scale is not always straight-

orward. The difficulty is exacerbated by the condensed nature ofhe Doushantuo Formation: ∼84 million years of geological his-ory is recorded in ∼200 m of sedimentary rocks, possibly with

ultiple hiatuses. It is therefore important to test biostratigraphicodels derived from one area through integrated analysis of high-

esolution stratigraphic data from others. Similarly, geochemicalodels derived from the Jiulongwan section also need to be tested

gainst new data from other sections. For example, Li et al. (2010)roposed a stratified redox model for the Doushantuo basin thatonsists of an oxic surface layer, a thin ferruginous layer at thehemocline, a euxinic wedge, and a ferruginous deep ocean. Thisodel was based on elemental and isotopic data from the Jiulong-an section, the Zhongling section (then interpreted to have beeneposited in the outer shelf on a ramp-like platform; see Fig. 1Bor section location), and two poorly sampled sections at Minle andonge representing slope and basinal facies. This model makes cer-ain predictions about the ocean redox conditions in the onshoreart of the basin, but these predictions cannot be tested withoutigh-resolution data from the Doushantuo Formation deposited inhallow-water environments.

To address the need for high-resolution, integrated data from ahallow-water inner shelf environment, we investigated the north-rn Xiaofenghe (NXF) section. This section was chosen because (1)etailed micropaleontological studies have already been publishedYin et al., 2007; McFadden et al., 2009), (2) the micropaleon-ological sample horizons were marked on the outcrop so thateochemical data can be precisely integrated with paleontologicalata, and (3) according to current basin architecture reconstruc-ion (Jiang et al., 2011), this section is about 35 km up-dip from theiulongwan section, allowing an exploration of paleoenvironmen-al conditions in shallower parts of the Yangtze Craton. Here weresent carbonate carbon, organic carbon, pyrite sulfur, carbonatessociated sulfate sulfur, and strontium isotopic data from the NXFection to address basinwide correlations (Jiang et al., 2011) and toest predictions of the stratified redox model (Li et al., 2010).

. Geological background and review of previoustratigraphic work

.1. Physical stratigraphy and radiometric age constraints

Neoproterozoic successions in South China developed on aouth–southeast facing (present-day position) passive marginn the Yangtze Craton during the breakup of Rodinia about830–750 Ma (Wang and Li, 2003; Zhao et al., 2011). The Cryo-enian glacial and interglacial successions (750–635 Ma) record aift-drift transition (Jiang et al., 2003a), and subsequent deposi-ion of the early Ediacaran Doushantuo Formation (635–551 Ma)ccurred in marginal marine environments and focused on topo-raphic lows created by block faulting during the developmentf the Nanhua Rift System (Cao et al., 1989; Wang and Li, 2003;ernhet, 2007). As a result, the Doushantuo Formation shows sig-ificant variations in facies and thickness (Zhu et al., 2007, 2011;

iang et al., 2011); its thickness ranges from 40 m at Weng’ano ∼300 m at Zhongling (see Fig. 1B for locations). The Yangtzelatform eventually evolved into an open shelf extending to a deep-ater basin during the deposition of the late Ediacaran Dengying

ormation (551–542 Ma) and its deepwater equivalent Liuchapo

ormation, which are overlain by basal Cambrian chert, phospho-ite, and shale (Dong et al., 2008, 2009; Ishikawa et al., 2008).

Although the Doushantuo Formation outcrops widely in Southhina, previous work has been centered on the Yangtze Gorges area,

h 192– 195 (2012) 125– 141 127

where the well exposed Jiulongwan section on the southern limbof the Huangling anticline is among the most extensively studied(Fig. 1). In the Yangtze Gorges area, the Doushantuo Formation canbe more than 200 m thick and generally it has been subdivided intofour lithostratigraphic members (e.g., Zhao et al., 1988; McFaddenet al., 2008, 2009). Member I is a 5-m-thick cap dolostone charac-terized by abundant sheet cracks, tepee structures, macropeloids,and barite fans (Jiang et al., 2006). Member II (∼80 m thick) is char-acterized by alternating organic-rich calcareous shale and shaleydolostone beds with abundant pea-sized fossiliferous chert nodules(Xiao et al., 2010). The boundary between Member II and MemberIII is transitional at the Jiulongwan section, but has been correlatedwith a mid-Doushantuo exposure surface in several shallow-watersections (e.g., Weng’an and Zhangcunping; see Fig. 1 for location)(Zhou and Xiao, 2007). This correlation that has been challengedby more recent studies (Yin et al., 2009, 2011; Liu et al., 2011; Zhuet al., 2011). Member III (∼60 m thick) consists of massive and lam-inated cherty dolostone overlain by interbedded lime mudstoneand dolomitic mudstone. Member IV (∼10 m thick) is composed ofblack organic-rich calcareous shale, siliceous shale, or mudstone.The four-fold division of the Doushantuo Formation, however, isnot easily applicable to sections beyond the Yangtze Gorges area.

Three depositional cycles (sequences) are recognized in theDoushantuo Formation in the Yangtze Gorges area (Zhu et al., 2007;McFadden et al., 2008). The boundary between the first two cyclesis marked by a mid-Doushantuo exposure surface in shallow-watersections elsewhere in the Yangtze Craton, such as the Weng’an andZhangcunping sections (Liu et al., 2009b; McFadden et al., 2009;Jiang et al., 2011; Zhu et al., 2011). In the Yangtze Gorges area,however, no easily recognizable exposure surfaces have been iden-tified, and various authors have correlated the mid-Doushantuoexposure surface at Weng’an with a lithofacies change near theMembers II–III boundary (Zhou and Xiao, 2007; McFadden et al.,2009; Jiang et al., 2011) or with a Member II horizon where a car-bonate unit is overlain by black shale with chert and phosphaticnodules (Zhu et al., 2007, 2011). The third cycle starts at a majorlithofacies change at the Members III–IV boundary and is termi-nated by a widespread karst surface in the Hamajing Member of theDengying Formation (Zhu et al., 2007), although Zhu et al. (2011)have recently discovered “a 30 cm thick interval of silty shale” inMember III at the Jiulongwan section and interpreted this silty shale“to mark a sequence boundary” and the beginning of the third cycle.These uncertainties highlight the challenge in identifying cycles orsequences in the Yangtze Gorges area where exposure surfaces arepoorly developed.

The depositional environment of the Doushantuo Formationin the Yangtze Gorges area is a matter of debate. The abundanceof saponitic smectite and unusually low concentration of redox-sensitive elements (Mo, U, V) in the lower Doushantuo Formation inthe Yangtze Gorges area has been interpreted as evidence for depo-sition in alkaline lacustrine environments (Bristow et al., 2009).This view, however, is at odds with the spatial continuity of theDoushantuo Formation across much of the Yangtze Craton, themarine affinity of Doushantuo fossil assemblages, and the similar-ity of Doushantuo lithostratigraphic and chemostratigraphic trendswith Ediacaran successions beyond the Yangtze Gorges area (Jianget al., 2003a, 2007; Kaufman et al., 2006; Zhou and Xiao, 2007; Zhouet al., 2007). Recent analysis of facies variations favors the interpre-tation that the Doushantuo Formation (with the exception of thecap dolostone) at Jiulongwan was deposited in a restricted shelflagoon (Ader et al., 2009; Vernhet and Reijmer, 2010; Jiang et al.,2011).

Several radiometric dates constrain the depositional age ofthe Doushantuo Formation. A U–Pb zircon age from the Cryo-genian Nantuo Formation (636.4 ± 4.9 Ma) (Zhang et al., 2008)places a maximum age constrain on the Doushantuo Formation.

1 esearc

OT(Mo2Jtb

atHTle2at6

2

YSip2a−uesveW�2ceWftDbtGh2Et2(JtbWc

−tY�ba

28 S. Xiao et al. / Precambrian R

n the southern limb of the Huangling anticline, zircon U–PbIMS ages have been reported from Member I cap dolostone635.2 ± 0.6 Ma), lower Member II (632.5 ± 0.5 Ma), and uppermost

ember IV (551.1 ± 0.7 Ma) (Condon et al., 2005). Two Re–Os agesf 598 ± 16 Ma (Kendall et al., 2009) and 593 ± 17 Ma (Zhu et al.,010) have been reported from the Member IV black shale atiulongwan, although Kendall et al. (2009) cautioned the interpre-ation of their Re–Os age because of possible heterogeneity of thelack shale samples.

Additionally, Liu et al. (2009b) reported a zircon U–Pb SHRIMPge of 614 ± 7.6 Ma from an ash bed in the Doushantuo Forma-ion at the Zhangcunping section on the northeastern limb of theuangling anticline, about 60 km to the northeast of Jiulongwan.his ash bed was collected within a dolostone that overlies theower Doushantuo black shale but underlies the mid-Doushantuoxposure surface. Finally, a Pb–Pb age of 599 ± 4 Ma (Barfod et al.,002) has been reported from the upper Doushantuo phosphoritet Weng’an, above the mid-Doushantuo exposure surface. Thesewo ages constrain the mid-Doushantuo exposure surface between14 ± 7.6 Ma and 599 ± 4 Ma.

.2. Chemostratigraphy

The chemostratigraphy of the Doushantuo Formation in theangtze Gorges area has been the focus of intensive investigation.everal sections on the southern limb of the Huangling anticline,ncluding the Jiulongwan section, show nearly identical �13Ccarbrofiles (Jiang et al., 2007; Zhou and Xiao, 2007; McFadden et al.,008; Sawaki et al., 2010). The cap dolostone of Member I is char-cterized by a negative �13Ccarb excursion (EN1) down to around4‰, with localized occurrence of extremely negative �13Ccarb val-es as low as −48‰ (Jiang et al., 2003b; Wang et al., 2008; Zhout al., 2010). Member II is characterized by a positive �13Ccarb excur-ion (EP1) with values up to +5‰. One or two negative �13Ccarbalues have been reported punctuating EP1 at Tianjiayuanzi (Chut al., 2003) and Jiulongwan (Sawaki et al., 2010; Zhu et al., 2011).e are uncertain whether such sporadic occurrences of negative

13Ccarb values are of secular or diagenetic origin. Zhu et al. (2007,011), however, believe that these negative �13Ccarb values havehemostratigraphic significance and they have named this negativexcursion WANCE (=Weng’an negative carbon isotope excursion).ANCE at Weng’an occurs at the mid-Doushantuo exposure sur-

ace. As discussed above, the correlation between Weng’an andhe Yangtze Gorges area is ambiguous: some correlate the mid-oushantuo exposure surface at Weng’an with the Members II–IIIoundary in the Yangtze Gorges area (Zhou and Xiao, 2007) andhus WANCE at Weng’an would be equivalent to EN2 in the Yangtzeorges area, whereas others correlate this exposure surface with aorizon within Member II in the Yangtze Gorges area (Zhu et al.,007, 2011), implying a negative �13Ccarb excursion within EP1.ven if one accepts the latter correlation, among the 11 Doushan-uo sections where WANCE has been identified by Zhu et al. (2007,011), only three actually have negative �13Ccarb values at WANCEthree data points at Weng’an, one at Xiangdangping, and two atiulongwan by Sawaki et al., 2010; the latter two are essentiallyhe same section). At all the other sections, WANCE is representedy positive �13Ccarb values. Thus, it remains to be tested whetherANCE is a truly negative excursion of chemostratigraphic signifi-

ance or just a kink within EP1.A more consistent negative �13Ccarb excursion (EN2, down to

7‰ and perhaps −9‰) occurs at the Members II–III boundary inhe Yangtze Gorges area (Jiang et al., 2007; Sawaki et al., 2010;

in et al., 2011; Zhu et al., 2011), followed by another positive13Ccarb excursion (EP2) in dolostone of lower Member III. Car-onates of upper Member III and carbonate nodules in Member IVre characterized by a strongly negative �13C excursion (EN3) with

h 192– 195 (2012) 125– 141

values as low as −10‰. The overall �13Ccarb chemostratigraphicprofile has been corroborated in multiple sections in the southernHuangling anticline although local variations exist (Condon et al.,2005; Jiang et al., 2007; McFadden et al., 2008; Lu et al., 2009; Yinet al., 2009, 2011; Sawaki et al., 2010; Zhu et al., 2011). Beyondthe southern Huangling anticline, �13Ccarb profiles of the Doushan-tuo Formation at the Zhangcunping and Weng’an sections, as wellas sections of the equivalent Lantian Formation in southern AnhuiProvince, are broadly similar to those near Jiulongwan (Zhou, 1997;Zhou and Xiao, 2007; Zhu et al., 2007; Zhao and Zheng, 2010; Yuanet al., 2011), but precise chemostratigraphic correlation is insecurebecause of the lack of independent stratigraphic markers. Majordeparture from the Jiulongwan profile has been described in deep-water slope and basinal sections, and these have been interpretedas evidence for major geochemical gradients in Ediacaran seawater(Jiang et al., 2007).

2.3. Biostratigraphy

The biostratigraphy of the Doushantuo Formation in SouthChina, and particularly in the Yangtze Gorges area, has simi-larly been a focus of previous studies (McFadden et al., 2009;Yin et al., 2009, 2011; Liu et al., 2011). Combining biostrati-graphic data from multiple sections in the Yangtze Gorges area,McFadden et al. (2009) recognized two possible acritarch bio-zones. The lower is dominated by the species Tianzhushania spinosa,whereas the upper is characterized by greater species evennessand includes abundant occurrence of Ericiasphaera rigida, T. spinosa,Eotylotopalla dactylos, Meghystrichosphaeridium “perfectum”, andTanarium conoideum. Yin et al. (2009, 2011) also recognizedtwo assemblages in the Doushantuo Formation in the YangtzeGorges area, with the lower assemblage dominated by the genusTianzhushania whereas the upper devoid of Tianzhushania but dom-inated by leiospheres, diverse acanthomorphs, and the tubularfossil Sinocyclocyclicus guizhouensis. Similarly, Liu et al. (2011) rec-ognized two assemblages in the Doushantuo Formation (the lowerT. spinosa assemblage and the upper Tanarium anozos–T. conoideumassemblage) and they also regarded that “the genus Tianzhushaniadoes not extend into the upper assemblage”. Thus, these studiesdiffer in whether Tianzhushania and T. spinosa extend to the upperbiozone. This discrepancy is a result of difference in methodol-ogy and stratigraphic correlation (see Fig. 1D–F for an example).First, McFadden et al.’s (2009) method was aimed at quantitativelytesting whether the lower and upper biozones – which are rec-ognized independent of biostratigraphic data, but on the basis ofsequence stratigraphic and chemostratigraphic correlation – aretaxonomically distinct. As emphasized by McFadden et al. (2009),their method was not designed to locate the optimal boundarybetween the two biozones. In contrast, Yin et al. (2009, 2011)started off defining the two assemblages by the presence/absenceof Tianzhushania, which may introduce circularity because thestratigraphic and environmental ranges of Tianzhushania has notbeen independently tested before it is used to define biostrati-graphic units. Thus, because McFadden et al. (2009) divided theirbiozones based on depositional cycles and chemostratigraphic fea-tures whereas Yin et al. (2009, 2011) separated their assemblagesbased on the presence/absence of Tianzhushania, the exact biozoneor assemblage boundaries may be different between the two stud-ies. Second, although McFadden et al. (2009), Yin et al. (2009, 2011),and Liu et al. (2011) all claimed that the two biozones/assemblagesoccur in lithostratigraphic units Members II and III and are asso-

ciated with EP1 and EP2, respectively, the correlation of theselithostratigraphic units and chemostratigraphic features beyondthe Yangtze Gorges area is different between these studies, result-ing in different conclusions with regard to the stratigraphic range

esearc

ossDe(a(pfzT(Wfmfw(

aptacYwTLTs2eabiiaeuotYYptIetw(beaattc(wuoniti

S. Xiao et al. / Precambrian R

f Tianzhushania. This pertains to the Weng’an and Zhangcunpingections, where acanthomorphs (including Tianzhushania and T.pinosa) occur in the upper Doushantuo Formation above a mid-oushantuo exposure surface (Xiao and Knoll, 1999; McFaddent al., 2009; Chen et al., 2010; Liu et al., 2011). McFadden et al.2009) correlated this exposure surface (and the associated neg-tive �13Ccarb excursion at Weng’an) with the II/III boundaryand associated EN2) at Jiulongwan on the basis of cyclostratigra-hy and chemostratigraphy, thus placing the Tianzhushania fossilsrom Weng’an and Zhangcunping acritarchs in their upper bio-one (Fig. 1E). In contrast, partly driven by the assumption thatianzhushania is restricted to their lower assemblage, Yin et al.2009, 2011) and Liu et al. (2011) followed Zhu et al.’s (2007, 2011)

ANCE concept and correlated the mid-Doushantuo exposure sur-ace at Weng’an and Zhangcunping with a poorly defined horizon in

iddle Member II at Jiulongwan, thus permitting the Tianzhushaniaossils from Weng’an and Zhangcunping acritarchs to be correlatedith the upper EP1 or the upper part of the lower assemblage

Fig. 1D).The biostratigraphic division proposed by Yin et al. (2009, 2011)

nd Liu et al. (2011) is appealing because of its simplicity: theresence/absence of a single genus (Tianzhushania) defines thewo assemblages. However, one needs to consider taphonomicnd environmental biases, as well as uncertainties in stratigraphicorrelation, when testing the biostratigraphic model proposed byin et al. (2009, 2011) and Liu et al. (2011). Certainly, at Jiulong-an and perhaps in the Yangtze Gorges area, Tianzhushania and

. spinosa are absent from Member III (McFadden et al., 2009;iu et al., 2011). But a more important question is whether theianzhushania-bearing strata above the mid-Doushantuo exposureurface at Weng’an, Zhangcunping, and Baokang (Xiao and Knoll,000; Yin et al., 2004; Zhou et al., 2007; McFadden et al., 2009; Chent al., 2010) can all be unambiguously correlated with Member IIt Jiulongwan. As discussed above, correlation of these sectionsased on sequence stratigraphic and chemostratigraphic data is

ndeed ambiguous. Furthermore, different taxa may have conflict-ng stratigraphic ranges at different sections due to environmentalnd taphonomic biases (Zhou et al., 2007). As an example, Yint al. (2009) claimed that S. guizhouensis first appears in theirpper assemblage; however, this taxon has so far been knownnly from the Weng’an biota (indeed first appearing right abovehe mid-Doushantuo exposure surface) and Member III in theangtze Gorges area (Xiao et al., 2000; Liu et al., 2008, 2009a).in et al.’s (2009) claim about S. guizhouensis would actually sup-ort a correlation of the Tianzhushania-bearing Weng’an biota withheir upper, not the lower, assemblage in the Yangtze Gorges area.ndeed, this correlation can better explain the taxonomic differ-nce between the Weng’an biota and the lower assemblage inhe Yangtze Gorges area: the latter is dominated by T. spinosahereas the former is not. Finally, even if one accepts Zhu et al.’s

2007, 2011) WANCE and sequence boundaries, the Tianzhushania-earing Weng’an and Zhangcunping biotas are restricted to Zhut al.’s sequence 2, whereas the Tianzhushania-dominated lowerssemblage in the Yangtze Gorges area (e.g., Jiulongwan, Tianji-yuanzi, and Wangfenggang) is mostly if not entirely restrictedo Zhu et al.’s sequence 1 (Liu et al., 2011). Thus, it is possibleo envision three Doushantuo assemblages when all evidence isonsidered (Fig. 1F): a lower assemblage dominated by T. spinosae.g., Member II in the Yangtze Gorges area), a middle assemblagehere T. spinosa is still present but become less abundant (e.g.,pper Doushantuo at Weng’an), and an upper assemblage with-ut T. spinosa (e.g., the upper assemblage of Liu et al., 2011). For

ow, we readily accept that the lower Doushantuo biozone is dom-

nated by the species Tianzhushania (McFadden et al., 2009). If inhe future it is confirmed that Tianzhushania exclusively occursn the lower Doushantuo Formation (Yin et al., 2009, 2011), our

h 192– 195 (2012) 125– 141 129

favored chemostratigraphic correlation of the NXF section to thelower Doushantuo at Jiulongwan would be further strengthenedby the dominance of Tianzhushania at NXF.

3. Methods

The NXF section is located at 30◦56.693′N, 111◦13.988′E (Fig. 1).It is the same as the lower ∼135 m of the Xiaofenghe section inYin et al. (2007) and Zhu et al. (2007), the lower ∼140 m of theXiaofenghe section in Jiang et al. (2011), and the northern hillsideXiaofenghe section of Zhu et al. (2011). The Xiaofenghe strato-columns in Yin et al. (2007, 2011), Liu et al. (2011), and Jiang et al.(2011) represent composite sections consisting of the NXF and SXF(southern Xiaofenghe section, 30◦55.945′N, 111◦13.266′E, Fig. 1),although splicing of these two sections is not straightforward.Because of the splicing problem, there are major inconsistencies inpublished stratigraphic thickness of the composite Xiaofenghe sec-tion, ranging from <100 m (Vernhet and Reijmer, 2010) to >200 m(Zhu et al., 2011). To avoid this problem and the stratigraphic rep-etitions resulting from modern landslides at the SXF section (Jianget al., 2011), we focused here only on the NXF section, which wascarefully measured and sampled at sub-meter scale (ca. 0.5 m inter-vals) for petrographic, paleontological, and geochemical analysis.

Petrographic thin sections were made for each sample andwere examined under a transmitted light microscope. Thin sectionexamination allowed us to quantitatively estimate the mineralog-ical (e.g., calcite, dolomite, silica, quartz, phosphate, and clay) andpetrological compositions (e.g., mudstone, wackestone, packstone,and grainstone). These data, along with sedimentary structuresobserved from field investigation, are plotted as three separatecolumns in the stratigraphic log (Fig. 2). All chert-phosphatic nod-ules in thin sections were thoroughly examined for microfossils(McFadden et al., 2009).

For �13Ccarb and �18Ocarb analysis, powders were micro-drilledfrom the finest-grained portions of polished slabs determined bypetrographic observation of corresponding thin sections. Approxi-mately 100 �g of carbonate powder was reacted for 10 min at 90 ◦Cwith anhydrous H3PO4 in a Multiprep inlet system connected toa GV Isoprime dual inlet mass spectrometer. Isotopic results areexpressed in the standard � notation as per mil (‰) deviationfrom V-PDB. Uncertainties determined by multiple measurementsof NBS-19 were better than 0.05‰ (1�) for both C and O isotopes.

Total organic carbon content (TOC) and �13Corg were deter-mined following procedures described by Kaufman et al. (2007).Approximately 1 g of whole rock powder was weighed and thenquantitatively digested with 6 M HCl. Residues were washed withdeionized water, centrifuged, isolated, dried, and weighed to deter-mine abundance of carbonate in the samples. TOC abundance(wt% of rock powder) was calculated from measured abundanceof carbon in the residues quantified by an elemental analyzer (EA)relative to the urea standard. Isotope abundances were determinedon a continuous flow GV Isoprime mass spectrometer. �13Corg val-ues are reported as per mil (‰) deviation from V-PDB. Uncertaintiesbased on multiple extraction and analyses of a standard carbonateare better than 0.15% (1�) for TOC and 0.3‰ (1�) for �13Corg.

Disseminated pyrite concentrations and �34Spy were analyzedusing both combustion and chromium reduction methods (Canfieldet al., 1986). For pyrite-rich samples, decalcified residues were sub-jected to direct EA combustion. For pyrite-poor samples, powderedsamples were reacted with 50 ml of 1 M CrCl2 and 20 ml of 10 MHCl in an N2 atmosphere. H2S produced from pyrite reduction by

CrCl2 was bubbled through a 1 M silver nitrate trap and precipitatedas Ag2S. The Ag2S was centrifuged, washed, dried, and weighed.Pyrite sulfur isotopes and concentrations were determined by EAcombustion at 1030 ◦C on a continuous flow GV Isoprime mass

130 S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141

F rvalsw e of mc igrap

sbwaWnwhypmbc

f(ttict

ig. 2. Lithostratigraphic log of the NXF section. Section was logged at 0.5 m inteackestone; P, packstone; G, grainstone. The left edge shows dominant percentag

olumn shows sedimentary structures, and the right edge dominant lithology. Strat

pectrometer. 100–500 �g of decalcified residue (for direct com-ustion samples) or 100 �g Ag2S (for chromium reduction samples)as dropped into a quartz reaction tube packed with quartz chips

nd elemental copper for quantitative oxidation and O2 resorption.ater was removed from combustion products with a 10-cm mag-

esium perchlorate trap, and SO2 was separated from other gasesith a 0.8 m PTFE GC column packed with Porapak 50–80 mesheated at 90 ◦C. Pyrite concentrations were calculated from SO2ields and reported as wt% pyrite. �34Spy values are reported aser mil (‰) deviation from V-CDT. Uncertainties determined fromultiple analyses of NBS 127 interspersed with the samples are

etter than 0.3% (1�) for concentration and 0.3‰ (1�) for isotopeomposition.

Extraction of CAS was conducted using techniques modifiedrom Burdett et al. (1989), Gellatly and Lyons (2005), and Shen et al.2008). Roughly 100 g of sample was leached with 10% NaCl solu-ion, washed, dried, powdered, and weighed. Dried powders were

reated with 30% hydrogen peroxide for 48 h to remove dissem-nated pyrite and organically bound sulfur that could potentiallyontaminate CAS extractions. Leached powders were quantita-ively dissolved in 3 M HCl and insoluble residues were separated

and drafted using the Dunham classification (Dunham, 1962). M, mudstone; W,ineralogy between calcite, dolomite, phosphate, and terrigenous clay. The center

hic datum at base of the Doushantuo Formation.

from the supernatant by gravity filtration through 8 �m What-man filters. Total volume of supernatant was measured, and 10 mlof 0.1 M BaCl2 was added to 40 ml of supernatant to precipitatebarite. The precipitates were then centrifuged, washed, dried, andweighed.

CAS concentrations were calculated from ICP–AES (inductivelycoupled plasma atomic emission spectrometer) analysis of smallaliquots of the leached solutions at Virginia Tech Soil Testing Labo-ratory. CAS concentrations were calculated from sulfur abundancesand reported as ppm SO4

2− in carbonate, with uncertainties bet-ter than 5%. Barite precipitates were combusted with a EurovectorEA and 34S/32S ratios were determined using a continuous flow GVIsoprime mass spectrometer. �34SCAS values are reported as per mil(‰) deviation from V-CDT. Uncertainties determined from multipleanalyses of NBS 127 are better than 0.3‰ (1�).

For strontium isotope analyses, micro-drilled carbonate pow-ders (ca. 5 mg) were repeatedly leached (3×) with ultrapure 0.2 M

ammonium acetate (pH ∼ 8.2) to remove exchangeable Sr from noncarbonate minerals (Montanez et al., 1996) and then treated withultrapure 0.5 M acetic acid for 12 h. The supernatant was separatedfrom insoluble residues by centrifugation, decanted, dried, and

S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141 131

Fig. 3. Sedimentary structures. (A) Field photograph of tepee-like structure in cap carbonate at ∼0.5 m. (B) Field photograph of quartz-filled sheet crack structures in capcarbonate at ∼1.5 m. (C and D) Thin section photomicrographs of chalcedony quartz and calcite filling sheet crack structures at 2.5 m, under non-polarized and cross-p microc ograpc

slstHRulSst7l(

4

ot

olarized transmitted light, respectively. Thin section XF-2.5. (E) Thin section photoross-polarized transmitted light. Thin section XF-21.9. (F) Thin section photomicrross-polarized transmitted light. Thin section XF-50.8.

ubsequently dissolved with 200 �l of 3 M HNO3. Small polyethy-ene columns were prepared with quartz wool and ∼1 cm of Srpec resin and washed with 400 �l of 3 M HNO3. After loading,he sample was sequentially eluted with 3 M HNO3 (200 �l), 7 MNO3 (600 �l) and 3 M HNO3 (100 �l) to primarily remove Ca andb, and then 400 �l of 0.05 M HNO3 was passed through the col-mn to collect the Sr fraction. The samples were then dried and

oaded with 0.7 �l of TaO onto a Re filament for analysis in a VGector 54 thermal ionization mass spectrometer at the Univer-ity of Maryland. Sr beams of 0.5 V on mass 88 were obtained atemperatures between 1450 ◦C and 1650 ◦C allowing for at least5 stable ratios. Repeated analysis of NBS 987 during the ana-

ytical session yielded an average value of 0.710238 ± 0.0000061�).

. Lithostratigraphy

Overall, the NXF section (Fig. 2) contains greater abundancesf carbonate, phosphatic clasts, silt and sand, but less shalehan the Jiulongwan section. An important marker bed of the

graph of phosphatic siltstone at 21.9 m. Phosphatic grains in total extinction underh of phosphatic wackestone at 50.8 m. Phosphatic grains in total extinction under

Doushantuo Formation, the basal Doushantuo cap dolostone(Member I), is present at the NXF section and overlies diamictiteof the Cryogenian Nantuo Formation. The cap dolostone at NXF isvery similar to Ediacaran cap dolostones described elsewhere inSouth China: it is a 3.9 m thick massive-bedded microlaminateddolomicrite, with peloids, abundant tepee-like structures (Fig. 3A),and quartz-filled sheet cracks (Fig. 3B–D). The crest of the tepee-like structures strikes at 90–110◦, consistent with measurementstaken at Jiulongwan. The upper part of the cap dolostone becomesthinner-bedded, more limey, and partially silicified. The cap dolo-stone passes into a 20-m thick organic-rich shale interbeddedwith fine-grained phosphatic siltstone (Fig. 3E). Phosphatic grainsare silt to fine sand-size, subangular to rounded, well sorted, andoccur in thin lenses and layers 1–3 cm thick. Small wave rippleswere observed at the base of this unit around 4.1 m above theformation. Thin dolostone interbeds also occur locally and con-

tain abundant pea-size chert nodules. The phosphatic componentincreases upsection from around 10% at the base of the unit togreater than 50% near the top. In outcrop, the phosphatic lensesincrease in density and thickness up to 5–7 cm thick.

132 S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141

Fig. 4. Chert nodules and microfossils. (A) Field photograph of chert nodules (black color) at 98.5 m. (B–H) Thin section photomicrographs under non-polarized light. (B) Chertnodule (brown color) at 96.2 m. Thin section XF-96.2. (C–E) Tianzhushania spinosa at 65.8 m. (D and E) Successively enlarged views (arrows in C and D) to show multilaminatemembrane and cylindrical processes characteristic of T. spinosa. Thin section XF-65.8. (F–H) Tianzhushania spinosa at 111.8 m. (G and H) Enlarged views (black and whitea istic ot

dltCtuTi

rrows in G) to show multilaminate membrane and cylindrical processes characterhis figure legend, the reader is referred to the web version of the article.)

Overlying the siltstone is a thick unit composed of chertyolomitic mudstone and wackestone (25–106 m). Dolostones are

ocally argillaceous, thin to medium bedded, and contain minoro moderate amounts of quartz and phosphate grains (Fig. 3F).hert nodules (Fig. 4A and B) are 1–5 cm in diameter, spheroidal

o oblong in shape and oriented parallel to bedding. Chert nod-les contain abundant acritarchs (Fig. 4C–E) and phosphatic grains.he abundance of phosphatic grains increases upsection, althought is lower than the underlying phosphatic siltstone. A sharp

f T. spinosa. Thin section XF-141.5. (For interpretation of the references to color in

lithofacies change occurs at ∼96 m, where phosphatic-dolomiticwackestone–packstone gives way to dolomitic mudstone and thenlime mudstone that dominate the rest of the section (96–140 m);this change may represent a flooding surface. The uppermost partof the NXF section (106–140 m) was measured along a satellite sec-

tion offset from the main section. Splicing of this satellite sectionwith the main section resulted in an approximate 5 m discrepancybetween our stratigraphic thickness measurements and those pub-lished in Yin et al. (2007) and Zhu et al. (2007, 2011). This unit is

S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141 133

F XiaofM ngyin

clu(

s2DchaDarJsDf

ig. 5. Biostratigraphic distribution of Ediacaran acanthomorphic acritarchs at thecFadden et al., 2009). NT, Nantuo Formation; DST, Doushantuo Formation; DY, De

haracterized by massive limestone with thin (possibly microbial)aminations interbedded with argillaceous limestone. Cherts in thisnit are oblong to banded, and contain well preserved acritarchsFig. 4F–H).

The upper NXF section is not exposed above 140 m in thetudy area. Others (e.g., Yin et al., 2007; Zhu et al., 2007,011) continued their stratigraphic measurements up to theoushantuo–Dengying boundary along the SXF section, which isomplicated by many faults/landslides and is inappropriate forigh-resolution chemostratigraphic sampling. Yin et al. (2007)nd Zhu et al. (2007) placed a black shale unit in the uppermostoushantuo Formation in their SXF stratocolumns, which weredopted by McFadden et al. (2009). This black shale unit would beegarded as equivalent to Doushantuo Member IV recognized at the

iulongwan section. However, we were unable to verify this blackhale unit at SXF (see also Zhu et al., 2011). Instead, the Doushantuo-engying boundary at SXF is represented by the transition

rom Doushantuo oolitic grainstone/packstone to Dengying cherty

enghe (NXF and SXF) and Jiulongwan sections (Yin et al., 2007; Zhou et al., 2007;g Formation.

dolostone with dissolution structures. It is possible that this blackshale unit is missing because of bedding-parallel faults, but it ismore likely that a facies change led to the disappearance of theblack shale in shallow water environments such as at the Weng’ansection (Zhou et al., 2005; Jiang et al., 2011).

Comparison between the NXF and Jiulongwan section suggeststhat the former is dominated by dolomitic phosphatic wackestoneand packstone whereas the latter by argillaceous mudstone, whichis consistent with the paleobathymetric inference that the NXFsection was more proximal than the Jiulongwan section (Jianget al., 2011). Although the Nantuo-Doushantuo boundary is unam-biguously identifiable at NXF, the three sedimentary cycles arenot easily recognizable because the exposure surfaces are poorlydeveloped. Sequence boundaries or erosional surfaces have been

marked on Xiaofenghe stratocolumns at ∼57 m (Fig. 2 in Yin et al.,2011; Liu et al., 2011), ∼62 m (Fig. 3 in Yin et al., 2011), ∼74 m(Jiang et al., 2011), and ∼88 m (Yin et al., 2007; Zhu et al., 2007,2011). These surfaces are all based on a single possible erosional

134 S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141

Fig. 6. Chemostratigraphic profiles of the NXF section. In the �34Spy and �34SCAS-py profiles, solid symbols represent samples with >2 mg Ag2S yield in CRS extraction (roughlyequivalent to >0.02% pyrite). In the 87Sr/86Sr profile, solid and empty symbols represent limestone and dolostone samples, respectively.

S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141 135

-15

-10

-5

0

5

10

-15 -10 -5 0 5

δ13C

carb

(‰ V

PB

D)

δ18Ocarb (‰ VPDB)

y = 0.9655x + 28.973(R² = 0.9423)

15

20

25

30

35

40

-10 -5 0 5 10

Δδ13

Cca

rb-o

rg (‰

VP

DB

)

δ13 Ccarb (‰ VPDB)

A BKnauth and Kennedy (2009)NXF (35-104 m)NXF (4-35 m)NXF (0-3.9 m cap dolostone)

0

1

2

3

4

5

6

7

8

0 2 4 6 8

Spy

(Wt %

)

Corg (wt %)

0

0.1

0.2

0.3

0.4

0.5

0 0.5 1 1.5 2

Longe (Li et al., 2010)Minle (Li et al., 2010)Zhongling (Li et al., 2010)Jiulongwan lower DST (0- 70 m) (McFadden et al., 2008)Jiulongwan upper DST (70- 155 m) (McFadden et al., 2008)Jiulongwan lower DST (0- 70 m) (Li et al., 2010)Jiulongwan upper DST (70- 155 m) (Li et al., 2010)NXF (This Study)

C

Fig. 7. Geochemical crossplots. (A) �13Ccarb vs �13Ccarb-org, showing positive correlation with a slope of nearly 1.0. (B) �13Ccarb vs �18Ocarb, showing poor correlation for mucho alterats with tp erted

sasNeeuiFtaldlMtaIic

f the Doushantuo Formation at NXF (35–140 m) and an offset from the meteoric

howing relatively low TOC, pyrite content, and Spy/Corg ratio at NXF as compared

lot. Pyrite contents reported in McFadden et al. (2008) and Table 1 have been conv

urface identified at SXF and correlated to NXF strata, and the vari-tion in stratigraphic height is an indication of the difficulty inplicing the NSF and SXF sections. Our stratigraphic analysis ofXF identifies a possible flooding surface at ∼96 m. However, norosional unconformity has been identified at NXF (see also Zhut al., 2011). Similarly, because of the absence of the black shalenit in the uppermost Doushantuo Formation, the surface separat-

ng the second and third cycles is also ambiguous at Xiaofenghe.rom the lithostratigraphic and cyclostratigraphic point of view,he cap carbonate and overlying organic-rich mudstone/siltstonend phosphatic wackestone/packstone (0–96 m) could be equiva-ent to the first cycle (Members I and II) at Jiulongwan, whereas theolomitic mudstone and overlying limestone (96–140 m) equiva-

ent to dolostone and overlying ribbon rocks of the second cycle orember III at Jiulongwan (Fig. 5; see also Jiang et al., 2011). Alterna-

ively, Zhu et al. (2007, 2011) propose that the sequence boundary

t ∼88 m in SXF could be correlated with a horizon within MemberI at Jiulongwan, implying that much if not all of the NXF sections equivalent to Member II at Jiulongwan. The biostratigraphic andhemostratigraphic data, reported below, indicate that the entire

ion trend identified in Knauth and Kennedy (2009). (C) Corg (wt%) vs Spyrite (wt%),he Jiulongwan section. Inset shows a close-up view of the lower left corner of the

to Spyrite contents.

NXF section is more likely correlative to the lower Doushantuo,consistent with Zhu et al.’s (2011) correlation. This correlation indi-cates that lithostratigraphic differences between the upper NXF andits potential correlative at Jiulongwan could be related to facieschanges between the inner shelf lagoon and nearshore subtidalcarbonates (Jiang et al., 2011).

5. Biostratigraphy

Acritarch biostratigraphy of the Doushantuo Formation atXiaofenghe has been previously published (Yin et al., 2007;McFadden et al., 2009). Yin et al. (2007) presented acritarch occur-rence data from the lower Doushantuo Formation at NXF and upperDoushantuo Formation at SXF. As discussed above, McFaddenet al. (2009) and Yin et al. (2009, 2011) recognized two potentialacritarch biozones in the Doushantuo Formation. Although there

are some differences between the two studies, both recognizedthat the lower biozone is dominated by Tianzhushania, particu-larly T. spinosa. The dominance of T. spinosa in the NXF section(McFadden et al., 2009; Yin et al., 2009), from ∼35 m to ∼110 m

136 S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141

Fig. 8.

esearc

(la

6s

psJaaSt−u−s+f��+va>s�s

o(22easiGt�m�o�spsmt(mtrSiai

FiJlt

S. Xiao et al. / Precambrian R

Figs. 4C–H and 5), indicates a biostratigraphic assignment to theower biozone and correlation to the lower Doushantuo Formationt Jiulongwan.

. Chemostratigraphy and correlation with the Jiulongwanection

Chemostratigraphic data are presented in Table S1 (see Sup-lemental Online Material) and plotted in Fig. 6. The data showeveral important features that are useful in correlation with theiulongwan section. First, the �13Corg profile is remarkably constantround −28.8‰, despite �13Ccarb variation up to 10‰, resulting in

strong correlation between �13Ccarb and �13Ccarb-org (Fig. 7A).econd, the �13Ccarb profile can be divided into three segments: (1)he cap dolostone (0–4 m), characterized by negative values around3‰; (2) the overlying 31 m strata (4–35 m), where �13Ccarb val-es hover around 0‰ (with three exceptions, −9.6‰ at 25.5 m,6.4‰ at 29.2 m, and +3.9‰ at 26.0 m); and (3) the rest of the

ection (35–140 m) with positive values mostly around +5‰ to6‰. This profile is similar to previously reported �13Ccarb datarom NXF (Liu et al., 2011; Yin et al., 2011; Zhu et al., 2011). Third,34Spy values are all positive and vary between +14.2‰ and +32.6‰.34SCAS values are also positive with irregular variation between6.6‰ and +39.1‰. �34SCAS-py varies from −6.1‰ to 16‰. Theariability in some intervals likely reflects diagenetic processes,lthough if only �34Spy values of pyrite-rich samples (those with2 mg Ag2S yield from chromium reduction extraction) are con-idered, sample-to-sample stratigraphic scattering in �34Spy and

34SCAS-py is reduced (Fig. 6). Finally, 87Sr/86Sr values of limestoneamples range between 0.7079 and 0.7081.

There has been extensive discussion on the diagenetic alterationf Proterozoic carbon, sulfur, and strontium isotope signaturesKaufman and Knoll, 1995; Lyons et al., 2004; Gellatly and Lyons,005; Halverson et al., 2007; Marenco et al., 2008; Shen et al.,008, 2011; Knauth and Kennedy, 2009; Derry, 2010; Halversont al., 2010). We interpret the �13C and �34S of the NXF sections representing mainly primary signatures. This interpretation isupported by the well defined stratigraphic trends (particularlyn �13Ccarb and �13Corg) and regional consistency in the Yangtzeorges area (when compared with the Jiulongwan section). Fur-

hermore, there is very little correlation between �13Ccarb and18Ocarb data from NXF (Fig. 7B), and these data plot outside theeteoric diagenetic trend (Knauth and Kennedy, 2009); at least

13Ccarb data from 35 to 104 m are not strongly altered by mete-ric diagenesis, although the primary origin of a few negative13Ccarb values at 25–30 m may be questioned because of the largeample-to-sample variation (i.e., stratigraphic inconsistency). NXFyrite is mostly disseminated and likely of authigenic origin. Eveno, the ultimate source of sulfur for pyrite precipitation came fromarine sulfate. The highly positive �34Spy values from the NXF sec-

ion are similar to those reported from other Ediacaran successionsShen et al., 2008, 2011; Ries et al., 2009) and may reflect unique

arine geochemistry of Ediacaran oceans. The strong sample-o-sample variation in �34SCAS and �18Ocarb time series likelyeflects variable degrees of alteration, although the range of sulfate

isotope compositions may also reflect the low sulfate availabil-ty in Ediacaran oceans (McFadden et al., 2008). 87Sr/86Sr ratiosre susceptible to diagenetic processes such as dolomitization andnfluence of detrital sediments. Thus, we deem that the 87Sr/86Sr

ig. 8. Comparison of chemostratigraphic profiles of the NXF and Jiulongwan sections, wison, corresponding chemostratigraphic profiles are plotted at the same scale. Highly radiulongwan were regarded as altered values (Sawaki et al., 2010) and are omitted from thocation) (Yang et al., 1999) are scaled to the thickness of the Jiulongwan section based onhe references to color in this figure legend, the reader is referred to the web version of th

h 192– 195 (2012) 125– 141 137

ratios from limestone samples more faithfully reflect the deposi-tional values.

Consistent with the biostratigraphic match between sections,chemostratigraphic profiles from NXF show similarities to thosefrom the lower Doushantuo Formation at Jiulongwan. Carbon iso-tope trends of the NXF section can be readily correlated withthose from the lower Doushantuo Formation (Members I and II)at Jiulongwan (McFadden et al., 2008). �13Corg trends are nearlyidentical between the two sections. The negative �13Ccarb excursionin the cap dolostone and the positive excursion at 35–140 m canbe matched with, respectively, EN1 in Doushantuo Member I andEP1 identified in Doushantuo Member II. �13Ccarb values at 5–35 mare also similar to the average �13Ccarb values in lower MemberII at Jiulongwan and other sections on the southern limb of theHuangling anticline despite local variations (Jiang et al., 2007; Zhouand Xiao, 2007; Zhu et al., 2007, 2011; McFadden et al., 2008; Yinet al., 2009, 2011; Sawaki et al., 2010). The negative �13Ccarb excur-sion at 25–30 m in the NXF section is only defined by two data pointsand could represent diagenetic alteration, although they may becorrelated with a brief negative �13Ccarb excursion, also defined byone or two data points, recovered from the lower Doushantuo For-mation (around 58 m from the base) in a drill core near Jiulongwan(Sawaki et al., 2010) and an equivalent horizon at Tianjiayuanzi inthe Yangtze Gorges area (Chu et al., 2003; Zhou and Xiao, 2007). Atthe present, we are inclined to interpret these negative values asdiagenetic artifacts with little chemostratigraphic significance.

�34Spy values at NXF are unusually positive, again similarto values from the lower Doushantuo Formation at Jiulongwan,although �34Spy and �34SCAS profiles at NXF show more consis-tent stratigraphic trends and less stratigraphic scattering than thelower Doushantuo Formation at the Jiulongwan section. The NXFsection also has higher �34Spy values but lower �34SCAS values,resulting in lower �34SCAS-py values relative to the lower Doushan-tuo Formation at Jiulongwan. From a chemostratigraphic pointof view, it is important to note that the prominent decrease inboth �34Spy and �34SCAS observed in the upper Doushantuo For-mation at Jiulongwan (arrows in Fig. 8) (McFadden et al., 2008;Li et al., 2010) is not seen at NXF, further strengthening theview that only the lower Doushantuo Formation is represented atNXF.

As a critical test of these regional correlations, we evaluatedvariations in strontium isotope signatures in limestone samplesthroughout the NXF section. Previous studies (Yang et al., 1999;Jiang et al., 2007; Sawaki et al., 2010) revealed a strong isotopic con-trast between lower and upper Doushantuo Formation limestonesat multiple sections in the Yangtze Gorges area. 87Sr/86Sr valuesof best-preserved samples rise from ca. 0.7078 to 0.7089 from thebottom to top of the formation (Fig. 8), a transition consistent withdata derived from other Ediacaran successions (Shields and Veizer,2002; Halverson et al., 2010). This rise cannot be explained by anupsection increase in the content of clays that contain more radio-genic Sr, because overall the lower Doushantuo is more argillaceousthan the upper Doushantuo and the highest 87Sr/86Sr values arefound in Member III limestone with little clay (McFadden et al.,2008; Sawaki et al., 2010). Our analyses (Figs. 6 and 8) of limestone

samples near the top NXF section are consistently low (rangingfrom 0.7079 to 0.7081). Matched with the trend defined by high Sr-content limestone samples from previous studies (including coresamples analyzed by Sawaki et al., 2010), the upper NXF samples are

th proposed chemostratigraphic correlation (yellow shading). To facilitate compar-iogenic 87Sr/86Sr values (0.709–0.713) from the cap carbonate at a drill core near

e Jiulongwan 87Sr/86Sr profile. 87Sr/86Sr data from the Tianjiayuanzi (see Fig. 1C for �13C chemostratigraphic correlation (Zhou and Xiao, 2007). (For interpretation ofe article.)

138 S. Xiao et al. / Precambrian Research 192– 195 (2012) 125– 141

Table 1Statistics of chemostratigraphic data from Doushantuo strata representing the �13Ccarb feature EP1 at NXF (this study) and Jiulongwan (Jiang et al., 2007; McFadden et al.,2008; Li et al., 2010). Statistics of 87Sr/86Sr data at Jiulongwan also includes data from Tianjiayuanzi (Yang et al., 1999) and a drill core near Jiulongwan (Sawaki et al., 2010).

NXF (34–140 m)(this study)

Jiulongwan (25.7–66 m) (Jiang et al.,2007; McFadden et al., 2008)

Jiulongwan (29–65 m) (Liet al., 2010)

TOC (wt%) 0.58 ± 0.39 (n = 163) 1.44 ± 0.81 (n = 52) 0.87 ± 0.21 (n = 9)Pyrite (wt%) 0.03 ± 0.01 (n = 44) 0.79 ± 1.08 (n = 40) 1.22 ± 0.70 (n = 6)Spy/Corg 0.04 ± 0.02 (n = 42) 0.3 ± 0.4 (n = 41) 0.6 ± 0.6 (n = 9)CAS (ppm) 310 ± 557 (n = 159) 591 ± 710 (n = 50) 131 ± 133 (n = 18)�13Corg (‰) −28.8 ± 0.6 (n = 163) −29.8 ± 0.4 (n = 52) −29.9 ± 0.3 (n = 20)�13Ccarb (‰) 5.5 ± 1.3 (n = 161) 4.8 ± 1.2 (n = 68) 4.0 ± 1.9 (n = 20)�13Ccarb-org 34.2 ± 1.3 (n = 155) 35.5 ± 0.7 (n = 17) 33.8 ± 2.0 (n = 20)�34Spy (‰) 24.6 ± 3.2 (n = 44) 18.3 ± 9.0 (n = 40) 7.0 ± 9.2 (n = 9)�34SCAS (‰) 28.8 ± 4.5 (n = 110) 32.5 ± 9.9 (n = 36) 38.0 ± 5.8 (n = 18)�34SCAS-py 3.3 ± 4.2 (n = 27) 12.7 ± 9.0 (n = 25) 28.9 ± 8.0 (n = 9)87Sr/86Sr 0.708004 ± 0.000071 (n = 5) 0.708028 ± 0.000156 (n = 24)

Fig. 9. Paleographic reconstruction (A; box diagram modified from Jiang et al., 2011) and stratified redox ocean model (B; cross section modified from Li et al., 2010), with thea el, euxt

cs

DvttOGert

pproximate location of Doushantuo sections marked on the diagrams. In this modhe open shelf and to deep oceans.

learly equivalent to those in the lower Jiulongwan strata, furtherupporting our correlation.

The established isotopic contrasts between the lower and upperoushantuo Formation at Jiulongwan are striking. The upper inter-al contains a carbon isotope anomaly (EN3) known globally ashe Shuram event, named after a long-lived negative carbon iso-ope excursion to values near −10‰ recorded in drill core fromman (Burns and Matter, 1993; Le Guerroue et al., 2006; Le

uerroue and Cozzi, 2010; Bergmann et al., 2011; Grotzingert al., 2011; Verdel et al., 2011). This biogeochemical anomaly isecorded in upper Member III and Member IV of the Doushan-uo Formation, and stratigraphic data from these intervals also

inia is restricted to the shelf lagoon. It remains uncertain whether euxinia extends

record simultaneous depletion of 34S in both sulfides and sul-fates (McFadden et al., 2008). Neither these significant anomalies,nor the high 87Sr/86Sr values, are recorded in samples from NXF.Thus, chemostratigraphic and biostratigraphic data indicate thatthe 140-m-thick NXF section is correlated with the ∼70-m-thicklower Doushantuo Formation at Jiulongwan (highlighted in Fig. 8),implying that the sedimentation rate at NXF was about twice that atJiulongwan, probably because NXF was closer to a carbonate factory

and detrital sediment sources. This correlation also suggests thatthe lithostratigraphic similarity between the dolostone–limestoneunit (96–140 m) at NXF and Member III at Jiulongwan does notimply chronostratigraphic equivalency.

esearc

7

DonacbaesFmspom

dmiettncttdiecaaiPe2tgb

NoalvTgNb(it(�(sttectFt

S. Xiao et al. / Precambrian R

. Test of Ediacaran ocean redox models

Li et al. (2010) proposed a stratified redox model for theoushantuo basin. In this model, the Doushantuo basin consistedf an oxic surface ocean that was underlain by a thin ferrugi-ous layer, a wedge of euxinic water resting on an open shelf,nd a deep ferruginous ocean. This model was derived from geo-hemical data from successions deposited in bathymetrically deepasinal ferruginous setting (Zhongling, Minle, and Longe sections)nd intermediate-depth euxinic setting (Jiulongwan section). How-ver, geochemical data from the Minle and Longe sections wereparse, and no data from proximal shallow facies were available.urther, the Zhongling section, previously interpreted to have accu-ulated on a distal ramp (Li et al., 2010), is now reinterpreted as

edimentation on the lagoonal side of a shelf margin shoal com-lex (Jiang et al., 2011; Zhu et al., 2011). This reinterpretationf the basin architecture has implications for the stratified redoxodel.Importantly, the basin architecture reinterpretation and our

ata from NXF indicate that both the NXF and Zhongling sectionsay have been deposited beneath non-euxinic (oxic or ferrug-

nous) waters in shallower environments surrounding a deeperuxinic lagoon, which developed shortly after the deposition ofhe Doushantuo cap dolostone (Jiang et al., 2011). The interpre-ation that the NXF sediments appear to have been deposited inon-euxinic waters is supported by their lower TOC and pyriteontents (Table 1) and lower Spyrite/Corg ratios as compared withhe lower Doushantuo Formation at Jiulongwan (Fig. 7C), althoughhis interpretation needs to be further tested by iron speciationata from NXF. A non-euxinic environment for the NXF section

s consistent with predictions of the Li et al. model. However, theuxinic condition at Jiulongwan spatially bracketed by non-euxiniconditions at NXF (this study) and Zhongling (Li et al., 2010) canlternatively be interpreted in the framework of a recent basinrchitecture reconstruction (Jiang et al., 2011): euxinia developedn a restricted lagoon and may have come and gone in the Ediacaraneriod (Fig. 9). This restricted euxinic lagoon is different in geom-try from the euxinic wedge that rested on an open shelf (Li et al.,010). More high-resolution chemostratigraphic data are needed toest whether euxinia also occurred in shelf margin and whether thelobal ocean was stratified with ferruginous/euxinic deep waterselow oxic surface waters.

In detail, some of the chemostratigraphic differences betweenXF and Jiulongwan can also be explained in the frameworkf the basin architecture presented by Jiang et al. (2011). First,lthough the NXF section and lower Doushantuo Formation at Jiu-ongwan have remarkably invariant �13Corg profiles, the �13Corg

alues are slightly greater at NXF than at Jiulongwan (Table 1).his subtle difference may indicate that photosynthesis had areater contribution to organic production at the non-euxinicXF section, whereas microbial recycling of aged organic car-on played a greater role in euxinic conditions at JiulongwanFike et al., 2006; McFadden et al., 2008). This interpretations also consistent with slightly greater �13Ccarb values at NXFhan Jiulongwan during the proposed EP1 positive excursionTable 1). Second, stratigraphic scattering observed in the �13Ccarb,34Spy, and �34SCAS profiles is greater at Jiulongwan than NXFFig. 8; Table 1). The most likely cause of this difference ishort-term perturbations related to sea-level and chemocline fluc-uations, which affected the remineralization of deep-water DOChrough bacterial sulfate reduction (Rothman et al., 2003; Lit al., 2010); the Jiulongwan section located at or just below the

hemocline would have been more affected by such fluctuationshan the NXF section located persistently above the chemocline.inally, the smaller �34SCAS-py values at NXF as compared withhe Jiulongwan section (Table 1) can be readily explained by

h 192– 195 (2012) 125– 141 139

authigenic pyrite formation at NXF (vs syngenetic pyrite forma-tion at Jiulongwan) where diffusion of seawater sulfate to porewaters limited the isotopic fractionation between sulfate andpyrite.

8. Conclusion

This study highlights the necessity of integrating basin archi-tecture, biostratigraphy, and chemostratigraphy when correlatingEdiacaran strata and reconstructing environmental changes inEarth’s distant past. Our biostratigraphic and chemostratigraphicdata highlight the potential of acanthomorph acritarchs and car-bonate carbon isotopes in Ediacaran correlation (Moczydłowska,2005; Willman and Moczydłowska, 2008; Vorob’eva et al., 2009;McFadden et al., 2009; Liu et al., 2011; Sergeev et al., 2011; Zhuet al., 2011), and they show that only lower Doushantuo Forma-tion is represented at the NXF section. A positive identification ofthe upper Doushantuo Formation at the SXF section remains to beverified by integrated stratigraphic data. Geochemical data indi-cate that the NXF sediments were laid down in non-euxinic waters.Considered with published basin architecture reconstruction (Jianget al., 2011; Zhu et al., 2011), the new data is consistent with eux-inia in a shelf lagoon bracketed by non-euxinic waters in the innershelf (e.g., NXF) and outer shelf margin (e.g., Zhongling), at leastduring the early Doushantuo time following the deposition of thecap dolostone. More data are needed to further test whether a eux-inic wedge, which appears to be an emerging feature in the redoxstructures of late Archean to early Neoproterozoic oceans (Reinhardet al., 2009; Johnston et al., 2010; Kendall et al., 2010; Poulton et al.,2010; Poulton and Canfield, 2011), also characterizes the Ediacaranoceans; and if so, whether the Ediacaran euxinic lens or wedge wascontrolled by oceanic factors such as a sulfate gradient (Li et al.,2010), paleogeographic factors such as basin restriction (Jiang et al.,2011), or a combination of both. Such a test requires geochemicaldata from Ediacaran successions beyond the Yangtze Gorges areaand beyond South China.

Acknowledgements

This research was supported by National Science Founda-tion, NASA Exobiology and Evolutionary Biology Program, ChineseAcademy of Sciences, National Natural Science Foundation of China,Chinese Ministry of Science and Technology, and the GuggenheimFoundation. We would like to thank Jinlong Wang for field assis-tance, James Farquhar for access to CRS extraction facilities, NatalieSievers, Benn Breeden, Rebecca Ohly, and Susan Drymala for techni-cal assistance, and two anonymous reviewers for critical but usefulcomments.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.precamres.2011.10.021.

References

Ader, M., Macouin, M., Trindade, R.I.F., Hadrien, M.-H., Yang, Z., Sun, Z., Besse, J., 2009.A multilayered water column in the Ediacaran Yangtze platform? Insights fromcarbonate and organic matter paired �13C. Earth Planet. Sci. Lett. 288, 213–227.

Barfod, G.H., Albaréde, F., Knoll, A.H., Xiao, S., Télouk, P., Frei, R., Baker, J., 2002. NewLu–Hf and Pb–Pb age constraints on the earliest animal fossils. Earth Planet. Sci.Lett. 201, 203–212.

Bergmann, K.D., Zentmyer, R.A., Fischer, W.W., 2011. The stratigraphic expression of

a large negative carbon isotope excursion from the Ediacaran Johnnie Formation,Death Valley. Precambrian Res. 188, 45–56.

Bristow, T.F., Kennedy, M.J., Derkowski, A., Droser, M.L., Jiang, G., Creaser, R.A., 2009.Mineralogical constraints on the paleoenvironments of the Ediacaran Doushan-tuo Formation. Proc. Natl. Acad. Sci. U.S.A. 106, 13190–13195.

1 esearc

B

B

C

C

C

C

C

D

D

D

D

F

G

G

H

H

I

J

J

J

J

J

J

K

K

K

K

K

K

L

L

40 S. Xiao et al. / Precambrian R

urdett, J., Arthur, M., Richardson, M., 1989. A Neogene seawater sulfur isotope agecurve from calcareous pelagic microfossils. Earth Planet. Sci. Lett. 94, 189–198.

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

anfield, D.E., Raiswell, R., Westrich, J.T., Reaves, C.M., Berner, R.A., 1986. The useof chromium reduction in the analysis of reduced inorganic sulfur in sedimentsand shale. Chem. Geol. 54, 149–155.

ao, R., Tang, T., Xue, Y., Yu, C., Yin, L., Zhao, W., 1989. Research on Sinian Stratawith ore deposits in the Yangzi (Yangtze) region, China. In: Nanjing Instituteof Geology and Palaeontology (Ed.), Upper Precambrian of the Yangzi (Yangtze)Region, China. Nanjing University Press, Nanjing, pp. 1–94.

hen, S., Yin, C., Liu, P., Gao, L., Tang, F., Wang, Z., 2010. Microfossil assemblagefrom chert nodules of the Ediacaran Doushantuo Formation in Zhangcunping,Northern Yichang, South China. Acta Geol. Sin. (Chin. Ed.) 84, 70–77.

hu, X., Zhang, Q., Zhang, T., Feng, L., 2003. Sulfur and carbon isotopic variationsin Neoproterozoic sedimentary rocks from southern China. Prog. Nat. Sci. 13,875–880.

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.

erry, L.A., 2010. A burial diagenesis origin for the Ediacaran Shuram-Wonoka car-bon isotope anomaly. Earth Planet. Sci. Lett. 294, 152–162.

ong, L., Xiao, S., Shen, B., Zhou, C., 2008. Silicified Horodyskia and Palaeopascichnusfrom upper Ediacaran cherts in South China: tentative phylogenetic interpreta-tion and implications for evolutionary stasis. J. Geol. Soc. Lond. 165, 367–378.

ong, L., Xiao, S., Shen, B., Zhou, C., Li, G., Yao, J., 2009. Basal Cambrian microfossilsfrom the Yangtze Gorges area (South China) and the Aksu area (Tarim Block,northwestern China). J. Paleontol. 83, 30–44.

unham, R.J., 1962. Classification of carbonate rocks according to depositional tex-ture. Am. Assoc. Petrol. Geol. Mem. 1, 108–121.

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

ellatly, A.M., Lyons, T.W., 2005. Trace sulfate in mid-Proterozoic carbonates andthe sulfur isotope record of biospheric evolution. Geochim. Cosmochim. Acta69, 3813–3829.

rotzinger, J.P., Fike, D.A., Fischer, W.W., 2011. Enigmatic origin of the largest-knowncarbon isotope excursion in Earth’s history. Nat. Geos. 4, 285–292.

alverson, G.P., Dudás, F.Ö., Maloof, A.C., Bowring, S.A., 2007. Evolution of the87Sr/86Sr composition of Neoproterozoic seawater. Palaeogeogr. Palaeoclimatol.Palaeoecol. 256, 103–129.

alverson, G.P., Wade, B.P., Hurtgen, M.T., Barovich, K.M., 2010. Neoproterozoicchemostratigraphy. Precambrian Res. 182, 337–350.

shikawa, T., Ueno, Y., Komiya, T., Sawaki, Y., Han, J., Shu, D., Li, Y., Maruyama,S., Yoshida, N., 2008. Carbon isotope chemostratigraphy of a Precam-brian/Cambrian boundary section in the Three Gorge area, South China:prominent global-scale isotope excursions just before the Cambrian Explosion.Gondwana Res. 14, 193–208.

iang, G., Sohl, L.E., Christie-Blick, N., 2003a. Neoproterozoic stratigraphic com-parison of the Lesser Himalaya (India) and Yangtze block (south China):paleogeographic implications. Geology 31, 917–920.

iang, G., Kennedy, M.J., Christie-Blick, N., 2003b. Stable isotopic evidence formethane seeps in Neoproterozoic postglacial cap carbonates. Nature 426,822–826.

iang, G., Kennedy, M., Christie-Blick, N., Wu, H., Zhang, S., 2006. Stratigraphy, sed-imentary structures, and textures of the late Neoproterozoic Doushantuo capcarbonate in South China. J. Sediment. Res. 76, 978–995.

iang, G., Kaufman, A.J., Christie-Blick, N., Zhang, S., Wu, H., 2007. Carbon isotopevariability across the Ediacaran Yangtze platform in South China: implicationsfor a large surface-to-deep ocean (13C gradient. Earth Planet. Sci. Lett. 261,303–320.

iang, G., Shi, X., Zhang, S., Wang, Y., Xiao, S., 2011. Stratigraphy and paleogeogra-phy of the Ediacaran Doushantuo Formation (ca. 635–551 Ma) in South China.Gondwana Res. 19, 831–849.

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: insightsfrom the Chuar Group, Grand Canyon, USA. Earth Planet. Sci. Lett. 290, 64–73.

aufman, A.J., Knoll, A.H., 1995. Neoproterozoic variations in the C-isotope compo-sition of sea water: stratigraphic and biogeochemical implications. PrecambrianRes. 73, 27–49.

aufman, A.J., Corsetti, F.A., Varni, M.A., 2007. The effect of rising atmospheric oxygenon carbon and sulfur isotope anomalies in the Neoproterozoic Johnnie Forma-tion, Death Valley, USA. Chem. Geol. 237, 47–63.

aufman, A.J., Jiang, G., Christie-Blick, N., Banerjee, D.M., Rai, V., 2006. Stable isotoperecord of the terminal Neoproterozoic Krol platform in the Lesser Himalayas ofnorthern India. Precambrian Res. 147, 156–185.

endall, B., Creaser, R.A., Selby, D., 2009. 187Re–187Os geochronology of Precambrianorganic-rich sedimentary rocks. Geol. Soc. Lond. Spec. Publ. 326, 85–107.

endall, B., Reinhard, C.T., Lyons, T., Kaufman, A.J., Poulton, S., Anbar, A.D., 2010. Per-vasive oxygenation along late Archaean ocean margins. Nat. Geosci. 3, 647–652.

nauth, L.P., Kennedy, M.J., 2009. The late Precambrian greening of the Earth. Nature460, 728–732.

e Guerroue, E., Cozzi, A., 2010. Veracity of Neoproterozoic negative C-isotope val-

ues: the termination of the Shuram negative excursion. Gondwana Res. 17,653–661.

e Guerroue, E., Allen, P.A., Cozzi, A., Etienne, J.L., Fanning, M., 2006. 50 Myr recoveryfrom the largest negative �13C excursion in the Ediacaran ocean. Terra Nova 18,147–153.

h 192– 195 (2012) 125– 141

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.

Liu, P., Xiao, S., Yin, C., Tang, F., Gao, L., 2009a. Silicified tubular microfossils fromthe upper Doushantuo Formation (Ediacaran) in the Yangtze Gorges area, SouthChina. J. Paleontol. 83, 630–633.

Liu, P., Yin, C., Chen, S., Tang, F., Gao, L., 2011. The biostratigraphic succession of acan-thomorphic acritarchs of the Ediacaran Doushantuo Formation in the YangtzeGorges area, South China and its biostratigraphic correlation with Australia.Precambrian Res., doi:10.1016/j.precamres.2011.07.009.

Liu, P., Yin, C., Gao, L., Tang, F., Chen, S., 2009b. New material of microfossils fromthe Ediacaran Doushantuo Formation in the Zhangcunping area, Yichang, HubeiProvince and its zircon SHRIMP U–Pb age. Chin. Sci. Bull. 54, 1058–1064.

Liu, P., Xiao, S., Yin, C., Zhou, C., Gao, L., Tang, F., 2008. Systematic description andphylogenetic affinity of tubular microfossils from the Ediacaran DoushantuoFormation at Weng’an, South China. Palaeontology 51, 339–366.

Lu, M., Zhu, M., Zhao, M., 2009. Litho- and carbon isotope stratigraphy of the Edi-acaran System in the Sixi section (Yangtze Gorges), Yichang, Hubei. J. Stratigr.33, 359–372.

Lyons, T.W., Walter, L.M., Gellatly, A.M., Martini, A.M., Blake, R.E., 2004. Sites ofanomalous organic remineralization in the carbonate sediment of South Florida,USA: the sulfur cycle and carbonate-associated sulfate. In: Amend, J.P., Edwards,K.J., Lyons, T.W. (Eds.), Sulfur Biogeochemistry: Past and Present (GSA SpecialPaper 379). Geological Society of America, Boulder, CO, pp. 161–176.

Marenco, P.J., Corsetti, F.A., Hammond, D.E., Kaufman, A.J., Bottjer, D.J., 2008. Oxi-dation of pyrite during extraction of carbonate associated sulfate. Chem. Geol.247, 124–132.

McFadden, K.A., Xiao, S., Zhou, C., Kowalewski, M., 2009. Quantitative evaluation ofthe biostratigraphic distribution of acanthomorphic acritarchs in the EdiacaranDoushantuo Formation in the Yangtze Gorges area, South China. PrecambrianRes. 173, 170–190.

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

Moczydłowska, M., 2005. Taxonomic review of some Ediacaran acritarchs from theSiberian Platform. Precambrian Res. 136, 283–307.

Montanez, I.P., Banner, J.L., Osleger, D.A., Borg, L.E., Bosserman, P.J., 1996. IntegratedSr isotope stratigraphy and relative sea-level history in Middle Cambrian plat-form carbonates. Geology 24, 917–920.

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

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.

Reinhard, C.T., Raiswell, R., Scott, C., Anbar, A.D., Lyons, T.W., 2009. A late Archeansulfidic sea stimulated by early oxidative weathering of the continents. Science326, 713–716.

Ries, J.B., Fike, D.A., Pratt, L.M., Lyons, T.W., Grotzinger, J.P., 2009. Superheavy pyrite(�34Spyr > �34SCAS) in the terminal Proterozoic Nama Group, southern Namibia:a consequence of low seawater sulfate at the dawn of animal life. Geology 37,743–746.

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

Sawaki, Y., Ohno, T., Tahata, M., Komiya, T., Hirata, T., Maruyama, S., Windley, B.F.,Han, J., Shu, D., Li, Y., 2010. The Ediacaran radiogenic Sr isotope excursion in theDoushantuo Formation in the Three Gorges area, South China. Precambrian Res.176, 46–64.

Sergeev, V.N., Knoll, A.H., Vorob’Eva, N.G., 2011. Ediacaran microfossils from the UraFormation, Baikal-Patom Uplift, Siberia: taxonomy and biostratigraphic signifi-cance. J. Paleontol. 85, 987–1011.

Shen, B., Xiao, S., Kaufman, A.J., Bao, H., Zhou, C., Wang, H., 2008. Stratification andmixing of a post-glacial Neoproterozoic ocean: evidence from carbon and sulfurisotopes in a cap dolostone from northwest China. Earth Planet. Sci. Lett. 265,209–228.

Shen, B., Xiao, S., Bao, H., Kaufman, A.J., Zhou, C., Yuan, X., 2011. Carbon andsulfur isotope evidence for a strong depth gradient and oceanic oxidationafter the Ediacaran Hankalchough glaciation. Geochim. Cosmochim. Acta 75,1357–1373.

Shields, G., Veizer, J., 2002. Precambrian marine carbonate isotope database: version1.1. Geochem. Geophys. Geosyst. 3, doi:10.1029/2001GC000266.

Verdel, C., Wernicke, B.P., Bowring, S.A., 2011. The Shuram and subsequent Edi-acaran carbon isotope excursions from southwest Laurentia, and implicationsfor environmental stability during the metazoan radiation. Geol. Soc. Am. Bull.123, 1539–1559.

Vernhet, E., 2007. Paleobathymetric influence on the development of the lateEdiacaran Yangtze Platform (Hubei, Hunan, and Guizhou Provinces, China). Sed-iment. Geol. 197, 29–46.

Vernhet, E., Reijmer, J.J.G., 2010. Sedimentary evolution of the Ediacaran Yangtzeplatform shelf (Hubei and Hunan provinces, Central China). Sediment. Geol. 225,99–115.

Vernhet, E., Heubeck, C., Zhu, M.-Y., Zhang, J.-M., 2006. Large-scale slope instabilityat the southern margin of the Ediacaran Yangtze platform (Hunan province,central China). Precambrian Res. 148, 32–44.

Vorob’eva, N.G., Sergeev, V.N., Knoll, A.H., 2009. Neoproterozoic microfossils fromthe margin of the East European Platform and the search for a biostratigraphicmodel of lower Ediacaran rocks. Precambrian Res. 173, 163–169.

Wang, J., Li, Z.-X., 2003. History of Neoproterozoic rift basins in South China: impli-cations for Rodinia break-up. Precambrian Res. 122, 141–158.

esearc

W

W

X

X

X

X

X

Y

Y

Y

Y

Y

S. Xiao et al. / Precambrian R

ang, J., Jiang, G., Xiao, S., Li, Q., Wei, Q., 2008. Carbon isotope evidence forwidespread methane seeps in the ∼635 Ma Doushantuo cap carbonate in SouthChina. Geology 36, 347–350.

illman, S., Moczydłowska, M., 2008. Ediacaran acritarch biota from the Giles 1 drill-hole, Officer Basin, Australia, and its potential for biostratigraphic correlation.Precambrian Res. 162, 498–530.

iao, S., Knoll, A.H., 1999. Fossil preservation in the Neoproterozoic Doushantuophosphorite Lagerstätte, South China. Lethaia 32, 219–240.

iao, S., Knoll, A.H., 2000. Phosphatized animal embryos from the Neoprotero-zoic Doushantuo Formation at Weng’an, Guizhou, South China. J. Paleontol. 74,767–788.

iao, S., Yuan, X., Knoll, A.H., 2000. Eumetazoan fossils in terminal Proterozoic phos-phorites? Proc. Natl. Acad. Sci. U.S.A. 97, 13684–13689.

iao, S., Zhang, Y., Knoll, A.H., 1998. Three-dimensional preservation of algae andanimal embryos in a Neoproterozoic phosphorite. Nature 391, 553–558.

iao, S., Schiffbauer, J.D., McFadden, K.A., Hunter, J., 2010. Petrographic and SIMSpyrite sulfur isotope analyses of Ediacaran chert nodules: implications formicrobial processes in pyrite rim formation, silicification, and exceptional fossilpreservation. Earth Planet. Sci. Lett. 297, 481–495.

ang, J., Sun, W., Wang, Z., Xue, Y., Tao, X., 1999. Variations in Sr and C isotopesand Ce anomalies in successions from China: evidence for the oxygenation ofNeoproterozoic seawater? Precambrian Res. 93, 215–233.

in, C., Bengtson, S., Yue, Z., 2004. Silicified and phosphatized Tianzhushanian,spheroidal microfossils of possible animal origin from the Neoproterozoic ofSouth China. Acta Palaeontol. Pol. 49, 1–12.

in, C., Liu, P., Chen, S., Tang, F., Gao, L., Wang, Z., 2009. Acritarch biostratigraphicsuccession of the Ediacaran Doushantuo Formation in the Yangtze Gorges. ActaPalaeontol. Sin. 48, 146–154.

in, C., Liu, P., Awramik, S.M., Chen, S., Tang, F., Gao, L., Wang, Z., Riedman, L.A.,

2011. Acanthomorph biostratigraphic succession of the Ediacaran Doushan-tuo Formation in the East Yangtze Gorges, South China. Acta Geol. Sin. 85,283–295.

in, L., Zhu, M., Knoll, A.H., Yuan, X., Zhang, J., Hu, J., 2007. Doushantuo embryospreserved inside diapause egg cysts. Nature 446, 661–663.

h 192– 195 (2012) 125– 141 141

Yuan, X., Chen, Z., Xiao, S., Zhou, C., Hua, H., 2011. An early Ediacaran assem-blage of macroscopic and morphologically differentiated eukaryotes. Nature470, 390–393.

Zhang, S., Jiang, G., Han, Y., 2008. The age of the Nantuo Formation and Nantuoglaciation in South China. Terra Nova 20, 289–294.

Zhao, J.-H., Zhou, M.-F., Yan, D.-P., Zheng, J.-P., Li, J.-W., 2011. Reappraisal of the agesof Neoproterozoic strata in South China: no connection with the Grenvillianorogeny. Geology 39, 299–302.

Zhao, Y.-Y., Zheng, Y.-F., 2010. Stable isotope evidence for involvement of deglacialmeltwater in Ediacaran carbonates in South China. Chem. Geol. 271, 86–100.

Zhao, Z., Xing, Y., Ding, Q., Liu, G., Zhao, Y., Zhang, S., Meng, X., Yin, C., Ning, B., Han,P., 1988. The Sinian System of Hubei. China University of Geosciences Press,Wuhan, 205 pp.

Zhou, C., 1997. Upper Sinian carbon isotope profile in Weng’an, Guizhou. J. Stratigr.21, 125–129.

Zhou, C., Xiao, S., 2007. Ediacaran �13C chemostratigraphy of South China. Chem.Geol. 237, 89–108.

Zhou, C., Xie, G., Xiao, S., 2005. New data of microfossils from Doushantuo Formationat Zhangcunping, Hubei Province. Acta Micropalaeontol. Sin. 22, 217–224.

Zhou, C., Bao, H., Peng, Y., Yuan, X., 2010. Timing the deposition of 17O-depletedbarite at the aftermath of Nantuo glacial meltdown in South China. Geology 38,903–906.

Zhou, C., Xie, G., McFadden, K., Xiao, S., Yuan, X., 2007. The diversification andextinction of Doushantuo-Pertatataka acritarchs in South China: causes andbiostratigraphic significance. Geol. J. 42, 229–262.

Zhu, B., Becker, H., Jiang, S.-Y., Pi, D.-H., Fischer-Gödde, M., 2010. Re–Os geochronol-ogy of black shales from the Doushantuo Formation, Yangtze Platform, SouthChina. Geol. Soc. Am. Prog. Abstr. 42 (5), 463.

Zhu, M., Zhang, J., Yang, A., 2007. Integrated Ediacaran (Sinian) chronostratigraphy

of South China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 254, 7–61.

Zhu, M., Lu, M., Zhang, J., Zhao, F., Li, G., Yang, A., Zhao, X., Zhao, M., 2011.Carbon isotope chemostratigraphy and sedimentary facies evolution of the Edi-acaran Doushantuo Formation in western Hubei, South China. Precambrian Res.,doi:10.1016/j.precamres.2011.07.019.