Available online at www.sciencedirect.com

www.elsevier.com/locate/gca

ScienceDirect

Geochimica et Cosmochimica Acta 150 (2015) 90–105

Marine redox conditions in the middle Proterozoic oceanand isotopic constraints on authigenic carbonate

formation: Insights from the Chuanlinggou Formation,Yanshan Basin, North China

Chao Li a,⇑, Noah J. Planavsky b, Gordon D. Love c, Christopher T. Reinhard d,Dalton Hardisty c, Lianjun Feng e, Steven M. Bates c, Jing Huang f, Qirui Zhang e,

Xuelei Chu e, Timothy W. Lyons c

a State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Wuhan 430074, Chinab Department of Geology and Geophysics, Yale University, New Haven, CT 06520, USA

c Department of Earth Sciences, University of California, Riverside, CA 92521, USAd School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA 30332, USA

e Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, Chinaf CAS Key Laboratory of Crust-Mantle Materials and Environments, School of Earth and Space Science, University of Science and

Technology of China, Hefei 230026, China

Received 6 April 2014; accepted in revised form 3 December 2014; Available online 10 December 2014

Abstract

To improve our understanding of ocean chemistry and biogeochemical cycling following the termination of large-scalePaleoproterozoic iron formation (IF) deposition (�1.85 billion years ago [Ga]), we conducted a Fe–S–C–Mo geochemicalstudy of the �1.65 Ga Chuanlinggou Formation, Yanshan Basin, North China. Despite the cessation of IF deposition,our results suggest the presence of anoxic but non-euxinic (ferruginous) conditions persisted below the surface mixed layerfor the deepest portion of the continental rifting basin and that this pattern is apparently independent of the local organiccarbon content. However, our paired S-isotope data of carbonate-associated sulfate and pyrite suggest presence of sulfatein pore fluids, which is not consistent with insufficient sulfate for bacterial sulfate reduction in the water column. Despite evi-dence for deposition under anoxic conditions, sedimentary molybdenum (Mo) concentrations are mostly not enriched relativeto average continental crust. This relationship is consistent with the notion that sulfide-dominated conditions in the watercolumn and/or the sediments are required for Mo enrichment and validates past assertions that Mo enrichment patternsin ancient shales track both the local presence and global distribution of euxinia specifically. In addition, we identified exten-sive diagenetic carbonate precipitation in the upper Chuanlinggou Formation with only moderately negative d13C values(�3.4 ± 1.4&). We propose, with support from a numerical model, that these diagenetic carbon isotope values were mostlikely derived from precipitation of carbonates dominantly in the methanic zone within the sediments. Diagenetic carbonateprecipitation in the methanic zone is likely to have been more extensive in the Proterozoic than the Phanerozoic due toporewater oxidant limitation.� 2014 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.gca.2014.12.005

0016-7037/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +86 27 67883606.E-mail address: chaoli@cug.edu.cn (C. Li).

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 91

1. INTRODUCTION

The last episode of major, economic iron formation (IF)deposition occurred around 1.85 Ga. Traditionally, thistransition in the sedimentary record was linked to a shiftfrom an anoxic and Fe-rich (ferruginous) ocean to one withmildly oxic deep waters (Holland, 2006; Slack et al., 2007,2009). However, because H2S can also titrate dissolvedFe2+ in anoxic marine environments to form pyrite, the glo-bal development of euxinic conditions in the deep ocean hasbeen suggested as being responsible for the disappearanceof large IFs at �1.85 Ga (Wilde, 1987; Canfield, 1998;Poulton et al., 2004). Indeed, multiple lines of evidencepoint to widespread euxinia in the mid-Proterozoic ocean(e.g., Shen et al., 2003; Brocks et al., 2005; Scott et al.,2008). Following from this model, delayed eukaryoticexpansion and diversification in mid-Proterozoic oceanshas been attributed to the limited availability of bioessentialand redox-sensitive metals (e.g., Fe, Mo, Cu, Zn, Cd) thatcould be readily removed from sulfidic waters (e.g.,Anbar and Knoll, 2002; Scott et al., 2008; Reinhardet al., 2013).

Rather than global-scale deep euxinia, recent studieshave proposed that Proterozoic ocean waters below the sur-face mixed layer remained largely ferruginous (Canfieldet al., 2008; Li et al., 2010; Poulton et al., 2010; Poultonand Canfield, 2011; Planavsky et al., 2011; Reinhardet al., 2013), although oxic deep waters may have prevailedin some oligotrophic basins (e.g., Sperling et al., 2014). Fur-ther, there is evidence that productive ocean margin settingswere commonly characterized by mid-depth euxinia withsurface, at least weakly oxic waters and deeper ferruginouswaters (Li et al., 2010, 2012; Poulton et al., 2010; Poultonand Canfield, 2011; reviewed in Lyons et al., 2012, 2014).Euxinia in the Proterozoic oceans spatially limited to inter-mediate depths, dominantly on ocean margins, may stillhave been sufficient to terminate the deposition of the largeIF at �1.85 Ga (e.g., Poulton et al., 2010; Kendall et al.,2011). Alternatively, the end of major IF deposition mayhave been linked to changes in hydrothermal fluid Fe-to-sulfide ratios (e.g., Kump and Seyfried, 2005; Planavskyet al., 2011) and the overall fluxes of those fluids to theocean (Isley and Abbott, 1999; Bekker et al., 2010). Moregenerally, expanding euxinia on the continental marginmay have been a factor leading to the patchy temporalrecord of early metazoan fossils (and inferred diversifica-tion rates) in the late Neoproterozoic (e.g., Li et al., 2010)and the early Cambrian (Feng et al., 2014; Jin et al.,2014), as well as to the mass extinctions at the end-Ordovi-cian (Hammarlund et al., 2012) and end-Permian (e.g.,Algeo et al., 2011), among others, suggesting a close rela-tionship between the spatial patterns of marine euxiniaand ecosystem stability through Earth history.

In order to investigate Proterozoic ocean chemistry dur-ing a key transitional period—the late Paleoproterozoic—and specifically to probe the nature of ocean redox struc-ture, we conducted a Fe–S–C–Mo biogeochemical studyof the �1.65 Ga Chuanlinggou Formation in the late Paleo-proterozoic semi-restricted continental rifting YanshanBasin, North China. In this study, we aim to address two

themes: (1) reconstructing seawater chemistry in a continen-tal rift basin and (2) the overarching factors that controlledseawater chemistry in the basin and in turn the broaderimplications for global ocean chemistry during the mid-Proterozoic.

2. GEOLOGIC BACKGROUND

2.1. Tectonics, stratigraphy, and paleogeography

The Yanshan Basin was a continental rift setting thatdeveloped within the northern margin of the North ChinaBlock (Fig. 1A) between �1.8 and 1.6 Ga. This continentalrifting was possibly linked to the early partial breakup ofthe supercontinent Columbia (also known as Nuna) in themid-Proterozoic from 1.6 to 1.2 Ga (Kusky and Li, 2003;Meng et al., 2011; Zhang et al., 2012, and referencestherein). The Changcheng Group is the earliest strati-graphic unit in the basin and consists of the Changzhougou,Chuanlinggou, Tuanshanzi, and Dahongyu formations(Fig. 1B). The Changcheng Group is distributed mainly inrelatively narrow fault-bounded zones of apparent grabenand half-graben geometries (Fig. 1A; Yan and Liu, 1998;Qiao and Gao, 2007).

In the depocenter area at Jixian, the basic geology of theProterozoic succession, including the Changcheng Group,has been studied systematically (e.g., Chen et al., 1980).The Changzhougou Formation, which unconformablyoverlies an Archean metamorphic complex, was dominatedby fluvial coarse sandstones and conglomerates in its lowerportion and by littoral-intertidal siltstones and sandstonesin its upper portion. Inferences of a predominantly flu-vial-littoral-intertidal depositional environment are sup-ported by abundant wave- and current-generatedsedimentary structures, such as ripples, swash marks, andcross bedding in the formation (Fig. 1B). The lower Chuan-linggou Formation consists mainly of silty illitic shales withinterbedded siltstones/sandstones and is characterized bysedimentary structures, such as cross bedding, indicativeof a shallow marine environment. The middle Chuanling-gou Formation is mainly composed of black illitic shaleswith clear planar bedding, suggesting a subtidal low-energyenvironment during deposition. The upper ChuanlinggouFormation consists mainly of calcareous black shales thatare intercalated with early- to mid-stage diagenetic carbon-ate beds, nodules, lenses and laminae (Fig. 2). The diage-netic carbonate laminae, lenses and nodules are typicallybedding-parallel with a wide range of lengths from a lessthan centimeters to almost 50 cm. Diagenetic beds/lami-nae/lenses/nodules can be characterized by gradual andirregular, wavy boundaries—similar to those commonlyobserved in Phanerozoic successions that have receivedextensive study (e.g., Miocene Monterey Formation, Loydet al., 2012). Some small carbonate veins penetrate theblack shale layers with wrinkles and are often terminatedby bedding-parallel carbonate beds or laminae (Fig. 2D);these features are similar to those sand veins and microbial-ly induced sedimentary structures (MISS) previously foundin the middle Chuanlinggou Formation (Shi et al., 2008),thus well consistent with their diagenetic origins. Some silty

Fig. 1. Geological background for the 1.65 Ga Chuanlinggou Formation. (A) Tectonic and paleogeographic context for the Yanshan Basinduring the deposition of the Chuanlinggou Formation (after Qiao and Gao, 2007, and Yan and Liu, 1998). YSTF: Yellow Sea TransformFault. Arrows represent invasion of open ocean water into the basin from an N and NE direction during the marine transgression as recordedin the middle Chuanlinggou Formation. (B) Generalized stratigraphic column with detailed sedimentary structures of the Changcheng Systemat the Jixian section (after Chen et al., 1980 and Chu et al., 2007). Age data sources: Lu and Li (1991), Wan et al. (2003), Gao et al. (2008,2009), Li et al. (2011). Formations: TSZ (Tuanshanzi); DHY (Dahongyu). CLG1, CLG2, CLG4 and CLG5 are the outcrop subsections at theJixian section.

Fig. 2. Diagenetic carbonate beds, nodules, lenses and laminae within the calcareous black shale facies of the upper Chuanlinggou Formationat the Jixian section. (A) A view of the upper subsection CLG5 showing carbonate beds. (B and C) Typical carbonate nodules, lenses andlaminae within the calcareous black shales. The red bar in B and the red marker pen in C are 5 cm and �15 cm, respectively, for scale. (D) Aclose look of the diagenetic carbonates in the dashed yellow frame in C showing perpendicular-to-high angle veins (to bedding layers) withwrinkles and wavy boundaries. The coin in D is 2.5 cm in diameter for scale.

92 C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 93

beds and lenses with relatively high-energy sedimentarystructures have been found locally in the middle-to-upperChuanlinggou Formation (Fig. 1), suggesting shoaling dur-ing deposition. Therefore, the middle-to-upper Chuanling-gou Formation was most likely deposited in apredominantly subtidal environment with episodic shoalinginto an intertidal environment.

The overlying Tuanshanzi Formation is dominantlycomposed of muddy dolostones in the lower portion andstromatolitic dolostones in the upper portion, which arelocally intercalated with siltstone or silty shale beds; abun-dant sedimentary structures such as mud cracks and crossbedding were found in the formation, indicating an inter-tidal to supratidal/lagoonal environment (Fig. 1B). Thus,the Changcheng Group is typically interpreted as recordinga transgressive–regressive cycle in its first three formations,with peak transgression coincident with deposition of theblack shales of the Chuanlinggou Formation (Yan andLiu, 1998).

The Chuanlinggou Formation shows a basin-scale waterdepth gradient running from north to south with deep-water facies deposited in the south along a narrow beltbounded by two sets of NE- to SW-trending rift faults (Bei-jing–Lingyuan and Luanxian–Jianchang in Fig. 1A). Dur-ing the transgressive phase of deposition for theChuanlinggou Formation, open ocean waters likely flowedinto the basin from the north (Qiao, 2002), the northeast(Yan and Liu, 1998), or both, and a water-depth gradientdeveloped from the north to the rift zone (Fig. 1A). Sucha depth gradient has been clearly identified in the transi-tional Xinglong area (Xu et al., 2002). Therefore, the Yan-shan Basin was generally a semi-restricted rift basin duringdeposition of the Chuanlinggou Formation, with maximumconnection to the open ocean at the time of the peaktransgression.

Our samples were collected from the Chuanlinggou For-mation in the depocenter area at Jixian from four outcropsubsections (CLG1, CLG2, CLG4, and CLG5 inFig. 1B). These comprise the middle to upper ChuanlinggouFormation in a section with a general lithologic transitionfrom silty black shales to calcareous black shales(Fig. 1B). The calcareous black shales without any evidencefor high-energy depositional conditions (CLG-5) are inter-preted as the deepest facies. These subtidal black shales rep-resent water depths beneath storm wave base from thedeepest portion of the Yanshan basin—but are certainlynot considered ‘deep’ in an open ocean sense.

2.2. Chronology

Zircon U–Pb ages of 1625 ± 6 million years (Ma) (Luand Li, 1991) and 1625.9 ± 8.9 Ma (Gao et al., 2008) fromthe Dahongyu volcanic rocks in the Jixian area and theyoungest U–Pb age of 1805 ± 25 Ma from the Changzhou-gou detrital zircons in the Beijing area (Wan et al., 2003)place the age of the Changcheng Group in the range ofca. 1.8–1.6 Ga. Recently, the age model for deposition ofsedimentary strata within the Changcheng Group wasrevised to 1.7–1.6 Ga to be consistent with new U–Pb vol-canic ages of 1673 ± 10 Ma for a granite-porphyry dike

(Li et al., 2011) and 1731 ± 4 Ma for a mafic dike (Penget al., 2012). These dikes were emplaced into the Archeanmetamorphic basement but were unconformably overlainby the Changzhougou Formation in Beijing area. Theseradiometric age constraints for the Changcheng Group(shown in Fig. 1B), including the new SHRIMP zirconage of 1638 ± 14 Ma for the diabase in the ChuanlinggouFormation reported by Gao et al. (2009), suggest that theChuanlinggou Formation was most likely deposited around�1.65 Ga—and thus well after the cessation of large-scaleIF deposition at �1.85 Ga.

3. SAMPLES AND METHODS

A total of 55 sedimentary rock samples were analyzed inthis study. For 27 out of the 55 samples, iron speciationdata were reported previously by Planavsky et al. (2011)as part of a sample compilation but without the detailedgeological and geochemical context that we provide here.All outcrop samples were collected in August 2009. Largeblocks (>200 g) of the freshest exposures were targetedfor sampling, and any surfaces that might have been weath-ered were trimmed prior to powdering. During rock pow-dering, we did not observe obvious visible pyrite nodulesor bands in our samples, nor were the samples significantlyrust stained. Details for the analytical methods are similarto those reported previously in Li et al. (2010, 2012). A con-cise summary of the Fe, Mo, and S geochemical methodsfollows.

Concentrations of total Fe (FeT) and Mo were measuredusing an Agilent 7500 quadrupole inductively coupledplasma mass spectrometer (ICP-MS) following a standardmulti-acid digestion (HNO3–HCl–HF). Replicate analysesof USGS standard (SDO-1, n = 7) were reproducible within1.4% and 10.4% for assays of Fe and Mo, respectively. TheFe contents in Fe-carbonates (Fecarb), ferric oxides (Feox),and magnetite (Femag) were determined using the sequentialextraction procedure described in Poulton and Canfield(2005) with quantification by ICP-MS. Disseminated pyritein samples was extracted using the chromium reductionmethod described in Canfield et al. (1986) with silver nitratetraps. Resulting Ag2S precipitates were carefully collectedand weighed for the stoichiometric calculation of pyrite sul-fur concentration and, in turn, the Fe content present as pyr-ite (Fepy) in the samples. The quality of Fepy datadetermined by weighing Ag2S precipitates was tested bytitration of ZnS collected in zinc acetate traps using the samechromium reduction method for five random samples withan average agreement of 5.2%. A subset of samples (10) weretested for the presence of pyrrhotite—given the possibility ofpyrite-to-pyrrhotite conversion at the metamorphic grade ofthe samples—using a HCl and SnCl2 boiling extraction (seedetails in Planavsky et al., 2011, and Asael et al., 2013).These tests confirmed that pyrrhotite was a minor compo-nent of the sulfide pool (<10% of the total sulfide pool).

Carbonate-associated sulfate (CAS) in our samples wasextracted using the HCl–BaCl2 procedure similar to thatdescribed in Wotte et al. (2012). A brief protocol is givenbelow: (1) 60–130 g of sample powder (200 mesh) were firstrinsed with a 10% NaCl solution to remove later stage

94 C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105

sulfate until no dissolved sulfate was observed to precipitateon addition of saturated BaCl2 solution (�250 g/L) in dec-anted supernatants. This step typically involved 2–3 rinses.(2) The samples were further rinsed with bleach to removeorganic matter, any potential organically bound sulfur, andmetastable pyrites until no sulfate was recovered as BaSO4.(3) The samples were then treated with 4 N HCl, which wasadded in small aliquots to minimize the likelihood that pHfell below 3.8. If a sample contained significant pyrite,which might have oxidized to some degree, 15% SnCl2was added to the HCl to inhibit possible ferric iron-medi-ated pyrite oxidation during CAS extraction (Planavskyet al., 2012). (4) Samples were processed by vacuum filtra-tion immediately following the HCl dissolution step toremove the CAS-bearing solution from the insoluble resi-due within 2 h to minimize further the potential for pyriteoxidation following dissolution. (5) Finally, liberated CASwas precipitated as BaSO4 for subsequent isotopic measure-ments by adding a saturated BaCl2 solution.

Resulting BaSO4 precipitates, as well as the Ag2S precip-itates resulting from the pyrite extractions, were combustedwith an excess of V2O5 on a Thermo Scientific Delta V Plusisotope ratio mass spectrometer coupled with a Costech ele-mental analyzer for determining sulfur isotope composi-tions of both pyrite (d34Spy) and CAS (d34SCAS). Sulfurisotope compositions are expressed in standard d-notationas permil (&) deviation from the V-CDT internationalstandard with an analytical error of 0.1& (1r) calculatedfrom replicate analyses of IAEA international standards(IAEA-S1, IAEA-S2, IAEA-S3 for d34Spy and NBS-127,IAEA-SO-5, IAEA-SO-6 for d34SCAS).

4. RESULTS AND DISCUSSION

4.1. Fe–C systematics

To investigate the marine redox chemistry of the�1.65 Ga Chuanlinggou Formation, we examined the con-centrations of total Fe (FeT) as well as extractable Fegrouped operationally by specific mineral types: Fecarb,Feox, Femag, Fepy, and remaining ‘unreactive’ Fe (FeU)mainly in clay minerals and other silicates. Unlike FeU,Fe in the other four mineral groups is reactive enough toform pyrite given sufficient exposure to H2S either in thewater column or during early sediment diagenesis and thusis defined as highly reactive iron (FeHR). Studies of modernsediments deposited beneath anoxic bottom waters showthat FeHR/FeT usually exceeds 0.38 in these settingsthrough inputs of additional FeHR tied to Fe cycling inthe oxygen-deficient water column (Raiswell and Canfield,1998; Raiswell et al., 2001; Poulton and Raiswell, 2002;reviewed in Lyons and Severmann, 2006, and Poultonand Canfield, 2011). This threshold value has been appliedwith success in numerous geochemical studies of ancientfine-grained siliciclastic rocks, although thermal alterationcan potentially convert FeHR to FeU during burial andreduce this threshold value (Raiswell et al., 2008).

Because the total Fe content remains constant despiteany internal redistribution, FeT/Al is another way to lookfor anoxic sedimentary Fe enrichments in rocks that might

have experienced burial/metamorphic alteration. Underanoxic bottom waters, modern analogues show that FeT/Al usually exceeds �0.5, nominally the ratio of averagecontinental crust and of most modern and ancient marinemuds and shales deposited in oxic environments (Lyonsand Severmann, 2006). Sediments deposited beneath oxy-gen-containing waters but immediately proximal to largeanoxic portions of the water column can also have enrich-ments in highly reactive iron (e.g., Scholz et al., 2014),although such settings are unlikely to leave a spatiallyand stratigraphically persistent record. Under anoxic andsulfidic (euxinic) bottom waters, FePy/FeHR exceeds 0.7and often 0.8 because the reactive Fe phases are titratedout of the system before appreciable H2S starts to accumu-late in the water column (Raiswell and Canfield, 1998;Poulton and Canfield, 2011).

Stratigraphic analysis of iron speciation data for theChuanlinggou Formation reveals two intervals with distinctfeatures (see Fig. 3). The first, a lower interval (LI), com-prising subsections CLG2, CLG4, and the lower �20 mof CLG5, is characterized by low FeHR/FeT [0.19 ± 0.08(SD)] and modestly elevated FeT/Al ratios (0.67 ± 0.40),relative to typical continental margin sediments (Lyonsand Severmann, 2006). This interval also shows moderateFePy/FeHR ratios (0.31 ± 0.18). This combination ofparameters could be consistent with either ferruginous oroxic bottom-water conditions for the deposition duringthe LI. The second distinct interval, an upper interval(UI) consisting of the upper part of CLG5 and CLG1, ischaracterized by higher FeHR/FeT ratios (0.68 ± 0.12) andhigher FeT/Al (1.27 ± 1.05) but lower FePy/FeHR

(0.10 ± 0.12) and similar FeT contents (see Fig. 3 and theAppendix), consistent with anoxic and ferruginous condi-tions for the deposition of the UI. Because of elevated car-bonate contents (>50%), some UI samples are not wellsuited to the Fe-based paleoredox proxies, which weredeveloped for and traditionally applied to fine-grained sili-ciclastics. However, a recent study based on a large compi-lation suggests that the FeHR/FeT threshold value of 0.38can also be valid for carbonate-rich rocks, provided thatFeT > 0.5 wt% (Clarkson et al., 2014), a threshold sur-passed by all of the samples examined here.

The UI contains the deepest water facies examined, indi-cating an increase in the amount of reactive Fe and in theFeHR/FeT ratios that correlate generally with the depth atwhich the sediments were deposited in the Yanshan Basin.This shift could be linked to the presence of a chemocline—i.e., a switch from oxic to ferruginous conditions withgreater depth. There is also a slight increase in averageTOC content during the transition to the UI—from0.67 wt% in LI to 1.16 wt% (Fig. 3 and Appendix). Theextent of organic matter loading is often invoked as a prox-imate driver for anoxic or euxinic water column chemistrybelow the mixed layer in Proterozoic oceans (e.g.,Johnston et al., 2010). However, the lack of iron enrich-ments in the LI, rather than reflecting deposition beneathoxic waters, could be explained by deposition under anoxicbut rapidly accumulating conditions, which can mask thesignatures of anoxia via clastic dilution (see Lyons andSevermann, 2006). This might be the case for the LI given

Fig. 3. Composite chemostratigraphy of the Jixian section with new iron speciation data from this study as well as data reported by Planavskyet al. (2011) and related geochemical proxies that provide information for water column redox conditions. UI: upper interval; LI: lowerinterval. Specific classes of chemically distinguishable iron mineral species are plotted relative to FeT in order to show their relative variationsfrom the LI into the UI. From left to right, the dashed lines in boxes represent average of TOC in this study, total Fe of average shales(Turekian and Wedepohl, 1961), average crustal value of FeT/Al, boundary values (0.38 for FeHR/FeT and 0.7–0.8 for FePy/FeHR) whichdifferentiate euxinic, ferruginous and oxic waters (see more details in text) and average of S-isotope compositions of pyrites from this study.

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 95

the presence of rippled surfaces (in CLG2, see Figs. 1 and 3)suggesting nearshore deposition at potentially high rates ofsedimentation.

Strong positive correlation was observed between FeHR/FeT ratios and TIC contents in all samples (R2 = 0.9;Fig. 4A). Furthermore, similar correlations were alsoobserved between Fecarb/FeT and TIC for both LI (linearR2 = 0.89) and UI (exponential R2 = 0.77) samples(Fig. 4B), consistent with appreciable ferrous carbonate(e.g., siderite) formation supported by an excess of Fe2+.These correlations are linked to varying amounts of detritalsediments versus authigenic (early diagenetic) carbonatephases.

Despite some ambiguity in the Fe speciation data,there is clearly no evidence for euxinic conditions in sub-surface waters for the deepest portion of the Yanshanbasin at Jixian. We also note that the internal consistencyof our Fe speciation data as demonstrated above (e.g.,elevated Fecarb with higher TIC; near zero Feox, and

Fig. 4. Crossplots of (A) FeHR/FeT and Fecarb/FeT (B), respectively, vChuanlinggou Formation at the Jixian section.

Femag in the most-easily-weathered LI black shales) indi-cate that recent surface weathering is not an issue forthese iron ratios and specifically that the pyrite concentra-tions that can be indicative of euxinia were not lostthrough later oxidation.

Based on Fe data it seems that non-euxinic conditionswere sustained in the subsurface water column of the�1.65 Ga semi-restricted Yanshan basin. In an anoxic mar-ine system, the development of euxinic versus ferruginousconditions will ultimately be controlled by the flux ratioof Fe2+/SO4

2� into the system, with a proximate controlexerted by the need for sufficient organic matter flux to fuelmicrobial H2S formation. We did not see an appreciablecorrelation between pyrite formation (expressed as Fepy/FeHR) and TOC contents for either our UI samples or allour Chuanlinggou samples (R2 = 0.02; Fig. 5), but this doesnot necessarily delineate whether sulfate levels or organicmatter fluxes were limiting sulfide production and the devel-opment of euxinia (more on this below).

ersus total inorganic carbon (TIC) content in samples from the

Fig. 5. Crossplot of FePy/FeHR versus total organic carbon (TOC)content in samples from the Chuanlinggou Formation at the Jixiansection.

Fig. 6. Frequency distribution of sulfur isotopic composition ofpyrite (d34Spy) preserved in the �1.65 Ga Chuanlinggou Forma-tion. For comparisons, the isotopic ranges of possible coevalseawater sulfate are shown above the histogram of d34Spy data. (1):the isotopic composition of carbonate-associate sulfate (d34SCAS)extracted from the Chuanlinggou Formation (Chu et al., 2007; thisstudy); (2): d34SCAS data from �1.65 Ga Paradise Creek Formation(Kah et al., 2004; Gellatly and Lyons, 2005); (3): the isotopiccomposition of coeval seawater sulfate (d34Ssw) suggested byStrauss (1993) based on sulfate evaporite records; (4) a d34Ssw

value of 21& suggested by Canfield and Farquhar (2009).

96 C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105

4.2. Sulfur isotopes

In order to explore the sulfur biogeochemistry in theYanshan basin and further inform the chemistry of latePaleoproterozoic oceans more broadly, we analyzed the iso-topic compositions of disseminated pyrites (d34Spy) andputative coexisting sulfate captured and preserved in thelattice of carbonate minerals (CAS and thus d34SCAS) ofthe same sample. Past work reveals that CAS can captureand preserve the isotopic composition of dissolved seawatersulfate (e.g., Burdett et al., 1989; Kah et al., 2004; Gill et al.,2007, 2008; Planavsky et al., 2012). In our case, however,use of the technique is atypical in that we are looking atCAS most likely in diagenetic carbonates (see also Loydet al., 2012). Therefore, the d34SCAS value is expected tocapture a pore fluid signal rather than the overlying seawa-ter value. An important implication is that our CAS dataare likely to be isotopically heavier than the coeval seawatergiven the preference for the light isotope, 32S, during bacte-rial sulfate reduction (BSR). Indeed, the d34SCAS values ofthe Chuanlinggou Formation, which are only availablefor the carbonate-rich UI samples in this study, range from27.3& to 40.2&. These values are significantly heavier thanthe estimated d34S of coeval seawater sulfate (d34Ssw) of21& (e.g., Strauss, 1993; Canfield and Farquhar, 2009)and comparable or heavier than previously reportedd34SCAS values for the Chuanlinggou Formation (Chuet al. (2007) and the �1.65 Ga Paradise Creek Formationin Australia (Kah et al., 2004; Gellatly and Lyons, 2005;Fig. 6).

In samples with d34SCAS and d34SPy values, their differ-ence (D34S = d34SCAS � d34SPy) ranges from �6.8& to4.8& (see Appendix; also see Fig. 6). This is a narrow rangecompared to most modern settings, and the large fractiona-tions often associated with BSR, and could reflect spuriousCAS records via pyrite oxidation. However, a few lines ofevidence suggest otherwise. First, as described above, wewere careful to avoid pyrite oxidation during extractionby using a protocol that minimizes ferric iron-mediated

reactions. We also rinsed our samples very carefully toremove any secondary sulfate that might have resulted fromoutcrop oxidation. Consistent with this level of care, somesamples, despite the large amount of material used (>50 g)and their relatively high pyrite contents (see Appendix),failed to yield measurable amounts of CAS. AlthoughCAS is low in our samples, we are confident that ourCAS extraction method effectively prevented significantcontamination from pyrite oxidation and other non-CASduring the CAS extraction. Also, 83% of our CAS-yieldingsamples had less than 0.3% Fepy (see Appendix), and manyof them contained only trace amounts of pyrite(Fepy < 0.1%), which would limit the potential impact ofoxidation during extraction. These samples still show thesmall D34S range noted above. Most importantly, if pyriteoxidation affected the d34SCAS significantly, during extrac-tion or through outcrop weathering, we should see a com-pressed D34S range—but not negative D34S values giventhe negligible isotope fractionation during pyrite oxidation(Taylor et al., 1984; Balci et al., 2007).

Peng et al. (2014) suggested that extracted sulfate couldbe contaminated by recent, secondary atmosphericallyderived sulfate (SAS). However, we used great care in rins-ing our samples, and our d34SCAS data are characteristicallyheavy and thus inconsistent with the much lighter putativeSAS contaminants from multiple locations suggested byPeng et al. (2014).

If the small-to-negative D34S values are genuine, as weassert, our observations are significant because they implythe presence of sulfate in pore fluids and sedimentary sulfatereduction under diagenetic conditions. This relationship in

Table 1Explanation and values for terms used in pore fluid dissolvedinorganic carbon model in this study.

Variable Explanation

x Sedimentation rate (cm/yr)G0 Utilizable organic carbon (moles � cm�3)Ds Sediment diffusion coefficient (1.92E�5 cm2 s�1 for

CO2, see Boudreau, 1996 table 4.3)k Reaction rate for anoxic remineralization (Tromp et al.,

1995)C0 Marine DIC concentration (2.2E�6 mol � cm�3 for

modern)z Depth (cm)d13Corg Carbon isotope value of organic carbon (�25&)d13Csw Carbon isotope value for marine seawater (0&)

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 97

turn indicates that the ferruginous, rather than euxinic, watercolumn implied for the UI was probably not a consequenceof sulfate being insufficiently abundant in the water columnto fuel BSR in excess of reactive Fe availability.

Given all these considerations, and assuming that thecarbonate is diagenetic, pyrite isotopically heavier thancoexisting sulfate is most easily explained, at least in ourcase, by formation that was delayed, from an evolved porefluid sulfate reservoir—with BSR driving the porewatersisotopically heavier (see also Loyd et al., 2012). Also impli-cit is earlier capture of the CAS in diagenetic carbonates.Specifically, we tentatively link these negative D34S valuesto calcium carbonate precipitation in the upper portion ofthe sediment pile and delayed pyrite formation via sulfida-tion of Fe carbonates deeper within the sediments. Detailsof this conceptual model remain unresolved, including therequired loss of earlier formed hydrogen sulfide via upwarddiffusion or some process other than pyrite formation, suchas reaction with organic matter. This model, however, isnot unreasonable given the expected slower reaction kinet-ics for iron-rich carbonates compared to labile Fe(III) oxy-hydroxide phases, which in this case may have yielded theirFe first, via bacterial reduction, to secondary ironcarbonates.

4.3. Mo enrichments

There is iron speciation evidence for ferruginous condi-tions in the deep-water facies of the Chuanlinggou Forma-tion. In some of the Chuanlinggou samples we investigated,the Fe proxy data yield ambiguous conclusions about depo-sitional redox conditions, but there is no clear evidence forextended periods of euxinia in the late PaleoproterozoicYanshan Basin—at least in the deepest Jixian area. Reac-tion with free hydrogen sulfide is a critical player in Mo sed-imentary enrichment under anoxic conditions; sulfide isrequired for the formation of particle reactive thiomolyb-dates or a Fe–Mo–S phase (e.g., Helz et al., 1996, 2011;Scott and Lyons, 2012). The bulk-rock Mo concentrations

Fig. 7. Compilation of molybdenum concentration versus FeHR/FeT (A)Chuanlinggou Fm. and euxinic Rove Fm. (1.84 Ga), Wollogorang Fm. (1Data sources: Chuanlinggou Formation (this study), Rove, Wollogoranget al., 2003).

in our Chuanlinggou samples are on average �1 ppm(Fig. 7A; Table 1), which is indistinguishable from averagecrustal values (1–2 ppm). This relationship is consistentwith our interpretation of dominantly ferruginous (sul-fide-limited) rather than euxinic conditions for the deepwaters (Scott et al., 2008) and a lack of significant pore-water sulfide accumulation in the upper, Fe-buffered por-tion of the sediment pile. The lack of Mo enrichmentpersists over a relatively wide range of TOC values, in con-trast to what is observed in typical euxinic shales (Fig. 7B).This relationship provides critical support for the idea thatsediments deposited below ferruginous marine settings arenot significant Mo sinks. Therefore, this work supportsthe argument that the size of the marine Mo reservoir canbe used to gauge the extent of euxinic conditions specifi-cally—rather than anoxic conditions more generally (seeReinhard et al., 2013).

4.4. Carbon isotope values in diagenetic carbonates

There are relatively few studies on Proterozoic earlydiagenetic carbonates despite the possibility that pore fluid

and total organic carbon (TOC) (B), respectively for the ferruginous.73 Ga), Velkerri Fm. (1.4 Ga), and modern Cariaco basin (14.5 ka).and Velkerri Formations (Scott et al., 2008), Cariaco basin (Lyons

98 C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105

carbonate precipitation may have been an important part ofthe global carbon cycle and global carbon isotope mass bal-ance during the Precambrian. Specifically, it was recentlyproposed that there was much more extensive precipitationof isotopically light diagenetic carbonates in the Precam-brian, when the oceans were more pervasively reducing(Higgins et al., 2009; Johnston et al., 2013; Macdonaldet al., 2013; Schrag et al., 2013). This phenomenon coulddramatically change estimates of organic carbon burialderived from the marine dissolved inorganic carbon (DIC)record (Schrag et al., 2013). Although the focus of this studyis not strictly on diagenetic carbonate formation, our sam-ples provide an excellent opportunity to explore their poten-tial impact on C isotope records. We provide additionaldiscussion on the mode and significance of the diageneticcarbonates of the Chuanlinggou Formation and explorethe possible links between marine oxidant levels, ferruginouspore fluids, and the authigenic carbonate sink.

The fact that the examined anoxic calcareous shales ofthe UI contain extensive diagenetic carbonates (Fig. 2) isconsistent with the idea that anoxic remineralization path-ways will drive extensive carbonate precipitation (Higginset al., 2009; Schrag et al., 2013). All of the carbonates fromthis study have negative d13C values, which is consistentwith at least partial pore fluid carbonate formation (e.g.,Loyd et al., 2012). However, the average d13C value(�3.4 ± 1.4&) is heavier and less variable than typicalmarine diagenetic carbonates formed within the zones ofsulfate and iron reduction or through anaerobic oxidationof biogenic methane in Phanerozoic sediments (usuallybetween �6& and �60&; e.g., Zhu et al., 2002; Meisteret al., 2007; Naehr et al., 2007; Loyd et al., 2012). Giventhe limited number of samples in this study, additionalwork will be needed to gauge if this is a common featureof Proterozoic diagenetic carbonates. However, if the Chu-anlinggou Formation samples are roughly representativeof the isotope values of typical mid-Proterozoic diageneticcarbonates, the data would imply that even extensiveauthigenic carbonate precipitation is unlikely to signifi-cantly alter our traditional view of the global isotope massbalance during this period. Heavier d13C values in the Pro-terozoic relative to comparable Phanerozoic environmentscould, in theory, be linked to a decrease in the size of themarine DIC reservoir through time—resulting in less effec-tive isotopic buffering of pore fluid DIC in younger sedi-ments (Bartley and Kah, 2004). Alternatively, limitedpore fluid oxidants—lower sulfate concentrations and(or) a limiting ferric iron supply (with non-bioturbatedsediments supporting little to no continuous Fe re-oxida-tion)—may have led to more extensive precipitation of car-bonates in the methanic zone, which is typicallycharacterized by positive d13CDIC values (see Naehret al., 2007, and references therein). We explore this possi-bility below.

A simple early diagenetic model (Berner, 1964, 1971) canbe used to gauge the factors controlling the d13C of diage-netic carbonates precipitated in anoxic pore fluids. With thisapproach it is straightforward to illustrate and roughlyquantify the main processes that will control the pore fluid

DIC carbon isotope values, allowing exploration of thepotential influence of oxidant limitation and methanogenesison the d13C of the diagenetic carbonates of the Chuanling-gou Formation.

We calculate pore fluid [DIC] using a modified versionof the one-dimensional general diagenetic equation for thechemical evolution of a fluid phase within the sediment(G-model; Berner, 1964):

@C@t¼ Ds

@2C@z2� x

@C@zþ kG0 exp � k

xz

� �ð1Þ

where the terms on the right hand side denote mass conser-vation due to fluxes associated with diffusion (term 1),advection (term 2), and reaction (term 3). These fluxesare, in turn, controlled by the sediment diffusion coefficienta given species (Ds), the concentration gradient as a func-tion of depth in the sediment [oC/oz, where z is depth (posi-tive downward)], the sediment advection velocity (x), and afirst-order reaction term for the conversion of labile organicmatter to DIC. In the latter term, k represents the intrinsicrate of organic carbon remineralization, and G0 representsthe concentration of metabolically available organic carbonat the sediment–water interface. For simplicity, we assumea scaling between k and x generalized to anoxic mineraliza-tion during sulfate reduction (e.g., Tromp et al., 1995).Assuming steady state (oC/ot = 0) and the boundary condi-tion C(z =1) = C1, we obtain the analytical solution:

CðzÞ ¼ � x2G0

Dsk þ x2exp � k

xz

� �þ C1 ð2Þ

If we further specify the boundary condition that theDIC concentration at the sediment–water interface is equalto that of the overlying seawater [C(z = 0) = C0 =[DIC]SW], we can obtain a solution for C1, and this canbe combined with Eq. (2) above to obtain a solution for[DIC]total as a function of depth in the sediment:

CðzÞ ¼ � x2G0

Dsk þ x2exp � k

xz

� �þ x2G0

Dsk þ x2þ C0 ð3Þ

We can then take the difference between [DIC]total and[DIC]SW to determine [DIC]remin, and through simple massbalance determine the carbon isotope composition of thepore fluid DIC:

½DIC�totald13Ctotal ¼ ½DIC�remind

13Cremin þ ½DIC�SWd13CSW

ð4Þ

All values/ranges for model parameters are listed inTable 1.

The model results are given in Fig. 8. Three significantpoints emerge from this basic analysis. First, availabilityof initial utilizable organic carbon (i.e., G0) has a significantimpact on pore fluid d13CDIC values (Fig. 8A). Although itis difficult to constrain values for G0, even in modern sys-tems, bulk TOC values in the Chuanlinggou sedimentshosting the diagenetic beds (upper Subsection CLG5) arenear 1% (Fig. 3, Appendix). We thus consider a reasonableminimum estimate to be >1% for initial utilizable organiccarbon for the Chuanlinggou diagenetic carbonates.

Fig. 8. Modeling for d13C of dissolved inorganic carbon (DIC) in pore fluids. (A) Sensitivity analyses for d13C of DIC versus sedimentationrate with contours representing utilizable organic carbon (G0). (B) Sensitivity analyses for d13C of DIC versus sedimentation rate withcontours representing marine DIC concentrations. G0 is equal to 1%. (C) d13C of DIC with G0 of 1%, modern marine DIC concentrations, andx of 0.1 cm/yr. For all contour plots, results are taken from 3 m depth, assuming oxidants will be exhausted below this depth. The horizontalblack lines in (A) and (B) represent the average d13C of the diagenetic carbonates of the Chuanlinggou Formation. See text for detaileddescription of model setup.

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 99

Second, it is clear from Fig. 8A,B that sedimentationrates also have the capacity to cause large variations inthe d13C of pore fluid DIC. However, with reasonablesedimentation rates for the examined settings over shorttime scales (e.g., �0.1 cm/yr on sub-million year timescales;Sadler, 1981), and with modern [DIC] and G0 of 1%, itbecomes difficult to explain the average d13C of the diage-netic carbonates if carbonate precipitation happened onlyin the sulfate and iron reduction diagenetic zones(Fig. 8C). The data would thus seem to imply a non-trivialportion of diagenetic carbonate formation in the deepermethanic zone of the sediment column.

Third, changes in seawater [DIC] result in significantshifts in pore fluid d13CDIC values (Fig. 8B and A). Withvery high marine DIC concentrations, there is significantisotopic buffering capacity, making it increasingly difficultto drive pore fluid DIC to lighter d13C values. However, rea-sonable estimates of sedimentation rate (e.g., between 0.05and 0.1 cm/yr) and inputs of organic carbon (for the exam-ined sections) yield pore fluid d13CDIC values significantlylighter than the d13C of the examined diagenetic carbonates(Fig. 8A); this prediction is true even with very elevated DICconcentrations (>5�modern [DIC]; Fig. 8B). This leaves uswith the intriguing possibility that carbonate formation inthe methanic zone, likely linked to pore fluid oxidant limita-tion (see more below), was an important factor behind therelatively heavy d13C of the diagenetic carbonates. Consis-tent with this model, diagenetic carbonates from the extre-mely organic-rich Jurassic Kimmeridge Clay have averaged13C values near �3& because of carbonate precipitationin the methanic zone (Irwin et al., 1977).

Significant carbonate precipitation in the methanic zonein the Proterozoic may be linked with low marine oxidant(primarily sulfate) concentrations and less net sulfate reduc-tion compared to ‘typical’ modern continental margin sed-iments. In other words, if seawater sulfate concentrationswere a small fraction of modern levels during the mid-Pro-terozoic and quickly consumed in the sediments, a relatively

thin zone of sulfate reduction may have caused limited evo-lution of d13C of DIC. Within the zone of methanogenesis,following sulfate consumption, the d13C of DIC would beexpected to shift to heavier values, as isotopically light car-bon is preferentially utilized for methane formation fromthe existing DIC pool. In this case, the result is relatively13C-enriched authigenic carbonate phases due fundamen-tally to oxidant limitation in Proterozoic seawater. Ofcourse, oxygen would have been absent beneath the anoxicbottom waters, and a number of studies point to low sulfatein the ocean (e.g., Kah et al., 2004; Scott et al., 2014) andlow atmospheric O2 at this time (e.g., Planavsky et al.,2014). We propose that porewater oxidant limitation islikely to have prevented authigenic carbonates from beinga major sink—in terms of the global mass balance—for iso-topically light carbon.

There is currently a lack of strong empirical evidence forwidespread isotopically light carbonate burial in marginalmarine environments during the Proterozoic, but if authi-genic carbonates were buried primarily in deep water envi-ronments one might expect this record to be poorlypreserved (e.g., Canfield and Kump, 2013). Given thatdecreasing the sedimentation rate and initial utilizableorganic carbon concentrations results in heavier pore fluidd13CDIC values, it may be difficult for distal marine environ-ments to support significant burial fluxes of authigenic car-bonate with markedly light d13C values. Given thisframework, the search for evidence of a significant isotopi-cally light carbon sink in the Proterozoic should focus onwell-preserved continental shelf settings. However, carbon-ate precipitation in the methanic zone may be less impor-tant in offshore than marginal settings due to lowerrelative organic carbon loading. If this is the case the resultthat offshore marine settings (compared to more organic-rich settings) may be more suited for burial of isolateddepleted carbonate despite heavier porewater DIC valuesin the sulfate reduction zone of the sediment column. Itcould be possible to resolve between these competing

Fig. 9. Compilation of FePy/FeHR versus total organic carbon(TOC) content reported for the mid-Proterozoic strata. (A) Datasetincludes both oxic and anoxic samples. (B) Dataset includes onlyunambiguous anoxic samples (i.e., FeHR/FeT > 0.38). Coefficient(R2) between FePy/FeHR and TOC and t-test p value for the datasetare given in brackets for each strata unit in legend. Data source:1.84 Ga Rove/Virginia Fm. (Poulton et al., 2004, 2010), 1.73 GaWollogorang and < 1.64 Ga Reward Fms. (Shen et al., 2002),1.64 Ga Mt. Isa Superbasin, 1.45 Ga Newland Fm. and 1.2 GaBorden Basin (Planavsky et al., 2011).

100 C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105

models with a more complex, spatially resolved coupledearly diagenetic and ocean modeling approach.

4.5. Broader implications for the middle Proterozoic ocean

chemistry

Ocean chemistry and redox structure in the late Paleo-proterozoic (1.8–1.6 Ga) is particularly intriguing becauseit is believed to mark a fundamental transition of oceanredox state after which major iron formations (IF) largelydisappeared from the rock record for a billion years. As sta-ted in Section 1, transitions from ferruginous to mildlyoxic-suboxic (Holland, 2006; Slack et al., 2007, 2009) oreuxinic (Canfield, 1998) deep ocean waters were the twomajor mechanisms proposed previously to explain the glo-bal termination of large IF deposition. In particular, theeuxinia hypothesis received support from sedimentary Fespeciation, S isotope systematics, and lipid biomarker datacollected from mid-Proterozoic marine basins (e.g., Brockset al., 2005; Poulton et al., 2004; Shen et al., 2002, 2003). Incontrast, new Fe speciation data for the <1.84 Ga RoveFormation (Animikie Group, Lake Superior region) sug-gest mid-depth euxinic waters associated with deeper ferru-ginous waters (Poulton et al., 2010), which is an emergingidea for the Precambrian ocean more generally, includingthe late Archean (Reinhard et al., 2009; Scott et al., 2011)and Neoproterozoic (Canfield et al., 2008; Li et al., 2010,2012; Johnston et al., 2010; see Lyons et al., 2014, forreview). Our Fe–S–C–Mo data from the ChuanlinggouFormation indicate non-euxinic (i.e., ferruginous and possi-bly oxic) subsurface marine conditions for the deepest por-tion of the �1.65 Ga Yanshan basin. We suggest that thiscondition may have been due to insufficient water columnsulfate reduction rates (relative to reactive Fe deliveryfluxes). This suggestion is supported by the occurrences ofironstones in the Chuanlinggou Formation in Jiaxian area(Planavsky et al., 2014) and the west part of basin (e.g.,Xuanhua area) where economic ironstones were depositedand mined historically. This situation is in contrast to manymodern semi-restricted euxinic marine settings (e.g., theBlack Sea and Cariaco Basin) where both sulfate andorganic matter fluxes are not limiting factors for the devel-opment of euxinia. This observation is also in contrast tothe Cryogenian interglacial semi-restricted Nanhua Basin(South China) where euxinia developed in the subsurfacewaters of the basin when the post-Sturtian transgressionbrought nutrient-rich deep seawater from the open oceaninto the basin, resulting in higher productivity in its surfacewater and, in turn, in high export of organic matter intosubsurface water (Li et al., 2012).

We observed, however, that the extent of euxinia seems tobe decoupled from local organic matter loading in most mid-Proterozoic settings. A compilation of paired Fepy/FeHR andTOC data reported for numerous mid-Proterozoic stratashows no significant correlation between relative pyritizationand TOC content (Fig. 9A). This also is true when wescreened the dataset to only include samples deposited underanoxic conditions (Fig. 9B). The suggestion is that other

factors beyond just local organic matter loading (which arelinked to TOC values) control dissolved sulfide-ferrous ironratios. These controls include broader, regional patterns oforganic carbon delivery and related rates of BSR relativeto pervasive availability of sulfide-buffering dissolved Fe.Other work leaves open the possibility that sulfate couldhave been a limiting factor in the development of Proterozoiceuxinia (Li et al., 2010).

ical

dat

ao

fth

eC

hu

anli

ngg

ou

Fo

rmat

ion

atJi

xian

sect

ion

.

Dep

tha

(m)

TIC

(%)

TO

C(%

)A

l(%

)M

n(%

)F

e Py

(%)

Fe c

arb

(%)

Fe o

x

(%)

Fe m

ag

(%)

Fe H

R

(%)

Fe U

(%)

Fe T

(%)

Fe H

R/

Fe T

Fe P

y/

Fe H

R

Fe T

/A

ld3

4S

Py

(&)

d34S

CA

S

(&)

D34S

(&)

d13C

Carb

(&)

d13O

Carb

(&)

Mo

(pp

m)

Mo

/TO

C(p

pm

/wt%

)

l(

UI)

12.5

6.4

0.4

3.5

0.16

0.01

2.46

0.19

0.05

2.72

1.08

3.80

0.72

0.00

1.08

31.2

32.4

1.2�

3.2

�8.

10.

30.

711

.310

.30.

40.

80.

330.

042.

690.

090.

553.

371.

354.

720.

710.

015.

9422

.60.

20.

510

.27.

30.

53.

00.

210.

043.

010.

120.

183.

351.

004.

350.

770.

011.

4722

.50.

20.

58.

96.

50.

43.

90.

170.

022.

360.

130.

252.

751.

083.

840.

720.

011.

0029

.132

.53.

4�

3.2

�8.

20.

20.

67.

69.

20.

42.

30.

270.

073.

120.

120.

533.

841.

565.

400.

710.

022.

3624

.628

.74.

1�

3.1

�8.

00.

30.

66.

58.

80.

42.

40.

230.

043.

280.

100.

363.

780.

754.

530.

830.

011.

8638

.733

.7�

5.0�

3.2

�8.

30.

30.

75.

88.

30.

82.

50.

240.

263.

680.

100.

324.

361.

165.

520.

790.

062.

2030

.533

.42.

9�

3.6

�8.

70.

70.

93.

20.

53.

26.

70.

080.

340.

260.

030.

851.

483.

795.

270.

280.

230.

7932

.91.

90.

61.

40.

24.

37.

50.

191.

170.

180.

740.

802.

902.

375.

260.

550.

400.

7036

.11.

30.

30

10.1

0.4

1.7

0.25

0.11

3.74

0.13

0.59

4.57

1.04

5.61

0.81

0.02

3.32

22.5

27.3

4.8�

3.3

�8.

50.

82.

293

.64.

91.

25.

00.

030.

440.

640.

020.

061.

161.

012.

170.

530.

380.

4337

.435

.5�

1.9�

2.4

�12

.31.

51.

283

.64.

01.

46.

30.

050.

510.

880.

080.

051.

531.

763.

280.

460.

340.

5242

.937

.5�

5.4�

2.6

�10

.50.

50.

380

.44.

71.

45.

10.

040.

391.

050.

240.

051.

731.

142.

860.

600.

230.

5640

.4�

2.4

�11

.50.

50.

479

.74.

31.

75.

10.

030.

410.

560.

560.

061.

581.

212.

790.

570.

260.

5536

.90.

40.

272

.85.

71.

34.

10.

030.

270.

580.

480.

061.

381.

012.

390.

580.

190.

5940

.237

.2�

3.0�

2.3

�12

.50.

50.

471

.47.

40.

83.

00.

070.

281.

860.

490.

242.

870.

983.

840.

750.

101.

2634

.438

.23.

8�

2.8

�11

.10.

60.

7

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 101

5. SUMMARY AND CONCLUSIONS

Our Fe–S–C–Mo biogeochemical investigation of stratafrom the �1.65 Ga Chuanlinggou Formation yields resultsthat are consistent with non-euxinic water column condi-tions (ferruginous and possibly oxic) persisting in the sub-surface waters of the semi-restricted Yanshan rift basin—at least in the deepest Jixian area. The non-correlative rela-tionships between relative pyritization (represented byFepy/FeHR) and TOC observed for both the ChuanlinggouFormation and other previously studied mid-Proterozoicstrata support that other factors beyond just local organicmatter loading must be considered for euxinic developmentin mid-Proterozoic oceans. However, paired CAS-pyrite S-isotope data from the Chuanlinggou Formation suggest thepersistence of sulfate into the pore fluids, which is not con-sistent with insufficient sulfate for BSR in water column. Inaddition, despite evidence for deposition under anoxic con-ditions, there are not significant Mo enrichments in theChuanlinggou Formation. This observation supports theidea that the presence of appreciable dissolved sulfide inthe water column is a key control in Mo enrichment andthat the size of the marine Mo reservoir can be used togauge the extent of euxinic conditions specifically—ratherthan anoxic conditions more generally (cf. Reinhardet al., 2013).

Lastly, there appears to have been extensive diageneticcarbonate precipitation during the deposition of the Chu-anlinggou Formation, consistent with the idea of enhancedprecipitation in anoxic pore fluids during the Precambrian(Schrag et al., 2013). However, these carbonates have onlymoderately negative d13C values (�3.4 ± 1.4&), which welink, with support from a simple numerical model, to exten-sive precipitation of carbonates in the methanic zone withinthe sediments. Oxidant limitation (sulfate in particular) inthe pore fluids could have led to more extensive precipita-tion of carbonates in the methanic zone, which is character-ized by positive d13CDIC values.

ACKNOWLEDGEMENTS

We thank Ben Gill, Bill Gilhooly, Jeremy Owens, Jiaxin Yanand Meng Cheng for laboratory or technical assistance. We arealso grateful to Erik Sperling, Graham Shields-Zhou, and Associ-ate Editor David Johnston for their valuable comments. Thisresearch was supported by the MOE of China (Grant # NCET-11-0724) and the Chinese 973 Program (Grants # 2013CB955704and # 2011CB808800) to C.L. N.J.P. acknowledges support fromthe Geological Society of America the NSF-EAR-PF and theNSF-ELT program. T.W.L and G.D.L. acknowledge funding fromthe NASA Astrobiology Institute, as well as the NSF-EAR-ELTtrack and the NSF-FESD program.

APPENDIX

Tab

leA

Key

geo

chem

Sam

ple

Up

per

Inte

rva

CL

G1-

13C

LG

1-12

CL

G1-

11C

LG

1-10

CL

G1-

09C

LG

1-08

CL

G1-

07C

LG

1-03

CL

G1-

02C

LG

1-01

CL

G5-

53C

LG

5-45

CL

G5-

44C

LG

5-43

CL

G5-

39C

LG

5-38

CLG5-37 70.3 5.9 0.7 4.0 0.07 0.56 1.85 0.66 0.36 3.44 1.08 4.52 0.76 0.16 1.12 40.6 37.8 �2.8 �2.4 �11.8 1.8 2.5CLG5-36 68.3 7.2 1.3 3.1 0.08 0.12 1.39 0.79 0.09 2.39 0.68 3.07 0.78 0.05 0.98 40.6 0.7 0.6CLG5-35 66.9 5.1 1.2 4.8 0.04 0.23 0.71 0.40 0.06 1.39 1.08 2.46 0.56 0.16 0.52 41.6 0.7 0.6CLG5-33 64.3 6.5 1.0 3.5 0.05 0.23 0.77 0.35 0.06 1.41 0.71 2.11 0.67 0.16 0.60 42.6 2.8 2.8CLG5-32 63.5 6.1 1.1 3.8 0.05 0.15 0.70 0.73 0.08 1.67 0.73 2.39 0.70 0.09 0.63 40.5 6.2 5.7CLG5-31 62.4 5.4 1.6 4.7 0.05 0.45 0.70 0.16 0.04 1.35 1.05 2.40 0.56 0.33 0.51 38.1 �2.4 �11.8 1.1 0.7CLG5-29 60.4 6.0 1.2 4.2 0.07 0.23 1.03 0.52 0.09 1.86 0.95 2.81 0.66 0.12 0.67 39.6 37.5 �2.1 �2.5 �10.9 0.9 0.7CLG5-27 58.3 6.1 1.1 3.8 0.06 0.16 1.05 0.78 0.22 2.21 0.77 2.98 0.74 0.07 0.79 41.4 0.4 0.4CLG5-25 56.1 6.5 1.3 3.6 0.05 0.05 0.64 1.00 0.03 1.72 0.73 2.44 0.70 0.03 0.68 40.4 37.8 �2.6 �2.5 �11.3 1.0 0.8CLG5-24 57.6 9.2 1.3 1.6 0.11 0.06 2.08 0.42 0.10 2.66 0.80 3.46 0.77 0.02 2.11 36.6 1.0 0.7CLG5-23 56.5 7.9 1.3 2.7 0.09 0.05 1.65 0.55 0.07 2.32 0.76 3.08 0.75 0.02 1.16 42.8 0.2 0.2CLG5-22 55.5 5.9 1.5 4.0 0.03 0.08 0.34 0.77 0.03 1.23 0.78 2.01 0.61 0.07 0.50 39.8 �2.4 �12.3 0.2 0.1CLG5-21 53.9 7.2 1.1 3.0 0.04 0.05 0.35 0.72 0.03 1.16 0.71 1.87 0.62 0.04 0.62 40.2 0.3 0.2CLG5-19 49.6 7.5 1.1 3.2 0.09 0.08 1.56 0.75 0.11 2.51 0.79 3.30 0.76 0.03 1.03 40.6 1.3 1.2CLG5-16 43.7 8.4 0.8 2.7 0.16 0.04 2.52 0.77 0.22 3.55 1.32 4.87 0.73 0.01 1.77 43.6 40.2 �3.4 �6.4 �8.4 1.6 1.9CLG5-15B 43.2 8.4 0.8 2.7 0.15 0.07 2.49 0.66 0.17 3.39 1.06 4.45 0.76 0.02 1.67 43.6 37.8 �5.8 0.4 0.5CLG5-15A 43.2 8.7 0.8 2.5 0.16 0.06 2.60 0.63 0.11 3.41 1.21 4.62 0.74 0.02 1.83 43.3 37.8 �5.5 �6.5 �8.3 0.3 0.4CLG5-13 41.45 8.1 1.1 2.5 0.13 0.06 1.93 0.98 0.04 3.01 0.69 3.69 0.81 0.02 1.49 41.0 39.9 �1.1 �4.9 �8.4 2.0 1.8CLG5-12 40.9 9.0 0.8 2.2 0.16 0.01 2.42 0.82 0.07 3.33 0.76 4.09 0.81 0.00 1.86 41.6 34.8 �6.8 �5.9 �8.0 3.3 4.1CLG5-09 36 2.4 1.6 6.4 0.05 0.05 0.50 1.19 0.10 1.84 2.06 3.90 0.47 0.03 0.61 40.4 0.5 0.3Lower Interval (LI)

CLG5-08B 10.4 0.3 1.1 7.4 0.02 0.63 0.30 0.04 0.18 1.15 4.08 5.23 0.22 0.55 0.71 38.7 0.5 0.4CLG5-08A 9.6 0.1 1.3 8.0 0.01 0.37 0.14 0.01 0.15 0.67 3.80 4.47 0.15 0.56 0.56 38.2 0.2 0.2CLG5-03 2.8 0.0 1.0 8.1 0.01 0.21 0.06 0.11 0.18 0.57 3.87 4.44 0.13 0.37 0.55 36.5 0.3 0.3CLG5-02 0.8 0.0 0.9 8.2 0.01 0.05 0.08 0.09 0.21 0.42 4.40 4.83 0.09 0.11 0.59 36.9 0.2 0.2CLG5-01 0 0.0 0.8 7.0 0.01 0.03 0.12 0.02 0.21 0.38 3.62 3.99 0.09 0.07 0.57 40.2 6.6 8.2CLG4-05 37.7 1.0 0.8 6.5 0.10 0.06 1.31 0.18 1.52 3.08 11.53 14.61 0.21 0.02 2.24 26.2 0.2 0.3CLG4-03 37.5 0.0 0.3 8.7 0.02 0.08 0.17 0.25 0.30 0.79 3.87 4.66 0.17 0.10 0.54 28.2 0.2 0.7CLG2-14 43.2 1.4 0.8 6.1 0.06 0.29 0.98 0.27 0.16 1.70 2.51 4.21 0.40 0.17 0.69 38.0 0.3 0.3CLG2-12 42.4 0.2 0.7 7.5 0.01 0.36 0.19 0.13 0.05 0.73 2.26 2.99 0.24 0.50 0.40 36.1 0.5 0.8CLG2-10 41.1 0.6 0.5 7.3 0.03 0.09 0.35 0.01 0.11 0.55 3.06 3.61 0.15 0.16 0.49 39.8 0.4 0.7CLG2-09 40.5 0.0 0.2 7.7 0.01 0.26 0.10 0.03 0.12 0.51 2.94 3.45 0.15 0.52 0.45 37.5 0.3 1.3CLG2-08 3.8 0.8 0.4 7.2 0.04 0.24 0.64 0.05 0.22 1.15 3.67 4.81 0.24 0.21 0.67 35.9 0.4 1.0CLG2-06 3.2 0.1 0.4 7.7 0.01 0.20 0.21 0.03 0.19 0.63 3.62 4.25 0.15 0.32 0.55 36.7 0.5 1.3CLG2-05 2.7 0.3 0.4 6.9 0.02 0.23 0.35 0.02 0.19 0.78 3.45 4.23 0.19 0.29 0.61 35.1 0.3 0.6CLG2-04B 1.6 0.0 0.4 7.8 0.01 0.17 0.11 0.09 0.15 0.51 3.72 4.24 0.12 0.33 0.55 36.6 0.4 1.0CLG2-04A 1.8 0.8 0.9 7.0 0.04 0.37 0.66 0.05 0.30 1.38 3.88 5.26 0.26 0.27 0.75 33.0 0.6 0.7CLG2-03 1.3 0.0 0.5 7.0 0.00 0.35 0.12 0.02 0.16 0.65 2.98 3.63 0.18 0.54 0.52 34.7 0.3 0.6CLG2-02 0.7 1.5 1.0 6.0 0.06 0.41 0.96 0.07 0.20 1.63 3.62 5.25 0.31 0.25 0.88 32.2 0.4 0.4CLG2-01 0 0.2 0.6 8.0 0.02 0.35 0.21 0.06 0.07 0.69 2.77 3.45 0.20 0.51 0.43 34.9 0.3 0.5

a Depths are relative to zero at the bottom of each subsection.

102C

.L

iet

al./G

eoch

imica

etC

osm

och

imica

Acta

150(2015)

90–105

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 103

REFERENCES

Algeo T. J., Chen Z., Fraiser M. L. and Twitchett R. J. (2011)Terrestrial–marine teleconnections in the collapse and rebuild-ing of Early Triassic marine ecosystems. Palaeogeogr. Palaeo-

climatol. Palaeoecol. 308, 1–11.Anbar A. D. and Knoll A. H. (2002) Proterozoic ocean chemistry

and evolution: a bioinorganic bridge. Science 297, 1137–1142.Asael D., Tissot F. L. H., Reinhard C. T., Rouxel O., Dauphas N.,

Lyons T. W., Ponzevera E., Liorzou C. and Cheron S. (2013)Coupled molybdenum, iron and uranium stable isotopes asoceanic paleoredox proxies during the Paleoproterozoic ShungaEvent. Chem. Geol. 362, 193–210.

Balci N., Shanks, III, W. C., Mayer B. and Mandernack K. W.(2007) Oxygen and sulfur isotope systematics of sulfateproduced by bacterial and abiotic oxidation of pyrite. Geochim.

Cosmochim. Acta 71, 3796–3811.Bartley J. K. and Kah L. C. (2004) Marine carbon reservoir, Corg–

Ccarb coupling, and the evolution of the Proterozoic carboncycle. Geology 32, 129–132.

Bekker A., Slack J. F., Planavsky N., Krapez B., Hofmann A.,Konhauser K. O. and Rouxel O. J. (2010) Iron formation: thesedimentary product of a complex interplay among mantle,tectonic, oceanic, and biospheric processes. Econ. Geol. 105,467–508.

Berner R. A. (1964) An idealized model of dissolved sulfatedistribution in recent sediments. Geochim. Cosmochim. Acta 28,1497–1503.

Berner R. A. (1971) Principles of Chemical Sedimentology.McGraw-Hill, New York.

Boudreau B. P. (1996) The diffusive tortuosity of fine-grainedunlithified sediments. Geochim. Cosmochim. Acta 60, 3139–3142.

Brocks J. J., Love G. D., Summons R. E., Knoll A. H., Logan G.A. and Bowden S. A. (2005) Biomarker evidence for green andpurple sulphur bacteria in a stratified Palaeoproterozoic sea.Nature 437, 866–870.

Burdett J. W., Arthur M. A. and Richardson M. (1989) A Neogeneseawater sulfur isotope age curve from calcareous pelagicmicrofossils. Earth Planet. Sci. Lett. 94, 189–198.

Canfield D. E. (1998) A new model for Proterozoic oceanchemistry. Nature 396, 450–453.

Canfield D. E. and Farquhar J. (2009) Animal evolution, biotur-bation, and the sulfate concentration of the oceans. Proc. Natl.

Acad. Sci. USA 106, 8123–8127.Canfield D. E. and Kump L. R. (2013) Carbon cycle makeover.

Science 339, 533–534.Canfield D. E., Raiswell R., Westrich J. T., Reaves C. M. and

Berner R. A. (1986) The use of chromium reduction in theanalysis of reduced inorganic sulfur in sediments and shales.Chem. Geol. 54, 149–155.

Canfield D. E., Poulton S. W., Knoll A. H., Narbonne G. M., RossG., Goldberg T. and Strauss H. (2008) Ferruginous conditionsdominated later Neoproterozoic deep-water chemistry. Science

321, 949–952.Chen J. B., Zhang H. M., Zhu S. X., Zhao Z. and Wang Z. G. (1980)

Research on Sinian Suberathem of Jixian Tianjin (in Chinese). InResearch on Precambrian Geology: Sinian Suberathem in China

(eds. Tianjin Institute of Geology and Mineral Resources andCAGS). Tianjin Sci. & Tech. Press, Tianjin, pp. 56–114.

Chu X. L., Zhang T. G., Zhang Q. R. and Lyons T. W. (2007)Sulfur and carbon isotope records from 1700 to 800 Macarbonates of the Jixian section, northern China: implicationsfor secular isotope variations in Proterozoic seawater andrelationships to global supercontinental events. Geochim.

Cosmochim. Acta 71, 4668–4692.

Clarkson M. O., Poulton S. W., Guilbaud R. and Wood R. A.(2014) Assessing the utility of Fe/Al and Fe-speciation to recordwater column redox conditions in carbonate-rich sediments.Chem. Geol.. http://dx.doi.org/10.1016/j.chemgeo.2014.05.031.

Feng L., Li C., Huang J., Chang H. and Chu X. (2014) A sulfatecontrol on marine mid-depth euxinia on the early Cambrian(ca. 529–521 Ma) Yangtze platform, South China. Precambrian

Res. 246, 123–133.Gao L. Z., Zhang C. H., Yin C. Y., Shi X. Y., Wang Z. Q., Liu Y.

M., Liu P. J., Tang F. and Song B. (2008) SHRIMP zircon ages:basis for refining the chronostratigraphic classification of theMeso- and Neoproterozoic strata in North China Old Land.Acta Geosci. Sin. 29, 366–376.

Gao L. Z., Zhang C. H., Liu P. J., Ding X. Z., Wang Z. Q. andZhang Y. J. (2009) Recognition of Meso- and Neoproterozoicstratigraphic framework in North and South China. Acta

Geosci. Sin. 30, 433–446.Gellatly A. M. and Lyons T. W. (2005) Trace sulfate in mid-

Proterozoic carbonates and the sulfur isotope record ofbiospheric evolution. Geochim. Cosmochim. Acta 69, 3813–3829.

Gill B. C., Lyons T. W. and Saltzman M. R. (2007) Parallel, high-resolution carbon and sulfur isotope records of the evolvingPaleozoic marine sulfur reservoir. Palaeogeogr. Palaeoclimatol.

Palaeoecol. 256, 156–173.Gill B. C., Lyons T. W. and Frank T. D. (2008) Behavior of

carbonate-associated sulfate during meteoric diagenesis andimplications for the sulfur isotope paleoproxy. Geochim.

Cosmochim. Acta 72, 4699–4711.Hammarlund E. U., Dahl T. W., Harper D. A. T., Bond D. P. G.,

Nielsen A. T., Bjerrum C. J., Schovsbo N. H., Schonlaub H. P.,Zalasiewicz J. A. and Canfield D. E. (2012) A sulfidic driver forthe end-Ordovician mass extinction. Earth Planet. Sci. Lett.

331–332, 128–139.Helz G. R., Miller C. V., Charnock J. M., Mosselmans J. F. W.,

Pattrick R. A. D., Garner C. D. and Vaughan D. J. (1996)Mechanism of molybdenum removal from the sea and itsconcentration in black shales: EXAFS evidence. Geochim.

Cosmochim. Acta 60, 3631–3642.Helz G. R., Bura-Nakic E., Mikac N. and Ciglenecki I. (2011) New

model for molybdenum behavior in euxinic waters. Chem. Geol.

284, 323–332.Higgins J. A., Fischer W. W. and Schrag D. P. (2009) Oxygenation

of the ocean and sediments: consequences for the seafloorcarbonate factory. Earth Planet. Sci. Lett. 284, 25–33.

Holland H. D. (2006) The oxygenation of the atmosphere andoceans. Philos. Trans. R. Soc. B 361, 903–915.

Irwin H., Curtis C. and Coleman M. (1977) Isotopic evidence forsource of diagenetic carbonates formed during burial oforganic-rich sediments. Nature 269, 209–213.

Isley A. E. and Abbott D. H. (1999) Plume-related mafic volcanismand the deposition of banded iron formation. J. Geophys. Res.

104, 15461–15477.Jin C., Li C., Peng X., Cui H., Shi W., Zhang Z., Luo G. and Xie S.

(2014) Spatiotemporal variability of ocean chemistry in theearly Cambrian, South China. Sci. China Earth Sci. 57, 579–591.

Johnston D. T., Poulton S. W., Dehler C., Porter S., Husson J.,Canfield D. E. and Knoll A. H. (2010) An emerging picture ofNeoproterozoic ocean chemistry: insights from the ChuarGroup, Grand Canyon, USA. Earth Planet. Sci. Lett. 290,64–73.

Johnston D. T., Poulton S. W., Tosca N. J., O’Brien T., HalversonG. P., Schrag D. P. and Macdonald F. A. (2013) Searching foran oxygenation event in the fossiliferous Ediacaran of north-western Canada. Chem. Geol. 362, 273–286.

104 C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105

Kah L. C., Lyons T. W. and Frank T. D. (2004) Low marinesulphate and protracted oxygenation of the Proterozoic bio-sphere. Nature 431, 834–838.

Kendall B., Gordon G. W., Poulton S. W. and Anbar A. D. (2011)Molybdenum isotope constraints on the extent of late Paleo-proterozoic ocean euxinia. Earth Planet. Sci. Lett. 307, 450–460.

Kump L. R. and Seyfried W. E. (2005) Hydrothermal Fe fluxesduring the Precambrian: effect of low oceanic sulfate concen-trations and low hydrostatic pressure on the composition ofblack smokers. Earth Planet. Sci. Lett. 235, 654–662.

Kusky T. M. and Li J. H. (2003) Paleoproterozoic tectonicevolution of the North China Craton. J. Asian Earth Sci. 22,383–397.

Li C., Love G. D., Lyons T. W., Fike D. A., Sessions A. L. andChu X. (2010) A stratified redox model for the Ediacaranocean. Science 328, 80–83.

Li C., Love G. D., Lyons T. W., Scott C. T., Feng L. J., Huang J.,Chang H. J., Zhang Q. R. and Chu X. L. (2012) Evidence for aredox stratified Cryogenian marine basin, Datangpo Forma-tion, South China. Earth Planet. Sci. Lett. 331, 246–256.

Li H. K., Su W. B., Zhou H. Y., Geng J. Z., Xiang Z. Q., Cui Y.R., Liu W. C. and Lu S. N. (2011) The base age of theChangchengian System at the northern North China Cratonshould be younger than 1670 Ma: constraints from zircon U–PbLA-MC-ICPMS dating of a granite-porphyry dike in MiyunCounty, Beijing. Earth Sci. Front. 18, 108–120.

Loyd S. J., Berelson W. M., Lyons T. W., Hammond D. E. andCorsetti F. A. (2012) Constraining pathways of microbialmediation for carbonate concretions of the Miocene MontereyFormation using carbonate-associated sulfate. Geochim. Cos-

mochim. Acta 78, 77–98.Lu S. N. and Li H. M. (1991) A precise U–Pb single zircon age

determination for the volcanics of the Dahongyu Formation,Changcheng System in Jixian. Bull. Chin. Acad. Geol. Sci. 22,137–146.

Lyons T. W. and Severmann S. (2006) A critical look at ironpaleoredox proxies: new insights from modern euxinic marinebasins. Geochim. Cosmochim. Acta 70, 5698–5722.

Lyons T. W., Werne J. P., Hollander D. J. and Murray R. W.(2003) Contrasting sulfur geochemistry and Fe/Al and Mo/Alratios across the last oxic-to-anoxic transition in the CariacoBasin, Venezuela. Chem. Geol. 195, 131–157.

Lyons T. W., Reinhard C. T., Love G. D. and Xiao S. (2012)Geobiology of the Proterozoic Eon. In Fundamentals of

Geobiology (eds. A. H. Knoll, D. E. Canfield and K. O.Konhauser). John Wiley & Sons Ltd, pp. 371–402.

Lyons T. W., Reinhard C. T. and Planavsky N. J. (2014) The riseof oxygen in Earth’s early ocean and atmosphere. Nature 506,307–315.

Macdonald F. A., Strauss J. V., Sperling E. A., Halverson G. P.,Narbonne G. M., Johnston D. T., Kunzmann M., Schrag D. P.and Higgins J. A. (2013) The stratigraphic relationship betweenthe Shuram carbon isotope excursion, the oxygenation ofNeoproterozoic oceans, and the first appearance of the Edia-cara biota and bilaterian trace fossils in northwestern Canada.Chem. Geol. 362, 250–272.

Meister P., McKenzie J. A., Vasconcelos C., Bernasconi S., FrankM., Gutjahr M. and Schrag D. P. (2007) Dolomite formation inthe dynamic deep biosphere: results from the Peru Margin.Sedimentology 54, 1007–1032.

Meng Q. R., Wei H. H., Qu Y. Q. and Ma S. X. (2011)Stratigraphic and sedimentary records of the rift to driftevolution of the northern North China craton at the Paleo- toMesoproterozoic transition. Gondwana Res. 20, 205–218.

Naehr T. H., Eichhubl P., Orphan V. J., Hovland M., Paull C. K.,Ussler, III, W., Lorenson T. D. and Greene H. G. (2007)Authigenic carbonate formation at hydrocarbon seeps incontinental margin sediments: a comparative study. Deep Sea

Res. Part II 54, 1268–1291.Peng P., Liu F., Zhai M. G. and Guo J. H. (2012) Age of the

Miyun dyke swarm: constraints on the maximum depositionalage of the Changcheng System. Chin. Sci. Bull. 57, 105–110.

Peng T., Bao H., Pratt L. M., Kaufman A. J., Jiang G., Boyd D.,Wang Q., Zhou X., Yuan C., Xiao S. and Loyd S. (2014)Widespread contamination of carbonate-associated sulfate bypresent-day secondary atmospheric sulfate: evidence from tripleoxygen isotopes. Geology 42, 815–818.

Planavsky N. J., McGoldrick P., Scott C. T., Li C., Reinhard C. T.,Kelly A. E., Chu X., Bekker A., Love G. D. and Lyons T. W.(2011) Widespread iron-rich conditions in the mid-Proterozoicocean. Nature 477, 448–451.

Planavsky N. J., Bekker A., Hofmann A., Owens J. D. and LyonsT. W. (2012) Sulfur record of rising and falling marine oxygenand sulfate levels during the Lomagundi Event. Proc. Natl.

Acad. Sci. USA 109, 18300–18305.Planavsky N. J., Reinhard C. T., Wang X., Thomson D.,

McGoldrick P., Rainbird R. H., Johnson T., Fischer W. W.and Lyons T. W. (2014) Low Mid-Proterozoic atmosphericoxygen levels and the delayed rise of animals. Science 346, 635–638.

Poulton S. W. and Canfield D. E. (2005) Development of asequential extraction procedure for iron: implications for ironpartitioning in continentally derived particulates. Chem. Geol.

214, 209–221.Poulton S. W. and Canfield D. E. (2011) Ferruginous conditions: a

dominant feature of the ocean through Earth’s history.Elements 7, 107–112.

Poulton S. W. and Raiswell R. (2002) The low-temperaturegeochemical cycle of iron: from continental fluxes to marinesediment deposition. Am. J. Sci. 302, 774–805.

Poulton S. W., Fralick P. W. and Canfield D. E. (2004) Thetransition to a sulphidic ocean �1.84 billion years ago. Nature

431, 173–177.Poulton S. W., Fralick P. W. and Canfield D. E. (2010) Spatial

variability in oceanic redox structure 1.8 billion years ago. Nat.

Geosci. 3, 486–490.Qiao X. F. (2002) Intraplate seismic belt and basin framework of

Sino-Korean Plate in Proterozoic. Earth Sci. Front. 9, 141–149.Qiao X. F. and Gao L. Z. (2007) Mesoproterozoic paleoearth-

quake and paleogeography in Yan-Liao Aulacogen. J. Palae-

ogeogr. 9, 337–352.Raiswell R. and Canfield D. E. (1998) Sources of iron for pyrite

formation in marine sediments. Am. J. Sci. 298, 219–245.Raiswell R., Newton R. and Wignall P. B. (2001) An indicator of

water-column anoxia: resolution of biofacies variations in theKimmeridge Clay (Upper Jurassic, UK). J. Sediment. Res. 71,286–294.

Raiswell R., Newton R., Bottrell S. H., Coburn P. M., Briggs D. E.G., Bond D. P. G. and Poulton S. W. (2008) Turbiditedepositional influences on the diagenesis of Beecher’s TrilobiteBed and the Hunsruck Slate; sites of soft tissue pyritization.Am. J. Sci. 308, 105–129.

Reinhard C. T., Raiswell R., Scott C., Anbar A. D. and Lyons T.W. (2009) A late Archean sulfidic sea stimulated by earlyoxidative weathering of the continents. Science 326, 713–716.

Reinhard C. T., Planavsky N. J., Robbins L. J., Partin C. A., GillB. C., Lalonde S. V., Bekkerd A., Konhauser K. O. and LyonsT. W. (2013) Proterozoic ocean redox and biogeochemicalstasis. Proc. Natl. Acad. Sci. USA 110, 5357–5362.

C. Li et al. / Geochimica et Cosmochimica Acta 150 (2015) 90–105 105

Sadler P. M. (1981) Sediment accumulation rates and thecompleteness of stratigraphic sections. J. Geol. 89, 569–584.

Scholz F., Severmann S., McManus J. and Hensen C. (2014)Beyond the Black Sea paradigm: the sedimentary fingerprint ofan open-marine iron shuttle. Geochim. Cosmochim. Acta 127,368–380.

Schrag D. P., Higgins J. A., Macdonald F. A. and Johnston D. T.(2013) Authigenic carbonate and the history of the globalcarbon cycle. Science 339, 540–543.

Scott C. and Lyons T. W. (2012) Contrasting molybdenum cyclingand isotopic properties in euxinic versus non-euxinic sedimentsand sedimentary rocks: refining the paleoproxies. Chem. Geol.

324–325, 19–27.Scott C., Lyons T. W., Bekker A., Shen Y., Poulton S. W., Chu X.

and Anbar A. D. (2008) Tracing the stepwise oxygenation ofthe Proterozoic ocean. Nature 452, 456–459.

Scott C. T., Bekker A., Reinhard C. T., Schnetger B., Krapez B.,Rumble, III, D. and Lyons T. W. (2011) Late Archean euxinicconditions before the rise of atmospheric oxygen. Geology 39,119–122.

Scott C., Wing B. A., Bekker A., Planavsky N. J., Medvedev P.,Bates S. M., Yun M. and Lyons T. W. (2014) Pyrite multiple-sulfur isotope evidence for rapid expansion and contraction ofthe early Paleoproterozoic seawater sulfate reservoir. Earth

Planet. Sci. Lett. 389, 95–104.Shen Y., Canfield D. E. and Knoll A. H. (2002) Middle Proterozoic

ocean chemistry: evidence from the McArthur Basin, northernAustralia. Am. J. Sci. 302, 81–109.

Shen Y., Knoll A. H. and Walter M. R. (2003) Evidence for lowsulphate and anoxia in a mid-Proterozoic marine basin. Nature

423, 632–635.Shi X., Jiang G., Zhang C., Gao L. and Liu J. (2008) Sand veins

and MISS from the Mesoproterozoic black shale (ca. 1.7 Ga) inNorth China: implication for methane degassing from micro-bial mats. Sci. China, Ser. D Earth Sci. 51, 1525–1536.

Slack J. F., Grenne T., Bekker A., Rouxel O. J. and Lindberg P. A.(2007) Suboxic deep seawater in the late Paleoproterozoic:evidence from hematitic chert and iron formation related toseafloor-hydrothermal sulfide deposits, central Arizona, USA.Earth Planet. Sci. Lett. 255, 243–256.

Slack J. F., Grenne T. and Bekker A. (2009) Seafloor-hydrothermalSi–Fe–Mn exhalites in the Pecos greenstone belt, New Mexico,and the redox state of ca. 1720 Ma deep seawater. Geosphere 5,302–314.

Sperling E. A., Rooney A. D., Hays L., Sergeev V. N., Vorob’evaN. G., Sergeeva N. D., Selby D., Johnston D. T. and Knoll A.

H. (2014) Redox heterogeneity of subsurface waters in theMesoproterozoic ocean. Geobiology 12, 373–386.

Strauss H. (1993) The sulfur isotopic record of Precambriansulfates: new data and a critical evaluation of the existingrecord. Precambrian Res. 63, 225–246.

Taylor B. E., Wheeler M. C. and Nordstrom D. K. (1984) Isotopecomposition of sulphate in acid mine drainage as measure ofbacterial oxidation. Nature 308, 538–541.

Tromp T. K., Van Cappellen P. and Key R. M. (1995) A globalmodel for early diagenesis of organic carbon and organicphosphorus in marine sediments. Geochim. Cosmochim. Acta

59, 1259–1284.Turekian K. K. and Wedepohl K. H. (1961) Distribution of the

elements in some major units of the Earth’s crust. Geol. Soc.

Am. Bull. 72, 175–191.Wan Y. S., Zhang Q. D. and Song T. R. (2003) SHRIMP ages of

detrital zircons from the Changcheng System in the MingTombs area Beijing: constraints on the protolith nature andmaximum depositional age of the Mesoproterozoic cover of theNorth China Craton. Chin. Sci. Bull. 48, 2500–2506.

Wilde P. (1987) Model of progressive ventilation of the latePrecambrian-early Paleozoic ocean. Am. J. Sci. 287, 442–459.

Wotte T., Shields-Zhou G. A. and Strauss H. (2012) Carbonate-associated sulfate: experimental comparisons of commonextraction methods and recommendations toward a standardanalytical protocol. Chem. Geol. 326, 132–144.

Xu D. B., Wang D. Z., Bai Z. D., Mei M. X. and Li Z. Z. (2002)Sedimentary environment and facies model of the Mesoprote-rozoic Chuanlinggou Formation in the Xinglong area, Hebei.Geol. China 29, 167–171.

Yan Y. Z. and Liu Z. L. (1998) The relationship betweenbiocommunities and Paleoenvironments in Changcheng Periodof the Yanshan Basin, North China. Acta MicroPaleontol. Sin.

15, 249–266.Zhang S., Li Z. X., Evan D. A. D., Wu H., Li H. and Dong J.

(2012) Pre-Rodinia supercontinent Nuna shaping up: a globalsynthesis with new paleomagnetic results from North China.Earth Planet. Sci. Lett. 353–354, 145–155.

Zhu Z., Aller R. C. and Mak J. (2002) Stable carbon isotopecycling in mobile coastal muds of Amapa, Brazil. Cont. Shelf

Res. 22, 2065–2079.

Associate editor: David Johnston