A continental- control on orbitally driven - cycling during Cretaceous Oceanic 2

Simon W. Poulton1, Susann Henkel2, Christian März3, Hannah Urquhart3, Sascha Flögel4, Sabine Kasten2, Jaap S. Sinninghe Damsté5, and Thomas Wagner3 1School of Earth and Environment, University of Leeds, Leeds LS2 9JT, UK 2Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, Germany 3School of Civil Engineering and Geosciences, Newcastle University, Drummond Building, Newcastle upon Tyne NE1 7RU, UK 4Helmholtz-Zentrum für Ozeanforschung Kiel (GEOMAR), Wischhofstrasse 1-3, 24148 Kiel, Germany 5NIOZ Royal Netherlands Institute for Sea Research, Department of Marine Organic Biogeochemistry, P.O. Box 59, NL-1790 AB Den Burg (Texel), Netherlands

ABSTRACT cycles broadly relate to fluctuations in total The Cretaceous period (~145–65 m.y. ago) was characterized by intervals of enhanced or- organic carbon (TOC) content (Kuhnt et al., ganic carbon burial associated with increased under greenhouse condi- 2005). Obliquity forcing has recently been sug- tions. The global consequences of these perturbations, oceanic anoxic events (OAEs), lasted up gested to be the dominant driver of this cyclicity, to 1 m.y., but short-term nutrient and climatic controls on widespread anoxia are poorly under- with eccentricity forcing possibly being a sec- stood. Here, we present a high-resolution reconstruction of oceanic redox and nutrient cycling ondary component in parts of the record (Mey- as recorded in subtropical shelf from Tarfaya, Morocco, spanning the initiation of ers et al., 2012). OAE2. Iron-sulfur systematics and biomarker evidence demonstrate previously undescribed We focus on the onset of OAE2, as docu- redox cyclicity on orbital time scales, from sulfidic to anoxic ferruginous (Fe-rich) water-col- mented in records of centimeter-scale (millen- umn conditions. Bulk geochemical data and sulfur isotope modeling suggest that ferruginous nial-scale) resolution from core S57 obtained conditions were not a consequence of nutrient or limitation, despite overall low sulfate from near the shelf basin center (Kuhnt et al., concentrations in the proto–North Atlantic. Instead, fluctuations in the weathering influxes of 2005). We utilize Fe-based redox proxies and sulfur and reactive iron, linked to a dynamic hydrological cycle, likely drove the redox cyclic- molecular biomarkers to provide a detailed eval- ity. Despite the potential for elevated phosphorus burial in association with Fe oxides under uation of oceanic redox conditions, and subse- ferruginous conditions on the Tarfaya shelf, porewater sulfide generation drove extensive phos- quently investigate controls on these conditions. phorus recycling back to the water column, thus maintaining widespread open-ocean anoxia. Finally, we evaluate the behavior of phosphorus cycling during redox fluctuations on the Tarfaya INTRODUCTION (März et al., 2008), whereby efficient recycling shelf to provide new mechanistic insight into the Major perturbations to the global Earth of phosphorus to the water column during eux- maintenance of widespread anoxia during OAE2. system occurred during the mid-Cretaceous, inic periods (positive productivity feedback) resulting in repetitive d13C isotope excursions contrasts with extensive burial of water-column METHODS in organic carbon and carbonate linked to en- phosphorus during ferruginous intervals (nega- hanced organic carbon burial (Jenkyns, 2010). tive productivity feedback). If prevalent on a Geochemical Analyses Although the precise driving mechanisms var- basinal or global scale, the development of fer- Water-column redox conditions were evalu- ied for each of these perturbations, extreme ruginous conditions with associated phosphorus ated using Fe speciation and molecular biomarker

greenhouse conditions were a common feature, burial would have had major implications for analyses. A “highly reactive” Fe fraction (FeHR) leading to enhanced hydrological cycling and the persistence of elevated marine productivity was quantified via a calibrated extraction scheme

oceanic nutrient (phosphorus) inputs, particu- and widespread anoxia. To date, however, no which includes carbonate-associated Fe (Fecarb),

larly in equatorial regions (Wagner et al., 2013). other studies have evaluated the potential for ferric (oxyhydr)oxides (Feox), magnetite (Femag),

Coupled with more restricted basinal conditions rapid redox cycling between euxinic and ferru- and (Fepy) (FeHR = Fecarb + Feox + Femag

and limited ocean circulation, enhanced primary ginous conditions during any of the Cretaceous + Fepy; Poulton and Canfield, 2005). Modern and production promoted extensive carbon burial, OAEs, and hence the prevalence, controls, and ancient sediments deposited from

ultimately resulting in the widespread develop- implications of such conditions are unknown. commonly have FeHR/total Fe (FeT) ratios >0.38, ment of anoxic oceanic conditions (Trabucho To address this, we present a redox and nu- in contrast to oxic depositional conditions, where Alexandre et al., 2010; Monteiro et al., 2012). trient reconstruction of an expanded shallow- ratios are consistently below this level (Poulton Redox-sensitive element and biomarker marine black section of Cenomanian- and Canfield, 2011). To provide further insight records imply widespread euxinic (sulfidic) Turonian age (OAE2) from the northwest into the chemical nature of an anoxic water col-

conditions during these oceanic anoxic events African shelf at Tarfaya, Morocco (see the GSA umn, the extent to which the FeHR pool is pyri- 1 (OAEs), which intermittently extended from Data Repository for details of the geologic set- tized distinguishes euxinic conditions (Fepy/FeHR bottom waters into the lower (Sin- ting). A positive organic carbon isotope excur- > 0.7) from anoxic, ferruginous water-column

ninghe Damsté and Köster, 1998; Hetzel et al., sion of ~3‰ (Fig. 1) marks the onset of OAE2 conditions (Fepy/FeHR < 0.7). 2009). There is evidence to suggest, however, (Tsikos et al., 2004), while distinct sedimentary Pyrite sulfur isotope compositions were de-

that euxinic conditions fluctuated with ferrugi- termined on Ag2S precipitates obtained through nous conditions on orbital time scales during 1 GSA Data Repository item 2015320, methods the Fe speciation techniques. Carbonate-associ- OAE3 (Coniacian-Santonian) in the deep-sea and data, including geologic setting, model parame- ated sulfate (CAS) was determined according to ters and mineralogical analyses, is available online at proto–North Atlantic (März et al., 2008). The www.geosociety.org/pubs/ft2015.htm, or on request refined techniques, and biomarkers for identify- OAE3 black highlight a classic effect from [email protected] or Documents Secre- ing photic zone euxinia (isorenieratane) were of redox fluctuations on phosphorus cycling tary, GSA, P.O. Box 9140, Boulder, CO 80301, USA. quantified by gas chromatography and gas chro-

GEOLOGY, November 2015; v. 43; no. 11; p. 963–966 | Data Repository item 2015320 | doi:10.1130/G36837.1 | Published online 24 September 2015 ©GEOLOGY 2015 Geological | Volume Society 43 | ofNumber America. 11 Gold | www.gsapubs.org Open Access: This paper is published under the terms of the CC-BY license. 963

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/43/11/963/3548007/963.pdf by guest on 25 September 2021 TOC (%) Fepy/FeHR µg Isoren./TOC ∆34S (‰) Sulfate Conc. (mM) 15 25 35 45 02468 510152050.2 0.4 0.6 0.8 1 0 10 15 50.0 Scenario 3 50.0 ++ + + ++ + + + + ++ ++ + + + Scenario 2 + +++ + + + + + + + + ++ + +++ + + ++ ++ pyrite + + + + + + + + + + + + + + 50.5 + +++ + + + + + + + + +++++ + + + ++ ++ + +++++ + + + + + + +++ + + + + + + + + + +++ ++ + + + 50.5 + +++ + + +++ + ++ + ++ ++ + ++ ++++ + ++ +++ +++ 51.0 + ++++ ++ + ++ + +++ +++ +++ + + ++++ + + + + + + +++ + + + +++ ++ + Cycle 1 ++++ + + + + + + + ++++ + + + ++ + + + + + + + + 51.5 51.0 ++ + + + + + + + + CAS +++++ +++ + + + + + + this study + ++ ++ + + + + + + + + ++++ ++ ++ + + ++ + +++ + + ++ ++ + +++++ + + ++ + + + ++ ++ + + ++ + + + +++ ++ + + + + 52.0 + +++ ++ + ++ + + ++ + ++ ++ ++ ++ ++ + + +++ ++ ++ + 51.5 + +++ + + ++ ++ OAE 2 + ++ ++ + ++ + + + ++ + + + Scenario1 + +++ + + + + + + +++ + + ++ +++ + + + + ++ + + + + + + 52.5 + + + + + + ++ + + + + ++ ++++ + + + + + + + +++ + +++ ++ +++ Depth (m) + + + + ferruginous + + + +++ +++ + + ++ + + 52.0 ++++ + + + + + + + + ++++ + + ++ + + ++ + + + ++++ + + + + + + + 53.0 + +++ + + + + + + + + + +++ + + ++ ++ + ++++ ++ + + ++ + +++ + + ++ ++ + + ++ + +++ + + + + + + OAE 2 ++++ + + + ++ + ++ ++++ ++ ++ ++ 52.5 + ++++ ++ ++ + + 53.5 + ++++ +++ +++ ++ ++++ + + ++ ++

+ + + + Cycle 0 +++ + + ++ + + + + ++ + +++ ++ + + Depth in section (m) 54.0 +

+ anoxic + +++++ oxic + ++ +++ ++ + + +++ + + + + + + 53.0 +++++ + + + ++ + + ++ + + ++++ + + + + + + + +++++ ++ + ++ + + + + + + +++ + + + ++ +++ ++ + + + + + + + + 54.5 ++++ + + + + + + + + +++ ++ + + ++ + + + -40-30-20-10 010 +++ ++ ++ ++ +++ + + + +++ +++ + ++ + + ++ ++ 53.5 + ++++ ++ + +++ +++ + + δ34S (‰) ++++ + + + + + + ++ + + + + ++ + ++++ + + + + + + + + +++++ +++ +++ + + + ++++ + + + + + + + ++ +++++ + + ++ ++ + + ++ ++ +++ +++ +++ + ++ + + + + + ++ + Figure 2. A: Sulfur (S) isotope profiles for py- +++ ++ + ++ ++ 54.0 +++ + ++ ++ ++ +++ + + + ++ + +++ + ++++ + ++ ++ + rite and carbonate-associated sulfate (CAS), +++ + ++ + + + + Cycle -1 +++++ +++ +++ + ++ +++ + ++ + + + + 34 + + +++ ++ + ++ + + + + and fractionations (D S) between CAS and 54.5 pyrite. B: Estimated seawater sulfate con- -28 -26 -24 0 0.4 0.6 0.8 1 0 0.2 0.4 0.6 centrations based on modeling of S isotope δ13Corg (‰) FeHR/FeT Feox/FeHR data, showing no fluctuations associated with redox cyclicity. Scenario 1: Organic S Figure 1. Bulk geochemical, Fe speciation, and isorenieratane data for onset of Oceanic An- burial flux set at 3 times pyrite burial flux; oxic Event 2 (OAE2), showing repetitive development of ferruginous conditions (gray shad- S weathering input flux increased by factor ing). Sedimentary cycles are as defined by Kolonic et al. (2005). Vertical dashed lines repre- of 3.2 compared to modern (Blättler et al., sent boundary for distinguishing anoxic conditions (Fe /Fe profile) and for distinguishing HR T 2011). Scenario 2: Organic S burial flux set between euxinic and ferruginous conditions (Fe /Fe profile). Inset graph shows interval py HR at 3 times pyrite burial flux; volcanic S input studied in context of entirety of OAE2. TOC—total organic carbon; Isonren.—isorenieratane; flux increased by factor of 7.6 compared to Fe —highly reactive Fe fraction; Fe —total Fe; Fe —pyrite; Fe —ferric (oxyhydr)oxides. HR T py ox modern (Adams et al., 2010). Scenario 3: Or- ganic S burial flux set at 6 times pyrite burial flux; S weathering input flux increased by factor of 3.2 compared to modern (Blättler et matography–mass spectrometry (see the Data bottom-water anoxia prior to, and during, the al., 2011); volcanic S input flux increased by Repository for full details of samples, methods, onset and initial maximum (as conventionally factor of 7.4 compared to modern (Adams et and data). defined by the initial positive shift in carbon al., 2010). OAE2—Oceanic Anoxic Event 2; isotopes) of OAE2 (Fig. 1). Throughout the ma- Conc.—concentration.

Sulfur Cycle Box Model jority of the interval, Fepy/FeHR ratios fall close A standard sulfur-isotope box model (Ad- to or above the 0.7 threshold, suggesting domi- temporal extent of euxinia. The modern ma- ams et al., 2010) was used to estimate seawater nantly euxinic conditions. This is supported by rine environment is delicately balanced with

sulfate concentrations in the proto–North At- the presence of the biomarker isorenieratane in regard to the fluxes of sulfate and FeHR to the

lantic during the onset of OAE2 (see the Data all samples studied, which implies at least peri- ocean, which are poised at a molar S:FeHR ra- Repository for full details of model parameters). odic incursions of sulfide into the lower photic tio of ~1.8:1 (Poulton and Canfield, 2011). This Measured organic sulfur contents are ~6 times zone, although the variability in concentration ratio is slightly less than the 2:1 ratio of these

higher than pyrite sulfur contents on the Tarfaya suggests that the temporal extent or intensity of elements in pyrite (FeS2; the main removal shelf, while in the deeper proto–North Atlantic photic zone euxinia may have fluctuated (Fig. mechanism for sulfate from the ocean), suggest- at Demerara Rise, organic sulfur burial was ~3 1). A pronounced feature is the cyclic develop- ing that extensive depletion of seawater sulfate times higher than pyrite burial (Hetzel et al., ment of ferruginous water-column conditions, under euxinic conditions could potentially have

2009). Therefore, we include an organic sulfur where intervals of low Fepy/FeHR (<0.7) corre- led to an excess of FeHR, hence driving the water component in the model and perform model spond to enhanced preservation of ferric oxides column ferruginous. This mechanism does not

runs where organic sulfur burial is set at 3 times (Feox/FeHR; Fig. 1). Ferruginous conditions oc- require complete removal of sulfate from the (scenarios 1 and 2) and 6 times (scenario 3) the curred every half sedimentary cycle and are not water column, but rather demands that rates of

pyrite burial flux (Fig. 2). To maintain steady related to changes in CaCO3 or TOC contents, sulfide production under low sulfate conditions state after the addition of the organic sulfur arguing for a robust mechanism that operated on were overwhelmed by the influx of reactive Fe, burial flux term, we perform model runs with orbital frequencies. thus restricting sulfide build-up to increased volcanic and/or weathering influxes porewaters (Meyers, 2007). of sulfur, consistent with the findings of previ- Controls on Redox Cycling We evaluate this mechanism by consider- ous studies (Adams et al., 2010; Blättler et al., Fluctuations in nutrient levels as a potential ing sulfur isotope systematics through the ana- 2011; Pogge von Strandmann et al., 2013). driver of the transitions to ferruginous water- lyzed section (Fig. 2). Our CAS analyses give column conditions can be ruled out due to con- a relatively constant seawater sulfate isotopic RESULTS AND DISCUSSION sistently high TOC (Fig. 1) and bio-essential composition (d34S) of 10‰ ± 2‰ (Fig. 2). This trace metal availability (see the Data Reposi- value is significantly lower than contemporane- Water-Column Redox Reconstruction tory). Instead, we first focus on the possibil- ous values of ~19‰ determined from barite in

Elevated FeHR/FeT ratios are apparent through- ity that enhanced burial of seawater sulfate (as Pacific sediments (Paytan et al., 2004) and CAS out the analyzed core, suggesting persistent pyrite) during euxinic intervals controlled the estimates that range from ~17‰ to 22‰ during

964 www.gsapubs.org | Volume 43 | Number 11 | GEOLOGY

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/43/11/963/3548007/963.pdf by guest on 25 September 2021 the early stages of OAE2 in relatively shallow- low sulfate. Sulfate is dominantly sourced from TOC/P water settings from northern Europe (Owens pyrite oxidation on land, which also releases 0 500 1000 1500 50.0 ++ et al., 2013). Our estimate is, however, closer Fe (Fe released during pyrite oxidation ulti- ++ HR RR (106:1) +++ ++ + ++ to CAS values recorded immediately prior to mately forms Fe oxides), but enhanced silicate +++ ++ ++ 50.5 +++ OAE2 in sediments from Italy (14‰–18‰; weathering would have provided an additional + ++ + ++ Ohkouchi et al., 1999) and the Western Interior source of FeHR over sulfate, as Fe release from + ++ ++ Seaway (10‰–13‰; Adams et al., 2010), sug- parent rocks increases with weathering intensity 51.0 + + ++ + ++ + +++ gesting significant spatial variability in seawater and continental runoff (Poulton and Raiswell, + + ++ ++ + + + sulfate isotope compositions at this time. The 2002). It is unclear whether a short-term weath- ++ + 51.5 + + + + + + + sulfate isotope record has also been shown to ering-induced increase in the relative flux of ++ + + + + ++ ++ vary temporally during OAE2, with isotopically FeHR would have been sufficient, in isolation, + 52.0 + ++ + ++ light compositions in the run-up to the OAE be- to drive the proto–North Atlantic shelf ferrugi- + ++ OAE 2 ++ +++ +++ ing attributed to enhanced volcanism (Adams nous. In addition, however, Fe minerals are + ++ HR + + + + 52.5 +++ et al., 2010). Subsequently, transitions to more dominantly (~85%) trapped in proximal settings + + + ++ + +++ positive values (increasing by 2‰–7‰) toward such as floodplains and lagoons in the modern Depth (m) + + + ++ the top of (or above) the OAE have been linked environment, whereas riverine dissolved sulfate 53.0 + + + +++ + + + +++ to extensive drawdown of seawater sulfate is effectively mixed into the ocean (Poulton and +++ + ++ +++ + ++ through enhanced pyrite burial (Ohkouchi et al., Raiswell, 2002). Increased rates of continental + ++ 53.5 ++ + +++ 1999; Adams et al., 2010; Owens et al., 2013). runoff during the Cretaceous would have di- ++ +++ ++ +++ Despite this spatial and temporal variability, our minished this proximal Fe sink by bypassing ++ HR ++ 54.0 + + high-resolution sulfur isotope analyses can be such settings, potentially resulting in a greatly ++ ++ ++ used to evaluate whether enhanced drawdown enhanced flux of Fe to the open shelf relative + HR AS (0.08) + of seawater sulfate also operated on orbital time to sulfate. A proportional increase in the flux 54.5 0 0.2 0.4 0.6 scales, thus driving the observed transitions to of Fe over sulfate would increasingly restrict HR P/Al ferruginous conditions. sulfidic conditions to the sediments (Meyers, When incorporated into the sulfur isotope 2007), ultimately promoting the development of Figure 3. P/Al and molar total organic carbon (TOC)/P profiles for Oceanic Anoxic Event 2 box model (Fig. 2), two robust conclusions can ferruginous conditions. (OAE2) sediments at site S57, Tarfaya, show- be drawn from our analyses. Firstly, consistent ing extensive release of phosphorus under with studies from other areas (Ohkouchi et al., Phosphorus Cycling and Implications for euxinic and ferruginous conditions. Arrows 1999; Paytan et al., 2004; Adams et al., 2010; the Maintenance of Widespread Anoxia represent average shale (AS, P/Al profile) and Owens et al., 2013), seawater sulfate concen- Euxinic-ferruginous redox cycling has ma- the Redfield ratio of 106:1 (TOC/P profile). trations significantly lower than at present (28 jor implications for phosphorus bioavailability, mM) are indicated for the Tarfaya shelf. Indeed, as demonstrated for black shales deposited dur- ter or Fe (oxyhydr)oxide minerals (Fig. 1), sul- our estimated range of ~3–7 mM (Fig. 2) is in ing OAE3 at Demerara Rise (März et al., 2008), fide production in sediment porewaters on the close agreement with estimates of ~2–4 mM with the possibility of extensive phosphorus Tarfaya shelf was sufficient to remobilize this (Adams et al., 2010) and 7 mM (Owens et al., burial under ferruginous conditions, in contrast sequestered phosphorus back into the water col- 2013) for the early stages of OAE2 in different to effective phosphorus recycling to the water umn, via either sulfide-promoted reduction of oceanic basins. Secondly, a particularly impor- column under euxinic conditions. This possibil- Fe oxides or preferential release of phosphorus tant factor for our study is that reconstructed ity is important to evaluate at Tarfaya, because from organic matter during bacterial sulfate re- seawater sulfate concentrations are relatively extensive phosphorus burial under ferruginous duction (Van Cappellen and Ingall, 1994). This constant throughout the analyzed interval (Fig. conditions would have limited marine produc- interpretation is supported by significant pyriti-

2). Thus, the sulfur isotope record does not tivity and hence call into question the role of zation of the FeHR pool throughout ferruginous support enhanced drawdown of sulfate under persistently elevated phosphorus in maintain- intervals on the Tarfaya shelf (Fig. 1), suggest- euxinic conditions as a driver for the repeated ing anoxia during OAE2 (cf. Mort et al., 2007). ing abundant sulfide availability during early development of ferruginous conditions, and an Consistent with a rise in reactive phosphorus diagenesis, which would also have preserved

alternative mechanism must be sought. burial at the onset of OAE2 observed in a num- FeHR enrichments in the sediment (Scholz et al., In this regard, we instead consider possible ber of global localities (Mort et al., 2007), P/Al 2014). In fact, the presence of isorenieratane

changes in the oceanic influx of sulfate and ratios are significantly enhanced at this point in and the occasional occurrence of Fepy/FeHR ra-

FeHR, which may have been driven by regular the Tarfaya core (Fig. 3). In contrast, the remain- tios >0.7 during ferruginous intervals suggest variability in continental weathering and run- der of the analyzed interval is characterized by that rates of bacterial sulfate reduction may have off. A climate-driven control is supported by P/Al ratios at or below average shale, with no been sufficient to result in short-lived episodes evidence for orbital-time-scale fluctuations in evidence for enhanced burial under ferruginous of water-column euxinia punctuating ferrugi- trade wind intensity and hydrological cycling, conditions. Building upon this observation, mo- nous intervals (Fig. 1), further promoting phos- translating into alternations between more hu- lar TOC/P ratios are consistently above the Red- phorus recycling within the water column itself. mid and drier periods (Wagner et al., 2013), and field ratio after the onset of OAE2, suggesting major element evidence for enhanced silicate effective recycling of phosphorus to the water CONCLUSIONS weathering during some of the ferruginous in- column (Van Cappellen and Ingall, 1994) dur- Our data provide the first evidence for dis- tervals apparent in the Cretaceous (see the Data ing both euxinic and ferruginous intervals on the tinct redox cyclicity between euxinic and ferru- Repository). This increase in weathering inten- Tarfaya shelf. ginous water-column conditions during OAE2, sity would have altered the proportional fluxes To explain these observations we suggest apparently driven by orbital fluctuations in con-

of sulfate and FeHR to the ocean on orbital time that although phosphorus was likely initially tinental hydrology and weathering. This redox scales, the effects of which would have been par- sequestered in the sediment during ferruginous cyclicity occurs both during and prior to the ticularly significant in an ocean characterized by intervals, in association with either organic mat- onset of OAE2, and has also been observed in

GEOLOGY | Volume 43 | Number 11 | www.gsapubs.org 965

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/43/11/963/3548007/963.pdf by guest on 25 September 2021 a deeper-water section during OAE3 (März et Kuhnt, W., Luderer, F., Nederbragt, A., Thurow, J., anic Anoxic Event 2: Nature Geoscience, v. 6, al., 2008), suggesting that such conditions were and Wagner, T., 2005, Orbital-scale record of p. 668–672, doi:10.1038/ngeo1875. likely a pervasive feature of anoxic water-col- the late Cenomanian–Turonian oceanic anoxic Poulton, S.W., and Canfield, D.E., 2005, Develop- event (OAE-2) in the Tarfaya Basin (Morocco): ment of a sequential extraction procedure for umn conditions in the Cretaceous greenhouse International Journal of Earth Sciences, v. 94, iron: Implications for iron partitioning in con- ocean. It remains to be seen, however, just how p. 147–159, doi:10.1007/s00531-004-0440-5. tinentally derived particulates: Chemical Geol- temporally and spatially expansive such condi- März, C., Poulton, S.W., Beckmann, B., Küster, K., ogy, v. 214, p. 209–221, doi:10.1016/j.chemgeo​ tions were. Nevertheless, our observation of Wagner, T., and Kasten, S., 2008, Redox sen- .2004.09.003. extensive remobilization of phosphorus from sitivity of P cycling during marine black shale Poulton, S.W., and Canfield, D.E., 2011, Ferrugi- formation: Dynamics of sulfidic and anoxic, non- nous conditions: A dominant feature of the the Tarfaya shelf under ferruginous conditions sulfidic bottom waters: Geochimica et Cosmo- ocean through Earth’s history: Elements (Que- highlights an efficient mechanism by which el- chimica Acta, v. 72, p. 3703–3717, doi:10.1016​ ​ bec), v. 7, p. 107–112, doi:10.2113/gselements​ evated primary productivity and anoxia could /j​.gca​.2008​.04.025. .7​.2.107. have been sustained in the open ocean. It is Meyers, S.R., 2007, Production and preservation of Poulton, S.W., and Raiswell, R., 2002, The low-tem- organic matter: The significance of iron: Pa- equally possible that these redox-nutrient feed- perature geochemical cycle of iron: From con- leoceanography, v. 22, PA4211, doi:​10.1029​ tinental fluxes to marine sediment deposition: backs may have operated as a major control on /2006PA001332. American Journal of Science, v. 302, p. 774– the spatial and temporal extent of ocean anoxia Meyers, S.R., Sageman, B.B., and Arthur, M.A., 805, doi:10.2475/ajs.302.9.774. during other periods of Earth history. 2012, Obliquity forcing of organic matter ac- Scholz, F., McManus, J., Mix, A.C., Hensen, C., and cumulation during Oceanic Anoxic Event 2: Schneider, R.R., 2014, The impact of ocean ACKNOWLEDGMENTS Paleoceanography, v. 27, PA3212, doi:10.1029​ ​ deoxygenation on iron release from continen- This work was funded by a Royal Society–Wolfson /2012PA002286. tal margin sediments: Nature Geoscience, v. 7, Research Merit Award (Wagner), a Natural Environ- Monteiro, F.M., Pancost, R.D., Ridgwell, A., and p. 433–437, doi:10.1038/ngeo2162. ment Research Council Research Fellowship (Poul- Donnadieu, Y., 2012, as the domi- Sinninghe Damsté, J.S., and Köster, J.A., 1998, A ton), and the German Science Foundation (SFB 754 nant control on the spread of anoxia and eux- euxinic southern North Atlantic Ocean during the sub-project A7 to Flögel). We thank Aubrey Zerkle for inia across the Cenomanian-Turonian oceanic Cenomanian/Turonian oceanic anoxic event: Earth helpful discussions, James Witts for SEM assistance, anoxic event (OAE2): Model-data comparison: and Planetary Science Letters, v. 158, p. 165–173, and Christian Bjerrum and anonymous referees for Paleoceanography, v. 27, PA4209, doi:10.1029​ ​ doi:10.1016​ /S0012​ -821X​ (98)​ 00052-1.​ very helpful comments that significantly improved the /2012PA002351. Trabucho Alexandre, J., Tuenter, E., Henstra, G.A., final manuscript. Mort, H.P., Adatte, T., Föllmi, K.B., Keller, G., Stein- van der Zwan, K.J., van de Wal, R.S.W., Dijkstra, mann, P., Matera, V., Berner, Z., and Stüben, D., H.A., and de Boer, P.L., 2010, The mid-Creta- REFERENCES CITED 2007, Phosphorus and the roles of productivity ceous North Atlantic nutrient trap: Black shales Adams, D.D., Hurtgen, M.T., and Sageman, B.B., and nutrient recycling during oceanic anoxic and OAEs: Paleoceanography, v. 25, PA4201, 2010, Volcanic triggering of a biogeochemical cas- event 2: Geology, v. 35, p. 483–486, doi:​10.1130​ doi:​10.1029​/2010PA001925. cade during Oceanic Anoxic Event 2: Nature Geo- /G23475A.1. Tsikos, H., et al., 2004, Carbon-isotope stratigraphy science, v. 3, p. 201–204, doi:10.1038/ngeo743.​ Ohkouchi, N., Kawamura, K., Kajiwara, Y., Wada, E., recorded by the Cenomanian/Turonian Oceanic Blättler, C.L., Jenkyns, H.C., Reynard, L.M., and Okada, M., Kanamatsu, T., and Taira, A., 1999, Anoxic Event: Correlation and implications Henderson, G.M., 2011, Significant increases Sulfur isotope records around Livello Bonarelli based on three key localities: Journal of the in global weathering during Oceanic Anoxic (northern Apennines, Italy) black shale at the Geological Society, v. 161, p. 711–719, doi:​ Events 1a and 2 indicated by calcium isotopes: Cenomanian-Turonian boundary: Geology, v. 27, 10.1144​/0016-764903-077. Earth and Planetary Science Letters, v. 309, p. 535–538, doi:10.1130​/0091​-7613​(1999)​027​<​ Van Cappellen, P., and Ingall, E.D., 1994, Benthic p. 77–88, doi:10.1016/j.epsl.2011.06.029. 0535​:SIRALB​>2.3.CO;2. phosphorus regeneration, net primary produc- Hetzel, A., Böttcher, M.E., Wortmann, U.G., and Owens, J.D., Gill, B.C., Jenkyns, H.C., Bates, S.M., tion, and ocean anoxia: A model of the coupled Brumsack, H.-J., 2009, Paleo-redox conditions Severmann, S., Kuypers, M.M.M., Woodfine, marine biogeochemical cycles of carbon and during OAE 2 reflected in Demerara Rise sedi- R.G., and Lyons, T.W., 2013, Sulfur isotopes phosphorus: Paleoceanography, v. 9, p. 677– ment geochemistry (ODP Leg 207): Palaeogeog- track the global extent and dynamics of euxinia 692, doi:10.1029/94PA01455. raphy, Palaeoclimatology, Palaeoecology, v. 273, during Cretaceous Oceanic Anoxic Event 2: Pro- Wagner, T., Hofmann, P., and Flögel, S., 2013, Marine p. 302–328, doi:10.1016/j​.palaeo​.2008​.11.005. ceedings of the National Academy of Sciences of black shale and Hadley Cell dynamics: A concep- Jenkyns, H.C., 2010, Geochemistry of oceanic anoxic the United States of America, v. 110, p. 18,407– tual framework for the Cretaceous Atlantic Ocean: events: Geochemistry Geophysics Geosystems, 18,412, doi:10.1073​/pnas​.1305304110. Marine and Petroleum Geology, v. 43, p. 222–238, v. 11, Q03004, doi:10.1029/2009GC002788. Paytan, A., Kastner, M., Campbell, D., and Thiemens, doi:10.1016/j.marpetgeo​ .2013​ .02.005.​ Kolonic, S., et al., 2005, Black shale deposition on M.H., 2004, Seawater sulfur isotope fluctuations the northwest African shelf during the Cenoma- in the Cretaceous: Science, v. 304, p. 1663–1665, Manuscript received 31 March 2015 nian/Turonian oceanic anoxic event: Climate doi:​10.1126​/science​.1095258. Revised manuscript received 24 August 2015 coupling and global organic carbon burial: Pa- Pogge von Strandmann, P.A.E., Jenkyns, H.C., Manuscript accepted 26 August 2015 leoceanography, v. 20, PA1006, doi:​10.1029​ and Woodfine, R.G., 2013, Lithium isotope /2003PA000950. evidence for enhanced weathering during Oce- Printed in USA

966 www.gsapubs.org | Volume 43 | Number 11 | GEOLOGY

Downloaded from http://pubs.geoscienceworld.org/gsa/geology/article-pdf/43/11/963/3548007/963.pdf by guest on 25 September 2021