Phosphorus and the roles of productivity and nutrient recycling during oceanic anoxic event 2 Haydon P. Mort* Institute of Geology, University of Neuchâtel, Rue Emile Argand 11, Case postale 158, Thierry Adatte CH-2009 Neuchatel, Switzerland Karl B. Föllmi Gerta Keller Department of Geosciences, Princeton University, Guyot Hall, Princeton, New Jersey 08544-1003, USA Philipp Steinmann Institute of Geology, University of Neuchâtel, Rue Emile Argand 11, Case postale 158, Virginie Matera CH-2009 Neuchatel, Switzerland Zsolt Berner Institut für Mineralogie und Geochemie, Universität Karlsruhe, 76128 Karlsruhe, Germany Doris Stüben ABSTRACT OAE 2, in order to evaluate the roles of primary productivity and nutrient Four sections documenting the impact of the late Cenomanian cycling and to test the outcome of numerical models by Van Cappellen and oceanic anoxic event (OAE 2) were studied in basins with different Ingall (1994, 1996), in which a relationship is postulated between anoxic paleoenvironmental regimes. Accumulation rates of phosphorus (P) conditions, P regeneration, and sustained primary productivity. bound to iron, organic matter, and authigenic phosphate are shown to The behavior of the carbon and P cycles during OAE 2 are recon- rise and arrive at a distinct maximum at the onset of OAE 2, with an structed in four localities representative of differing paleoceanographic associated increase in δ13C values. Accumulation rates of P return to pre- regimes, based on whole-rock stable carbon isotopes, quantities and types excursion values in the interval where the δ13C record reaches its fi rst of organic matter, and phosphorus mass accumulation rates (P MARs) maximum. An offset in time between the maximum in P accumulation for P bound to oxyhydroxides, authigenic minerals, detrital material, and peaks in organic carbon burial, hydrogen indices, and Corg/Preact and organic matter. The sections studied are at Pueblo (Colorado, USA), molar ratios is explained by the evolution of OAE 2 in the following Eastbourne (UK), Furlo (Italy), and Manilva (Spain). The Pueblo sec- steps. (1) An increase in productivity increased the fl ux of organic mat- tion is located in the relatively shallow Western Interior Seaway and, as ter and P into the sediments; the preservation of organic matter was low the Cenomanian-Turonian stratotype section and point, forms a refer- and its oxidation released P, which was predominantly mineralized. (2) ence point for dating other sections based on high-resolution planktic Enhanced productivity and oxidation of organic matter created dysoxic foraminiferal biostratigraphy and the typical δ13C record for OAE 2 bottom waters; the preservation potential for organic matter increased, (Fig. 1, Site 1) (Keller et al., 2004). A new chronostratigraphic frame- whereas the sediment retention potential for P decreased. (3) The lat- work was applied to the Pueblo section (Sageman et al., 2006) generat- ter effect sustained high primary productivity, which led to an increase ing relative ages based on orbital frequencies identifi ed above bed 63 in the abundance of free oxygen in the ocean and atmosphere system. (see Pueblo bed number in Fig. 2). The Eastbourne section (Site 2) was After the sequestration of CO2 in the form of black shales, this oxygen deposited in a continental shelf setting and has also been studied exten- helped push the ocean back into equilibrium, terminating black shale sively (e.g., Paul et al., 1999; Gale et al., 2005). The sections at Furlo deposition and removing bioavailable P from the water column. (Site 3), located in the Umbria-Marche Basin, Italy, and Manilva (Site 4), in Andalusia, southern Spain (Reicherter et al., 1994), were deposited Keywords: phosphorus, feedbacks, biogeochemistry, Cenomanian- in deep pelagic environments marked by organic-rich deposition (i.e., Turonian, oceanic anoxic event 2. the Bonarelli level; Luciani and Cobianchi, 1999). INTRODUCTION Periods during which oceanic bottom waters became oxygen depleted 90o -60o -30o o o 0 are well documented for the Cretaceous, and their study has led to an 60 improved understanding of the global carbon cycle and climate change in a Western Interior greenhouse world (Schlanger and Jenkyns, 1976; Jenkyns, 1980). The late Seaway Cenomanian oceanic anoxic event (OAE 2) is one of the most prominent North America anoxic episodes of the Mesozoic, characterized by the widespread accu- 1 Norwegian Seaway mulation of organic-carbon rich sediment and diagnostic redox-sensitive 2 o metals (e.g., Orth et al., 1993). OAE 2 was accompanied by a 2‰ posi- 30 tive δ13C excursion in marine carbonates and organic matter with a plateau North Atlantic 4 3 of high δ13C values that persisted during the anoxic episode. Biotic effects resulting in the extinction as well as the evolution of new species are clearly observed (e.g., Leckie, 1985). Various OAE triggering mechanisms have been proposed, ranging from increased oceanic CO2 derived from large igneous province activity (Kerr, 1998) to the increased nutrient delivery 0o Continents into surface waters from continental sources (e.g., Larson and Erba, 1999). Continental However, there are large gaps in our understanding of these events, particu- Shelves Africa larly with respect to the rate of primary productivity, its interactions with Present Coastline S. America the global nutrient cycles, and feedback mechanisms caused by O2-defi cient bottom waters. This study focuses on the behavior of phosphorus (P) during Figure 1. Paleogeography of western Northern Hemisphere at Ceno- manian-Turonian boundary (93.5 Ma). Adapted from Kuypers et al. *E-mail: [email protected] (2002). Study sites: 1—Pueblo, 2—Eastbourne, 3—Furlo, 4—Manilva. © 2007 The Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or [email protected]. GEOLOGY,Geology, June June 2007; 2007 v. 35; no. 6; p. 483–486; doi: 10.1130/G23475A.1; 3 fi gures; Data Repository item 2007118. 483 TOC (%) 0 1.0 2.0 3.0 01020 10 20 TOC] HI [mg HC/g Pueblo Furlo Manilva 600 400 600 R2 = 0.74 400 400 Sedimentation Rate 200 200 (cm/kyr) R2 = 0.45 R2 = 0.78 Stage R2 = 0.56 Biozone Depth (m) 0246 Figure 2. Phosphorus (P) 8 2 Sed. rate speciation with accumulation 13 Limestone Chalk 6 δ C rates (mg/cm2/k.y.) in four W. archaeo W. Black Shale Marls sections, plotted against 4 1 corresponding δ13C curves. Chert 2 Gray shaded area highlights Cenomanian Bentonite interval in time between peak Eastbourne, UK R. cushmani 0 OC AR in P mass accumulation rates 2.5 3.5 4.5 5.5 0.1 0.3 579 1 2 0.1 0.2 (Pueblo) and fi rst peak in δ13C values. 17 012 0 24 H. hel Percentages for total organic Turon. 86 Bed No. carbon, hydrogen indices, 84 15 and Corg /Preactive molar ratios 79 are also shown. Smoothed 77 13 lines are 3-point moving 69 2 W. archeocretacea W. averages. Scatter plots are 63 1 correlations between total 11 organic carbon (TOC) and Pueblo, USA hydrogen indices. Two corre- 9 lations are made for Pueblo Cenomanian R. cushmani (black circles and gray 7 squares represent terrestrial -1 012 .1 .2 0246 0 0.5 1.0 00.2 1.0 2.0 100 300 and/or oxidized and marine 0.0 0.5 1.0 11 organic matter, respec- tively). Rotalipora cushmani 10 2 biozones in Furlo and W. archaeo W. Manilva are extended with 1 gray bars, which represent 9 probable extinction level. All Cenomanian Furlo, Italy organic carbon P speciation data can be R. cushmani 8 found in Appendix 1 (also -27 -25 -23 0 .02 .04 0.2.4.6 0.2 0.4 .01.02 .03 0 5 10 15 20 0 500 1000 0.0 0.4 0.8 0200400 see Appendix 2 for more 5 details; see footnote 1). 2 Tur.—Turonian; OC AR— W. archaeo W. 4 organic carbon accumula- 1 tion rate; VPDB—Vienna Peedee belemnite. 3 TOC Cenomanian R. cushmani OC AR Manilva, Spain 2 organic carbon -26 -24 -22 0.4 0.8 1.2 0.5 1.5 0 1 2 0.5 1.5 01020 0 1000 TOC (%) δ13 P P P P Corg/Preact C ‰ Fe authi detrital org 10 30 50 (VPDB) (mg/cm2/kyr) OC AR (g/cm2/kyr) METHODS scales to be used to construct MARs. Total reactive P (Preactive) was calcu- The SEDEX sequential extraction method (originally constructed by lated by an addition of all the measured phases, under the assumption that Ruttenberg, 1992) was used to obtain P contents for phases associated most detrital P represents authigenic P (Appendix 2; see footnote 1). Total with oxyhydroxides (PFe), authigenic francolite (Pauthigenic), detrital apatite organic carbon (TOC) and hydrogen indices (HI) were measured by Rock- (Pdetrital ), and organic remains (Porganic). Instrumental accuracy was <5% Eval, with precisions of <1%. Molecular Corg/Preactive ratios were calculated using a Perkin Elmer Lambda 10 spectrophometer. The P MAR was cal- to assess the relative importance of the vertical fl ux of P to and from the culated in mg/cm2/k.y. Rock densities were calculated in the laboratory sediment. Intersite age control was based on the well-dated sequences at by measuring the displacement of water by a sample of known mass. The Pueblo and Eastbourne. Correlations with Furlo and Manilva were based accumulation rates used for the Pueblo section were based on dating from on microfossil biostratigraphy and the δ13C curves derived from organic Keller et al. (2004 and references therein) as well as the new cyclostrati- carbon. For a more detailed methodology, see Appendix 2. graphic framework from Sageman et al. (2006). GSA Data Repository Appendix 11 contains the raw concentrations of P to enable other time RESULTS In all sections, concentrations and MARs of all measured P species 1GSA Data Repository item 2007118, Appendices 1 and 2, containing raw (except for PFe in Eastbourne and Porg in Furlo and Manilva) reach maxi- phosphorus data, mass accumulation rates, and more detailed descriptions of δ13 the methods used in this study, is available online at www.geosociety.org/pubs/ mum values within or close to the interval in which the onset of the C ft2007.htm, or on request from [email protected] or Documents Secretary, positive excursion is registered.