Extinction and recovery of benthic foraminifera across the Paleocene–Eocene Thermal Maximum at the Alamedilla section (Southern Spain

Extinction and recovery of benthic foraminifera across the Paleocene–Eocene Thermal Maximum at the Alamedilla section (Southern Spain
of 15
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


Publish on:

Views: 57 | Pages: 15

Extension: PDF | Download: 0

  Extinction and recovery of benthic foraminifera across the Paleocene – EoceneThermal Maximum at the Alamedilla section (Southern Spain) L. Alegret a, ⁎ , S. Ortiz a,b , E. Molina a a Dpto. Ciencias de la Tierra (Paleontología), Facultad de Ciencias, Universidad de Zaragoza, 50009 Zaragoza, Spain b Department of Earth Sciences, University College London, WC1E 6BT London, UK  a b s t r a c ta r t i c l e i n f o  Article history: Received 30 December 2008Received in revised form 25 April 2009Accepted 10 May 2009 Keywords: Paleocene – EoceneWarmingExtinctionBenthic foraminiferaDeep-sea A complete succession of lower bathyal – upper abyssal sediments was deposited across the Paleocene – Eocene Thermal Maximum (PETM) at Alamedilla (Betic Cordillera, Southern Spain), where the benthicforaminiferal turnover and extinction event associated with the negative carbon isotope excursion (CIE)across the PETM have been investigated. Detailed quantitative analyses of benthic foraminifera allowed us todistinguish assemblages with paleoecological and paleoenvironmental signi 󿬁 cance: pre-extinction fauna,extinction fauna, survival fauna (including disaster and opportunistic fauna) and recovery fauna. Theseassemblages have been associated with signi 󿬁 cant parts of the  δ 13 C curve for which a relative chronology hasbeen established. The correlation between the benthic turnover, the  δ 13 C curve, the calcite and silicatemineral content, and sedimentation rates, allowed us to establish the sequence of events across the PETM.At Alamedilla, the benthic extinction event (BEE) affected ~37% of the species and it has been recorded over a30-cm-thick interval that was deposited in c.a.10 ky, suggesting a gradual but rapid pattern of extinction. Thebeginning of the BEE coincides with the onset of the CIE (+0 ky) and with an interval with abundant calcite,and it has been recorded under oxic conditions at the sea 󿬂 oor (as inferred from the benthic foraminiferalassemblages and the reddish colour of the sediments). We conclude that dissolution and dysoxia were notthe cause of the extinctions, which were probably related to intense warming that occurred before the onsetof the CIE.The BEE is immediately overlain by a survival interval dominated by agglutinated species (the  Glomospira Acme). The lowcalcite content recorded within the survival interval may result from the interaction betweendilution of the carbonate compounds by silicicate minerals (as inferred from the increased sedimentationrates),andtheeffectsofcarbonatedissolutiontriggeredbytheshoalingoftheCCD.Wesuggestthat Glomospira species (disaster fauna) may have bloomed opportunistically in areas with methane dissociation, in andaround the North Atlantic. The disaster fauna was rapidly replaced by opportunistic taxa, which point to oxicand, possibly, oligotrophic conditions at the sea 󿬂 oor. The CCD gradually dropped during this interval, andcalcite preservation improved towards the recovery interval, during which the  δ 13 C values and the calcitecontent recovered (c.a. +71.25 to 94.23 ky) and stabilized ( N 94.23 ky), coeval with a sharp decrease insedimentation rates.© 2009 Elsevier B.V. All rights reserved. 1. Introduction The generally warm, greenhouse world of the Paleogene under-went signi 󿬁 cant disruption at the Paleocene – Eocene transition, c.a.55 Ma ago, when temperatures increased by 5° to 9 °C in the oceansand on land within less than 10,000 years (e.g., Röhl et al., 2000;Thomas, 2007). A major perturbation of the global carbon cycleoccurred during this episode that is commonly known as thePaleocene – Eocene Thermal Maximum (PETM; Zachos et al., 1993),including a negative 2.5 to 6 ‰  carbon isotope excursion (CIE) inmarine and terrestrial  δ 13 C values of carbonate and organic carbon(e.g., Kennett and Stott, 1991; Thomas and Shackleton, 1996; Zachoset al., 2001; Bowen et al., 2006) and a ~2 km shoaling of the calcitecompensation depth (CCD) in the deep-sea (e.g., Zachos et al., 2005).The onset of these anomalies occurred during a time period  b 20 ky(Röhl et al., 2000, 2007), whereas subsequent recovery of   δ 13 C valuestook 170 ky (Röhl et al., 2007). This global perturbation of the carboncycleisinterpretedintermsofarapidinputofisotopicallylightcarboninto the ocean – atmosphere system, possibly related to the massivedissociation of marine methane hydrates, although the triggeringmechanism is still under debate (e.g., Dickens et al.,1997; Katz et al.,2001; Cramer and Kent, 2005; Pagani et al., 2006; Sluijs et al., 2007a).Rapid changes in terrestrial and marine biota occurred during thePETM, including the most severe extinction of deep-sea benthicforaminifera recorded during the Cenozoic (e.g., Tjalsma and Lohmann, Palaeogeography, Palaeoclimatology, Palaeoecology 279 (2009) 186 – 200 ⁎  Corresponding author. Fax: +34 976 761106. E-mail address:  laia@unizar.es (L. Alegret).0031-0182/$  –  see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.palaeo.2009.05.009 Contents lists available at ScienceDirect Palaeogeography, Palaeoclimatology, Palaeoecology  journal homepage: www.elsevier.com/locate/palaeo  1983; Thomas, 2007), a global acme of the dino 󿬂 agellate genus  Apec-todinium  and its migration to high latitudes (e.g., Crouch et al., 2001;Sluijs et al., 2007a,b), an acme of the planktic foraminiferal genus  Acarinina  and its migration towards higher latitudes (e.g., Arenillas andMolina, 1996; Kelly et al., 1998; Molina et al., 1999), the rapidevolutionary turnover of calcareous nannoplankton (e.g., Bralower,2002; Stoll, 2005), the extinction and srcination of shallow-waterlarger benthic foraminifera (e.g., Pujalte et al., 2003), latitudinalmigration of plants (Wing et al., 2005) and a rapid radiation of mammals on land (e.g., Koch et al.,1992). The relation between carboncycle perturbation, global warming and coeval biotic changes is notcompletely understood, because of the different response that indivi-dual ecosystems may have had to the effects of carbon release (Bowenet al., 2006). In particular, the causes of the rapid ( b 10 ky) benthicextinction event (BEE) in the deep-sea are not yet clear (e.g., Thomas,2007). Many deep-sea species that had survived the global environ-mental crisis of the Cretaceous/Paleogene boundary (e.g., Alegret andThomas,2005,2007;Alegret,2007)wentextinctatthePETM,andwerereplaced by post-extinction faunas of paleogeographically varyingtaxonomical composition (e.g., Thomas, 1998). In addition, in manysites and sections (including the Tethys area) the BEE coincides withcarbonate-depleted intervals that are almost barren of calcareousforaminifera (Ortiz, 1995; Thomas, 1998; Alegret et al., 2005, 2009).Whereas several studies on deep-sea benthic foraminifera from thecentral Tethys have been recentlycarried out (e.g., Galeotti et al., 2004;Giusberti et al., 2007, 2009), published data from the western Tethyshave only been reported from the Caravaca section in Southern Spain(Ortiz,1995), where the benthic foraminiferal record of the basal CIE isstrongly affected by dissolution. In contrast, a continuous succession of upper Paleocene and lower Eocene sediments is very well exposed atAlamedilla (33°N, Southern Spain; e.g., Arenillas and Molina, 1996; Luet al., 1996; Monechi et al., 2000). We document in detail the benthicforaminiferal turnover and extinction event associated with the carbonisotope excursion (CIE) at Alamedilla. In order to infer paleoenviron-mental consequences of the PETM, the benthic foraminiferal resultshavebeenintegratedwithgeochemicalandmineralogicaldata(Luetal.,1996, 1998); the recognition of several benthic foraminiferal assem-blageswithindistinctportionsof the δ 13 Ccurvemaybeausefultoolforcorrelation and paleoenvironmental reconstruction across the Paleo-cene – Eocene warming event. 2. Materials and methods A continuous succession of upper Paleocene and lower EocenepelagicsedimentsisverywellexposedatAlamedilla(SouthernSpain;Fig. 1), in the central Subbetic Cordillera (e.g., Arenillas and Molina,1996; Monechi et al., 2000).Upper Paleocene sediments consist of gray marls, with a 15-cm-thick turbiditic layer intercalated in the lower part of the studiedsection.Theonsetof thecarbonisotopeexcursion(CIE)wasidenti 󿬁 edby Lu et al. (1996,1998) in sample 13.25, in a level of pink marls thatgrade into a 30-cm-thick red clay interval (meters 13.50 to 13.80;Fig. 2). These authors used X-ray diffraction to determine the min-eral composition of bulk sediment, and distinguished calcite, detritus(de 󿬁 ned as quartz, K-feldspar and plagioclase) and phyllosilicates(wt.%). In coincidence with the onset of the CIE, these authorsdocumented a decrease in the percentage of calcite and an increase inthe percentage of silicate minerals, which make up to 12% of thewhole rock composition during the core of the CIE. However, due tothe low sampling resolution in the lowermost Eocene in the study byLu et al. (1996), we carried out analyses of the %CaCO 3  in somesamples across the critical BEE interval (Figs. 2, 5). The red clay Fig.1. A)LocationoftheAlamedillasection,andotherPaleocene – Eocenesectionsreferredtointhetext.B)Paleogeographicalreconstructionmodi 󿬁 edfromHayetal.(1999).GSSP=Global Stratotype Section and Point.187 L. Alegret et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 279 (2009) 186 –  200  Fig.2. δ 13 Ccurve (inbulk sediment) andpercentages ofCaCO 3 anddetritus(silicateminerals)at Alamedilla (modi 󿬁 edfromLuet al.,1996).Agemodel (A – F)according toRöhletal.(2007).Numberofbenthic foraminifera pergramof washedresidue 63 µm – 1 mm, percentages of agglutinated and infaunal taxa, and diversity and heterogeneity indices across the upper Paleocene and lower Eocene. White triangles represent the %CaCO 3  analyses carried out across the critical BEEintervalinthis study.(1)ArenillasandMolina(1996)andMolinaetal.(1999); (2)Monechi etal.(2000). M. = Morozovella; A. = Acarinina; P. wilcox .= Pseudohastigerina wilcoxensis; (3) Highestoccurrence of  Stensioeina beccariiformis .CIE=Carbon Isotope Excursion. PEB = Paleocene/Eocene boundary. 1   8   8   L   .A  l    e   g r  e  t   e  t   a  l    . /  P   a  l    a  e  o   g  e  o   g r  a   p h    y  ,P   a  l    a  e  o  c  l    i    m a  t   o  l    o   g   y  ,P   a  l    a  e  o  e  c  o  l    o   g   y 2   7   9    (  2   0   0   9    )  1   8   6   – 2   0   0    Fig. 3.  Stratigraphic distribution and relative abundance of benthic foraminiferal taxa across the upper Paleocene and lower Eocene at Alamedilla. Note that species that disappear in the Paleocene – Eocene transition have been represented inFig. 4.  δ 13 C curve (inbulk sediment) and percentages of calcite modi 󿬁 ed from Lu et al.(1996). White triangles represent the %CaCO 3 analysescarried out across the critical BEEinterval in this study.(1) Arenillas and Molina (1996) andMolina et al. (1999); (2) Monechi et al. (2000).  M. = Morozovella; A. = Acarinina; P. wilcox . =  Pseudohastigerina wilcoxensis . 1   8   9   L   .A  l    e   g r  e  t   e  t   a  l    . /  P   a  l    a  e  o   g  e  o   g r  a   p h    y  ,P   a  l    a  e  o  c  l    i    m a  t   o  l    o   g   y  ,P   a  l    a  e  o  e  c  o  l    o   g   y 2   7   9    (  2   0   0   9    )  1   8   6   – 2   0   0    interval is overlain by red marls and, higher up in the section, by graymarls (Fig. 2).In addition to geochemical and mineralogical studies (Lu et al.,1996, 1998), the planktic foraminiferal and nannofossil faunal turn-over across the PETM at Alamedilla have been documented (Arenillasand Molina, 1996; Lu et al., 1996; Monechi et al., 2000), but benthicforaminifera have not been studied in detail, so far. A total of 24samples have been studied across the studied interval, encompassingthe uppermost 2.25 m of the Paleocene and the lowermost 4 m of theEocene. Samples were picked at 10 to 20 cm intervals across the onsetandinthe “ core ” oftheCIE,andat~40cmintervalsbelowandaboveit(Fig. 2). The position of the studied samples was correlated to theresults by Lu et al. (1996,1998). Preservation of benthic foraminiferaltestsismoderateacrossthestudiedsection,althoughthepreservationof calcareous tests is poor in the lowermost Eocene, improving up-section.Quantitative studies of benthic foraminifera were based on repre-sentative splits (using a modi 󿬁 ed Otto micro-splitter)of approximately300 specimens larger than 63  μ  m (Plate I). The number of benthic foraminifera per gram of residue (63 µm – 1 mm), the percentages of benthic foraminiferal taxa (Fig. 3), the calcareous/agglutinated ratio,and several proxies for benthic foraminiferal diversity (Fig. 2) such asthe Fisher-aindex and theH(S)Shannon – Weaverinformationfunction(Murray,1991) were calculated. Probable microhabitat preferences andenvironmentalparameterswereinferredfromthebenthicforaminiferalmorphotype analysis (e.g., Corliss, 1985, 1991; Jones and Charnock,1985; Corliss and Chen,1988; Jorissen et al.,1995). Paleobathymetricalinferences were based on benthic foraminiferal data, especially on theoccurrence and abundance of depth-related species, their upper depthlimits, and through the comparison between fossil and recentassemblages (e.g., Van Morkhoven et al., 1986; Alegret et al., 2001,2003).Wefollowedthebathymetricdivisionde 󿬁 nedinVanMorkhovenet al. (1986): upper bathyal=200 – 600 m, middle bathyal=600 – 1000 m, lower bathyal=1000 – 2000 m and abyssal  N 2000 m.For biostratigraphical control, we follow the planktic foraminiferaland calcareous nannofossil zones identi 󿬁 ed by Arenillas and Molina(1996), Molina et al. (1999) and Monechi et al. (2000) at Alamedilla (Figs. 2, 3). In order to infer the timing across the CIE interval atAlamedilla, we studied the shape of the bulk carbonate  δ 13 C curve atAlamedillaandcomparedittothebulkcarbonate δ 13 CcurvefromOceanDrillingProgramSite(ODP)690,forwhichahigh-resolutionagemodelhas been recently developed by Röhl et al. (2007). The main peaks andplateaus of this curve may be related to the rapid changes in the inputandremovalofisotopicallylightcarbonacrossthePETM(e.g.,Röhletal.,2000).Basedonthisidea,wefollowedtheorbitallycalibratedtimescaleby Röhl et al. (2007) and chose six tie points (A – F; Fig. 2): Paleocene/Eocene boundary (PEB)=0 ky, A=0.75 kr, B=21.90 ky, C=42.38 ky,D=71.25ky,E=81.17ky,F=94.23ky.ThecorrelationbetweentheCIEat Alamedilla and ODP Site 690 is problematic because the onset of theexcursionappearstobelessabruptatAlamedilla.Thismightresultfroma lower time resolution of the bulk carbonate record in Lu et al. (1996)than in the Bains et al. (1999) record for Site 690. Provided the smallinitial shift in bulk  δ 13 C at Alamedilla is indeed coeval with the largerinitial shift at Site 690, one could argue that the record is much morecompleteatAlamedilla,atleastoverthebeginningof theCIE, making itone of the most complete marine records over that interval.In addition, we showa newage model and sedimentation rates forthe sections below and above the CIE at Alamedilla (Fig. 4), based ondepthsofdatumlevelsandnumericalagesaccordingtothetimescalesapplied to ODP Leg 208 (Shipboard Scienti 󿬁 c Party, 2004). This newage model updates that proposed by Lu et al. (1996) for Alamedilla. 3. Faunal turnover  Quantitative results of the benthic foraminiferalanalyses are shownin Figs. 2 and 3, and in Table 1. Benthic foraminiferal assemblages are strongly dominated bycalcareoustaxa(~90% of the assemblages)in allsamples but two (13.50 and 13.60), where agglutinated taxa make up47% and 96% of the assemblages, respectively. Infaunal morphogroupsaredominantthroughthewholestudiedinterval,makingup52 – 79%of theassemblages. The following benthic foraminiferalassemblageshavebeen recognised across the Paleocene – Eocene transition at Alamedilla.  3.1. Pre-extinction fauna Assemblages just below the extinction event (samples 11 to 13.20)are diverse (Fisher- α   index~31)and heterogeneous(Shannon – Weaverindex ~3.7; Fig. 2), and dominated by buliminids ( Bolivinoidesdelicatulus ,  Bulimina kugleri ,  Bulimina trinitatensis ), polymorphinids,  Anomalinoides  spp. (  A. ammonoides, A.  cf.  zitteli ),  Cibicidoides hyphalus , C. pseudoperlucidus ,  Nuttallides truempyi ,  Osangularia ,  Stensioeinabeccariiformis ,  Siphogenerinoides brevispinosa ,  Stilostomella  spp., etc.(Fig.3).Thecompositionoftheassemblagesisslightlydifferent(higherpercentages of agglutinated taxa such as  Clavulinoides amorpha  and Gaudryinapyramidata )insample11.45,whichisassociatedtoa~15-cm-thick turbidite deposited 2 m below the P/E boundary.The relative abundance of infaunal morphogroups increases fromthe bottom of the section (52%) towards the horizon of the main BEE(sample 13.50), where they show maximum values (79% of theassemblages). Diversity decreases, and the percentage of   Abyssaminaquadrata  starts to increase (making up to 10% of the assemblages)~20 – 25 cm below the onset of the CIE, coeval with a decrease in therelative abundance of   Anomalinoides  spp.  3.2. The benthic foraminiferal extinction event  The highest occurrences of 32 benthic foraminiferal species are re-corded in the uppermost Paleocene at Alamedilla, 6 of which correspond Fig. 4.  Newagemodelandsedimentation ratesforthesections belowandabovetheCIEatAlamedilla,mainlybasedondepthsofdatumlevelsofplankticforaminifera(ArenillasandMolina,1996),andcalcareousnannofossils(Monechietal.,2000),andonnumerical ages according to the timescales applied to ODP Leg 208 (Shipboard Scienti 󿬁 c Party,2004) and in Berggren and Pearson (2005). LAD: Last Appearance Datum; FAD: First Appearance Datum; X: Abundance crossover. Age of the P/E boundary according toWesterhold et al. (2008).  A. = Acarinina; I. = Igorina; M. = Morozovella; Ps. =Pseudohastigerina; S. = Subbotina; T. = Tribrachiatus; T. cont. = T. contortus . A1.8 cm/kyand 0.4 cm/ky sedimentation rate has been inferred for the uppermost Paleocene andlower Eocene, respectively, compared to the 2.4 cm/ky and 0.7 cm/ky inferred by Luet al. (1996). These results are somewhat different, but in the same orderof magnitude.Note the increase in sedimentation rates (4.7 cm/ky) at the P/E boundary (13.25 m),through the interval with a high percentage of silicate minerals (15 m). From that levelupwards,sedimentationratesremainlow(0.4cm/ky)fortherestofthestudiedinterval.190  L. Alegret et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 279 (2009) 186 –  200
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks