Most of our information about the evolution of Earth’s ocean–climate system comes from the analysis of sediments laid down in the past. For example, the microfossil assemblage reflects the temperature, salinity and nutrient abundance of the water in which the organisms lived, while the chemical and isotopic composition of biogenic carbonates may be used to reconstruct past variations in the operation of the carbon cycle, as well as changes in ocean circulation.
Nevertheless, understanding the link between these sediment variables (or ‘proxies’) and environmental conditions is not straightforward. This volume adopts a novel approach by bringing together palaeontologists, geochemists and palaeoceanographers, who contribute evidence that is required to better constrain these proxies. Topics include: (i) processes of biomineralization, and their effect on the chemical and isotopic composition of different organisms; (ii) proxy validation, including field, laboratory and theoretical studies; (iii) the links between modern and fossil organisms.
Biogeochemical Controls on Palaeoceanographic Environmental Proxies
The Geological Society of London Books Editorial Committee Chief Editor
BOB PANKHURST (UK) Society Books Editors
JOHN GREGORY (UK) JIM GRIFFITHS (UK) JOHN HOWE (UK) PHIL LEAT (UK) NICK ROBINS (UK) JONATHAN TURNER (UK) Society Books Advisors
MIKE BROWN (USA) ERIC BUFFETAUT (FRANCE ) JONATHAN CRAIG (ITALY )
RETO GIERE´ (GERMANY ) TOM MC CANN (GERMANY ) DOUG STEAD (CANADA ) RANDELL STEPHENSON (NETHERLANDS )
Geological Society books refereeing procedures The Society makes every effort to ensure that the scientific and production quality of its books matches that of its journals. Since 1997, all book proposals have been refereed by specialist reviewers as well as by the Society’s Books Editorial Committee. If the referees identify weaknesses in the proposal, these must be addressed before the proposal is accepted. Once the book is accepted, the Society Book Editors ensure that the volume editors follow strict guidelines on refereeing and quality control. We insist that individual papers can only be accepted after satisfactory review by two independent referees. The questions on the review forms are similar to those for Journal of the Geological Society. The referees’ forms and comments must be available to the Society’s Book Editors on request. Although many of the books result from meetings, the editors are expected to commission papers that were not presented at the meeting to ensure that the book provides a balanced coverage of the subject. Being accepted for presentation at the meeting does not guarantee inclusion in the book. More information about submitting a proposal and producing a book for the Society can be found on its web site: www.geolsoc.org.uk.
It is recommended that reference to all or part of this book should be made in one of the following ways: AUSTIN , W. E. N. & JAMES , R. H. (eds) 2008. Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303. SCHMIDT , D. N., ELLIOTT , T. & KASEMANN , S. A. 2008. The influences of growth rates on planktic foraminifers as proxies for palaeostudies – a review In: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 73 –85.
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 303
Biogeochemical Controls on Palaeoceanographic Environmental Proxies
EDITED BY
W. E. N. AUSTIN University of St. Andrews, UK and
R. H. JAMES The Open University, UK
2008 Published by The Geological Society London
THE GEOLOGICAL SOCIETY The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world and the largest in Europe. It was incorporated under Royal Charter in 1825 and is Registered Charity 210161. The Society is the UK national learned and professional society for geology with a worldwide Fellowship (FGS) of over 9000. The Society has the power to confer Chartered status on suitably qualified Fellows, and about 2000 of the Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent European title, European Geologist (EurGeol). One fifth of the Society’s fellowship resides outside the UK. To find out more about the Society, log on to www.geolsoc.org.uk. The Geological Society Publishing House (Bath, UK) produces the Society’s international journals and books, and acts as European distributor for selected publications of the American Association of Petroleum Geologists (AAPG), the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the Society for Sedimentary Geology (SEPM) and the Geologists’ Association (GA). Joint marketing agreements ensure that GSL Fellows may purchase these societies’ publications at a discount. The Society’s online bookshop (accessible from www.geolsoc.org.uk) offers secure book purchasing with your credit or debit card. To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other societies worldwide, consult www.geolsoc.org.uk, or contact the Fellowship Department at: The Geological Society, Burlington House, Piccadilly, London W1J 0BG: Tel. þ44 (0)20 7434 9944; Fax þ44 (0)20 7439 8975; E-mail:
[email protected]. For information about the Society’s meetings, consult Events on www.geolsoc.org.uk. To find out more about the Society’s Corporate Affiliates Scheme, write to
[email protected]. Published by The Geological Society from: The Geological Society Publishing House, Unit 7, Brassmill Enterprise Centre, Brassmill Lane, Bath BA1 3JN, UK (Orders: Tel. þ44 (0)1225 445046, Fax þ44 (0)1225 442836) Online bookshop: www.geolsoc.org.uk/bookshop The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. # The Geological Society of London 2008. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/08/$15.00. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 978-1-86239-257-1 Typeset by Techset Composition Ltd., Salisbury, UK Printed by The Cromwell Press Ltd, Wiltshire, UK Distributors North America For trade and institutional orders: The Geological Society, c/o AIDC, 82 Winter Sport Lane, Williston, VT 05495, USA Orders: Tel þ1 800-972-9892 Fax þ1 802-864-7626 E-mail
[email protected] For individual and corporate orders: AAPG Bookstore, PO Box 979, Tulsa, OK 74101-0979, USA Orders: Tel þ1 918-584-2555 Fax þ1 918-560-2652 E-mail
[email protected] Website http://bookstore.aapg.org India Affiliated East-West Press Private Ltd, Marketing Division, G-1/16 Ansari Road, Darya Ganj, New Delhi 110 002, India Orders: Tel þ91 11 2327-9113/2326-4180 Fax þ91 11 2326-0538 E-mail
[email protected]
Contents AUSTIN , W. E. N. & JAMES , R. H. Biogeochemical controls on palaeoceanographic environmental proxies: an introduction
1
JAMES , R. H. & AUSTIN , W. E. N. Biogeochemical controls on palaeoceanographic environmental proxies: a review
3
WILLIAMS , R. J. P. Some fundamental features of biomineralization
33
ZEEBE , R. E., BIJMA , J., HO¨ NISCH , B., SANYAL , A., SPERO , H. J. & WOLF -GLADROW , D. A. Vital effects and beyond: a modelling perspective on developing palaeoceanographic proxy relationships in foraminifera
45
PEARSON , P. N. & BURGESS , C. E. Foraminifer test preservation and diagenesis: comparison of high latitute Eocene sites
59
SCHMIDT , D. N., ELLIOTT , T. & KASEMANN , S. A. The influences of growth rates on planktic foraminifers as proxies for palaeostudies – a review
73
CUIF , J. P., DAUPHIN , Y., MEIBOM , A., ROLLION -BARD , C., SALOME´ , M., SUSINI , J. & WILLIAMS , C. T. Fine-scale growth patterns in coral skeletons: biochemical control over crystallization of aragonite fibres and assessment of early diagenesis
87
GOODAY , A. J., NOMAKI , H. & KITAZATO , H. Modern deep-sea benthic foraminifera: a brief review of their morphology-based biodiversity and trophic diversity
97
MACKENSEN , A. The use of benthic foraminiferal d13C in palaeoceanography: constraints from primary proxy relationships
121
MC CORKLE , D., BERNHARD , J. M., HINTZ , C. J., BLANKS , J. K., CHANDLER , G. T. & SHAW , T. J. The carbon and oxygen stable isotopic compostion of cultured benthic foraminifera
135
CAGE , A. G. & AUSTIN , W. E. N. Seasonal dynamics of coastal water masses in a Scottish fjord and their potential influence on benthic foraminiferal shell geochemistry
155
CRAVEN , K. F., BIRD , M. I., AUSTIN , W. E. N. & WYNN , J. Isotopic variability in the intertidal acorn barnacle Semibalanus balanoides: a potentially novel sea-level proxy indicator
173
Index
187
Biogeochemical controls on palaeoceanographic environmental proxies: an introduction WILLIAM E. N. AUSTIN1 & RACHAEL H. JAMES2 1
University of St Andrews, UK (e-mail:
[email protected]) 2
Open University, UK (e-mail:
[email protected])
The current volume samples a selection of papers presented at the Geological Society of London meeting on ‘Biogeochemical Controls on Palaeoceanographic Proxies’, held at Burlington House, London, UK on 3–4 October 2005. The aim of the meeting was to bring together palaeontologists, geochemists and palaeoceanographers who could contribute evidence that, when considered together, would better constrain the proxies that are used for palaeoclimate reconstruction. An improved understanding and quantification of past climate change, and the processes that force climate to change, has a fundamental role to play in constraining model projections of future climate (e.g. Hegerl et al. 2006) but it remains a huge challenge. This is because key climate variables, such as temperature and ocean salinity, cannot be observed in a world which no longer exists, but must instead be teased from proxies in the geological and ice records. There are numerous proxy archives, but one of the most important, currently lying at the forefront of palaeoceanographic research, is the biogeochemical composition of sediment records. This publication consists of 11 papers which deal with various aspects of biogeochemical proxies and their interpretation in terms of past climate. Seven of these specifically focus on the Foraminifera. What are proxies? Primarily, these are biogenic components which have a close relationship to environmental parameters and may be identified as so-called ‘proxy variables’ (Wefer et al. 1999), providing measurable descriptors of key climatic and environmental variables. The methods commonly employed in palaeoceanography have their origins in the biological, chemical and physical sciences; palaeoceanography therefore represents a relatively young and truly crossdisciplinary field of research. At the time of writing, an excellent new book entitled Proxies in Late Cenozoic Paleoceanography has been published (Hillaire-Marcel & De Vernal 2007), providing a comprehensive review of the subject. This volume begins with an overview by James & Austin, highlighting some of the most important biological and geochemical proxies, and outlining their contribution to our understanding of the
ocean –climate system. We anticipate that this review article will provide an accessible introduction to the topic as well as an insight into a wide range of analytical methods. The second paper by Williams highlights some of the fundamental features of the biomineralization process, initially contrasting these with inorganic mineral formation, before selecting examples from corals, foraminifera and coccoliths. The paper by Zeebe et al. develops a modellingbased approach to improve our understanding of stable carbon, oxygen, and boron isotopes as well as magnesium incorporation into foraminiferal calcite shells. The approach adopted by these authors is based on an integrated understanding of the inorganic chemistry, inorganic isotope fractionation, and biological controls that determine palaeo-tracer signals in organisms used in climate reconstructions. They argue very convincingly that the integration of laboratory experiments, field and culture studies, theoretical considerations, and numerical modelling holds the key to the method’s success. These authors demonstrate that a mechanism-based understanding is often required before primary climate signals can be extracted from the geological record, because these signals can be heavily overprinted by secondary, nonclimate related phenomena. One such phenomenon is foraminifer test diagenesis, which Pearson & Burgess highlight in their study of high-latitude Eocene sites. These authors illustrate the textures using a combination of reflected light microscopy and high-resolution scanning electron microscopy, highlighting the fact that foraminifer tests are prone to diagenetic recrystallization on a micron scale, which can affect their geochemical composition. For example, diagenetic calcite added on the sea floor or in shallow burial at low temperature, may result in the severe underestimation of the apparent tropical sea surface temperatures obtained from calcite foraminifera (e.g. Pearson et al. 2001). Next, two papers consider the growth rates and growth patterns in biogenic calcite. The first of these, by Schmidt et al. provides an overview of the effect of growth rate on the trace element
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 1 –2. DOI: 10.1144/SP303.1 0305-8719/08/$15.00 # The Geological Society of London 2008.
2
W. E. N. AUSTIN & R. H. JAMES
composition of planktonic foraminifera. Crucially, they show that up to 40% of the glacial-interglacial variation in foraminiferal Sr/Ca may be linked to changes in foraminifera test size, as opposed to the Sr/Ca ratio of seawater. The second, by Cuif et al. illustrates very fine-scale growth patterns in coral skeletons, suggesting a biogeochemical control on the crystallization of aragonite fibres and providing an assessment of early diagenesis. The next four papers focus exclusively on benthic foraminifera. An interesting review of deep-sea benthic foraminifera is provided by Gooday et al., highlighting the numerical importance of abyssal taxa below the calcite compensation depth and their likely role in global carbon cycling. These foraminiferal assemblages have a low fossilization potential and the authors highlight the substantial loss of apparent biomass and biodiversity that may result from estimates of palaeoproductivity based upon deep-sea benthic foraminiferal assemblage data. On a related theme, the paper by Mackensen considers the use of benthic foraminiferal stable isotopes and provides a thorough review of this important palaeoceanographic tracer. While epifaunal species largely reflect the carbon isotope composition of the dissolved inorganic carbon in sea water, this paper highlights the roles of organic carbon fluxes to the sea floor, pore-water chemistry and methanogens upon the stable carbon isotope composition of infaunal species. A novel and highly innovative approach which may ultimately improve our understanding and calibration of foraminiferal proxies is presented by McCorkle et al., in which the authors present results on the carbon and oxygen stable isotopic composition of cultured benthic foraminifera. Offsets between the chemistry of the cultured foraminifera shells and that of the system water are thought to reflect at least three factors: speciesdependent vital effects; ontogenetic variations in shell chemistry; and the aqueous carbonate chemistry of the experimental system. The authors suggest that our understanding of some key factors influencing benthic foraminiferal shell chemistry remains incomplete; culture systems are therefore likely to prove extremely important in isolating the most important biogeochemical controls. The theme of non-equilibrium isotope behaviour in foraminiferal shell carbon and oxygen stable isotopic composition is continued by Cage & Austin, who present evidence for seasonally driven temperature
changes in mid-latitude shelf seas that explain a so-called ‘seasonal effect’. Finally, a novel proxy is introduced by Craven et al., who provide empirical data to show that stable carbon isotopic variability in organic tissue of the inter-tidal acorn barnacle, Semibalanus balanoides, is related to immersion/emersion times. With more work, this technique could potentially provide critical information about small-scale, but rapid, variations in past sea-level; such information is crucial for assessing the likely impact of anthropogenic-induced climate change. We would like to thank all the participants at the Geological Society of London meeting on ‘Biogeochemical Controls on Palaeoceanographic Proxies’, held at Burlington House, London on 3–4 October 2005. We thank our fellow meeting organisers, Ros Rickaby (Oxford) and Leon Clarke (Bangor), together with Jann Matela and Andy Lloyd of the Open University, who dealt expertly with the administration and graphic design, respectively. Staff of the Geological Society Publishing House, notably Angharad Hills and Helen Floyd-Walker, have provided invaluable guidance in the production of this volume and we particularly acknowledge the expert advice of our referees, who gave their time to review the papers published here. Finally, we acknowledge the generous support of our various sponsors: The Marine Studies Group, The Challenger Society for Marine Science, The Quaternary Research Association, The Micropalaeontological Society, The Geochemistry Group, The UK Integrated Ocean Drilling Program, New Wave Research, Varian and GV Instruments.
References H EGERL , G. C., C ROWLEY , T. J., H YDE , W. T. & F RAME , D. J. 2006. Climate sensitivity constrained by temperature reconstructions over the past seven centuries. Nature, 440, 1029–1032. H ILLAIRE -M ARCEL , C. & D E V ERNAL , A. 2007. Proxies in Late Cenozoic Paleoceanography. Elsevier. P EARSON , P. N., D ITCHFIELD , P. W., S INGANO , J. ET AL . 2001. Warm tropical sea surface temperatures in the Late Cretaceous and Eocene epochs. Nature, 413, 481–487. W EFER , G., B ERGER , W. H., B IJMA , J. & F ISCHER , G. 1999. Clues to ocean history; A brief overview of proxies. In: F ISHER , G. & W EFER , G. (eds) Uses of Proxies in Paleoceanography: Examples from the South Atlantic. Springer-Verlag, Berlin Heidelberg, 1– 68.
Biogeochemical controls on palaeoceanographic environmental proxies: a review RACHAEL H. JAMES1 & WILLIAM E. N. AUSTIN2 1
Department of Earth Sciences, The Open University, Walton Hall, Milton Keynes, England, MK7 6AA (e-mail:
[email protected])
2
School of Geography & Geosciences, University of St Andrews, St Andrews, Fife, Scotland, KY16 9AL (e-mail:
[email protected]) Abstract: Scientific observations of our oceans and climate go back no more than a couple of hundred years. Most of our information about the evolution of Earth’s ocean-climate system relies instead on proxies – primarily measurements of sediment components that respond to changes in environmental parameters. This paper provides an overview of some of the most important biological and geochemical proxies and outlines their contribution to our understanding of the ocean-climate system. We also discuss some of the challenges that need to be overcome to obtain accurate records. These include: better understanding of the controls on the mechanisms of biomineralization; the impacts of post-depositional dissolution and diagenesis on primary proxy relationships; proxy validation; and analytical considerations.
The methods commonly employed in palaeoceanography have their origins in the biological, chemical and physical sciences; palaeoceanography therefore represents a relatively young and truly crossdisciplinary field of research. Where sediments and biogenic components have a close relationship to environmental parameters, so-called ‘proxy variables’ may be identified (Wefer et al. 1999). Such proxy records provide measurable descriptors of key climatic and environmental variables, surface ocean temperature being among the most important of these. In recent decades, proxies have been combined and the multi-proxy approach has gained considerable favour, providing a robust test of past climatic and environmental reconstructions and enabling important variables such as surface ocean salinity (e.g. Duplessy et al. 1991) to be reconstructed for the first time. Perhaps one of the most widely acknowledged early examples of the use of proxy-methods in palaeoceanography were the synoptic temperature reconstructions of the CLIMAP group (1976, 1981), who provided surface-ocean temperatures for the Last Glacial Maximum (Fig. 1). In this approach, a transform equation is derived from a calibration set, the latter comprising both the proxy variable (planktonic foraminifera) and target parameter of interest (sea surface temperature (SST)). Imbrie & Kipp’s (1971) method, upon which the CLIMAP reconstructions are based, use abundance data from palaeontological samples in order to estimate palaeotemperature. Their approach, which proved particularly important in the development of quantitative palaeoceanography
for a whole range of microfossil groups, utilizes factor regression to develop transform equations and works well at estimating temperatures from the mid-range of the calibration set. However, as Wefer et al. (1999) highlight in their comprehensive review of this subject, a number of assumptions are built into these proxy methodologies, so that the sources of error in the final target parameter may be large or even poorly known. Critically, the calibration data often relate surface sediment data (so-called ‘core-top’ samples) to variables in the overlying water column, such as SST. The aim of this review chapter is to highlight the role of biogeochemical processes upon some of the most commonly employed palaeoceanographic proxies.
Biomineralization: a globally significant process Biomineralization is the process whereby organisms form minerals, and research in this field provides an excellent example at the interface between the earth and life sciences. Early research focussed on biologically formed calciumcontaining minerals and the field of calcification was born (e.g. Schmidt 1924). However, as the diversity of known biogenic minerals has grown, currently 64 different minerals are known (Knoll 2003), the general field of research has become known as biomineralization (Weiner & Dove 2003). The term ‘biomineral’, as Weiner & Dove (2003) highlight in their excellent review of the subject, refers not only to the mineral itself, but
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 3 –32. DOI: 10.1144/SP303.2 0305-8719/08/$15.00 # The Geological Society of London 2008.
4
R. H. JAMES & W. E. N. AUSTIN
Fig. 1. The CLIMAP reconstruction of Last Glacial Maximum (LGM) August sea surface temperatures (SSTs) for the global ocean based on planktonic foraminiferal assemblage data. Modified after CLIMAP (1984; http://iridl.ldeo.columbia.edu).
also to the organic components associated with the process of biomineralization. These processes, reflecting the degree of biological control, are divided into two fundamentally different groups (Lowenstam 1981): ‘biologically induced’ and ‘organic matrix-mediated’. In the case of biologically induced mineralization, we recognize a process of secondary precipitation of minerals resulting from the interactions between biological activity and the environment (e.g. McConnaughey, 1989). In biologically controlled mineralization, a generalized term for organic matrix-mediated mineralization adopted by Mann (1983), cellular activity locates and controls the nucleation, growth and morphology of the biomineral. These processes occur extra-, inter- or intracellularly and have a profound effect upon the morphology and the compositional signatures contained in biogenic minerals. As such, both biological processes and environmental signals are encoded into biominerals, the former often know as ‘vital effects’. One of the major challenges now facing the palaeoceanographic community is the need to improve our understanding of some of the fundamental controls on the mechanisms of biomineralization and an understanding of vital effects will be central to this effort.
The processes of biomineralization reflected in the form and function of skeletons have been shaped by natural selection through geologic time. As Knoll (2003) highlights, the skeletal export of calcium carbonate and silica from today’s oceans did not exist during early Earth history, yet the supply of calcium and silica from weathering introduced a continuous supply to the Proterozoic oceans. Proterozoic and Phanerozoic sedimentary successions therefore differ more in terms of the facies distributions rather than in the abundance of carbonate and silica in preserved deposits. During the Proterozoic, before the widespread Phanerozoic evolution of mineralized skeletons in the eukaryotes, CaCO3 and SiO2 precipitation and accumulation was driven by evaporation in shallow shelf seas (e.g. Knoll & Swett 1990). Although primarily driven by physical processes, microscopic crystal structures preserved in silicified stromatolites highlight the role of bacteria as facilitators of this deposition (Knoll & Semikhatov 1998). The oldest evidence from which protistan biomineralization has been inferred comes from 724 + 6 million year old rocks, where the fossils are preserved as pyrite veneers (Porter & Knoll 2000). Skeletal biomineralization clearly played a part in the early diversification of animals, but it
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
is the so-called ‘Cambrian explosion’ that marks the most remarkable period of skeletal evolution in the metazoans (see Knoll 2003). Indeed, all the major skeletal biominerals appear in the geological record at this time, most likely providing their owners with a diverse range of protective skeletons in reponse to the appearance of effective bilaterian animal predators at this time (Bengtson 1994). Biologically produced carbonates represent the biosphere’s largest carbon reservoir, hence calcareous organisms affect the ocean’s pH and CO2 content, directly influencing atmospheric CO2 and ultimately global temperatures (Cohen & McConnaughey 2003). A growing interest in the short-term redistribution of anthropogenic CO2 has arisen because of these climatic effects (see Sarmiento & Gruber 2002). The sequestration of anthropogenic CO2 by the oceans occurs primarily through two key mechanisms, the so-called ‘biological’ and ‘carbonate’ pumps, which together account for the export to the deep ocean of organic matter and calcium carbonate produced in the surface ocean. The processes involved in the ocean carbon cycle and its interaction with anthropogenic CO2 are complex (Royal Society 2005), for example the production of CaCO3, while providing a net sink for dissolved inorganic carbon (DIC), generates CO2: Ca2þ þ 2HCO 3 $ CaCO3 þ H2 O þ CO2 Dissolution of CaCO3 below the photic zone, i.e. the above reaction in reverse, is therefore a net sink for CO2 and recent studies (e.g. Milliman et al. 1999) have suggested that significant dissolution may be coupled with organic matter respiration at shallower depths in the ocean i.e. above 1000 m – potentially providing an important sink for anthropogenic CO2 on short-term timescales. While the weathering of carbonate rocks does not affect long-term atmospheric CO2 concentration because of carbonate burial in the ocean (which releases CO2), the process is different when silicate rocks are weathered to provide ions for marine carbonate formation (e.g. Berner 1990), because there is a net consumption of CO2: CaSiO3 þ CO2 $ CaCO3 þ SiO2 The long time intervals that separate the deposition of marine carbonates and the return of volcanic CO2, i.e. the above reaction in reverse, as a result of the thermal decomposition of carbonates deep within the lithosphere, can lead to an imbalance in atmospheric CO2 concentrations.
5
Biological proxy records For convenience, the fossil world is generally divided according to the rather arbitrary terminology of larger macrofossils and smaller microfossils (e.g. Armstrong & Brasier 2005). Micropalaeontology is a collective term for the study of microfossils, often applied to those groups with mineral walls, and palynology is applied to organic-walled microfossils. Whilst numerous groups of organisms have contributed to the field of palaeoceanography, the current review is restricted to the following microfossil groups: foraminifera, coccolithophores and diatoms. Our review of macrofossils is similarly restricted and deals with corals and molluscs; the study of the latter groups can conveniently be classified as ‘sclerochronology’. Sclerochronology is the study of physical and chemical variations in the accretionary hard tissues of organisms, and the temporal context in which they formed.
Foraminifera The Foraminiferida (informally referred to as foraminifera) are eukaryotic unicellular organisms, classified in the Kingdom Protoctista, Phylum Granuloreticulosa, Class Foraminifera (Sen Gupta 1999). They range from the Early Cambrian to the present day and are either planktonic or benthic in life habitat. Of the approximately 10 000 extant foraminiferal species known, only 50 are planktonic, the remainder being benthic. Adult specimens span almost four orders of magnitude in size, ranging from approximately 50 mm to greater than 12 cm in some tropical forms. More recently, multi-nucleated deep-sea Xenophyophores, previously placed in a separate class, have been suggested to belong to the phylum Granuloreticulosa, some of which measure more than 25 cm in diameter (Gooday 1999). As a group, the foraminifera have considerable biological significance; their live density can exceed 106 per m2 and wet weight biomass estimates of more than 10 g per m2 are reported (Culver 1993). The soft tissue (cytoplasm) of the cell is largely enclosed with a shell or test, which may be composed of organic matter (tectin) (see Gooday et al. 2008), secreted minerals (calcite, aragonite or silica) or agglutinated particles. Tests may consist of single (unilocular) or multiple (multilocular) chambers; the latter display episodic growth, so that each new chamber can be considered a separate growth phase (Reiss 1958). As new chambers form, calcification occurs on both the inner and outer surface of an organic template, leading in most cases to the formation of a thin inner lamella and a thicker outer lamella
6
R. H. JAMES & W. E. N. AUSTIN
(Debenay et al. 2000). Wall ultra-structure is often complex (e.g. Hansen 1999) so that as new chambers are added to the test, the lamellar mode of construction may result in walls and septae formed by the juxtapositioning of calcite of different ages that are interspersed with organic material (Allison & Austin 2003). These skeletal organic materials are known to have substantial metal binding capacities (Mitterer 1978) and their presence in foraminiferal tests may affect trace metal determinations (Allison & Austin 2003).
Coccolithophores The term coccolithophores is used to describe living marine, unicellular phytoplankton, belonging to the phylum Haptophyta and the class Prymnesiophyceae (Edvardsen et al. 2000), the latter also including non-calcifying organisms. Living coccolithophores typically have an exoskeleton composed of numerous minute calcite platelets called coccoliths, which typically range from 3–15 mm in diameter and which are readily preserved in the sedimentary record. These minute algal protists made their first appearance in the Late Triassic (e.g. Bown et al. 2004), marking a transition in the locus of global calcification from the continental shelves towards the deep ocean and driving fundamental changes in ocean chemistry and patterns of sedimentation (Baumann et al. 2005; Hay 2004). Today, the group represent an extremely abundant and important constituent of the ocean phytoplankton and are gaining increasing attention because of their dual role in the ocean biological and carbonate pumps, together with their influence on climate feedback mechanisms driven by their release of dimethylsulphide, whose molecules act as a source of cloud nucleation (Westbroek et al. 1993). Approximately 200 extant coccolithophore species (diploid stage) are recognized (Young et al. 2003). Coccolithophore taxonomy is based upon coccolith morphology and two major groups are recognised, the holo- and heterococcoliths, the latter stage being the most common. There are some important differences in the style and mechanisms of biomineralization in these groups: holococcoliths are formed from numerous, minute euhedral calcite crystallites and are thought to form outside the cell (Young et al. 1999), whereas heterococcoliths are formed intracellularly from a radial array of variably shaped crystal units (Young et al. 1997). In general, there is relatively little data available on the relative contributions made by different planktonic calcifying organisms (i.e. coccolithophores, foraminifera, pteropods) to the global production of calcium carbonate. Estimates of the coccolith contribution to pelagic marine carbonate
production range from 20 –60% (e.g. Ziveri et al. 1999), although this figure may exceed 70wt-% in the surface sediments of the oligotrophic gyres of the South Atlantic (e.g. Baumann et al. 2005) and increase to .80wt-% in mid-Brunhes time at Site 1082 off SW Africa (Baumann & Freitag 2005). Historically, coccolithophore biogeography has been based on the recognition of rather broadly defined floral zones (e.g. McIntyre & Be´ 1967): Subarctic, Temperate, Subtropical, Tropical and Subantarctic. More recently, species-specific biogeographies have been established (e.g. Ziveri et al. 2004), suggesting that the early floral zones, while broadly useful, are probably too simplistic. Based on ocean-scale distributions of nannoplankton, Young (1994) recognized three distinct assemblages based on coccolith type and coccosphere morphology that characterize particular environments (Fig. 2). Placolith-bearing species characterise meso- to eutrophic environments, such as upwelling areas and include Emiliania huxleyi, Gephyrocapsa spp., Umbilicosphaera spp. In this assemblage, the coccoliths comprise a proximal and distal shield joined by a central column. The oligotrophic and mid-ocean environments are dominated by umbelliform assemblages e.g. Umbellosphaera spp. and Discosphaera tubifera, whose coccoliths bear large processes and which flare distally to produce a double-layered coccosphere. The stable water column of the deep photic zone of the low- to mid-latitudes are dominated by floriform species, whose coccospheres are characterized by a dense assymetrical mass of coccoliths. A number of factors are thought to influence the transformation of living coccolithophores from the surface ocean into the sedimentary record, many of which have an important bearing on the fossil record (see review by Baumann et al. 2005). These include incorporation into fast sinking ‘marine snow’, breakage and dissolution due to grazing. The calcareous nannoplankton are among the most important microfossils used in Mesozoic and Cenozoic biostratigraphy, because of: (i) their very high abundance in most marine sediments; (ii) their wide geographic distribution; and (iii) their rapid evolutionary development which allow for narrow subdivisions of geological time.
Diatoms Diatoms are unicellular algae, coloured goldenbrown by photosynthetic pigments and encased in an opaline silica skeleton: the frustule. Unlike calcareous nannoplankton and dinoflagellates, the vegetative cells of diatoms lack flagella, although flagellated gametes are produced during sexual reproduction. Planktonic diatoms are widely
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
Fig. 2. Four palaeoecologically significant extant coccolithophores. Each specimen is represented by an image of the complete coccosphere (left) and detail of the coccoliths (right). Figure courtesy of Jeremy Young and Craig Koch, The Natural History Museum, London. From the top (dimensions are coccosphere diameters): 1, Coccolithus pelagicus ssp. pelagicus (coccosphere diameter 12.6 mm), one of the most useful cold water markers; 2, Gephyrocapsa oceanica (6.9 mm), a high productivity and warm water indicator. Gephyrocapsa and Emiliania are also responsible for production of alkenone biomarkers used in the U37 K palaeothermometry technique; 3, Braarudosphaera bigelowii (16.1 mm) a low salinity indicator; and 4, Florisphaera profunda (12.6 mm), a deep photic species which acts as a useful proxy for surface-water oligotrophy.
7
8
R. H. JAMES & W. E. N. AUSTIN
distributed in the surface ocean (photic zone) and have adaptations to promote floatation (frustule shapes and processes that increase the ratio of surface area:volume, formation of colonies, storage of fats and oils to reduce the specific gravity of the cell). The oldest reliable records of diatoms is from the Early Jurassic (Toarcian stage), although abundant, well-preserved diatoms do not appear until the Aptian-Albian. Diatoms are taxonomically placed in a separate class, the Bacillariophyceae and, together with the Chrysophyceae and Xanthophyceae, belong in the division Chrysophyta. As a group, they all form endoplasmic cysts, secrete silica, store oils rather than starch, and process a bipartite wall. Diatoms themselves are divided into two orders: the Centrales, which exhibit radial symmetry and the Pennales, which exhibit bilateral symmetry. The taxonomy of the group is based upon the external opaline skeleton of the frustule, which consists of two overlapping valves: the larger epivalve and the smaller hypovalve, the latter forms internally adjacent to the parental epivalve during vegetative division of the cell. Between these valves, lies the girdle, itself comprising the epicingulum and hypocingulum (Fig. 3). Diatoms are autotrophic and, as such, form the basis of food chains in many aquatic ecosystems (e.g. Austin et al. 2005). Their requirements of light and so-called biolimiting nutrients (e.g. silica, nitrates, phosphates and iron), mean that the seasonal changes in diatom abundance in the surface ocean tend to be marked, often
Epivalve + Epicingulum = Epitheca Epivalve
Epicingulum Girdle Hypocingulum
Hypovalve
Hypovalve + Hypocingulum = Hypotheca Epitheca + Hypotheca = Frustule Fig. 3. Diatom frustule components in schematic cross-section view (modified from Barron 1993).
characterised by spring and late summer blooms. Planktonic diatoms under bloom conditions may number as many as 1000 million cells per m3 of water (Armstrong & Brasier 2005). However, diatoms living under high nutrient availability, such as regions of upwelling, often face an acute shortage of dissolved silica, leading to weakly silicified frustules (e.g. Conley et al. 1994). The degree of silicification of the diatom frustule will directly influence its preservation potential, often leading to the loss of a sedimentary record from the most productive season (e.g. Sancetta 1992). Indeed, modern seawater is undersaturated with respect to silica, largely because of the vast amounts taken up during diatom biomineralization. For this reason, diatom frustules are prone to dissolution, for example by pressure at depth or alkaline conditions, and therefore exhibit marked regional differences in preservation (Nelson et al. 1995).
Corals Corals belong to a large class of organisms, the anthozoans, within the phylum Cnidaria and have a fossil record spanning more than 400 million years. As in other cnidarians, most corals contain symbiotic algae (zooxanthellae). The extraordinary biological productivity and calcium carbonate synthesis in reef-building corals owes much to the mutual exchange of algal photosynthates and cnidarian metabolites (Barnes 1987). For example, calcification rates are significantly reduced in corals from which zooxanthellae are removed or expelled (as in ‘bleaching’ events) or under reduced light conditions (Lalli & Parsons 1995). Corals may grow as solitary or colonial forms. The process of calcification in stony corals (scleractinians) occurs in the lower portion of the polyp, producing a cup-like structure, the calyx, in which the polyp (typically 1–3 mm in diameter) sits. Structural elements of calyx comprise the surrounding walls (theca), the floor (basal plate), and, extending upwards from the basal plate a series of thin, radial elements (sclerosepta). As a consequence of the rather restrictive environmental conditions required to satisfy the algal-cnidarian symbiotic relationship, reefbuilding corals are restricted in their geographical distribution. Reefs are generally confined to the euphotic zone (,70 m) of the tropics, typically to latitudes of around 308, growing optimally at water temperatures between 238 and 29 8C and at salinities above 32. While most coral reefs are associated with shallow tropical seas, recent discoveries have highlighted the scale and abundance of cold-water coral
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
ecosystems throughout the world’s oceans (e.g. Roberts et al. 2006). They are known to comprise stony corals (Scleractinia), soft corals (Octocorallia), black corals (Antipatharia) and hydrocorals (Stylasteridae). Unlike most of the tropical reefforming species, cold-water corals are azooxanthellate. Typically occurring at water temperatures between 48 and 12 8C, their depth distributions are variable, ranging from c. 50 m at high latitudes to nearly 4000 m in the low latitudes. In the case of scleractinian cold-water corals, their global biogeography also appears to critically depend on the depth of aragonite saturation, which may help to explain the paucity of records in the North Pacific, where waters are generally undersaturated with respect to aragonite. Roberts et al. (2006) provide a very useful overview of the biology and geology of cold-water coral ecosystems.
Molluscs The Mollusca are a large and diverse invertebrate phyla, making their first appearance during the early Cambrian, with a major radiation during late Cambrian –early Ordovician times (Runnegar 1996). Perhaps the most widely studied molluscs in palaeoceanography are the Bivalvia, a largely marine group with paired, unequilateral calcareous shells joined by a dorsal hinge. While making an early appearance in the Cambrian, bivalves exhibit a relatively limited abundance throughout the Palaeozoic, becoming much more abundant in the Mesozoic. The shells of bivalves are multilayered, consisting of two intermixed phases (Wilbur 1961), namely an organic matrix and crystalline calcium carbonate in the form of calcite or aragonite. Many bivalve shells contain a mixture of both calcite and aragonite, often occurring in recurrent patterns and discrete shell layers. However, no aragonite shells are known before the Carboniferous, presumably reflecting diagenetic changes, rather than original shell mineralogy. These structural features have an important role to play in evolutionary studies (e.g. Taylor et al. 1969) and are usefully summarized by Clarkson (1998). The incremental nature of shell growth in bivalves provides an opportunity for detailed temporal reconstruction of past environments (e.g. Weidman et al. 1994) and their widespread distribution from freshwater to deep-sea environments make them particularly useful for palaeoenvironmental reconstructions. However, as Freitas et al. (2006) highlight, the potential of bivalves as useful geochemical proxies remains to be fully realized and these authors cite the limited number of calibration and validation studies completed to date.
9
Geochemical proxy records As we have seen in the previous section, the microfossil assemblage of marine sediments is a key palaeoceanographic proxy. Other important proxies are based on the chemical composition of these microfossils, in terms of the chemical elements and the relative proportions of the different isotopes of those elements. Here, we provide a brief, fossil-group by fossil-group overview of some of these chemical proxies, and what they can tell us about past oceanic conditions and longterm climate change.
Foraminifera The science of palaeoceanography can be traced back to the publication of oxygen isotope analyses of Foraminifera from deep-sea cores by Cesare Emiliani in 1955 (Fig. 4; Emiliani 1955). Emiliani’s study demonstrated that the d18O value of foraminiferal calcite underwent periodic oscillations during the Pleistocene. Crucially, a few years previously, Harold Urey had calculated that ‘accurate determinations of the 18O content of carbonate rocks could be used to determine the temperature at which they were formed’ (Urey 1947); Emiliani was therefore able to show that sea surface temperatures had varied by c. 6 8C during the Pleistocene and, furthermore, that ‘good correlation exists between times of temperature minima . . . and times of insolation minima in high northern latitudes’. Emiliani’s paper demolished the notion that the marine environment is constant over periods of thousands of years, and provided firm evidence for past climate change. Analysis of the chemical and isotopic composition of foraminiferal tests remains one of the most important methods for obtaining information about past marine environments today; good examples of their utility can be found in McCorkle et al. (2008) and Cage & Austin (2008). Here, we give a brief overview of: (i) d18O, an established foraminiferal proxy; and (ii) foraminiferal Mg/Ca, which is emerging as an important palaeothermometer. A more complete summary of foraminiferal proxies can be found in Wefer et al. (1999) and in Henderson (2002). In the decades since Emiliani’s pioneering study, there have been numerous publications of high resolution records of foraminiferal d18O across the Cenozoic. Figure 5 shows a recent compilation of oxygen isotope data for bottom-dwelling, deep sea foraminifera from over 40 Deep Sea Drilling Program (DSDP) and Ocean Drilling Program (ODP) sites (Zachos et al. 2001). Over the long term, these data reveal a
10
R. H. JAMES & W. E. N. AUSTIN
Fig. 4. First records of foraminiferal d18O values and isotopic temperatures obtained from a sediment core from the Caribbean (168360 N, 748480 W, water depth 2965 m) (Emiliani 1955). In accordance with Emiliani’s original paper (Emiliani 1955), d18O values are related to temperature (T ) and the isotopic composition of the seawater in which the foraminiferan grew (A) by: T ¼ 16.5 2 4.3(d18O 2 A) þ 0.14(d18O 2 A)2.
1.8‰ decrease in d18O, from the Mid-Palaeocene (59 Ma) to early Eocene (50 Ma), followed by a 17 My-long trend towards higher d18O. In the absence of large ice sheets (see below) the d18O data chart the transition from ‘greenhouse’ conditions, peaking with the early Eocene Climatic Optimum (EECO) at 52 to 50 Ma, to ‘icehouse’ conditions that culminated in the appearance of the first permanent ice sheets on Antarctica at c. 34 Ma. Since that time, the presence of large ice sheets, whether ephemeral or permanent, has complicated the interpretation of the d18O record. This is because foraminiferal d18O also records changes in seawater d18O, itself an indicator of global ice volume and salinity. Continental ice sheets are depleted in 18O relative to seawater (d18O ¼ 230 to 240‰); melting of these ice sheets decreases seawater d18O by 0.1‰ for every c. 10 m increase in sea-level. Separating the relative contributions of temperature and seawater d18O to foraminiferal d18O remains a challenge, but measurements of foraminiferal Mg/Ca may be
useful (e.g. Lear et al. 2000) and are discussed below. Thus, the rapid .1‰ step in d18O at c. 34 Ma is thought to reflect both global cooling and a rapid expansion of Antarctic ice sheets (Zachos et al. 1994; Lear et al. 2000). These ice sheets are believed to have persisted until the latter part of the Oligocene (26 to 27 Ma), when a warming trend (represented by rising d18O) reduced the extent of Antarctic ice. This warm phase peaked in the late middle Miocene (17 to 15 Ma), and was followed by gradual cooling and reestablishment of a major ice sheet on Antarctica. Since c. 3.2 Ma, d18O has declined more rapidly, reflecting the build-up of northern hemisphere ice sheets. Figure 5 also reveals considerable variability in d18O on short (103 to 106 years) timescales. For intervals characterized by glaciation, much of this variability is concentrated in the Milankovitch bands. For example, over the last 800 to 900 ky d18O variance in the 100-ky (eccentricity) frequency band has been exceptionally pronounced, but is weaker through the early Pleistocene and
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
11
Fig. 5. Global deep-sea oxygen isotope record after Zachos et al. (2001). Curve represents the smoothed five-point running mean of data compiled from more than 40 DSDP and ODP sites. Vertical bars depict a rough estimation of ice volume in each hemisphere relative to the LGM. Dashed bar denotes periods of minimal ice coverage (50%); full bar represents close to maximum ice coverage (.50% of present).
Pliocene when the signal was dominated by variance in the 41-ky (obliquity) band (e.g. Ruddiman et al. 1986). Crucially, a small component of the short-term variability are the aberrations in d18O; loosely defined as brief (c. 103 to 105 yr) anomalies that stand out well above ‘normal’ background variability in terms of rate and/or amplitude. The
most prominent of these is the Late Palaeocene Thermal Maximum (LPTM) which is defined by a .1‰ negative d18O excursion that occurred in less than 10 ky near the Palaeocene/Eocene boundary. The excursion corresponds to a 5 8 to 6 8C rise in deep-sea temperature, which correlates with an abrupt negative carbon isotope excursion as well
12
R. H. JAMES & W. E. N. AUSTIN
as widespread dissolution of seafloor carbonate and mass extinction of benthic foraminifera (e.g. Zachos et al. 2005). Considered together, these observations implicate a rise in greenhouse gas concentrations, possibly from the dissociation and subsequent oxidation of methane from marine clathrates (Dickens et al. 1995). Other oxygen isotope anomalies can be seen at the Oligocene/ Miocene boundary (23 Ma) and in the middle Miocene (14 Ma); their random distribution suggests that they can arise through a number of different mechanisms, not only catastrophic methane release. Temperature dependence of foraminiferal Mg/ Ca ratios was first reported back in the 1950s (Chave 1954; Blackmon & Todd 1959). In recent years there have been various attempts to calibrate this, including culture, sediment trap and core-top approaches that have given consistent results for both planktonic and benthic foraminifera (Fig. 6); thanks to careful studies such as these, foraminiferal Mg/Ca is now emerging as an important temperature proxy. There are a number of advantages of foraminiferal Mg/Ca thermometry over other palaeotemperature proxies (Barker et al. 2005). Firstly, as discussed above, foraminiferal d18O is additionally controlled by the oxygen isotope composition of
seawater; separating the temperature component can be difficult. In most cases, Mg/Ca responds exclusively to temperature. Moreover, new laser ablation techniques permit analyses of d18O and Mg/Ca on the same foraminifera test, allowing both temperature and seawater d18O to be determined. Secondly, the oceanic residence times for Ca and Mg are relatively long (106 and 107 years, respectively) so the Mg/Ca ratio of seawater can (unlike seawater d18O) be considered to be constant over glacial/interglacial timescales. Thirdly, since temperature estimates based on Mg/Ca ratios are specific to the species employed they may be used to reconstruct temperatures from different depths in the water column depending on the species’ habitat preferences. Finally, measurement of Mg/Ca ratios is quite straightforward so high resolution records may be obtained in a relatively short time. There are now a number of reports that demonstrate the power of paired foraminiferal Mg/Ca and stable isotope records (e.g. Elderfield & Ganssen 2000; Lear et al. 2000; Lear et al. 2004; Medina-Elizalde & Lea 2005). Figure 7 shows the benthic foraminiferal Mg/Ca record associated with two d18O excursions at the Oligocene/ Miocene boundary (23 Ma) (Lear et al. 2004). The data indicate that bottom water temperatures decreased by c. 2 8C over the 150 ka immediately
Fig. 6. Relationship between Mg/Ca and calcification temperature determined for two species of planktonic foraminifera (Globigerina bulloides and Globigerinoides sacculifer) recovered from core-tops, culture experiments and sediment traps. Data are from Elderfield & Ganssen (2000), Anand et al. (2003), Lea et al. (1999) and Nu¨rnberg et al. (1996).
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
13
Fig. 7. Records of (a) benthic foraminiferal d18O and (b) benthic foraminiferal Mg/Ca across the Oligocene/Miocene boundary. Note the abrupt positive excursion in the benthic foraminiferal d18O record; the excursion consists of a total increase of around 1‰ occurring in two phases, reaching maximum values of 2.2‰ at 23 Ma; (c) Mg temperatures calculated from (b) using the calibration equation Mg/Ca ¼ 1.008 exp 0.114 T (Lear et al. 2002). The vertical shaded bars represent intervals of cooling inferred from the benthic foraminiferal Mg/Ca record.
prior to both d18O excursions. However, the Mg/Ca data also reveal that the excursions themselves, which are both positive and are consistent with transient expansion of the Antarctic ice sheet, are synchronous with the onset of a c. 2 8C warming over c. 150 ka. It is suggested that the warming during these glacial expansions reflects increased
greenhouse forcing prompted by a sudden decrease in global chemical weathering rates as Antarctic basement silicate rocks became blanketed by an ice sheet. As is the case for most proxies, there are a number of issues that potentially affect the application of foraminiferal Mg/Ca palaeothermometry.
14
R. H. JAMES & W. E. N. AUSTIN
As shown in Figure 6, small differences exist in Mg/Ca-temperature calibrations between species, reducing its accuracy for extinct species. Differences in Mg/Ca can also exist between different size fractions of the same species (Anand et al. 2003). Moreover, laser ablation ICP-MS analyses indicate that substantial variations in Mg/Ca can occur within individual foram tests (e.g. Hathorne et al. 2003). This is partly because many species migrate vertically throughout their life cycle (and thus calcify at a variety of water depth/ temperature), and partly because many species secrete a secondary crust of calcite at the time of gametogenesis that may have a Mg/Ca ratio distinct from other regions of the test as a result of differing biophysiological controls on Mg uptake (e.g. Nu¨rnberg et al. 1996). Other issues relate to the modification of foraminiferal Mg/Ca during sample preparation (e.g. reductive cleaning techniques can reduce Mg/Ca; Barker et al. 2003) or during post-depositional dissolution under the influence of undersaturated bottom waters or pore waters (e.g. Rosenthal et al. 2000). Finally, it is now clear that carbonate ion saturation state affects Mg incorporation into foraminiferal calcite at temperatures below c. 3 8C (Martin et al. 2002; Elderfield et al. 2006). This can significantly affect the application of Mg/Ca palaeothermometry in cold deep waters; for example, Elderfield et al. (2006) report that the Mg/Ca ratio of benthic foraminifera from the core top is higher than it is in benthic foraminifera from the Last Glacial Maximum (LGM) in a core from the North Atlantic.
It is, of course, impossible that the LGM Mg/Ca value can be a measure only of temperature. Rather, these authors suggest that the LGM Mg/Ca ratio implies higher glacial carbonate ion saturation, by about 50 mmol/kg. Interpreting foraminiferal compositions requires an understanding of the incorporation of an element or isotope into the calcium carbonate shell. Measurements of the element partition coefficients in foraminiferal calcite relative to seawater have made it abundantly clear that partition coefficients for inorganic equilibrium precipitation of calcite are inappropriate for foraminiferal calcite (Morse & Bender 1990). This is also apparent in that foraminifera that share the same habitat display significant differences in their trace element concentrations (Fig. 8; Blackmon & Todd 1959), there is intrashell variability in trace element concentrations (e.g. Hathorne et al. 2003), and in that foraminiferal shells have complex shape, structure, texture and crystallography (e.g. Towe & Cifelli 1967). Nevertheless, the mechanism of calcification in Foraminifera remains poorly understood. Recent observations of perforate Foraminifera, which dominate the oceans today, suggest that biomineralization occurs in large intracellular vacuoles that are partially isolated from seawater (Erez 2003). Because there is potential for transportation of ions into or out of the vacuole (by active pumping, diffusion and/or leaking), the chemical composition of the calcification fluid may be different from that of seawater (Elderfield et al. 1996; Chang et al. 2004; Bentov & Erez 2006). Processes
Fig. 8. Variation in Mg/Ca ratio of different Foraminifera groups for water temperatures of between approximately 268 and 28 8C. Data are from Blackmon & Todd (1959). Also shown, on the right, is the range of values determined for inorganic precipitation of calcite from seawater at c. 25 8C (Mucci 1987).
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
15
of foraminiferal biomineralization are explored further by Williams (2008).
Coccolithophores Coccolithophores produce long-chain alkenones that have proved particularly useful for palaeotemperature reconstruction; they can be found in almost all marine sediments from the present day back to at least 140 million years, and are remarkably unaffected during sedimentation and burial. It turns out that in order to maintain the flexibility of their cell membranes, coccolithophores alter the number of double bonds in the alkenone chain in response to temperature. Laboratory experiments using cultures of Emiliani huxleyi have shown that the relative abundance of alkenones containing 37 carbon atoms and two double bonds (C37:2) is higher in cooler waters, while the relative abundance of alkenones containing 37 carbon atoms and three double bonds (C37:3) is higher in warmer waters (Prahl & Wakeham 1987). The alkenone unsaturation index, U37 K , is thus defined as: U37 K ¼ [C37:2 ]=([C37:2 ] þ [C37:3 ]) and is related to temperature (T) as follows (Prahl et al. 1988): U37 K ¼ 0:034 T þ 0:039
Coccolith-forming algae tend to bloom in the spring when light intensity and nutrient supply are at an optimum. This dominates the supply of alkenones to the sediments, so U37 K tends to be a measure of seasonal temperature, rather than average annual temperature. Figure 9 compares records of sea surface temperature derived from the U37 K index and planktonic foraminiferal assemblages for a core recovered from the eastern North Atlantic (Chapman et al. 1996). Records derived from the U37 K index are generally in good agreement with those derived from warm (August) foraminiferal assemblages between 28 000 and 8000 years ago, which suggests that maximum coccolithophore production occurred in the summer months in the glacial ocean. The relationship between these two proxies breaks down for sediments deposited during the last 8000 years. This may be linked to a switch in the seasonal timing of the coccolithophore bloom- from mid-summer in the glacial ocean, to late spring in the modern ocean. While the alkenone paleothermometer has been remarkably successful, there are few well-developed coccolithophore proxies based on carbonate chemistry. This is largely because of the difficulty of separating different species: coccolithophores
Fig. 9. Comparison of planktonic foraminiferal SST estimates for August (open circles, right) and February (closed circles, left), with the U37 K (black squares) palaeotemperature index, for a site in the eastern North Atlantic (Chapman et al. 1996).
grown in culture exhibit species-specific disequilibrium stable isotope effects (e.g. Dudley et al. 1980) that lead to ambiguities in interpreting downcore records (e.g. Paull & Thierstein 1990). However, important advances have been made recently through the development of nannofossil separation techniques (e.g. Minoletti et al. 2001; Stoll & Ziveri 2002) that may well lead to an explosion in the use of nannofossil geochemical proxies. To date, the coccolithophore carbonate proxy that shows most promise is Sr/Ca. Field studies of sediments (e.g. Stoll & Schrag 2000) and culture experiments (Rickaby et al. 2002; Stoll et al. 2002) indicate that coccolith Sr/Ca correlates with rates of both organic carbon fixation and calcification. This is important, because reconstruction of the history of coccolithophore productivity is crucial for defining past levels of atmospheric carbon dioxide (Westbroek et al. 1993). In a recent study, Rickaby et al. (2007) report large changes in coccolithophore Sr/Ca over the past
16
R. H. JAMES & W. E. N. AUSTIN
1 Ma that appear to be driven by changing production of bloom species with unusually high Sr/Ca ratios. Periods of high Sr/Ca and high bloom production are also associated with high global coccolithophore production, and correlate inversely with the low amplitude 100 ka and higher amplitude 400 ka eccentricity orbital frequency. The reason for the link between production of coccolithophore blooms and eccentricity is not certain, but it may be related to eccentricity regulation of the length of the growing season and light intensity (Rickaby et al. 2007). The seawater-calcite partition coefficient for strontium in coccolithophore calcite ranges from 0.1 to 0.6, depending on calcification rate (Rickaby et al. 2002; Stoll et al. 2002), which is an order of magnitude greater than the seawatercalcite partition coefficient for inorganic calcite (0.021 to 0.14, depending on precipitation rate (Lorens 1981; Tesoriero & Pankow 1996)). This discrepancy indicates a strong biological control on Sr uptake into coccolith calcite. Precipitation of calcite takes place within an intracellular vesicle that is completely isolated from the cytosol (Young et al. 1999), so seawater is separated from the site of calcification by at least two membranes. The strong link between coccolith Sr/Ca and rates of organic carbon fixation and calcification suggests that the Sr/Ca transported to the vesicle increases with increased rates of ion pumping across these membranes (e.g. Rickaby et al. 2002). Coccolith calcification is further discussed by Williams (2008).
Diatoms While the overwhelming majority of palaeoceanographic records are based on analysis of the shells of calcareous plankton, these are often incomplete because of dissolution: c. 80% of the ocean floor lies below the carbonate compensation depth. In such environments (which include much of the Southern Ocean) it is imperative to seek alternative sources of palaeodata, and in recent times efforts have been increasingly focused on developing geochemical proxies based on diatom opal. These proxies include: Ge/Si, which may provide information about the weathering intensity of continental silicates (e.g. Froelich et al. 1992); Zn, which is thought to record changes in oceanic free Zn2þ concentrations (e.g. Ellwood & Hunter 2000); as well as a number of stable isotopes, some of which are explored further below. Earliest measurements of the oxygen isotope composition of fossilized diatom silica suffered from problems with analytical reproducibility (e.g. Labeyrie 1979) because biogenic opal contains significant amounts of loosely-bound water, up to c. 12
weight %, that has a different 18O/16O ratio from the silica (Si-O lattice) oxygen. This problem seems to have been largely overcome with the introduction of new analytical techniques such as stepwise fluorination (for a summary of these techniques see Schmidt et al. (1997)) and a number of studies have utilised diatom d18O to produce records of seawater d18O and SST (e.g. Labeyrie 1974; Shemesh et al. 1992). Nevertheless, substantial uncertainty still exists in the temperature relationship of the oxygen isotope fractionation between biogenic silica and seawater. Schmidt et al. (1997) found that the d18O values of living diatom samples were 3 to 10‰ lower than expected from the diatom silica– water calibration based on fossil diatoms. The reason for this discrepancy is not known, but it could be related to internal structural and compositional changes in the silica arising from the in situ condensation of Si-OH groups to Si-O-Si bonds as opal-A ripens (Schmidt et al. 2001). The utility of diatom d30Si and d15N as proxies for nutrient utilization in the past has recently been reviewed by De La Rocha (2006). Briefly, diatoms discriminate against 15N as they take up and utilize nitrate (e.g. Altabet & Franc¸ois 2001), and they also discriminate against 30 Si during silicic acid uptake (e.g. De La Rocha et al. 1997). This fractionation results in an increase of d30Si and d15N of silicic acid and nitrate in the euphotic zone as phytoplankton draw the nutrient concentrations down during a bloom or other period of growth; this rise in nutrient d30Si and d15N with nutrient depletion is recorded in the opal produced by growing diatoms. High values of d30Si and d15N in sedimentary diatoms are thus taken to indicate that a greater fraction of available nutrients has been consumed over the course of a growing season. Studies of diatom d30Si and d15N from the opal belt south of the present-day Antarctic Polar Front (APF) produce a pattern of high interglacial d30Si and low interglacial d30Si (De La Rocha et al. 1998; Brzezinski et al. 2002), but high glacial d15N and low interglacial d15N (Crosta & Shemesh 2002). Thus, in the Southern Ocean, it seems that Si and N consumption are decoupled: high levels of nitrate consumption by the whole phytoplankton community appear to coincide with a minimum in silicic acid utilization by diatoms. Although at first glance this seems counterintuitive, there are a number of possible explanations. Firstly, Si and N cycling could be decoupled over glacial-interglacial timescales if the bulk of nitrate uptake during glacials is accomplished by non-siliceous phytoplankton, rather than by diatoms (De La Rocha et al. 1997). Alternatively, increased Fe availability in
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
the Southern Ocean during glacials could increase diatom utilization of nitrate while decreasing diatom uptake of silicic acid (Brzezinski et al. 2002). Finally, it must be remembered that there are a number of unresolved problems with both proxies (De La Rocha 2006). Diatom d30Si and d15N could also be affected by changes in the species composition of the sediments, there are a variety of artefacts associated with cleaning and analytical procedures, and there is concern that the d30Si and d15N values of seawater could vary on glacial-interglacial timescales.
Corals Corals are particularly useful paleoclimate recorders because they contain a broad array of geochemical tracers within their skeletons and they have annual growth bands (Knutson et al. 1972) that can be accurately dated using radiometric techniques (e.g. Bard et al. 1990). Moreover, corals provide enhanced time resolution (biweekly to annual) because of their high growth rate (from c. 0.2 to c. 20 mm/yr; e.g. Adkins et al. 2004), so coralline records can be integrated with other high resolution palaeoclimate data derived from tree rings, ice cores and varved sediments. Finally, they are unaffected by the mixing processes that compromise records derived from oxic marine sediments (bioturbation). Tropical corals. Massive reef corals represent a rich archive of tropical palaeoclimates. Their aragonitic skeleton contains various geochemical tracers that can be potentially linked to environmental parameters. These include Sr/Ca, d18O, d44Ca, U/Ca and Mg/Ca, all of which can be linked to sea surface temperature, 14C, which has
17
been used to determine the decadal variability of ocean circulation (e.g. Druffel & Griffin 1993), and Ba/Ca, which has been linked to upwelling activity (e.g. Ourbak et al. 2006). Recent studies have also explored the possibility of using d18O and U/Ca as tracers of sea surface salinity (e.g. Gagan et al. 2000; Ourback et al. 2006). Two coralbased proxies, Sr/Ca and U/Ca, are discussed in more detail below. Early work on Sr/Ca in coral skeletons failed to recognise any temperature effect (e.g. Bowen 1956). However, improvements in analytical instrumentation have dramatically increased the signal-to-noise ratio for this proxy (e.g. Beck et al. 1992) and there is now abundant evidence that coralline Sr/Ca can be a high-fidelity recorder of sea surface temperature. For example, Figure 10 shows a high-resolution record of the variation in Sr/Ca measured parallel to the growth axis of a Porites lobata coral head from offshore of Rabaul, East New Britain, Papua New Guinea, together with measured values for SST (Quinn et al. 2006). There is a strong correspondence between positive excursions in coral Sr/Ca and cool excursions in sea surface temperature, and the timing of these excursions is entirely consistent with the timing of El Nino events. The records suggest that the climate of Rabaul is thus significantly impacted by El NinoSouthern Oscillation (ENSO) dynamics: ENSO warm phase (El Nino) events are characterized by conditions that are cooler (and also drier) than normal. Although there are more than thirty published Sr/Ca-temperature calibrations for the genus Porites (see Corre`ge 2006), no ‘universal’ calibration exists and it seems likely that temperature may not be the only control on coralline Sr/Ca
Fig. 10. Variations in (a) sea-surface temperature in the Rabaul region and (b) Sr/Ca ratio of a Rabaul coral. Shaded vertical bars denote ENSO warm phase (El Nino) events. Data from Rayner et al. (2003) and Quinn et al. (2006).
18
R. H. JAMES & W. E. N. AUSTIN
ratios. Ion microprobe analyses (see Cuif et al. 2008) have revealed small scale (sub-micron) heterogeneity in coralline Sr/Ca that must be related to biological and/or kinetic effects, rather than temperature (e.g. Allison & Finch 2004; Meibom et al. 2007). Nevertheless, work by Allison & Finch (2004) suggests that these small-scale heterogeneities become dampened and the effect of SST becomes progressively more important as Sr/Ca data are integrated over longer timescales. A further complication is that the concentration of both Sr2þ and Ca2þ in seawater affects the Sr/Ca ratio measured in corals. It is therefore a requirement for Sr/Ca palaeothermometry that the Sr/Ca ratio of seawater is constant over time and space. This is probably not the case: seawater Sr/Ca exhibits a range of 8.51 to 8.55 mmol/mol (e.g. DeVilliers et al. 1994) which could account for about 0.6 8C of the variation in coralline Sr/Ca-temperature calibrations. Over glacialinterglacial cycles, this effect could be much greater (up to 2 8C) because the erosion of exposed carbonate shelves potentially alters seawater Sr/Ca (Stoll & Schrag 1998). Finally, Sr/Ca ratios of fossilised corals can also be affected by diagenesis. Precipitation of secondary aragonite can significantly increase coralline Sr/Ca leading to erroneously low estimates of past SST (Allison et al. 2005a), while infilling of skeletal voids with calcite and transformation of coral aragonite to calcite (usually in the vadose zone) lowers Sr/Ca resulting in apparent SST anomalies as high as 115 8C (McGregor & Gagan 2003). There is a tremendous amount of data on uranium concentrations of corals (e.g. Swart & Hubbard 1982) because it is required for 230Th dating studies. However, 230Th dating requires relatively large samples (.100 mg) so information on short-term (weekly to seasonal) variations in U/Ca is generally lost. The first measurements of coralline U/Ca at around monthly temporal resolution were published by Min et al. in 1995. This study revealed seasonal changes in U/Ca that anticorrelated strongly with temperature. Crucially, these authors found that the fractional change in U/Ca per degree Celsius was c. 5%, about 6 times the fractional change exhibited in Sr/Ca per degree Celsius. More recent studies, however, suggest that the relationship between U/Ca and temperature is more complex than originally thought, and U/Ca may also be affected by other parameters such as sea surface salinity (Ourback et al. 2006), growth rate or carbonate ion concentration (Shen & Dunbar 1995). Deep water corals. Although reef corals have provided a wealth of palaeoceanographic information, this is restricted to shallow tropical waters. On the other hand, non-reef building, non-
photosynthetic corals have widespread distribution (from surface to abyssal depths, and from polar to equatorial latitudes), but their utility as recorders of deep ocean variability has not been extensively studied, mainly because stable isotope (d18O, d13C) and trace element analyses suggest that their skeletons are precipitated far out of equilibrium with seawater (e.g. Smith et al. 1997; Cohen et al. 2006). Nevertheless, a number of promising deep water coral-based proxies have recently emerged in the literature. These include 14 C, which provides information as to deep water ventilation (Adkins et al. 1998), phosphorus, which can be used to reconstruct changes in the nutrient status of the oceans (Montagna et al. 2006), and Nd isotopes, which are explored in more detail below. Measurements of the Nd isotopic composition of D. dianthus corals from the Drake Passage and the southwest Pacific indicate that deep-sea corals record the Nd isotopic composition of surrounding seawater (van de Flierdt et al. 2006). This is important, because the Nd isotopic composition of seawater provides information on deep water circulation patterns. The residence time of Nd (600 to 2000 years; e.g. Jeandel et al. 1995) is similar to the ocean mixing time and different deep water masses possess distinct Nd isotope compositions that reflect local input from the surrounding continental land masses, and subsequent dispersal by ocean circulation. Accordingly, North Atlantic Deep Water (NADW) has the least radiogenic isotope composition (1Nd ¼ –13.5 + 0.5 (Piepgras & Wasserburg 1987); where 1Nd is the measured 143Nd/144Nd ratio relative to that of the chondritic uniform reservoir) reflecting erosion of old continental material into the Labrador Sea area while deep water in the Pacific Ocean possesses the most radiogenic isotope composition (1Nd ¼ þ3.5 + 2.0; summarised in van der Flierdt et al. 2004) due to weathering of young mantlederived material with high 1Nd values (þ20) surrounding this ocean. Because Nd isotopes are not known to be affected by biological activity, or physical processes such as temperature and salinity, 1Nd values of ancient deep-water corals should therefore reflect past changes in ocean circulation. To test this idea, van de Flierdt et al. (2006) measured the 1Nd value of 10 deep-sea corals from the New England seamounts in the northwest Atlantic ranging in age from modern to 91 ka ago. They found that the Nd isotopic composition of North Atlantic deep and intermediate water has remained nearly constant through the last glacial cycle. The constancy of the 1Nd value of NADW on glacial-interglacial timescales considerably simplifies interpretation of Nd isotopic variability through time in other locations as a tracer of water mass mixing and ocean circulation.
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
Molluscs Because of their widespread distribution (from the tropics to high latitudes, from the coastal zone to the deep ocean), incremental shell deposition (enabling sampling at sub-annual resolution; Fig. 11) and good preservation, marine molluscs are potentially important archives of environmental information. In contrast to corals, they appear to precipitate their shell close to isotopic equilibrium with seawater and so mollusc d18O can provide information about past variations in water temperature and water d18O (e.g. Elliot et al. 2003). The utility of other geochemical mollusc proxies is much less certain. Mollusc d13C appears to reflect a number of environmental variables, ranging from seawater d13C to salinity, as well as physiological factors such as growth rate (e.g. Elliot et al. 2003). There appears to be no direct relationship between temperature and Sr/Ca ratios in mollusc shells; rather, mollusc Sr/Ca seems to vary with growth rate and/or metabolism (e.g. Gillikin et al. 2005). This relationship is, however, not simple and is different for different species. Gillikin et al. (2005) found a strong relationship between average annual Sr/Ca ratios and annual growth rate in one clam species (Saxidomus giganteus) but not in another (Mercenaria mercenaria), while Klein et al. (1996a) concluded that while calcite secreted along the lateral margins of the mussel Mytilus trossulus was strongly influence by metabolic activity, calcite secreted at the central margin of the shell was in near-equilibrium with seawater. Meanwhile, a temperature dependence of Mg/Ca ratios has been reported for the mussel Mytilus trossulus (Klein et al. 1996b) but not for
19
the scallop Pecten maximus (Lorrain et al. 2005). Moreover, Freitas et al. (2006) report four-fold variations in shell Mg/Ca between different bivalve species for the same temperature range, and conclude that mollusc Mg/Ca is an unreliable temperature proxy. Work on the calibration and validation of mollusc geochemical proxies is ongoing (for example, it has recently been shown that the d44Ca value of fossil rudist calcite may reflect past changes in temperature; Immenhauser et al. 2005) but there is clearly some way to go before their potential can be fully assessed.
Geochemical proxy measurements In this Section, we outline some of the challenges that need to be overcome in order to obtain accurate geochemical proxy records, and we describe some of the analytical tools that are available to us.
Presence of contaminant phases Palaeoceanographic reconstructions based on the chemical analysis of microfossils can be biased by the presence of detrital silicate material and other contaminant phases. Siliceous particles, such as clays, are ubiquitous in marine sediments and are significantly enriched in many trace elements relative to carbonate material. Adhering clays can be removed by crushing the sample and agitating in methanol (e.g. Martin & Lea 2002). However, there is some suggestion that silicate grains can be incorporated into the biogenic matrix (e.g. Vigier et al. 2007); if this proves to be the case, then isolation of this material is likely to be far more tricky. Removal of organic matter and diagenetic Mn-oxide coatings is also required for some fossil groups, notably foraminifera. Respectively, these can be achieved by so-called ‘oxidative’ and ‘reductive’ cleaning of the foram test (e.g. Martin & Lea 2002). Reductive cleaning is known to lower Mg/Ca (Rosenthal et al. 2004) as well as other trace element/calcium ratios (Fig. 12). This may be due either to removal of the contaminant phase or to partial dissolution of the fossil shell itself which occurs as a side-effect of reductive cleaning. Therefore, for proper comparison of data from different laboratories, standard cleaning procedures must be adopted (e.g. Barker et al. 2005).
Fossil preservation Fig. 11. Direct light image of a Mercenaria mercenaria bivalve shell, showing that the carbonate is precipitated in layers; each dark/light band represents one year of growth. These layers can be sampled at sub-annual resolution using laser ablation techniques, or high precision microdrill apparatus. Image courtesy of Mary Elliot (University of Edinburgh).
Preservation of fossilized shell material is commonly assessed under light microscope or under low power with the scanning electron microscope but it is increasingly apparent that these techniques cannot pick up recrystallization features on the micron scale (see Pearson & Burgess, 2008). This is important because diagenetic alteration of
20
R. H. JAMES & W. E. N. AUSTIN
(a)
(b)
(c)
Fig. 12. Trace element/Ca ratios of foraminifera subject to different cleaning procedures. (a) Mn/Ca, (b) Mg/Ca and (c) Sr/Ca. Samples cleaned using procedure A were subject to ultrasonication in methanol, oxidation and a weak acid leach. Those cleaned using procedure B were subject to additional reductive cleaning, and those cleaned using procedure C were subject to additional reductive cleaning as well as cleaning in DTPA (diethylene triamine pentaacetic acid). DTPA removes refractory phases rich in Ba and the rare earth elements (e.g. Haley & Klinkhammer 2002). Note that the introduction of the reductive cleaning step significantly lowers Mn/Ca and also lowers Mg/Ca. Reductive cleaning has no resolvable effect on Sr/Ca. DTPA has little effect on Mn/Ca, Mg/Ca or Sr/Ca. Error bars (where visible) represent the standard deviation of the mean of two (O. universa) or three (G. conglomerata) separate analyses. Data are from Hathorne (2004).
ancient biogenic material can cause substantial changes to its original chemistry (e.g. Sexton et al. 2006). For example, in a recent study, Pearson et al. (2001) demonstrated that recrystallized foraminiferal material from the Late
Cretaceous and Eocene epochs had significantly higher d18O and d13C than well-preserved foraminifera of approximately the same age (and latitude), because of addition of diagenetic calcite. Crucially, this study showed that the presence of diagenetic
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
calcite decreased apparent sea surface temperatures estimated from foraminiferal d18O by at least 13 8C; correction for this effect indicates that tropical temperatures in the Late Cretaceous and Eocene were at least as warm as today, and probably several degrees warmer, in accordance with the geographical distributions of temperature-sensitive organisms such as corals, mangroves and reptiles. Several studies support the idea that the chemical composition of biogenic carbonate is modified by partial dissolution under the influence of bottom waters or porewaters that are undersaturated with respect to calcium carbonate (e.g. Rosenthal et al. 2000). In a study of planktonic foraminifera conducted by Brown & Elderfield (1996), artificial partial dissolution of G. tumida tests resulted in a decrease in both Mg/Ca and Sr/Ca, but there was no significant change for G. sacculifer. The cause of this reduction is thought to be preferential dissolution of high-Mg inner (chamber) calcite, which forms in warmer waters than the low-Mg calcite crust (keel). In theory, it should be possible to correct for such effects if the extent of dissolution can be quantified. A number of dissolution indicators, including foraminiferal test weights (e.g. Broecker & Clark 2001) and calcite crystallinity (Bassinot et al. 2004), have been tested, but none have been widely applied to date.
Proxy calibration and validation The ideal biogeochemical proxy should correlate with a single environmental parameter, and this correlation should be preserved in perpetuity in the sedimentary record. Sadly, this is rarely the case, and all proxies need to undergo a comprehensive validation exercise to confirm their veracity. A tongue-in-cheek view of the proxy verification process is shown in Figure 13. There are three main approaches to proxy validation, each of which has both strengths and weaknesses. The first is culturing, where an organism is grown in the laboratory under controlled conditions. This allows us to distinguish the effects of the different physical (e.g. temperature, light) and chemical environmental parameters that can vary in the ocean. Culturing is not often straightforward, but controlled growth of corals (e.g. Reynaud-Vaganay et al. 1999), foraminifera (e.g. Lea et al. 1999) and coccolithophores (e.g. Rickaby et al. 2002) has contributed crucial information for validating a number of proxy relationships. The drawbacks with culturing studies are that the controlled environment may not reproduce natural growth fully and that only a limited number of species can be studied. A second way to establish a calibration is based on sampling organisms in
21
Fig. 13. ‘Palaeoceanographic Proxy Confidence Factor Phase Chart’ (PPCFPC), as proposed by Elderfield (2002). The chart shows the relationship between time since the proxy was proposed, and confidence in the proxy. Initially, confidence is high. As the proxy relationship is explored in more detail, problems start to emerge and proxy confidence falls. At some point in time, many of these issues are ironed out, and proxy confidence can be reliably assessed. It is generally never as great as first thought!
the water column at sites where the chemical and physical characteristics of the water are monitored. For example, corals can be stained with dye in their ambient environment and then left to grow while temperature is continually monitored (e.g. Smith et al. 1979), while planktonic organisms can be collected either by net (with the advantage of regional coverage) or by sediment trap (with the advantage of seasonal and interannual records) at the same time as temperature, salinity and water chemistry data (e.g. Anand et al. 2003). The disadvantage with water column studies is that, for non-sessile organisms, they do not reflect the material that forms the sedimentary record. For example, Erez & Honjo (1981) demonstrated that the d18O value of planktonic foraminifera is systematically greater in material collected in tows that in material collected in sediment traps (located below the depth of the photic zone) and sediments. This is because, on average, 45% of the total weight of the foram test is deposited below the depth of the photic zone, where water temperature is lower. A third approach involves analysis of modern core-top material. This has the great advantage in that it does reflect the material that forms the sedimentary record but the disadvantage of potential artefacts, specifically from dissolution (see above), and of difficulties in establishing links to specific environmental variables. Proxy calibrations are often imperfect, because several more or less independent variables will determine the behaviour of the proxy. The best way to overcome this is to estimate a variable
22
R. H. JAMES & W. E. N. AUSTIN
using more than one proxy: the ‘multiproxy’ approach. Nevertheless, there are two outstanding problems with all proxies, both of which are intractable. The first is that we must make the critical assumption that the proxy calibration established today has applied in the past. The second is that there is no way to calibrate proxy relationships for species that are extinct.
Analysis of major, minor and trace elements The first analyses of the chemical elements in biogenic material were made using a combination of different methodologies, usually atomic absorption or emission spectrophotometry, because of the large differences in the abundance of chemical constituents and the limited linear range of these methods. For example, determination of foraminiferal Cd/Ca was typically accomplished by measuring Cd by graphite furnace atomic absorption spectrophotometry (GFAAS) and Ca by flame AAS (Boyle & Keigwin 1985). The precision of spectrophotometric methods was generally 2–4%. More precise determination (as low as ,0.1%) can be achieved using isotope dilution thermal ionisation mass spectrometry (ID-TIMS) (e.g. Rickaby et al. 2000) but this technique is extremely time consuming. All of these methods are limited to a single-element analysis. Today, analyses of bulk fossil chemistry are largely conducted by inductively coupled plasma (ICP) atomic emission spectrometry (AES), ICP mass spectrometry (MS) or ICP optical emission spectrometry (OES) techniques (e.g. Rosenthal et al. 1999; DeVilliers et al. 2002; Green et al. 2003). These techniques permit rapid and precise (c. 1% for many elements) measurement of a number of chemical constituents simultaneously. ICP-MS offers higher sensitivities than AES and OES, enabling measurement of more elements and smaller sample sizes. As palaeoceanographic reconstructions based on bulk fossil chemistry proliferate, comparability of measurements between laboratories has become an important issue. A recent calibration exercise involving thirteen laboratories by Rosenthal et al. (2004) evaluated the reproducibility of analyses of synthetic standard solutions within and between laboratories. The study found that intralaboratory instrumental precisions were generally better than 0.5% for both Mg/Ca and Sr/Ca measurements, but interlaboratory precisions (r.s.d) were significantly worse (up to 3.4% and 1.8%, respectively). This could be a result of differences between calibration standards used by laboratories, or because the circulated standard solutions had become contaminated in the interval between their
preparation and analyses, or a combination of both. A suitable solid standard for interlaboratory calibration would help to overcome some of these problems (Greaves et al. 2005). While measurements of bulk fossil chemistry are crucial for acquiring palaeoceanographic records, information as to biomineralization processes and the effects of recrystallization require assessment of the variability of chemical constituents within a fossil. Measurements of the intrafossil variability of Mg and Sr have been achieved by electron microprobe (e.g. Brown & Elderfield 1996), but this technique is only suitable for a limited range of elements. Secondary ion mass spectrometry (SIMS), or ion microprobe, has been used to provide both high spatial resolution (c. 10 mm) and precise analyses of a wider range of elements (B, F, Mg, Sr, Rb, Ba and U) in corals (Allison 1996; Hart & Cohen 1996) but, to date, only Mg, Sr and Li data have been reported for foraminiferal calcite (Allison & Austin 2003; Bice et al. 2005; Vigier et al. 2007). Laser ablation ICP-MS is increasingly the method of choice for in-situ analyses of biogenic material, because the instrumentation is relatively cheap, and a large number of elements can be measured simultaneously at low levels (e.g. Hathorne et al. 2003). This technique has been applied, for example, to remove surface contaminants (Fig. 14), to determine the daily variation in the Mo and Ba concentration in a scallop shell (Barats et al. 2007) and to determine the effect of the day-night, photosynthesis-respiration cycle of algal symbionts on foraminiferal Mg/Ca (Eggins et al. 2004). The main problem with in-situ chemical analysis of biogenic material is the absence of suitable solid certified standard reference materials. Fractionation of the chemical elements can occur during sample ablation (by laser or electron beam), during transport to the mass spectrometer and, in the case of LA-ICP-MS, in the plasma (e.g. Kuhn & Gu¨nther 2004). Fractionation is matrix-dependent, so standardization of biogenic material to, for example, a glass reference material (such as NIST 612) may not be appropriate (Fig. 15).
Isotope analysis Simultaneous measurements of the abundance of two (or more) isotopes of the same element in biogenic materials are usually obtained by mass spectrometry. This technique requires ionisation of atoms from the sample; ions can be either positively or negatively charged, and they can be produced in a number of different ways: 1. Gas bombardment. Gaseous samples (hydrogen, carbon and oxygen (CO2), sulphur
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
23
Fig. 14. Change in trace element/Ca ratios through the wall of the final ‘kummerform’ chamber of a G. sacculifer test. Note that the surface of the test is coated in a Mn-rich contaminant phase that is also enriched in most other trace elements. At depths of .1.5 mm, element/Ca ratios are relatively constant, suggesting that only calcite-bound trace elements are present. The Mn-rich surface veneer can thus be removed by pre-ablation of the test. After Hathorne et al. (2003).
2.
Fig. 15 Mg/Ca ratio of a calcite grain measured by LA-ICP-MS. Data are shown standardized both to a pressed calcite pellet (Hathorne et al. 2003) (circles) and to the certified NIST 612 glass reference material (Pearce 1997) (squares). Error bars show the standard deviation (1s) of c. 70 0.9 s integrations. The shaded horizontal bar represents encompasses the mean and standard deviation (1s; n ¼ 18) of the Mg/Ca ratio obtained by solution ICP-MS analysis of the same sample. Note that there is a systematic offset between the solution ICP-MS data and the LA-ICP-MS data standardized to NIST 612; this is presumably related to matrix-induced fraction. Data are from Hathorne et al. (2008).
3.
(SO2), and also the inert gases) are sprayed with electrons, stripping a peripheral electron from the gas molecule. In palaeoceanography, gas source mass spectrometry is commonly used to measure carbon and oxygen isotope ratios in biogenic carbonate; samples are dissolved in acid (usually phosphoric), producing CO2 gas. Recent technical advances mean that the sample dissolution procedure can be automated (for example, the ‘Kiel device’ runs a carousel containing multiple (typically c. 50) reaction vessels; Ostermann & Curry 2000), which allows for a high sample throughput and good external long-term precision (0.05– 0.06‰ for d13C and 0.065–0.075‰ for d18O; Spo¨tl & Vennemann 2003), even for small samples (,50 mg). Thermal ionization. Solid samples are heated to a high temperature (1000–1800 8C) in a vacuum, producing either positive or negative ions. The TIMS method has been used, for example, to obtain records of seawater 87 Sr/88Sr and 1Nd from biogenic carbonate (e.g. McArthur et al. 2001; Burton & Vance 2000). This technique permits precise isotope ratio measurements (external reproducibility ,5ppm for 143Nd/144Nd and 87Sr/86Sr ratios), but it is restricted to those elements with a relatively low first ionization potential. Inductively-coupled plasma. Sample solutions, or laser ablation products, are ionized in a stream of argon within a plasma torch. The advantage of this technique is that the plasma
24
4.
R. H. JAMES & W. E. N. AUSTIN
torch can ionize all elements, including those too refractory to be analysed by TIMS (such as Mo, Hf and Fe). Unfortunately, a standard (quadrupole) ICP mass spectrometer yields rather poor isotope ratios (at best 2%), as the tops of the mass peaks are not flat. However, in the mid-1990s, the ICP source was combined with a magnetic sector and an array of multiple ion collectors (MC-ICP-MS). This development has vastly improved the precision of isotope ratio measurements from an ICP source (external reproducibility is typically better than ,25 ppm for 143Nd/144Nd and 87 Sr/86Sr ratios), making this method the likely successor to TIMS. Secondary ion sputtering. The surface of a solid sample is sprayed with a primary ion beam (usually negative oxygen ions or positive caesium ions), producing secondary ions (from the sample), which are sputtered into the mass spectrometer. SIMS (ion probe) is particularly useful for obtaining in situ measurements. While SIMS has been widely used to measure element concentrations in biogenic material, there are only a handful of reports on its application to the study of isotope ratios to date (e.g. Ale´on et al. 2001).
Once the atoms from the sample have been ionized, the ion beam is accelerated by a high voltage and then separated in a magnetic field according to their mass. The ion beams are collected in Faraday cups, converted to voltages by a high-value resistor, and these voltages, which are
proportional to isotopic abundances, are analysed by an array of voltmeters. The main challenge in making precise and accurate measurements of isotope ratios is mass bias, or instrumental mass fractionation. In simple terms, the measured isotope ratio is never the same as the true isotope ratio; this is because of variable transmission of the ion beam into the mass spectrometer. There are two main approaches to correcting for this effect. The first is empirical and is based on comparison with isotopic analysis of a known standard (e.g. Dixon et al. 1993). This approach assumes that standards and samples fractionate to the same degree during isotopic analysis, requiring carefully controlled analytical conditions. A second approach is the ‘double spike’ method, which allows for rigorous calculation of instrumental mass fractionation (e.g. Galer 1999) but can only be utilised for elements with three or more isotopes. This approach is required if the precision of an isotope ratio measurement needs to be greater than about 1‰ per mass unit. Accurate and precise measurements of isotope ratios can also be compromised by matrix effects. Some elements have isotopes of the same mass (e.g. 54Cr and 54Fe), so they must be separated from one another with care prior to analysis. Sample matrix can also have non-isobaric effects. These are largely associated with changes in the sensitivity of an analyte due to the presence of other elements. Changes in sensitivity result in a change in instrumental mass bias (Fig. 16); for this reason, it is important to ensure that the sample matrix is the same as that of the standard. Preferably, the analyte should be completely separated
Fig. 16. Matrix effect on Li isotope ratios measured by MC-ICP-MS. ‘Pure’ solution contains 100 ppb Li and 0 ppb Na. With increasing addition of Na,7Li/6Li decreases. Note also that the internal error of the analysis (2s) increases with increasing Na. For accurate and precise measurement of Li isotope ratios, Li must be properly separated from Na. This can be achieved by cation exchange chromatography (e.g. James & Palmer 2000).
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW
from the sample matrix. Finally, plasma sources can produce molecular species, such as 40Ca16O which affects 56Fe, and double-charged species, such as 48 Ca2þ which affects 24Mg. These interferences can be minimized in a variety of ways (see Albare`de & Beard (2004) for a summary), and some of them can now actually be resolved by the new generation high-resolution MC-ICP-MS instruments (e.g. FinniganTM NEPTUNE; Nu Plasma HR; Nu Plasma 1700).
25
average intra-laboratory precision (,1.6 8C). However, the interlaboratory reproducibility of alkenone concentration measurements was far higher than theoretical estimates (c. 32%). There was no clear reason for this (for example, alkenone concentration did not appear to be affected by differences in sample clean-up procedures), but the authors concluded that a reduction in the interlaboratory reproducibility of alkenone concentrations is needed to improve confidence in the U37 K palaeotemperature proxy.
Analysis of organic constituents The principal method used to analyse organic constituents in marine sediments is gas chromatography mass spectrometry (GCMS). First, volatile components are extracted from the sediment using ultra-pure solvents. Solvents are usually selected on the basis of their polarity; solvents most readily dissolve solutes of approximately the same polarity. However, marine sediments contain complex mixtures of organic compounds, so glasscolumn and high-performance liquid chromatography are commonly used to clean and separate them into fractions before they are introduced into the GCMS system. The solvent extract is injected into the gas chromatograph, where it is vaporized and mixed with an inert carrier gas (such as hydrogen or helium). Organic macromolecules that are not readily extracted by solvents, such as cellulose and chitin, may be vaporized by pyrolysis prior to injection. The gaseous mixture is then passed through a capillary column which separates hydrocarbons according to either relative volatility, shape or polarity, depending on the nature of the stationary phase. Eluting compounds pass directly from the column to the mass spectrometer source, where they are ionized to form molecular ions. Molecular ions can also undergo fragmentation or rearrangement to form other ions and these fragment ions may also dissociate further to form other electrically charged molecular or atomic species of successively lower formula weight. The result is a characteristic fragmentation pattern or mass spectrum of the parent molecule. Structures of unknown compounds can be determined by comparison of their mass spectra with library spectra of known compounds. A number of research groups now utilise such techniques for palaeoceanographic studies, so comparability of data between laboratories is crucial. In a recent study, twenty four laboratories were provided with eight test samples and asked to measure U37 K , the alkenone unsaturation index, and alkenone concentrations (Rosell-Mele´ et al. 2001). The interlaboratory difference in U37 K temperature estimates was ,2.1 8C, which is slightly higher than the
Palaeoceanographic proxies: a forward look Recent advances in analytical instrumentation and methodology promise to add many more geochemical proxies to the palaeoceanographer’s toolbox. MC-ICP-MS has opened up a number of isotopic systems previously considered intractable, such as zinc (which shows potential as a tracer of nutrient utilization; Vance et al. 2006) and germanium, whose isotopic composition has recently been measured in biogenic opal (Rouxel et al. 2006). Meanwhile improvements in gas source mass spectrometry techniques now permit precise measurement of the abundance of 13C – 18O bonds in biogenic carbonate. 13C – 18O abundance shows promise as a palaeothermometer which, unlike d18O, is independent of the stable isotope composition of seawater (Ghosh et al. 2006). New technology is playing an important role also in understanding processes of biomineralization. The new generation nanoSIMS ion probe now permits chemical analysis of biogenic material at the sub-micrometer length scales that are characteristic of the skeletal ultrastructure; this has provided new insights into transport pathways of trace elements from seawater into the biogenic skeleton (e.g. Meibom et al. 2007). Development of new UV-femtosecond laser systems, which apparently ablate solid material with no chemical or isotopic fractionation (Horn et al. 2006), will also be useful for in-situ analyses of small-scale variations in trace elements and isotopes in biogenic material. Intra-shell heterogeneity can also now be explored using sequential leaching techniques (Haley & Klinkhammer 2002). Furthermore, the precise location of trace elements in biogenic minerals can now be determined by X-ray Absorption Spectroscopy (XAS) methods; for example, Allison et al. (2005b) report that Sr is ideally substituted in coral aragonite, with no evidence of strontianite or partly ordered structural states. This indicates that the Sr in coral skeletons is substituted randomly for Ca in the aragonite structure, which is a requirement for the coral Sr palaeothermometer.
26
R. H. JAMES & W. E. N. AUSTIN
The recent discovery of cryptic genetic diversity in modern planktonic foraminifera (e.g. Darling et al. 2000) has potentially significant implications for their use as geochemical proxies. There is growing evidence that these cryptic genetic types are ecologically different (e.g. Bauch et al. 2003) so some of the noise associated with proxy reconstructions based on apparently ‘single species’ (i.e. those with similar shell morphology) may in fact be due to the lumping of cryptic genetic types. For example, DNA studies of the rightcoiling variant of N. pachyderma in the Nordic Seas has revealed two distinct genotypes; one that thrives in relatively warm North Atlantic waters, and another that dominates in polar waters (Bauch et al. 2003). The offset in d18O between the two species is c. 0.5‰, which translates to a temperature of c. 28C or 2 salinity units. If the relative proportion of these two species at a particular site has changed over time, this effect needs to be considered for accurate reconstruction of past sea surface temperatures.
References A DKINS , J. F., C HENG , H., B OYLE , E. A., D RUFFEL , E. R. M. & E DWARDS , R. L. 1998. Deep-sea coral evidence for rapid climate change in ventilation of the deep North Atlantic 15 400 years ago. Science, 280, 725– 728. A DKINS , J. F., H ENDERSON , G. M., W ANG , S.-L., O’S HEA , S. & M OKADEM , F. 2004. Growth rates of the deep-sea scleractinia Desmophyllum cristagalli and Enallopsammia rostrata. Earth and Planetary Science Letters, 227, 481– 490. A LBARE` DE , F. & B EARD , B. 2004. Analytical methods for non-traditional isotopes. Reviews in Mineralogy & Geochemistry, 55, 113–152. A LE´ ON , J., C HAUSSIDON , M., C HAMPENOIS , M. & M AGIN , D. 2001. Quantitative imaging of stable isotopes by ion microprobe. Geostandards Newsletter, 25, 417 –429. A LLISON , N. 1996. Comparative determinations of trace and minor elements in coral aragonite by ion microprobe analysis, with preliminary results from Phuket, southern Thailand. Geochimica et Cosmochimica Acta, 60, 3457–3470. A LLISON , N. & A USTIN , W. E. N. 2003. The potential of ion microprobe analysis in detecting geochemical variations across individual foraminifera tests. Geochemistry, Geophysics, Geosystems, 4, 8403, doi:10.1029/2003GC000539. A LLISON , N. & F INCH , A. A. 2004. High-resolution Sr/ Ca records in modern Porites lobata corals: Effects of skeletal extension rate and architecture. Geochemistry, Geophysics, Geosystems, 5, Q05001, doi:10.1029/ 2004 GC000696. A LLISON , N., F INCH , A. A., T UDHOPE , A. W., N EWVILLE , M., S UTTON , S. R. & E LLAM , R. M. 2005a. Reconstruction of deglacial sea surface
temperatures in the tropical Pacific from selective analysis of a fossil coral. Geophysical Research Letters, 32, L17609, doi:10.1029/2005GL023183. A LLISON , N., F INCH , A. A., N EWVILLE , M. & S UTTON , S. R. 2005b. Strontium in coral aragonite: 3. Sr coordination and geochemistry in relation to skeletal architecture. Geochimica et Cosmochimica Acta, 69, 3801– 3811. A LTABET , M. A. & F RANC¸ OIS , R. 2001. Nitrogen isotope biogeochemistry of the Antarctic Polar Frontal Zone at 1708W. Deep Sea Research Part II, 48, 4247– 4273. A NAND , P., E LDERFIELD , H. & C ONTE , M. H. 2003. Calibration of Mg/Ca thermometry in planktonic foraminifera from a sediment trap time series. Paleoceanography 18, 1050, doi:10.1029/2002PA000846. A RMSTRONG , H. A. & B RASIER , M. D. 2005. Microfossils. Blackwell Publishing, Oxford. A USTIN , H. A., A USTIN , W. E. N. & P ATERSON , D. M. 2005. Extracellular cracking and content removal of the benthic diatom Pleurosigma angulatum (Quekett) by the benthic foraminifera Haynesina germanica (Ehrenberg). Marine Micropaleontology, 57, 68– 73. B ARATS , A., P ECHEYRAN , C., A MOUROUX , D., D UBASCOUX , S., C HAUVAUD , L. & D ONARD , O. F. X. 2007. Matrix-matched quantitative analysis of trace-elements in calcium carbonate shells by laser-ablation ICP-MS: Application to the determination of daily scale profiles in scallop shell (Pecten maximus). Analytical and Bioanalytical Chemistry, 387, 1131–1140. B ARD , E., H AMELIN , B., F AIRBANKS , R. G. & Z INDLER , A. 1990. Calibration of the 14C timescale over the past 130 000 years using mass spectrometric U-Th ages from Barbados corals. Nature, 345, 405– 410. B ARKER , S., G REAVES , M. & E LDERFIELD , H. 2003. A study of cleaning procedures used for foraminiferal Mg/Ca paleothermometry. Geochemistry, Geophysics, Geosystems, 8407, doi:10.1029/2003GC000559. B ARKER , S., C ACHO , I., B ENWAY , H. & T ACHIKAWA , K. 2005. Planktonic foraminiferal Mg/Ca as a proxy for past oceanic temperatures: a methodological overview and data compilation for the Last Glacial Maximum. Quaternary Science Reviews, 24, 821– 834. B ARNES , R. D. 1987. Invertebrate Zoology. Fifth Edition. Fort Worth, Texas: Harcourt Brace Jovanovich College Publishers. B ARRON , J. A. 1993. Diatoms. In: L IPPS , J. E. (ed.) Fossil Prokaryotes and Protists. Blackwell Scientific Publications, Oxford, 155–167. B ASSINOT , F., M ELIERES , F., G EHLEN , M., L EVI , C. & L ABEYRIE , L. 2004. Crystallinity of foraminifer shells: a proxy to reconstruct past bottom water CO322 changes? Geochemistry, Geophysics, Geosystems, 5, doi:10.1029/2003GC000668. B AUCH , D., D ARLING , K., S IMSTICH , J., B AUCH , H. A., E RIENKEUSER , H. & K ROON , D. 2003. Palaeoceanographic implications of genetic variation in living North Atlantic Neogloboquadrine pachyderma. Nature, 424, 299– 302. B AUMANN , K. H. & F REITAG , T. 2004. Pleistocene fluctuations in the northern Benguela Current system as revealed by coccolith assemblages. Marine Micropaleontology, 52, 195 –215.
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW B AUMANN , K. H., A NDRULEIT , H., B O¨ CKEL , B., G EISEN , M. & K INKEL , H. 2005. The significance of extant coccolithophores as indicators of ocean water masses, surface water temperature, and palaeoproductivity: a review. Pala¨ontologische Zeitschrift, 79, 93–112. B ECK , J. W., E DWARDS , R. L. ET AL . 1992. Sea-surface temperature from coral skeletal strontium/calcium ratios. Science, 257, 644–647. B ENGTSON , S. 1994. The advent of animal skeletons. In: B ENGTSON , S. (ed.) Early Life on Earth. Nobel Symposium 84. Columbia University Press, New York, 421–425. B ENTOV , S. & E REZ , J. 2006. Impact of biomineralization process on the Mg content of foraminiferal shells: a biological perspective. Geochemistry, Geophysics, Geosystems, 7, Q01P08. B ERNER , R. A. 1990. Atmospheric carbon dioxide levels over Phanerozoic time. Science, 249, 1382– 1386. B ICE , K. L., L AYNE , G. D. & D AHL , K. 2005. Application of secondary ion mass spectrometry to the determination of Mg/Ca in rare, delicate, or altered planktonic foraminifera: Examples from the Holocene, Paleogene, and Cretaceous. Geochemistry, Geophysics, Geosystems, 6, Q12P07, doi:10.1029/2005GC000974. B LACKMON , P. D. & T ODD , R. 1959. Mineralogy of some foraminifera as related to their classification and ecology. Journal of Paleontology, 33, 1 –15. B OWEN , H. J. M. 1956. Strontium and barium in seawater and marine organisms. Journal of the Marine Biological Association of the UK, 35, 451– 460. B OWN , P., L EES , J. A. & Y OUNG , J. R. 2004. Calcareous nannoplankton evolution and diversity through time. In: T HIERSTEIN , H. R. & Y OUNG , J. R. (eds) Coccolithophores – From Molecular Processes to Global Impacts. Springer, Berlin, 481–508. B OYLE , E. A. & K EIGWIN , L. D. 1985. Comparison of Atlantic and Pacific paleochemical records for the last 215,000 years: changes in deep ocean circulation and chemical inventories. Earth and Planetary Science Letters, 76, 135– 150. B ROECKER , W. S. & C LARK , E. 2001. An evaluation of Lohmann’s foraminiferal weight dissolution index. Paleoceanography, 16, 534 –534. B ROWN , S. J. & E LDERFIELD , H. 1996. Variations in Mg/Ca and Sr/ca ratios of planktonic foraminifera caused by postdepositional dissolution: Evidence of shallow Mg-dependent dissolution. Paleoceanography, 11, 543–551. B RZEZINSKI , M. A., P RIDE , C. J., F RANCK , V. M., S IGMAN , D. M., S ARMIENTO , J. L., M ATSUMOTO , K., G RUBER , N., R AU , G. H. & C OALE , K. H. 2002. A switch from Si(OH)4 to NO32 depletion in the glacial Southern ocean. Geophysical Research Letters, 29, 1564, doi: 10.1029/2001GL014349. B URTON , K. W. & V ANCE , D. 2000. Glacial-interglacial variations in the neodymium isotope composition of seawater in the Bay of Bengal recorded by planktonic foraminifera. Earth and Planetary Science Letters, 176, 425– 441. C HANG , V. T.-C., W ILLIAMS , R. J. P., M AKISHIMA , A., B ELSHAW , N. S. & O’N IONS , R. K. 2004. Mg and Ca isotope fractionation during CaCO3 biomineralisation. Biochemical and Biophysical Research Communications, 323, 79–85.
27
C HAPMAN , M. R., S HACKLETON , N. J., Z HAO , M. & E GLINTON , G. 1996. Faunal and alkenone reconstructions of subtropical North Atlantic surface hydrography and paleotemperature over the last 28 ka. Paleoceanography, 11, 343– 357. C HAVE , K. E. 1954. Aspects of the biogeochemistry of magnesium I. Calcareous marine organisms. Journal of Geology, 62, 266–283. C LARKSON , E. N. K. 1998. Invertebrate Palaeontology and Evolution. Allen and Unwin, London, UK. CLIMAP-Project Members 1976. The surface of the ice-age earth. Science, 191, 1131– 1137. CLIMAP-Project Members 1981. Seasonal Reconstructions of the Earth’s surface at the Last Glacial Maximum. Geological Society America Map and Chart Series MC-36. CLIMAP-Project Members 1984. The last interglacial ocean. Quaternary Research, 21, 123– 224. C OHEN , A. L., G AETANI , G. A., L UNDA¨ LV , T., C ORLISS , B. H. & G EORGE , R. Y. 2006. Compositional variability in a cold-water scleractinian, Lophelia pertusa: new insights into ‘vital effects’. Geochemistry, Geophysics, Geosystems, 7, Q12004, doi:10.1029/ 2006GC001354. C OHEN , A. L. & M C C ONNAUGHEY , T. 2003. Geochemical perspectives on coral mineralization. In: D OVE , P. M., D E Y OREO , J. J. & W EINER , S. (eds) Biomineralization. Reviews in Mineralogy and Geochemistry, 54, 151– 187. C ONLEY , D. J., Z IMBA , P. V. & T HERIOT , E. 1994. Silica content of freshwater and marine benthic diatoms. In: K OCIOLEK , J. P. (ed.) Proceedings of the 11th International Diatom Symposium, San Francisco, 1990. Memoir Californian Academy of Science, 17, 95–101. C ORRE` GE , T. 2006. Sea surface temperature and salinity reconstruction from coral geochemical tracers. Palaeogeography, Palaeoclimatology, Palaeoecology, 232, 408 –428. C ROSTA , X. & S HEMESH , A. 2002. Reconciling down core anticorrelation of diatom carbon and nitrogen isotopic ratios from the Southern Ocean. Paleoceanography, 17, 1010, doi:10.1029/2000 PA000565. C ULVER , S. J. 1993. Foraminifera. In: L IPPS , J. H. (ed.) Fossil Prokaryotes and Protists. Blackwell Scientific, Oxford, 203 –247. D ARLING , K. F., W ADE , C. M., S TEWARD , I. A., K ROON , D., D INGLE , R. & L EIGH B ROWN , A. J. 2000. Molecular evidence for genetic mixing of Arctic and Antarctic subpolar populations of planktonic foraminfers. Nature, 405, 43–47. D E L A R OCHA , C. L. 2006. Opal-based proxies of paleoenvironmental conditions. Global Biogeochemical Cycles, 20, GB4S09, doi:10.1029/2005 GB002664. D E L A R OCHA , C. L., B RZEZINSKI , M. A. & D E N IRO , M. J. 1997. Fractionation of silicon isotopes by marine diatoms during biogenic silica formation. Geochimica et Cosmochimica Acta, 61, 5051–5056. D E L A R OCHA , C. L., B RZEZINSKI , M. A., D E N IRO , M. J. & S HEMESH , A. 1998. Silicon-isotope composition of diatoms as an indicator of past oceanic change. Nature, 395, 680–683.
28
R. H. JAMES & W. E. N. AUSTIN
D EBENAY , J. P., G UILLOU , J. J., G ESLIN , E. & L ESOURD , M. 2000. Crystallization of calcite in foraminiferal tests. Micropaleontology, 46, 87– 94. D E V ILLIERS , S., S HEN , G. T. & N ELSON , B. K. 1994. The Sr/Ca-temperature relationship in coralline aragonite: Influence of variability in (Sr/Ca)seawater and skeletal growth parameters. Geochimica et Cosmochimica Acta, 58, 197–208. D E V ILLIERS , S., G REAVES , M. & E LDERFIELD , H. 2002. An intensity ratio calibration method for the accurate determination of Mg/Ca and Sr/Ca of marine carbonates by ICP-AES. Geochemistry, Geophysics, Geosystems, 3, 1001, doi:10.1029/2001GC000169. D ICKENS , G. R., O’N EIL , J. R., R EA , D. K. & Z ACHOS , J. C. 1995. Dissociation of oceanic methane hydrate as a cause of the carbon isotope excursion at the end of the Paleocene. Paleoceanography, 10, 965–972. D IXON , P. R., P ERRIN , R. E., R OKOP , D. J., M AECK , R., J ANECKY , D. R. & B ANAR , J. P. 1993. Measurements of iron isotopes (54Fe, 56Fe, 57Fe and 58Fe) in sub micron quantities of iron. Analytical Chemistry, 65, 2125–2130. D RUFFEL , E. R. M. & G RIFFIN , S. 1993. Large variations of surface ocean radiocarbon: evidence of circulation changes in the southwestern Pacific. Journal of Geophysical Research, 98, 20 249– 20 259. D UDLEY , W. C., D UPLESSY , J. C., B LACKWELDER , P., B RAND , L. E. & G UILLARD , R. R. L. 1980. Coccoliths in Pleistocene-Holocene nanofossil assemblages. Nature, 285, 222 –223. D UPLESSY , J.-C., L ABEYRIE , L. D., J UILLET -L ECLERC , A., M AITRE , F., D UPRAT , J. & S ARNTHEIN , M. 1991. Surface salinity reconstruction of the North Atlantic Ocean during the last glacial maximum. Oceanologica Acta, 14, 311– 324. E DVARDSEN , B., E IKREM , W., G REEN , J. C., A NDERSEN , R. A., VAN DER S TAAY , S. Y. M. & M EDLIN , L. K. 2000. Phylogenetic reconstructions of the Haptophyta inferred from 18S ribosomal DNA sequences and available morphological data. Phycologia, 39, 19–35. E GGINS , S. M., S ADEKOV , A. & D E D ECKKER , P. 2004. Modulation and daily banding of Mg/Ca in Orbulina universa tests by symbiont photosynthesis and respiration: a complication for seawater thermometry? Earth and Planetary Science Letters, 225, 411–419. E LDERFIELD , H. 2002. Foraminiferal Mg/Ca paleothermometry: expected advances and unexpected consequences. Geochimica et Cosmochimica Acta, 66, A213. E LDERFIELD , H., B ERTRAM , C. J. & E REZ , J. 1996. A biomineralization model for the incorporation of trace elements into foraminiferal calcium carbonate. Earth and Planetary Science Letters, 142, 409–423. E LDERFIELD , H. & G ANSSEN , G. 2000. Past temperature and d18O of surface ocean waters inferred from foraminiferal Mg/Ca ratios. Nature, 405, 442–445. E LDERFIELD , H., Y U , J., A NAND , P., K IEFER , T. & N YLAND , B. 2006. Calibrations for benthic foraminiferal Mg/Ca paleothermometry and the carbonate ion hypothesis. Earth and Planetary Science Letters, 250, 633 –649. E LLIOT , M., DE M ENOCAL , P. B., L INSLEY , B. K. & H OWE , S. S. 2003. Environmental controls on the
stable isotopic composition of Mercenaria mercenaria: potential application to paleoenvironmental studies. Geochemistry, Geophysics, Geosystems, 4, 1056, doi:10.1029/2002GC000425. E LLWOOD , M. J. & H UNTER , K. A. 2000. The incorporation of zinc and iron into the frustule of the marine diatom Thalassiosira pseudonana. Limnology and Oceanography, 45, 1517–1524. E MILIANI , C. 1955. Pleistocene temperatures. Journal of Geology, 63, 538– 578. E REZ , J. 2003. The source of ions for biomineralization in Foraminifera and their implications for paleoceanographic proxies. In: D OVE , P. M., D E Y OREO , J. J. & W EINER , S. (eds) Biomineralization. Reviews in Mineralogy and Geochemistry, 54, 115 –149. E REZ , J. & H ONJO , S. 1981. Comparison of isotopic composition of planktonic foraminifera in plankton tows, sediment traps and sediments. Palaeogeography, Palaeoclimatology, Palaeoecology, 33, 129– 156. F REITAS , P. S., C LARKE , L. J., K ENNEDY , H., R ICHARDSON , C. A. & A BRANTES , F. 2006. Environmental and biological controls on elemental (Mg/Ca, Sr/Ca and Mn/Ca) ratios in shells of the king scallop Pecten maximus. Geochimica et Cosmochimica Acta, 70, 5119– 5133. F ROELICH , P. N., B LANK , V. R., M ORTLOCK , R. A., C HILLRUND , S. N., D USTAN , W., U DOMIAT , A. & P ENG , T. H. 1992. River fluxes of dissolved silica were higher during glacials: Ge/Si in diatoms, rivers and oceans. Paleoceanography, 7, 739–767. G AGAN , M. K., A YLIFFE , L. K., B ECK , J. W., C OLE , J. E., D RUFFEL , E. R. M., D UNBAR , R. B. & S CHRAG , D. P. 2000. New views of tropical paleoclimates from corals. Quaternary Science Reviews, 19, 45–64. G ALER , S. J. 1999. Optimal double and triple spiking for high precision lead isotopic measurement. Chemical Geology, 157, 255 –274. G HOSH , P., A DKINS , J., A FFEK , H., B ALTA , B., G UO , W., S CHAUBLE , E. A., S CHRAG , D. & E ILER , J. M. 2006. 13 C– 18O bonds in carbonate minerals: A new kind of paleothermometer. Geochimica et Cosmochimica Acta, 70, 1439– 1456. G ILLIKIN , D. P., L ORRAIN , A., N AVEZ , J., T AYLOR , J. W., A NDRE´ , L., K EPPENS , E., B AEYENS , W. & D EHAIRS , F. 2005. Strong biological controls on Sr/Ca ratios in aragonitic marine bivalve shells. Geochemistry, Geophysics, Geosystems, 6, Q05009, doi:10.1029/2004GC000874. G OODAY , A. J. 1999. Biodiversity of foraminifera and other protists in the deep sea: scales and patterns. Belgian Journal of Zoology, 129, 61–80. G OODAY , A. J., N OMAKI , H. & K ITAZATO , H. 2008. Modern deep-sea benthic foraminifera: a brief review of their morphology-based biodiversity and trophic diversity. In: A USTIN , W. E. N. & J AMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, 303, 97–119. G REAVES , M., B ARKER , S., D AUNT , C. & E LDERFIELD , H. 2005. Accuracy, standardization, and interlaboratory calibration standards for foraminiferal Mg/Ca thermometry. Geochemistry, Geophysics, Geosystems, 6(2), Q02D13, doi:10.1029/2004GC000790.
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW G REEN , D. R. H., C OOPER , M. J., G ERMAN , C. R. & W ILSON , P. A. 2003. Optimization of an inductively coupled plasma-optical emission spectrometry method for the rapid determination of highprecision Mg/Ca and Sr/Ca in foraminiferal calcite. Geochemistry, Geophysics, Geosystems, 4, 8404, doi:10.1029/2002GC000488. H ALEY , B. A. & K LINKHAMMER , G. P. 2002. Development of a flow-through system for cleaning and dissolving foraminiferal tests. Chemical Geology, 185, 51– 69. H ANSEN , H. J. 1999. Shell construction in modern calcareous Foraminifera. In: S EN G UPTA , B. K. (ed.) Modern Foraminifera. Kluwer Academic, Norwell, Mass., 57–70. H ART , S. R. & C OHEN , A. L. 1996. An ion probe study of annual cycles of Sr/Ca and other trace elements in corals. Geochimica et Cosmochimica Acta, 60, 3075–3084. H ATHORNE , E. C. 2004. The Trace Element and Lithium Isotope Composition of Planktonic Foraminifera. PhD thesis, The Open University, p. 247. H ATHORNE , E. C., A LARD , O., J AMES , R. H. & R OGERS , N. W. 2003. Determination of intratest variability of trace elements in foraminifera by laser ablation inductively coupled plasma mass-spectrometry. Geochemistry, Geophysics, Geosystems, 4, 8408. H ATHORNE , E. C., J AMES , R. H., S AVAGE , P. & A LARD , O. 2008. Physical and chemical characteristics of particles produced by laser ablation of biogenic calcium carbonate. Journal of Analytical Atomic Spectrometry, 23, 240–243. H AY , W. W. 2004. Carbonate fluxes and calcareous nannoplankton. In: T HIERSTEIN , H. R. & Y OUNG , J. R. (eds) Coccolithophores – From Molecular Processes to Global Impacts. Springer, Berlin, 509– 528. H ENDERSON , G. M. 2002. New proxies for paleoclimate. Earth and Planetary Science Letters, 203, 1– 13. H ORN , I., VON B LANCKENBURG , F., S CHOENBERG , R., S TEINHOEFEL , G. & M ARKL , G. 2006. In-situ iron isotope ratio determination using UV-femtosecond laser ablation with application to hydrothermal ore formation processes. Geochimica et Cosmochimica Acta, 70, 3677–3688. I MBRIE , J. & K IPP , N. G. 1971. A new micropaleontological method for quantitative paleoclimatology: Application to a Late Pleistocene Caribbean core. In: T UREKIAN , K. (ed.) The Late Cenozoic Glacial Ages. Yale University Press, New Haven, Conn., 71–181. I MMENHAUSER , A., N A¨ GLER , T. F., S TEUBER , T. & H IPPLER , D. 2005. A critical assessment of mollusk 18 O/16O, Mg/Ca, and 44Ca/40Ca ratios as proxies for Cretaceous seawater temperature seasonality. Palaeogeography, Palaeoclimatology, Palaeoecology, 215, 221– 237. J AMES , R. H. & P ALMER , M. R. 2000. The lithium isotope composition of international rock standards and solutions. Chemical Geology, 166, 319– 326. J EANDEL , C., B ISHOP , J. K. & Z INDLER , A. 1995. Exchange of neodymium and its isotopes between seawater and small and large particles in the Sargasso Sea. Geochimica et Cosmochimica Acta, 59, 535– 547. K LEIN , R. T., L OHMANN , K. C. & T HAYER , C. W. 1996a. Sr/Ca and 13C/12C ratios in skeletal calcite of Mytilus
29
trossulus: covariation with metabolic rate, salinity, and carbon isotopic composition of seawater. Geochimica et Cosmochimica Acta, 43, 1505– 1520. K LEIN , R. T., L OHMANN , K. C. & T HAYER , C. W. (1996b). Bivalve skeletons record sea-surface temperatures and d18O via Mg/Ca and 18O/16O ratios. Geology, 24, 415– 418. K NOLL , A. H. 2003. Biomineralization and evolutionary history. In: D OVE , P. M., D E Y OREO , J. J. & W EINER , S. (eds) Biomineralization. Reviews in Mineralogy and Geochemistry, 54, 329–356. K NOLL , A. H. & S EMIKHATOV , M. A. 1998. The genesis and time distribution of two distinctive Proterozoic stromatolite microstructures. Palaios, 13, 407 –421. K NOLL , A. H. & S WETT , K. 1990. Carbonate deposition during the late Proterozoic era: an example from Spitsbergen. American Journal of Science, 290A, 104–132. K NUTSON , D. W., B UDDEMEIER , R. W. & S MITH , S. V. 1972. Coral chronologies: seasonal growth bands in reef corals. Science, 177, 270–272. K UHN , H. R. & G U¨ NTHER , D. 2004. Laser ablation ICP-MS: Particle size dependent elemental composition studies on filter-collected and online measured aerosols from glass. Journal of Analytical Atomic Spectrometry, 19, 1158–1164. L ABEYRIE , L. D. 1974. New approach to surface seawater paleotemperatures using 18O/16O ratios in silica of diatom frustules. Nature, 248, 40–42. L ABEYRIE , L. D. 1979. La composition isotopique de l’oxyge`ne de la silice des valves de diatome´es. Mise au point d’une nouvelle me´thode de pale´oclimatologie. The`se d’Etat, Universite´ de Paris XI, Note CEA #2086. L ALLI , C. M. & P ARSONS , T. R. 1995. Biological Oceanography: An Introduction. Oxford, UK. ButterworthHeinemann Ltd., 220– 233. L EA , D. W., M ASHIOTTA , T. A. & S PERO , H. J. 1999. Controls on magnesium and strontium uptake in planktonic foraminifera determined by live culturing. Geochimica et Cosmochimica Acta, 63, 2369–2379. L EAR , C. H., E LDERFIELD , H. & W ILSON , P. A. 2000. Cenozoic deep-sea temperatures and global ice volumes from Mg/Ca in benthic foraminiferal calcite. Science, 287, 269 –272. L EAR , C. H., R OSENTHAL , Y. & S LOWEY , N. 2002. Benthic foraminiferal Mg/Ca-paleothermometry: a revised core-top calibration. Geochimica et Cosmochimica Acta, 66, 3375–3387. L EAR , C. H., R OSENTHAL , Y., C OXALL , H. K. & W ILSON , P. A. 2004. Late Eocene to early Miocene ice shett dynamics and the global carbon cycle. Paleoceanography, 19, PA4015, doi:10.1029/ 2004PA001039. L ORRAIN , A., G ILLIKIN , D., P AULET , Y.-M., C HAUVAUD , L., L EMERCIER , A., N AVEZ , J. & A NDRE , L. E´. 2005. Strong kinetic effects on Sr/Ca ratios in the calcitic bivalve Pecten maximus. Geology, 33, 965– 968. L ORENS , R. B. 1981. Sr, Cd, Mn, and Co distribution coefficients in calcite as a function of calcite precipitation rate. Geochimica et Cosmochimica Acta, 45, 553– 561. L OWENSTAM , H. A. 1981. Minerals formed by organisms. Science, 211, 1126–1131.
30
R. H. JAMES & W. E. N. AUSTIN
M ANN , S. 1983. Mineralization in biological systems. Structural Bonding, 54, 125 –174. M ARTIN , P. A. & L EA , D. W. 2002. A simple evaluation of cleaning procedures on fossil benthic foraminiferal Mg/Ca. Geochemistry, Geophysics, Geosystems, 3, doi:10.1029/2001GC000280. M ARTIN , P. A., L EA , D. W., R OSENTHAL , Y., S HACKLETON , N. J., S ARNTHEIN , M. & P APENFUSS , T. 2002. Quaternary deep sea temperature histories derived from benthic foraminiferal Mg/Ca. Earth and Planetary Science Letters, 198, 193–209. M C A RTHUR , J. M., H OWARTH , R. J. & B AILEY , T. R. 2001. Strontium isotope stratigraphy: LOWESS Version 3: Best fit to the marine Sr-isotope curve for 0 –509 Ma and accompanying look-up table for deriving numerical age. Journal of Geology, 109, 155– 170. M C C ONNAUGHEY , T. 1989. 13C and 18O isotopic disequilibrium in biological carbonates: I. Patterns. Geochimica et Cosmochimica Acta, 53, 151–162. M C C ORCKLE , D. C., B ERNHARD , J. M. ET AL . 2008. The carbon and oxygen stable isotopic composition of cultured benthic Foraminifera. In: A USTIN , W. E. N. & J AMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, 303, 135 –154. M C C ULLOCH , M. T., G AGN , M. K., M ORTIMER , G. E., C HIVAS , A. R. & I SDALE , P. J. 1994. A highresolution Sr/Ca and d18O coral record from the Great Barrier Reef, Australia, and the 1982– 1983 El Nino. Geochimica et Cosmochimica Acta, 58, 2747–2754. M C G REGOR , H. V. & G AGAN , M. K. 2003. Diagenesis and geochemistry of Porites corals from Papua New Guinea: Implications for paleoclimate reconstruction. Geochimica et Cosmochimica Acta, 67, 2147–2156. M C I NTYRE , A. & B E´ , A. 1967. Modern Coccolithophoridae in the Atlantic Ocean. I. Placoliths and crytholiths. Deep-Sea Research, 14, 561–597. M EDINA -E LIZALDE , M. & L EA , D. W. 2005. The midPleistocene transition in the tropical Pacific. Science, 310, 1009– 1012. M EIBOM , A., M OSTEFAOUI , S., C UIF , J.-P., D AUPHIN , Y., H OULBREQUE , F., D UNBAR , R. & C ONSTANZ , B. 2007. Biological forcing controls the chemistry of reef-building coral skeleton. Geophysical Research Letters, 34, L02601, doi:10.1029/2006GL028657. M ILLIMAN , J. D., T ROY , P. J., B ALCH , W. M., A DAMS , A. K., L I , Y.-H. & M ACKENZIE , F. T. 1999. Biologically mediated dissolution of calcium carbonate above the chemical lysocline? Deep-Sea Research, 46, 1653–1669. M IN , G. R., E DWARDS , R. L., T AYLOR , F. W., R ECY , J., G ALLUP , C. D. & B ECK , J. W. 1995. Annual cycles of U/Ca in coral skeletons and U/Ca thermometry. Geochimica et Cosmochimica Acta, 59, 2025–2042. M INOLETTI , F., G ARDIN , S., N ICTO , E., R ENARD , M. & S PEZZAFERRI , S. 2001. Mise au point d’un protocol expe´rimental de separation granulome´trique d’assemblages de nannofossiles calcaires; applications paleoe´cologiques et geochimiques. Bulletin de la Societe Geologique de France, 172, 437–446.
M ITTERER , R. M. 1978. Amino acid composition of metal binding capability of the skeletal protein of corals. Bulletin Marine Science, 28, 173– 180. M ONTAGNA , P., M C C ULLOCH , M., T AVIANI , M., M AZZOLI , C. & V ENDRELL , B. 2006. Phosphorus in cold-water corals as a proxy for seawater nutrient chemistry. Science, 312, 1788–1791. M ORSE , J. W. & B ENDER , M. L. 1990. Partition coefficients in calcite: Examination of factors influencing the validity of experimental results and their application to natural systems. Chemical Geology, 82, 265–277. M UCCI , A. 1987. Influence of temperature on the composition of magnesian calcite overgrowths precipitated from seawater. Geochimica et Cosmochimica Acta, 51, 1977– 1984. N ELSON , D. M., T RE´ GUER , P., B RZEZINSKI , M. A., L EYNAERT , A. & Q UE´ GUINER , B. 1995. Production and dissolution of biogenic silica in the ocean – revised global estimates, comparison with regional data and relationship to biogenic sedimentation. Global Biogeochemical Cycles, 9, 359 –372. N U¨ RNBERG , D., B IJMA , J. & H EMBELEN , C. 1996. Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperatures. Geochimica et Cosmochimica Acta, 60, 803– 814. O STERMANN , D. R. & C URRY , W. B. 2000. Calibration of stable isotopic data: An enriched d18O standard used for source gas mixing detection and correction. Paleoceanography, 15, 353–360. O URBACK , T., C ORRE` GE , T., M ALAIZE´ , B., L E C ORNEC , F., C HARLIER , K. & P EYPOUQUET , J. P. 2006. A high-resolution investigation of temperature, salinity, and upwelling activity proxies in corals. Geochemistry, Geophysics, Geosystems, 7, Q03013, doi:10.1029/2005GC001064. P AULL , C. K. & T HIERSTEIN , H. R. 1990. Comparison of fine fraction with monospecific foraminiferal stable isotope stratigraphies from pelagic carbonates across the last Glacial termination. Marine Micropaleontology, 16, 207–217. P EARCE , N. J. G. 1997. A compilation of new and published major and trace elements data for NIST SRM 610 and NIST SRM 612 glass reference materials. Geostandards Newsletter, 21, 115–144. P EARSON , P. N., D ITCHFIELD , P. W., S INGANO , J., H ARCOURT -B ROWN , K. G., N ICHOLAS , C. J., O LSSON , R. K., S HACKLETON , N. J. & H ALL , M. A. 2001. Warm tropical sea surface temperatures in the Late Cretaceous and Eocene epochs. Nature, 413, 481– 487. P IEPGRAS , D. J. & W ASSERBURG , G. J. 1987. Rare earth element transport in the western North Atlantic inferred from Nd isotopic observations. Geochimica et Cosmochimica Acta, 51, 1257– 1271. P ORTER , S. M. & K NOLL , A. H. 2000. Testate amoebae in the Neoproterozoic Era: evidence from vase-shaped microfossils in the Chuar Group, Grand Canyon. Paleobiology, 26, 360–385. P RAHL , F. G. & W AKEHAM , S. G. 1987. Calibration of unsaturation patterns in long-chain ketone compositions for paleotemperature assessment. Nature, 320, 367–369.
PALAEOCEANOGRAPHIC ENVIRONMENTAL PROXIES: A REVIEW P RAHL , F. G., M UEHLHAUSED , L. A. & Z AHNLE , D. L. 1988. Further evaluation of long-chain alkenones as indicators of paleoceanographic conditions. Geochimica et Cosmochimica Acta, 52, 2303– 2310. Q UINN , T. M., T AYLOR , F. W. & C ROWLEY , T. J. 2006. Coral-based climate variability in the western Pacific Warm Pool since 1867. Journal of Geophysical Research, 111, C11006, doi:10.1029/2005JC003243. R AYNER , N. A., P ARKER , D. E., H ORTON , E. M., F OLLAND , C. K., A LEXANDER , L. V., R OWELL , D. P., K ENT , E. C. & K APLAN , A. 2003. Global analyses of SST, sea ice, and night marine air temperature since the late nineteenth century. Journal of Geophysical Research, 108, 4407, doi:10.1029/2002JD002670. R EISS , Z. 1958. Classification of lamellar foraminifera. Micropaleontology, 4, 51–70. R EYNAUD -V AGANAY , S., G ATTUSO , J. P., C UIF , J. P., J AUBERT , J. & J UILLET -L ECLERC , A. 1999. A novel culture technique for scleractinian corals: application to investigate changes in skeletal d18O as a function of temperature. Marine Ecology– Progress Series, 180, 121–130. R ICKABY , R. E. M., B ARD , E., S ONZOGNI , C., R OSTEK , F., B EAUFORT , L., B ARKER , S., R EES , G. & S CHRAG , D. P. 2007. Coccolith chemistry reveals secular variations in the global carbon cycle? Earth and Planetary Science Letters, 253, 83–95. R ICKABY , R. E. M., S CHRAG , D. P., R IEBESALL , U. & Z ONDERVAN , I. 2002. Growth-rate dependence of Sr incorporation during calcification of Emiliani huxleyi. Global Biogeochemical Cycles, 16, (2001GB001408). R ICKABY , R. E. M., G REAVES , M. J. & E LDERFIELD , H. 2000. Cd in planktonic and benthic foraminiferal shells determined by thermal ionisation mass spectrometry. Geochimica et Cosmochimica Acta, 64, 1229– 1236. R OBERTS , J. M., W HEELER , A. J. & F REIWALD , A. 2006. Reefs of the Deep: The biology and geology of coldwater coral ecosystems. Science, 312, 543– 547. R OSELL -M ELE´ , A., B ARD , E., E MEIS , K.-C. ET AL . 2001. Precision of the current methods to measure the alkenone proxy UK37 and absolute alkenone abundance in sediments: Results of an interlaboratory comparison study. Geochemistry, Geophysics, Geosystems, 2, # 2000GC000141. R OSENTHAL , Y., F IELD , M. P. & S HERRELL , R. M. 1999. Precise determination of element/calcium ratios in calcareous samples using sector field inductively coupled plasma mass spectrometry. Analytical Chemistry, 71, 3248– 3253. R OSENTHAL , Y., L OHMANN , G. P., L OHMANN , K. C. & S HERRELL , R. M. 2000. Incorporation and preservation of Mg in Globigerinoides sacculifer: implications for reconstructing O-18/O-16 of seawater. Paleoceanography, 15, 135 –145. R OSENTHAL , Y., P ERRON -C ASHMAN , S., L EAR , C. H. ET AL . 2004. Laboratory intercomparison study of Mg/Ca and Sr/Ca measurements in planktonic foraminifera for paleoceanographic research. Geochemistry, Geophysics, Geosystems, 5, Q04D09, doi:10.1029/2003GC000650. R OUXEL , O., G ALY , A. & E LDERFIELD , H. 2006. Germanium isotopic variations in igneous rocks and
31
marine sediments. Geochimica et Cosmochimica Acta, 70, 3387–3400. ROYAL S OCIETY 2005. Ocean Acidification due to Increasing Atmospheric Carbon Dioxide. Clyvedon Press, Cardiff, UK. R UDDIMAN , W. F., R AYMO , M. & M C I NTYRE , A. 1986. Matuyama 41 000-year cycles: North Atlantic Ocean and northern hemisphere ice sheets. Earth and Planetary Science Letters, 80, 117– 129. R UNNEGAR , B. 1996. Early evolution of Mollusca: the fossil record. In: T AYLOR , J. (ed.) Origin and Evolutionary Radiation of the Mollusca. Oxford University Press, Oxford, 77–87. S ANCETTA , C. 1992. Primary production in the glacial North Atlantic and North Pacific Oceans. Nature, 360, 249–251. S ARMIENTO , J. L. & G RUBER , N. 2002. Sinks for anthropogenic carbon. Physics Today, 55, 30–36. S CHMIDT , M., B OTZ , R., S TOFFERS , P., A NDERS , T. & B OHRMANN , G. 1997. Oxygen isotopes in marine diatoms: a comparative study of analytical techniques and new results on the isotope composition of recent marine diatoms. Geochimica et Cosmochimica Acta, 61, 2275–2280. S CHMIDT , M., B OTZ , R., R ICKERT , D., B OHRMANN , G., H ALL , S. R. & M ANN , S. 2001. Oxygen isotopes of marine diatoms and relations to opal-A maturation. Geochimica et Cosmochimica Acta, 65, 201–211. S CHMIDT , W. J. 1924. Die Bausteine des Tierkorpers in Polarisiertem Lichte. F. Cohen Verlag, Bonn. S EN G UPTA , B. K. (ed.) 1999. Modern Foraminifera. Kluwer Academic Publishers, Boston. S EXTON , P. F., W ILSON , P. A. & P EARSON , P. N. 2006. Microstructural and geochemical perspectives on planktic foraminiferal preservation: ‘Glassy’ versus ‘Frosty’. Geochemistry, Geophysics, Geosystems, 7, Q12P19, doi:10.1029/2006GC001291. S HEMESH , A., C HARLES , C. D. & F AIRBANKS , R. G. 1992. Oxygen isotopes in biogenic silica: global changes in ocean temperature and isotopic composition. Science, 26, 1434– 1436. S HEN , G. T. & D UNBAR , R. B. 1995. Environmental controls on uranium in reef corals. Geochimica et Cosmochimica Acta, 59, 2009–2024. S MITH , S. V., B UDDEMEIER , R. W., R EDALJE , R. C. & H OUCK , J. E. 1979. Strontium-calcium thermometry in coral skeletons. Science, 204, 404–407. S MITH , J. E., R ISK , M. J., S CHWARCZ , H. P. & M C C ONNAUGHEY , T. A. 1997. Rapid climate change in the North Atlantic during the Younger Dryas recorded by deep-sea corals. Nature, 386, 818– 820. S PO¨ TL , C. & V ENNERMANN , T. W. 2003. Continuousflow isotope ratio mass spectrometric analysis of carbonate minerals. Rapid Communications in Mass Spectrometry, 17, 1004–1006. S TOLL , H. M. & S CHRAG , D. P. 1998. Effects of Quaternary sea level cycles on strontium in seawater. Geochimica et Cosmochimica Acta, 62, 1107–1118. S TOLL , H. M. & S CHRAG , D. P. 2000. Coccolith Sr/Ca as a new indicator of coccolithophorid calcification and growth rate. Geochemistry, Geophysics, Geosystems, 1, doi:10.1029/1999GC000015.
32
R. H. JAMES & W. E. N. AUSTIN
S TOLL , H. M. & Z IVERI , P. 2002. Separation of monospecific and restricted coccolith assemblages from sediments using differential settling velocity. Marine Micropaleontology, 46, 209– 221. S TOLL , H. M., Z IVERI , P., G IESEN , M., P ROBERT , I. & Y OUNG , J. R. 2002. Potential and limitations of Sr/Ca ratios in coccolith carbonate: new perspectives from cultures and monospecific samples from sediments. Philosphical Transactions of the Royal Society of London, 360, 719– 747. S WART , P. K. & H UBBARD , J. A. E. B. 1982. Uranium in scleractinian coral skeletons. Coral Reefs, 1, 13–19. T AYLOR , J. D., K ENNEDY , W. J. & H ALL , A. 1969. The shell structure and mineralogy of the Bivalvia. Introduction, Nuculacea – Trigoniacea. Bulletin of the British Museum of Zoology, Supp. 3, 1 –125. T ESORIERO , A. J. & P ANKOW , J. F. 1996. Solid solution partitioning of Sr2þ, Ba2þ, and Cd2þ to calcite. Geochimica et Cosmochimica Acta, 60, 1053–1063. T OWE , K. M. & C IFELLI , R. 1967. Wall structure in the calcareous foraminifera: Crystalographic aspects and a model for calcification. Journal of Paleontology, 41, 742 –762. U REY , H. C. 1947. The thermodynamic properties of isotopic substances. Journal of the Chemical Society, 1947, 562– 581. VAN DE F LIERDT , T., F RANK , M., L EE , D.-C., H ALLIDAY , A. N., R EYNOLDS , B. C. & H EIN , J. R. 2004. New constraints on the sources and behaviour of neodymium and hafnium in seawater from Pacific Ocean ferromanganese crusts. Geochimica et Cosmochimica Acta, 68, 3827–3843. VAN DE F LIERDT , T., R OBINSON , L. F., A DKINS , J. F., H EMMING , S. R. & G OLDSTEIN , S. L. 2006. Temporal stability of the neodymium isotope signature of the Holocene to glacial North Atlantic. Paleoceanography, 21, PA4102, doi:10.1029/2006PA001294. V ANCE , D., A RCHER , C., B ERMIN , J., K ENNAWAY , G., C OX , E. J., S TATHAM , P. J., L OHAN , M. C. & E LLWOOD , M. J. 2006. Zn isotopes as a new tracer of metal micronutrient usage in the oceans. Geochimica et Cosmochimica Acta, 70, A666. V IGIER , N., R OLLION -B ARD , C., S PEZZAFERRI , S. & B RUNET , F. 2007. In situ measurements of Li isotopes in foraminifera. Geochemistry, Geophysics, Geosystems, 8, Q01003, doi:10.1029/2006GC001432. W EFER , G., B ERGER , W. H., B IJMA , J. & F ISCHER , G. 1999. Clues to ocean history; a brief overview of proxies. In: F ISHER , G. & W EFER , G. (eds) Uses of Proxies in Paleoceanography: Examples from the South Atlantic. Springer-Verlag Berlin Heidelberg, 1 –68. W EIDMAN , C. R., J ONES , G. A. & L OHMAN , K. C. 1994. The long-lived mollusk A.islandica: a new paleoceanographic tool for the reconstruction of bottom water temperatures for the continental shelves of the northern North Atlantic Ocean. Journal of Geophysical Research, Oceans, 99, 18305–18314.
W EINER , S. & D OVE , P. M. 2003. An overview of biomineralization processes and the problem of the vital effect. In: D OVE , P. M., D E Y OREO , J. J. & W EINER , S. (eds) Biomineralization. Reviews in Mineralogy and Geochemistry, 54, 1 –29. W ESTBROEK , P., B ROWN , C. W., VAN B LEUSWIJK , J., B ROWNLEE , C., B RUMMER , G. J., C ONTE , M., E GGE , J., F ERNANDE´ Z , E., J ORDAN , R., K NAPPERTSBUSCH , M., S TEFELS , J., V ELDHUIS , M., VAN DER W AAL , P. & Y OUNG , J. R. 1993. A model system approach to biological climate forcing, the example of Emiliania huxleyi. Global Planetary Change, 8, 27–46. W ILBUR , K. M. 1961. Shell formation and regeneration. In: W ILBUR , K. M. & Y ONGE , C. M. (eds) Physiology of Mollusca, Vol. 1. Academic Press, New York, 243–281. W ILLIAMS , R. J. P. 2008. Some fundamental features of biomineralization. In: A USTIN , W. E. N. & J AMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, 303, 33–44. Y OUNG , J. R. 1994. Functions of coccolithophores. In: W INTER , A. & S IESSER , W. G. (eds) Coccolithophores. Cambridge University Press, UK, 63– 82. Y OUNG , J. R., D AVIS , S. A., B OWN , P. R. & M ANN , S. 1999. Coccolith ultrastructure and biomineralisation. Journal of Structural Biology, 126, 195–215. Y OUNG , J. R., B ERGEN , J. A., B OWN , P. R., B URNETT , J. A., F IORENTINO , A., J ORDAN , R. W., K LEIJNE , A., V AN N IEL , B. E., R OMEIN , A. J. T. & V ON S ALIS , K. 1997. Guidelines for coccolith and Calcareous nonnofossil terminology. Palaeontology, 40, 875– 912. Y OUNG , J. R., G EISEN , M., C ROS , L., K LEIJNE , A., S PRENGEL , C., P ROBERT , I. & O STERGAARD , J. B. 2003. A guide to extant coccolithophore taxonomy. Journal of Nannoplankton Research, Special Issue, 1, 1–125. Z ACHOS , J. C., S TOTT , L. D. & L OHMANN , K. C. 1994. Evolution of early Cenozoic marine temperatures. Paleoceanography, 9, 353– 387. Z ACHOS , J., P AGANI , M., S LOAN , L., T HOMAS , E. & B ILLUPS , K. 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science, 292, 686–693. Z ACHOS , J. C., R O¨ HL , U., S CHELLENBERG , S. A., S LUIJS , A., H ODELL , D. A., K ELLY , D. C., T HOMAS , E., N ICOLO , M., R AFFI , I., LOURENS , L. J., M C C ARREN , H. & K ROON , D. 2005. Rapid acidification of the ocean during the Paleocene-Eocene thermal maximum. Science, 308, 1611– 1615. Z IVERI , P., Y OUNG , J. R. & V AN H INTE , J. E. 1999. Coccolithophore export production and accumulation rates. GeoResearch Forum, 5, 41–56. Z IVERI , P., B AUMANN , K.-H., B O¨ CKEL , B., B OLLMANN , J. & Y OUNG , J. 2004. Present-day coccolithophore biogeography in the Atlantic Ocean. In: T HIERSTEIN , H. R. & Y OUNG , J. R. (eds) Coccolithophores – From Molecular Processes to Global Impacts. Springer, Berlin, 403 –428.
Some fundamental features of biomineralization R. J. P. WILLIAMS Inorganic Chemistry Laboratory, Oxford University, South Parks Road, Oxford OX1 3QR, UK (e-mail:
[email protected]) Abstract: This contribution summarizes the considerations that are of major importance in inorganic mineral formation before we look at specific biological minerals. Some factors which have to be taken into account (other than those that are well-known from inorganic (abiological) precipitations) are: the nature of the biological organic matrix; the restricted volume, outside or inside the cytoplasm, which can cause differences in impurity content (Mg); crystal morphology; and isotopic fractionation. Cases such as those of corals, foraminifera and coccoliths are taken as examples.
Factors affecting mineralization Equilibria The formation of a mineral from a solution is under a number of different controlling factors (Crick 1986; Frau´sto da Silva & Williams 1991). The simplest case we can consider is an equilibrium between a particular form of a mineral, say calcite, and the activities, not concentrations, of the ions, for example: Ca2þ þ CO2 3 CaCO3 (calcite) The activities depend upon the salt concentration and the equilibrium constant depends upon the temperature and the pressure. (Pressure becomes especially important in the ocean depths.) This equation is made complicated by the interaction in solution of Ca2þ ions with certain with cations anions and the interaction of CO22 3 (especially the proton) so that at least one, a second, equilibrium is very important: Hþ þ CO2 3 HCO3
where we must use activities again and they are dependent on salt concentration while the equilibrium constant of acid dissociation, pKa, is again dependent on temperature and pressure. The acidity (alkalinity) of the sea and its salt concentration, as well as its temperature, are known to have varied with time so that the equilibrium itself is not an easy historical problem to study. Keeping strictly to equilibrium conditions we turn to the free energy, stability, of CaCO3. It is never 100% pure and, especially when precipitated from seawater, is always contaminated with Mg2þ ions, even to a considerable percentage, to say
nothing of other cations and even other anions. The concentrations/activities of these ions are temperature and pressure dependent at equilibrium. Now we must also ask about the shape and size of the CaCO3 crystals. The surface energy is always less than the internal energy of a crystal so that final equilibrium is one large crystal. Again the degree of incorporation of impurities is different on different surfaces and all surfaces are different in this respect from the interior of a crystal. In most circumstances we do not observe true equilibrium, one single crystal, but a mass of small crystals and this leads us to the consideration of the features of mineral formation approaching but not quite at equilibrium. It adds the complication that the surfaces can be a considerable controlling contributor to the whole. Clearly also the equilibrium condition is of one crystal shape but the shapes in organisms are controlled. Given that the equilibrium is dependent on so many factors we may well choose to just define equilibrium model inorganic mineral formation empirically. Concentrating attention on calcite grown from water solutions we should determine experimentally certain close to equilibrium parameters at different temperatures, pH and salt concentrations, assuming that all crystals are large enough in size and that ion exchange from and to the crystals and the solution are closely balanced. The crystallization from seawater can only be done at supersaturation so the equilibrium conditions can only be estimated. It is possible to vary the nature of ‘seawater’ experimentally in other ways, e.g. Mg2þ content. We can then obtain certain parameters that can be given approximately in a completely abiotic crystallization system. I have not been involved in such experimental work but it is easy to see that a multidimensional grid can be set up which can be used to solve unknown temperature conditions in which
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 33–44. DOI: 10.1144/SP303.3 0305-8719/08/$15.00 # The Geological Society of London 2008.
34
R. J. P. WILLIAMS
a given crystal has grown ideally at equilibrium using a basic set of analytical data including impurity (Mg2þ, Naþ and certain anions) and isotopic (Ca, C, O, Mg etc.) contents. In principle the unknown, the temperature of growth, is then determinable provided variables such as to the pH and salt concentrations at which the crystals were formed is known. We have used this approach in defining the nature of CaCO3 in a study of this mineral using literature data on preparations of inorganic minerals and experimental data for coral and foraminifera (Chang et al. 2004). I turn to some observations below after an examination of the more difficult kinetic factors controlling the rate of nucleation and growth of crystals. (Note that supersaturation is an unstable condition and strictly-speaking comparisons in such conditions are so sensitive to impurities as to make an accurate grid of the above kind difficult to obtain.)
Rate control Unfortunately all mineralization is made complicated in that the crystals grow at a finite rate from a supersaturated solution where growth depends upon initial nucleation as well as increase in crystal size (Crick 1986; Frau´sto da Silva & Williams 1991; Mann 2001). Very high supersaturation is necessary for CaCO3 nucleation in the absence of surfaces. The CaCO3 that forms is then likely to be initially amorphous, and far from equilibrium since growth is very rapid after its nucleation. Parameters of such material may be very variable and little can be learnt from them. At lower degrees of supersaturation the problems of nucleation and subsequent growth have to be separated and I treat nucleation first assuming that the degree of supersaturation is fixed though this is clearly dependent on salt and impurity concentrations, pH, pressure and temperature. Nucleation has an activation free energy barrier since any initial crystal has a very high surface:internal volume ratio. So the rate of nucleation is dependent upon fluctuations, the chance that say 100 ions of bind together when further Ca2þ and CO22 3 growth is down-hill in energy. You may say that this is of no consequence, as we shall only analyze large crystals where much of the material is only incorporated at a later stage. However, observations show that, in fact, all nucleation is dependent on surfaces of particles, dust or on the containing vessel and the surface decides not just the size of the crystals and their shape but even the allotrope which forms, that is vaterite, aragonite or calcite and, hence, what is observed depends critically on nucleation. Now why should this be, when calcite is the most stable? I believe this is especially problematic in organism mineralization.
The height of the barrier to nucleation in the absence of a surface is calcite . aragonite . vaterite, hence raising supersaturation considerably, before the point of formation of an amorphous precipitate, will lead to vaterite not calcite (since large degrees of supersaturation favour the most soluble form). However the surface predicates what is observed by lowering nucleation barriers selectively so that the presence of a particular surface leads to crystallization of any particular one of the three allotropic forms even at lower degrees of supersaturation. We can understand this in terms ions of the different bindings of Ca2þ and CO22 3 on different surfaces leading to different Ca2þ/ packing in very small nucleating bodies. It CO22 3 is entirely possible that other foreign ions such as Sr2þ, of similar size to Ca2þ, will assist this nucleation while others such as Mg2þ, much smaller than Ca2þ, may seriously inhibit it. The coordination geometry of Sr2þ and Ca2þ are similar while that of Mg2þ and Ca2þ are quite different. Mg2þ is most stable in a highly symmetric distribution of six atoms of anions or water around it while Ca2þ and Sr2þ are usually to be found in seven or eight coordination sites of low symmetry, see MgO, CaCO3 and SrCO3. These differences are common to binding proteins so that these polymers can, and do, control ion binding and migration, e.g. by pumping selectively (Williams 1999). Other factors such as salt, pressure and temperature also affect nucleation. The surfaces and filaments of organisms cause very selected crystal forms, see below. Let us assume that nucleation of a small calcite crystal has been achieved, then growth occurs especially on particular planes of crystals and hence the crystal form may not be that of an equilibrium crystal. This preferential shape of a crystal depends on impurities and surfaces as well as on temperature and pressure. Growth is maintained by supply of ions above saturation so supply is extremely important. The supply from seawater is very large so that, once nucleated, we have to consider what stops growth especially in a deposit of fixed structure. This cessation of growth is most apparent in biological shells and is due to organic material binding to crystal surfaces. One consequence is that most shells are made of very small crystallites and surfaces are of considerable importance to growth.
Inorganic precipitates: experimental observations Seawater has different physical and chemical properties in different locations, which must be given in any study (Table 1). It contains about 50 times more Mg than Ca at 1 1023 M. Two different
SOME FUNDAMENTAL FEATURES OF BIOMINERALIZATION
35
Table 1. Sample sources Sample Seawater(IAPSO2) Seawater Seawater Coral (Acropora) Coral (Acropora Pocillopora) Coral (Acropora) Foraminifera Foraminifera Foraminifera Foraminifera Foraminifera Foraminifera Foraminifera
Core
Depth (m)
– – – – – – BOFS5K RC8-39 28K RCIO-140 MD85 87KG KNR110-80BC
– – – – – – 3547 4330 4900 1679 3001 1782 2952
Location
SST1 (8C)
North Atlantic3 Mediterranean3 Atlantic3 – c. 27.5 Barbados3 Red Sea3 222 Mauritius3 c. 25.5 N. Atlantic (50841.30 N, 21851.90 W) 14 12.5 S. Pacific (428530 S, 428210 W) N. Atlantic (248N, 228W) 22.5 W. Eq. Pacific (28390 S, 1568590 E) 29 0 0 S. E. Atlantic (29842 S, 12856 E) 21 25 Arabian Sea (23835.330 N, 64813.020 E) N. Atlantic (4837.860 N, 43839.030 W) 27.5
1
Typical average for sea surface temperature at locality from Levitus 94. Standard seawater from Ocean Scientific International Ltd, Southampton. Salinity: 34.998. Batch: 35NI. 3 No further information about longitude and latitude of localities. 2
properties of precipitates of CaCO3 obtained from seawater have been analyzed repeatedly under conditions close to equilibrium (Chang et al. 2004 and references therein). They are the magnesium content and the isotope ratios of which we discuss here only those of Ca and Mg and not of carbonate. The Mg2þ content of calcite is found to be close to 80 mmol/ mol Ca at 258. The content of Mg2þ in aragonite is much lower with a value close to 4 mmol/mol Ca. It is known that calcite can take up much larger quantities of Mg2þ in dolomite and that the uptake is sensitive to many parameters, for example it is sensitive to temperature (see Fig. 1), pressure and salt. The isotopic composition of calcium in aragonite and calcite precipitated from seawater, shows a lower content of the heavier 44Ca relative to 42Ca of some 20.6‰ to 20.7‰ as compared with the ratio in seawater (Gussone et al. 2003). Fewer measurements of 26Mg relative to 24Mg show a lowering of 21.0 to 21.1‰ relative to the ratio in seawater (Chang et al. 2004). This difference is consistent with the difference in absolute masses of Mg and Ca. We need to account for the negative values. Isotopes in compounds have no chemical differences but the mass differences lead to differences in zero point energy. The light isotope has the higher frequency, higher zero point energy, and is found to show the greater ability to go to the less stable condition following the heat content change: highly stable state
lowly stable state
As the difference in zero point energy will be greater, i.e. more positive, in the most stable condition favouring the heavy isotope. From this rough point of view we consider the formation of CaCO3. The overall
equilibrium condition is based here upon the free energy change of the reaction: 2 Ca(H2 O)2þ n þ CO3 (H2 O)m
CaCO3 þ (n þ m)H2 O where n and m are integers in the range n ¼ 6–8, m ¼ 4–8. Analysis shows that it is the large entropy change which favours CaCO3 and the heat change is unfavourable to CaCO3. Thus as expected, it is the light isotope of calcium that more readily enters CaCO3 and the heavy isotope content is correspondingly reduced (negative change). The same description applies to Ca in both aragonite and calcite, and they have the same isotope fractionation, and even more forcefully applies to the Mg isotopes in MgCO3 but the actual larger positive heat term for Mg2þ uptake into CaCO3 is not known accurately. (The analysis is most easily made starting from the gaseous ions in all cases.) In passing note that the kinetic selection of isotopes in a chemical reaction going over an opposed activation energy (kinetic) barrier will also favour the light isotope as the vibration frequency is higher in the ground state for the lighter isotope. The difference in this energy in the transition state is very small and the ground state difference therefore gives a greater positive fractionation at equilibrium in a stable product.
Biologically-controlled mineralization Immediately we can see there are possible novel factors that a biological system provides as
36
R. J. P. WILLIAMS
Fig. 1. Some data concerning the temperature dependence of magnesium in calcium carbonate.
potential controls (Frau´sto da Silva & Williams, 1991; Mann 2001). They are surfaces, control of supersaturation in controlled volumes, control of impurity and salt concentrations, and of pH. These will be considered in turn. There is no biomineralization by organisms that is not under some organism control, since all nucleation of it occurs on a biological surface, often a protein or a polysaccharide. Its particular biological selectivity decides if calcite or aragonite appear. (It is not often the case that the supersaturation reaches that required for vaterite formation.) If we assume that the crystals grow on the surface directly from seawater at a given pH and if we wish to know the unknown temperature at which they were grown then, we need to observe several factors including the size and shape of the crystals since empirical model studies tell us that these parameters may affect the very analytical chemical content that we wish to use to estimate temperature. (We shall assume the crystals are large enough so that bulk internal considerations, rather than nucleation or surfaces, control the chemical content and that growth has been slow, though this may not be true.) The parameters we shall measure here are isotope ratios and Mg2þ content (possibly also Sr2þ content, and
that of other impurities and their isotopes). Before turning to examples we need to make some observations about all cells independent from whether or not they are associated with mineralization (Fig. 2). The general ways in which Ca2þ and Mg2þ are handled in cells are shown in Figs 3 and 4. Note that in seawater both Ca2þ and Mg2þ ions are pumped out of all cells, Mg2þ, 50 –fold; Ca2þ ¼ 104 –fold.
Cell properties All cells have a cytoplasmic concentration of about 1023 M Mg2þ and of about 1027 M Ca2þ concentration. In seawater Ca2þ concentration exceeds 1023 M while that of Mg2þ approaches 1021 M. However while cells do not usually accumulate Mg2þ in vesicles the content of Ca2þ there can be at least as high as 1023 M allowing precipitation. Thus there are different mechanisms for handling the two ions in the cells before crystallization. Given these complications, as well as those stated above, it is then simplest to consider extreme cases where we know that mineralization occurs only on the surface of the organisms (as is the case of coral) or only in vesicles (as is true of
SOME FUNDAMENTAL FEATURES OF BIOMINERALIZATION
37
Fig. 2. The uptake of elements into a eukaryote cell, which is a large cell with numbers of compartments, evolving about 2 billion years ago. The functions of metal ions are extremely varied, controlled by gene expression of proteins, P, including pumps, and in part they control genetic expression by feedback to the gene. Some ions, especially Ca2þ and Mn2þ, are often exported or put into vesicles while Mg2þ is also pumped out of cells in seawater. Details of the functions and activities of all the ions are given in Frau´sto da Silva & Williams (1991).
coccoliths). In Sections about coral aragonite and Foraminifera, description of the sample sources are in Table 1 and the Mg content and isotope data are in Table 2.
at close to inorganic equilibrium but as it forms on a surface it is not exactly comparable with abiotic precipitates. Growth is stopped by organic excretions from the cells as far as we know.
Coral aragonite
Coccoliths
The indications are that control over the nucleation is important in coral since some produce aragonite, others calcite. The corals do not appear to control subsequent growth in that the minerals do not have a shell-like biological form. We take the Mg content of coral CaCO3 first and will assume that seawater universally has a Ca2þ content of 1023 M while that of Mg2þ is 50 1023 M. The Mg-content of coral aragonite is around 4 mmol/mol, which is very similar to that known for inorganic aragonite (Table 2). The 44/40Ca isotope fraction relative to seawater is close to 20.6‰ and that for 26/24Mg is about 21.0‰ both very similar to that in inorganic crystals. We conclude that coral CaCO3 is formed
The second example chosen is the coccolith. The coccolith biological shell is based on extreme biological control. It is quite remarkably a set of single crystal calcite plates the growth of which is an extraordinary architectural feat. In one species, the architecture which is species-dependent, looks somewhat like a wheel with rims and some dozen spokes, implying timed particular controlled growth on different crystallographic axis (Mann et al. 1989; Mann 2001). I initiated some of this investigation but Mann must be given the credit for the elucidation of the nature of the single crystal structure. Parallel extremely controlled crystal growth forms are seen in many other
38
R. J. P. WILLIAMS
Fig. 3. The distribution of calcium ions in a eukaryotic cell. Note especially transport to and from the environment and into many organelles and vesicles. In part of one such vesicle, see ER, calcite is laid down in coccoliths. Ca2þ ions are constantly entering the cytoplasm in signalling of events and leaving again via pumps. ER, endoplasmic reticulum; MITO, mitochondria; chloro, chloroplast.
relatively simple organisms, e.g. the SrSO4 of Acantharia (Wilcock et al. 1988) but note the amorphous silica networks of diatoms. The coccolith crystal plates grow internally in vesicles on particular crystal faces before being deposited externally. Hence the components of the crystals are handled in a series of steps. In the case of the coccolith there is calcium export from the cytoplasm to the environment and there is export from the cytoplasm to the vesicle followed by directed contained growth on surfaces. The calcium ion therefore moves from the sea, 1023 M, through the cytoplasm, 1027 M, and then to the vesicle at .1023 M to form a nucleus and then in a deliberate crystalline growth pattern. This means that the calcium ion suffers outward pumping at the outer membrane and inward pumping at the second membrane. It also diffuses inward and is deliberately allowed in to activate the cell. We expect the two pumps to have not very different isotopic fractionating powers. We explain that the 44/40Ca isotope fractionation, which is larger, 22.15‰ (Table 3), than that from simple inorganic fractionation from seawater (20.7‰; Gussone et al. 2003), as follows. The calcium of the cytoplasm is probably affected by inward diffusion of seawater ions added to by stimulated activity pulses (Fig. 3). It is then inward pumping into vesicles that increases the light isotope fractionation relatively before the
Fig. 4. The distribution of activities of the Mg2þ ion which is largely excluded from vesicles e.g. the reticula and the ER but has very many activities in the cytoplasm. Note that pumping can be outward to seawater but is inward in fresh water to maintain the 1023 M concentration in all cells. A-F, different proteins. ATP acts at pumps.
Table 2. Mg and Ca isotope and Mg/Ca analyses of seawater, reef corals and perforate Foraminifera Species
T2 (8C)
d26Mg3 (‰)
2s
d44Ca (‰)
Mg/Ca4 (mmol/mol)
N. Atlantic Mediterranean Atlantic Barbados Red Sea Red Sea Mauritius Mauritius BOFS5K RC8-39 RC8-39 28K RC10-140 87KG 87KG BOFS5K BOFS5K MD85 MD85 MD85 MD85 BOFS5K BOFS5K BOFS5K MD85 MD85 MD85 MD85 RC8-39 RC8-39 87KG 87KG 87KG KNR110-80BC RC10-140 RC10-140
– – – 27.5 22 22 25.5 25.5 14 12.5 12.5 22.5 29 25 25 14 14 21 21 21 21 14 14 14 21 21 21 21 12.5 12.5 25 25 25 27.5 29 29
2.60 2.65 2.72 1.72 1.67 1.72 1.65 1.80 21.40 22.14 22.79 21.78 21.56 21.68 – 22.35 22.05 21.54 – – – 22.31 – – 22.69 – – – 22.24 22.10 21.77 21.53 – 22.14 21.39 –
0.15 0.13 0.13 0.15 0.15 0.15 0.43 0.43 0.45 0.45 0.42 0.45 0.45 0.42 – 0.43 0.45 0.43 – – – 0.45 – – 0.43 – – – 0.43 0.45 0.43 0.43 – 0.43 0.43 –
0.92 – – 0.39 0.24 0.16 0.33 0.37 0.37 0.30 0.34 0.28 0.29 0.31 – 0.42 0.32 0.34 – – – 0.38 – – 0.28 – – – 0.30 0.30 0.31 0.28 – 0.27 0.36 –
– – – 3.44 3.82 2.84 4.58 3.57 – – – – – 2.54 2.28 1.75 – 1.95 1.99 1.59 1.89 – 1.78 1.32 1.74 2.20 2.00 2.42 1.17 – 3.61 3.76 3.64 3.64 3.92 4.41
39
(Continued)
SOME FUNDAMENTAL FEATURES OF BIOMINERALIZATION
– – – Acropora palmata Pocillopora sp. Acropora sp. Acropora sp. Acropora sp. Globogerina bulloides Globogerina bulloides Globogerina bulloides Globorotalia conglobatus Globorotalia conglobatus Neogloboquadrina dutertrei Neogloboquadrina dutertrei Globorotalia hirsuta Globorotalia hirsuta Globorotalia hirsuta Globorotalia hirsuta Globorotalia hirsuta Globorotalia hirsuta Globorotalia inflata Globorotalia inflata Globorotalia inflata Globorotalia inflata Globorotalia inflata Globorotalia inflata Globorotalia inflata Globorotalia inflata Globorotalia inflata Globorotalia menardii Globorotalia menardii Globorotalia menardii Globorotalia menardii Globorotalia menardii Globorotalia menardii
Core1
40
Table 2. Continued Species
T2 (8C)
d26Mg3 (‰)
2s
d44Ca (‰)
Mg/Ca4 (mmol/mol)
87KG KNR110-80BC RC10-140 28K 28K 28K 28K 87KG 87KG 87KG KNR110-80BC KNR110-80BC RC10-140 RC10-140 RC10-140 87KG RC10-140 RC10-140 28K KNR110-80BC RC10-140 BOFS5K KNR110-80BC KNR110-80BC MD85
25 27.5 29 22.5 22.5 22.5 22.5 25 25 25 27.5 27.5 29 29 29 25 29 29 22.5 27.5 29 14 27.5 27.5 21
22.00 22.35 21.43 – 21.62 – 21.36 21.36 – – 21.59 – 22.07 22.04 – 21.93 22.06 22.12 22.35 21.89 – 21.71 22.71 – 21.88
0.42 0.45 0.42 – 0.43 – 0.42 0.42 – – 0.43 – 0.43 0.42 – 0.45 0.43 0.45 0.43 0.45 – 0.45 0.45 – 0.43
0.40 0.27 0.36 – 0.29 – 0.31 – – – 0.36 – 0.26 0.26 – 0.39 0.27 0.27 0.30 0.30 – 0.32 0.35 – 0.40
– – – 3.23 2.19 2.58 – – 4.48 4.09 3.84 4.05 4.94 – 4.88 – 3.59 – 1.97 – 3.23 – – 3.90 1.99
R. J. P. WILLIAMS
Globorotalia menardii Globorotalia menardii Globorotalia menardii Globorotalia menardii Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (with sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (with sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (with sac) Hastigerina pelagica Hastigerina pelagica Hastigerina pelagica Globigerinoides ruber Globigerinoides ruber Globigerinoides ruber Globorotalia truncatulinoides Globorotalia truncatulinoides Globorotalia truncatulinoides Globorotalia truncatulinoides
Core1
1
MD85 RC8-39 BOFS5K BOFS5K 28K 28K 28K 87KG 87KG KNR110-80BC BOFS5K RC10-140 RC10-140 87KG RC10-140 87KG 87KG 87KG 87KG MD85 28K 28K 87KG 87KG 87KG KNR110-80BC KNR110-80BC
21 12.5 14 14 22.5 22.5 22.5 25 25 27.5 14 29 29 25 29 25 25 25 25 21 22.5 22.5 25 25 25 27.5 27.5
– 22.48 – – – 21.05 21.17 20.78 – 21.10 – – – – – – 21.81 – – 21.66 21.80 – 21.36 – – 21.59 –
Information about location and depth given in Table 1. Typical sea surface temperature at locality. Expressed relative to the SRM980 Mg and SRM915a Ca isotope standards. The 2s error of d44Ca is +0.13‰. 4 Determined by MC-ICP-MS. External reproducibility +0.6 mmol/mol (2s). 2 3
– 0.43 – – – 0.43 0.45 0.43 – 0.43 – – – – – – 0.45 – – 0.42 0.45 – 0.42 – – 0.43 –
– 0.40 – – – 0.34 0.21 0.36 – 0.29 – – – – – – 0.43 – – – 0.23 – 0.31 – – 0.36 –
1.83 1.56 1.49 1.53 3.25 7.64 – 7.29 8.82 – 5.86 3.60 3.56 3.56 2.89 2.80 – 2.48 1.94 – – 2.58 – 4.448 4.09 3.84 4.05
SOME FUNDAMENTAL FEATURES OF BIOMINERALIZATION
Globorotalia truncatulinoides Globorotalia truncatulinoides Globorotalia truncatulinoides Globorotalia truncatulinoides Globorotalia truncatulinoides Orbulina universa Orbulina universa Orbulina universa Orbulina universa Orbulina universa Orbulina universa Pulleniatina obliquiloculata Pulleniatina obliquiloculata Pulleniatina obliquiloculata Sphaeordina dehiscens Globorotalia siphonifera Uvigerina sp. Uvigerina sp. Uvigerina sp. Cibicidoides wuellerstorfi Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (with sac) Globigerinoides sacculifer (no sac) Globigerinoides sacculifer (with sac)
41
42
R. J. P. WILLIAMS
Table 3. Calcium light isotope enrichment 40
Tissue CaCO3 of foraminifera formed directly from seawater. CaCO3 of coccoliths formed by internal precipitation of calcium in vesicles. Ca2(OH)PO4 of deer bones. Note passage of calcium through many steps in plants and animals.
Ca/Ca44 enrichment (‰) 1.00 2.15 2.50
Data are from Zhu & Macdougall (1998) and De La Rocha & DePaolo (2000).
formation of calcite itself, which is known to favour the light isotope in inorganic crystals. Pumping them more than doubles the fractionation. This will follow if pumping has an opposed heat term (as has crystallization) when the light isotope is preferentially pumped. These steps may have a small temperature dependence. It would be good to know the isotope composition of the cytoplasmic calcium. A more striking observation is the Mg content of the coccolith which is extremely low, 0.1 to 0.2 mmol/mol Ca. Now Mg2þ ions are reduced somewhat in the cytoplasm (1023 M) relative to seawater (nearly 1021 M) by pumping out but in effect are altogether excluded from coccolith vesicles. It is this exclusion of Mg2þ which is typical of in-vesicle growth of many minerals, but note that the content here is close to the order of a thousand times less than the Mg content of inorganic calcite and is much less than expected from equilibria with the cytoplasm 1023 M Mg2þ. We have no available data on Mg isotopes in coccolith shells but we shall not be surprised if they are very different from those in inorganic calcite.
Foraminifera Before turning to these organisms we note dominant features of totally-controlled not just nucleated minerals (as illustrated by coccoliths) as compared to inorganic precipitation from the same solution, seawater (as illustrated by corals) but controlled in form by surfaces. The points to note in the case of the coccoliths are: (a) The morphology of the shell is fixed and species dependent. (b) Individual crystal growth is on defined axes. (c) The foreign ion contamination is very low (Mg2þ in calcite). (d) The isotope fractionation is accentuated. The existing data under these headings led Elderfield to write a note on “Carbonate Mysteries” (Elderfield 2002). The data for foraminifera are that: (a) the morphology of the shells is fixed but species specific;
(b) the Mg2þ content is low, not very low, and species related, see Table 2; (c) isotope fractionation of Mg is different (larger) than in coral and species samples selective, see Table 2. Sample data are given in detail, and for many identical species, as it is necessary to have confidence in what is a relatively recent mode of analysis. Variation between individual organisms is seen to be possible due to minor genetic variations in populations. Note in particular that in the species Orbulina universa the Mg2þ content and the isotopic fractionation are most unusual relative to all others and in the same direction – closer to inorganic crystals (Fig. 5; see also Fig. 1). However we also note that the Ca isotope data of the calcite are not readily distinguishable from those of an inorganic precipitate and that the Mg content is at least 10-fold greater than in coccoliths. How can we explain the data? We must accept that there is a high degree of biological control but it is not to the same degree as that in the coccolith. Special attention is drawn to the Mg2þ content and the Mg isotope fractionation as they are useful parameters. The solution from which the calcite is formed cannot be simply seawater. Given that the shells have speciesselected shape and some features of the Mg data are species specific we conclude that the shells must be grown in a biologically controlled volume, a vesicle, of some kind. Now the Mg2þ content of the calcite is lowered by some 40 to 100 fold from that in inorganic calcite values. This is close to the expected concentration difference between calcite grown not in seawater, 50 mM, but at cytoplasmic Mg2þ concentrations, 1 mM. The suggestion is clearly that Mg2þ is pumped out of the solution in which the calcite is formed. The isotope fractionation indicates that this is an uphill process with stronger binding to the membrane on the vesicle inside and weaker on the cytoplasmic outside so that energy input from ATP, adenosine triphosphate, will cause the heavy isotope to be removed more. A possible solution is that the vesicle in which the calcite grows is not similar to that in coccoliths where the vesicle is formed from the endoplasmic
43
δ
δ
SOME FUNDAMENTAL FEATURES OF BIOMINERALIZATION
Fig. 5. The species dependencies of the Mg and Ca isotope ratios in a variety of foraminifera. Note Orbulina universa. Some species appear to show a wide variety of values. Mg and Ca isotope ratios of seawater and coral are also given for comparison.
reticulum, ER, and it grows slowly as calcite is formed. In the coccolith vesicle magnesium is not pumped from the vesicle but cannot enter to seawater levels while calcium is pumped in. In the foraminifera it is possible that the vesicle is an enclosure formed from seawater and some extra calcium is pumped in while the magnesium is pumped out to cytoplasmic levels. Alternatively the vesicle is formed from the ER but allows in
Mg2þ to cytoplasmic levels. Depending on the temperature and the species the leakage of seawater to the vesicle will increase the Mg content of calcite as shown in Figure 1. It is possible that one part of the membrane of the vesicle is very close to the outer cytoplasmic membrane. It should be possible to determined the origin of the calcite, whether it is from seawater or not, by an analysis of all the trace elements in it and by comparison with
44
R. J. P. WILLIAMS
coccolith and coral trace element content. It would be helpful to know the isotope composition of cytoplasmic magnesium.
for many exchanges with the above coworkers, particularly Dr Chang. The author is a chemist and apologises for any mistakes with the use of conventions well-known to micropalaentologists but unknown to him.
Genetic variation There is an important feature of genetics which we must have in mind as it could make for difficulty when using biologically-controlled parameters of mineralization to uncover the history for example of the temperature at which the crystals grew. While it is clear that it is essential to use one species in the analysis there is the complication of genetic drift within a species due to a series of single base changes. Such changes could affect directly the Ca or Mg pumps or they could adjust more remote protein functions. It is for this reason that the number of samples given in Table 2 is so large and apparently repetitive. It could be that some of the differences are real. For a perfect parameter characterisation there should be mineral data from organism samples with identical DNA sequences. This problem lies on top of changes with time of pH, CO2, temperature or salt concentrations.
Other organisms The uptake of trace elements and both bulk and trace element isotope distribution into many biological minerals have been studied and the extremes of behaviour between systems in which the mineral is close to being in equilibrium with the external fluid, e.g. bone, (Table 3) and some shells, and those which are formed from cytoplasmic fluids as above have been noted. The isotopic fractionation is usually in favour of the lighter metal isotope in precipitates. The indication of a rate controlled process is in the magnitude of isotope fractionation. The insolubility of most calcium salts is entropy driven against a heat uptake (giving an explanation of this selection but not of the magnitude of these data.) Temperature dependence is expected for all trace element and isotope data but it is often confounded by the uncertainty of other environmental changes and their effect on ionic activities. This is not to say that an effective method cannot be found for following temperature at different historical times using metal data from minerals, but great care will be essential. All the data presented here were obtained in the Earth Sciences Department at Oxford. The measurements were directed by Prof. K. O’Nions and the experimental work was carried out jointly by V. T.-C. Chang, A. Makishima and N. S. Belshaw. The interpretation of the data is my own. Full details of all experimental results are given in Chang (2002) and full analytical information of some 80 organisms are given (Tables 1 and 2). I am most grateful
References B URTON , E. A. & W ALTER , L. M. 1991. The effects of pCO2 and temperature on magnesium incorporation in calcite in seawater and MgCl2-CaCl2 solutions. Geochimica et Cosmochimica Acta, 55, 777– 785. C HANG , V. T.-C. 2002. Mg and Ca isotope fractionation during CaCO3 biomineralisation. PhD thesis, Oxford University. C HANG , V. T.-C., W ILLIAMS , R. J. P., M AKISHIMA , A., B ELSHAW , N. S. & O’N IONS , R. J. 2004. Mg and Ca isotope fractionation during CaCO3 biomineralization. Biochemistry and Biophysics Research Reports, 323, 79–85. C RICK , R. E. 1986. (ed.) Origin, Evolution and Modern Aspects of Biomineralisation. Plenum Press, New York. D E L A R OCHA , C. L. & D E P AOLO , D. J. 2000. Isotopic evidence for variations in the marine calcium cycles over the Cenozoic. Science, 289, 1176– 1178. E LDERFIELD , H. 2002. Carbonate Mysteries. Science, 296, 1618–1620. F RAU´ STO DA S ILVA , J. J. R. & W ILLIAMS , R. J. P. 1991. The Biological Chemistry of the Elements. Oxford University Press, Oxford, 467– 496. (Note: not in the 2nd edition, 2001.) G USSONE , N., E ISENHAUER , A., H EUSER , A., D IETZEL , M., B OCK , B., B O¨ HM , F., S PERO , H. J., L EA , D. W., B IJMA , J. & N A¨ GLER , T. F. 2003. Model for kinetic effects on calcium isotope fractionation (d44Ca) in inorganic aragonite and cultured planktonic foraminifera. Geochimica et Cosmochimica Acta, 67, 1375–1382. L EA , D. W., M ASHIOTTA , T. A. & S PERO , H. J. 1999. Controls on magnesium and strontium uptake in planktonic foraminifera. Geochimica et Cosmochimica Acta, 63, 2369. M ANN , S., W EBB , J. & W ILLIAMS , R. J. P. (eds) 1989. Biomineralisation. VCH Weinheim, Germany. M ANN , S. 2001. Biomineralisation. Oxford University Press, Oxford. M UCCI , A. 1987. Influence of temperature on magnesium calcite overgrowths from seawater. Geochimica et Cosmochimica Acta, 51, 1977– 1986. N U¨ RNBURG , D., B IJIMA , J. & H EMLEBEN , C. 1996. Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperature. Geochimica et Cosmochimica Acta, 60, 803– 809. W ILCOCK , J. R., P ERRY , C. C., W ILLIAMS , R. J. P. & M ANTOURA , R. F. C. 1988. Crystallographic and morphological studies of the celestite skeleton of the acantharian species Phyllostaurus siculus. Proceedings of The Royal Society, London, B233, 393–405. W ILLIAMS , R. J. P. 1999. Calcium as a cell regulator. In: C ARAFOLI , E. & K LEE , C. (eds) Proceedings of The Royal Society, London, Oxford University Press, Oxford. Z HU , P. & M ACDOUGALL , J. D. 1998. Calcium isotopes in the marine environment and the oceanic calcium cycle. Geochimica et Cosmochimica Acta, 62, 1691–1698.
Vital effects and beyond: a modelling perspective on developing palaeoceanographical proxy relationships in foraminifera ¨ RBEL HO ¨ NISCH3,4, ABHIJIT SANYAL4, RICHARD E. ZEEBE1, JELLE BIJMA2, BA 5 HOWARD J. SPERO & DIETER A. WOLF-GLADROW2 1
School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii at Manoa, 1000, Pope Road, MSB 504, Honolulu, HI 96822, USA (e-mail:
[email protected]) 2
Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, D-27570 Bremerhaven, Germany 3
Marum, Bremen University, Leobener Strasse, 28359 Bremen, Germany
4
Lamont-Doherty Earth Observatory of Columbia University, Geochemistry Building, 61 Route 9W, Palisades, NY, 10964, USA 5
Geology Department, University of California, One Shields Avenue, Davis, CA 95616-8605, USA
Abstract: This paper mainly reviews our recent work on the biology and geochemistry of foraminifera with respect to their use as palaeoceanographic proxies. Our approach to proxy validation and development is described, primarily from a modeler’s point of view. The approach is based on complementary steps in understanding the inorganic chemistry, inorganic isotope fractionation, and biological controls that determine palaeo-tracer signals in organisms used in climate reconstructions. Integration of laboratory experiments, field and culture studies, theoretical considerations and numerical modelling holds the key to the method’s success. We describe effects of life-processes in foraminifera on stable carbon, oxygen, and boron isotopes as well as Mg incorporation into foraminiferal calcite shells. Stable boron isotopes will be used to illustrate our approach. We show that a mechanism-based understanding is often required before primary climate signals can be extracted from the geologic record because the signals can be heavily overprinted by secondary, non-climate related phenomena. Moreover, for some of the proxies, fundamental knowledge on the thermodynamic, inorganic basis is still lacking. One example is stable boron isotopes, a palaeo-pH proxy, for which the boron isotope fractionation between the dissolved boron compounds in seawater was not precisely known until recently. Attempts to overcome such hurdles are described and implications of our work for palaeoceanographic reconstructions are discussed.
Development and validation of palaeoceanographic proxy relationships in foraminifera have evolved rapidly over the past few years. During the early years of palaeoceanography, offsets from isotopic and elemental geochemical equilibrium that were attributed to life processes were often referred to as biological ‘vital effects’. In the case of stable carbon and oxygen isotopes, the black box was opened, resulting in a precise characterization of biological effects on geochemical signals recorded in the calcite shells of foraminifera. Interspecific variations have long been recognized in the stable carbon and oxygen isotope system (for review, see Wefer & Berger 1991; Spero 1998). However, the breakthrough in understanding inter- as well as intraspecific isotope variability came with culture experiments of live foraminifera under controlled
laboratory conditions (Bijma et al. 1998; BouvierSoumagnac & Duplessy 1985; Hemleben et al. 1985; Spero & DeNiro 1987; Spero & Williams 1988; Spero & Lea 1993, 1996) as pioneered by Be´ et al. (1977) and Hemleben et al. (1977). The profound consequences of controlled culture experiments for palaeoceanographic interpretations were widely recognized in 1997, when Spero and coworkers demonstrated that the seawater carbonate chemistry significantly affects d13C and d18O in planktonic foraminifera (Spero et al. 1997; Bijma et al. 1999). This phenomenon has been referred to as the ‘carbonate ion effect’. While palaeoceanographers had long been aware that temperature and seawater d18O affect foraminiferal d18O (Emiliani 1955; Shackleton 1967), another important player, the ocean’s CO2 chemistry, had to be added to the
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 45–58. DOI: 10.1144/SP303.4 0305-8719/08/$15.00 # The Geological Society of London 2008.
46
R. E. ZEEBE ET AL.
list. As a result, d18O-based temperature estimates are likely too low for geologic periods in the more distant past of high atmospheric CO2 concentrations and low oceanic pH (Zeebe 2001; Royer et al. 2004; Bice et al. 2006). With respect to quantitative modelling of life processes in foraminifera, a first attempt to open the ‘vital effect’ black box by means of a mathematical approach was provided by Spero et al. (1991). Their work outlined an abstracting concept, transforming geometry and fluxes in the living organism (Fig. 1) into mathematical equations which allowed the calculation of stable carbon isotope fractionation in a model foraminifer. But it was not until after the discovery of the carbonate ion effect that more sophisticated tools such as numerical models of the chemical microenvironment (Fig. 2) were developed to understand life processes, stable isotope fractionation, the carbonate ion effect and, prospectively, trace metal incorporation into foraminiferal calcite (Wolf-Gladrow & Riebesell 1997; Wolf-Gladrow et al. 1999; Zeebe et al. 1999; Zeebe 1999). Elderfield et al. (1996) proposed a Rayleigh distillation model for trace element incorporation into foraminiferal CaCO3, which is consistent with data in benthic but not in planktonic foraminifera. As suggested by Zeebe & Sanyal (2002), the process of metal incorporation is, particularly in the case of Mg2þ, most likely intimately intertwined with the energetics of the precipitation mechanism itself (cf. also Erez 2003). Development of a comprehensive theory of element incorporation in foraminifera by means of mathematical and numerical modelling is currently an active area of palaeoceanographic research. Parallel to the refinement of well-established proxy relationships such as d13C and d18O in foraminifera, other important geochemical proxies have been revitalized or newly developed over the past years, including metal/Ca ratios of Mg, Sr, U, Li and stable calcium and boron isotopes. Stable boron isotope ratios in foraminifera provide a tool for reconstructing the pH of ancient seawater (e.g. Spivack et al. 1993; Hemming & Hanson 1992; Sanyal et al. 1995; Pearson & Palmer 2000; Ho¨nisch & Hemming 2005). The biogeochemical and physicochemical aspects of this ‘palaeo-acidimetry’ proxy have been intensively examined over the past years by culture studies with live planktonic species, inorganic precipitation experiments, and theoretical means including ab initio molecular orbital theory (Sanyal et al. 1996, 2000, 2001; Ho¨nisch et al. 2003; Ho¨nisch & Hemming 2004; Zeebe et al. 2001, 2003; Zeebe 2005a). In this paper, we highlight some of our recent work on the development and validation of palaeoceanographic proxy relationships in foraminifera, primarily from a modeller’s point of view. ‘Vital
effects and beyond’ briefly describes the philosophy of our approach which is spelled out in terms of stable isotope fractionation in ‘Stable isotope fractionation’. In ‘Foraminifera dramatically alter their chemical and isotopic micro-environment’ we show that foraminifera strongly perturb their chemical and isotopic microenvironment, which has immediate consequences for the palaeoceanographic interpretation of stable isotopes in foraminifera from the fossil record. Application of our method to stable boron isotopes and a downcore reconstruction of Late Pleistocene glacialinterglacial cycles in surface ocean pH is presented in ‘Planktonic forminifera are reliable recorders of the ocean’s palaeo-pH’. Finally, ‘Foraminifera appear to control their shell-Mg/Ca ratio by a luxurious method’ describes our findings that foraminifera seem to control their Mg/Ca ratio by a rather expensive method in terms of energy requirements. The final section also points to several gaps in our understanding of biomineralization in foraminifera.
Vital effects and beyond: the approach Our approach to proxy validation and development is based on complementary steps in exploring the inorganic chemistry, inorganic isotope fractionation and biological controls on proxy relationships in organisms relevant to climate reconstructions. In many cases, the integration of laboratory experiments, field and culture studies, theoretical considerations and numerical modelling has turned out to be a successful method for this task. The foremost goal of this research is to improve climate reconstructions. Climate signals extracted from the geological record can be heavily overprinted by secondary, non-climate related phenomena because in the case of foraminifera, climate fluctuations are recorded by living organisms rather than by chemical compounds of inorganic origin. The long-term prospect of this work is to achieve refined palaeoceanographic interpretations of proxy relationships and to apply those relationships to the actual down-core record. The practical field application of our work to deep-sea sediment records has been documented by several authors of the present paper (e.g. Sanyal & Bijma 1999; Zeebe 2001; Spero & Lea 2002; Ho¨nisch & Hemming 2005).
Stable isotope fractionation A great deal of our recent efforts has been focused on developing a comprehensive theory of stable isotope fractionation in foraminifera, focusing on the elements of carbon, oxygen and boron. While the inorganic CO2 chemistry and isotope
PALAEOCEANOGRAPHICAL PROXY RELATIONSHIPS IN FORAMINIFERA
47
Fig. 1. Light microphotographs of the symbiotic, planktonic foraminifer Orbulina universa (top) and the non-symbiotic species Globigerina bulloides (bottom). After Spero (1998).
fractionation of carbon in the CO2-H2O-CaCO3 system has been rather well known for quite some time now (for summary, see Millero 1995; Zhang et al. 1995; Zeebe & Wolf-Gladrow 2001), this is not the case for oxygen (Usdowski & Hoefs 1993;
Zeebe 1999; Zeebe 2005b) and less so for boron. In the following, the inorganic chemistry and stable isotope fractionation of dissolved boron in aqueous solution is used as an example to illustrate the steps taken in the process of understanding the
48
R. E. ZEEBE ET AL.
Fig. 2. Schematic of life processes. Photosynthesis, respiration, and calcification perturb the chemical and isotopic microenvironment of the organism. The distance to the centre of the shell is denoted by r, while R1 and R2 refer to the radius of the foraminiferal shell and edge of symbiont halo, respectively. Dissolved inorganic carbon depleted in 13C is taken up during photosynthesis, while 13C-depleted CO2 is released during respiration (after Wolf-Gladrow et al. 1999; Zeebe et al. 1999).
thermodynamic basis of a proxy such as the d11B palaeo-pH proxy.
Inorganic chemistry Dissolved boron in seawater comes mainly in two forms 2 as boric acid, B(OH)3, and borate ion, B(OH)2 4 . The boric acid – borate equilibrium can be written as: þ B(OH)3 þ H2O B(OH)2 4 þH
with stoichiometric equilibrium constant KB: þ KB ¼ [B(OH) 4 ][H ]=[B(OH)3 ]
(1)
while the total boron concentration BT is given by BT ¼ [B(OH) 4 ] þ [B(OH)3 ]
(2)
The concentration of the dissolved boron species as a function of pH is shown in Figure 3a. There is little discussion regarding the chemical thermodynamics of the boron equilibrium (cf. Zeebe & Wolf-Gladrow 2001). A value frequently used for the dissociation constant of boric acid, pKB, is 8.60 at T ¼ 258C, S ¼ 35 (DOE 1994). It is also noted that at typical total boron concentration of c. 420
mmol kg21 in seawater, polynuclear boron species can probably safely be ignored. Cotton & Wilkinson (1988) state that polynuclear boron species are negligible at concentrations ,25 mM [see also Su & Suarez (1995) and references therein]. Using pK’s for the polynuclear B3 species given in Kakihana et al. (1977), the concentration of 211 M at typical total B3O3(OH)2 4 , e.g. is 3 10 seawater boron concentration. On the other hand, the kinetics of the boric acid –borate equilibrium are less well known. Yet in order to calculate fluxes, pH gradient and boron isotope distribution in the vicinity of a foraminifer, the kinetics (i.e. the speed of the conversion between the two dissolved boron species) is crucial. At the time when we developed the numerical models of the chemical microenvironment of foraminifera there was, to the best of our knowledge, no measured value for this rate constant available in the literature. The problem was eventually solved by considering sound absorption data in seawater, which is described in Zeebe et al. (2001).
Inorganic isotope fractionation Boron has two stable isotopes, 10B and 11B, which make up 19.82% and 80.18% of the total boron (IUPAC 1998). As can be seen in Figure 3a, at
PALAEOCEANOGRAPHICAL PROXY RELATIONSHIPS IN FORAMINIFERA
Concentration (μmol kg−1)
(a) 400 −
B(OH)3
300
B(OH)4
200 100 0
(b) 70
δ11B (‰)
60 B(OH)3
50 Sea water
40
−
B(OH)4
30 20 10
7
8
9
10
pH Fig. 3. (a) The concentration of dissolved boron species as a function of pH at T ¼ 258C, S ¼ 35, and total boron concentration of 416 mmol kg21 (DOE 1994). (b) Boron isotopic composition of B(OH)3 and B(OH)2 4 as a function of pH assuming a(B(OH)23 -B(OH)24 ) ¼ 1.030 (cf. Hemming & Hanson 1992; Zeebe 2005a).
low pH all dissolved boron is essentially boric acid, B(OH)3, whereas at high pH all dissolved boron is essentially borate ion, B(OH)2 4 . Because the stable isotope 11B is enriched in B(OH)3 compared to B(OH)2 4 , the isotopic composition of the dissolved species change with pH (Fig. 3b). At low pH the isotopic composition of B(OH)3 is equal to the isotopic composition of the total dissolved boron, 39.5‰. On the other hand, at high pH the isotopic composition of B(OH)2 4 is equal to the isotopic composition of the total dissolved boron. In between, the d11B of the two species increase. Based on the assumption that the charged species, B(OH)2 4 , is incorporated into foraminiferal calcite (Hemming & Hanson 1992), the d11B of calcite also increases with pH and a palaeo-pH proxy is created (Fig. 3b). One uncertainty regarding the inorganic basis of this proxy is the value of the thermodynamic equilibrium fractionation factor between B(OH)3 and 2 2 B(OH)2 4 , a(B(OH)3 -B(OH)4 ) or, in short, a(B32B4). Based on the theory of thermodynamic properties
49
of isotopic substances (Urey 1947), Kakihana & Kotaka (1977) calculated a(B32B4) ¼ 1.0193 at 300 K. Due to the absence of an experimental value, this theoretical value has been widely used in geochemical applications over the past 25 years or so. However, recent theoretical work suggests a larger fractionation factor. In 2005, two theoretical articles were published indicating a(B32B4) . 1.030 and a(B32B4) ¼ 1.027 at 300 K, respectively (Zeebe 2005a; Liu & Tossell 2005). These results were based on various theoretical, analytical methods and on numerical approaches using ab initio molecular orbital theory and point towards a larger value for a(B32B4), as also indicated by Oi (2000). Thus, theory predicts a value for the boron isotope fractionation factor between B(OH)3 and B(OH)2 4 at 258C of about 30‰ rather than 20‰. In fact, in the following year an experimental value of 28.5‰ was published (Byrne et al. 2006). It is emphasized that this is the fractionation factor between the dissolved boron compounds in solution and is not to be confused with fractionation factors involving stable boron isotope ratios in carbonates. The latter is discussed in the next section. In that context, it is important to note that when a fractionation of 28.5‰ is used to calculate the isotopic composition of borate, the shape and inflection point of the borate curve does not match the shape of the empirical carbonate data (Fig. 4). The above example illustrates an aspect of a proxy relationship which requires more fundamental work because it is of basic, thermodynamic nature. Such hurdles need to be overcome by experimental and theoretical efforts. However, it would be erroneous to draw the general conclusion that a proxy approach whose inorganic basis is not yet completely understood is per se invalid. In the case of boron, e.g. uncertainties in a(B32B4) do not bias pH reconstructions provided that empirical organism-specific calibrations are used.
Biological controls Some organism-specific calibrations are shown in Figure 4. They include results from controlled culture experiments with the two planktonic foraminiferal species Globigerinoides sacculifer and Orbulina universa (Sanyal et al. 1996, 2001) and the coral species Porites cylindrica and two species of Acropora (Ho¨nisch et al. 2004; Reynaud et al. 2004). First, the d11B of boron incorporated into the biogenic carbonates of all these coral and foraminiferal species increase with pH. This is the basis of the d11B-pH proxy. Second, there are offsets between different groups and species. The corals appear to be isotopically heavier
50
R. E. ZEEBE ET AL.
30 –
B(OH)4 28
(Kakihana et al. 1977)
26
P. cylindrica A. nobilis (Hönisch et al. 2004) Acropora sp. (Reynaud et al. 2004) G. sacculifer (Sanyal et al. 2001) Synthetic calcite (Sanyal et al. 2000) O. universa (Sanyal et al. 1996)
d11B (‰)
24
22
20
18 –
B(OH)4 with e = 30‰ (Zeebe 2005) –
B(OH)4 with e = 28.5‰ (Byrne et al. 2006)
16 –
B(OH)4 with e = 27‰ (Liu & Tossell 2005) 14 7.5
8
8.5
9
pH (SW scale) 11
Fig. 4. Empirical relationships between d B and culture seawater pH measured in three species of corals (Porites and Acropora), two species of planktonic foraminifera (Globigerinoides sacculifer and Orbulina universa), and inorganically precipitated calcite. Palaeoceanographic reconstructions use the empirical curves for reconstructing past seawater pH. The upper black solid, the dotted, and the dashed black line represent the d11B of B(OH)2 4 using a(B32B4) ¼ 1.019, 1.027, and 1.030 (Kakihana & Kotaka 1977; Liu & Tossell 2005; Zeebe 2005a). The dot-dashed 11 black line represents the experimental a(B32B4) ¼ 1.0285 of Byrne et al. (2006). Note that d B B(OH)24 and the d11B in carbonates are two different quantities and that one cannot be deduced from the other (see text).
(enriched in 11B) compared to the foraminifera. It is interesting to note that the coral skeletons consist of the CaCO3 polymorph aragonite, while the foraminifera G. sacculifer and O. universa produce calcite shells. The offset between the two foraminiferal species is about 2‰. It is discussed in the next section that changes in the microenvironment of foraminifera can lead to light/dark shifts in shell d11B. However, the offset between G. sacculifer and O. universa is difficult to explain with this mechanism (Zeebe et al. 2003). Also shown in Figure 4 are results for inorganically precipitated calcite (Sanyal et al. 2000) which falls between the foraminifera. In summary, the d11B-pH relationship has been found in the biogenic carbonates tested. The foraminifera are offset from the
inorganic calcite and the corals seem to be generally enriched relative to that. So far only the isotopic fractionation between a standard and the carbonates as a function of pH has been discussed. Now let us look at the dissolved species of boron in aqueous solution. The upper black solid, the dotted, and the dashed black lines in Figure 4 represent the d11B 11 ) as a function of pH of B(OH)2 4 (d BB(OH) 2 4 calculated using a(B32B4) ¼ 1.019, 1.027, and 1.030 (Kakihana & Kotaka 1977; Liu & Tossell 2005; Zeebe 2005a). The dot-dashed black line represents the experimental a(B32B4) ¼ 1.0285 of Byrne et al. (2006). It is obvious that no matter what the true value of a(B32B4) is, the is exclusively assumption that B(OH)2 4
PALAEOCEANOGRAPHICAL PROXY RELATIONSHIPS IN FORAMINIFERA
incorporated into the carbonates without further fractionation cannot hold for all biogenic and inorganic carbonates. If this assumption was correct, then all carbonates would fall on a single line and this line would be the d11B of 11 cannot be B(OH)2 4 . As a corollary, d B B(OH)2 4 deduced from the d11B of the carbonates and vice versa. In the future, some remaining issues of the d11B-pH proxy need to be addressed: (1) how does temperature, seawater salinity/composition, and total boron concentration affect the results for a(B32B4) published by Byrne et al. (2006); and (2) what causes the offsets between the d11B of the boron species in solution and in the carbonates. Meanwhile, neither of these questions compromises the use of d11B in carbonates as a palaeo-pH indicator.
Foraminifera dramatically alter their chemical and isotopic micro-environment If one considers an organism of the size of a foraminifer (R , 1 mm), being surrounded by a comparatively large volume of seawater, it may be difficult to see that the organism would have any significant influence on its environment. One would rather assume that the environmental properties the organism sees, and thus records in its shell, are dictated by the bulk seawater properties. Strictly, this is not the case. Although not independent of ambient conditions, the chemistry and isotopic ratios in the vicinity of the shell are primarily set by the properties within the diffusive boundary layer of the organism. The reason is that the typical length scale of the organism, R, is smaller than the so-called Kolmogorov scale, h, roughly the ‘size of the smallest eddie’: h ¼ (n3 =1)1=4
(3)
where n is the kinematic viscosity and 1 is the energy dissipation rate. In most parts of the ocean, h is typically .10 mm, and .1 mm in the wind mixed layer (Lazier & Mann 1989). This means that the transport within the boundary layer of a foraminifer, for instance, is dominated by slow molecular diffusion rather than rapid turbulent mixing. The diffusion time scale on the millimetre scale (L ¼ 1023 m) is given by t ¼ L 2/D ffi 1000 s where D ¼ 1029 m2 s21 is a diffusion coefficient characteristic for small molecules in seawater (note that diffusion on the length scale of single symbiotic algae, say a few micrometres, is much quicker). The combination of long diffusion time scales with high concentrations of symbionts and large rates of respiration and calcification of the
51
foraminifera can drastically alter the microenvironment of the organism. As a result, the signal it encounters can be significantly different from that of the bulk medium. But how different? Regarding foraminifera, this question has only recently been addressed using microsensor studies and numerical modelling (Jørgensen et al. 1985; Rink et al. 1998; Wolf-Gladrow et al. 1999; Zeebe et al. 1999; Zeebe et al. 2003). Figure 5 shows an example of a model experiment simulating a foraminifer under dark conditions (for details, see Wolf-Gladrow et al. 1999). Due to respiration, the CO2 concentration at the shell increases, while the pH drops (panels a and d). Model results agree well with microsensor pH transects (diamonds in panel d, B. B. Jørgensen, pers. comm.). It is interesting to note that microsensor observations of pH and our model simulations are generally in good agreement also with more recent pH electrode studies (Rink et al. 1998; WolfGladrow et al. 1999). However, recent microelectrode measurements of dissolved CO2 show smaller dark/light CO2 amplitudes at the shell surface than the model (Ko¨hler-Rink & Ku¨hl 2005). One possible explanation is that reaction rate constants as implemented in the model (based on inorganic chemistry) and those in the vicinity of the organism are different. Another is that in comparison to pH electrodes, the full potential of microsensor CO2 technology is yet to be reached. Nevertheless, the important message is that at typical radii and life process fluxes of planktonic foraminifera, a boundary layer with strong chemical gradients is developed (this is not necessarily the case for other plankton, cf. Wolf-Gladrow & Riebesell 1997). This leads to substantial differences between e.g. pH and O2/CO2 concentrations in the vicinity of the organism and the ambient seawater. For example, in symbiotic foraminifera under high-light conditions, O2 has been measured to be 2 to 2.5 times higher at the shell than in the bulk medium, while measured and simulated pH rises by more than 0.4 units at the shell (Jørgensen et al. 1985; Rink et al. 1998; Wolf-Gladrow et al. 1999; Ko¨hler-Rink & Ku¨hl 2005). During night time, and when ambient pH is lowered below typical seawater values ( pH , 7.7, HJS and JB, unpublished results), acidic conditions at the shell due to CO2 respiration can lead to calcification inhibition or actual dissolution of calcite chambers. But not only is the chemistry within the foraminiferal boundary layer drastically perturbed. Stable isotope ratios are also affected, which bears directly on palaeoceanographic interpretations of stable isotopes in fossil foraminifera. For example, preferential uptake of 12C during symbiont photosynthesis under light conditions leads to enrichment of
52
R. E. ZEEBE ET AL.
(a)
(b)
40
20
→ Distance to shell centre
0
1950 1900 1850 1800
(d)
400 350
8.4
pH
8.3
300
8.2
2−
[CO3 ] (µmol kg−1)
(c)
2000
R [HCO−3] (µmol kg−1)
[CO2] (µmol kg−1)
60
8.1
250 200
0
500
1000 r (µm)
1500
8
2000
0
500
1000 r (µm)
1500
2000
Fig. 5. Results of a diffusion-reaction model of the foraminiferal microenvironment under dark conditions. 2 (a) Respiration raises CO2 at the shell, while [CO32 2 ] and pH decrease (c, d). (b) The HCO3 pool is large and shows relatively small changes (, 7%) under dark conditions (this is different under high-light, Wolf-Gladrow et al. 1999). Microsensor pH transects (diamonds in panel d) were measured by B. B. Jørgensen and co-workers.
2
O. universa/HL O. universa/LL Model/HL200 Model/Dark
1 Δδ11B (º/oo)
shell-d13C by up to 1.5‰ (Spero & Williams 1988), in agreement with results of diffusion-reaction models which include stable carbon isotopes (Zeebe et al. 1999). Likewise, it is not difficult to imagine that stable boron isotope ratios at the shell are different under light vs. dark conditions, considering the substantial pH variations at the shell (Fig. 5d). In fact, controlled laboratory experiments and numerical modelling has shown this to be the case (Ho¨nisch et al. 2003; Zeebe et al. 2003). Figure 6 illustrates the differences in shell d11B in the dark and light, respectively. In the dark, pH at the shell is lowered, and, considering that d11BB(OH)24 decreases with pH (Fig. 3b), shell d11B is lowered as well 2 provided that B(OH)2 4 is preferentially incorporated into the calcite. The opposite applies to high-light conditions. Although the sign of the light/dark induced d11B shift is clear, its magnitude could not be calculated until the kinetics of the boric acid –borate ion reaction were understood. (Note that the boundary layer chemistry
− 0 Bulk B(OH)4
−1 8.11
8.16 8.21 pH Fig. 6. Light/dark induced shift in shell d11B of O. universa as measured in culture experiments (stars) and numerically simulated (diamonds). Dd11B is the difference relative to bulk d11BB(OH)24 at pH 8.16; HL ¼ High Light; LL ¼ Low Light. Experimental and modeled total dissolved boron concentration were 10 times elevated over natural seawater concentrations. In the model, a symbiont halo thickness of 200 mm was assumed (HL200).
PALAEOCEANOGRAPHICAL PROXY RELATIONSHIPS IN FORAMINIFERA
53
Fig. 7. Reconstruction of surface ocean pH over glacial cycles based on d11B (Ho¨nisch & Hemming 2005). (a) Past atmospheric CO2 concentration from the Vostok ice core (Petit et al. 1999). (b) d11B in Globigerinoides sacculifer from ODP core 668B in the eastern equatorial Atlantic, Sierra Leone Rise at 2693 m water depth (right axis, red symbols) superimposed on d18O of Globigerinoides ruber (left axis, black symbols). (c) Record of d18O of seawater, reflecting changes in global ice volume (Waelbroeck et al. 2002). Note that surface pH reconstructions track the glacial-interglacial climate oscillations in agreement with inferred changes from ice-core CO2.
is properly described by a steady-state of fluxes involving diffusion and reaction kinetics. It is not chemical equilibrium.) Taking advantage of our previous work on the boric acid –borate ion kinetics described in the ‘Inorganic Chemistry’ section (Zeebe et al. 2001), the light/dark induced d11B shift was calculated (Fig. 6). The model results using 10 times elevated total boron concentration (as in culture experiments) and a symbiont halo thickness of 200 mm match experimental results well (Ho¨nisch et al. 2003; Zeebe et al. 2003).
Planktonic foraminifera are reliable recorders of the ocean’s palaeo-pH Reliable proxies for the ocean’s CO2 chemistry are of utmost importance because they can provide information about past atmospheric CO2 concentrations and clues to the causes of carbon cycle variations. In turn, such information is crucial to comprehending feedbacks of Earth’s climate
system. For example, due to lack of adequate CO2 records, it is still controversial whether CO2 was the primary driver of the Cenozoic cooling trend. Moreover, we still lack a sound understanding of the link between glacial-interglacial changes in atmospheric CO2 and deep ocean chemistry. These problems need to be solved by reliable CO2system reconstructions. As described earlier, stable boron isotope ratios in foraminifera provide a tool for reconstructing the pH of ancient seawater (Spivack et al. 1993; Sanyal et al. 1995; Pearson & Palmer 2000; Ho¨nisch & Hemming 2005); combined with information on one other parameter of the carbonate system (e.g. CO22 3 , total CO2, or total alkalinity), past atmospheric pCO2 may be estimated (e.g. Tyrrell & Zeebe 2004). Efforts to understand and calibrate this proxy, including culture experiments, inorganic precipitation studies, and theoretical approaches hitherto indicate that planktonic foraminifera are reliable recorders of the ocean’s palaeo-pH (Sanyal et al. 1995, 1996, 2000, 2001; Ho¨nisch et al. 2003;
54
R. E. ZEEBE ET AL.
CaCO3 Origin
Planktonic foraminifera Abiotic marine calcite
Inorganic precipitates 10 mol % MgCO3
20
Fig. 8. Typical values of mole % MgCO3 in planktonic foraminifera, abiotic marine calcite, and inorganic precipitates from laboratory studies.
Ho¨nisch & Hemming 2004; Zeebe et al. 2001, 2003; Zeebe 2005a). Assuming the modern relationship between alkalinity and salinity remained constant over the course of the Pleistocene, Sanyal et al. (1995) and Ho¨nisch & Hemming (2005) translated their d11B -pH reconstructions (Fig. 7) and estimated alkalinities into aqueous pCO2 estimates, which quantitatively reflect atmospheric pCO2 reconstructions measured in ice cores (Petit et al. 1999; Siegenthaler et al. 2005). Whereas surface ocean pH estimates have never been questioned, boron isotope estimates of a dramatic glacial deep-sea increase in pH and [CO22 3 ] (Sanyal et al. 1995) could not be confirmed by sedimentary records of carbonate preservation and other geochemical proxy records. Deep sea pH estimates have therefore been criticized (Broecker & Henderson 1998). The major uncertainty of those estimates was the use of mixed benthic foraminifer species, which are likely to record a mix of pH conditions from pore and bottom waters. A recent sediment study now focusing on the use of the single epibenthic foraminifer species Cibicidoides wuellerstorfi, found glacial deep water pH in the Atlantic similar to or no higher than þ 0.08 pH units relative to interglacials (Ho¨nisch et al. in press). These new data no longer support the hypothesis of a much more basic deep ocean which could explain the entire glacial drop in atmospheric pCO2. More validation studies for the Pacific Ocean are underway but the studies mentioned above demonstrate that the boron isotope proxy is a useful tool, if carefully applied.
Foraminifera appear to control their shell-Mg/Ca ratio by a luxurious method Mg/Ca ratios in carbonates and seawater affect the thermodynamic equilibrium between solution and
the solid state, as well as the kinetics during crystal precipitation and dissolution. These properties are relevant for global carbon, calcium, and magnesium fluxes (Morse & Mackenzie 1990) and were likely important drivers of switches between Phanerozoic calcite and aragonite seas (Stanley & Hardie 1998). More recently, Mg/Ca ratios in foraminifera have received great attention because of their use as a palaeothermometer (e.g. Nu¨rnberg et al. 1996; Lea et al. 2000; Tripati & Elderfield 2005). Planktonic foraminifera seem to have strong control over their shell Mg concentration because the Mg/Ca ratio is significantly smaller than, for instance, of abiotic marine calcite or inorganically precipitated calcite in the laboratory (Fig. 8). The latter two fall in the category of high-magnesian calcites. Because Mg2þ is also known to be an inhibitor of calcite growth, one viable strategy of planktonic foraminifera to initiate calcite precipitation may be the removal of Mg2þ ions from the site of calcification. If there are other advantages to produce low- instead of high-magnesian calcite (related to thermodynamic stability, for instance), then Mg2þ removal would serve two purposes at the same time. Zeebe & Sanyal (2002) investigated such a scenario by means of inorganic precipitation experiments. The purpose of the experiments was to mimic the chemistry of a calcifying fluid during precipitation, analogous to a simple calcification scheme as depicted in Figure 9 (for a recent
Seawater/ Cytoplasm
Diffusion
Ion Transport
H2O, CO2
H+, Ca2+, Mg2+?
Membrane
Calcifying Fluid
Precipitation
CaCO3 Fig. 9. Simple calcification scheme of CaCO3 precipitation from a calcifying fluid. Organisms may control precipitation by separating a certain space from the ambient seawater by a membrane which is permeable to H2O and CO2 diffusion. Ion transport across the membrane may be mediated by Hþ-ATPase and Ca2þ-ATPase. Note that whether such transport systems are active in foraminifera remains to be tested; even less is known about magnesium transporters.
PALAEOCEANOGRAPHICAL PROXY RELATIONSHIPS IN FORAMINIFERA
review of biomineralization in foraminifera, see Erez 2003). The evaluation of the experimental results plus consideration of Hþ, CO2, and Ca2þ fluxes indicate that it is energetically much more efficient to initiate calcite precipitation by removal of protons, rather than Mg2þ ions (Zeebe & Sanyal 2002). This result is puzzling because the low Mg concentrations in planktonic foraminifera are then difficult to explain by considering costeffectiveness during ‘house building’. Of course, it is well known that the cheaper house is not necessarily the better one and other factors may be important for the low Mg/Ca ratios in planktonic foraminifera. Alternatively, calcification mechanisms could also involve Mg2þ-binding organic ligands. Such avenues should be explored in the future in order to solve the puzzle of biomineralization in foraminifera. Recent advances in measurement techniques allow analysis of ever smaller samples. Eggins et al. (2004) used high-resolution microanalysis techniques to study the spatial distribution of Mg in the final chamber of the planktonic foraminifer Orbulina universa. They found paired bands of low and high Mg/Ca ratios which were interpreted as diurnal growth bands reflecting pH changes in the foraminiferal microenvironment driven mainly by variations in photosynthesis and respiration of the symbionts. They speculate that the Mg banding may be accompanied by similar variations in oxygen, carbon and boron isotopes. If this is true and measurable (cf. Rollion-Bard 2005) it would open the door to investigating stable isotope variations at the sub-shell/sub-chamber scale of foraminifera. Element and isotope variations across a single foraminiferal shell pose a new challenge for models of biomineralization.
Conclusions In this paper we have primarily reviewed some of our recent work on developing and validating palaeoceanographic proxy relationships in foraminifera. Several aspects of the biology and geochemistry of planktonic foraminifera relevant to climate reconstructions have been explored in great detail over the last 10 years or so. A few conclusions derived from this research were discussed in the current paper: (1) foraminifera dramatically alter their chemical and isotopic micro-environment; (2) planktonic foraminifera are reliable recorders of the ocean’s pH; and (3) foraminifera appear to control their shell-Mg/Ca ratio by a luxurious method. Our results were obtained by a team effort, combining culture studies of live foraminifera, laboratory studies and theoretical work. In order to employ the full potential of
55
palaeoceanographic proxies that involve once living organisms, the approach outlined in the current paper has turned out to be successful in many cases. We conclude that future research should continue to employ this approach. David Lea, Ann Russell and the many students and research colleagues are acknowledged who participated in various field campaigns that produced the experimental data discussed in this paper. Experimental research described here was supported by the National Science Foundation and DFG (Palaeoprox: BI 432/3) and the European Commission (6C: EVK2-CT-2002-00135) to J. B. Support to R. E. Z. was provided by the National Science Foundation (NSF-OCE05-25647).
References B E´ , A. W. H., H EMLEBEN , C., A NDERSON , O. R., S PINDLER , M., H ACUNDA , J. & T UNTIVATE -C HOY , S. 1977. Laboratory and field observations of living planktonic foraminifera. Micropaleontology, 23, 155– 179. B ICE , K. L., B IRGEL , D., M EYERS , P. A., D AHL , K. A., H INRICHS , K.-U. & N ORRIS , R. D. 2006. A multiple proxy and model study of Cretaceous upper ocean temperatures and atmospheric CO2 concentrations. Paleoceanography, 21, PA2002, doi:10.1029/ 2005PA001203. B IJMA , J., H EMLEBEN , C., H UBER , B. T., E RLENKEUSER , H. & K ROON , D. 1998. Experimental determination of the ontogenetic stable isotope variability in two morphotypes of Globigerinella siphonifera (d’Orbigny). Marine Micropaleontology, 35, 141–160. B IJMA , J., S PERO , H. J. & L EA , D. W. 1999. Reassessing foraminiferal stable isotope geochemistry: impact of the oceanic carbonate system (experimental results). In: F ISCHER , G. & W EFER , G. (eds) Use of Proxies in Paleoceanography: Examples from the South Atlantic, Springer-Verlag, Berlin, 489–512. B YRNE , R. H., Y AO , W., K LOCHKO , K., T OSSELL , J. A. & K AUFMAN , A. J. 2006. Experimental evaluation of the isotopic exchange equilibrium 10B(OH)3 þ 11 11 B(OH)2 B(OH)3 þ 10B(OH)2 4 ¼ 4 in aqueous solution. Deep-Sea Research I, 53, 684 –688. B OUVIER -S OUMAGNAC , Y. & D UPLESSY , J.-C. 1985. Carbon and oxygen isotopic composition of planktonic foraminifera from laboratory culture, plankton tows and recent sediment: implications for the reconstruction of paleoclimatic conditions and of the global carbon cycle. Journal of Foraminifera Research, 15, 302– 320. B ROECKER , W. S. & H ENDERSON , G. M. 1998. The sequence of events surrounding Termination II and their implications for the cause of glacial-interglacial changes. Paleoceanography, 13, 352–364. C OTTON , F. A. & W ILKINSON , G. 1988. Advanced Inorganic Chemistry: A Comprehensive Text (5th ed.). Wiley, New York. DOE. 1994. Handbook of Methods for the Analysis of the Various Parameters of the Carbon Dioxide System in Sea Water; Version 2. In: D ICKSON , A. G. & G OYET , C. (eds) ORNL/CDIAC-74.
56
R. E. ZEEBE ET AL.
E GGINS , S. M., S ADEKOV , A. & DE D ECKKER , P. 2004. Modulation and daily banding of Mg/Ca in Orbulina universa tests by symbiont photosynthesis and respiration: a complication for seawater thermometry? Earth and Planetary Science Letters, 225, 411–419. E LDERFIELD , H., B ERTRAM , C. J. & E REZ , J. 1996. A biomineralization model for the incorporation of trace elements into foraminiferal calcium carbonate. Earth and Planetary Science Letters, 142, 409–423. E MILIANI , C. 1955. Pleistocene temperatures. Journal of Geology, 63, 538–578. E REZ , J. 2003. The source of ions for biomineralization in foraminifera and their implications for paleoceanographic proxies. Review in Mineralogy and Geochemisty, 54, 115– 149. H EMLEBEN , C., B E , A. W. H., A NDERSON , O. R. & T UNTIVATE -C HOY , S. 1977. Test morphology, organic layers and chamber formation of the planktonic foraminifer Globorotalia menardii (d’Orbigny). Journal of Foraminifera Research, 1 –25. H EMLEBEN , C., S PINDLER , M., B REITINGER , I. & D EUSER , W. G. 1985. Field and laboratory studies on the ontogeny and ecology of some Globorotaliid species from the Sargasso Sea off Bermuda. Journal of Foraminifera Research, 15, 254– 272. H EMMING , N. G. & H ANSON , G. N. 1992. Boron isotopic composition and concentration in modern marine carbonates. Geochimica et Cosmochimica Acta, 56, 537– 543. H O¨ NISCH , B. & H EMMING , N. G. 2004. Ground-truthing the boron isotope paleo-pH proxy in planktonic foraminifera shells: partial dissolution and shell size effects. Paleoceanography, 19 PA4010, doi: 10.1029/2004PA001026. H O¨ NISCH , B. & H EMMING , N. G. 2005. Surface ocean pH response to variations in pCO2 through two full glacial cycles. Earth and Planetary Science Letters, 236, 305– 314. H O¨ NISCH , B., B IJMA , J., R USSELL , A. D., S PERO , H. J., P ALMER , M. R., Z EEBE , R. E. & E ISENHAUER , A. 2003. The influence of symbiont photosynthesis on the boron isotopic composition of foraminifera shells. Marine Micropaleontology, 49, doi:10.1016/ S0377– 8398(03)00030– 6. H O¨ NISCH , B., H EMMING , N. G., G ROTTOLI , A. G., A MAT , A., H ANSON , G. N. & B IJMA , J. 2004. Assessing scleractinian corals as recorders for paleo-pH: empirical calibration and vital effects. Geochimica et Cosmochimica Acta, 68, 3675–3685. H O¨ NISCH , B., B ICKERT , T. & H EMMING , N. G. 2008. Modern and pleistocene boron isotope composition of the benthic foraminifer Cibicidoides Wuellerstrorfi, Earth and Planetary Science Letters, in press. IUPAC. 1998. Isotopic compositions of the elements 1997. Pure Applied Chemistry, 70, 217–235. J ØRGENSEN , B. B., E REZ , J., R EVSBECH , N. P. & C OHEN , Y. 1985. Symbiotic photosynthesis in a planktonic foraminiferan, Globigerinoides sacculifer (Brady), studied with microelectrodes. Limnology Oceanography, 30, 1253– 1267. K AKIHANA , H. & K OTAKA , M. 1977. Equilibrium constants for boron isotope-exchange reactions. Bulletin of Research Laboratory for Nuclear Reactions, 2, 1 –12.
K AKIHANA , H., K OTAKA , M., S ATOH , S., N OMURA , M. & O KAMOTO , M. 1977. Fundamental studies on the ion-exchange separation of boron isotopes. Bulletin of the Chemical Society of Japan, 50, 158–163. K O¨ HLER -R INK , S. & K U¨ HL , M. 2005. The chemical microenvironment of the symbiotic planktonic foraminifer Orbulina universa. Marine Biology Research, 1, 68– 78. L AZIER , J. R. N. & M ANN , K. H. 1989. Turbulence and the diffusive layers around small organism. Deep-Sea Research, 36, 1721–1733. L EA , D. W., P AK , D. K. & S PERO , H. J. 2000. Climate impact of late Quaternary equatorial Pacific sea surface temperature variations. Science, 289, 1719–1724. L IU , Y. & T OSSELL , J. A. 2005. Ab initio molecular orbital calculations for boron isotope fractionations on boric acids and borates. Geochimica et Cosmochimica Acta, 69, 3995–4006. M ILLERO , F. J. 1995. Thermodynamics of the carbon dioxide system in the oceans. Geochimica et Cosmochimica Acta, 59, 661–677. M ORSE , J. W. & M ACKENZIE , F. T. 1990. Geochemistry of Sedimentary Carbonates. Developments in Sedimentology, 48, Elsevier, Amsterdam. N U¨ RNBERG , D., B IJMA , J. & H EMLEBEN , C. 1996. Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperatures. Geochimica et Cosmochimica Acta, 60, 803–814; Erratum 2483–2484. O I , T. 2000. Calculations of reduced partition function ratios of monomeric and dimeric boric acids and borates by the ab initio molecular orbital theory. Journal of Nuclear Science and Technology, 37, 166–172. P EARSON , P. N. & P ALMER , M. R. 2000. Atmospheric carbon dioxide concentrations over the past 60 million years. Nature, 406, 695– 699. P ETIT , J. R., J OUZEL , J., R AYNAUD , D. ET AL . 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature, 399, 429– 436. R EYNAUD , S., H EMMING , G. N., J UILLET -L ECLERC , A. & G ATTUSO , J.-P. 2004. Effect of pCO2 and temperature on the boron isotopic composition of the zooxanthellate coral Acropora sp. Coral Reefs, 23, 539–546. R INK , S., K U¨ HL , M., B IJMA , J. & S PERO , H. J. 1998. Microsensor studies of photosynthesis and respiration in the symbiotic foraminifer Orbulina universa. Marine Biology, 131, 583– 595. R OLLION -B ARD , C. 2005. Microanalysis by ion microprobe of oxygen isotopic compositions in foraminifera. Geophysical Research Abstracts, 7, 1607– 7962. R OYER , D. L., B ERNER , R. A., M ONTANEZ , I. P., T ABOR , N. J. & B EERLING , D. J. 2004. CO2 as a primary driver of Phanerozoic climate change. GSA Today, 14, 4–10. S ANYAL , A. & B IJMA , J. A. 1999. Comparative study of the northwest Africa and eastern equatorial Pacific upwelling zones as sources of CO2 during glacial periods based on boron isotope paleo-pH estimation. Paleoceanography, 14, 753–759. S ANYAL , A., H EMMING , N. G., H ANSON , G. N. & B ROECKER , W. S. 1995. Evidence for a higher pH in the glacial ocean from boron isotopes in foraminifera. Nature, 373, 234– 236.
PALAEOCEANOGRAPHICAL PROXY RELATIONSHIPS IN FORAMINIFERA S ANYAL , A., H EMMING , N. G., B ROECKER , W. S., L EA , D. W., S PERO , H. J. & H ANSON , G. N. 1996. Oceanic pH control on the boron isotopic composition of foraminifera: evidence from culture experiments. Paleoceanography, 11, 513–517. S ANYAL , A., N UGENT , M., R EEDER , R. J. & B IJMA , J. 2000. Seawater pH control on the boron isotopic composition of calcite: evidence from inorganic calcite precipitation experiments. Geochimica et Cosmochimica Acta, 64, 1551– 1555. S ANYAL , A., B IJMA , J., S PERO , H. J. & L EA , D. W. 2001. Empirical relationship between pH and the boron isotopic composition of Globigerinoides sacculifer: Implications for the boron isotope paleo-pH proxy. Paleoceanography, 16, 515 –519. S HACKLETON , N. J. 1967. Oxygen isotope analysis and Pleistocene temperatures re-assessed. Nature, 215, 15– 17. S IEGENTHALER , U., S TOCKER , T. F., M ONNIN , E., L U¨ THI , D., S CHWANDER , J., S TAUFFER , B., R AYNAUD , D., B ARNOLA , J.-M., F ISCHER , H., M ASSON -D ELMOTTE , V. & J OUZEL , J. 2005. Stable carbon cycle-climate relationship during the Late Pleistocene. Science, 310, 1313–1317. S PERO , H. J. 1998. Life history and stable isotope geochemistry of planktonic foraminifera. In: N ORRIS , R. D. & C ORFIELD , R. M. (eds) Isotope Paleobiology and Paleoecology; Paleontological Society Papers, 4, 7–36. S PERO , H. J. & D E N IRO , M. J. 1987. The influence of symbiont photosynthesis on the d18O and d13C values of planktonic foraminiferal shell calcite. Symbiosis, 4, 213–228. S PERO , H. J. & L EA , D. W. 1993. Intraspecific stable isotope variability in the planktic foraminifera Globigerinoides sacculifer: Results from laboratory experiments. Marine Micropaleontology, 22, 221–234. S PERO , H. J. & L EA , D. W. 1996. Experimental determination of intraspecific stable isotope variability in Globigerina bulloides: Implications for paleoceanographic reconstructions. Marine Micropaleontology, 28, 231–246. S PERO , H. J. & L EA , D. W. 2002. The cause of carbon isotope minimum events on glacial terminations. Science, 296, 522–525. S PERO , H. J. & W ILLIAMS , D. F. 1988. Extracting environmental information from planktonic foraminiferal d13C data. Nature, 335, 717– 719. S PERO , H. J., L ERCHE , I. & W ILLIAMS , D. F. 1991. Opening the carbon isotope “vital effect” black box, 2, quantitative model for interpreting foraminiferal carbon isotope data. Paleoceanography, 6, 639–655. S PERO , H. J., B IJMA , J., L EA , D. W. & B EMIS , B. E. 1997. Effect of seawater carbonate concentration on foraminiferal carbon and oxygen isotopes. Nature, 390, 497– 500. S PIVACK , A. J., Y OU , C.-F. & S MITH , H. J. 1993. Foraminiferal boron isotope ratios as a proxy for surface ocean pH over the past 21 Myr. Nature, 363, 149–151. S TANLEY , S. M. & H ARDIE , L. A. 1998. Secular oscillations in the carbonate mineralogy of reef-building and sediment-producing organisms driven by tectonically forced shifts in seawater chemistry. Palaeogeography, Palaeoclimatology, Palaeoecology, 144, 3–19.
57
S U , C. & S UAREZ , D. L. 1995. Coordination of adsorbed boron: a FTIR spectroscopic study. Environmental Science and Technology, 29, 302–311. T RIPATI , A. & E LDERFIELD , H. 2005. Deep-sea temperature and circulation changes at the Paleocene-Eocene thermal maximum. Science, 308, 1894– 1898. T YRRELL , T. & Z EEBE , R. E. 2004. History of carbonate ion concentration over the last 100 million years. Geochimica et Cosmochimica Acta, 68, 3521– 3530. U REY , H. C. 1947. The thermodynamic properties of isotopic substances. Journal of the Chemical Society, 562– 581. U SDOWSKI , E. & H OEFS , J. 1993. Oxygen isotope exchange between carbonic acid, bicarbonate, carbonate, and water: a re-examination of the data of McCrea (1950) and an expression for the overall partitioning of oxygen isotopes between the carbonate species and water. Geochimica et Cosmochimica Acta, 57, 3815– 3818. W AELBROECK , C., L ABEYRIE , L., M ICHEL , E., D UPLESSY , J. C., M C M ANUS , J. F., L AMBECK , K., B ALBON , E. & L ABRACHERIE , M. 2002. Sea-level and deep water temperature changes derived from benthic foraminifera isotopic records. Quaternary Science Review, 21, 295– 305. W EFER , G. & B ERGER , W. H. 1991. Isotope paleontology: growth and composition of extant calcareous species. Marine Geology, 100, 207–248. W OLF -G LADROW , D. A. & R IEBESELL , U. 1997. Diffusion and reactions in the vicinity of plankton: a refined model for inorganic carbon transport. Marine Chemistry, 59, 17–34. W OLF -G LADROW , D. A., B IJMA , J. & Z EEBE , R. E. 1999. Model simulation of the carbonate system in the microenvironment of symbiont bearing foraminifera. Marine Chemistry, 64, 181–198. Z EEBE , R. E. 1999. An explanation of the effect of seawater carbonate concentration on foraminiferal oxygen isotopes. Geochimica et Cosmochimica Acta, 63, 2001–2007. Z EEBE , R. E. 2001. Seawater pH and isotopic paleotemperatures of Cretaceous oceans. Palaeogeography, Palaeoclimatology, Palaeoecology, 170, 49–57. Z EEBE , R. E. 2005a. Stable boron isotope fractionation between dissolved B(OH)3 and B(OH)2 4 . Geochimica et Cosmochimica Acta, 69, 2753– 2766. Z EEBE , R. E. 2005b. Large effect of hydration on 18O fractionation between H2O and CO2 Implications for the pH-carbonate-d18O relationship and inferred climate changes. Geophysical Research Abstracts, 7, 1607– 7962. Z EEBE , R. E. & S ANYAL , A. 2002. Comparison of two potential strategies of planktonic foraminifera for house building: Mg2þ or Hþ removal? Geochimica et Cosmochimica Acta, 66, 1159– 1169. Z EEBE , R. E. & W OLF -G LADROW , D. A. 2001. CO2 in Seawater: Equilibrium, Kinetics, Isotopes. Elsevier Oceanography Series, Amsterdam, 346. Z EEBE , R. E., B IJMA , J. & W OLF -G LADROW , D. A. 1999. A diffusion-reaction model of carbon isotope fractionation in foraminifera. Marine Chemistry, 64, 199– 227.
58
R. E. ZEEBE ET AL.
Z EEBE , R. E., S ANYAL , A., O RTIZ , J. D. & W OLF -G LADROW , D. A. 2001. A theoretical, study of the kinetics of the boric acid – borate equilibrium in seawater. Marine Chemistry, 74, 113– 124. Z EEBE , R. E., W OLF -G LADROW , D. A., B IJMA , J. & H O¨ NISCH , B. 2003. Vital effects in planktonic
foraminifera do not compromise the use of d11B as a paleo-pH indicator: evidence from modelling. Paleoceanography, 18, 1043. Z HANG , J., Q UAY , P. D. & W ILBUR , D. O. 1995. Carbon isotope fractionation during gas-water exchange and dissolution of CO2. Geochimica et Cosmochimica Acta, 59, 107–114.
Foraminifer test preservation and diagenesis: comparison of high latitude Eocene sites PAUL N. PEARSON & CATHERINE E. BURGESS School of Earth, Ocean and Planetary Sciences, Cardiff University, Main Building, Park Place, Cardiff, CF10 3YE, UK (e-mail:
[email protected]) Abstract: Foraminifer tests are prone to diagenetic recrystallization on a micron scale that can affect their geochemical composition, hence it is important to identify fossil material that is well-preserved. Here we illustrate the textures of tests from several high-latitude Eocene sites using a combination of Reflected Light Microscopy and high-resolution Scanning Electron Microscopy. The sites are Ocean Drilling Program (ODP) Site 647 in the Labrador Sea (538N), ODP Site 689 at Maud Rise in the Weddell Sea (648S), ODP Site 1135 (Kerguelen Plateau c. 598S) and outcrop samples from Hampden Beach, New Zealand (palaeolatitude c. 558S). The foraminifera studied from Site 647 and Hampden Beach have glassy, transparent tests that show only minor signs of diagenetic alteration, whereas the foraminifera from Sites 690 and 1135 are opaque and recrystallized. We associate the better state of foraminifer preservation at the former sites with the high clay content of the sediments. Our observations suggest that Site 647 and Hampden Beach may be useful for the establishment of high-latitude sea-surface temperatures and hence global temperature gradients in the Eocene.
Planktonic foraminifer tests are widely used to investigate palaeoenvironmental signals from the surface oceans by the analysis of their isotope ratios and trace element composition. Obviously if a past environmental signal is to be reliably reconstructed it is necessary either that the tests have survived with little or no diagenetic alteration or, if diagenetic alteration has occurred, that it has not greatly affected the test chemistry. If systematic diagenetic biases operate on foraminifer tests then they may lead to misinterpretation of past conditions. In recent years there has been a growing awareness that fossil foraminifer tests from classic deep-sea ooze and chalk sites that have previously been regarded as well-preserved may actually have been substantially affected by diagenesis, and that a more reliable geochemical record may be obtained from better preserved material from clays (Huber et al. 1995; Norris & Wilson 1998; Wilson & Norris 2001; Pearson et al. 2001, 2007; Wilson et al. 2002; Stewart et al. 2004; Williams et al. 2005a, b). Pearson et al. (2001, 2002) argued that for typical Cretaceous and Palaeogene ooze/chalk sites, foraminifer tests might consist of as much as 50% diagenetic calcite added on the sea floor or in shallow burial at low temperature, and consequently that the apparent tropical sea surface temperatures obtained from recrystallized foraminifera may be severe underestimates. One of the most fundamental aspects of the global climate system is the sea-surface temperature gradient from the equator to the polar seas. In order
to establish this latitudinal temperature gradient for the Eocene it is desirable to obtain well-preserved material from the high latitudes (.508) to complement low latitude data (Pearson et al. 2001). The most important Eocene sites that have been previously studied for stable isotope geochemistry are from Kerguelen Plateau (ODP Sites 738, 748, 750, 1135) and Maud Rise (ODP Sites 277, 690), both in the Southern Ocean. Unfortunately the Eocene sediment at these sites is carbonate-rich and has questionable foraminifer preservation, making it desirable to locate and study clay-rich sites. In order to further the study of high latitude palaeoceanography for the Eocene we have re-examined foraminifera from Kerguelen Plateau (Site 1135) and Maud Rise (Site 689) as well as two clay-rich sites, ODP Site 647 and Hampden Beach, New Zealand; see below for details. We have used a combination of Reflected Light Microscopy (RLM) and SEM to more fully evaluate the preservation state of selected samples than has usually been attempted elsewhere. We have interpreted our observations in the light of what is known from studies of biomineralization in modern planktonic foraminifera and diagenesis, which we summarize in the following sections.
Foraminifer biomineralization and test textures Planktonic foraminifera secrete tests of Low Magnesium Calcite. Each test consists of a series of
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 59–72. DOI: 10.1144/SP303.5 0305-8719/08/$15.00 # The Geological Society of London 2008.
60
P. N. PEARSON & C. E. BURGESS
chambers that (in most species) are added episodically in a spiral arrangement as the organism grows. Hemleben et al. (1986, 1989) and Be´ et al. (1979) have reviewed the process of chamber formation, which is slightly different in spinose and non-spinose species. Chamber formation in both groups starts with the extrusion of a cytoplasmic bulge from the primary aperture, which then takes on the shape of the chamber-to-be. Calcification takes place on both sides of a Primary Organic Membrane (POM) which lies within the cytoplasm, surrounded by an Outer Protective Envelope. The test is constructed from minute calcareous plaques or ‘microgranules’ (Blow 1979, p. 320) that are initially secreted by rhizopodia and then apparently transported to the precise site of test formation. These microgranules are about 0.1 mm in diameter and have an irregular shape with no clear crystal faces (Blow 1979; Hemleben et al. 1989). When in place they fuse laterally to form the test wall, which grows both inward and outward from the POM (predominantly in an outward direction in spinose species) until chamber formation is complete. Further test layers may be added subsequently to the outer surface as new chambers are formed. A large adult chamber with a radius of 100 mm and a thickness of 10 mm is constructed from over a billion microgranules. When the test is complete, the inner and outer surfaces are generally very smooth, even on a submicron scale, which implies an additional process of surface rendering and shaping. It may have finely sculpted structures such as pores, ridges, spines, apertural lips, pustules and keels. Such features differ markedly between species and are related to life habit, including modes of cytoplasm flow and feeding activity. During life, the foraminifer is also able to resorb calcite and repair damaged parts of the test (Hemleben et al. 1977, 1989). Hemleben et al. (1977, 1986, 1989) have illustrated the internal wall structure of broken chambers from a number of species. The layer of the POM is often clearly visible in cross section, as are internal layers associated with test thickening during chamber formation. Despite the fact that the test surfaces are usually very smooth (except where pustlose), the relict internal microgranular structure is still clearly visible with SEM when the tests are broken open. Some species of foraminifera resorb parts of their tests while substantially thickening the outer surface in the hours before sexual reproduction (gametogenesis) (Be´ 1980). The appearance of this ‘gametogenic crust’ differs between species. It may be irregular and prismatic, as in some Globorotalia, or may appear smooth as in Globigerinoides. Observations suggest that the gametogenic crust in some species is more distinctly crystalline
than the rest of the test (Hemleben et al. 1989). Often the crust is formed after the foraminifer sinks to deep waters so it will have a distinct isotopic and chemical composition (Spero & Lea 1993; Hastings et al. 1998; Benway et al. 2003). The foraminiferal test, when well preserved, is remarkably strong under compressive stress, for example in resisting moderate pressure under a glass slide. When viewed under moderately high magnification with RLM, the calcite is typically translucent, especially in small specimens and thinwalled species, and the outer surfaces are smooth and reflective except where covered with pustules. In small specimens it is usually possible to see the tubular pore channels running through the test as well as internal features such as inner chamber partitions. Thus fresh and well-preserved specimens, especially from the fine fraction, tend to have a ‘glassy’ appearance. However large specimens and those with a thick gametogenic crust can be more opaque, and adhering sediment including coccoliths can add to their opacity. The degree of transparency is increased when the foraminifer tests are placed in water or oil which reduces refraction from the surface. Hemleben & Olsson (2006) have described the test ultrastructures of Eocene planktonic foraminifera. Test textures are broadly similar and sometimes virtually identical to living species, indicating that models of biomineralization based on observations of living forms are applicable to the fossil record.
Diagenesis A variety of processes may affect a foraminifer test as it falls through the water column or is buried in sea-floor sediment. These processes can loosely be divided into dissolution, overgrowth and recrystallization, although the processes are inter-related. Parts of the test may be more susceptible to dissolution and recrystallization than others. Alteration of the biogenic structures often begins as the test sinks through the water column, especially if the seafloor is deep. Most species live and calcify in the top 100 m, but deeper waters are colder and more corrosive to calcite. In very deep oceanic settings all the tests dissolve before reaching the sediment. Above the carbonate compensation depth tests can survive into the geological record, but even so they can be highly dissolved and fragmented in deeper sites (Hemleben et al. 1989). Once buried, the tests are susceptible to further dissolution, overgrowth and recrystallization. Overgrowth occurs when inorganic calcite is precipitated from solution on the outer or internal surface of the test. Such crystals can grow to large
FORAMINIFER TEST PRESERVATION
sizes. In limestones and some chalks it is common for the foraminifer tests to be filled with diagenetic precipitate. Such inorganic calcite will generally be easily distinguished from the biogenic calcite because it consists of large equant crystals, often growing outward or inward from the test surface with radially directed c-axes. In some species, the gametogenic crust has the appearance of inorganic overgrowth, although as a rule overgrowth will affect all tests regardless of the species. Recrystallization is an equally important but less obvious process, in which the internal microgranular structure of the tests is replaced by larger crystals. The fresh biogenic calcite has a high surface energy because it is composed of small (c. 0.1 mm), very irregular plaques without clear crystal faces. We suggest that the recrystallization is driven by energetic gain involved in replacing these microgranules with larger, more equant crystals with lower surface energy. The process is poorly understood, but it must involve both dissolution and precipitation, and it probably occurs within aqueous films on a very local (submicron) scale. These aqueous films may have a very different chemical composition from the flowing pore fluids that surround the tests and fill the empty pore space (Pingitore 1982). It is probably the case that the constituents are repeatedly dissolved and re-precipitated as the original texture degrades and larger diagenetic crystals form. Our observations of foraminifers from many sites suggests that the end state of this process typically consists of loosely packed blocky calcite crystals of a micron or more in diameter, often with dissolved pore space between them. It may be that the process slows or stops when this more stable crystal structure has been obtained. The result is the complete replacement of the test by neomorphic calcite which may nevertheless retain some of the original geochemical fingerprint. Our observations indicate that the recrystallization causes significant weakening of the test, although quantitative studies have not yet been conducted. Recrystallized tests crumble much more easily under moderate compressive stress and they are less able to withstand ultrasonic cleaning. Most obviously, the tests are no longer transparent, but opaque and chalky white (the so-called ‘frosty’ appearance described by Sexton et al. 2006). It may be that organic membranes such as the POM and pore linings are destroyed in the recrystallization process, because we have only observed them in non-recrystallized material. Foraminifer tests may be wholly recrystallized without suffering significant overgrowth or infilling, and equally they may be completely infilled while retaining a near-pristine microgranular structure. Hence if one observes that a test is free from
61
overgrowth it is not sufficient guarantee that it is has not been neomorphosed/recrystallized. For geochemical studies it is critically important that foraminifer test textures are illustrated at high magnification so that the degree of recrystallization can be assessed. We recommend the use of several criteria to assess the state of preservation as follows: 1. 2. 3.
4.
Are the tests translucent and reflective under RLM, especially smaller thin walled specimens and when in water or oil? Do ultrafine features such as spines (if originally possessed) survive? Under SEM, are originally smooth parts of the test such as the outer surface (in some species), the inner surface (in most species), apertural lips and sutures still smooth on a submicron scale? Can the submicron microgranular texture be identified in cross section when the test is broken?
The mode of diagenetic alteration presumably depends on many factors including time, initial water depth, burial depth history, sediment composition, burial temperature and pore water chemistry. No simple diagenetic model is likely to be able to take all these factors into account, so preservation has to be assessed on a site-by-site basis. In general, ancient tests from relatively impermeable clay-rich sediments, which are often in more shallow settings, tend to exhibit near-pristine microgranular textures, whereas tests from carbonate oozes in deeper oceanic settings are almost always recrystallized to some degree. An important question is to what extent the recrystallization process affects test chemistry, including trace elements and isotopic ratios. Pearson et al. (2001) represented the recrystallization as occurring along a mixing line between original and diagenetic end-member compositions, at least with respect to the carbon and oxygen isotope systems, and argued that typical neomorphosed tests are perhaps halfway along this trajectory. However there is no reason why the process should affect all aspects of test chemistry to an equal degree. Trace elements with a positive partition coefficient from pore fluids into calcite may be less susceptible to alteration (i.e. being removed from the test during the recrystallization process) than others (Sexton et al. 2006). Even the concept of pore fluids may not be applicable because the dissolution and recrystallization must be taking place in very small cavities or aqueous films within the ‘solid’ test wall, and these solutions may have a very different chemical composition to the true pore fluids that circulate between the tests and within the chambers (Pingitore 1982). More work involving multiple geochemical tracers is
62
P. N. PEARSON & C. E. BURGESS
required to develop an adequate model for this process.
We selected representative specimens of planktonic foraminifera from four sites, discussed below, for comparative textural study. Between them the sites contain a wide variety of species. In order to facilitate comparisons between the sites we have chosen specimens of the same species and approximately the same size for illustration, but all species were not necessarily available from all sites. Specimens were mounted adjacent to one another on adhesive pads on SEM stubs. The tests were immersed in water and the specimens from each species group were photographed together under RLM, so that the illumination conditions and exposure time were consistent (Table 1). They were then dried and the stubs were coated with gold-palladium for SEM study. After imaging the external surface at low and high resolution, the tests were removed from the stub, broken using moderate pressure under a glass slide and the fragments were remounted on SEM stubs and re-coated for investigation of the internal surfaces and the test walls in cross section.
Our initial observations indicate that the foraminifers are well-preserved and ‘glassy’ through much of the Eocene and Oligocene with little variation is preservation (see also Arthur et al. 1989). We selected a typical sample for more detailed investigation (Sample 647A/50R/5, 103– 105 cm), which comes from 479.53 mbsf. The sediment from this core is a relatively homogeneous greenish gray (5GY 5/1) nannofossil claystone. The clay content of the sediment varies from approximately 60 –80% based on the smear slide summary (Srivastava et al. 1987) and the calcium carbonate has been measured at 29.0–33.8%, with the highest recorded clay content in the same section as the sample was taken. We chose the sample from Core 50 because it coincides with a rare incursion of the warm-water genus Hantkenina into the site. The species in question is Hantkenina australis, which possesses recurved tubulospines (Coxall & Pearson 2006) and was first described by Finlay (1939) from Hampden Beach, New Zealand. However further work is required to determine whether the samples from Site 647 and Hampden Beach are from a similar stratigraphic level in the middle Eocene. Other planktonic foraminifer genera present include Subbotina, Globorotaloides and Catapsydrax.
ODP Site 647
ODP Site 689
ODP Site 647 is located in the Labrador Sea about half way between the southern tip of Greenland and the northern tip of Newfoundland (Srivastava et al. 1987). A total of 736.0 m was penetrated in Hole 647A (53819.8760 N, 45815.7170 W), terminating in basalt. The oldest sediments at the site have been dated to the early Eocene using calcareous nannofossils, foraminifers and dinocysts. A thick middle Eocene succession was recovered that extends from about 450 metres below sea floor (mbsf) to 650 mbsf. The average sedimentation rate in this interval is 36 m/Ma (Srinivasta et al. 1987). The high clay content led Srinivasta et al. (1987) to conclude that it had been deposited in a hemipelagic environment, with the clays being derived from North America and Greenland.
ODP Site 689 was cored on the eastern flank of Maud Rise in the Weddell Sea (Barker et al. 1988). Hole 689B (64831.0090 S; 03805.9960 E) was drilled to 297.3 m. Age control is from foraminifera, nannofossils, diatoms and palaeomagnetics. The Eocene extends from about 130 to 200 mbsf, with a sediment accumulation rate of 4– 4.5 m/m.yr. The sediment is described as nannofossil ooze and chalk deposited in a pelagic environment (Barker et al. 1988). Our initial observations suggest that the foraminifers from this site and neighbouring ODP Site 690 are free from infilling but chalky and opaque. We selected one (Sample 689B/15H/2, 42.5 –43 cm, 222.92 mbsf) from the upper Eocene for further analysis. Four samples from Core 15H were
Materials and methods
Table 1. Exposure times and lighting for RLM photographs Photograph Exposure Time/ms Lighting Style
Figure 1(e –h) 270 Directed light
Figure 2(d –f) 1200 Ring light
Figure 3(d – f) 320 Ring þ directed light
Figure 4(c – d) 674 Directed light
All RLM photographs were taken with a Lecia DFC 480 camera on a Lecia MZ16 microscope using Earth Basic software for image capture. The diaphragm was set to half open on the microscope and the gain and colour saturation set to x1 on the camera for all photographs.
FORAMINIFER TEST PRESERVATION
recorded by Barker et al. (1988) as having between 87– 89% CaCO3. The sediment is described as diatom bearing nannofossil ooze. Sample 689B/ 15H/2, 42.5–43 cm contains abundant radiolaria and foraminifera. Common planktonic foraminifer genera present are Globoquadrina, Subbotina, Globigerinatheka, Globorotaloides and Chiloguembelina. The presence of Globigerinatheka index confirms the Eocene age.
ODP Site 1135 ODP Site 1135 (59842.00 S, 84816.40 E) was drilled on the southern Kerguelen Plateau in the southern Indian Ocean (Coffin et al. 2000). Hole 1135A was drilled to a total depth of 526 mbsf. Age control is primarily from foraminifers, nannofossils and palaeomagnetics. Eocene sediments described as nannofossil ooze underlie a thin Pliocene cover, from about 10–200 mbsf. The sediments have a very constant carbonate percentage of about 95% CaCO3. Sediment accumulation rates for the Eocene have been estimated at 15 m/m.yr (Coffin et al. 2000). Our initial study of Eocene foraminifera from this site and others on the Kerguelen Plateau is that they are opaque and chalky in appearance. We selected one typical sample (Sample 1135A/ 6R/CC, from 56.1 mbsf) for more detailed analysis. A sample from this core was recorded as having 96% CaCO3 (Coffin et al. 2000). The sediment was described as foraminifer bearing nannofossil ooze. Common genera include Acarinina, Globigerinatheka, Subbotina, Globorotaloides and Pseudohastigerina.
Hampden Beach The Hampden Beach section is a cliff section on the east coast of South Island, New Zealand (45817.00 S, 170850.750 E). Several important species of planktonic foraminifera were described from this location by Finlay (1939) including Globigerinatheka index, Acarinina collactea and Hantkenina australis. The exposure is approximately 2.5 km long and represents a total stratigraphic thickness of around 153 m. The whole thickness of the section is exposed for sampling in outcrop, as it has a gentle dip of around 5 degrees along its length. Biostratigraphy based on dinoflagellates (Wilson 1985) and foraminifers (Jenkins 1971), dates it to the middle Eocene, and suggests an average sedimentation rate for the section of approximately 30 m/m.y although this figure is subject to considerable uncertainty because of possible repetition and/or missing section as a result of faulting.
63
Observations on the samples suggest that the foraminifers are generally well-preserved, although some samples are apparently affected by surface weathering and dissolution. The sediment is a slightly calcareous mudstone to very fine sandstone that is highly glauconitic at the base in the south of the exposure. The sediment is dominated by clays, fine quartz grains, abundant larger grains of glauconite and micas with subsidiary feldspars and siderite. Representative samples from several places in the section have been chosen for more detailed analysis of foraminifer preservation. Those reported here are samples CB050 and CB058 from towards the southern end of the section and Sample PPO2-HB14 from the middle of the section.
Comparative textural study The results of our textural investigation are shown in Figures 1 – 4. Figure 1 shows four specimens of Subbotina linaperta from ODP Sites 689, 1135, Hampden Beach and Site 647 respectively. This species is interpreted as occupying a relatively deep thermocline habitat (Poore & Matthews 1984; Pearson et al. 1993), although it may have been a surface-dwelling form at high latitudes. It had a cancellate (honeycomb) spinose wall in life (Olsson et al. 2006) but presumably resorbed the spines at gametogenesis like modern species with the same wall texture. Figure 2 shows three specimens of Globoturbototalita sp. from ODP Site 1135, Hampden Beach and Site 647 respectively. This species was probably a shallow-water dwelling form and had a cancellate spinose wall similar to Subbotina (Olsson et al. 2006). Figure 3 shows specimens of Globigerinatheka index from ODP Sites 689, 1135 and Hampden Beach respectively. This species is regarded as a surface mixed-layer dweller (Boersma et al. 1979; Pearson et al. 1993, 2001), which probably lived with symbiotic algae and had a cancellate spinose wall on which a thick gametogenic crust often developed in adult forms (Premoli Silva et al. 2006). Figure 4 shows two specimens of Acarinina bullbrooki from ODP Site 1135 and Hampden Beach respectively. This species is regarded as a surface mixed-layer dwelling form (Boersma et al. 1987; Pearson et al. 2001) and had an association with symbiotic algae (Pearson et al. 1993). It was non-spinose with a muricate (pustulose) wall (Blow 1979; Berggren et al. 2006). Under RLM (top row in the plates), the specimens of all species from ODP 689 and 1135 were chalky, white and nearly opaque. Internal features
64
P. N. PEARSON & C. E. BURGESS
Fig. 1. Comparison of four specimens of Subbotina linaperta from four localities, ODP Site 689 – Maud Rise (a, e, i, m, q, u), ODP Site 1135 – Kergeulen Plateau (b, f, j, n, r, v), Hampden Beach (c, g, k, o, s, w), ODP Site 647 – North Atlantic (d, h, l, p, t, x). RLM photos taken with the specimens submerged in water (a– d) are compared to SEM photographs of the same specimens (e–x). Scale bar is 200 mm for (a– h). Scale bars are 10 mm in (i –l) and 2 mm in (m–x).
FORAMINIFER TEST PRESERVATION
65
Fig. 2. Comparison of three specimens of Globoturborotalita sp. from three localities, ODP Site 1135 – Kergeulen Plateau (a, d, g, j, m), Hampden Beach (b, e, h, k, n), ODP Site 647 – North Atlantic (c, f, i, l, o). RLM photos taken with the specimens submerged in water (a–c) are compared to SEM photographs of the same specimens (d –o). Scale bar is 200 mm for (a–f). Scale bars are 10 mm in (g –i) and 2 mm in (j– o).
66
P. N. PEARSON & C. E. BURGESS
Fig. 3. Comparison of three specimens of Globeriginatheka index from three localities, ODP Site 689 – Maud Rise (a, d, g, j, m), ODP Site 1135 – Kergeulen Plateau (b, e, h, k, n), Hampden Beach (c, f, i, l, o). RLM photos taken with the specimens submerged in water (a– c) are compared to SEM photographs of the same specimens (d –o). Scale bar is 200 mm for Scale bars are 10 mm in (g –i) and 2 mm in (j– o).
FORAMINIFER TEST PRESERVATION
67
Fig. 4. Comparison of two specimens of Acarinina bullbrooki from two localities, ODP Site 1135 – Kergeulen Plateau (a, c, e, g, i), Hampden Beach (b, d, f, h, j). RLM photos taken with the specimens submerged in water (a, b) are compared to SEM photographs of the same specimens (c–j). Scale bar is 200 mm in (a–d). Scale bars are 10 mm in (e, f) and 2 mm in (g –j).
68
P. N. PEARSON & C. E. BURGESS
were difficult to discern, which we attribute to a combination of the recrystallized nature of the wall and the presence of some internal sediment. In contrast, the specimens from Hampden Beach and ODP Site 647 were ‘glassy’ and translucent, with reflective surfaces. Pore channels were clearly visible through the chamber walls and features such as the apertural lip on Globoturborotalita sp. (Fig. 2b) and Acarinina bullbrooki (Fig. 4b) were sharply delineated. Under low power SEM (second row in the plates) the various tests appear similar to one another. The principle features of the wall textures of the various species can be observed in all specimens, such as the porous nature of the wall and some surface ornamentation such as inter-pore ridges and muricae. However the partially peeled surface of the final chamber of Subbotina linaperta from ODP Site 689 is apparent through the loss of cancellation on the test surface (Fig. 1e) and it is also possible to see the trace of a broken final chamber on the specimen from Hampden Beach (Fig. 1g). Under moderate power with the SEM it is possible to see differences in the preservation that are consistent between the sites (images in the third row, with a 10 micron scale bar on the plates; note that adhering coccoliths are visible in some images and there is also some clay on the specimens from ODP Site 647). The specimens from Hampden Beach and ODP Site 647 all show smoother external surfaces, in which the original wall texture is better preserved, than the other sites. A degree of dissolution is evident in some of these specimens, especially Subbotina linperta from ODP Site 647. This is manifested as seemingly cracked and eroded inter-pore ridges, revealing the internal test texture in places (Fig. 1l). Specimens from ODP Site 689 and 1135 show a rougher, more granular texture that we attribute to both recrystallization of the test wall and dissolution in some specimens. The Subbotina linaperta specimen from ODP Site 869 shows rosettes of calcite centered on the pores which results from the complete loss of the inter-pore ridges. The inter-pore ridges survive in the S. linaperta specimen from ODP Site 1135 but the texture is not as smooth as in the better-preserved sites. When the external surfaces are viewed with high magnification with SEM (images in the fourth row of the plates) the fundamentally different character of the preservation style is most apparent. The Hampden Beach and ODP Site 647 specimens are either smooth at this scale (as in the case of G. index, Fig. 3l) or very small sub-micron crystals are visible on the surface (as in most of the other images). We interpret the small crystals visible in these specimens as representing the internal
microstructure (see below) that has been revealed by a small amount of etching, but there may also be a degree of overgrowth and/or recrystallization, even in this ‘glassy’ material. In contrast, however, the specimens from ODP Sites 689 and 1135 all show a blocky texture of equant crystals of approximately 1–2 microns in diameter. The similarity of these textures to those observed at other sites (e.g. Huber et al. 1995; Pearson et al. 2001) suggest that this may be the typical end-stage of a micron-scale recrystallization process. When the test walls are broken open and imaged under high magnification there are also clear differences between the preservation styles (images in the fifth row on the figures). The specimens from Hampden Beach and ODP Site 647 show a submicron granularity that is very similar to that illustrated from modern tests in cross section (e.g. Hemleben et al. 1989), hence we interpret this texture as being biogenic in origin. However the specimens from ODP Sites 689 and 1135 show 1–2 micron scale crystals that are similar to those visible on the external surface, revealing that the process of recrystallization is pervasive. Images of the internal chamber surfaces of the Subbotina specimens are shown on Figure 1 in the sixth row. The internal test walls of the wellpreserved specimens from Hampden Beach and ODP Site 647 are smooth whereas the specimens from ODP Sites 689 and 1135 are crystalline at a micron scale. (This was also observed on the other species but the illustrations were omitted because of space constraint.) In summary, the samples from ODP 647 and Hampden Beach meet most of the criteria set out for good foraminiferal preservation: 1. 2.
3.
4.
the tests are translucent and ‘glassy’ under RLM both when dry and to an even greater extent when immersed in water; no ultrafine features such as spines are seen on these specimens but, since they are only seen in some species and during restricted parts of the life cycle, their absence does not indicate poor preservation; under SEM, originally smooth parts of the test, particularly the sutures, apertural lips and inner surface are still smooth on a submicron scale; and the submicron microgranular texture can be identified in cross section showing that substantial recrystallization has not occurred on a micron scale within the test. However, a small amount of recrystallization cannot be ruled out.
In contrast, the tests we have examined from ODP Sites 689 and 1135 are partly dissolved and pervasively recrystallized with neomorphic calcite. We
FORAMINIFER TEST PRESERVATION
relate the good state of preservation in the former sites to the high clay content of the sediment, which is related to their hemipelagic settings and the rapid sedimentation rates. The poorer preservation comes from sites with very high carbonate contents and greater permeability. At the present time we do not know exactly why recrystallization is inhibited in the clay-rich sites, and more work is needed to develop a quantitative physicochemical model of the diagenetic process.
Implications for geochemical studies Two sites have previously been studied on Maud Rise to obtain Cretaceous and Palaeogene isotopic records, namely ODP Sites 689 and 690 (Barrera & Huber 1990; Stott & Kennett 1990; Stott et al. 1990; Huber et al. 1995; Diester-Haass & Zahn 1996). These studies have been invaluable in revealing aspects of high latitude climate history, including long-term trends in temperature and productivity, short-term events including the isotope excursions at the Cretaceous/Palaeogene (Stott & Kennett 1990) and Palaeocene/Eocene boundary (Stott et al. 1990), and the presence orbital cycles in the 100 kyr and 40 kyr wavebands (Diester-Haass & Zahn 1996). Barrera & Huber (1990), Stott & Kennett (1990) & Huber et al. (1995) have illustrated foraminifer test preservation. Their illustrations show that, at high magnification, the tests show signs of recrystallization. Our work underlies the need to interpret the magnitude of isotopic excursions and the downcore trends with caution. Geochemical work at the Kerguelen Plateau sites has mostly been conducted on ODP Site 748 and 744 (Zachos et al. 1992a, b; Salamy & Zachos 1999). The lithology of these sites is deep-water nannofossil ooze, as it is at ODP Site 1135, and the preservation state appears similar to Site 1135 under RLM. The Kerguelen records have also been valuable in highlighting aspects of Palaeogene climate, especially the early Oligocene glacial maximum where there is an isotope excursion at the same level as a peak in Ice-Rafted Debris (Zachos et al. 1992b). To date no geochemical work has been published on the record from ODP Site 1135. As for Maud Rise above, the interpretation of the geochemical records should take into account the recrystallized nature of the material. Our observations suggest very good preservation of foraminiferal calcite at ODP Site 647, which may potentially provide a highly expanded record of Eocene climate change in the North Atlantic. However investigations need to proceed with caution in the light of geochemical work carried out in the wake of ODP Leg 105 by scientists
69
associated with that cruise. Nielsen et al. (1989) analyzed the sedimentary facies, mineralogy and geochemistry of the Palaeogene section at ODP Site 647. They found that approximately the bottom 30 m of the section (from approximately 670–700 mbsf) had been affected by diagenetic alteration which they interpreted as hydrothermal alteration originating from the basaltic basement. These effects resulted in an increase in the chlorite content and iron and manganese concentration of the sediment in the bottom part of the core, suggesting that it should be avoided for geochemical work. Above that level, the sediment composition is much more constant and is similar to that expected from deep marine clays; therefore it does not seem to have been affected by the same process, although complex diagenetic concretions occur through much of the Eocene succession (Bohrmann & Thiede 1989). The sample discussed here is from 479.53 mbsf which is well above the zone of alteration recorded by Nielsen et al. (1989). More puzzling is the geochemical and stable isotope study of Arthur et al. (1989), especially their oxygen and carbon isotope analyses of selected species of planktonic and benthic foraminifers in the Eocene and across the Eocene/Oligocene boundary. Arthur et al. (1989) found that there was little difference in the oxygen isotope ratios of the planktonic species ‘Globigerina’ (¼Subbotina) eocanea and Catapsydrax unicavus, whereas in other sites they are often isotopically distinct (with C. unicavus representing a truly deep-water planktonic environment in contrast to the thermocline habitat of S. eocaena). This could be explained because there was little temperature stratification in the Labrador Sea water mass at the time (Arthur et al. 1989). However, while planktonic foraminifer isotope values are roughly what might be expected from this location (Arthur et al. 1989), the benthic foraminifer values are surprising: most measurements are more negative than the planktonics. Arthur et al. (1989) considered the idea of a temperature inversion in the water column (but dismissed this as unlikely), and suggested instead that an unusual diagenetic regime had operated, in which the benthic tests had been preferentially recrystallized. However it is not obvious why the benthic foraminifera alone should be recrystallized, nor why their tests should have become more negative with respect to d18O under these conditions. Furthermore Arthur et al. observed that test preservation of both planktonics and benthics is usually good by visual inspection. We suggest a third explanation for the isotope data, namely that the tests are all generally very well-preserved and faithfully record the original geochemical fingerprint. The lack of the expected oxygen isotope differential between planktonics and benthics can
70
P. N. PEARSON & C. E. BURGESS
be explained by: (1) only deep-dwelling species of planktonic foraminifera were analyzed by Arthur et al. (1989); or (2) there may have been little difference in the original temperatures of calcification between benthics and deep planktonics, because the site is not in very deep water and the vertical temperature gradient may have been shallow. Under these conditions, species-specific vital effects among the benthic foraminifera, which are poorly known in the Palaeogene, could cause a reversal of the expected planktonic-benthic isotope differentials. The greater scatter in the benthic isotope data might reflect fluctuation in deepwater sources. However we agree with Arthur et al. (1989, p. 134) that more detailed geochemical studies may be necessary to resolve these issues. Observations under RLM and SEM show that samples from Hampden Beach have generally good to excellent preservation with only occasional minor dissolution and no recrystallization of foraminifer tests. This suggests that the foraminifera will record a primary isotopic signal rather than being overprinted by inorganic calcite during diagenesis. This makes the site potentially important for climatic studies, as the only Eocene clay site from the southern high latitudes that is known to yield well-preserved foraminifera. The site at Hampden Beach was extensively sampled during field seasons in 2002 and 2005. Isotopic and trace element studies are currently in progress and it is hoped that they will provide a Southern Ocean climate record for the middle Eocene that is less affected by diagenesis than those currently in existence. Together with ODP Site 647 in the northern hemisphere, these sites may help in the reconstruction of global climate gradients in the Eocene. We thank Hugh Morgans and colleagues at the Institute of Geological and Nuclear Sciences, Lower Hutt, New Zealand, for assistance in the field at Hampden Beach.
References A RTHUR , M. A., D EAN , W. E., Z ACHOS , J. C., K AMINSKI , M., H AGERTY R EIG , S & E LMSTROM , K. 1989. Geochemical expression of early diagenesis in middle Eocene – lower Oligocene pelagic sediments in the southern Labrador Sea, Site 647, ODP Leg 105. In: S RIVASTAVA , S. P., A RTHUR , M., C LEMENT , B. ET AL . (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 105, 111–136. B ARKER , P. F., K ENNETT , J. P. ET AL . 1988. Proceedings of the Ocean Drilling Program, Initial Reports, 113, College Station, TX (Ocean Drilling Program). B ARRERA , E. & H UBER , B. T. 1990. Evolution of Antarctic waters during the Maestrichtian: foraminifer oxygen and carbon isotope ratios, Leg 113. In:
B ARKER , P. F., K ENNETT , J. P. ET AL . (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 113, 813– 828. B E´ , A. W. H. 1980. Gametogenic calcification in a spinose planktonic foraminifer, Globigerinoides sacculifer (Brady). Journal of Foraminiferal Research, 10, 117–128. B E´ , A. W. H., H EMLEBEN , C., A NDERSON , O. R. & S PINDLER , M. 1979. Chamber formation in planktonic foraminifera. Micropalaeontology, 5, 283–310. B ENWAY , H. M., H ALEY , G. P., K LINKHAMMER , G. P. & M IX , A. C. 2003. Adaptation of a flow-through leaching procedure for Mg-Ca paleothermometry. Geochemistry, Geophysics, Geosystems, 4, 8403, doi:10.1029/2002GC000312. B ERGGREN , W. A., P EARSON , P. N. & H UBER , B. T. 2006. Taxonomy, biostratigraphy and phylogeny of Eocene Acarinina. In: P EARSON , P. N. ET AL . (eds) Atlas of Eocene Planktonic Foraminifera. Cushman Foundation Special Publication, No 41. B LOW , W. H. 1979. The Cainozoic Globigerinida. E. J. Brill, Leiden. B OERSMA , A., S HACKLETON , N. J. & G IVEN , Q. 1979. Carbon and oxygen isotopic records at DSDP Site 384 (North Atlantic) and some Paleocene paleotemperatures and carbon isotope variations in the Atlantic Ocean. Initial Reports of the Deep Sea Drilling Project, 43, 695–717. B OERSMA , A., P REMOOLI S ILVA , I. & S HACKLETON , N. J. 1987. Atlantic Eocene planktonic foraminiferal paleohydrographic indicators and stable isotope paleoceanography. Paleoceanography, 2, 287– 331. B OHRMANN , G. & T HIEDE , J. 1989. Diagenesis in Eocene claystones, ODP Site 647, Labrador Sea: Formation of complex authigenic carbonates, smectites and apatite. In: S RIVASTAVA , S. P., A RTHUR , M., C LEMENT , B. ET AL . Proceedings of the Ocean Drilling Program, Scientific Results, 105, 137–154. C OFFIN , M. F., F REY , F. A., W ALLACE , P. J. ET AL . 2000. Proceedings of the Ocean Drilling Program, Initial Reports, 183 (online). Available from World Wide Web: ,http://www-odp.tamu.edu/publications/ 183_IR/183ir.htm. C OXALL , H. K. & P EARSON , P. N. 2006. Taxonomy, Biostratigraphy, and Phylogeny of the Hantkeninidae (Clavigerinella, Hantkenina, and Cribrohantkenina). In: P EARSON , P. N. ET AL . (eds) Atlas of Eocene Planktonic Foraminifera. Cushman Foundation Special Publication, 41, 205– 246. D IESTER -H AASS , L. & Z AHN , R. 1996. EoceneOligocene transition in the Southern Ocean: history of water mass circulation and biological productivity. Geology, 24, 163– 166. F INLAY , H. J. 1939. New Zealand Foraminifera: the occurrence of Rzehakina, Hantkenina, Rotaliatina and Zeauvigerina. Transactions of the Royal Society of New Zealand, 68, 538– 542. H ASTINGS , D. W., R USSELL , A. D. & E MERSON , S. R. 1998. Foraminiferal magnesium in Globigerinoides sacculifer as a paleotemperature proxy. Paleoceanography, 13, 161– 169. H EMLEBEN , C., B E´ , A. W. H., A NDERSON , O. R. & T UNTIVATE -C HOY , S. 1977. Test morphology, organic layers and chamber formation of the planktonic
FORAMINIFER TEST PRESERVATION foraminifer Globorotalia menardii (d’Orbigny). Journal of Foraminiferal Research, 7, 1– 25. H EMLEBEN , C., A NDERSON , O. R., B ERTHOLD , W. U. & S PINDLER , M. 1986. Calcification and chamber formation in foraminifera – a brief overview. In: L EADBEATER , B. S. C. & R IDING , R. (eds) Biomineralization in Lower Plants and Animals. The Systematics Association, Special Volume, 30, 237–249. H EMLEBEN , C., S PINDLER , M. & A NDERSON , O. R. 1989. Modern Planktonic Foraminifera. SpringerVerlag, New York. H EMLEBEN , C. & O LSSON , R. K. 2006. Wall textures of Eocene planktonic foraminifera. In: P EARSON , P. N. ET AL . (eds) Atlas of Eocene Planktonic Foraminifera. Cushman Foundation Special Publication, 41, 47– 66. H UBER , B. T., H ODELL , D. A. & H AMILTON , C. P. 1995. Middle-Late Cretaceous climate of the southern high latitudes: Stable isotopic evidence for minimal equator-to-pole thermal gradients. Geological Society of America Bulletin, 107, 1164–1191. J ENKINS , D. G. 1971. New Zealand Cenozoic Planktonic Foraminifera. New Zealand Geological Survey, Paleontological Bulletin, No. 42, 1 –278. N IELSEN , O. B., C REMER , M., S TEIN , R., T HIEBAULT , F. & Z IMMERMAN , H. 1989. Analysis of sedimentary facies, clay mineralogy, and geochemistry of the Paleogene sediments of Site 647, Labrador Sea. In: S RIVASTAVA , S. P., A RTHUR , M., C LEMENT , B. ET AL . Proceedings of the Ocean Drilling Program, Scientific Results, 105, 101–110. N ORRIS , R. D. & W ILSON , P. A. 1998. Low latitude seasurface temperatures from the mid-Cretaceous and the evolution of planktic foraminifera. Geology, 26, 823–826. O LSSON , R. K., H EMLEBEN , C., H UBER , B. T. & B ERGGREN , W. A. 2006. Taxonomy, biostratigraphy, and phylogeny of Eocene Globigerina, Globoturborotalita, Subbotina, and Turborotalita. In: P EARSON , P. N. ET AL . (eds) Atlas of Eocene Planktonic Foraminifera. Cushman Foundation Special Publication, 41, 111–168. P EARSON , P. N., S HACKLETON , N. J. & H ALL , M. A. 1993. Stable isotope paleoecology of middle Eocene planktonic foraminifera and multi-species isotope stratigraphy, DSDP Site 523, South Atlantic. Journal of Foraminiferal Research, 23, 123–140. P EARSON , P. N., D ITCHFIELD , P. W., S INGANO , J., H ARCOURT -B ROWN , K. G., N ICHOLAS , C. J., OLSSON, R. K., S HACKLETON , N. J. & H ALL , M. A. 2001. Warm tropical sea surface temperatures in the Late Cretaceous and Eocene epochs. Nature, 413, 481–487. P EARSON , P. N., D ITCHFIELD , P. W. & S HACKLETON , N. J. 2002. Tropical temperatures in greenhouse episodes - reply. Nature, 419, 898–898. P EARSON , P. N., VAN D ONGEN , B. E., N ICHOLAS , C. J., P ANCOST , R. D., S CHOUTEN , S., S INGANO , J. M. & W ADE , B. S. 2007. Stable warm tropical climate through the Eocene epoch. Geology, 35, 211–214. P INGITORE , N. E. 1982. The role of diffusion during carbonate diagenesis. Journal of Sedimentary Petrology, 52, 27– 39.
71
P OORE , R. Z. & M ATTHEWS , R. K. 1984. Oxygen isotope ranking of Late Eocene and Oligocene planktonic foraminifers: implications for Oligocene sea-surface temperatures and global ice volume. Marine Micropaleontology, 9, 111–134. P REMOLI S ILVA , I., W ADE , B. S. & P EARSON , P. N. 2006. Taxonomy, biostratigraphy and phylogeny of Globigerinatheka and Orbulinoides. In: P EARSON , P. N. ET AL . (eds) Atlas of Eocene Planktonic Foraminifera. Cushman Foundation Special Publication, 41, 169– 212. S ALAMY , K. A. & Z ACHOS , J. C. 1999. Latest Eocene – Early Oligocene climate change and Southern Ocean fertility: Inferences from sediment accumulation and stable isotope data. Palaeogeography, Palaeoclimatology, Palaeoecology, 145, 61–77. S EXTON , P. F., W ILSON , P. A. & P EARSON , P. N. 2006. Microstructural and geochemical perspectives on planktic foraminiferal preservation: ‘Glassy’ versus ‘Frosty’. Geochemistry, Geophysics, Geosystems, 7, Q12P19, doi:10.1029/2006GC001291. S PERO , H. J. & L EA , D. W. 1993. Intraspecific stable isotope variability in the planktic foraminifera Globigerinoides sacculifer – results from laboratory experiments. Marine Micropaleontology, 22, 221–234. S RIVASTAVA , S. P., A RTHUR , M. ET AL . 1987. Proceedings of the Ocean Drilling Program, Initial Reports, 105, College Station, TX (Ocean Drilling Program). S TEWART , D. R. M., P EARSON , P. N., D ITCHFIELD , P. W. & S INGANO , J. M. 2004. Miocene tropical Indian Ocean temperatures: evidence from three exceptionally preserved foraminiferal assemblages from Tanzania. Journal of African Earth Sciences, 40, 173– 190. S TOTT , L. D. & K ENNETT , J. P. 1990. The paleoceanographic and paleoclimatic signature of the Cretaceous/Paleogene boundary in the Antarctic: Stable isotope results from ODP Leg 113. In: B ARKER , P. F., K ENNETT , J. P. ET AL . Proceedings of the Ocean Drilling Program, Scientific Results, 113, 829–848. S TOTT , L. D., K ENNETT , J. P., S HACKLETON , N. J. & C ORFIELD , R. M. 1990. The evolution of Antarctic surface waters during the Paleogene: inferences from the stable isotopic composition of planktonic foraminifers, ODP Leg 113. In: B ARKER , P. F., K ENNETT , J. P. ET AL . Proceedings of the Ocean Drilling Program, Scientific Results, 113, 849 –864. W ILLIAMS , M., H AYWOOD , A. M., T AYLOR , S. P., V ALDES , P. J., S ELLWOOD , B. W. & H ILLENBRAND , C.-D. 2005a. Evaluating the efficacy of planktonic foraminifer calcite d18O data for sea surface temperature reconstruction for the Late Miocene. Geobios, 38, 843– 863. W ILLIAMS , M., H AYWOOD , A. M., H ILLENBRAND , C.-D. & W ILKINSON , I. P. 2005b. Efficacy of d18O data from Pliocene planktonic foraminifer calcite for spatial sea surface temperature reconstruction: comparison with a fully coupled ocean-atmosphere GCM and fossil assemblage data for the mid-Pliocene. Geological Magazine, 142, 399–417. W ILSON , G. J. 1985. Dinoflagellate biostratigraphy of the Eocene Hampden Section, North Otago, New Zealand. New Zealand Geological Survey Record, 8, 93–101.
72
P. N. PEARSON & C. E. BURGESS
W ILSON , P. A. & N ORRIS , R. D. 2001. Warm tropical ocean surface and global anoxia during the midCretaceous period. Nature, 412, 425–429. W ILSON , P. A., N ORRIS , R. D. & C OOPER , M. J. 2002. Testing the Cretaceous greenhouse hypothesis using glassy foraminiferal calcite from the core of the Turonian tropics on Demerara Rise. Geology, 30, 607– 610. Z ACHOS , J. C., B ERGGREN , W. A., A UBRY , M. P. & M ACKENSEN , A. 1992a. Isotope and trace element
geochemistry of Eocene and Oligocene foraminifers from Site 748, Kerguelen Plateau. In: W ISE , S. W. J R ., S CHLICH , R. ET AL . (eds) Proceedings of the Ocean Drilling Program, Scientific Results, 120, 839–854. Z ACHOS , J. C., B REZA , J. R. & W ISE , S. W. 1992b. Early Oligocene ice-sheet expansion on Antarctica: Stable isotope and sedimentological evidence from Kerguelen Plateau, Southern Ocean. Geology, 20, 569–573.
The influences of growth rates on planktic foraminifers as proxies for palaeostudies – a review D. N. SCHMIDT1, T. ELLIOTT1 & S. A. KASEMANN2 1
Department of Earth Sciences, University of Bristol, Wills Memorial Building, BS8 1RJ, Bristol, UK (e-mail:
[email protected]) 2
Grant Institute of Earth Science, The University of Edinburgh, The King’s Buildings, West Mains Road, EH9 3JW, Edinburgh, UK Abstract: Size dependent changes in element concentrations in planktic foraminifers have long been recognized to influence their reliability as an archive for climate change. Traditionally, these changes have been interpreted as changes in element partitioning during the ontogeny of the organism with faster growth rates in the earlier part of the development. These changes, in the light of new culture experiments, can also be interpreted as changes in growth rate throughout the entire life of the organism, with larger, faster-growing specimens discriminating less efficiently against trace element incorporation into the calcite shell. Growth rates of foraminifera are influenced by the environment and hence change geographically and temporally at various scales, e.g. glacial-interglacial or rapid millennial events such as the Younger Dryas. These changes in growth rate can account for some changes in element to calcium ratio between glacial and interglacials, which were previously linked to changes in seawater element ratios.
Environmental variables such as temperature, salinity or nutrients cannot be directly measured for the past, but are determined via proxy data (see Fischer & Wefer 1999). The minute tests of planktic foraminifera represent one of our best archives of these proxies. Their excellent preservation, global occurrence and high abundance in carbonate rich marine sediments above the lysocline are prime reasons for their extensive use in palaeoceanographic and palaeoclimatic studies. Analyses of elemental and isotopic compositions in addition to species abundances (see Fischer & Wefer 1999 for review) have proven to be valuable proxies to understand past climate. However, the relationship between environmental parameters and the chemical compositions of foraminiferal tests are cannot be fully modelled using simple, inorganic equilibria, and are at least partly-controlled by less well understood process related to the biology of the foraminifers (Erez 2003; Weiner & Dove 2003). Hence, quantifying ecological and biological effects on proxy incorporation into foraminifers is vital for the understanding of the reliability of species as records of ocean conditions. For example, carbon isotopes in foraminifers change with different size fractions (Spero & DeNiro 1987). Oppo & Fairbanks (1989) and Ravelo & Fairbanks (1995) showed that the difference in d13C between size factions in a single sample is greater than the glacial-interglacial amplitude. This difference in d13C at specific sizes can create significant bias if the size fraction of the
analysed specimens changes due to the absence of large specimens in glacial samples. Along the same lines, Brown & Elderfield (1996) and Elderfield et al. (2002) documented a change in trace element ratios with increasing size. Most trace element abundances relative to calcium are significantly greater in seawater than in any part of the foraminiferal test. Elemental partition coefficients (D) of trace elements between foraminiferal calcite and seawater are much smaller than those predicted by laboratorydetermined partition coefficients for inorganic calcite in seawater (see Elderfield et al. 1996 and references therein for various elements and isotopes). These small partition coefficients suggests that the organism actively discriminates against uptake of these elements from seawater (Brown & Elderfield 1996; Erez 2003; Eggins et al. 2004) emphasising a strong biological control on foraminiferal calcite precipitation (see Bentov & Erez 2006 for a comprehensive review). This discrimination does not seem to be constant (Fig. 1) since element ratios change with the test size of the foraminifer (Elderfield et al. 2002). The distribution coefficients suggest calcite precipitation that follows a Raleigh-like distillation model from a partially restricted reservoir (Elderfield et al. 1996; Erez 2003). Several mechanisms have been used to explain such test size related variability in isotopic and trace element compositions, e.g. incorporation changes during growth with faster growing early
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 73–85. DOI: 10.1144/SP303.6 0305-8719/08/$15.00 # The Geological Society of London 2008.
74
D. N. SCHMIDT ET AL.
(a) 8.5
8.0
(b) 1.44
Mg/Casw 5145 mmol/mol
Sr/Ca [mmol/mol]
Mg/Ca [mmol/mol]
7.5 7.0 6.5 6.0 5.5
1.40 1.38 1.36 1.34
5.0
1.32
4.5
1.30
4.0 200
300
400 500 600 Size fraction [µm]
Sr/Casw 8 mmol/mol
1.42
700
1.28 200
300
400 500 600 Size fraction [µm]
700
Fig. 1. (a) Bulk analyses of Mg/Ca ratios in different size classes of Orbulina universa from Atlantic core top samples (open circles) (Elderfield et al. 2002) and culture experiments (filled squares) (Russell et al. 2004). The cultures of O. universa by Russell et al. (2004) were maintained at 228 and 301 [CO22 3 ] [mmol/kg] to exclude the influences of temperature or carbonate ion on the trace element incorporation. The change in Mg/Ca in the cultures would equate to a temperature change of c. 2 8C. (b) Culture experiments on O. universa show a similar increase in Mg/Ca ratios with increasing size, while the Sr/Ca ratio in the culture experiments is opposite to the results found in sediment samples.
chambers and slower growing later chambers (Elderfield et al. 2002). Empirical partitioning coefficients for Sr based on culture experiments of benthic foraminifers decrease with increasing growth rates, in contrast to predicted relationships based on inorganic growth (Erez 2003). In this review, we will discuss the potential influence of changing average growth rates of foraminifers on trace element incorporation into foraminiferal calcite and its consequences for our ability to quantify past climate.
Spatial heterogeneity of trace elements in foraminiferal tests Planktic foraminifers grow by adding chambers on average once every 24– 48 h in their adult stage (Hemleben et al. 1989). Foraminifer calcify out of an internal seawater reservoir (Anderson & Faber 1984) the size of which is related to the size of the organism (Erez 2003). Trace elements have to pass through the membrane that envelopes the seawater containing vacuole to get to the site of calcification (see Erez 2003 for a recent extensive review). Calcification of the chamber wall starts with small plaques of calcite on the primary organic membrane (POM) which coalesce to form the wall (Hemleben et al. 1986). The POM is the site of initial calcification and serves as the template.
The organic layer must have two active surfaces since calcification happens on both sides of the POM creating a bilamellar wall (Fig. 1). The inner lamella forms the interior of the new chamber and the outer one covers the new chamber and the rest of the test (Hemleben et al. 1986). Each subsequent calcite layer is preceded by the formation of an organic layer (Hemleben et al. 1977 and Fig. 2b). Growth towards the inside is minor while layers are added periodically toward the outside (Hemleben 1969). Each time a new chamber is formed the older chambers are covered by new calcite, resulting in a thick wall of calcite between successive organic layers (Fig. 2; Hemleben et al. 1986). Thus, the history of calcification events is reflected by pattern of the calcite layers, including their number and thickness. Different chemical compositions are found in the different layers of the foraminiferal calcite (Nu¨rnberg et al. 1996; Eggins et al. 2003; Gehlen et al. 2004; see also Figs 3 & 4), in part because they are formed at different water depths where temperature, illumination and water chemistry vary, and in part because the mechanisms of calcification may differ (Erez 2003). The calcite directly adjacent to the POM has different concentrations for some elements than the bulk of the calcite e.g. higher Mg concentrations (Fig. 3). It has been suggested that organic matrices may alter the partition coefficient of ions in the biogenic calcite and therefore, trace element concentrations may
INFLUENCES OF GROWTH RATES ON PLANKTIC FORAMINIFERS
Fig. 2. Backscatter electron image of Globorotalia truncatulinoides from the mid-latitude South Atlantic showing the lamellar construction of foraminiferal test. (a) Overview of the foraminiferal calcite, indicating the multilayered structures of the foraminiferal calcite. Highlighted areas indicate enlargement of the chamber fourth from last (ch f-4 (b)) and penultimate chamber (c). Dark areas, of low reflectance, indicate grain boundaries and organic material (POM ¼ primary organic membrane, oOM ¼ outer organic membrane). Since the samples have not been chemically cleaned, the inside of the test is covered by sediment. The numbers on the layers indicate the generations of carbonate covering during growths, cc indicates the calcite crust/gametogenetic calcite. The last chamber shows no gametogenetic calcite, whereas ch f-4 has a thick gametogenetic cover. Notice the significant differences in porosity and crystal size in the various parts of the test.
75
be different near the organic layers (Bentov & Erez 2006). It has also been put forward that calcification in foraminifers has a transient amorphous calcite phase similar to other organisms (Politi et al. 2004) resulting in a change in partitioning coefficients and precipitation of a high Mg phase at the beginning of chamber formation (Bentov & Erez 2006). The outermost layer, the gametogenic calcite, has different concentrations for some elements than the ontogenetic proportion of the calcite wall. Magnesium calcite ratios, e.g. are lower in the gametogenic calcite than the ontogenetic (Fig. 3; Eggins et al. 2003; Eggins et al. 2004; Klinkhammer et al. 2004; Anand & Elderfield 2005; Sadekov et al. 2005). An earlier study indicated the reverse trend of Mg/Ca between gametogenic and ontogenetic calcite (Nu¨rnberg 1995; Nu¨rnberg et al. 1996), but has not been reproduced by subsequent studies. It is also notable that high resolution work has indicated an extremely high Mg/Ca outermost rim that is attributed to surficial contamination. The relative proportions of the chemicallydistinct layers in the foraminiferal test have significant influences on the average trace element/Ca ratios. Some species, such as G. ruber, do not deposit gametogenic calcite and are therefore chemically more homogeneous than species with gametogenic calcite (Eggins et al. 2003, Fig. 4). Other species have large amounts of gametogenic calcite, e.g. up to 250% in O. universa (Caron et al. 1990). However, the amount of gametogenic calcite reported in the literature can vary even for the same species e.g. 30% (Be´ 1980) vs. 70 –115% (Caron et al. 1990) in G. sacculifer. This variability is evident in the relationship between size and weight of O. universa (Fig. 5; Russell et al. 2004). At sizes around 600 mm the variability in weight at a given test size indicates variable test thickness. The thickness of the test may influence the relative proportions of gametogenic to ontogenetic calcite in the foraminiferal test. To address this question, we have measured some cross section of G. truncatulinoides (see Fig. 2 for an example) and determined the size of the specimen, the thickness of the gametogenic calcite and the thickness of the wall (Fig. 6). These measurements are rare since the preparation of these cross-sections is time consuming. Though there is evidence that the thickness of the gametogenic calcite layer increases with size for G. truncatulinoides (Fig. 6), the ratio between the thickness of the GAM and the test size does not show a trend for this species. The proportional contribution, however, of the gametogenic to nongametogenic calcite changes with increasing test weight per unit length (Caron et al. 1990). Small
76
D. N. SCHMIDT ET AL.
Fig. 3. Electron microprobe elemental map of magnesium (left) and strontium (middle) concentrations of the penultimate chamber (ch f-1) of G. truncatulinoides. Bright areas indicate high concentrations while dark areas, most notably the outer gametogenic layer, indicate low concentrations. The backscatter electron picture (right) of the same area clearly shows the position of two outer organic membranes (oOM) building the base of the final ontogenetic and for the gametogenic calcite (GAM), while the primary organic membrane is less distinct and tentatively placed using the faintly dotted line. The dark perpendicular lines crosscutting the wall are sections along pores.
individuals will have a higher relative proportion of gametogenic calcite compared to large specimen. Considering the low Mg concentrations in the gametogenic calcite, small individuals are consequently expected to have lower Mg/Ca ratios than larger ones. The gametogenic and ontogenetic calcite are not just geochemically but also crystallographically distinct. The main body of calcite crystals have poorly-defined crystal faces (anhedral) that mainly grow along (0001) while the final outer calcite layer has rhombohedral structures with faces mainly along (1014) (Hemleben et al. 1977).
Inorganic calcite experiments have shown that this rhombohedral face has up to 4 times lower Mg concentrations that the other faces, while the Sr concentrations are inversely correlated with crystal growth (Paquette & Reeder 1990) although the variations are not as strong as for Mg (Reeder & Grams 1987). If non-equivalent faces advance at different rates then the style of crystal growth can influence elemental concentrations in foraminiferal calcite (Paquette & Reeder 1990; Reeder et al. 1997). Currently, there is no information on the relationship between crystal growth rates and the growth rates of the foraminifers but a relationship between the
2.5
7.0
G. sacculifer
6.0 5.0 4.0
Sr/Ca [mmol/mol]
Mg/Ca [mmol/mol]
8.0
2.0 G. ruber 1.5 G. sacculifer
1.0 0.5
G. ruber 3.0
0.0 1
3
outside
5
7 9 11 13 Depth [µm]
15
17
inside
1
3
outside
5
7 9 11 13 Depth [µm]
15
17
inside
Fig. 4. Element to calcium ratios in Globigerinoides sacculifer and G. ruber determined by laser ablation ICPMS indicating the spatial heterogeneity in Mg concentrations in G. sacculifer but not in G. ruber (modified after Eggins et al. 2003). The Mg concentrations are higher in the inner part of G. sacculifer but constant in G. ruber which lacks Mg-poor gametogenic calcite. In contrast, Sr concentrations are relatively stable throughout the test and very similar for both species. Note that the 1 mm layer with very high Mg values, which we interpret as contamination on the outside of the test, has been excluded from this plot to highlight the internal variability. The different duration of the analysis is determined by the thickness of the foraminiferal test.
INFLUENCES OF GROWTH RATES ON PLANKTIC FORAMINIFERS
Variations in growth rate and its potential effect on trace element incorporation
700 650
Size [µm]
600 550 500 450 400 350 300 0
20
40
60
80
100
Weight [µg] Fig. 5. Size weight relationship of Orbulina universa in culture experiments indicating the increase in weight with increasing size and the higher variability in this relationship at sizes larger 600 mm (Russell et al. 2004).
environment and crystal growth has been suggested (Towe & Cifelli 1967). Furthermore, it has been suggested that low temperatures lead to slow crystal growth resulting in ‘good crystallinity’ while at warm temperatures crystal growth is fast resulting in ‘low crystallinity’ (Meliers 1977). A potential link between crystallinity, temperature and trace element incorporation might help us to disentangle biotic from abiotic influences on trace element incorporation.
Recent multi-proxy reconstructions of past climates indicate that other factors influence trace element information. Palaeotemperature estimates, based on alkenones, the composition of microfossil assemblages and Mg/Ca ratios, disagree with each other in complex ecosystems such as the coastal upwelling areas or high latitude regions (Kucera et al. 2005). These environmental settings influence physiological processes which influence trace element incorporation into foraminiferal calcite (see Erez 2003 for review). Amongst these processes are respiration of the foraminifer or photosynthesis of the symbionts. Symbiont activity, mainly by dinoflagellates, can be responsible for major changes in the chemistry of the microenvironment around the foraminiferal test (e.g. Jørgensen et al. 1985; Zeebe et al. 2003). Furthermore Ortiz et al. (1996) inferred that isotopic disequilibrium increases with metabolic rates related to temperature and food supply both highly variable in these settings. These environments are also exceptional since foraminifers tend to be much smaller than inferred from global temperature-size relationships and hence indicate that these environments limit growth of foraminifers (Schmidt et al. 2004a) and potentially affect elemental incorporation. It has been suggested that growth rates influence proxy incorporation (Elderfield et al. 1996). Growth rate controlled element incorporation has been documented in biogenic (Rickaby et al. 2002) and inorganic carbonate minerals (Lorens 1981).
45
0.07
40
0.06
35 Ratio of GAM/size
Thickness of GAM [µm]
77
30 25 20 15
0.05 0.04 0.03 0.02
10 0.01
5 0 300
400
500 Size [µm]
600
700
0.00 300
400
500 Size [µm]
600
700
Fig. 6. Relationship between thickness of gametogenic calcite (GAM) and test size in several specimens of G. truncatulinoides from the sample GeoB1726 (South Atlantic). The thickness was measured on the f-2 chamber of cross sections through the test of nine specimens (see Fig. 2 for an example).
78
D. N. SCHMIDT ET AL.
In the past, however, growth rate attributed changes in element incorporation have been linked to changing calcification rates during the life cycle of the organisms and rarely attributed to different overall growth rate changes due to variable environmental conditions. Here, we focus on changes in overall foraminiferal growth rates and how these influence the incorporation of trace elements.
The relationship between element concentrations and test size Changes in element/calcium ratios in specimens of different sizes from the same samples are traditionally interpreted to reflect ontogenetic changes, i.e. small specimens to reflect younger specimens with premature death. Most specimens form sediment sample however display gametogenic calcite, which implies they are adults that have undergone gametogenesis. Foraminiferal growth ends at reproduction, during which gametes are released and the empty test sinks to the ocean floor. In most planktic foraminifers reproduction is triggered by the synodic lunar cycle (Spindler et al. 1979; Bijma et al. 1990a; Schiebel et al. 1997). Assuming such externally triggered reproduction, different final test sizes suggest different average growth rates during the overall life. The final size is thus determined by environmental factors that control the growth rate of the foraminifers during a lunar cycle (Caron et al. 1981; Caron et al. 1987a; Bijma et al. 1990b). Similarly, most specimens cultured to analyse the relationships between environmental parameters and palaeoproxies are reported to have undergone gametogenesis (e.g. Lea et al. 1999; Russell et al. 2004). Therefore, similar to the sediment analogues, larger-sized specimens cultured under the same environmental conditions are likely to result from faster growth. The culture study of Russell et al. (2004) reported size and weight of their cultured specimens and hence allows us to test our hypothesis that the overall growth rate during the life of the specimen may influence trace element incorporation. All their specimens have undergone gametogenesis and, hence, have reached the adult reproductive phase of their life cycle and do not represent different ontogenetic stages. Figure 1 (squares) shows a subset of their analysis, cultured at 22 8C and 301 mmol/kg [CO22 3 ]. The specimens reached final size between 427 – 653 mm (Russell et al. 2004) and show an increase in Mg/Ca and Sr/Ca ratios with increasing size (Fig. 1). Although there is no clear explanation why these individual specimens grew to a different final test size, it is inherent that those reaching
larger size must have grown fastest. The change in Mg/Ca is equivalent to a temperature change of c. 4 8C from 22 – 18 8C using the calibration in Lea et al. (1999). Experiments with Orbulina universa and Globigerina bulloides also indicate increasing Sr/Ca with increasing temperatures and pH (Lea et al. 1999), potentially reflecting a calcification rate effect on Sr incorporation as suggested for inorganic calcite (Lorens 1981; Carpenter & Lohmann 1992). The difference in Sr/Ca ratios would be the equivalent of a 3.3 8C temperature change using the calibration in Lea et al. (1999). Intriguingly, Sr/Ca analysis of O. universa from a sediment sample (BOFS core 31 K, Elderfield et al. 2002) show a decrease in Sr/Ca ratios with increasing size, while the Mg/Ca relationship has the same positive correlation on both, sediments and cultures (Elderfield et al. 2002). This apparent contradiction demands more Sr/Ca culture work to resolve the problem.
Environmental controls on growth rate If the overall growth rate of the foraminifers can influence element uptake, it is important to assess potential factors influencing growth. Several physical and chemical properties of the ambient sea water, such as temperature, carbonate saturation, nutrients and oxygen availability have been shown to influence size (see Schmidt et al. 2004a; Schmidt 2006 for a detailed discussion) (Fig. 7) and weight (Barker & Elderfield 2002; Barker et al. 2004). Every species has a set of environmental conditions which lead to optimum growth (Fig. 7; Bradshaw 1961). Under more optimal environmental conditions, planktic foraminifers will grow faster, achieve larger sizes and will be more abundant (Hecht 1976; Schmidt et al. 2004a). Outside these optima, growth rates will decrease and reproduction will become impossible. A subpolar species, such as one of the Globigerina bulloides ecotypes, will reach its largest size in subpolar waters, whereas Globorotalia menardii attains its maximum size in tropical environments (Fig. 8). The presence of an environmental optimum results in a complex behaviour of size with regards to geography or temperature. The maximum size (and thus overall growth rates) of foraminifera will be attained at a location of the environmental optimum. The size will decrease with either higher or lower temperature and hence, negative and positive size vs. temperature correlations are possible within one species. Because size can be used to characterize environmental optima, the stability of these optima over time and the environmental adaptation in the younger geological past can be monitored using
INFLUENCES OF GROWTH RATES ON PLANKTIC FORAMINIFERS
Upper reproductive limit Upper growth limit
Minimum lethal temp. Growth rate
Growth rate
Lower reproductive limit Lower growth limit
maximum growth rate ‘optimum’
Maximum lethal temperature Warm water species
Cold water species 5 15 25 35 Temperature [°C]
Fig. 7. Stylized temperature dependent growth model for hypothetical warm water and cold water species (Schmidt et al. 2006) based on Bradshaw (1961). The environmental niche is dived into optimum growth, a reproductive range, the minimum temperature for growth and the physiological limit leading to lethal conditions.
size distributions (Schmidt et al. 2003). N. pachyderma, G. bulloides, G. ruber, G. sacculifer, G. tumida, G. menardii and O. universa have had the same optimum temperature adaptation over the last 300 kyrs as today, while both deeper-dwelling species, G. truncatulinoides and G. inflata, had slightly higher optimum temperatures in the last 300 kyrs prior to the Holocene. The change in optimum temperatures is unsurprising. Both, G. inflata and G. truncatulinoides are very recent (latest Pliocene) originations. G. truncatulinoides just recently colonized subpolar waters in two successive phases of expansion at c. 300 and 200 kyr (Kennett 1970; Pharr & Williams 1987) which was associated with two genetic splitting events, i.e. origination of new species, at c. 300 kyrs (warm-water and cold water ‘species’) which were subsequently further differentiated at c. 170 kyrs (warm) (de Vargas et al. 2001) (Fig. 8). and c. 128 kyrs (cold) (Renaud & Schmidt 2003). Therefore, the species analysed in modern samples is not the same one as those analysed during the last 300 kyrs. Because the new species are adapted to colder environments, such as the subpolar Atlantic (Healy-Williams et al. 1985), the modern temperature adaptation of the species complex has changed towards colder temperatures. Temperature influences foraminifers directly, since cell physiology accelerates with temperature and approximately doubles when temperature increases by 10 8C (Caron et al. 1987a, 1987b; Bijma et al. 1990b; Spero et al. 1991). Temperature can also act indirectly on size, e.g. via increased
79
stratification which leads to larger vertical temperature and density gradients and results in higher number of distinct niches and minimized interspecific competition (Schmidt et al. 2004b) all of which allow growth to larger size. Enhanced calcification, i. e. fast biomineralization rates, in foraminifer tests could be a direct consequence of higher carbonate supersaturation (Lohmann 1995; Barker & Elderfield 2002; Bijma et al. 2002). Since CO2 is less soluble in warmer waters, calcium carbonate supersaturation increases from the poles to the tropics (Buddemeier & Fautin 1994), paralleling the observed trend in planktic foraminifera size but also making it difficult to disentangle from metabolically linked temperature effects. Highly variable habitats such as frontal systems and upwelling areas lead to smaller size in foraminifers documented in plankton tows (Ortiz et al. 1995) and sediment samples (Hecht 1974) maybe as a result of high turbulence, frequent eddies (Beckmann et al. 1987). Even symbiont bearing species tend to be smaller in these areas, as increased light attenuation inhibits growth via lowered symbiotic activity (Bijma et al. 1992; Ortiz et al. 1995). High primary productivity leads to a higher oxygen consumption that, combined with the reduced oxygen production by the symbionts, may lead to oxygen deficiency and reduced foraminiferal respiration. Difference in oxygen level may also explain why fertile areas, such as the Zaire river plumes, allow growth to large size (Bijma et al. 1992), since this area has anomalously high oxygen saturation.
Controls on growth rate over glacial interglacial cycles The size of several foraminiferal species has been shown to change during glacial-interglacial cycles e.g. Orbulina universa (Be´ & Duplessy 1976) and G. bulloides (Malmgren & Kennett 1978a, 1978b), mimicking the modern latitudinal environmental size changes (Schmidt et al. 2003). The amplitude of this size change matches the amplitude of palaeoclimatic fluctuations, with significant size changes in highly variable environments and no significant size change in stable environments, such as the Caribbean (see Schmidt et al. 2006 for discussion). Based on our assumption that the final test size represents overall growth rates, larger size implies higher growth rate and vice versa. Martin et al. (1999) interpreted glacialinterglacial changes in Sr/Ca ratios in G. bulloides and N. pachyderma in the sub-antarctic Indian respectively Pacific Oceans to reflect changes in
80
D. N. SCHMIDT ET AL.
1500
1500 N. pachyderma (s)
G. bulloides 1300 Maximum size [µm]
Maximum size [µm]
1300 1100 900 700 500
1100 900 700 500
300 –5
0
5
10 15 20 SST [°C]
25
300 –5
30
1500
10 15 20 SST [°C]
25
30
10 15 20 SST [°C]
25
30
G. menardii 1300 Maximum size [µm]
1300 Maximum size [µm]
5
1500 G. truncatulinoides
1100 900 700 500 300 –5
0
1100 900 700 500
0
5
10 15 20 SST [°C]
25
30
300 –5
0
5
Fig. 8. Sea surface temperature (SST)-size relationship for species with different environmental adaptations (data from Schmidt et al. 2004a): the polar species Neogloboquadrina pachyderma sinistral, the subpolar species Globigerina bulloides, Globorotalia truncatulinoides showing two environmental optima one for the subtropical type (after de Vargas et al. 2001) and one for the temperate type and for the tropical species Globorotalia menardii.
ocean Sr/Ca ratios. Based on the relationship between size and Sr/Ca ratios, this data can be reinterpreted. G. bulloides at a nearby Site (E49-19, Malmgren & Kennett 1978b) show larger size, more optimal environmental conditions and faster average growth rate during glacials. Faster growth rates influence the Sr/Ca incorporation (as seen in the Holocene and in culture experiments) and, may explain part of the differences in Sr/Ca ratios. We have used the size and trace element data by Cronblad & Malmgren (1981) to make a rough assessment of the potential influences of size changes on Sr/Ca ratios (Fig. 9). Their Sites E49-19 in the central Subantarctic Indian Ocean (43853.20 S and 90806.00 E) and E48-22 in the subtropical convergence of the Indian Ocean (39853.70 S
and 85824.60 E) are in a similar environmental setting to the sites analysed by Martin et al. (1999). During glacials, G. bulloides at these sites are larger and Sr concentrations are lower, mimicking the inverse relationship between size and Sr seen in core top samples (Fig. 1; Elderfield et al. (2002). Although we believe the sense of change of Sr/Ca presented by Cronblad & Malmgren (1981) is significant, we note that their Sr/Ca ratios vary by a factor of two between glacials and interglacials in contrast to typical 10% change (e.g. Billups et al. 2004). Instead we examine data from the study of Martin et al. which show that the Sr/Ca ratios of bulloides in RC11-120 vary by 0.08 mmol/mol from 1.28 to 1.36 mmol/mol over the last
INFLUENCES OF GROWTH RATES ON PLANKTIC FORAMINIFERS
Size [µm] 250 270 290 310
Size [µm] 300 320 340 360
0
1 3
Age [kyr]
100
5
200
7
300
400
9
E48-22
E49-19
11
Fig. 9. Size of Globigerina bulloides from E49-19 in the central subantarctic Indian Ocean (43853.20 S and 90806.00 E) and E48-22 in the subtropical convergence zone of the Indian Ocean (39853.70 S and 85824.60 E) (Cronblad & Malmgren 1981).
300 kyrs (Martin et al. 1999). We then used the Sr/ Ca to size relationship of Elderfield et al. (2002) for G. bulloides (size range between 250 and 350 mm, equivalent to the size range analysed by Martin et al. (1999)) to calculate Sr/Ca variations potentially linked to size changes [Sr/ Ca ¼ 20.0005*size þ 1.5238]. The calculated size effect on Sr/Ca is 0.03 mmol/mol and can therefore explain 40% of the glacial-interglacial variation seen in the Martin et al. (1999) record. We acknowledge that the use of a subtropical calibration conducted on material from 198 N in the Atlantic (Elderfield et al. 2002) is not ideal for the comparison to these subpolar cores due to possible effects of different genotypes (Kucera & Darling 2002; Kucera et al. 2005) and their specific environmental adaptation and/or element incorporation (Steinke et al. 2005). This calculation, however, represents a proof of concept and suggests a need for more rigorous testing.
Outlook Understanding and linking both the organism’s growth rate and crystal growth rates to geochemistry is vitally important for proxy calibration. Looking at a sediment record through time, the growth rates of foraminifers are changing because of the changing environmental conditions (Schmidt et al. 2003). Therefore, picking the same size range, as is usually done, does not represent the same life stage of the foraminifers and
81
potentially not the same mechanisms of trace element incorporation. More importantly, size ranges of species vary dramatically geographically and temporally. Consequently, the size range picked should reflect these size changes. For example Al-Sabouni et al. (2007) suggest to stop using the standard .150 mm sieve size given that size of planktic foraminifera is geographically variable (see Fig. 8). In contrast they suggest to use the .63 mm fraction in polar oceans and the .125 mm elsewhere. In the late Quaternary, it should be tested if adjusting the analysed size range to reflect smaller or larger final size of the species may reduce some noise. Furthermore, the stability of species adaptation and hence their applicability to palaeoclimate questions has just been tested for few species and a restricted time range. It is vitally important for our understanding of critical time intervals older than 300 kyrs to determine the end of the stable environmental adaptation for important proxy carriers. More work, using non-foraminiferal temperature proxies are necessary to identify times and regions where foraminifer are in non-analogue to modern adaptations. This type of work has been started by the MARGO project for the last glacial maximum (Kucera et al. 2005), but it is absolutely necessary to extend the time span analysed to times of large environmental changes such as the Mid-Brunhes event and the Mid-Pleistocene Transition. On glacial interglacial timescales, ecophenotypic size variability and adaptive responses are the most important driving factor for size changes. Our understanding of the effects of exchange of cryptic species in a sediment core through time (Renaud & Schmidt 2003) on palaeoproxies is in its infancy (Huber et al. 1997; Bauch et al. 2003; Steinke et al. 2005). On long time scales, evolutionary size variations become important and may influence elemental incorporation. Foraminifers in the Late Neogene are significantly larger than earlier in the geological records (Schmidt et al. 2004b) with the largest changes in the Late Miocene/Early Pliocene of the subtropics and tropics. These changes have been documented in a number of species (Malmgren et al. 1983; Malmgren et al. 1996; Huber et al. 2000) and may influence their use as proxy carriers on longer time-scales. Although traditional mono-specific measurements of populations of dozens of specimens have small measurement errors, they also average over the wealth of fine scale information stored in the test. In contrast, high-spatial resolution measurements of trace element proxies on individual foraminifera (e.g. Eggins et al. 2003; Hathorne et al. 2003; Eggins et al. 2004; Gehlen et al. 2004; Rathmann et al. 2004) provide an opportunity
82
D. N. SCHMIDT ET AL.
to assess the role of changing life conditions on the reliability of the chemical archive. Such in-situ analyses will provide constraints for models of biomineralisation and will quantify spatial heterogeneity of trace elements and isotopes that is integral to interpretation of less well-preserved material in the geological record where partial dissolution influence the relative proportions of the chemically distinct layers in the foraminiferal calcite and hence the reliability of the climatic interpretation. DNS would like to thank NERC for founding (NE/ B500874/1). We would like to thank Stuart Kerns for help with electron microprobe and Barbara Donner (Bremen) for samples. We would also like to thank Paula McDade (University of Edinburgh) for help with the Scanning Electron Microscope and Michael Hall for the sample preparation.
References A L -S ABOUNI , N., K UCERA , M. & S CHMIDT , D. N. 2007. Vertical niche separation control of diversity and size disparity in planktonic foraminifera. Marine Micropaleontology, 63, 75– 90. A NAND , P. & E LDERFIELD , H. 2005. Variability of Mg/Ca and Sr/Ca between and within the planktonic foraminifers Globigerina bulloides and Globorotalia truncatulinoides. Geochemistry, Geophysics, Geosystems, 6, Q11D15, http://dx.doi.org/10.1029/2004GC000811. A NDERSON , O. R. & F ABER , W. W. 1984. An estimation of calcium carbonate deposition rate in a planktonic foraminifer Globigerinoides sacculifer using 45Ca as a tracer: a recommended procedure for improved accuracy. Journal of Foraminiferal Research, 14, 303– 308. B ARKER , S. & E LDERFIELD , H. 2002. Foraminiferal calcification response to glacial-interglacial changes in atmospheric CO2. Science, 297, 833 –836. B ARKER , S., K IEFER , T. & E LDERFIELD , H. 2004. Temporal changes in North Atlantic circulation constrained by planktonic foraminiferal shell weights. Paleoceanography, 19(PA3008), doi:10.1029/2004PA001004. B AUCH , D., D ARLING , K., S IMSTICH , J., B AUCH , H. A., E RLENKEUSER , H. & K ROON , D. 2003. Palaeoceanographic implications of genetic variation in living North Atlantic Neogloboquadrina pachyderma. Nature, 424, 299 –302. B E´ , A. W. H. 1980. Gametogenic calcification in a spinose planktonic foraminifer, Globigerinoides sacculifer (Brady). Marine Micropaleontology, 5, 283–310. B E´ , A. W. H. & D UPLESSY , J. C. 1976. Subtropical convergence fluctuations and Quaternary climates in the middle latitudes of the Indian Ocean. Science, 194, 419–422. B ECKMANN , W., A URAS , A. & H EMLEBEN , C. 1987. Cyclonic coldcore eddy in the eastern North Atlantic. III. Zooplankton. Marine Ecology Progress Series, 39, 165 –173. B ENTOV , S. & E REZ , J. 2006. Impact of biomineralization processes on the Mg content of foraminiferal shells: A biological perspective. Geochemistry, Geophysics,
Geosystems, 7, Q01P08, http://dx.doi.org/10.1029/ 2005GC001015. B IJMA , J., E REZ , J. & H EMLEBEN , C. 1990a. Lunar and semi-lunar reproductive cycles in some spinose planktonic foraminifers. Journal of Foraminiferal Research, 20, 117–127. B IJMA , J., F ABER , W. W. & H EMLEBEN , C. 1990b. Temperature and salinity limits for growth and survival of some planktonic foraminifers in laboratory cultures. Journal of Foraminiferal Research, 20, 95–116. B IJMA , J., H EMLEBEN , C., O BERHA¨ NSLI , H. & S PINDLER , M. 1992. The effects of increased water fertility on tropical spinose planktonic foraminifers in laboratory cultures. Journal of Foraminiferal Research, 22, 242 – 256. B IJMA , J., H O¨ NISCH , B. & Z EEBE , R. 2002. Impact of the oceanic carbonate chemistry on living foraminiferal shell weight: Comment on “Carbonate ion concentration in glacial-age deep waters of the Caribbean Sea” by W. S. Broecker and E. Clark. Geochemistry, Geophysics, Geosystems, 3, doi:10.1029/ 2002GC000388. B ILLUPS , K., R ICKABY , R. E. M. & S CHRAG , D. P. 2004. Cenozoic pelagic Sr/Ca records: Exploring a link to paleoproductivity. Paleoceanography, 19, doi:10.1029/2004PA001011. B RADSHAW , J. S. 1961. Laboratory experiments on the ecology of foraminifera. Cushman Foundation of Foraminiferal Research, 12, 87–106. B ROWN , S. J. & E LDERFIELD , H. 1996. Variations in Mg/Ca and Sr/Ca ratios of planktonic foraminifera caused by postdepositional dissolution: Evidence of shallow Mg-dependent dissolution. Paleoceanography, 11, 543– 551. B UDDEMEIER , R. W. & F AUTIN , D. G. 1994. Saturation state and the evolution and biogeography of symbiotic calcification. In: A LLEMAND , D. & C UIF , J.-P. (eds) 7th International Symposium on Biomineralization, 14, Muse´e Oce´anographique, Monaco, 23–32. C ARON , D. A., B E´ , A. W. H. & A NDERSON , O. R. 1981. Effects of variations in light intensity on life processes of the planktonic foraminifer Globigerinoides sacculifer in laboratory culture. Journal of the Marine Biological Association of the United Kingdom, 62, 435–451. C ARON , D. A., F ABER , W. W. J. & B E´ , A. W. H. 1987a. Effects of temperature and salinity on the growth and survival of the planktonic foraminifer Globigerinoides sacculifer. Journal of the Marine Biological Association of the United Kingdom, 67, 323–342. C ARON , D. A., F ABER , W. W. J. & B E´ , A. W. H. 1987b. Growth of the spinose planktonic foraminifer Orbulina universa in laboratory culture and the effect of temperature on life processes. Journal of the Marine Biological Association of the United Kingdom, 67, 343–358. C ARON , D. A., A NDERSON , R. O., L INDSEY , J. L., F ABER , J., W ALTER , W. & L IN L IM , E. E. 1990. Effects of gametogenesis on test structure and dissolution of some spinose planktonic foraminifera and implications for test preservation. Marine Micropaleontology, 16, 93– 116.
INFLUENCES OF GROWTH RATES ON PLANKTIC FORAMINIFERS C ARPENTER , S. J. & L OHMANN , K. C. 1992. Sr/Mg ratios of modern marine calcite: Empirical indicators of ocean chemistry and precipitation rate. Geochimica et Cosmochimica Acta, 56, 1837–1849. C RONBLAD , H. G. & M ALMGREN , B. A. 1981. Climatically controlled variation of Sr and Mg in Quaternary planktonic foraminifera. Nature, 291, 61–64. DE V ARGAS , C., R ENAUD , S., H ILBRECHT , H. & P AWLOWSKI , J. 2001. Pleistocene adaptive radiation in Globorotalia truncatulinoides: genetic, morphologic and environmental evidence. Paleobiology, 27, 104–125. E GGINS , S., D E D ECKKER , P. & M ARSHALL , J. 2003. Mg/Ca variation in planktonic foraminifera tests: implications for reconstructing palaeo-seawater temperature and habitat migration. Earth and Planetary Science Letters, 212, 291– 306. E GGINS , S. M., S ADEKOV , A. & D E D ECKKER , P. 2004. Modulation and daily banding of Mg/Ca in Orbulina universa tests by symbiont photosynthesis and respiration: a complication for seawater thermometry? Earth and Planetary Science Letters, 225, 411– 419. E LDERFIELD , H., B ERTRAM , C. J. & E REZ , J. 1996. A biomineralization model for the incorporation of trace elements into foraminiferal calcium carbonate. Earth and Planetary Science Letters, 142, 409– 423. E LDERFIELD , H., V AUTRAVERS , M. & C OOPER , M. 2002. The relationship between shell size and Mg/Ca, Sr/Ca, d18O and d13C of species in planktonic foraminifera. Geochemistry, Geophysics, Geosystems, 3, 10.1029/2001GC000194. E REZ , J. 2003. The source of ions for biomineralization in foraminifera and their implications for paleoceanographic proxies. Reviews in Mineralogy and Geochemistry, 54, 115–149. F ISCHER , G. & W EFER , G. (eds) 1999. Use of Proxies in Paleoceanography. Springer-Verlag, Berlin, Heidelberg, New York. G EHLEN , M., B ASSINOT , F., B ECK , L. & K HODJA , H. 2004. Trace element cartography of Globigerinoides ruber shells using particle-induced X-ray emission. Geochemistry, Geophysics, Geosystems, 5, Q12D12, doi: 10.1029/2004GC000822. H ATHORNE , E. C., A LARD , O., J AMES , R. H. & R OGERS , N. W. 2003. Determination of intratest variability of trace elements in foraminifera by laser ablation inductively coupled plasma-mass spectrometry. Geochemistry, Geophysics, Geosystems, 4, 10.1029/ 2003GC000539. H EALY -W ILLIAMS , N., E HRLICH , R. & W ILLIAMS , D. F. 1985. Morphometric and stable isotopic evidence for subpopulations of Globorotalia truncatulinoides. Journal of Foraminiferal Research, 15, 242–253. H ECHT , A. D. 1974. Intraspecific variation in recent populations of Globigerinoides ruber and Globigerinoides trilobus and their application to paleoenvironmental analysis. Journal of Paleontology, 48, 1217– 1234. H ECHT , A. D. 1976. An ecologic model for test size variation recent planktonic foraminifera: application to the fossil record. Journal of Foraminiferal Research, 6, 295 –311. H EMLEBEN , C. 1969. Zur Morphogenese planktonischer Foraminiferen. Zitteliana, 1, 91–133.
83
H EMLEBEN , C., A NDERSON , O. R., B ERTHOLD , W. & S PINDLER , M. 1986. Calcification and chamber formation in Foraminifera – a brief overview. In: L EADBEATER , B. S. C. & R IDING , R. (eds) Biomineralization in lower plants and animals. The Systematics Association, Special Volume, 30, 237–249. H EMLEBEN , C., B E´ , A. W. H., A NDERSON , R. O. & T UNTIVATE , S. 1977. Test morphology, organic layers and chamber formation of the planktonic foraminifers Globorotalia menardii (d’Orbigny). Journal of foraminiferal Research, 7, 1 –25. H EMLEBEN , C., S PINDLER , M. & A NDERSON , O. R. 1989. Modern Planktonic Foraminifera. Springer, New York, Berlin, Heidelberg. H UBER , B. T., B IJMA , J. & D ARLING , K. 1997. Cryptic speciation in the living planktonic foraminifer Globigerinella siphoniphera (d’Orbigny). Paleobiology, 23, 33–62. H UBER , R., M EGGERS , H., B AUMANN , K.-H., R AYMO , M. E. & H ENRICH , R. 2000. Shell size variation of the planktonic foraminifer Neogloboquadrina pachyderma sin. in the Norwegian-Greenland Sea during the last 1.3 Myrs: implications for paleoceanographic reconstructions. Palaeogeography, Palaeoclimatology, Palaeoecology, 160, 193–212. J ØRGENSEN , B. B., E REZ , J., R EVSBACH , N. P. & C OHEN , Y. 1985. Symbiotic photosynthesis in a planktonic foraminifera, Globigerinoides sacculifer (Brady), studied with microelectrodes. Limnology and Oceanography, 30, 1253– 1267. K ENNETT , J. P. 1970. Pleistocene paleoclimates and foraminiferal biostratigraphy in subantarctic deep-sea cores. Deep Sea Research, 17, 125– 140. K LINKHAMMER , G. P., H ALEY , B. A., M IX , A. C., B ENWAY , H. M. & C HESEBY , M. 2004. Evaluation of automated flow-through time-resolved analysis of foraminifera for Mg/Ca paleothermometry. Paleoceanography, 19, PA4030, doi:10.1029/ 2004PA001050. K UCERA , M. & D ARLING , K. F. 2002. Cryptic species of planktonic foraminifera: their effect on palaeoceanographic reconstructions. Philosophical Transaction of the Royal Society of London Series A, 360, 695 –718. K UCERA , M., R OSELL -M ELE´ , A., S CHNEIDER , R., W AELBROECK , C. & W EINELT , M. S. 2005. Multiproxy approach for the reconstruction of the glacial ocean surface (MARGO). Quaternary Science Reviews, 24, 813–819. L EA , D. W., M ASHIOTTA , T. A. & S PERO , H. J. 1999. Controls on magnesium and strontium uptake in planktonic foraminifera determined by live culturing. Geochimica Cosmochimica Acta, 63, 2369– 2379. L OHMANN , G. P. 1995. A model for variation in the chemistry of planktonic foraminifera due to secondary calcification and selective dissolution. Paleoceanography, 10, 445– 457. L ORENS , R. B. 1981. Sr, Cd, Mn, and Co distribution coefficients in calcite as a function of calcite precipitation rate. Geochimica et Cosmochimica Acta, 45, 553– 561. M ALMGREN , B. & K ENNETT , J. P. 1978a. Test size variation in Globigerina bulloides in response to Quaternary palaeoceanographic changes. Nature, 275, 123–124.
84
D. N. SCHMIDT ET AL.
M ALMGREN , B., K UCERA , M. & E KMAN , G. 1996. Evolutionary changes in supplementary apertural characteristics of the Late Neogene Sphaeroidinella dehiscens lineage (Planktonic Foraminifera). Palaios, 11, 192 –206. M ALMGREN , B. A., B ERGGREN , W. A. & L OHMAN , G. P. 1983. Evidence for punctuated gradualism in the Late Neogene Globorotalia tumida lineage of planktonic foraminifera. Paleobiology, 9, 377–389. M ALMGREN , B. A. & K ENNETT , J. P. 1978b. Late Quaternary paleoclimatic applications of mean size variations in Globigerina bulloides d’Orbigny in the Southern Indian Ocean. Journal of Paleontology, 52, 1195–1207. M ARTIN , P. A., L EA , D. W., M ASHIOTTA , T. A., P APENFUSS , T. & S ARNTHEIN , M. 1999. Variation of foraminiferal Sr/Ca over Quaternary glacialinterglacial cycles: Evidence for changes in mean ocean Sr/Ca? Geochemistry, Geophysics, Geosystems, 1, 1999GC000006. M ELIERS , F. 1977. X-ray mineralogy studies, Leg 41, Deep Sea Drilling Project, Eastern North Atlantic Ocean. Initial Reports of the Deep Sea Drilling Program, Leg 41. Washington, D C, 1065– 1086. N U¨ RNBERG , D. 1995. Magnesium in tests of Neogloboquadrina pachyderma sinistral from high northern and southern latitudes. Journal of Foraminiferal Research, 25, 350 –368. N U¨ RNBERG , D., B IJMA , J. & H EMLEBEN , C. 1996. Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperatures. Geochimica et Cosmochimica Acta, 60, 803–814. O PPO , D. W. & F AIRBANKS , R. G. 1989. Carbon isotopic comparison of tropical surface water during the past 20 000 years. Paleoceanography, 4, 333– 351. O RTIZ , J. D., M IX , A. C. & C OLLIER , R. W. 1995. Environmental control of living symbiotic and asymbiotic foraminifera of the California Current. Paleoceanography, 10, 987– 1009. O RTIZ , J. D., M IX , A. C., R UGH , W., W ATKINS , J. M. & C OLLIER , R. W. 1996. Deep-dwelling planktonic foraminifera of the northeastern Pacific Ocean reveal environmental control of oxygen and carbon isotopic disequilibria. Geochimica et Cosmochimica Acta, 60, 4509–4523. P AQUETTE , J. & R EEDER , R. J. 1990. New type of compositional zoning in calcite – insights into crystalgrowth mechanisms. Geology, 18, 1244–1247. P HARR , R. B. & W ILLIAMS , D. F. 1987. Shape changes in Globorotalia truncatulinoides as a function of ontogeny and paleobiogeography in the Southern Ocean. Marine Micropaleontology, 12, 348– 355. P OLITI , Y., A RAD , T., K LEIN , E., W EINER , S. & A DDADI , L. 2004. Sea urchin spine calcite forms via a transient amorphous calcium carbonate phase. Science, 306, 1161–1164. R ATHMANN , S., H ESS , S., K UHNERT , H. & M ULITZA , S. 2004. Mg/Ca ratios of the benthic foraminifera Oridorsalis umbonatus obtained by laser ablation from core top sediments: relationship to bottom water temperature. Geochemistry, Geophysics, Geosystems, 5, Q12013, doi:10.1029/2004GC000808. R AVELO , A. C. & F AIRBANKS , R. G. 1995. Carbon isotopic fractionation in multiple species of planktonic
foraminifera from core-tops in the tropical Atlantic. Journal of Foraminiferal Research, 25, 53– 74. R EEDER , R. J. & G RAMS , J. C. 1987. Sector zoning in calcite cement crystals: Implications for trace element distributions in carbonates. Geochimica et Cosmochimica Acta, 51, 187–194. R EEDER , R. J., V ALLEY , J. W., G RAHAM , C. M. & E ILER , J. M. 1997. Ion microprobe study of oxygen isotopic compositions of structurally nonequivalent growth surfaces on synthetic calcite. Geochimica et Cosmochimica Acta, 61, 5057– 5063. R ENAUD , S. & S CHMIDT , D. N. 2003. Habitat tracking as a response of the planktic foraminifer Globorotalia truncatulinoides to environmental fluctuations during the last 140 kyr. Marine Micropaleontology, 49, 97–122. R ICKABY , R. E. M., S CHRAG , D. P., Z ONDERVAN , I. & R IEBESELL , U. 2002. Growth rate dependence of Sr incorporation during calcification of Emiliania huxleyi. Global Biogeochemical Cycles, 16, 10.1029/2001GB001408. R USSELL , A. D., H O¨ NISCH , B., S PERO , H. J. & L EA , D. W. 2004. Effects of seawater carbonate ion concentration and temperature on shell U, Mg, and Sr in cultured planktonic foraminifera. Geochimica et Cosmochimica Acta, 68, 4347–4361. S ADEKOV , A. Y., E GGINS , S. M. & D E D ECKKER , P. 2005. Characterization of Mg/Ca distributions in planktonic foraminifera species by electron microprobe mapping. Geochemistry Geophysics Geosystems, 6, Q12P06, http://dx.doi.org/10.1029/2005 GC000973. S CHIEBEL , R., B IJMA , J. & H EMLEBEN , C. 1997. Population dynamics of the planktic foraminifer Globigerina bulloides from the eastern North Atlantic. Deep Sea Research Part II, 44, 1701– 1713. S CHMIDT , D. N., R ENAUD , S. & B OLLMANN , J. 2003. Response of planktic foraminiferal size to late Quaternary climate change. Paleoceanography, 18, 10.1029/2002PA000831. S CHMIDT , D. N., R ENAUD , S., B OLLMANN , J., S CHIEBEL , R. & T HIERSTEIN , H. R. 2004a. Size distribution of Holocene planktic foraminifer assemblages: biogeography, ecology and adaptation. Marine Micropaleontology, 50, 319–338. S CHMIDT , D. N., T HIERSTEIN , H. R., B OLLMANN , J. & S CHIEBEL , R. 2004b. Abiotic Forcing of Plankton Evolution in the Cenozoic. Science, 303, 207–210. S CHMIDT , D. N., L AZARUS , D., Y OUNG , J. & K UCERA , M. 2006. Biogeography and evolution of body-size of marine plankton. Earth-Science Reviews, 78, 239–266. S PERO , H. J. & D E N IRO , M. J. 1987. The influence of symbiont photosynthesis on the d18O and d13C values of planktonic foraminiferal shell calcite. Symbiosis, 4, 213– 228. S PERO , H. J., L ERCHE , I. & W ILLIAMS , D. F. 1991. Opening the carbon isotope “vital effect” box, 2. Quantitative model for interpreting foraminiferal carbon isotope data. Paleoceanography, 6, 639 –655. S PINDLER , M., H EMLEBEN , C., B AYER , U., B E´ , A. W. H. & A NDERSON , O. R. 1979. Lunar periodicity of reproduction in the planktonic foraminifer Hastigerina pelagica. Marine Ecology Progress Series, 1, 61–64.
INFLUENCES OF GROWTH RATES ON PLANKTIC FORAMINIFERS S TEINKE , S., C HIU , H.-Y., Y U , P.-S., S HEN , C.-C., L O¨ WEMARK , L., M II , H.-S. & C HEN , M.-T. 2005. Mg/Ca ratios of two Globigerinoides ruber (white) morphotypes: Implications for reconstructing past tropical/subtropical surface water conditions. Geochemistry, Geophysics, Geosystems, 6, Q11005, http://dx.doi.org/10.1029/2005GC 000926. T OWE , K. M. & C IFELLI , R. 1967. Wall ultrastructure in the calcareous foraminifera: crystallographic aspects
85
and a model for calcification. Journal of Paleontology, 41, 742– 763. W EINER , S. & D OVE , P. M. 2003. An Overview of Biomineralization Processes and the Problem of the Vital Effect. Reviews in Mineralogy and Geochemistry, 54, 1–29. Z EEBE , R. E., W OLF -G LADROW , D. A., B IJMA , J. & H O¨ NISCH , B. 2003. Vital effects in foraminifera do not compromise the use of delta B-11 as a paleo-pH indicator: Evidence from modeling. Paleoceanography, 18, 1043, doi:10.1029/2003PA000881.
Fine-scale growth patterns in coral skeletons: biochemical control over crystallization of aragonite fibres and assessment of early diagenesis J. P. CUIF1, Y. DAUPHIN1, A. MEIBOM2, C. ROLLION-BARD3, M. SALOME´4, J. SUSINI4 & C. T. WILLIAMS5 1
UMR 8148 IDES, Faculte´ des Sciences, Baˆt 504, F-91405 Orsay, France (e-mail:
[email protected]) 2
MNHN, LEME - Nanoanalyse, 57 rue Cuvier, F-75423 Paris, France
3
CRPG, Rue N-D des Pauvres, F-54501 Vandoeuvre les Nancy, France 4
ESRF ID21, BP 220, F-38043 Grenoble Cedex 9, France
5
NHM, Analytical Department, Cromwell Road, SW7 5BD London, UK
Abstract: Recent works have paved the way for an understanding of the scale at which environmental signals might be recorded in coral skeletons. In this paper, the resulting structural and chemical insights are exemplified by a Goniastrea corallite. The bulk of the coral skeleton consists of fibrous aragonite, which in turn is constructed by sequential growth of micrometre thick layers, oriented parallel to the local growth direction. These growth layers consist of nanograins (50–100 nm) of aragonite that appear to be crystallized in close association with a matrix, conceivably proteoglycans, which seem to coat individual nanograins. These observations contradict the traditional notion that coral fibre consists of ‘a single crystal of orthorhombic aragonite’. Additionally, the ultrastructural observations provide us with criteria to assess early diagenetic effects. Some Lower Norian corals from South Anatolia (Turkey) display extremely wellpreserved mineralogy and structures. They have also preserved the organic components of their skeletons from which it has been demonstrated, through a study of the Nitrogen isotopic composition, that photosynthesis was involved in the metabolism of these early Scleractinia. But even in these remarkably preserved corals, we find evidence for diagenetic changes at the nanometre scale, concerning both the amount of organic matrices and the appearance of the aragonitic nanogranular units. Such micro-structural observations call for caution when interpreting isotopic effects in the fossil coral record.
The polyps of Scleractinia (Hexacoralla) built calcareous hard parts that have been used as environmental archives since the middle of last century (Epstein et al. 1953). To date, the interpretation of the chemical or isotopic measurements relies on a widely shared concept of the origin of the different units from which the skeleton is formed. The bulk of the skeleton consists of what appear to be several hundreds of micrometres-long fibres of aragonite. These fibres, which were first observed by Pratz (1882), are often arranged in three-dimensional fan-systems, which have been compared with the spherulitic crystallization of inorganic chemical precipitations (Bryan & Hill 1941, p. 84). For example, Bryan & Hill concluded that ‘each coral fibre is a single crystal of orthorhombic aragonite’ a viewpoint that was reinforced by the interpretation of Barnes (1970) who described the spatial arrangements of coral fibres as a result of ‘crystal growth competition’. Thus,
in contrast to the ‘matrix mediated’ calcification process in which growth of skeleton units is controlled by specifically secreted organic compounds (as in Mollusc shells, for instance), the paradigm in coral skeletogenesis has been that the bulk of the skeleton is produced by aragonite precipitation from a supersaturated solution, close in composition to seawater. A statement by Veis (2005, p. 1419) epitomizes this view: ‘In “biologically induced” mineralization – for example, in corals – the minerals adopt crystal shapes similar to those formed by inorganic processes and have essentially random crystal orientations’. Reflecting this viewpoint, geochemists work from the basis of a purely physio-chemical models of skeletogenesis for scleractinian corals (e.g. Gaetani & Cohen 2006). Recently, however, it has become clear that purely physiochemical models cannot explain the observed structural and biochemical properties of coral skeletons and are also unable to account for
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 87–96. DOI: 10.1144/SP303.7 0305-8719/08/$15.00 # The Geological Society of London 2008.
88
J. P. CUIF ET AL.
their chemical and isotopic variability. This paper aims first at summarizing the main features of coral skeletal ultra-structure, which is the scale at which environmental signals is recorded in coral skeletons. Then, by comparing data from living corals with observations and measurements made on ancient, but extremely well-preserved aragonite corals from the Triassic, it is proposed that observations of the fine-scale growth patterns may provide us with new criteria for assessing the diagenetic state of fossil samples. This has important implications for the level of confidence in the interpretation of geochemical data obtained from ancient coralline aragonites.
Materials and methods The studied corals were collected alive: Goniastrea from New Caledonia, Favia from Moorea island (Polynesia) and Merulina from New-Caledonia (Noumea lagoon). The polyp tissues were removed following a three-hour immersion in pure water that causes the cells to be destroyed, after which the remaining tissues are easily removed by a waterjet. In all cases, only the upper parts, i.e. the most recently formed skeleton, of the corallite structures were used to assess the structural and compositional properties of the skeletons. Triassic corals (Lower Norian) were collected in the upper valley of the Alakir Cay river near the Dere Ko¨y village (Lycian Taurus, South Anatolia, Turkey). Growth structures of fibrous aragonite were made visible by etching of polished surfaces as follows: 45 seconds, continuously stirred in a 1 per mil formic acid solution with 5% weight glutaraldehyde added (Cuif & Dauphin 1998). SEM observations were carried out on an XL30 Philips microscope. Synchrotron radiation mapping of the intra skeletal organic sulphates was carried out at the ID21 beam line of the ERF-Grenoble. The X-ray beam was tuned at 2.4285 keV and focused to an about 0.3 micrometre diameter using specifically designed lenses. Scanning displacement of the sample driven by piezo electric devices allows submicrometre resolution mapping of polished surface and in-situ distribution of sulphated polysaccharides to be visualized. A full technical description is available in Cuif et al. (2003). Microprobe mappings and measurements of minor element concentrations were obtained at the Analytical Centre of the Mineralogical Laboratory of the Natural History Museum (London) using a CAMECA SX 50 instrument equipped with four wave-length dispersive detectors. Atomic force observations were carried out with a Dimension 3000 microscope (Digital Instrument) using the
‘tapping’ mode. In this mode, three different images are simultaneously produced: (1) height images allow the actual topography of the sample surface to be visible at the nanometre scale; (2) Amplitude images are derived from the height images and emphasize the changes in topography; (3) Phase images are based on the lag that result from interaction of the microscope tip with the sample material. Surfaces were diamond polished and cleaned by very short exposition (about 5 seconds) to light acidic solution followed by ultra-pure water rinsing (Cuif & Dauphin 2005). Thermogravimetry coupled to infrared spectrometry has been carried out at the analytical centre of the CNRS at Lyon-Vernaison. Weight loss of about 100 mg of skeleton powder during heating from ambient to 500 8C was continuously recorded. Simultaneously, emitted water and CO2 were measured by infrared absorption at 1508 cm21 and 2363 cm21 respectively (Cuif et al. 2004). Oxygen isotopic ratio were measured at the Centre de Recherche Pe´trographique et Ge´ochimique de Nancy (CRPG-France), using a CAMECA SIMS 1270 instrument. For technical details we refer to Rollion-Bard et al. (2003). High resolution chemical profiles using the NanoSIMS (Cameca TM) were made at the Museum National d’Histoire Naturelle de Paris (LEME, nanoanalyse group). For technical details, refer to Meibom et al. (2004).
Results The coordinated stepping growth mode for the fibrous skeleton of Goniastrea Figures 1a–c show the usual view of fibres as elongate units arranged in a ‘fan system’ (Fig. 1b), originating from crystalline spherulites that are visible in polarized light (Fig. 1c) on a Goniastrea corallite. However, the real growth mode of coral fibres is revealed by SEM observations of a polished and etched surface, as illustrated Figure 1d. Under these observational conditions, a completely different pattern appears on the etched surfaces. The original radiating arrangement of fibres is still visible after etching, but the predominant feature is now a sequence of narrow crests and grooves on length scales of a few micrometres. Furthermore, Figures 1e and 1 f show that overall organization of this groove-system is not independent of the orientation of the fibres. Crests and grooves are developed perpendicular or strongly oblique to the direction of fibre growth and are organized concentrically around restricted domains that exhibit specific micro structural patterns. In the centre of these concentric zones, one finds the early mineralization zones (EMZ). The
FINE-SCALE GROWTH PATTERNS IN CORAL SKELETONS
89
Fig. 1. Fibrous fan-systems versus stepping growth-mode of fibres in the Goniastrea corallites (a) Calicinal view of a Goniastrea corallite; (b) SEM view of the radiating arrangement of fibres; (c) Polarized light view of a thin section; (d) SEM view of a polished and etched surface in the Goniastrea wall and septal system. Early Mineralization Zones (EMZ) appear as dot-lines; (e) Surrounding the EMZ, the superimposed layered growth steps are revealed by etching; (f) Superposition of a fibre fan-system (polarized light) to picture of the stepping growth pattern. Crystal like fibres include regularly produced layers with higher sensitivity to etching (white dots), leading to the concentric banding transversal to fibre orientations.
90
J. P. CUIF ET AL.
EMZ are constantly formed at the growing tips of the septal structures and are subsequently overgrown by layers of fibrous aragonite. Therefore, the EMZ always occupy a central position with respect to orientation of the fibrous fan-systems. This is the reason why Ogilvie-Gordon (1896), when she coined the term ‘centres of calcification’, described them as ‘the points from which fibres diverge’. As Figure 1f illustrates, etching reveals the high degree of control and coordination of fibre crystallization: only a perfect synchronism in growth of adjacent fibres allows layered structure to be built. Applying etching solution to this layered structure results in the crests and grooves system that emphasizes chemical heterogeneity of the crystal-like fibrous units.
The layered organization of sulphated polysaccharide within fibres Figure 2 shows synchrotron based X-ray fluorescence micro-analyses, which reveal the distribution of sulphated polysaccharides within the Goniastrea walls and septa (Fig. 2a). The EMZ regions are characterized by high concentrations of
sulphated polysaccharides (Fig. 1b; low resolution). Additionally, concentration of sulphated polysaccharides within the fibrous skeleton is not constant. High resolution mapping shows that variation in sulphur concentrations (Fig. 2c) corresponds to the banded organization of the fibrous layers (Fig. 2d). The linear dimensions and the layered arrangement of the sulphated organic appear to be in close correspondence to the mineral layering. This implies that Ca-carbonate and organic compounds are closely associated at a sub-micrometre scale.
The organo-mineral interplay at the nanometre scale Atomic force microscopy allows this relationship between organics and mineral components to be directly observed at the sub-micrometre length scale (Fig. 3). Each layer of the fibrous aragonite is built by carbonate granular units of mean dimension on the order of a few tens of nanometres (Fig. 3a and 3d; n). AFM ‘tapping mode’ imaging does not allow a well-resolved picture of the surface topography to be obtained, but the derivative images (Figs 3b and 3e) clearly show that each granules is embedded in a material that seems to coat individual granules (c). AFM phase contrast images (Figs 3c and 3f) provide evidence of the specific properties of this grain embedding material. Phase contrast arises from differences in the interaction between the AFM tip and the observed materials on the basis of physical characteristics such as elasticity, viscosity and adhesiveness. In this regard, the material that coats each individual carbonate granules exhibits strikingly specific properties. As growth of coral fibres involves two different components, aragonite and organics, which are co-existing at a submicrometre-scale (as shown by synchroron radiation X-ray fluorescence), the Merulina and Goniastrea skeletons exemplify the organo-mineral fine scale organization of the fibre growth layers.
Structure and composition of Lower Norian corals from the Alakir Cay valley Fig. 2. Low and high resolution maps of polysaccharides distribution in the Goniastrera skeleton (a) Polished surface of wall and septa; (b) Low resolution X-ray fluorescence map: higher concentrations of S-polysaccharides are visible in EMZ of wall and septa; (c) High resolution X-ray fluorescence map in fibrous tissue. Evidence of the layered distribution of S-polysaccharides; (d) Banding pattern within the mineral phase. Correspondence between mineral stepping growth and layered distribution of polysaccharides indicates that the two phases (organic and mineral) interplay at a submicrometre scale.
Most of the fossil corals from the Alakir Cay outcrops exhibit remarkable fossilization patterns from both morphological, microstructural and mineralogical view-points (summary in Cuif 1980). However, they have been submitted to mineralizing solutions: crystallization of blocky calcite has occurred within the corallite internal cavities (Figs 4a–b). In spite of this crystallization process, fibrous aragonite is still well-preserved. For example, the average concentration of Sr is about 0.7% by weight (Fig. 4c), which is similar to the Sr concentration in recent coral fibres. Iron
FINE-SCALE GROWTH PATTERNS IN CORAL SKELETONS
91
Fig. 3. Examples of the three different AFM pictures obtained by using ‘tapping mode’ on 200 to 2450 nanometre wide field views of the fibrous skeletons of Merulina (a –c) and Goniastrea (d –f). Figures 3a and 3d: height images; Figures 3b and 3e: amplitude images; Figures 3c and 3f: phase images. The high contrast between nanograin (n) and cortex (c) indicates a strong diffferences with respect to physical parameters such as elasticity, viscosity, adhesiveness, leading to a strong phase lag in the cortex region. This suggests an organic nature of the cortex material.
(0.45% weight) is located exclusively in the infilling blocky calcite (Fig. 4d). No trace of Fe is observed within the skeleton, with exception of the median line of the septa, i.e. in the EMZ, which has long been recognized to be very sensible to dissolution and therefore easily modified by diagenesis. Distribution of sulfur within the fibrous skeleton show exact similarity with the Sr distribution (Fig. 4e) and concentration of sulphur is about 0.25% by weight. In addition to these quantitative microprobe measurements, synchrotron fluorescence at 2.4825 keV (the energy of the sulphated bond of sulphur in polysaccharides) indicate that S is still in the sulphated form, for which XANES analyses indicate sulphated polysaccharides (Cuif et al. 2003). Mapping at the sub-micrometre scale of these sulphated polysaccharides (ESRF–ID21 beam line) summarizes the remarkable preservation of skeletons in these ancient corals (Fig. 4f).
Measurements of skeletal d18O variations Amongst the many members of the highly differentiated coral fauna from the Lower Norian Alakir outcrops, the thick-walled solitary coral Pachythecalis major (Figs 4g–h) offers the opportunity to
obtain long-term records of environmental conditions on a single specimen. In this species, the compact wall may reach up to three centimetres in thickness. On polished surfaces perpendicular to the corallite growth direction, the Pachythecalis wall exhibits a clear structural periodicity (Fig. 4i). The wall is built by densely packed fibrous fascicles with radial orientation and inward growth (Fig. 4k). Accordingly, d18O measurements made on a radial line perpendicular to the major growth rings, show a strong oscillation, the measured values being distributed between 20.5‰ and 24‰ relative to the PDB standard (original result).
Discussion Attention must be drawn first on the multiple controls of the preservation status of these fossils previous to measurements. Not only are the mineral units exclusively aragonitic, allowing the fibrous microstructure to be clearly described, but the minor element concentrations are in agreement with what can be expected from well-preserved coral skeletons. In addition, the skeletal organic contents have been so well protected that, studying
92
J. P. CUIF ET AL.
Fig. 4. Preservation status of the Triassic corals from the Alakir Cay outcrops (Lower Norian). Figures 4a to 4f: Minor element concentrations and distribution of organic content in a Retiophyllia corallite. (a) Overview of a transversal section; (b) Skeletal fibrous tissue and blocky calcite filling of the internal cavity; (c– e) Electron microprobe maps of minor element distributions in skeleton and calcite filling; (f) X-ray XANES map of sulphated polysaccharides in corallite skeleton. The lower organic content in the median line of the septa corresponds to calcite filling after dissolution of the EMZ structures (see also distribution of minor elements); (g– k) Skeletal features of Pachythecalis major. (g) Sketch of an axial section of a Pachythecalis corallite; (h –i) Two sections in the Pachythecalis skeleton perpendicular to the growth axis; (j) Arrangement of fibres in Pachythecalis wall: thin section, polarized light; (k) Major growth rings in the wall of Pachythecalis.
FINE-SCALE GROWTH PATTERNS IN CORAL SKELETONS
93
0 δ18O –1 –2 –3 –4 Growth direction –5 Limits of major growth layers 18
Fig. 5. Oscillation of the d O in correspondence to major growth rings of the Pachythecalis wall (see location on Fig. 4k).
the d15N from isolated organic matrix, it as been possible to find a clear indication of a symbiotic metabolism for this Lower Norian species (Muscatine et al. 2005). Previously, characterization of amino acids in the organic compounds extracted from the Pachythecalis wall (Gautret & Marin 1993) has demonstrated the importance of the acidic fraction (aspartic and glutamic acids). Both characterizations support the authenticity of the skeletal organic content as the remains of the mineralizing matrices that have driven crystallization of the Pachythecalis walls during the specimen life. Clearly, the overall quality of these fossil materials has been accurately checked. In any study dealing with aragonites from ancient corals, such a series of positive results concerning the preservation status would probably led investigators to
(a) %WL
admit a full reliability of the measured values (as for any other proxy-bearing biogenic material). But surprisingly, evidences also exist that changes have occurred in both chemical and structural properties of the Pachythecalid skeletons. Thermogravimetric studies of recent coral skeletons coupled to infrared measurement of the gas emitted during heating have shown that the mineralizing organic compounds that remain entrapped in the skeleton growth layers represent about 2.2– 2.6% of the total weight (Cuif et al. 2004). This nonmineral component includes a noticeable proportion of water: about 1% of the skeleton weight is due to water structurally linked to the matrix proteoglycans. Figure 6a shows the coordination of the three weight-loss profiles for the recent coral Favia stelligera.
(b) %WL
weight water CO2
1
1
2
2
3
Temperature °C 100
200
300
400
500
600
3
Temperature °C 100
200
300
400
500
600
Fig. 6. Profiles of weight losses, water emission and CO2 emission during heating of a recent Favia stelligera skeleton (a) compared to a Triassic Pachythecalis wall skeleton (b). Note the synchronism between water emissions (around 300 8C) and the first CO2 (arrows) in both the recent (a) and Triassic corals (b).
94
J. P. CUIF ET AL.
Similar measurements made on Pachythecalis wall show a good consistency of the three weight-loss profiles, but the recorded weights are substantially lower than in the recent corals. To-date, no such weak values have been recorded concerning the hydro-organic content in skeletons of recent corals. Hypothesis can be made that the biomineralization process in Triassic corals did not request the same amount of mineralizing matrices than the recent ones. However, two converging observations suggest that the quantitative difference of mineralizing matrices between recent and Triassic corals may be related to fine compositional and nanostructural changes. SEM surface examination of the Pachythecalis fibres shows that a structural transformation of fibres has occurred (Figs 7a –b). Atomic force microscopy confirms that if some traces of nanogranular organization of the fibres are still visible from place to place, most of the skeleton nanograins seem gathered into more massive units (Fig. 7c, 7e). At a larger observation level, hypothesis of a disorganization of the aragontitic fibres is also supported. In some samples, microscopic observation of Pachythecalid wall in transmitted light, shows a layered incorporation of pigmented traces within the skeleton (Fig. 8a). This shows that the micron-scaled growth mode was probably present in this coral family, in addition to other similarities to modern Scleractinia (biochemical and biological). But polished and etched Pathythecalid fibres (Fig. 8b) reveal that the micron-scale layered organization has been disrupted: growth layering
is no longer visible (compare to Fig. 1e–f ). Recent chemical measurements have shown that when the layered structure of fibres is well visible, a clearly correlated Mg signal can be measured and correlated to the growth layering (Fig. 8c, from Meibom et al. 2004). NanoSIMS measurements show that Sr and Mg have distinct behaviors with respect to the biomineralization process (Fig. 8c), suggesting a different position of these elements within the skeletal material and, consequently, distinct sensitivities to the diagenetic influences. In the Pachythecalid skeletons the loss of mineralizing matrix and the correlative reorganization of skeletal aragonite illustrate the very early diagenetic process. Presently, no information is available concerning the distinct mobility of Sr and Mg during the early diagenesis, prior to aragonite recrystallization, information that could be useful when using Sr/Ca and Mg/Ca ratio as environmental proxies. The Pachythecalis d18O oscillation (Fig. 5), probably seasonal, can be compared to a recent example analyzed by Ivany et al. (2004) in the massive skeletons of an Oligocene coral (Archohelia) from the Byram formation (Mississippi). The d18O was found to oscillate between 22 and 24.8 permil versus PDB (rather similar to the Pachythecalis variations). Assuming that seawater is characterized by d18O ¼ 20.5% and that the water salinity was ‘normal’, Ivany et al. (2004) obtained a surprisingly high seasonal temperature variation of 12 to 24 8C in the Archohelia skeletons. A parallel approach applied to minor
Fig. 7. Fine scale structural changes in the Pachythecalis wall. (a) Close SEM view of fibre morphology in a recent coral skeleton (Favia stelligera); (b) Equivalent picture in the skeleton of the Triassic Pachythecalis. The overall direction of the fibre tufts is preserved (allowing precise microstructural studies to be made), but the scaly surface of fibres suggests occurrence of internal changes; (c) Nanoscale structural patterns of the skeleton. Individual nanograins are still recognizable in some places, but are frequently fused into larger units (tapping mode, phase image); (d– e) Skeleton nanograins can be visible in some places (d), but the large and composite units support hypothesis of a modification of the fine structure of the skeleton, correlated to the loss of nineralizing matrices (height images).
FINE-SCALE GROWTH PATTERNS IN CORAL SKELETONS
95
Fig. 8. Disorganization of the growth layering in aragonitic fibres of Pachythecalids. (a) Layered growth mode in the wall of some Pachythecalid (thin section, transmitted light); (b) SEM view of the aragonitic fibres of Pathythecalis after etching: no visible growth layering; (c) Difference in Sr/Ca and Mg/Ca ratio during fibre growth in a recent Porites.
element fractionation leads to temperature oscillation of the same magnitude, but with mean values much lower than the d18O signal, suggesting that Archohelia might have lived exposed to remarkably low seawater temperatures. Conclusively, applying the empirical equations from the current literature to ancient corals lead to results that justify the claim for a ‘new strategy’ (Lough 2003) in order to improve the use of coral records. This is still more obvious for materials from the Lower Norian specimens from the Alakir Cay river, in spite of similarity in biomineralization mechanism and their excellent preservation status. As recently stated by Nothdurft & Webb (2007), taking care of the specific growth modes of coral skeletons is essential to ensure a good correlation between sampling locations and growth timing. In parallel, a better understanding of the chemical and isotopical changes that occur during the very early diagenetic process might improve the value of the palaeo-environmental proxies.
Conclusion The following conclusions are reached: † In contrast to common opinion, crystallization of coral fibres is submitted to a biogeochemical control during crystallization; † High resolution measurements must be carried out according to growth layering;
† Preservation status of the skeletons must be checked at the same level: controls based on overall mineralogic status do not ensure validity of the numerical results; and † Evaluating the consequences of early diagenetic changes, prior to aragonite recrystallization, should be studied to increase the validity of the interpretations of the measured environmental signals.
References A DKINS , J. F., B OYLE , E. A., C URRY , W. B. & L UTRINGER , A. 2003. Stable isotopes in deepsea corals and a new mechanism for ‘vital effects.’ Geochimica et Cosmochimica Acta, 67, 1129– 1143. B ARNES , D. J. 1970. Coral skeletons: an explanation of their growth and structure. Science, 170, 1305–1308. B RYAN , W. B. & H ILL , D. 1941. Spherulitic crystallization as a mechanism of skeletal growth in the hexacorals. Proceedings of the Royal Society of Queensland, 52, 78–91. C UIF , J. P. 1980. Microstructure versus morphology in the skeletons of Triassic scleractinian corals. Acta Palaeontologica Polonica, 25/3 –4, 361–374. C UIF , J. P. & D AUPHIN , Y. 1998. Microstructural and physico-chemical characterizations of the ‘centers of calcification’ in the septa of some recent Scleractinian corals. Palaeontologische Zeischritft, 72, 257– 270. C UIF , J. P. & D AUPHIN , Y. 2005. The environmental recording unit in coral skeletons – a synthesis of structural and chemical evidences for a biochemically driven, stepping-growth process in fibres. Biogeoscience, 2, 61–73.
96
J. P. CUIF ET AL.
C UIF , J. P., D AUPHIN , Y., D OUCET , J., S ALOME´ , M. & S USINI , J. 2003. XANES mapping of organic sulfate in three scleractinian coral skeletons. Geochimica et Cosmochimica Acta, 67, 75– 83. C UIF , J. P., D AUPHIN , Y., B ERTHET , P. & J EGOUDEZ , J. 2004. Associated water and organic compounds in coral skeletons: quantitative thermogravimetry coupled to infrared absorption spectrometry. Geochemistry, Geophysics, Geosystems, 5, 11. doi: 10.1029/2004/GC000783. E PSTEIN , S., B UCHSBAUM , R., L OWENSTAM , H. A. & U REY , H. C. 1953. Revised carbonate-water isotopic temperature scale. Geological Society of America Bulletin, 64, 1315–1326. G AETANI , G. A. & C OHEN , A. L. 2006. Element partitioning during precipitation of aragonite from seawater: A framework for understanding paleoproxies. Geochimica et Cosmochimica Acta, 70, 4617– 4634. G AUTRET , P. & M ARIN , F. 1993. Tendances diage´ne´tiques des structures aragonitiques fibreuses produites par des Spongiaires et des Madre´poraires du trias supe´rieur de Turquie. Compte-Rendus de l’Acade´mie des Sciences de Paris, 316, se´ries II, 1319–1325. I VANY , L. C., P ETERS , S. C., W ILKINSON , B. H., L OHMANN , K. C. & R EIMER , B. A. 2004. Composition of the early Oligocene ocean from coral stable isotope and elemental chemistry. Geobiology, 2, 97–106. L OUGH , J. M. 2003. A strategy to improve the contribution of coral data to high-resolution
paleoclimatology. Paleoceanography, Paleoclimatology, Paleoecology, 204, 115–143. M EIBOM , A., C UIF , J. P., H ILLION , F., C ONSTANTZ , B., J UILLET -L ECLERC , A., D AUPHIN , Y., W ATANABE , T. & D UNBAR , R. B. 2004. Distribution of magnesium in coral skeleton. Geophysical Research Letters, 31, L23306, doi:10.1029/2044GL021313. M USCATINE , L., G OIRAN , C., L AND , L., J AUBERT , J., C UIF , J. P. & A LLEMAND , D. 2005. Stable isotopes (d13C and d15N) of organic matrix from coral skeletons. Proceedings of the National Academy of Science, 102/5, 525–1530. N OTHDURFT , L. D. & W EBB , G. E. 2007. Microstructure of common reef-building coral genera Acropora, Pocillopora, Goniastrea and Porites: constratints on spatial resolution in geochemical sampling. Facies, 53, 1– 26. O GILVIE -G ORDON , M. 1896. Microscopic and systematic study of madreporarian types of corals. Royal Society of London Philosophical Transactions, 187/B, 83–345. P RATZ , E. 1882. Uber die vervandschaftlichen Beziehungen einigen Korallengattungen. Palaeontographica, 29, 81–123. R OLLION -B ARD , C., B LAMART , D., C UIF , J. P. & J UILLET -L ECLERC , A. 2003. Microanalysis of C and O isotopes of azooxanthellate and zooxanthellate corals by ionmicroprobe. Coral Reefs, 22/4, 405–415. V EIS , A. 2005. A window on biomineralization. Science, 307, 1419.
Modern deep-sea benthic foraminifera: a brief review of their morphology-based biodiversity and trophic diversity A. J. GOODAY1, H. NOMAKI2 & H. KITAZATO2 1
National Oceanography Centre, Southampton, Empress Dock, European Way, Southampton SO14 3ZH, UK (e-mail:
[email protected])
2
Institute for Research on Earth Evolution, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka 237-0061, Japan Abstract: Most fossil deep-sea foraminifera are multichambered and have relatively robust, calcareous or agglutinated shells. Modern assemblages, on the other hand, include many fragile monothalamous (single-chambered) forms and komokiaceans (a superfamily of protist currently placed within the foraminifera) with soft test walls. These groups are poorly known and most of the hundreds of morphospecies recognized in deep-sea samples are undescribed. The relative abundance of robust and fragile taxa varies with water depth and food supply. Calcareous and other hard-shelled species tend to predominate in relatively eutrophic areas, particularly on continental margins, but decrease as a proportion of the ‘entire’ live fauna (i.e. including soft-shelled species) with increasing water depth, even above the CCD (carbonate compensation depth). Most of the species on which the foraminiferal proxies used in palaeoceanography are based live in these bathyal regions. At abyssal depths, and particularly below the CCD, faunas are largely agglutinated and dominated by monothalamous forms. These assemblages have a much lower fossilization potential than those found on continental margins. In addition to carbonate dissolution, these patterns probably reflect adaptations to increasingly oligotrophic conditions on the ocean floor with increasing depth and distance from land. Bathyal species include herbivores and opportunistic deposit feeders (omnivores) that consume labile organic material, in addition to deep-infaunal deposit feeders, and must contribute significantly to carbon cycling. Many abyssal monothalamous foraminifera, in constrast, accumulate stercomata (waste pellets composed of fine sediment particles) and probably ingest sediment, associated bacteria and more refractory organic matter. Some monothalamous species without stercomata may be bacteriovores. Although they probably process organic carbon at a slower rate than calcareous species, the shear abundance of monothalamous taxa at abyssal depths suggests that they are important in carbon cycling on a global scale. The loss of a substantial proportion of foraminiferal biomass and biodiversity from the fossil record should be considered when using foraminifera to reconstruct palaeoproductivity, for example, by using the Benthic Foraminiferal Accummulation Rate (BFAR). Different dietary preferences among calcareous species have implications for the stable carbon isotope signal preserved in their shells.
Foraminifera are by far the most abundant benthic organisms to be preserved in the deep-sea fossil record. Geologists analyse modern foraminiferal faunas in order to refine and calibrate palaeoceanographic proxies based on the fossilized shells of these protists (Gooday 2003; Jorissen et al. 2007). Calcareous species, which are often the only faunal component to survive the fossilization process, are the main focus of interest but some agglutinated species also have fairly good fossil records in deep-sea sediments (Kaminski & Gradstein, 2005) and may be useful in reconstructions of palaeoceanography (e.g. Kuhnt & Moullade 1991). Yet studies of a more biological nature suggest that these well-known groups are the tip of the iceberg of foraminiferal biodiversity. Many different kinds of foraminifera inhabit deep-sea environments. Fragile taxa, some of them with soft, flexible
test walls, represent an important but largely undescribed component of deep-sea foraminiferal assemblages and indeed of the entire ocean-floor biota (Tendal & Hessler 1977; Nozawa et al. 2006). In parallel with these developments, a clearer understanding of the ecology of benthic foraminifera in the deep ocean has emerged in recent years (Gooday 2003). Shallow water environments are often dynamic and heterogeneous so that a variety of abiotic factors, among them temperature, salinity, substrate characteristics, energy (current flow and wave action), food supply and bottomwater oxygenation, influence the distribution, abundance and life processes of benthic foraminifera (Murray 1991, 2006). In deeper water, where temperature and salinity are more or less constant, food and oxygen availability become crucial factors together with, in some areas, current flow,
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 97–119. DOI: 10.1144/SP303.8 0305-8719/08/$15.00 # The Geological Society of London 2008.
98
A. J. GOODAY ET AL.
sediment characteristics and carbonate dissolution (Mackensen et al. 1995; Gooday 2003; Murray 2006; Jorissen et al. 2007). This makes knowledge of diets and feeding habits important for understanding the ecology of deep-sea foraminifera. In this brief review, we focus on these two aspects of deep-sea foraminiferal faunas – their biodiversity and trophic diversity. It is important to note here that the deep sea is not a single habitat and strong environmental and biological contrasts exist between bathyal continental margins and the abyssal plains (Etter et al. 2005). In a general sense, the species that interest geologists are most common at bathyal depths, whereas the less familiar fragile taxa dominate at abyssal depths. We argue that contrasts in the overall taxonomic character of assemblages on continental margins and in central oceanic regions are mirrored by changes in the predominant trophic strategies. To a large extent, our review is written from a biological perspective and not all of it is relevant to the development and interpretation of palaeoceanographic proxies. This
reflects the often divergent aims and approaches of biologists and geologists. Nevertheless, there are many points of contact between the two disciplines, and some of the issues discussed here have potentially important implications in both palaeoceanography and biogeochemistry, as well as for the evolutionary development of foraminifera.
Trends in assemblage composition with water depth Larger foraminifera (macrofaunal size fraction) Macrofaunal foraminiferans (those retained on a 300 mm mesh) include fairly robust, singlechambered agglutinated forms such as Bathysiphon, Hyperammina, Pilulina, Rhabdammina, Saccammina and Saccorhiza, as well as more delicate, sack-like saccamminids (Fig. 1). These typically tubular or spherical forms are often extremely
Fig. 1. Shipboard photographs of large (macrofaunal) saccamminids from the Southern Ocean, collected during Polarstern Cruise ANT XXII/2. All are undescribed. (a) Silver species with two apertures; Stn, 153#4, 638 19.350 S, 648 36.790 W, 2079 m water depth; (b) Very delicate, greyish-coloured species; Stn 102#11: 658 35.510 S, 368 28.830 W, 4800 m; (c) Spherical species with multiple apertures; Stn 102#13; (d) White, flask-shaped species; St 94#14: 668 380 S, 278 100 W, 4890 m.
DEEP-SEA BENTHIC FORAMINIFERA
common on productive continental margins (Gooday et al. 1997) where they occur together with large, multichambered calcareous and agglutinated genera such as Biloculina, Cyclammina, Hormosina, Globobulimina, Lenticulina and Nodosinum (Herb 1971; Gooday et al. 2001). On abyssal plains, however, the macrofaunal size fraction is dominated by Komokiacea (a superfamily of protist currently placed within the foraminifera) and similar delicate taxa (Tendal & Hessler 1977; Schro¨der et al. 1989; Gooday et al. 1997, 2007a), some examples of which are illustrated in Figure 2. This important group has a soft, flexible test wall, large intra-test accumulations of stercomata (waste pellets), and relatively sparse cytoplasm, in other words a ‘large volume/low biomass’ organization (Tendal 1979). Gooday et al. (1997) present data showing that komokiaceans are much more abundant on the Porcupine Abyssal Plain (PAP), and particularly on the more oligotrophic Madeira Abyssal Plain (MAP), than robust monothalamous foraminifera. Other studies summarized by Gooday et al. (1997, table 4) suggest that komokiaceans are a dominant faunal group in oligotrophic abyssal settings. When the test weight is included, komokiaceans alone constitute more than 90% of biomass of benthic organisms collected in grab samples (.500 mm fraction) from some parts of the Pacific Ocean (Kamenskaya 1988). In the central regions of the southern Atlantic Ocean, komokiaceans and xenophyophores (including tests) together contribute 70% of the total benthic biomass (Kamenskaya 1987). Macrofaunal foraminifera are typically quite diverse. Shires (1994; data in Gooday et al. 1998) recognized well over 100 morphospecies (complete and fragmented) in box cores from the Porcupine and Madeira Abyssal Plains (NE Atlantic). Komokiaceans (including chain-like forms), psammosphaerids and hormosinaceans were the most specious groups. Tendal & Hessler (1977) recognized dozens of komokiacean species in single deep-sea samples and guessed that the total number of species in a few bathyal, abyssal and hadal samples ‘may be in the hundreds’. Samples collected during Polarstern Cruise XXII/3 (ANDEEP III) from 15 stations (1820–4930 m water depth) in the SE Atlantic and Weddell Sea yielded around 50 morphospecies of komokiaceans and other relatively large, stercomata-bearing testate protists (Gooday et al. 2007a). Large tubular, spherical and other agglutinated foraminifera, some of which were apparently fairly delicate, sometimes fossilize (e.g. Kaminski & Gradstein 2005; Miller 2005). There are no confirmed records of fossil komokiaceans, although Aschemocella carpathica (Neagu 1964), as illustrated by Kaminski & Gradstein (2005), resembles some of
99
the chain-like komokiaceans of Gooday (1983, 1990b), Gooday et al. (2007a) and Schro¨der et al. (1988). Kaminski & Gradstein (2005, p. 24) report ‘fragile abyssal tubular forms, some of which resemble modern Komokiaceans from Cretaceous to Palaeogene ‘Scaglia-type’ biofacies in Italy and Spain. Komokiacean-like foraminifera are also reported from the Late Cretaceous Plantagenet Formation in the North Atlantic (Kuhnt et al. 1989).
Smaller foraminifera (meiofaunal size fraction) Assemblages of smaller deep-sea foraminifera (32 –300 mm) typically comprise calcareous (rotaliids, buliminids, miliolids, lageniids, robertinids) and multichambered agglutinated (e.g., trochamminaceans and hormosinaceans) species, together with a diversity of monothalamous taxa. These include simple flask-like tests with fragile but rigidly cemented test walls (Lagenammina spp.) and very delicate tests with ‘soft’, flexible test walls. The soft-walled, monothalamous species make up at least 20%, and often substantially more, of all ‘live’ (rose-Bengal stained) foraminifera in deep-sea samples (Table 1). They include three main types (Gooday et al. 2004b; Nozawa et al. 2006): (1) species with transparent, organic-walls (‘allogromiids’ in the traditional sense); (2) flask-shaped or oval forms with agglutinated walls and with a single aperture or two apertures at opposite ends of the test (‘soft-walled saccamminids’); and (3) simple spherical, agglutinated tests without apertures (‘psammosphaerids’). In addition, organic-walled species of Nodellum and Resigella, which are subdivided into more or less clearly defined chambers, are widespread in the deep sea (Gooday et al. 2004b, 2007a). Distinguishing between the fossilizable and nonfossilizable faunal components of deep-sea foraminiferal assemblages is somewhat subjective (Schro¨der 1986). The calcareous and multichambered agglutinated groups with calcareous cement usually fossilize more readily than agglutinated species with organic cement (Mackensen et al. 1990). Fossil occurrence of soft-walled monothalamous foraminifera are rare. However, a fairly diverse assemblage of well-preserved species is reported from Lower Ordovician sediments near St Petersberg, Russia (Nestell & Tolmacheva 2004). It includes members of the genus Amphitremoida in which the test is shaped like a rugby ball with apertures located at the pointed ends. Some of these fossil species are morphologically similar to some modern deep-sea saccamminids (e.g. Fig. 4d). Smaller foraminifera are very diverse in welloxygenated deep-sea habitats. Core samples of 25 cm2 surface area often yield .100 stained
100
A. J. GOODAY ET AL.
Fig. 2. Komokiaceans from the Southern Ocean, collected during Polarstern Cruise ANT XXII/2 (Gooday et al. 2007a, b). All are undescribed, in most cases at the generic level. (a) Small bunch of chambers
DEEP-SEA BENTHIC FORAMINIFERA
species, many of them non-calcareous (Gooday et al. 1998; Gooday & Hughes 2002). A compilation of data acquired using the same wet-sorting method reveals that, broadly speaking, there is an inverse relation between the abundance of monothalamous (allogromiids, saccamminids, psammosphaerids) and calcareous foraminifera (Table 1). The proportions of calcareous foraminifera at particular North Atlantic stations may be quite variable, reflecting both spatial (sites A-C of the Deep Ocean Benthic Boundary Layer [BENBO] thematic programme) and temporal (e.g. PAP) variations in the presence of phytodetritus. Nevertheless, calcareous foraminifera are generally more abundant at sites located at water depths ,2000 m than they are in deeper water. Monothalamous foraminifera usually constitute .20% of the fauna in all samples but percentages exceed 40% only on the abyssal plains. A high proportion of monothalamous foraminifera, and a low proportion of calcareous foraminifera, characterize the Cape Verde Abyssal Plain (4545 m) and the 3400 m site off Oman. An increasing proportion of monothalamous foraminifera, and a corresponding decline in calcareous foraminifera, with increasing depth is clearly evident in the Weddell Sea data (Table 1). This trend is generated largely by the sharp decrease in the absolute abundance of calcareous foraminifera at the two deepest stations (Fig. 4). Foraminifera from the 0 –1 cm sediment layers (.32 mm fraction) of samples from the Kaplan East site in the central equatorial Pacific (c. 4100 m water depth) were overwhelmingly dominated by agglutinated fragments and monothalamous species (Nozawa et al. 2006) (Table 2; Fig. 3). About two-thirds of stained specimens were either obvious fragments, mainly of tubular foraminifera, or chambers believed to be fragments of komokiaceans. Complete individuals, which represented the remaining third of the total, were dominated by psammosphaerids which constituted 27% of all foraminifera (complete plus fragments). Only about 8% of specimens were complete individuals that could be assigned to species (either described or undescribed) and most of these were monothalamous forms. Although the Kaplan East site is situated above the CCD at 4100 m water depth, calcareous foraminifera were very rare. Foraminifera living below the CCD are almost entirely agglutinated. Over vast areas of the
101
abyssal seafloor, particularly in the Pacific Ocean where the CCD is relatively shallow, foraminiferal assemblages consist mainly of monothalamous species (Saidova 1967). These often dominate the benthic biota in terms of biovolume (Snider et al. 1984). On a global scale, these protists are probably the most important eukaryotic members of abyssal sediment communities.
Trophic biology of deep-sea foraminifera Inferences about the diets of deep-sea foraminifera can be drawn from faunal observations. For example, Gooday & Lambshead (1989) recognized three different temporal and spatial patterns in relation to seasonal phytodetrital inputs at a bathyal site in the Porcupine Seabight subject to seasonal phytodetritus inputs. Some species were more abundant in May, before the arrival of phytodetritus. Some were more abundant after the spring bloom (July) and lived embedded within phytodetritus aggregates. Others exhibited a similar temporal pattern but were not directly associated with phytodetritus. These divergent patterns suggest that several trophic types with different diets can coexist at one site. More direct information is available from light and electron microscopic observations, the analysis of lipid biomarkers to identify food sources, and in-situ experimental studies addressing the responses of deep-sea benthic foraminiferal species and communities to isotopically-labelled food inputs.
Responses to experimental food inputs Recent experiments have done much to improve our knowledge of deep-sea foraminiferal ecology. Some have been conducted in shore-based laboratories, usually at 1 atm and using unlabled food sources, others in situ on the seafloor using isotopically labelled food. Laboratory experiments on samples from the bathyal NW Mediterranean revealed increases in foraminiferal population densities at both the assemblage and species level following the addition of different kinds of algal food (Heinz et al. 2001, 2002). There was no upward migration in response to the addition of food. Levin et al. (1999) undertook the first in-situ tracer experiments using 13C-labelled diatoms in order to track benthic trophic pathways. They
Fig. 2. (Continued) resembling Edgertonia; Stn 59#9 678 30.990 S, 08 0.020 W, 4650 m; (b) Komokiacean-like foraminiferan; Stn 94#14, 668 39.100 S, 278 9.250 W, 4889 m; (c) Ipoa pennata; Stn 88#8, 688 3.64 S, 208 27.500 W, 4932 m; (d) Elongate, Lana-like komokiacean; Stn 127#7, 638 35.660 S, 508 42.860 W, 2618 m; (e) Skeletonia variabilis; Stn 102#13, 658 34.310 S, 368 31.040 W, 4803 m; (f) Chain-like? komokiacean; Stn 102#13; (g) Undescribed komokiacean genus; Stn 21#5, 478 39.370 S, 48 15.650 E, 4566 m; (h) Komokiacean-like foraminiferan; Stn 102#13.
102
Location North Atlantic BENBO B: Aug. 1997 May 1998 July 1998 PSB: April July BENBO C: May 1998 July 1998 BENBO A: Aug. 1997 May 1998 July 1998 PAP: 11908#33 11908#39 11908#70
Water depth
Size fraction
1100 m
.125 mm
1340 m 1930 m 3570 m
4850 m
Allogromiids Saccamminids Psammosphaerids
Lagenammina
Hormosinacea
MAF
Calcareous
Total numbers
28.6 34.7 32.8
10.9 21.0 3.9
7.0 6.4 3.7
4.8 3.8 9.7
27.3 32.9 49.4
560 1221 1017
*39.3 *24.4
ND ND
8.18 2.01
11.0 11.7
25.7 55.7
2663 3565
37.6 30.0
8.56 6.82
18.1 8.18
9.45 6.89
17.5 45.3
2265 6547
.45 mm .63 mm .125 mm 31.0 24.7 19.8
18.1 27.0 20.3
40.5 31.5 25.7
2.7 4.8 5.1
5.7 11.7 28.2
331 504 369
29.4 33.7 42.2
11.2 11.2 16.9
12.0 10.7 8.3
3.39 4.59 6.52
23.6 23.4 9.57
638 762 651
.63 mm
A. J. GOODAY ET AL.
Table 1. Percentage abundance of foraminiferal groups in samples from different depths in the Atlantic, Indian and Pacific Oceans. MAF ¼ multichambered agglutinated taxa including Trochamminacea but excluding the Hormosinacea. Data derived from following sources: BENBO Sites A and B: Hughes (2004); BENBO Site C: Gooday & Hughes (2002); PSB: Gooday & Lambshead (1989); PAP, MAP, CVAP: Gooday (1996); Indian Ocean: Gooday et al. (2000); Weddell Sea: Cornelius & Gooday (2004); Pacific Ocean: Gooday et al. (2004). All studies employed the same wet-sorting methodology. Not all groups are included and the total percentages therefore do not add up to 100%
MAP: 12174#15 12174#24 12174#88
4950 m
CVAP
4545 m
Indian Ocean Oman margin
.63 mm 47.9 44.3 44.9
6.01 8.08 7.61
7.13 7.27 14.17
1.56 5.17 1.81
7.57 9.53 6.29
450 605 394
.63 mm
56.8
7.45
6.09
3.93
1.03
560
3400 m
.63 mm
52.9
5.93
22.3
3.90
6.74
1604
1100 m 2080 m 3050 m 4065 m 4975 m
.63 mm .63 mm .63 mm .63 mm .63 mm
6.3 10.8 17.2 22.6 27.3
0.3 4.9 7.3 7.3 3.6
0.9 3.5 8.7 5.9 0.8
12.3 9.6 18.0 18.5 24.5
64.9 52.0 31.3 14.2 15.7
350 596 412 287 249
5350 m 5289 m 4263 m
.32 mm .32 mm .32 mm
36.7 31.8 40.9
4.6 5.8 6.5
19.7 14.6 5.3
27.4 28.5 24.3
4.9 14.4 16.0
549 738 337
W. Central Pacific: Stn 64 sample A Stn 64 sample B
5570 m 5570 m
.32 mm .32 mm
57.4 62.8
12.0 13.6
15.3 5.9
5.9 5.4
4.7 4.4
469 337
*Includes Lagenammina.
DEEP-SEA BENTHIC FORAMINIFERA
Weddell Sea Stn 133 Stn 132 Stn 131 Stn 134 Stn 137 Pacific Ocean N. Pacific: Stn 2/3 Stn 6 Stn 15
103
104
A. J. GOODAY ET AL.
Fig. 3. Numbers of rose Bengal stained specimens of different faunal groups at different water depths in the Weddell Sea. Data from Cornelius & Gooday (2004).
focussed on macrofaunal species at Sites I (off Cape Fear) and III (off Cape Hatteras), both located at 850 m water depth on the North Carolina margin. At Site I, more than half of the 39 specimens of large agglutinated protists examined, including Rhizammina, Hyperammina, Bathysiphon and psamminid xenophyophores, exhibited label uptake after 1.0 and 1.5 days. A similar proportion of specimens (14 out of 26) retained the 13C signature after 14 months. At Site III, tracer uptake was dominated by annelids. Moodley et al. (2002) used a benthic lander to conduct in-situ experiments at a 2170 m deep site on the Iberian margin of the NE Atlantic. 13 C-labeled algae were processed by the examined compartments of the benthic community (bacteria and fauna .300 mm in the upper 5 cm of sediment plus community respiration) within a 35-hr period. Foraminifera were responsible for processing 29.1% of the added carbon and bacteria 21.7%. The highest 13C enrichment values (þ2952 and þ3993 ‰) were found among foraminifera; ‘familiar’ taxa (calcareous, agglutinated, soft-shelled) yielded much higher values than ‘unfamiliar’ taxa (komokiaceans, large astrorhizids including Rhizammina). However, bacteria comprised most (95%) of the benthic biomass whereas foraminiferal biomass was ,1.5%. Moodley et al. (2002) concluded that foraminifera have highly efficient food-gathering capabilities and play a crucial role in the short-term packaging and initial processing of labile organic food at their study site. Nomaki et al. (2005a) reached a similar conclusion based on seafloor incubations at a bathyal site in Sagami Bay, Japan (1449 m water depth) using 13C-labeled algae (Dunaliella) deployed from a submersible. As on the Iberian margin, uptake of carbon by benthic foraminifera was rapid and substantially higher than for
metazoans. However, the responses varied from species to species. Two shallow infaunal species (Uvigerina akitaensis and Bulimina aculeata) showed more rapid and higher uptake rates, as well as faster increases in abundance, than intermediate infaunal species (Bolivina pacifica and Textularia kattegatensis) and deep infaunal species (Globobulimina affinis and Chilostomella ovoidea). In B. aculeata, 16 –18% of protoplasmic biomass was derived from the labled algae after only 2 days; after 6 and 11 days the proportion rose to more than one third. The other shallow infaunal species, U. akitaensis, exhibited rather slower uptake rates, but concentrations reached 22% and 40% after 6 and 11 days respectively. Uptake of the labeled carbon by the two intermediate infaunal species, and one of the deep infaunal species, was slow, but after 11 days it accounted for 33% (Bolivina pacifica), 13% (G. affinis) and 5.3% (T. kattegatensis) of the biomass. Further in situ experiments involving 13C-labelled bacteria and algae confirmed these results (Nomaki et al. 2006). They showed also that none of the above mentioned species selectively consumed bacteria, although B. aculeata, T. kattegatensis, Cyclammina cancellata, Globobulimina affinis and Chilostomella ovoidea appeared to ingest them randomly. In laboratory experiments using sediments from the same Sagami Bay site, there was no overall increase in foraminiferal densities between aquaria fed with unlabled algae and unfed aquaria (Nomaki et al. 2005b). However, shallow and intermediate infaunal species migrated upwards within the sediment following the addition of food while G. affinis reacted very slowly and C. ovoidea showed no migratory response. In some earlier pressurized experiments (Ohga 1995), these two species added a new chamber every 90–120 days, a slow growth rate suggesting that a full life cycle occupied two years (Ohga & Kitazato 1997). Most deep-sea foraminiferal experiments, both laboratory-based and in situ, have focussed on continental margins. Witte et al. (2003a) carried out the first tracer experiment at an abyssal site remote from land. They used landers to track prokaryotic and macrofaunal responses to a pulse of 13 C-labeled phytodetritus on the PAP (NE Atlantic, 4850 m water depth). The sediment community oxygen consumption (SCOC) showed a significant increase after only 2.5 days, probably arising mainly from bacterial respiration. Bacterial extracellular enzyme activity, and the incorporation of labelled carbon into bacterial phospholipids-derived fatty acids, however, increased at a slower rate while total bacterial numbers did not increase significantly, possibly due to grazing by larger organisms. Foraminifera responded more slowly than bacteria. Uptake was low after 2.5 days but then increased
DEEP-SEA BENTHIC FORAMINIFERA
105
Fig. 4. Monothalamous foraminifera from the equatorial East Pacific (Nozawa et al. 2006). All are undescribed.
to dominate carbon uptake after 23 days. In comparison, uptake by metazoans, particularly the macrofauna, was much faster with substantial 13 C-labeling evident after 2.5 days. These experiments suggest that, in contrast to the continental margin sites of Moodley et al. (2002) and Nomaki et al. (2005a), metazoan macrofauna, rather than
bacteria and foraminifera, dominate the initial processing of labile food at this abyssal site. This may reflect a higher proportion of sediment feeding foraminifera at the PAP, combined with physiological differences between foraminifera living at bathyal and abyssal depths (Nomaki et al. 2005a) and the effects of low bottom-water temperatures at
106
A. J. GOODAY ET AL.
Table 2. Total numbers of specimens and species of complete and fragmented foraminifera in 7 replicate samples from Kaplan East site the eastern Equatorial Pacific (15 8N, 119 8W; 4100 m water depth). Data from Nozawa (2005)
Agglutinated fragments Indet. agglutinated spheres Other complete monothalamous Complete multichambered agglutinated Complete calcareous Totals
4850 m (Moodley et al. 2005). Since the PAP is a relatively eutrophic abyssal plain, we predict that the response of foraminifera will be even slower in other, more oligotrophic abyssal settings. Labelled carbon uptake as a proportion of biomass in different faunal compartments, based on deep-sea and shallow-water experiments, is summarized in Figure 5 (based on Nomaki 2005). Foraminiferal biomass is often lower than bacterial biomass (Moodley et al. 2002; Nomaki 2005; Witte et al. 2003a; Kro¨nke et al. 2000). Nevertheless, with the exception of the abyssal site of Witte et al. (2003a), foraminifera always show higher ingestion rates of carbon relative to biomass than metazoans and bacteria. This relationship applies in shallow water and in the bathyal deep sea, despite differences in temperature, a parameter that exerts a strong control on the speed and magnitude of benthic cycling of fresh organic matter (Moodley et al. 2005). The overall conclusion that we draw from these experiments is that foraminifera are very active in cycling organic carbon on continental margins, but less active on abyssal plains.
Diets Lipps & Valentine (1970) concluded that foraminifera feed mainly at a low trophic level on minute organisms and detritus and form a vital link to higher levels of marine food chains. The relatively small number of coastal species that have been studied consume mainly diatoms, other algae and bacteria (reviewed by Lee 1974, 1980; Goldstein 1999). The pioneering 14C tracer studies of J. J. Lee and colleagues in the 1960s and 1970s revealed that littoral species feed selectively when offered a range of food items and that differences in the nutritional quality of the food are reflected in reproductive and growth responses (e.g. Lee et al. 1966; Lee & Muller 1973; Lee 1980). The extent to which these results can be extrapolated to deeperwater habitats is unclear, although it seems inevitable that deep-sea foraminifera, particularly those
Specimens
Species
4266 (49.1%) 3455 39.8%) 793 (9.1%) 144 (1.66%) 26 (0.30%) 8684
84 (33.5%) indeterminate 124 (49.4%) 34 (13.5%) 9 (3.58%) 251
living in central oceanic regions, have fewer potential food types to choose from. Attempts to categorize the diets and feeding habits of benthic foraminifera often involve a considerable degree of speculation. In the most recent review, Murray (2006) recognized the following foraminiferal feeding strategies: herbivory and bactivory, passive suspension feeding, detritivory, carnivory, utilization of dissolved organic material, omnivory and parasitism. Gooday et al. (1992b) inferred the existence of a number of trophic groups among deep-sea species. They distinguished between: (1) feeding on rapidly sedimented material derived from surface phytoplankton production by species living near the sediment surface; (2) deposit feeding on more degraded material by infaunal species; (3) omnivorous feeding on organic detritus; (4) deposit feeding combined with the accumulation of stercomata; and (5) osmotrophy. Gooday et al. (1997) drew a distinction between large, robust, monothalamous foraminifera (e.g. Bathysiphon, Rhabdammina) living on euthrophic continental margins, and stercomata-accumulating taxa, which are more common in oligotrophic regions. In their Sagami Bay experiments, Nomaki et al. (2006) used diatoms to represent food derived from the water column (algal-based phytodetritus) and bacteria to represent food derived from the sediment. On this basis, they recognized three trophic types: (1) phytophagous species (herbivores) that consume only phytodetritus; (2) seasonal-phytophagous species that consume phytodetritus when it is available and sedimentary organic matter at other times; and (3) non-selective deposit feeders that accumulate and ingest degraded organic matter, bacteria and other organisms present in the sediment. Fontanier et al. (2003, 2005) also distinguished between species that feed on labile and on more refractory organic matter. Based on these ideas, and considering evidence published since the review of Gooday et al. (1992b), we suggest that deep-sea foraminifera can be divided into the following trophic types. We emphasize that these
DEEP-SEA BENTHIC FORAMINIFERA
107
(a) 100
Carbon uptake (mg C m–2 d–1)
10
on
=
10
rb
d
1
on
of
ad
ed
ca
cti
a Fr 0.1
Bacteria Foraminifera Meiofauna Macrofauna Fauna
1%
0.01
1%
0. 0.001 1
10
100
1000
104
105
Biomass (mg C m–2)
Bacteria Foraminifera Meiofauna Macrofauna Fauna
Fraction of added carbon in biomass (%)
(b) 10
1
0.1
0.01
0
1000
2000
3000
4000
5000
Water depth (m) Fig. 5. (a) Relationship between organism biomass and assimilation rate of algal carbon in different faunal groups derived from in-situ 13C-labelling experiments at the deep-sea floor; (b) labelled algal fraction in different faunal groups derived from in-situ 13C-labelling experiments. Data for both panels are derived from Middelburg et al. (2000), Moodley et al. (2000, 2002, 2005), Witte et al. (2003a, b) and Nomaki et al. (2005b). Assimilation rates and labelled algal fraction are normalization to a 1 day incubation period, based on original incubation periods ranging from 12 hours to 4 days. Modified from Nomaki (2005).
108
A. J. GOODAY ET AL.
types are based on a mixture of direct evidence and speculation and that the boundaries between them are often blurred. In addition to the feeding types discussed below, it is possible that some large bathyal foraminifera are capable of carnivory. Herbivorous species. These are specialist consumers of seasonally-pulsed phytodetritus. They are most common on bathyal continental margins but also occur on some eutrophic abyssal plains. Nomaki et al. (2005a, 2006) recognized three herbivorous (phytophagous) species, Uvigerina akitaensis, Bolivina spissa and Bolivina pacifica, of which the first two occupied shallow-infaunal microhabitats while the third was intermediate infaunal. Faunal studies suggest that other epifaunal/ shallow infaunal species may be phytophagous. They include species that exploit phytodetritus deposits, such as Alabaminella weddellensis, Bolivina spathulata, Cassidulina carinata, Epistominella exigua, E. arctica, and Eponides pusillus in the Atlantic and Southern Oceans (Gooday 1988, 1993; Gooday & Lambshead 1989; Gooday & Turley, 1990; Gooday & Hughes 2002; Fontanier et al. 2003; Cornelius & Gooday 2004). Like some coastal species (Lee 1980), deep-sea herbivorous foraminifera probably choose nutritious food yielding high rates of growth and reproduction. However, there is evidence from fluorescence and transmission electron microscopy that E. exigua ingests cyanobacteria, bacteria and flagellate scales in addition to algae (Turley et al. 1992), and so the diet of this and similar species should not be categorized too rigidly. Opportunistic deposit feeders (omnivores). We include here a rather broad assortment of species that selectively ingest phytodetritus but can also feed on other substrates when this high quality food source is not available. There is probably no clear boundary between herbivorous foraminifera and these more omnivorous species. Heeger (1990; summarized by Gooday et al. 1992b) observed a range of recognizable food items within the cytoplasm of calcareous species in the deep GreenlandNorwegian Sea. They included pennate diatoms, dinoflagellates, coccoid and thecate algae, bacteria, copepod cuticle and siliceous aggregates. Species living on elevated substrates contained fresher food particles than species living in the sediment. The variety of vacuole contents observed by Heeger (1990) suggests that the species he studied were fairly unselective. Goldstein & Corliss (1994) studied the diets of two deep-sea species, Uvigerina peregrina (shallow infaunal) and Globobulimina pacifica (deep infaunal), from 710 m on the California Borderland. The cytoplasm contained parcels of sediment and associated organic detritus and bacteria.
Fragments of diatom frustules were present in the sediment parcels of U. peregrina but not those of G. pacifica. Goldstein & Corliss (1994) regarded these species as deposit feeders because they gather together, and then ingest, sediment and a wide variety of associated organic particles. Nomaki et al. (2006) considered three species in Sagami Bay to be ‘seasonal phytophagous species’, which selectively consumed algal food if it is available and sedimentary organic matter at other times of the year. A mixture of shallow, intermediate and deep infaunal species (Bulimina aculeata, Textularia kattegatensis, Globobulimina affinis respectively) exhibited this behaviour. The species examined by Goldstein & Corliss (1994) were probably characterized by a similar degree of trophic flexibility. Suhr et al. (2003) analysed fatty acids in three benthic foraminiferal species: the rotaliid Globocassidulina subglobosa, the miliolid Quinqueloculina seminula, and the monothalamous species Thurammina albicans, from the deep Antarctic Peninsula shelf (560 m water depth). Multidimensional scaling plots of replicate fatty acid profiles revealed four clusters corresponding to the three species and the surrounding phytodetritus. The clusters were tighter for the two calcareous species than for Thurammina albicans, suggesting that they have more distinct and therefore selective diets. The calcareous species yielded significantly more polyunsaturated fatty acids (PUFA) than the phytodetritus and appeared to be feeding selectively on different components of this labile material. Moreover, G. subglobosa contained significantly higher amounts of PUFA before the spring bloom than in post-bloom samples and the fatty acid composition also differed significantly between the two seasons. Thus, G. subglobosa seems to have a stronger response to labile food derived from the spring phytoplankton bloom than Q. seminula. These three species probably represent a trophic spectrum reflecting a greater through to a lesser dependence on phytodetritus. Some large agglutinated foraminifera that are common on continental margins, including Bathysiphon spp., also appear to be omnivores (Gooday et al. 1992a, b). Gooday et al. (2002) present fatty acid data for one such species, Bathysiphon capillare, a long, slender tubular infaunal foraminiferan that lives in the upper 5 cm of sediment on the Scottish margin. The cytoplasm contains numerous small pellets (stercomata) consisting of small mineral particles, presumably derived from ingested sediment. Fatty acid analyses suggests that it consumes bacteria together with some fresh algal material (phytodetritus). Rapid uptake of 13 C-labelled diatoms by other large agglutinated species, including Pelosina sp., Hyperammina, Bathysiphon rufum, and an astrorhiziid-like
DEEP-SEA BENTHIC FORAMINIFERA
foraminiferan, was reported at Site I on the North Carolina margin (Levin et al. 1999). Intermediate- and deep-infaunal sediment deposit feeders. The in-situ experiments of Nomaki et al. (2006) in Sagami Bay indicated that the deep infaunal species Chilostomella ovoidea ingested bacteria and algae unselectively. They regarded this species as a sediment deposit feeder. Fontanier et al. (2005) proposed that high abundances of intermediate- and deep-infaunal species (C. oolina, Globobulimina spp., M. barleeanum) at a 2300-m deep canyon site in the Bay of Biscay with oxic bottom water and a relatively low downward organic flux reflected high inputs of degraded organic matter. They suggested that these species consume bacteria associated with redox fronts within the sediments (see also Jorissen et al. 1998). Earlier, Caralp (1989) had suggested that M. barleeanum feeds on more refractory material than Bulimina exilis. Reports that G. affinis ingested 13C-labelled diatoms after 6 days during in-situ tracer experiments, and that Chilostomella oolina underwent significant increases in abundance 4 weeks after phytodetritus inputs at station M in the NE Pacific (Drazen et al. 1998), sounds a cautionary note and suggest that some degree of trophic flexibility exists among these deep infaunal foraminifera. Stercomata-bearing foraminifera. Many deep-sea monothalamous taxa and komokiaceans, in addition to xenophyophores and gromiids, accumulate stercomata (Tendal 1979; Gooday et al. 1992b). These pellets of waste material consist largely of clay minerals, presumably derived from ingested sediment, suggesting that stercomata-bearing foraminifera are deposit feeders and digest organic matter and bacteria associated with the sediment. As indicated below, it is also possible that they capture suspended particles. Stercomata-bearing taxa are characteristic inhabitants of oligotrophic parts of the deep sea, for example, the Challenger Deep (Todo et al. 2005), the central Arctic Ocean (Wollenburg & Mackensen, 1998) and the central North Pacific (Bernstein et al. 1978), eastern equatorial Pacific and other parts of the Pacific Ocean (Kamenskaya 1988). Laureillard et al. (2004) report that the xenophyophore Syringammina corbicula contained a higher level of fatty acids characteristic of bacteria than the surrounding sediment. However, the precise relationship between the xenophyophore and the bacteria is obscure. This giant foraminiferan (Pawlowski et al. 2003b) may feed on and digest bacteria directly, or it could house symbiotic bacteria in the cytoplasm or, as suggested by Tendal (1979), it could use the stercomata for cultivating bacteria that are subsequently harvested as a food source. This latter strategy may extend to other
109
taxa that accumulate stercomata, although there is currently no direct evidence for bacterial farming by any of these foraminifera. Transmission electron microscopy (TEM) provides some tantilizing clues about the trophic biology of these enigmatic groups. Two preliminary studies suggest that in Rhizammina algaeformis (a komokiacean-like species) and the xenophyophore Aschemonella ramuliformis, the strand-like cytoplasm is invaded by open spaces bounded by cell membranes (Cartwright et al. 1989; Hopwood et al. 1997). These seem to be ‘gut-like’ invaginations of the cell surface which must increase the surface area substantially and may be involved in food uptake. These observations were based on formalin-fixed material and need to be confirmed by TEM studies of optimally-fixed samples. Nevertheless, the closely similar appearance of the cytoplasm in R. algaeformis and A. ramuliformis suggests a common trophic strategy involving the capture of fine, particulate material, either directly from the sediment or from suspension. We speculate that this kind of cell structure is typical of xenophyophores, komokiaceans and other large, stercomata-bearing taxa. Bacteriverous monothalamous foraminifera? Experimental studies revealed no evidence for the selective uptake of bacteria by 8 multichambered foraminiferal species in Sagami Bay (Nomaki et al. 2006). Similarly, tracer experiments suggest that bacteria are generally not eaten by littoral foraminifera (Lee et al. 1966; Lee 1980), although some species appear to require a bacterial component for sustained reproduction (Muller & Lee 1969). On the other hand, Allogromia sp. (NF) consumes bacteria as well as algae and can survive for some years in cultures containing a single bacterial species (Lee & Pierce 1963). Bernhard & Bowser (1992) observed this species, and A. laticollaris, harvesting and consuming bacterial biofilms in laboratory cultures. Langezaal et al. (2005) also reported that A. laticollaris, and the calcareous foraminiferan Ammonia beccarii, collected fluorescently-stained bacteria using their pseudopodia and incorporated them into the cell body. The above-mentioned species live in shallow water. A predominantly bacteriovorous diet is also possible for some small, monothalamous foraminifera in the deep sea. Turley et al. (1992) describe pressurized experiments during which abyssal sediment samples were incubated with three species of microalgae and bacteria-sized (0.73 mm) fluorescent microspheres. The microspheres were the only particles ingested by an undescribed tinogullmid allogromiid with well-developed terminal apertures. A similar tinogullmid is sometimes found within crustacean moults (Gooday 1990a) and long vermiform monothalamous
110
A. J. GOODAY ET AL.
foraminifera occur within organic tubes (part of the granellare system) that formerly enclosed the protoplasm of the xenophyophore Syringammina fragilissima (Hughes & Gooday 2004). These are environments where bacteria are likely to flourish. Based on these and other observations, Gooday (2002) argued that some deep-sea monothalamous foraminifera (‘allogromiids’) are bacteriovores. Here, we speculate that they ingest bacteria in the same way as deep-sea benthic nanoflagellates (Turley et al. 1988). This hypothesis could be tested using ultrastructural and experimental approaches. Suspension feeders. These are well documented among erect, epifaunal species (e.g. Lipps 1983). Schemes that assign foraminifera to functional morphogroups have classified Rhizammina, komokiaceans and other basically tubular deep-sea forms as suspension feeders (Jones & Charnock 1985; Kaminski & Gradstein 2005). Large erect species are almost certainly able to feed in this way; examples from soft-bottom, deep-sea habitats include Arborammina hilaryae, Pelosina arborescens and Saccorhiza ramosa (Shires et al. 1994 and references therein). Similar species occur on hard substrates such as manganese nodules (Mullineaux 1987) and hydrothermal constructions (Kamenskaya et al. 2002). It is likely that some stercomata-bearing foraminifera, which we interpret as deposit feeders, are able to intercept particles above the sediment surface. The fan-like and reticulate morphologies found among large xenophyophores may facilitate particle trapping where currents are enhanced (Levin & Thomas 1988). Interestingly, Sokolova (2000) considers the oligotrophic, central oceanic regions, where stercomata-bearing foraminifera are abundant, to be dominated by suspension feeders. Nevertheless, the existence of infaunal komokiaceans (Tendal & Hessler 1977; Kuhnt & Collins 1995), xenophyophores (Gooday et al. 2001), and large tubular foraminiferans (e.g. Gooday et al. 2002) implies that at least some of these forms are deposit feeders. In general, the mixture of infaunal and epifaunal life positions adopted by tubular and branching agglutinated foraminifera in the deep sea suggests that, except for the erect, often arborescent morphotypes, test morphology is a relatively poor guide to their trophic biology.
Distribution of trophic types in the deep ocean Many deep-sea foraminifera can probably utilize different food sources as they become available and hence the trophic types outlined above cannot be considered as rigid categories. To a greater or
lesser extent, they must grade into each other. The different responses to food pulses reported for Chilostomella species (Drazen et al. 1998; Nomaki et al. 2005a), and the consumption of different food types by Globobulimina species, illustrates this point. Nevertheless, the endpoints of the spectrum from species (often calcareous) that feed predominately on labile organic matter derived from phytoplankton production to species, including those which accumulate stercomata, that seem to feed on low quality food, must surely reflect a real ecological contrast. Herbivorous and other phytodetritus feeders occur mainly on continental margins and on more eutrophic abyssal plains (Table 1). The stercomata-bearing foraminifera are typically dominant in the more oligotrophic parts of the deep sea. Rates of growth, reproduction and metabolic activity in deposit-feeding foraminifera are probably slower than those of herbivorous species that consume more labile food. Faunal studies suggest that species feeding on phytodetritus grow and feed more rapidly than those that exclusively inhabit the sediments (Gooday & Lambshead 1989; Gooday & Rathburn 1999; Gooday & Hughes 2002). In Sagami Bay, Japan, two shallow infaunal species, Bolivina pacifica and Textularia kategattensis, both grew faster after phytodetritus deposition whereas deep-infaunal species (Chilostomella ovoidea and Globobulimina spp.) grew slowly and did not show clear seasonal growth patterns (Ohga & Kitazato 1997). Bertram & Cowan (1999) distinguished between two foraminiferal assemblages colonising experimental substrates deployed at 800 to 1985 m water depth on the Cross Seamount, south of Hawaii. Many agglutinated foraminifera (termed ‘stable assemblage’) colonized the substrates at uniform rates whereas the colonization rates of calcareous species (termed ‘transient assemblage’) fluctuated over time in a way that was probably related to the flux of particles from the upper ocean. These authors conclude that members of the stable assemblage feed mainly on bacteria whereas members of the transient assemblage feed on organic detritus derived from surface production. Cedhagen & Mattson (1991) report some intriguing observations of live Globopelorhiza sublittoralis, a ‘mudball’ foraminiferan from 60– 720 m off the Swedish west coast. Pseudopodial movements and burrowing activity, and by inference metabolic rate, were much slower than in co-occurring species. Although G. sublittoralis is not a komokiacean, it has a similar ‘high volume/low biomass’ organization. We agree with Cedhagen & Mattson (1991) that deep-sea komokiaceans probably have similar metabolic characteristics. In summary, we suggest that there are contrasts in the trophic biology and ecology, as well as in the
DEEP-SEA BENTHIC FORAMINIFERA
overall taxonomic composition, of foraminiferans living on relatively eutrophic continental margins and more oligotrophic abyssal plains.
Wider implications The development and refinement of proxies based on the faunal characteristics and shell isotope chemistry of benthic foraminifera depends on a detailed understanding of foraminiferal ecology, a requirement that has driven recent advances in this field. Most of this effort has focused on familiar calcareous species which inhabit continental margins and on which proxies are largely based. The relevance of the observations reviewed above to the development of palaeoceanographic proxies is rather mixed. Trophic preferences of deep-sea foraminifera, including calcareous species, are probably most important, for example, in trying to understand benthic foraminiferal responses to organic fluxes, their abundance and distribution in relation to productivity gradients (Altenbach et al. 1999), and shell isotope chemistry. On the other hand, the softshelled foraminifera, which are particularly abundant in abyssal regions, are of little interest to most geologists. The vast majority of these species are unlikely to fossilize, and assemblages preserved in ancient deep-sea sediments therefore do not represent the entire original foraminiferal fauna. This does not necessarily undermine the use of foraminifera as proxies in palaeoceanography, since these are based on empirical relationships between the faunal characteristics of hard-shelled assemblages and environmental parameters. In this sense, the soft-shelled species are no different from nematodes, polychaetes and other soft-bodied organisms that abound in the deep sea. They are part of the total benthic community that must have been present in ancient oceans. Nevertheless, we suggest that these protists are not entirely irrelevant in the Earth sciences. They may have some bearing on the development of palaeoproductivity proxies. They probably play a significant role in carbon cycling at abyssal depths and are of crucial importance for understanding the early evolutionary history of foraminifera. In the following sections, we discuss these issues in more detail.
Interpretation of the palaeo-record A sizable proportion of deep-sea foraminiferal biodiversity must be lost from the Quaternary and Cenozoic fossil record. This applies particularly in the oligotrophic abyss where the soft-shelled component dominates. Nevertheless, the wide utility of foraminifera in palaeoceanography reflects the
111
considerable amount of environmental information conveyed by those species that are preserved. Murray & Alve (1999, 2001 reviewed by Gooday 2003) used acid to dissolve the calcareous species and agglutinated species with calcareous cement, thereby reducing diversity still further. They found that the environmental signal survived the removal of 95% of the original assemblage. Hence the loss of monothalamous species and komokiaceans from the fossil record does not necessarily compromise the use of foraminiferal faunal data in palaeoceanographic reconstructions. In any case, these reconstructions tend to be concentrated in bathyal, continental margin settings where delicate foraminifera are relatively less abundant. The Benthic Foraminiferal Accumulation Rate (BFAR) has been proposed as a palaeoproductivity proxy (Herguera & Berger 1991; Herguera 2000). This approach depends on a number of assumptions, including a linear relationship between the flux of organic matter to the seafloor organic and surface productivity, a similar relationship between the organic flux and the number of foraminiferal shells, a constant sedimentation rate, and no significant carbonate dissolution (Herguera & Berger 1991). When used with care, BFAR seems to provide a semi-quantitative proxy for organic flux to the seafloor, although extrapolations to primary productivity are problematic (Jorissen et al. 2007) and the method is ineffective in low-oxygen settings (Naidu & Malmgren 1995). However, BFAR values will reflect the nature and quality of the organic matter as well as its quantity (Guichard et al. 1999). Fossilizable foraminifera are relatively more abundant in eutrophic areas with substantial inputs of labile material compared to nonfossilizable agglutinated taxa which dominate in oligotrophic areas. The partitioning of the organic carbon flux between these two faunal compartments is likely to vary accordingly. This could lead to a bias in BFAR values that may be important when comparing areas of different productivity (Herguera 2000). On a more positive note, the fact that, in a general sense, hard-shelled, calcareous foraminifera are linked more closely to the organic flux than the delicate taxa that do not fossilize, tends to support the BFAR approach. The stable carbon isotope signal preserved in benthic foraminiferal carbonate tests reflects global ocean circulation and organic carbon fluxes. It is also subject to local influences, notably gradients in the isotope ratios of sediment pore-waters (‘microhabitat effect’), and ‘vital effects’ possibly related to the incorporation of metabolic carbon into the shell material (McCorkle et al. 1990; Mackensen & Bickert 1999; Schmiedl et al. 2004). Mackensen et al. (1993) recognized a ‘phytodetritus effect’ whereby deviations in d13C
112
A. J. GOODAY ET AL.
values derived from tests of the epifaunal species Cibicidoides wuellerstorfi were attributed to calcification within the 13C-depleted microenvironments of phytodetritual aggregates. This can be seen as an extension onto the sediment surface of the ‘microhabitat effect’. Differences in the d13C values of some shallow-infaunal species (e.g. Uvigerina peregrina, Bulimina aculeata) from areas characterized by different organic-matter fluxes have also been reported (reviewed by Schmiedl et al. 2004). Another process contributing to these patterns may be the incorporation into foraminiferal shells of isotopically light carbon derived from the ingestion of fresh phytodetritus (Gooday 1996). The isotopic signature of bacteria reflect values of sedimentary organic matter (Coffin et al. 1994; Kelley et al. 1998) and therefore differ from those of phytodetritus. Thus, contrasting dietary preferences (e.g. Nomaki et al. 2005a, 2006) may enhance differences in carbon isotope signatures due to other factors. It is also likely that carbon isotope values of sedimentary detritus consumed by deposit feeders are influenced on continental margins by carbon derived from terrestrial plants.
although macrofaunal metazoans and particularly polychaetes may contribute significantly to the initial processing of labile material through the rapid ingestion and bioturbation of freshly deposited phytodetritus (Levin et al. 1997; 1999). We argue above that herbivorous and omnivorous foraminifera which consume labile food sources derived from photosynthetic production are more abundant on continental margins. Because they exploit phytodetritus inputs, it is probably these trophic types that contribute most to the remineralization of organic carbon on continental margins (Moodley et al. 2002; Nomaki et al. 2006). Foraminifera, including monothalamous forms, which we suggest depend on more refractory organic material and bacteria predominate in central oceanic regions. Although they must process organic matter at a slower rate than herbivorous species (Moodley et al. 2002), and probably have slower rates of reproduction, growth and metabolism, the shear abundance of these poorly-known forms on abyssal plains (Tendal & Hessler 1977; Bernstein et al. 1978) suggest that they play a substantial role in the decomposition of organic carbon at abyssal depths.
Foraminiferal phylogeny Global biogeochemical cycling Intensified primary productivity, high rates of vertical fluxes, and lateral inputs from the shelf and slope make continental margins important areas for biogeochemical cycling (Jahnke et al. 1989; Walsh 1991; Romankevich et al. 1999; Laws et al. 2000; Antia et al. 2001; Lochte et al. 2002). According to Jahnke et al. (1989), the highest benthic remineralization rates, and by inference the highest input rates of organic carbon, occur within a 100 km-wide zone adjacent to the base of the continental slope in their study area off central California. They conclude that more than half of the flux of material to the seafloor across a .3000 km transect in the NE Pacific occurs within 500 km of the continental slope. Jahnke (1996) compiled data for benthic oxygen flux rates in the deep ocean (.1000 m water depth) and then extrapolated between the sites using an empirically derived relationship between benthic oxygen fluxes and the carbonate-free organic carbon burial rate. This approach yielded low flux rates in the central oceanic gyres of all ocean basins and high rates for continental margins. However, because they occupy a greater area, the central oceanic gyres accounted for about half of the total benthic oxygen flux below 1000 m depth. The pulse-tracking experiments reviewed above suggest that foraminifera play a key role in short-term, organic matter cycling on continental margins (Heip et al. 2001; Moodley et al. 2002),
Describing the ‘foraminiferal biodiversity iceberg’ in modern oceans is an important objective for understanding the early evolution of foraminifera. Foraminifera probably originated in the Neoproterozoic, long before the earliest known foraminiferal fossils from the basal Cambrian (Pawlowski et al. 2003a). Much of the early history of these protists is hidden from view and can only be glimpsed through the application of molecular techniques. Analysis of SSU sRNA gene sequences confirms what was assumed by some, although by no means all, earlier authors (e.g. Tappan & Loeblich (1988); Cifelli 1990) – that monothalamous agglutinated and organic-walled forms lie at the base of the foraminiferal evolutionary tree (Pawlowski et al. 2003a). However, the molecular evidence also reveals that, rather than a linear progression from organic-walled to agglutinated forms, the early evolution of foraminifera involved a bush-like radiation of lineages in which these two wall types evolved repeatedly, in some cases within the same clade. Studies of the molecular genetics of deep-sea monothalamous taxa has already contributed to our knowledge in this area (Gooday & Pawlowski 2004; Gooday et al. 2004a). More direct evidence about the evolutionary history of ‘primitive’ foraminifera could come from fossil occurrences. The fossil record of soft-shelled foraminifera is poor (Tappan & Loeblich 1988). Many genera are Palaeozoic and, with some exceptions (Nestell & Tolmacheva 2004), do not resemble modern forms
DEEP-SEA BENTHIC FORAMINIFERA
(Loeblich & Tappan 1987). We know of no convincing deep-sea records, although Palaeozoic and Mesozoic cherts preserved within ophiolite complexes in parts of Japan (Kato et al. 2002) are worth investigating for possible examples. Vaseshaped microfossils (VSMs) occur widely in the Neoproterozoic of Sweden. There is good morphological evidence that well-preserved VSMs from the Grand Canyon, and probably from other localities, are testate amoebae (Porter & Knoll 2000). These records enhance the likelihood that soft-walled monothalamous foraminifera, which are morphologically similar to testate amoebae, will be found in ancient sediments. In addition to body fossils, evidence for the existence of softbodied foraminifera in the past might be provided by chemical biomarkers. Other studies have examined the molecular genetics of deep-sea calcareous foraminifera. Ertan et al. (2004) analysed phylogenetic relationships between the main foraminiferal groups, as well as within the buliminids and rotaliids, based on material from intertidal to bathyal depths. This and similar investigations may provide important insights into the evolutionary history of deep-sea foraminifera in relation to palaeoceanographic changes during the Mesozoic and Cenozoic. Other studies reveal conflicting evidence regarding genetic differentiation within deep-water morphospecies. Schweizer et al. (2005) found almost no genetic divergence within a morphologically highly variable population of Uvigerina peregrina from Oslo Fjord. Grimm et al. (2007) reported a wide distribution for some bathyal genotypes of Chilostomella. They also found evidence for cryptic speciation within this genus, a phenomenon very prevailent in the coastal genus Ammonia (Hayward et al 2004). Another recent study has revealed a high degree of genetic homogeneity over a wide geographical range at abyssal depths in Cibicidoides wuellerstorfi, Epistominella exigua and Oridorsalis umbonatus (Pawlowski et al. 2007). These results provide palaeoceanographers with confidence that at least some important morphospecies display consistent ecological responses to environmental gradients across their entire range. At the same time, they sound a warning about the possible existence of cryptic species. These are likely to be most common on topographically complex continental margins (Etter et al. 1999, 2005).
Conclusions and future perspectives Much of the foraminiferal research undertaken by geologists has focused on bathyal continental margins where calcareous and other hard-shelled species with a high fossilization potential are
113
common. Recent advances in our knowledge of the trophic biology and ecology of these species is important for understanding their responses to organic fluxes to the seafloor and the generation of the stable carbon isotope signal in calcareous shells. Many of the common shallow- and intermediate-infaunal species have diets consisting, at least in part, of labile organic matter (phytodetritus) derived from primary production. At greater depths on abyssal plains, delicate, softbodied taxa often dominate foraminiferal assemblages. We speculate that they feed on more degraded organic matter and/or bacteria, and have slower metabolic and growth rates than species that consume phytodetritus. Although important for biologists, these delicate, often monothalamous forms have little fossilization potential and are of less interest to geologists. However, they have considerable phylogenetic importance. Also, because they are abundant over a vast area of abyssal seafloor, they probably play an important global role in organic matter cycling at the sediment surface and may need to be considered when formulating palaeoceanographic proxies for the flux of organic matter to the seafloor. In an excellent review, Jorissen et al. (2007) highlight the many problems that frustrate the refinement of palaeoceanographic proxies based on foraminifera. These include a lack of very large datasets of species distributions in relation to environmental variables and an incomplete understanding of taphonomic processes and fluxes of organic matter through the water column, as well as an imprecise knowledge of foraminiferal ecology. Biologists would also like to know more about deep-sea foraminifera and, although their aims and approaches often differ, they share with geologists some common interests in the ecology of these protists. The following are some examples of approaches that may help to advance our general knowledge of bathyal and abyssal foraminiferal biology, to the benefit of both disciplines.
New experimental techniques Recent laboratory and in-situ experiments, reviewed above, have provided valuable insights into the ecology of deep-sea foraminifera including their trophic preferences. However, except for preliminary studies by Turley et al. (1992) and Ohga (1995), all laboratory-based experiments have been conducted at 1 atm on samples originating from water depths less than about 2200 m (Hemleben & Kitazato 1995). The use of large pressurized chambers, such as the IPOCAMP vessels, would enable more sophisticated and replicated experiments to be carried out on deep-sea samples under ambient pressures. At the same
114
A. J. GOODAY ET AL.
time, the increasing availability and use of Remote Operated Vehicles (ROVs) in the deep sea is providing new possibilities for in-situ experimentation on the seafloor. Another approach is to test novel tracer methods using shallow-water foraminiferal species which are easy to collect and culture. Once new methods are perfected under optimal conditions, they can be applied in the more challenging deep-sea environment.
In-situ monitoring using cabled seafloor observatories The current development of seafloor observatories connected to laboratories on land by means of an underwater cable, or via direct satellite telemetry using a surface buoy, offer exciting possibilities for real time monitoring of the seafloor environment. Such systems have already been established off the coasts of North America (e.g. NEPTUNE) and Japan (ARENA) (Asakawa et al. 2004) and there are plans to develop similar systems around the European margin under the ESONET initiative (Priede et al. 2004). The incorporation of planar optodes into observatories in order to measure the small-scale temporal and spatial variations in oxygen concentrations in the surface sediments will be of particularly interest for understanding the relationship between foraminiferal species and oxygen (Glud et al. 2005; Oguri et al. 2006).
Studies of the cell body The focus of most foraminiferal research has been on the test rather than on the cell body. Yet it is the cell body, and particularly the system of reticulopodia, that provide the defining characteristic of foraminifera (Bowser & Travis 2002), responsible for undertaking basic life functions such as locomotion, the collection of particles for food, test construction and cyst formation, and in metabolic exchange (Travis & Bowser 1991). A comparison of pseudopodial morphology and activity in ‘advanced’ multilocular foraminifera and ‘primitive’ monothalamous foraminifera may improve understanding of the function of these different groups in deep-sea biological communities, including their respective roles in carbon cycling. Transmission electron microscopic studies of bathyal calcareous foraminifera have yielded useful information about ingested food particles (Heeger 1990; Goldstein & Corliss 1994). However, these studies have been limited and there is very little equivalent information about other kinds of deep-sea foraminifera. A detailed ultrastructural examination of xenophyophores and
komokiaceans could provide crucial insights into their trophic biology.
Molecular genetic studies Despite recent advances, the study of the molecular genetics of deep-sea foraminifera remains in its infancy. Further work may help to clarify a number of issues that have potential relevance in palaeoceanography, notably possible genetic differentiation within morphospecies, the existence of genetically distinct cryptic species, and phylogenetic relationships between species. We thank Dr Bill Austin for inviting one of us (AJG) to present a paper at the Geological Society of London, Marine Studies Group meeting on Biological and Biogeochemical Controls on Palaeoceanographic Proxies. We are grateful to Mrs Kate Davis for her help with the figures and to Drs. Naohiko Ohkouchi (JAMSTEC), Olga Kamenskaya (Moscow), Kate Larkin and Mike Zubkov (Southampton) for useful discussions and an anonymous review for comments that helped to considerably improve the manuscript.
References A LTENBACH , A. V., P FLAUMANN , U., S CHIEBEL , R., T HIES , A., T IMM , S. & T RAUTH , M. 1999. Scaling percentages and distributional patterns of benthic foraminifera with flux rates of organic carbon. Journal of Foraminiferal Research, 29, 173– 185. A NTIA , A. N., K OEVE , W., F ISHER , G. ET AL . 2001. Basin-wide particulate carbon flux in the Atlantic Ocean: regional export patterns and potential for atmosphaeric CO2 sequestration. Global Biogeochemical Cycles, 15, 845– 862. A SAKAWA , K., M IKADA , H., K AWAGUCHI , K., S HIRASAKI , Y., K OJIMA , J., M URAMATSU , J. & H ORIUCHI , Y. 2004. ARENA: A New Scientific Cable-network for Real-time and Long-term Underwater Observation. Proceedings of the Fourteenth International Offshore and Polar Engineering Conference, Toulon, France, May 23– 28, 2004, 363–370. B ERNHARD , J. M. & B OWSER , S. S. 1992. Bacterial biofilms as a trophic resource for certain foraminifera. Marine Ecology Progress Series, 83, 263 –272. B ERNSTEIN , B. B., H ESSLER , R. R., S MITH , R. & J UMARS , P. A. 1978. Spatial dispersion of benthic Foraminifera in the abyssal central North Pacific. Limnology and Oceanography, 23, 401 –416. B ERTRAM , M. A. & C OWAN , J. P. 1999. Temporal variations in the deep-water colonization rates of small benthic foraminifera: the results of an experiment on Cross Seamount. Deep-Sea Research I, 46, 1021– 1049. B OWSER , S. S. & T RAVIS , J. L. 2002. Reticulopodia: structural and behavioural basis for the suprageneric placement of granuloreticulosan protist. Journal of Foraminiferal Research, 32, 440–447.
DEEP-SEA BENTHIC FORAMINIFERA C ARALP , M. H. 1989. Abundance of Bulimina exilis and Melonis barleeanum: relationship to the quality of marine organic matter. Geo-Marine Letters, 9, 37– 43. C ARTWRIGHT , N. G., G OODAY , A. J. & J ONES , A. R. 1989. The morphology, internal organization, and taxonomic position of Rhizammina algaeformis Brady, a large, agglutinated, deep-sea foraminifer. Journal of Foraminiferal Research, 19, 115– 125. C EDHAGEN , T. & M ATTSON , S. 1991. Globopelorhiza sublittoralis gen. et sp. n., a komokiacean (Protozoa: Foraminiferida) from the Scandinavian sublittoral. Sarsia, 76, 209– 213. C IFELLI , R. 1990. A history of the classification of Foraminifera (1826–1933). Part I. Foraminiferal classification from d’Orbigny to Galloway. Cushman Foundation for Foraminiferal Research, Special Publication, 27, 1–88. C OFFIN , R. B., C IFUENTES , L. A. & E LDRIDGE , P. M. 1994. The use of stable isotopes to study microbial processes in estuaries. In: L AJTHA , K. & M ICHENER , R. H. (eds) Stable Isotopes in Ecology and Environmental Sciences. Blackwell Scientific Publications, 222–240 C ORNELIUS , N. & G OODAY , A. J. 2004. ‘Live’ (stained) deep-sea benthic foraminifera in the western Weddell Sea: trends in abundance, diversity and taxonomic composition in relation to water depth. Deep-Sea Research II, 51, 1571– 1602. D RAZEN , J. C., B ALDWIN , R. J. & S MITH , K. L. 1998. Sediment community response to a temporally varying food supply at an abyssal station in the NE Pacific. Deep-Sea Research II, 45, 893–913. E TTER , R. J., R EX , M. A., C HASE , M. C. & Q UATTRO , J. M. 1999. A genetic dimension to deep-sea biodiversity. Deep-Sea Research I, 46, 1095–1099. E TTER , R. J., R EX , M. A., C HASE , M. R. & Q UATTRO , J. M. 2005. Population differentiation decreases with depth in deep-sea bivalves. Evolution, 59, 1479–1491. E RTAN , K. T., H EMLEBEN , V. & H EMLEBEN , C. 2004. Molecular evolution of some selected benthic foraminifera as inferred from sequences of the small subunit ribosomal DNA. Marine Micropaleontology, 53, 367–388. F ONTANIER , C., J ORISSEN , F. J., C HAILLOU , G., D AVID , C., A NSCHUTZ , P. & L AFON , V. 2003. Seasonal and interannual variability of benthic foraminiferal faunas at 550 m depth in the Bay of Biscay. Deep-Sea Research I, 50, 457– 494. F ONTANIER , C., J ORISSEN , F. J., C HAILLOU , G., A NSCHUTZ , P., G RE´ MARE , A. & G RIVEAUD , C. 2005. Live foraminiferal faunas from a 2800 m deep lower canyon station from the Bay of Biscay: faunal response to focusing of refractory organic matter. Deep-Sea Research I, 52, 1189–1227. G OLDSTEIN , S. T. 1999. Foraminifera: a biological overview. In: S EN G UPTA , B. K. (ed.) Modern Foraminifera. Kluwer Academic Publishers, Dordrecht, 37– 55. G OLDSTEIN , S. T. & C ORLISS , B. H. 1994. Deposit feeding in selected deep-sea and shallow-water benthic foraminifera. Deep-Sea Research, 41, 229–241. G OODAY , A. J. 1988. A response by benthic Foraminifera to the deposition of phytodetritus in the deep sea. Nature, 332, 70–73.
115
G OODAY , A. J. 1990a. Tinogullmia riemanni sp. nov. (Allogromiina; Foraminiferida), a new species associated with organic detritus in the deep sea. Bulletin British Museum Natural History, (Zoology), 56, 93–103. G OODAY , A. J. 1990b. Recent deep-sea agglutinated Foraminifera: a brief review. In: H EMLEBEN , C., K AMINSKI , M. A., K UHNT , W. & S COTT , D. B. (eds) Paleoecology, Biostratigraphy, Paleoceanography and Taxonomy of Agglutinated Foraminifera. NATO ASI Series C: Mathematical and Physical Sciences, Vol. 327. Kluwer Academic Publishers, 271– 304. G OODAY , A. J. 1993. Deep-sea benthic foraminiferal species which exploit phytodetritus: characteristic features and controls on distribution. Marine Micropaleontology, 22, 187– 205. G OODAY , A. J. 1996. Epifaunal and shallow infaunal foraminiferal communities at three abyssal NE Atlantic sites subject to differing phytodetritus input regimes. Deep-Sea Research I, 43, 1395– 1421. G OODAY , A. J. 2002. Organic-walled allogromiids: aspects of their occurrence, diversity and ecology in marine habitats. Journal of Foraminiferal Research, 32, 384– 399. G OODAY , A. J. 2003. Benthic foraminifera (Protista) as tools in deep-water palaeoceanography: a review of environmental influences on faunal characteristics. Advances in Marine Biology, 46, 1–90. G OODAY , A. J. & L AMBSHEAD , P. J. D. 1989. The influence of seasonally deposited phytodetritus on benthic foraminiferal populations in the bathyal northeast Atlantic: the species response. Marine Ecology – Progress Series, 58, 53– 67. G OODAY , A. J. & T URLEY , C. M. 1990. Responses by benthic organisms to inputs of organic material to the ocean floor: a review. Philosophical Transactions of the Royal Society of London, A331, 119– 138. G OODAY , A. J. & R ATHBURN , A. E. 1999. Temporal variability in living deep-sea benthic foraminifera. Earth Science Reviews, 46, 187–212. G OODAY , A. J. & H UGHES , J. A. 2002. Foraminifera associated with phytodetritus deposits at a bathyal site in the northern Rockall Trough (NE Atlantic): seasonal contrasts and a comparison of stained and dead assemblages. Marine Micropaleontology, 46, 83–110. G OODAY , A. J. & P AWLOWSKI , J. 2004. Conqueria laevis gen. and sp. nov., a new soft-walled, monothalamous foraminiferan from the deep Weddell Sea. Journal of the Marine Biological Association of the United Kingdom, 84, 919 –924. G OODAY , A. J., L EVIN , L. A., T HOMAS , C. L. & H ECKER , B. 1992a. The distribution and ecology of Bathysiphon filiformis and B. major de Folin (Protista, Foraminiferida) on the continental slope off North Carolina. Journal of Foraminiferal Research, 22, 129– 146. G OODAY , A. J., L EVIN , L. A., L INKE , P. & H EEGER , T. 1992b. The role of benthic foraminifera in deep-sea food webs and carbon cycling. In: R OWE , G. T. & P ARIENTE , V. (eds) Deep-Sea Food Chains and the Global Carbon Cycle. Kluwer Academic Publishers, The Netherlands, 63–91.
116
A. J. GOODAY ET AL.
G OODAY , A. J., S HIRES , R. & J ONES , A. R. 1997. Large, deep-sea agglutinated foraminifera: two differing kinds of organization and their possible ecological significance. Journal of Foraminiferal Research, 27, 278– 291. G OODAY , A. J., B ETT , B. J., S HIRES , R. & L AMBSHEAD , P. J. D. 1998. Deep-sea benthic foraminiferal diversity in the NE Atlantic and NW Arabian sea: a synthesis. Deep-Sea Research II, 45, 165– 201. G OODAY , A. J., H UGHES , J. A. & L EVIN , L. A. 2001. The foraminiferan macrofauna from three North Carolina (U.S.A.) slope sites with contrasting carbon flux: a comparison with the metazoan macrofauna. Deep-Sea Research I, 48, 1709–1739. G OODAY , A. J., P OND , D. W. & B OWSER , S. S. 2002. Ecology and nutrition of the large agglutinated foraminiferan Bathysiphon capillare in the bathyal NE Atlantic: distribution within the sediment profile and lipid biomarker composition. Marine Ecology Progress Series, 245, 69–82. G OODAY , A. J., H OLZMANN , M., G UIARD , J., C ORNELIUS , N. & P AWLOWSKI , J. 2004a. A new monothalamous foraminiferan from 1000 to 6300 m water depth in the Weddell Sea: morphological and molecular characterisation. Deep-Sea Research II, 51, 1603–1616. G OODAY , A. J., H ORI , S., T ODO , Y., O KAMOTO , T., K ITAZATO , H. & S ABBATINI , A. 2004b. Soft-walled, monothalamous benthic foraminiferans in the Pacific, Indian and Atlantic Oceans: aspects of biodiversity and biogeography. Deep-Sea Research I, 51, 33– 53. G OODAY , A. J., C EDHAGEN , T., K AMENSKAYA , O. E. & C ORNELIUS , C. 2007a. The biodiversity and biogeography of komokiaceans and other enigmatic foraminiferan-like protists in the deep Southern Ocean. Deep-Sea Research II, 54, 1691– 1719. G OODAY , A. J., K AMENSKAYA , O. E. & C EDHAGEN , T. 2007b. New and little-known Komokiacea (Foraminifera) from the Weddell Sea and adjacent areas of the Southern Ocean. Zoological Journal of the Linnean Society. G OODAY , A. J., T ODO , Y., U EMATSU , K. & K ITAZATO , H. (2007c). New organic-walled Foraminifera (Protista) from the ocean’s deepest point, the Challenger Deep (western Pacific Ocean). Zoological Journal of the Linnean Society. G LUD , R. N., W ENZHO¨ FER , F., T ENGBERG , A., M IDDELBOE , M., O GURI , K. & K ITAZATO , H. 2005. Distribution of oxygen in surface sediments from central Sagami Bay, Japan: In situ measurements by microelectrodes and planar optodes. Deep-Sea Research I, 52, 1974– 1987. G RIMM , G. W., S TO¨ GERER , K., E RTAN , K. T., K ITAZATO , H., K UCˇ ERA , M., H EMLEBEN , V. & H EMLEBEN , C. 2007. Diversity of rDNA in Chilostomella: molecular differentiation patterns and putative hermit types. Marine Micropaleontology, 62, 75–90. G UICHARD , S., J ORISSEN , F. & P EYPOUQUET , J.-P. 1999. Late Quaternary benthic foraminiferal records testifying lateral variability of the Cap Blanc upwelling signal. C.R. Acade´mie des sciences, Paris, Sciences de la terre et des planets, 329, 295 –301. H AYWARD , B. W., H OLZMANN , M., G RENFELL , H. R., P AWLOWSKI , J. & T RIGGS , C. M. 2004.
Morphological distinction of molecular types in Ammonia – towards a taxonomic revision of the world’s most commonly misidentified foraminifera. Marine Micropaleontology, 50, 237–271. H EEGER , T. 1990. Elektronmikrokopische Untersuchungen zur Erna¨hrungsbiologie benthischer Foraminiferen. Berichte aus dem Sondersforschungsbereich 313, 31, 1 –139. H EINZ , P., S CHMIEDL , G., K ITAZATO , H. & H EMLEBEN , C. 2001. Response of deep-sea benthic foraminifera from the Mediterranean Sea to simulated phytoplankton pulses under laboratory conditions. Journal of Foraminiferal Research, 31, 210–227. H EINZ , P., H EMLEBEN , C. & K ITAZATO , H. 2002. Timeresponse of cultured deep-sea benthic foraminifera to different algal diets. Deep-Sea Research I, 49, 517–537. H EIP , C. H. R., D UINEVELD , G., F LACH , E. ET AL . 2001. The role of the benthic biota in sedimentary metabolism and sediment-water exchange processes in the Goban Spur area (NE Atlantic). Deep-Sea Research II, 48, 3223–3243. H ERB , R. 1971. Distribution of Recent benthonic foraminifera in the Drake Passage. In: L LANO , G. A. & W ALLEN , I. E. (eds) Biology of Antarctic Seas IV. Antarctic Research Series, 17. American Geophysical Union, Washington DC, 251– 300. H EMLEBEN , C. & K ITAZATO , H. 1995. Deep-sea foraminifera under long time observation in the laboratory. Deep-Sea Research, 42, 827– 832. H ERGUERA , J. C. & B ERGER , W. H. 1991. Paleoproductivity from benthic foraminifera abundance: glacial to postglacial change in the west-equatorial Pacific. Geology, 19, 1173–1176. H ERGUERA , J. C. 2000. Late glacial paleoproductivity patterns in the eastern Equatorial Pacific: benthic foraminiferal records. Marine Micropaleontology, 40, 259–275. H OPWOOD , J. D., M ANN , S. & G OODAY , A. J. 1997. The crystallography and possible origin of barium sulphate in deep-sea rhizopod protists (class Xenophyophorea). Journal of the marine biological Association of the United Kingdom, 77, 969 –987. H UGHES , J. A. & G OODAY , A. J. 2004. The influence of dead Syringammina fragilissima (Xenophyophorea) tests on the distribution of benthic foraminifera in the Darwin Mounds region (NE Atlantic). Deep-Sea Research I, 51, 1741–1758. J AHNKE , R. A. 1996. The global ocean flux of particulate organic carbon: areal distribution and magnitude. Global Biogeochemical Cycles, 10, 71–88. J AHNKE , R. A., E MERSON , S. R., R EIMERS , C. E., S CHUFFERT , J., R UTTENBERG , K. & A RCHER , D. 1989. Benthic recycling of biogenic debris in the eastern tropical Atlantic Ocean. Geochimica et Cosmochimica Acta, 53, 2947– 2960. J ONES , R. W. & C HARNOCK , M. A. 1985. ‘Morphogroups’ of agglutinated foraminifera. Their life positions and feeding habits and potential applicability in (paleo)ecological studies. Revue de Pale´obiologie, 4, 311–320. J ORISSEN , F. J., W ITTLING , I., P EYPOUQUET , J. P., R ABOUILLE , C. & R ELEXANS , J. C. 1998. Live benthic foraminiferal faunas off Cap Blanc, NW
DEEP-SEA BENTHIC FORAMINIFERA Africa: community structure and microhabitats. Deep-Sea Research I, 45, 2157–2188. J ORISSEN , F. J., F ONTANIER , C. & T HOMAS , E. 2007. Paleoceanographical proxies based on deep-sea benthic foraminiferal assemblage characteristics. In: H ILLAIRE -M ARCEL , C. & D E V ERNAL , A. (eds) Proxies in Late Cenozoic Paleoceanography. Volume 1 - Methods in Late Cenozoic Paleoceanography IIIBiological Tracers and Biomarkers. Elsevier, pp. 263–325. K AMENSKAYA , O. E. 1987. Xenophyophoria and Komokiacea in trophic chains of the deep-sea benthos. In: Feeding of Marine Invertebrates and its Significance in Formation of Communities. Academy of Sciences of the USSR. P. P. Shirshov Institute of Oceanology, Moscow, 15– 22. K AMENSKAYA , O. E. 1988. Quantitative distribution of Komokiacea and Xenophyophora in the Southern Atlantic. In: K UZNETSOV , A. P. & S OKOLOVA , M. N. (eds) Structural and Functional Researches of the Marine Benthos. Academy of Sciences of the USSR. P. P. Shirshov Institute of Oceanology, Moscow, 15– 20. K AMENSKAYA , O. E., B AGIROV , N. E. & S IMDIANOV , T. G. 2002. Luffammina atlantica gen. et sp. nov. (Foraminifera, Arboramminidae) from the hydrothermal area Rainbow (Mid-Atlantic Ridge). In: K UZNETSOV , A. P. & Z EZINA , O. N. (eds) Adapation Aspects of Evolution of Marine Fauna. VNIRO Publishing House, Moscow, 144–152. K AMINSKI , M. A. & G RADSTEIN , F. M. 2005. Atlas of Paleogene Cosmopolitan Deep-water Agglutinated Foraminifera. Grzybowski Foundation Special Publication, 10, 547. K ATO , Y., N AKAO , K. & I SOZAKI , Y. 2002. Geochemistry of Late Permian to Early Triassic pelagic cherts from southwestern Japan: implications for an oceanic redox change. Chemical Geology, 182, 15–34. K ELLEY , C. A., C OFFIN , R. B. & C IFUENTES , L. A. 1998. Stable isotope evidence for alternative bacterial carbon sources in the Gulf of Mexico. Limnology and Oceanography, 43, 1962– 1969. K RO¨ NCKE , I., V ANREUSEL , A., V INCX , M., W OLLENBURG , J., M ACKENSEN , A., L IEBEZEIT , G. & B EHRENDS , B. 2000. Different benthic sizecompartments and their relationship to sediment chemistry in the deep Eurasian Arctic Ocean. Marine Ecology Progress Series, 199, 31– 41. K UHNT , W. & C OLLINS , W. 1995. Fragile abyssal foraminifera from the northwestern Sargasso Sea: distribution, ecology and paleoceanographic significance. In: K AMINSKI , M. A., G EROCH , S. & G ASINSKI , M. A. (eds) Proceedings of the Fourth International Workshop on Agglutinated Foraminifera. Grzybowski Foundation Special Publication, 3, 159–172. K UHNT , W. & M OULLADE , M. 1991. Quantitative analysis of Upper Cretaceous abyssal agglutinated foraminiferal distribution in the North Atlantic – paleoceanographic implications. Revue de Micropale´ontologie, 34, 313– 349. L ANGEZAAL , A. M., J ANNINK , N. T., P IERSON , E. S. & VAN DER Z WAAN , G. J. 2005. Foraminiferal selectivity towards bacteria: an experimental approach using
117
cell-permeant stain. Journal of Sea Research, 54, 256– 275. L AUREILLARD , J., M E´ JANELLE , L. & S IBUET , M. 2004. Use of lipids to study the trophic ecology of deep-sea xenophyophores. Marine Ecology Progress Series, 270, 129–140. L AWS , E. A., F ALKOWSKI , P. G., S MITH J R , W. O. & M C C ARTHY , J. J. 2000. Temperature effects on export production in the open ocean. Global Biogeochemical Cycles, 14, 1231–1246. L EE , J. J. 1974. Towards understanding the niche of foraminifera. In: H EDLEY , R. H. & A DAMS , G. (eds) Foraminifera. Academic Press, New York, 1, 208– 257. L EE , J. J. 1980. Nutrition and physiology of the Foraminifera. In: L EVANDOWSKI , M. & H UTNER , S. (eds) Biochemistry and Physiology of Protozoa, Vol. 3. Academic Press, London, 43– 66. L EE , J. J. & P IERCE , S. 1963. Growth and physiology of foraminifera in the laboratory: Part 4 – monoalgal culture of an allogromiid with notes on its morphology. Journal of Protozoology, 10, 404–411. L EE , J. J. & M ULLER , W. A. 1973. Trophic dynamics and niches of salt marsh foraminifera. American Zoologist, 15, 215– 223. L EE , J. J., M C E NERY , M., P IERCE , S., F REUDENTHAL , H. D. & M ULLER , W. A. 1966. Tracer experiments in feeding littoral foraminifera. Journal of Protozoology, 13, 659–670. L EVIN , L. A. & T HOMAS , C. L. 1988. The ecology of xenophyophores (Protista) on eastern Pacific seamounts. Deep-Sea Research, 35, 2003– 2027. L EVIN , L. A., B LAIR , N., D E M ASTER , D., P LAIA , G., F ORNES , W., M ARTIN , C. & T HOMAS , C. 1997. Rapid subduction of organic matter by maldanid polychaetes on the North Carolina slope. Journal of Marine Research, 55, 595– 611. L EVIN , L. A., B LAIR , N. E., M ARTIN , C. M., D E M ASTER , D. J., P LAIA , G. & T HOMAS , C. J. 1999. Macrofaunal processing of phytodetritus at two sites on the Carolina margin: in situ experiments using 13C-labeled diatoms. Marine Ecology Progress Series, 182, 37– 54. L IPPS , J. H. 1983. Biotic interactions in benthic foraminifera. In: T EVESZ , M. J. S. & M C C ALL , P. L. (eds) Biotic Interactions in Recent and Fossil Benthic Communities. Plenum Publishing Corporation, 331– 376. L IPPS , J. H. & V ALENTINE , J. W. 1970. The role of foraminifera in the trophic structure of marine communities. Lethaia, 3, 279–286. L OCHTE , K., A NDERSON , R., F RANCOIS , R., J AHNKE , R. A., S HIMMIELD , G. & V ETROV , A. 2002. Benthic processes and the burial of carbon. In: F ASHAM , M. J. R. (ed.) Ocean Biogeochemistry. The Role of the Ocean Carbon Cycle in Global Change. Springer, Berlin, 195– 216. L OEBLICH , A. J. R. & T APPAN , H. 1987. Foraminiferal Genera and their Classification, Vols 1– 2. Van. Nostrand Reinhold, N.Y. M ACKENSEN , A. & B ICKERT , T. 1999. Stable carbon isotopes in benthic foraminifera. In: F ISCHER , G. & W EFER , G. (eds) Use of Proxies in Paleoceanography. Springer, Berlin, 229– 254.
118
A. J. GOODAY ET AL.
M ACKENSEN , A., G ROBE , H., K UHN , G. & F U¨ TTERER , D. K. 1990. Benthic foraminiferal assemblages from the eastern Weddell Sea between 68 and 738S: distribution, ecology and fossilization potential. Marine Micropaleontology, 16, 241– 283. M ACKENSEN , A., H UBBERTEN ,, B ICKERT , T., F ISCHER , G. & F U¨ TTERER , D. K. 1993. The d13C in benthic foraminiferal tests of Fontbotia wuellerstorfi (Schwager) relative to the d13C of dissolved inorganic carbon in Southern Ocean deep water: implications for glacial ocean circulation models. Paleoceanography, 8, 587– 610. M ACKENSEN , A., S CHMIEDL , G., H ARLOFF , J. & G IESE , M. 1995. Deep-sea foraminifera in the South Atlantic Ocean: Ecology and assemblage generation. Micropaleonotology, 41, 342– 358. M C C ORKLE , D. C., K EIGWIN , L. D., C ORLISS , B. H. & E MERSON , S. R. 1990. The influence of microhabitats on the carbon isotopic composition of deep-sea benthic foraminifera. Paleoceanography, 5, 161–185. M IDDLEBURG , J. J., B ARRANGUET , C., B OSCHKER , H. T. S., H ERMAN , P. M. J., M OENS , T. & H EIP , C. H. R. 2000. The fate of intertidal microphytobenthos carbon: an in situ 13C-Labeling Study. Limnology and Oceanography, 45, 1224–1234. M ILLER , W. III 2005. Giant Bathysiphon (Astrorhizina: Foraminifera) from the Late Cretaceous Hunter Cove Formation, southwestern Oregon. Journal of Paleontology, 79, 389– 394. M OODLEY , L., B OSCHKER , H. T. S., M IDDELBURG , J. J., P EL , R., H ERMAN , P. M. J., DE D ECKERE , E. & H EIP , C. H. R. 2000. Ecological significance of benthic foraminifera: 13C labeling experiments. Marine Ecology Progress Series, 202, 289–295. M OODLEY , L., M IDDELBURG , J. J., B OSCHKER , H. T. S., D UINEVELD , G. C. A., P EL , R., H ERMAN , P. M. J. & H EIP , C. H. R. 2002. Bacteria and Foraminifera: key players in a short-term deep-sea benthic response to phytodetritus. Marine Ecology Progress Series, 182, 37–54. M OODLEY , L., M IDDELBURG , J. J., S OETAERT , K., B OSCHKER , H. T. S., H ERMAN , P. M. J. & H EIP , C. H. R. 2005. Similar rapid response to phytodetritus deposition in shallow and deep-sea sediments. Journal of Marine Research, 63, 457–469. M ULLER , W. A. & L EE , J. J. 1969. Apparent indispensability of bacteria in foraminiferan nutrition. Journal of Protozoology, 16, 471– 478. M ULLINEAUX , L. S. 1987. Organisms living on manganese nodules and crusts: distribution and abundance at three North Pacific sites. Deep-Sea Research, 34, 165– 184. M URRAY , J. W. 1991. Ecology and Palaeoecology of Benthic Foraminifera. Longman Scientific & Technical. M URRAY , J. W. 2006. Ecology and Applications of Benthic Foraminifera. Cambridge University Press. M URRAY , J. W. & A LVE , E. 1999. Taphonomic experiments on marginal marine foraminiferal assemblages: how much ecological information is preserved? Palaeogeography, Palaeoclimatology, Palaeoecology, 149, 183– 197. M URRAY , J. W. & A LVE , E. 2001. Do calcareous dominated shelf foraminiferal assemblages leave
worthwhile ecological information after their dissolution? In: H ART , M. B., K AMINSKI , M. A. & S MART , C. W. (eds) Proceedings of the Fifth International Workshop on Agglutinated Foraminifera, (Plymouth, U. K., September 6– 16, 1997). Grzybowski Foundation Special Publication, 7, 311–331. N AIDU , P. D. & M ALMGREN , B. A. 1995. Do benthic foraminifer records represent a productivity index in oxygen minimum zone areas? An evaluation from the Oman Margin, Arabian Sea. Marine Micropaleontology, 26, 49– 55. N EAGU , T. 1964. Large size agglutinated foraminifera from the Carpathian of Rumania. Rocznik Polskiego Towarzystwa Geologicznego, 34, 579–588. N ESTELL , G. P. & T OLMACHEVA , T. Y. 2004. Early Ordovician foraminifers from the Lava River Section, northwestern Russia. Micropaleontology, 50, 253–280. N OMAKI , H. 2005. The fate and degradation process of phytodetritus by benthic communities: in situ 13 C-tracer experiments. PhD Thesis, University of Tokyo. N OMAKI , H., H EINZ , P., N AKATSUKA , T., S HIMANAGA , M. & K ITAZATO , H. 2005a. Species-specific ingestion of organic carbon by deep-sea benthic foraminifera and meiobenthos: in situ tracer experiments. Limnology and Oceanography, 50, 134 –146. N OMAKI , H., H EINZ , P., H EMLEBEN , C. & K ITAZATO , H. 2005b. Behaviour and response of deep-sea foraminifera to freshly supplied organic matter: a feeding experiment in micrososm environments. Journal of Foraminiferal Research, 35, 103–113. N OMAKI , H., H EINZ , P., N AKATSUKA , T. ET AL . 2006. Different ingestion patterns of 13C labelled bacteria and algae by deep-sea benthic foraminifera. Marine Ecology Progress Series, 310, 95–108. N OZAWA , F. 2005. Deep-sea benthic foraminiferal fauna from the equatorial Pacific Ocean. MSc Thesis, Graduate School of Science, Chiba University [in Japanese]. N OZAWA , F., K ITAZATO , H., T SUCHIYA , M. & G OODAY , A. J. 2006. ‘Live’ benthic foraminifera at an abyssal site in the equatorial Pacific nodule province: abundance, diversity and taxonomic composition. Deep-Sea Research I, 53, 1406– 1422. O HGA , T. 1995. Seasonal changes in the population ecology of deep-sea benthic foraminifera in Sagami Bay. Unpublished MSc thesis, Institute of Geosciences, Shizuoka University, Japan. O HGA , T. & K ITAZATO , H. 1997. Seasonal changes in bathyal foraminiferal populations in response to the flux of organic matter (Sagami Bay, Japan). Terra Nova, 9, 33– 37. O GURI , K., K ITAZATO , H. & G LUD , R. N. 2006. Platinum octaetylporphyrin based planar optodes combined with an UV-LED excitation light source: an ideal tool for high-resolution O2 depleted environments. Marine Chemistry, 100, 95– 106. P AWLOWSKI , J., H OLZMANN , M., B ERNEY , C. ET AL . 2003a. The evolution of early foraminifera. Proceedings of the National Academy of Sciences, 100, 11494–11498.
DEEP-SEA BENTHIC FORAMINIFERA P AWLOWSKI , J., H OLZMANN , M., F AHRNI , J. & R ICHARDSON , S. L. 2003b. Small subunit ribosomal DNA suggests that the xenophyophoran Syringammina corbicula is a foraminiferan. Journal of Eukaryotic Microbiology, 50, 483–487. P AWLOWSKI , J., F AHRNI , J., L ECROQ , B. ET AL . 2007. Bipolar gene flow in deep-sea benthic Foraminifera. Molecular Ecology, 16, 4089– 4096. P ORTER , S. M. & K NOLL , A. H. 2000. Testate amoebae in the Neoproterozoic Era: evidence from vase-shaped microfossils in the Chuar Group, Grand Canyon. Paleobiology, 26, 360– 385. P RIEDE , I. G., T HOMPSEN , L. ET AL . 2004. ESONET – European Sea Floor Observatory Network. Proceedings of the Oceans ’04 Conference, Kobe, Japan 9–12 November 2004, 2155–2163. R OMANKEVICH , E. A., V ETROV , A. A. & K ORNEEVA , G. A. 1999. Geochemistry of organic carbon in the ocean. In: G RAY , J. S., A MBROSE , W. Jr. & S ZANIAWSKA , A. (eds) Biogeochemical Cycling and Sediment Ecology. NATO ASI Series, Kluwer Academic Publishers, Dordrecht, Boston, London, 1–27. S AIDOVA , K. M. 1967. Sediment stratigraphy and paleogeography of the Pacific Ocean by benthonic foraminifera during the Quaternary. Progress in Oceanography, 4, 143–151. S CHMIEDL , G., P FEILSTICKER , M., H EMLEBEN , C. & M ACKENSEN , A. 2004. Environmental and biological effects on the stable isotope composition of recent deep-sea benthic foraminifera from the western Mediterranean Sea. Marine Micropaleontology, 51, 129–152. S CHRO¨ DER , C. J. 1986. Deep-water arenaceous foraminifera in the northwest Atlantic Ocean. Canadian Technical Report of Hydrography and Ocean Sciences, 71, 1–191. S CHRO¨ DER , C. J., S COTT , D. B. & M EDIOLI , F. S. 1989. Fragile abyssal foraminifera (including new Komokiacea) for the Nares Abyssal Plain. Micropaleontology, 35, 10– 48. S CHWEIZER , M., P AWLOWSKI , J., D UIJNSTEE , I. A. P., K OUWENHOVEN , T. J. & VAN DER Z WAAN , G. J. 2005. Molecular phylogeny of the foraminiferal genus Uvigerina based on ribosomal DNA sequences. Marine Micropaleontology, 57, 51– 67 S HIRES , R. 1994. The taxonomy, morphology and ecology of novel deep-sea agglutinated foraminifers in the northeast Atlantic. PhD Thesis, University of Reading. S HIRES , R., G OODAY , A. J. & J ONES , A. R. 1994. A new large agglutinated foraminifer (Arboramminidae n. fam.) from a oligotrophic site in the abyssal
119
northeast Atlantic. Journal of Foraminiferal Research, 24, 149– 157. S NIDER , L. J., B URNETT , B. R. & H ESSLER , R. R. 1984. The composition and distribution of meiofauna and nanobiota in a central North Pacific deep-sea area. Deep-Sea Research, 31, 1225– 1249. S OKOLOVA , M. N. 2000. Feeding and Trophic Structure of the Deep-Sea Macrobenthos. Smithsonian Institution Libraries, Washington, D. C., 264 pp. S UHR , S. B., P OND , D. W., G OODAY , A. J. & S MITH , C. R. 2003. Selective feeding by foraminifera on phytodetritus on the western Antarctic Peninsula shelf: evidence from fatty acid biomarker analysis. Marine Ecology Progress Series, 262, 153– 162. T APPAN , H. & L OEBLICH , A. R. 1988. Foraminiferal evolution, diversification and extinction. Journal of Paleontology, 62, 695–714. T ENDAL , O. S. 1979. Aspects of the biology of Komokiacea and Xenophyophoria. Sarsia, 64, 13–17. T ENDAL , O. S. & H ESSLER , R. R. 1977. An introduction to the biology and systematics of Komokiacea. Galathea Report, 14, 165–194. T ODO , Y., K ITAZATO , H., H ASHIMOTO , J. & G OODAY , A. J. 2005. Simple foraminifera flourish at the ocean’s deepest point. Science, 307, 689. T RAVIS , J. L. & B OWSER , S. S. 1991. Motility of foraminifera. In: L EE , J. J. & A NDERSON , O. R. (eds) Biology of Foraminifera. Academic Press, New York, 91–155. T URLEY , C. M., L OCHTE , K. & P ATTERSON , D. J. 1988. A barophilic flagellate isolated in the mid-North Atlantic. Deep-Sea Research, 35, 1079– 1092. T URLEY , C. M., G OODAY , A. J. & G REEN , J. C. 1992. Maintenance of abyssal Foraminifera under high pressure and low temperature: some preliminary results and their biological and palaeoceanographical implications. Deep-Sea Research, 40, 643– 652. W ALSH , J. J. 1991. Importance of the continental margins in the marine biogeochemical cycling of carbon and nitrogen. Nature, 350, 53–55. W ITTE , U., W EINHO¨ FER , F., S OMMER , S. ET AL . 2003a. In situ experimental evidence of the fate of a phytodetritus pulse at the abyssal sea floor. Nature, 424, 763– 766. W ITTE , U., A BERLE , N., S AND , M. & W EINHO¨ FER , F. 2003b. Rapid response of a deep-sea benthic community to POM enrichment: an in situ experimental study. Marine Ecology Progress Series, 251, 27–36. W OLLENBURG , J. E. & M ACKENSEN , A. 1998. Living benthic foraminifers from the central Arctic Ocean: faunal composition, standing stock and diversity. Marine Micropaleontology, 34, 153– 185.
On the use of benthic foraminiferal d13C in palaeoceanography: constraints from primary proxy relationships ANDREAS MACKENSEN Alfred Wegener Institute for Polar and Marine Research, 27568 Bremerhaven, Germany (e-mail:
[email protected]) Abstract: Recent findings are reviewed from observations in the field on the generation of the d13C signal in shells of live (Rose Bengal stained) benthic foraminifera, and end up with implications for the interpretation of fossil signatures. The d13C values of calcite tests of preferentially epifaunal foraminifera principally reflect the d13C of dissolved inorganic carbon (DIC) of ambient seawater, whereas infaunal species record a porewater signal, both with an offset from equilibrium calcite. Species occupying the deepest average living depth in the sediment usually exhibit lowest d13C test values, but d13C values of conspecific specimens at a single site do not decrease with increasing subbottom depth and decreasing porewater d13CDIC. Organic carbon fluxes to the sediment surface are generally reflected by infaunal species such that lowered d13C values coincide with high fluxes, but even strictly epifaunal species may reflect seasonally pulsed phytodetritus supply by depleted test d13C. In high-productivity environments, however, where dissolved oxygen and sedimentary carbonate contents are low, benthic foraminiferal tests show 13C enrichment probably due to carbonate-ion undersaturation. Ontogenetic increase in d13C values of certain infaunal species suggests a slow-down of metabolic rates during test growth and decreasing fractionation with age. At sites of active methane discharge d13C values of infaunal species reflect low pore water d13CDIC values, documenting active methane release in the sediment, whereas lowered d13C values of strictly epifaunal species are most probably the result of incorporation of 13C depleted methanotrophic biomass.
Since it was realized that the carbon isotopic composition of benthic foraminiferal tests carried an imprint of the nutrient content of deepwater masses, measuring the 13C/12C ratio of benthic foraminiferal tests is probably the most widespread method for reconstructing past ocean circulation and dissolved nutrient contents (Shackleton 1977; Curry & Lohmann 1982; Duplessy et al. 1984). Particularly, it is used for reconstructions of deep and bottom water circulation during climate cycles in the late Pleistocene (e.g. Curry et al. 1988; Duplessy et al. 1988; Sarnthein et al. 1994; Mackensen et al. 2001; Zahn & Stu¨ber 2002; Bickert & Mackensen 2004; Curry & Oppo 2005). Similarly, on time scales further back in the geological past, benthic foraminiferal d13C was used extensively to gain information on deep-water flow, nutrient content and the global carbon cycle (e.g. Woodruff et al. 1981; Miller & Fairbanks 1985; Woodruff & Savin 1985; Zachos et al. 2001). Last but not least, negative d13C excursions measured in benthic foraminifera from the Palaeocene/Eocene Thermal Maximum at about 55 Ma, and repeatedly in the last 80 000 years fostered hypotheses and speculations on the release of methane into the biosphere by rapid decomposition of submarine gas hydrates (e.g. Dickens et al. 1995; Kennett et al. 2000; Katz et al. 2001).
During the last 25 years, knowledge of past global ocean circulation has significantly increased. This is mainly due to the combined use of a number of various geochemical proxies in addition to d13C, such as for example Cd/Ca ratios and radiogenic tracers, as well as faunal indicator assemblages and species. However, a consistent picture of the global ocean circulation during even the last glacial maximum has yet to emerge (review in Lynch-Stieglitz 2003). Some of the disagreement about water masses and circulation patterns is due to insufficient data coverage as for example in the Southern Ocean and the Pacific, some is the result of conflicting interpretations of the various proxies, and some is the result of only limited knowledge of how the various geochemical signals are formed and recorded in the shells of benthic foraminifera. Synsedimentary and early diagenetic processes such as dissolution and authigenic calcite precipitation, as well as later geological processes connected to sediment subsidence and compaction further overprint the original signature. In this paper, the d13C signal and the primary process of its recording in live benthic foraminifera are focussed on, and findings from observations in the field are reviewed, which have to be considered when interpreting fossil d13C values.
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 121– 133. DOI: 10.1144/SP303.9 0305-8719/08/$15.00 # The Geological Society of London 2008.
122
A. MACKENSEN
Methods
Signal forming
Modern research on deep-sea benthic foraminifera is based on sampling gear that is able to obtain virtually undisturbed surface sediment samples from the ocean floor. Large vented box corers are still the equipment of choice when heavily compacted, sandy or gravelly sediments from areas with low sedimentation rates, re-deposition or erosion are to be sampled. All other deep ocean sediment surfaces are best sampled with the aid of a multiple corer, which might be equipped with a video camera for better control and precise positioning. To precisely sample small-scale features in cold seep environments or at other submarine vent sites remotely operated vehicles (ROVs) or deep-sea submersibles are employed that with the aid of a manipulator are able to handle pushcores. To decipher the ecology and its influence on stable isotope fractionation during calcification of benthic deep-sea foraminifera from field observation it is necessary to identify specimens that have been living at, or at least until recently before the time of sampling. Except for Bernhard & Bowser (1992), all of the papers referred to in this overview, are based on Rose Bengal stained benthic foraminiferal faunas. Preservation of live benthic foraminifera in Rose Bengal stained alcohol is known to have its defects. Particularly, in low-temperature and low-oxygen environments, this method does not exactly discriminate between non-vital protoplasm that may have died weeks or even months ago, and living cells active at the time of sampling (e.g. Bernhard 2000). However, if only brightly coloured and pristine tests are used for analyses, this method reliably indicates live individuals and remains the most appropriate method for broad-scale distributional studies (Murray & Bowser 2000). The stable carbon isotope ratios of dissolved inorganic carbon (DIC) and benthic foraminiferal calcite generally are determined with isotope ratio gas mass spectrometers calibrated via NBS 19 international standard to the VPDB (Vienna Pee Dee Belemnite) scale. All values are given in d-notation versus VPDB with an overall precision of measurements including sample preparation usually better than +0.06 and +0.1‰ for calcite and DIC carbon isotopes, respectively. Except one single-specimen based dataset (Hill et al. 2004), all stable isotope data from papers referred to in this overview are from speciesspecific multi-specimens analyses. The number of specimens used for a single analysis depended on size and weight of species but usually varied between 2 and 25.
The use of d13C values of benthic foraminiferal tests as proxy for past deep-water circulation fundamentally is based on the kinetic isotope fractionation during the uptake of carbon by ocean’s plankton during photosynthesis resulting in organic matter d13C values commonly ranging from 218 to 223‰ VPDB, and a surface water DIC pool enriched in 13C. However, higher kinetic fractionation of carbon isotopes by southern high-latitude plankton communities can cause deviations from “Redfield” d13C stoichiometry (Sackett et al. 1965; Rau et al. 1982). A range of Recent d13CDIC surface water values between 2.5‰ in the mid-latitude Atlantic and 0.7‰ in the northern Pacific has been determined (Kroopnick 1985). In addition to kinetic fractionation during photosynthesis, equilibrium fractionation between atmospheric CO2 and DIC during air/sea gas exchange influences the ocean’s d13CDIC (Broecker & Maier-Reimer 1992; Lynch-Stieglitz et al. 1995; Mackensen et al. 1996). At isotopic equilibrium between atmospheric CO2 and the DIC of the surface ocean, the latter would be enriched in 13C by about 8‰ at 20 8C relative to atmospheric CO2, becoming more enriched by about 1‰ per 10 8C cooling. Thus, given a modern d13C of atmospheric CO2 of 28‰, equilibrium fractionation would result in surface water d13CDIC values ranging from 21 to 2‰. Because of slow air-sea gas exchange rates relative to mass transport of DIC within the ocean, however, isotopic equilibrium usually is not achieved, and an equilibrium temperature effect most importantly applies under specific oceanographic settings, such as sites of sea-ice or bottom-water formation in high latitudes (Mackensen 2001). After isolation from the atmosphere a water mass ages, and as it reflects biological cycling its d13CDIC decreases and its nutrient content increases (Kroopnick 1985). Values of deep-water mass d13CDIC range from 1.2‰ in the core of North Atlantic Deep Water, via 0.4‰ in Circumpolar Deep Water, to 21.0‰ in northern Pacific Deep Water.
Signal recording Microhabitat and average living depth Deep-sea benthic foraminifera live on, and within, the sediment down to a depth of 10 cm below the seafloor (Corliss 1985; Gooday 1986; Mackensen & Douglas 1989). After two decades of intensive research and collecting high-quality samples with the aid of multiple corers, there is no doubt that
BENTHIC FORAMINIFERAL d13C IN PALAEOCEANOGRAPHY
food availability and oxygen content of bottom and interstitial waters most substantially control the benthic foraminiferal microhabitat (e.g. Mackensen et al. 1995; Gooday & Rathburn 1999; Jorissen 1999; Gooday 2003; Licari & Mackensen 2005). Generally, it is assumed that the oxygen penetration depth controls the maximum habitat depth as long as food is available. Consequently, a model was proposed that determines the microhabitat depth in eutrophic environments by a critical oxygen level within the sediment; and by a critical level of food supply in case of oligotrophic environments (Corliss & Emerson 1990; Jorissen et al. 1995). Since the pore-water oxygen concentration and food availability, including bacterial communities, are coupled to seasonal fluctuations in the ocean’s surface productivity, microhabitat depth preferences do not only vary between different species,
123
but also within a single species, depending on season and food supply (cf. Linke & Lutze 1993). Only few species of the marine calcareous benthos precipitate calcite near isotopic equilibrium with DIC of ambient bottom water (Wefer & Berger 1991). Basically, two mechanisms are responsible for carbon isotope disequilibria seen in benthic foraminifera: metabolic and kinetic effects (McConnaughey 1989a, 1989b). Metabolic effects reduce the d13C values due to the incorporation of respired CO2 into the foraminiferal test (Spero & Lea 1996; McConnaughey et al. 1997; Wilson-Finelli et al. 1998). Kinetic isotope fractionation occurs during the hydration and hydroxylation of CO2 and is particularly strong during stages of rapid test calcification. Metabolic and kinetic effects can be encompassed by the term ‘vital effect’. On top of species-specific vital effects the microhabitat effect causes benthic foraminiferal d13C to
0
2
Subbottom depth (cm)
4
6
8
10
12 –1.5
–1
–0.5 δ13C
13
0
0.5
1
(‰VPDB)
Fig. 1. Mean d C values of live epifaunal F. wuellerstorfi (squares), C. pachyderma (diamonds), and L. lobatula (dots) from five stations, as well as d13C values of live infaunal G. affinis (crosses), F. mexicana (triangles), and B. mexicana (open circles) from seven, four, and ten stations, respectively, plotted versus sediment depth. Error bars give standard deviations of means of stations. Stations are in the South Atlantic Ocean along the southwest African continental margin. Bottom water d13CDIC values at stations lie exactly within the range of F. wuellerstorfi d13C, i.e. between 0.45 and 0.65‰ (after Mackensen & Licari 2004).
124
A. MACKENSEN
deviate from the d13CDIC of the bottom water mass. This effect usually refers to low d13C values of benthic foraminifera calcifying within the sediment porewater (McCorkle et al. 1997), or other specific microhabitats, for instance within a phytodetrital layer directly on the sediment surface (Mackensen et al. 1993). Generally, d13C values of preferentially epifaunal foraminiferal species are affected by d13CDIC values of the bottom water mass, whereas those of infaunal species by the d13CDIC of the pore water (Grossman 1984a; McCorkle et al. 1985). The pore water d13CDIC depends on the decomposition rate of sedimentary organic matter, which is largely driven by the flux of particulate organic matter to the sea floor (Belanger et al. 1981; Woodruff & Savin 1985; McCorkle et al. 1990). The d13C values of epifaunal species including for example Fontbotia wuellerstorfi, Cibicidoides pachyderma, and Lobatula lobatula are consistently and significantly higher than d13C values of infaunal species such as for example Globobulimina affinis, Fursenkoina mexicana and Bulimina mexicana (Fig. 1). This is in good agreement with the hypothesis that lower d13C values of species commonly considered to dwell in subbottom microhabitats simply reflect the calcification depth within the sediment and its depleted d13CDIC porewater values (Woodruff et al. 1980; Grossman 1984a; Mackensen & Douglas 1989; McCorkle et al. 1990; Rathburn et al. 1996; McCorkle et al. 1997; Mackensen & Licari 2004; Schmiedl et al. 2004). Alternatively, different taxa may feed on different diets including methanotrophs or even live in symbiosis with bacteria, instead of eating them (Bernhard & Bowser 1992; Bernhard et al. 2000). This eventually, via metabolism and the internal CO2 pool of the foraminifera, affects the stable carbon isotopic composition of their calcareous tests (Jorissen et al. 1995; Kitazato & Ohga 1995; Hill et al. 2004; Mackensen et al. 2006). At a single site specimens of most benthic foraminiferal species record the same d13C signal with only little variation of about +0.10‰, regardless of the actual sediment depth they were separated from (Mackensen & Douglas 1989; McCorkle et al. 1990; McCorkle & Keigwin 1994; Rathburn et al. 1996; McCorkle et al. 1997; Mackensen et al. 2000; Mackensen & Licari 2004). This implies that if specimens of a particular species vertically migrate within the sediment (as is known from culture experiments; Severin 1987; Moodley 1992; Ernst et al. 2002), they obviously do not record different d13C signals in accordance with downwards rapidly decreasing d13C values of pore water DIC (Fig. 1). It was suggested that, by moving within the sediment and successive chamber building, individuals of infaunal species
record an average isotope signal of the pore water (Mackensen & Licari 2004). Alternatively, it is possible that all individuals of the same species at a single site calcify at one specific depth (McCorkle et al. 1990; McCorkle et al. 1997; Mackensen et al. 2000; Tachikawa & Elderfield 2002; Holsten et al. 2004; Mackensen & Licari 2004).
Sustained and pulsed organic matter flux It is long known that d13C values of infaunal Uvigerina peregrina depend on the organic carbon decomposition rate within the sediment and thus on the organic matter flux from the upper ocean to the sediment surface (Zahn et al. 1986). More recent studies corroborate these early findings in suggesting that indeed most, if not all, infaunal species reflect in their d13C values the site-specific organic carbon flux, in spite of having relatively constant d13C values throughout their distribution within the sediments at a single site (Rathburn et al. 1996; Mackensen et al. 2000; Holsten et al. 2004; Mackensen & Licari 2004; Fontanier et al. 2006). In a detailed comparison of open slope and canyon environments in the western Mediterranean Sea, Schmiedl et al. (2004) for example calculated that an increase of organic carbon fluxes of some 7 g C m22year21 results in about 0.6‰ lower d13C values in shallow infaunal Uvigerina mediterranea. Similar to infaunal species, F. wuellerstorfi, although generally a reliable indicator of bottom water d13CDIC (McCorkle & Keigwin 1994; Mackensen & Licari 2004), shows significant negative shifts under conditions of strongly seasonal productivity and phytoplankton blooms with subsequent rapid sedimentation and the development of a phytodetritus layer on the sea floor (Mackensen et al. 1993; Mackensen & Bickert 1999) (Fig. 2). It was suggested that, in areas with strongly seasonal production where particulate organic matter is rapidly deposited, a phytodetritus layer is developed that causes a d13C gradient within this layer above the actual sediment/water interface. This layer affects epifaunal foraminifera in a twofold way: (1) the food supply triggers chamber building and reproduction (cf. Gooday & Turley 1990; Corliss & Silva 1993; Ohga & Kitazato 1997); and (2) calcification will take place in times of very strong d13C gradients at the sediment/water interface.
Methane and methanotrophic food Live benthic foraminifera and their stable isotopic composition at sites of active methane discharge have been studied at the Pacific margins (Rathburn et al. 2000; Bernhard et al. 2001; Rathburn et al. 2003; Torres et al. 2003; Herguera et al. 2004;
BENTHIC FORAMINIFERAL d13C IN PALAEOCEANOGRAPHY
125
2 1:1
δ13CCaCO3 Cibicidoides (‰PDB)
1.5
1
0.5
0
–0.5 –1.5
0
0.5 δ13CΣCO2
1
1.5
bottom water (‰PDB)
Fig. 2. The d13C values of F. wuellerstorfi and closely related cibicidoids plotted versus d13CDIC of bottom water on a global scale. Filled symbols indicate live specimens, triangles and diamonds samples from the Southern and the Arctic Oceans, respectively. Open circles are data from all oceans used for original calibration, and open squares are data from the South Atlantic Ocean. Negative deviations of foraminiferal calcite d13C from one-to-one relationship . 0.2‰ are caused by the phytodetritus effect (after Mackensen & Bickert 1999).
Hill et al. 2004), and the Atlantic margins (Sen Gupta et al. 1997; Mackensen et al. 2006). However, it is not yet clear whether methane venting at the seafloor directly affects the d13C values of live benthic foraminifera at all and, if so, what processes are involved (Stott et al. 2002). Torres et al. (2003) showed that, at sites with anaerobic methane oxidation, tests of live benthic foraminifera have d13C values not significantly lower than those observed in non-venting sediments in spite of extremely low pore water d13CDIC. Herguera et al. (2004) reported on live specimens from active methane venting sites that did not show distinctly lower d13C values than expected from bottom and pore water d13CDIC values. In a study of live benthic foraminifera from a cold methane seep in Monterey Bay off California, Rathburn et al. (2003) concluded that high variability in the stable carbon isotopic composition between conspecific specimens and between different sites can be used to infer past methane seepage. A recent study on live foraminifera from Hydrate Ridge off Oregon reports on single specimens recording d13C values as low as 221.2‰ and average values of multiple specimens of the same species reaching 25.6‰ (Hill et al. 2004). This is
in contrast to the results of Torres et al. (2003) from the same seep area, who measured live foraminifera with d13C values within the range expected from local organic matter decomposition (0 to 24‰). Mackensen et al. (2006) analyzed tests of Fontbotia wuellerstorfi, Cassidulina neoteretis and Cassidulina reniforme from the Ha˚kon Mosby Mud Volcano (HMMV), a cold methane-venting seep off northern Norway at the Barents Sea continental slope. There, in areas densely populated by pogonophoran tubeworms, d13C values of live epifaunal F. wuellerstorfi are up to 4.4‰ lower than at the reference site not affected by methane venting, thus representing the lowest values hitherto reported for this species. Live C. neoteretis and C. reniforme reach d13C values of 27.5 and 25.5‰, respectively (Fig. 3). Specimens of suspension feeder F. wuellerstorfi are almost exclusively found attached to pogonophores, which protrude up to three centimetres above the sediment where d13C values of bottom water DIC are not significantly depleted. Therefore, Mackensen et al. (2006) conclude that low test d13C values of this species are the result of incorporation of strongly 13C depleted methanotrophic biomass the specimens in this particular area feed on, rather
126
A. MACKENSEN
0
–1
δ13C (tests)–δ13CDIC
–2
–3
–4
–5
–6
–7
–8 –0.2
0
0.2 δ18O
0.4
0.6
(tests)–δ18O
0.8
1
1.2
(equ. calcite)
Fig. 3. Stable isotopic composition of live (filled symbols) and dead (open symbols) F. wuellerstorfi (circles), C. neoteretis (diamonds), and C. reniforme (squares) from all sites within the HMMV, and the reference site (R) not influenced by methane seepage. Error bars give standard deviation of replicates at single sites, and R is positioned next to plotted values from the reference site. The d13C and d18O values are adjusted to bottom water d13CDIC (1.11‰) and equilibrium d18O, respectively (after Mackensen et al. 2006).
than due to low bottom water d13CDIC. Lowest d13C of live specimens of infaunal C. neoteretis and C. reniforme exceed the range of pore water d13CDIC values, which usually are due to organic matter decomposition and thus, in fact, might document active methane release in the sediment.
Ontogenetic effects There is increasing evidence that the carbon isotope fractionation factor for calcification in deep-sea benthic foraminifera decreases during individual lifetime, i.e. specimens show a 13C enrichment with increasing test size (Schmiedl et al. 2004 and Schumacher, Mackensen, Jorissen, unpubl. data) (Fig. 4). This is surprising since most previous studies on deep-sea benthic foraminifera suggest no size-dependent changes (e.g. Grossman 1984b; Wefer & Berger 1991; Loubere et al. 1995). However, the only early study so far that reported a size dependent enrichment of benthic d13C values explicitly states that this was the case only
for Uvigerina peregrina (Dunbar & Wefer 1984), and both the data of Schmiedl et al. (2004) and Schumacher et al. (unpubl.) analyzed Uvigerina mediterranea and U. akitaensis, respectively. So, it seems possible that only Uvigerina spp. and closely related taxa show this ontogenetic effect. On the contrary, there is no doubt that most planktic foraminiferal species get enriched in heavy isotopes with increasing test size (Berger 1979; Oppo & Fairbanks 1989; Spero & Lea 1996). The ontogenetic increase in d13C is suggested to be due to high metabolic activity during early life stages and subsequent approach to equilibrium during the later stages of growth and shell thickening with decreased metabolic rates (e.g. Spero & Lea 1996).
Carbonate ion effect The saturation state of the ambient water with respect to carbonate may also influence the isotopic signal recorded in the benthic foraminiferal shell.
BENTHIC FORAMINIFERAL d13C IN PALAEOCEANOGRAPHY
127
Fig. 4. Mean d13C values of live U. mediterranea plotted versus test size from two sites on the continental slope of the western Mediterranean Sea located close to each other, but differing in the amount of organic carbon fluxes. Higher organic carbon flux and more depleted d13C of U. mediterranea are found in ‘Canyon Axis’ samples (after Schmiedl et al. 2004). ROS and RCA give correlation coefficients of calculated logarithmic regression lines for ‘Open Slope’ and ‘Canyon Axis’ sites, respectively.
Culturing experiments revealed that the stable isotopic composition of planktic foraminiferal tests responds to changes in seawater carbonate ion concentration (Spero et al. 1997). Recently, this so-called carbonate ion effect (CIE) was applied to interpret deviations of d13CDIC values of planktic foraminifera Neogloboquadrina pachyderma from surface water d13CDIC values in the Ochotsk Sea (Bauch et al. 2002). If the responses of benthic and planktic foraminifera are similar, a decrease in carbonate ion concentration by 10 mmol/kg would be equivalent to an increase in calcitic test d13C of 0.1‰ (Lea et al. 1999). It was argued that the significant gradients in pH and CO22 3 concentration in deep-sea pore waters may partly explain the differences in stable isotopic compositions of epifaunal and infaunal foraminifera (Bemis et al. 1998; Mackensen & Licari 2004; Schmiedl et al. 2004). In a study on the d13C distribution in tests of live and dead benthic foraminifera in surface and subsurface sediments from 16 stations of intermediate water depths along the SW African continental
margin, Mackensen & Licari (2004) showed that tests of live, infaunal species as well as epifaunal F. wuellerstorfi under high-productivity areas with low bottom water oxygen, and sedimentary carbonate contents below 15% are clearly enriched in 13C (Fig. 5). Off SW Africa, the low sedimentary carbonate contents are related to the oxidation of particulate organic matter within the water column, which is indicated by a well-developed oxygen minimum zone, as well as by enhanced organic matter decomposition in the sediments of the continental slope beneath. Both processes cause the water to become more acidic and lower the carbonate saturation state through the consequent decrease of the carbonate ion concentration, which results in carbonate aggressive bottom and pore waters. In this environment low carbonate ion concentration results in enhanced isotopic fractionation during calcification of foraminiferal tests (Spero et al. 1997) or in an initial dissolution of calcitic tests as early as during the lifetime of the organisms. Although both of these processes induce a shift toward higher d13C values, the latter process,
128
A. MACKENSEN
1
80
70 0.5 60
50
–0.5
40
30
CaCO3 (weight%)
δ13C (‰VPDB)
0
–1 20 –1.5 10
–2 –30
0 –25
–20
–15
–10
–5
0
5
Latitude (°) 13
Fig. 5. Mean d C values of live Fontbotia wuellerstorfi (open squares), Lobatula lobatula (filled squares), Cibicidoides pachyderma (filled diamonds), Bulimina mexicana (crosses), Globobulimina affinis (dots), Fursenkoina mexicana (filled triangles), Uvigerina auberiana (open triangles), and U. peregrina (open diamonds), as well as CaCO3 contents (open circles) of the sediments from stations along the African continental margin plotted versus latitude. Error bars indicate standard deviations of means of all down-core values at each single station. Bottom water d13CDIC values are subtracted. Horizontal lines at +0.2‰ give range of 1:1 relationship between d13C values of bottom water DIC and foraminiferal calcite as commonly tolerated for palaeonutrient reconstructions (from Mackensen & Licari 2004).
however, seems rather unlikely, since during lifetime benthic foraminiferal tests are protected by an outer organic lining against calcite dissolution. Therefore, the enriched calcite d13C values were suggested to be caused by the CIE due to severe carbonate ion undersaturation. This may be best illustrated by the shallow infaunal Bulimina mexicana: Mean d13C values from stations between 26 and 20 8S with .50% sedimentary CaCO3 content, vary between 21.2 and 21.5‰, between 19 and 15 8S with ,5% CaCO3 values increase to 20.7 and 21.2‰, and between 12 8S and the Equator with .15% carbonate, values decrease again below 21.7‰ (Fig. 5). A scatter plot of the sedimentary carbonate content versus mean d13C values of F. wuellerstorfi, Fursenkoina mexicana and Uvigerina auberiana further illustrates that, below a threshold of about 15% sedimentary carbonate content, d13C values are enriched by up to 0.7‰ relative to values of
samples containing 50– 80 % carbonate (Fig. 6). Mackensen & Licari (2004) conclude that in high productivity areas where decomposition of organic matter within both the water column and the interstitial waters is not compensated for by a sufficient supply of carbonate from the oceanic surface layer, the carbonate ion concentration decreases below a threshold (as indicated by the 15 % sedimentary carbonate content), which finally results in a 13C enrichment of both epifaunal F. wuellerstorfi and infaunal species. Above this threshold d13C values of infaunal species only reflect the influence of strongly depleted pore water d13CDIC values. Similar 13C enrichments have been observed earlier in the oxygen minimum zone off Peru (Dunbar & Wefer 1984), but in unstained samples only, so it is not clear whether dissolution or a possible CIE was responsible. With the limited data available, it seems reasonable that under natural conditions at the sediment/water interface
BENTHIC FORAMINIFERAL d13C IN PALAEOCEANOGRAPHY
129
1
0.5
δ13C (‰VPDB)
0
–0.5
–1
–1.5
–2 0
10
20
30
40
50
60
70
80
CaCO3 (weight%) Fig. 6. Mean d13C values of live F. wuellerstorfi (squares), F. mexicana (diamonds), and Uvigerina auberiana (circles) are plotted versus CaCO3 contents of the sediments from stations along the African continental margin. Error bars indicate standard deviations of means of down-core values at each single station. Bottom water d13CDIC values are subtracted (after Mackensen & Licari 2004).
a CIE is operating during calcification of benthic foraminifera (Mackensen & Licari 2004) like in the upper ocean in planktic foraminifera (Bauch et al. 2002).
Palaeoceanographic implications Summarizing the above in terms of palaeoceanographic applicability, I conclude the following: Generally, infaunal species reflect enhanced organic carbon fluxes to the seafloor by decreased test d13C values. However, more quantitative species-specific calibrations are needed before this relationship is reliably applicable in palaeoproductivity reconstructions. Conspecific live individuals of infaunal species display consistent d13C values throughout their distribution within the sediments. This makes the interpretation of fossil benthic foraminiferal d13C values most reliable, since only one signal at one particular site at one particular time is recorded. In areas of seasonally high primary
production epifaunal d13C values may be significantly depleted relative to bottom water d13CDIC: suggesting a too old age of the water mass relative to others. This effect needs to be compensated for when interpreting fossil d13C values from seasonally high-production areas such as for example high-latitude frontal systems. Ontogenetic enrichment in d13C values of certain infaunal species suggests a slow-down of metabolic rates during test growth. Thus, for determining the stable isotopic composition of fossil material, size-class dependent measurements should be performed, similar to the procedure applied in plankton analyses. At sites of methane release, live specimens of infaunal species may record porewater 13CDIC depletions, which exceed those usually caused by organic matter decomposition. A similarly depleted signal in fossil infaunal species compared to site-specific organic matter decomposition rates consequently indicates methane venting into the sediment, but does not say anything about a possible release into the water column. At sites of methane release into
130
A. MACKENSEN
the sediment, tests of live epifaunal species may be significantly depleted in 13C in spite of unchanged bottom water d13CDIC, probably reflecting the incorporation of extremely depleted methanotrophic biomass. Again, a similarly depleted signal in fossil epifaunal species indicates the availability of extremely depleted food and thus methane venting in the vicinity of the sampled site, but does not indicate a possible release of methane into the water column that affected bottom water DIC isotope values. If decomposition of organic matter within an oxygen minimum zone and the porewater below is not compensated for by a sufficient supply of carbonate from the oceanic surface production, then the carbonate ion concentration drops below a threshold that finally results in 13C enrichment of epifaunal F. wuellerstorfi. Even preferentially infaunal species are 13C enriched, although porewater d13CDIC values are low. This paper was the basis of a talk given at the Geological Society Meeting on ‘Biogechemical Controls on Palaeoceanographic Proxies’ in London, 3– 4 October 2005. I thank Rachael James for inviting me, Sylvia Bru¨ckner and Astrid Eberwein for discussion, and two anonymous referees for thorough and helpful reviews.
References B AUCH , D., E RLENKEUSER , H., W INCKLER , G., P AVLOVA , G. & T HIEDE , J. 2002. Carbon isotopes and habitat of polar planktic foraminifera in the Okhotsk Sea: the ‘Carbonate Ion Effect’ under natural conditions. Marine Micropaleontology, 45, 83–99. B ELANGER , P. E., C URRY , W. B. & M ATTHEWS , R. K. 1981. Core-top evaluation of benthic foraminiferal isotopic ratios for paleo-oceanographic interpretations. Palaeogeography, Palaeoclimatology, Palaeoecology, 33, 205–220. B EMIS , B. E., S PERO , H. J., B IJMA , J. & L EA , D. W. 1998. Reevaluation of the oxygen isotopic composition of planktonic foraminifera: Experimental results and revised paleotemperature equations. Paleoceanography, 13, 150–160. B ERGER , W. H. 1979. Stable isotopes in foraminifera. In: L IPPS , J. H., B ERGER , W. H., B UZAS , M. A., D OUGLAS , R. & R OSS , C. A. (eds) Foraminiferal Ecology and Paleoecology. Society of Economic Paleontologists and Mineralogists, Houston, 6, 156–198. B ERNHARD , J. M. 2000. Distinguishing live from dead foraminifera: Methods review and proper applications. Micropaleontology, 46, 38–46. B ERNHARD , J. M. & B OWSER , S. S. 1992. Bacterial biofilms as a trophic resource for certain benthic foraminifera. Marine Ecology Progress Series, 83, 263– 272. B ERNHARD , J. M., B UCK , K. R., F ARMER , M. A. & B OWSER , S. S. 2000. The Santa Barbara Basin is a symbiosis oasis. Nature, 403, 77– 80.
B ERNHARD , J. M., B UCK , K. R. & B ARRY , J. P. 2001. Monterey Bay cold-seep biota: assemblages, abundance, and ultrastructure of living foraminifera. Deep Sea Research Part I: Oceanographic Research Papers, 48, 2233– 2249. B ICKERT , T. & M ACKENSEN , A. 2004. Last glacial to Holocene changes in South Atlantic deep water circulation. In: W EFER , G., M ULITZA , S. & R ATMEYER , V. (eds) The South Atlantic in the Late Quaternary – Reconstruction of Material Budget and Current Systems. Springer-Verlag, Berlin, Heidelberg, 671–693. B ROECKER , W. S. & M AIER -R EIMER , E. 1992. The influence of air and sea exchange on the carbon isotope distribution in the sea. Global Biogeochemical Cycles, 6, 315–320. C ORLISS , B. H. 1985. Microhabitats of benthic foraminifera within deep-sea sediments. Nature, 314, 435–438. C ORLISS , B. H. & E MERSON , S. 1990. Distribution of Rose Bengal stained deep-sea benthic foraminifera from the Nova Scotian continental margin and Gulf of Maine. Deep-Sea Research, 37, 381–400. C ORLISS , B. & S ILVA , K. A. 1993. Rapid growth of deep-sea benthic foraminifera. Geology, 21, 991– 994. C URRY , W. B. & L OHMANN , G. P. 1982. Carbon isotopic changes in benthic foraminifera from the western South Atlantic: reconstruction of abyssal circulation patterns. Quaternary Research, 18, 218– 235. C URRY , W. B. & O PPO , D. W. 2005. Glacial water mass geometry and the distribution of d13C of SCO2 in the western Atlantic Ocean. Paleoceanography, 20, doi:10.1029/2004PA001021. C URRY , W. B., D UPLESSY , J. C., L ABEYRIE , L. D. & S HACKLETON , N. J. 1988. Changes in the distribution of d13C of deep water SCO2 between the last glaciation and the Holocene. Paleoceanography, 3, 317–341. D ICKENS , G., O’N EIL , J. R., R EA , D. K. & O WEN , R. M. 1995. Dissociation of oceanic methane hydrate as a cause of the carbon isotope excursion at the end of the Paleocene. Paleoceanography, 10, 965– 972. D UNBAR , R. B. & W EFER , G. 1984. Stable isotope fractionation in benthic foraminifera from the Peruvian continental margin. Marine Geology, 59, 215–225. D UPLESSY , J. C., S HACKLETON , N. J., M ATTHEWS , R. K., P RELL , W., R UDDIMAN , W. F., C ARALP , M. H. & H ENDY , C. H. 1984. 13C record of benthic foraminifera in the last interglacial ocean: implications for the carbon cycle and the global deep water circulation. Quaternary Research, 21, 225– 243. D UPLESSY , J. C., S HACKLETON , N. J., F AIRBANKS , R. G., L ABEYRIE , L., O PPO , D. & K ALLEL , N. 1988. Deepwater source variations during the last climatic cycle and their impact on the global deepwater circulation. Paleoceanography, 3, 343– 360. E RNST , S. R., D UIJNSTEE , I. A. P. & VAN DER Z WAAN , G. J. 2002. The dynamics of the benthic foraminiferal microhabitat: recovery after experimental disturbance. Marine Micropaleontology, 46, 343–362. F ONTANIER , C., M ACKENSEN , A., J ORISSEN , F. J., A NSCHUTZ , P., L ICARI , L. & G RIVEAUD , C. 2006. Stable oxygen and carbon isotopes of live benthic foraminifera from the Bay of Biscay: microhabitat
BENTHIC FORAMINIFERAL d13C IN PALAEOCEANOGRAPHY impact and seasonal variability. Marine Micropaleontology, 58, 159– 183. G OODAY , A. J. 1986. Meiofaunal foraminiferans from the bathyal Porcupine Seabight (north-east Atlantic): size structure, standing stock, taxonomic composition, species diversity and vertical distribution in the sediment. Deep-Sea Research, 33, 1345– 1373. G OODAY , A. J. 2003. Benthic Foraminifera (Protista) as tools in deep-water palaeoceanography: environmental influences on faunal characteristics. Advances in Marine Biology, 46, 2– 90. G OODAY , A. J. & R ATHBURN , A. E. 1999. Temporal variability in living deep-sea benthic foraminifera: a review. Earth Science Reviews, 46, 187 –212. G OODAY , A. J. & T URLEY , C. M. 1990. Responses by benthic organisms to inputs of organic material to the ocean floor: a review. Philosophical Transactions of the Royal Society London, A 331, 119–138. G ROSSMAN , E. L. 1984a. Carbon isotopic fractionation in live benthic foraminifera—comparison with inorganic precipitate studies. Geochimica et Cosmochimica Acta, 48, 1505–1512. G ROSSMAN , E. L. 1984b. Stable isotope fractionation in live benthic foraminifera from the southern California borderland. Palaeogeography, Palaeoclimatology, Palaeoecology, 47, 301–327. H ERGUERA , J., P EREZ , E., P AULL , C., U SSLER , B., P ELTZER , E., B REWER , P. & C ORTINA , A. 2004. Carbon isotopic signatures in living benthic foraminifera from methane-soaked gas hydrate-bearing sediments in Guaymas Basin, Gulf of California. EOS Transactions, American Geophysical Union, 85, Abstract B43B-0144. H ILL , T. M., K ENNETT , J. P. & V ALENTINE , D. L. 2004. Isotopic evidence for the incorporation of methanederived carbon into foraminifera from modern methane seeps, Hydrate Ridge, Northeast Pacific. Geochimica et Cosmochimica Acta, 68, 4619– 4627. H OLSTEN , J., S TOTT , L. & B ERELSON , W. 2004. Reconstructing benthic carbon oxidation rates using d13C of benthic foraminifers. Marine Micropaleontology, 53, 117–132. J ORISSEN , F. J. 1999. Benthic foraminiferal microhabitats below the sediment-water interface. In: S EN G UPTA , B. K. (ed.) Modern Foraminifera. Kluwer Academic Publishers, Dordrecht, Boston, London, 161–179. J ORISSEN , F. J., DE S TIGTER , H. C. & W IDMARK , J. G. V. 1995. A conceptual model explaining benthic foraminiferal microhabitats. Marine Micropaleontology, 26, 3 –15. K ATZ , M. E., C RAMER , B. S., M OUNTAIN , G. S., K ATZ , S. & M ILLER , K. G. 2001. Uncorking the bottle: what triggered the Paleocene/Eocene thermal maximum methane release? Paleoceanography, 16, 549–562. K ENNETT , J. P., C ANNARIATO , K. G., H ENDY , I. L. & B EHL , R. J. 2000. Carbon isotopic evidence for methane hydrate instability during Quaternary interstadials. Science, 288, 128–133. K ITAZATO , H. & O HGA , T. 1995. Seasonal changes in deep-sea benthic foraminiferal populations: results of long-term observations at Sagami Bay, Japan. In: S AKAI , H. & N OZAKI , Y. (eds) Biogeochemical Processes and Ocean Flux in the Western Pacific.
131
Terra Scientific Publishing Company (Terrapub), Tokyo, 331– 342. K ROOPNICK , P. M. 1985. The distribution of 13C of total CO2 in the world oceans. Deep-Sea Research, 32, 57–84. L EA , D. W., B IJMA , J., S PERO , H. J. & A RCHER , D. E. 1999. Implications of a carbonate ion effect on shell carbon and oxygen isotopes for glacial ocean conditions. In: F ISCHER , G. & W EFER , G. (eds) Use of Proxies in Paleoceanography. Springer-Verlag, Berlin Heidelberg New York, 513–522. L ICARI , L. & M ACKENSEN , A. 2005. Benthic foraminifera off West Africa (18N to 328S): do live assemblages from the topmost sediment reliably record environmental variability? Marine Micropaleontology, 55, 205– 233. L INKE , P. & L UTZE , G. F. 1993. Microhabitat preferences of benthic foraminifera – a static concept or a dynamic adaptation to optimize food acquisition? Marine Micropaleontology, 20, 215– 134. L OUBERE , P., M EYERS , P. & G ARY , A. 1995. Benthic foraminiferal microhabitat selection, carbon isotope values, and association with larger animals: a test with Uvigerina peregrina. Journal of Foraminiferal Research, 25, 83–95. L YNCH -S TIEGLITZ , J. 2003. Tracers of past ocean circulation. In: E LDERFIELD , H. (ed.) The Oceans and Marine Geochemistry. Elsevier Ltd., Amsterdam, 6, 433– 451. L YNCH -S TIEGLITZ , J., S TOCKER , T. F., B ROECKER , W. S. & F AIRBANKS , R. G. 1995. The influence of air-sea exchange on the isotopic composition of oceanic carbon: observations and modeling. Global Biogeochemical Cycles, 9, 653– 665. M ACKENSEN , A. 2001. Oxygen and carbon stable isotope tracers of Weddell Sea water masses: new data and some paleoceanographic implications. Deep-Sea Research I, 48, 1401–1422. M ACKENSEN , A. & B ICKERT , T. 1999. Stable carbon isotopes in benthic foraminifera: proxies for deep and bottom water circulation and new production. In: F ISCHER , G. & W EFER , G. (eds) Use of Proxies in Paleoceanograpy: Examples from the South Atlantic. Springer-Verlag, Berlin Heidelberg, 229 –254. M ACKENSEN , A. & D OUGLAS , R. G. 1989. Down-core distribution of live and dead deep-water benthic foraminifera in box cores from the Weddell Sea and the California continental borderland. Deep-Sea Research, 36, 879– 900. M ACKENSEN , A. & L ICARI , L. 2004. Carbon isotopes of live benthic foraminifera from the eastern South Atlantic Ocean: sensitivity to bottom water carbonate saturation state and orgnaic matter rain rates. In: W EFER , G., M ULITZA , S. & R ATMEYER , V. (eds) The South Atlantic in the Late Quaternary – Reconstruction of Material Budget and Current Systems. Springer-Verlag, Berlin, Heidelberg, 623–644. M ACKENSEN , A., H UBBERTEN , H.-W., B ICKERT , T., F ISCHER , G. & F U¨ TTERER , D. K. 1993. The d13C in benthic foraminiferal tests of Fontbotia wuellerstorfi (Schwager) relative to the d13C of dissolved inorganic carbon in southern ocean deep water: Implications for glacial ocean circulation models. Paleoceanography, 8, 587–610.
132
A. MACKENSEN
M ACKENSEN , A., S CHMIEDL , G., H ARLOFF , J. & G IESE , M. 1995. Deep-sea Foraminifera in the South Atlantic Ocean: ecology and assemblage generation. Micropaleontology, 41, 342– 358. M ACKENSEN , A., H UBBERTEN , H.-W., S CHEELE , N. & S CHLITZER , R. 1996. Decoupling of d13CSCO2 and phosphate in Recent Weddell Sea Deep and Bottom Water: implications for glacial southern ocean paleoceanography. Paleoceanography, 11, 203–215. M ACKENSEN , A., S CHUMACHER , S., R ADKE , J. & S CHMIDT , D. N. 2000. Microhabitat preferences and stable carbon isotopes of endobenthic foraminifera: clue to quantitative reconstruction of oceanic new production? Marine Micropaleontology, 40, 233– 258. M ACKENSEN , A., R UDOLPH , M. & K UHN , G. 2001. Late Pleistocene deep-water circulation in the subantarctic eastern Atlantic. Global and Planetary Change, 30, 195– 226. M ACKENSEN , A., W OLLENBURG , J. & L ICARI , L. 2006. Low d13C in tests of live epibenthic and endobenthic foraminifera at a site of active methane seepage. Paleoceanography, 21, PA2022, doi:10.1029/ 2005PA001196. M C C ONNAUGHEY , T. 1989a. 13C and 18O isotopic disequilibrium in biological carbonates: I. Patterns. Geochimica et Cosmochimica Acta, 53, 151–162. M C C ONNAUGHEY , T. 1989b. 13C and 18O isotopic disequilibrium in biological carbonates: II. In vitro simulation of kinetic isotope effects. Geochimica et Cosmochimica Acta, 53, 163– 171. M C C ONNAUGHEY , T., B URDETT , J., W HELAN , J. F. & P AULL , C. K. 1997. Carbon isotopes in biological carbonates: respiration and photosynthesis. Geochimica et Cosmochimica Acta, 61, 611– 622. M C C ORKLE , D. C., E MERSON , S. R. & Q UAY , P. D. 1985. Stable carbon isotopes in marine porewaters. Earth and Planetary Science Letters, 74, 13– 26. M C C ORKLE , D. C., K EIGWIN , L. D., C ORLISS , B. H. & E MERSON , S. R. 1990. The influence of microhabitats on the carbon isotopic composition of deep-sea benthic foraminifera. Paleoceanography, 5, 161–185. M C C ORKLE , D. C. & K EIGWIN , L. D. 1994. Depth profiles of d13C in bottom water and core-top C. wuellerstorfi on the Ontong-Java Plateau and Emperor Seamounts. Paleoceanography, 9, 197–208. M C C ORKLE , D. C., C ORLISS , B. H. & F ARNHAM , C. A. 1997. Vertical distributions and stable isotopic compositions of live (stained) benthic foraminifera from the North Carolina and California continental margins. Deep-Sea Research I, 44, 983– 1024. M ILLER , K. G. & F AIRBANKS , R. G. 1985. Oligocene to Miocene carbon isotope cycles and abyssal circulation changes. The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present. American Geophysical Union, Washington DC, 32, 469– 486 M OODLEY , L. 1992. Laboratory experiments on the infaunal activity in benthic foraminifera. In: M OODLEY , L. (ed.) Experimental Ecology of Benthic Foraminifera in Soft Sediments and its (Paleo) Environmental Significance. Vrije Universiteit, Amsterdam, Dissertation, 89–106. M URRAY , J. W. & B OWSER , S. S. 2000. Mortality, protoplasm decay rate, and reliability of staining techniques
to recognize ‘living’ foraminifera: a review. Micropaleontology, 30, 66– 77. O HGA , T. & K ITAZATO , H. 1997. Seasonal changes in bathyal foraminiferal populations in response to the flux of organic matter (Sagami Bay, Japan). Terra Nova, 9, 33– 37. O PPO , D. W. & F AIRBANKS , R. G. 1989. Carbon isotope composition of tropical surface water during the past 22,000 years. Paleoceanography, 4, 333 –351. R ATHBURN , A. E., C ORLISS , B. H., T APPA , K. D. & L OHMANN , K. C. 1996. Comparisons of the ecology and stable isotopic compositions of living (stained) benthic foraminifera from the Sulu and South China Seas. Deep-Sea Research I, 43, 1617– 1646. R ATHBURN , A. E., L EVIN , L. A., H ELD , Z. & L OHMANN , K. C. 2000. Benthic foraminifera associated with cold methane seeps on the northern California margin: ecology and stable isotopic composition. Marine Micropaleontology, 38, 247–266. R ATHBURN , A. E., P EREZ , M. E., M ARTIN , J. B., D AY , S. A., M AHN , C., G IESKES , J., Z IEBIS , W., W ILLIAMS , D. & B AHLS , A. 2003. Relationships between the distribution and stable isotopic composition of living benthic foraminifera and cold methane seep biogeochemistry in Monterey Bay, California. Geochemistry, Geophysics, Geosystems, 4, 1106, doi:1110.1029/ 2003GC000595. R AU , G. H., S WEENEY , R. E. & K APLAN , I. R. 1982. Plankton 13C:12C ratio changes with latitude: differences between northern and southern oceans. Deep-Sea Research I, 29, 1035–1039. S ACKETT , W. M., E CKELMANN , W. R., B ENDER , M. L. & B E´ , A. W. 1965. Temperature dependence of carbon isotope composition in marine plankton and sediments. Science, 148, 235–237. S ARNTHEIN , M., W INN , K., J UNG , S. J. A., D UPLESSY , J.-C., L ABEYRIE , L., E RLENKEUSER , H. & G ANSSEN , G. 1994. Changes in east Atlantic deepwater circulation over the last 30,000 years: eight time slice reconstructions. Paleoceanography, 9, 209–267. S CHMIEDL , G., P FEILSTICKER , M., H EMLEBEN , C. & M ACKENSEN , A. 2004. Environmental and biological effects on the stable isotope composition of Recent deep-sea benthic foraminifera from the Mediterranean Sea. Marine Micropaleontology, 51, 129– 152. S EN G UPTA , B. K., P LATON , E., B ERNHARD , J. M. & A HARON , P. 1997. Foraminiferal colonization of hydrocarbon-seep bacterial mats and underlying sediment, Gulf of Mexico slope. Journal of Foraminiferal Research, 27, 292– 300. S EVERIN , K. P. 1987. Laboratory observations of the rate of subsurface movement of a small miliolid foraminifer. Journal of Foraminiferal Research, 17, 110–116. S HACKLETON , N. J. 1977. Tropical rainforest history and the equatorial Pacific carbonate dissolution cycles. In: A NDERSEN , N. R. & M ALAHOFF , A. (eds) Fate of Fossil Fuel CO2 in the Oceans. Plenum Press, New York, 401 –427. S PERO , H. J., B IJMA , J., L EA , D. W. & B EMIS , B. E. 1997. Effect of seawater carbonate concentration on foraminiferal carbon and oxygen isotopes. Nature, 390, 497– 500.
BENTHIC FORAMINIFERAL d13C IN PALAEOCEANOGRAPHY S PERO , H. J. & L EA , D. W. 1996. Experimental determination of stable isotope variability in Globigerina bulloides: implications for paleoceanographic reconstructions. Marine Micropaleontology, 28, 231– 246. S TOTT , L. D., B UNN , T., P ROKOPENKO , M., M AHN , C., G IESKES , J. & B ERNHARD , J. M. 2002. Does the oxidation of methane leave an isotopic fingerprint in the geologic record? Geochemistry, Geophysics, Geosystems, 3, 1012; doi:1010.1029/2001 GC000196. T ACHIKAWA , K. & E LDERFIELD , H. 2002. Microhabitat effects on Cd/Ca and d13C of benthic foraminifera. Earth and Planetary Science Letters, 202, 607– 624. T ORRES , M. E., M IX , A. C., K INPORTS , K., H ALEY , B., K LINKHAMMER , G. P., M C M ANUS , J. & DE A NGELIS , M. A. 2003. Is methane venting at the seafloor recorded by d13C of benthic foraminifera shells? Paleoceanography, 18, 1062; doi: 1010.1029/ 2002PA000824. W EFER , G. & B ERGER , W. H. 1991. Isotope paleontology: growth and composition of extant calcareous species. Marine Geology, 100, 207– 248. W ILSON -F INELLI , A., C HANDLER , G. T. & S PERO , H. J. 1998. Stable isotope behavior in paleoceanographically important benthic foraminifera: results from microcosm culture experiments. Journal of Foraminiferal Research, 28, 312–320.
133
W OODRUFF , F. & S AVIN , S. 1985. d13C values of Miocene Pacific benthic foraminifera: correlations with sea level and biological productivity. Geology, 13, 119– 122. W OODRUFF , F., S AVIN , S. & D OUGLAS , R. G. 1980. Biological fractionation of oxygen and carbon isotopes by recent benthic foraminifera. Marine Micropaleontology, 5, 3 –11. W OODRUFF , F., S AVIN , S. & D OUGLAS , R. G. 1981. Miocene stable isotope record: a detailed deep Pacific Ocean study and its paleoclimatic implications. Science, 212, 665– 668. Z ACHOS , J. C., P AGANI , M., S LOAN , L., T HOMAS , E. & B ILLUPS , K. 2001. Trends, Rhythms, and Aberrations in Global Climate 65 Ma to Present. Science, 292, 686– 693. Z AHN , R. & S TU¨ BER , A. 2002. Suborbital intermediate water variability inferred from paired benthic foraminiferal Cd/Ca and d13C in the tropical West Atlantic and linking with North Atlantic climates. Earth and Planetary Science Letters, 200, 191– 205. Z AHN , R., W INN , K. & S ARNTHEIN , M. 1986. Benthic foraminiferal d13C and accumulation rates of organic carbon: Uvigerina peregrina group and Cibicidoides wuellerstorfi. Paleoceanography, 1, 27–42.
The carbon and oxygen stable isotopic composition of cultured benthic foraminifera DANIEL C. MC CORKLE1, JOAN M. BERNHARD1, CHRISTOPHER J. HINTZ2, JESSICA K. BLANKS3, G. THOMAS CHANDLER2 & TIMOTHY J. SHAW4 1
Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA, 02543, USA (e-mail:
[email protected])
2
Department of Environmental Health Sciences, Arnold School of Public Health, University of South Carolina, Columbia, SC, 29208, USA
3
Marine Science Program, University of South Carolina, Columbia, SC, 29208, USA 4
Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC, 29208, USA
Abstract: Laboratory cultures of several species of benthic foraminifera were grown under controlled physical and chemical conditions during months-long experiments carried out at the University of South Carolina in 2001 and 2002. A dozen experimental culture chambers contained a c. 1– 3 mm layer of trace-metal free silica substrate, and were continuously flushed with water from a large (1600 L) seawater reservoir with known, constant temperature and composition (d18O(water), carbonate system chemistry, and trace element concentrations). Each year, in most of the culture chambers, one or more species reproduced, producing hundreds of juveniles which grew into size classes ranging from 100 to 500 microns. Bulimina aculeata was the most successful species in the 2001 cultures, and both B. aculeata and Rosalina vilardeboana were abundant in 2002. We determined the shell C and O isotopic composition of the cultured foraminifera, and compared these isotopic values with the water chemistry of the culture chambers, and also with the shell chemistry of field specimens collected from sites on the North Carolina and South Carolina (USA) continental margin. The cultured foraminifera showed substantial offsets from the d13C of system water dissolved inorganic carbon (20.5 to 22.5‰, depending on species) and smaller offsets (0 to 20.5‰) from the predicted d18O of calcite in equilibrium with the culture system water at the growth temperature. These offsets reflect at least three factors: species-dependent vital effects; ontogenetic variations in shell chemistry; and the aqueous carbonate chemistry ([CO2 3 ] or pH) of the experimental system.
The isotopic and elemental compositions of the calcium carbonate shells of fossil foraminifera are widely used to estimate ocean chemistry and temperature at the time of calcification. Numerous fieldbased calibration studies have shown the reliability of foraminifera as proxy recorders, and these studies have taught us a great deal about the biological processes and environmental factors that influence foraminiferal shell chemistry. In general, benthic foraminiferal d18O values reflect bottom water temperature and the oxygen isotopic composition of water (Shackleton & Opdyke 1973), the d13C values of epibenthic species reflect the d13C of bottom water dissolved inorganic carbon (DIC) (Duplessy et al. 1984; Curry et al. 1988), and the metal:calcium ratios of benthic foraminifera reflect the trace element content of bottom water (e.g. cadmium: Boyle 1988, 1992; barium: Lea & Boyle 1989, 1990; zinc: Marchitto et al. 2000),
and also bottom water temperature (magnesium: Lear et al. 2002; Martin et al. 2002). However, differences between paired foraminiferal shell chemistry records demonstrate that our understanding of these proxies remains incomplete (e.g. Boyle 1992; Rosenthal et al. 1995; Boyle & Rosenthal 1996). Field calibration studies have revealed potential artifacts that can complicate proxy interpretations (McCorkle et al. 1990; Mackensen et al. 1993; McCorkle et al. 1995), and staining studies have documented the microhabitats (i.e. epibenthic (at the sediment-water interface) and endobenthic (within the sediments)) occupied by different benthic foraminiferal species (Corliss 1985; Mackensen & Douglas 1989; Corliss & Emerson 1990). Staining methods have been used both to focus palaeochemical calibration efforts on live (or ‘recently’ live) specimens, and to use species’ microhabitat preferences
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 135– 154. DOI: 10.1144/SP303.10 0305-8719/08/$15.00 # The Geological Society of London 2008.
136
D. C. MC CORKLE ET AL.
to help interpret their shell chemistry (Grossman 1987; McCorkle et al. 1990; Mackensen et al. 1993; Rathburn et al. 1996; McCorkle et al. 1997; Mackensen et al. 2000; Schmeidl et al. 2004). Although most calibrations of foraminiferal proxy records have been carried out using core-top sediments, the inherent environmental variability in the ocean makes interpretation of these empirical calibration studies difficult. In addition, important environmental, ecological, and physiological factors often co-vary in the field. An alternative way to calibrate geochemical proxies in foraminifera is to grow the foraminifers in the laboratory under known environmental conditions. Culturing studies with planktic foraminifera provide the best-known examples of this approach, including the culture-based calibration of the planktic isotopic and elemental thermometers (Erez & Luz 1982; Bijma et al. 1990; Nurnberg et al. 1996), and a set of planktic foraminifera culture experiments carried out by Spero, Lea and co-workers (Lea & Spero 1992; Spero & Lea 1993; Lea et al. 1995; Spero et al. 1997; Bemis et al. 1998; Bijma et al. 1999). These latter studies demonstrated the influence of the carbonate chemistry of the growth medium (seawater carbonate ion concentration or pH) on shell C and O isotopic composition. Culturing work has also investigated the controls on foraminiferal boron isotopic composition (Sanyal et al. 1996; Ho¨nisch et al. 2003) and metal:calcium ratios (Russell et al. 1994, 2004; Mashiotta et al. 1997; Lea et al. 1999). Culturing studies of benthic foraminifera are still in their infancy compared with the planktic experiments noted above, although successful studies have investigated the C and O isotopic composition (Chandler et al. 1996; Wilson-Finelli et al. 1998) and trace metal content (Russell et al. 1994; Toyufuku et al. 2000; Havach et al. 2001; Segev & Erez 2006; Hintz et al. 2006a, b) of cultured benthic foraminifera. Here, we present carbon and oxygen stable isotopic results from two benthic foraminiferal culturing experiments (Hintz et al. 2004, 2006a, b), and compare these results with the physical and chemical conditions in the culture system, and with new and previously published C and O isotopic data from field specimens.
Methods Live benthic foraminifera were collected from a 770 m site off the coast of Cape Hatteras, North Carolina and from a 220 m site on the ‘Charleston Bump’ off the coast of South Carolina (R/V Cape Henlopen; June 2001), and from a 1020 m site south of Cape Hatteras (R/V Oceanus; April 2002). Bottom water samples for chemical analyses
were collected using a corer-mounted Niskin bottle, and also with CTD rosette casts at nearby sites. At sea, flocculent surface sediments were siphoned off the tops of Soutar box cores and gently sieved, taking care to keep the sediment material at in-situ temperatures. The . 75 mm fraction was kept in bottom water at in-situ temperatures for the duration of each cruise, and returned to the laboratory in Columbia, SC, where stock foraminiferal cultures were maintained in a 7 8C environmental chamber. Separate sediment samples from the top 0.5 cm of cores collected at each site were incubated on board the ship at in-situ temperature with the vital probe CellTracker Green (Molecular Probes, Invitrogen Detection Technologies) to label foraminifera that were alive at the time of core collection (Bernhard 2000), and these sediments were then preserved for later foraminiferal identification and shell chemistry analyses on shore. The University of South Carolina culturing system and the physical and chemical conditions of the August 2001 through December 2001 experiment (‘2001 experiment’) and the June 2002 through January 2003 experiment (‘2002 experiment’) have been described by Hintz et al. (2004). The 1600 liter culture medium reservoir for these two experiments contained a mixture of seawater and artificial seawater with a salinity of approximately 35‰ and the culture system temperature was held constant at approximately 7 8C + 0.5 8C (Hintz et al. 2004). Prior to each experiment, foraminifers were incubated in calcein (Bernhard et al. 2004). Calcein staining prior to culture chamber inoculation provided a way to distinguish preexperimental individuals (and pre-experimental chambers within an individual) from individuals or chambers that calcified under controlled conditions during the experiments. At the start of each culturing experiment 80–100 calcein-stained specimens (multiple species from a single site, . 90 mm in diameter) were added to cell tissue culture cups (8 mL volume, 8 mm nominal pore diameter) housed in acrylic culture chambers. There were nine culture chambers in the 2001 experiment and twelve in the 2002 experiment. Each culture cup contained a 1–2 mm-thick layer of silt-sized silicon dioxide to provide the physical substrate required for foraminiferal growth; the culture chamber design and use of an artificial substrate minimize formation of sedimentary microhabitats within the culture cups (Wilson-Finelli et al. 1998; Havach et al. 2001; Hintz et al. 2004). Samples of the culture system water were collected bi-weekly for determination of total alkalinity and total dissolved inorganic carbon (DIC) concentrations, the carbon isotopic composition of DIC (d13C(DIC)), and the oxygen isotopic composition
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
of water (d18O(water)). The alkalinity and DIC data discussed in this report were obtained by closed vessel titration of large volume (c. 100 mL) samples using an automated titration system (Bradshaw et al. 1981; Brewer et al. 1986). Alkalinity and DIC concentrations were determined using a nonlinear curve fitting approach (Department of Energy 1994), and standardized using certified reference materials obtained from Dr A. Dickson (Scripps Institution of Oceanography). The standard deviation of pairs of replicate analyses of culture water was 3.9 meq/kg for alkalinity and 4.1 mmol/ kg for DIC in the 2001 experiment (n ¼ 17), and 2.8 meq/kg for alkalinity and 5.6 mmol/kg for DIC in the 2002 experiment (n ¼18). The d13C of DIC was determined by acidification and automated vacuum extraction of small samples (3 –5 mL), followed by isotope ratio measurements on the VG-PRISM mass spectrometer in the National Ocean Science Accelerator Mass Spectrometer laboratory at WHOI (McCorkle et al. 1990; McCorkle & Klinkhammer 1991). The standard deviation of replicate extractions and isotopic analyses was 0.04‰ (2002, n ¼ 14); d13C values are reported relative to VPDB. The d18O of water was determined in the laboratory of Dr D. Schrag (Harvard University) using a VG Optima mass spectrometer with a VG Isoprep 18 automated shaker/equilibrator (Schrag et al. 2002). The standard deviation of replicate d18O analyses was 0.02‰ (2002, n ¼ 21); d18O(water) values are reported relative to SMOW. The carbon and oxygen isotopic composition of foraminiferal calcite was determined using the Finnigan 252-Kiel Device mass spectrometer in Dr W. Curry’s lab at WHOI. Because the cultured foraminifera were generally fairly small and lightly calcified (0.8 to 4.0 mg CaCO3 per specimen), multiple specimens were pooled for each analysis (8 to 100 individuals/analysis for B. aculeata in the 2001 experiment, 10 to 32 individuals/analysis for B. aculeata in the 2002 experiment, and 10 to 43 individuals/analysis for R. vilardeboana in the 2002 experiment). These multi-specimen samples insured that the total mass of carbonate analysed ranged from roughly 30 to 130 ug of CaCO3. The CellTracker Green-labelled field specimens were larger and more massive; each analysis of field specimens used 11 to 23 individuals. Foraminiferal d13C and d18O values are reported relative to VPDB.
Results Chemistry of culture system water The 1600 L of culture system water in the 2001 experiment was a roughly 3:1 mixture of Instant
137
Ocean artificial seawater and filtered Gulf Stream seawater, and for the 2002 experiment the 2001 water was amended by replacing approximately 480 L of the Instant Ocean-seawater mixture with Gulf Stream seawater (Hintz et al. 2004). The salinity data show considerable scatter, both on each sampling date and through time (Fig. 1). We suspect that these apparent salinity fluctuations reflect analytical problems with the osmometerbased salinity estimates, and take the constant d18O(water) and alkalinity values (below) as evidence of the system’s chemical stability (Hintz et al. 2004). The low absolute values of the culture water d18O reflect the oxygen isotopic composition of the deionized water used to prepare the Instant Ocean in 2001; the increase between 2001 and 2002 reflects the addition of surface seawater to the system (Fig. 1). The d18O(water) values of the culture system remained essentially constant during 2001, and decreased slightly between June and September of 2002, after which they remained constant. The average d18O(water) values were 24.36 + 0.12‰ in 2001, and 23.39 + 0.07‰ in 2002. In 2001 the alkalinity remained constant at 3176 + 7 meq/kg, while DIC increased slightly during the course of the experiment, from 2867 to 2883 mmol/kg; in 2002 the alkalinity was 2959 + 6 meq/kg, and DIC again increased slightly, from 2675 to 2687 mmol/kg (Fig. 2). The long-term variability in alkalinity during each several-month experiment (+0.2%) was slightly larger than the precision of our alkalinity measurements (+0.1%), but was nonetheless small, and the alkalinity values did not show a significant trend in either year. The alkalinity and DIC of the culture system water in 2001 was substantially higher than normal seawater values because the system water was ‘polished’ prior to the start of the experiment by passing it through columns of manganese fibre, activated charcoal, and crushed coral (Hintz et al. 2004). For comparison, the measured bottom water alkalinity and DIC values at the 220 m site were 2324 + 4 meq/kg and 2196 + 8 mmol/kg, respectively, at the 770 m site alkalinity and DIC values were 2323 meq/kg and 2191 mmol/kg, and at the 1020 m site alkalinity and DIC values were 2319 + 8 meq/kg and 2159 + 1 mmol/kg. The culture system alkalinity and DIC values in the 2002 experiment were lower than in the 2001 experiment (though still higher than seawater) due to the addition of surface seawater to the system. The culture water DIC values increased by 0.6% in 2001 and 0.8% in 2002. The DIC stable isotope data, discussed below, suggest that these small DIC increases reflect the slow partial equilibration of the 1600 L culture system with the fall-winter
D. C. MC CORKLE ET AL.
–3.0
35
–3.5
34
–4.0
Salinity (ppt)
(a) 36
33
Salinity
δ18O (water) ‰
138
–4.5
δ18O 9/1/2001
12/1/2001
–5.0 3/1/2002 –2.0
35
–2.5
34
–3.0
33
–3.5
Salinity (ppt)
(b) 36
32 6/1/2002
9/1/2002
12/1/2002
δ18O (water) ‰
32 6/1/2001
–4.0 3/1/2003
Date Fig. 1. Time series of the salinity (dotted diamonds) and d18O(water) (solid diamonds) of the culture system water during the 2001 experiment (a) and 2002 experiment (b). The apparent salinity fluctuations reflect analytical (calibration) problems; the relatively constant d18O(water) values are a better indication of the system’s chemical stability.
pCO2 increase during the annual cycle of atmospheric pCO2. The alkalinity and DIC data were used to calculate the carbonate ion concentration and pH of the culture medium during the second half of the first experiment, and throughout the second experiment (Fig. 3). These calculations used the program CO2SYS (Lewis & Wallace 1998) with the carbonate dissociation constants of Roy et al. (1993), the calcite solubility of Mucci (1983), and the assumption that the boron/salinity ratio of the culture system water was equal to the seawater ratio. Because much of the culture system water in both years was Instant Ocean, it may not be correct to estimate the total borate concentration from the whole-ocean boron/salinity relationship. However, trends in the concentrations of carbonate system species during each year will be independent of the actual absolute total borate concentration. Because of the high alkalinity of the reservoir water, the culture experiment carbonate ion
concentration (230– 240 mmol/kg in 2001, 205– 225 mmol/kg in 2002 (Fig. 3)) was more than twice that of the bottom water at the two sampling sites (100 to 115 mmol/kg). The calculated culture system pH (8.2) was correspondingly higher than bottom water pH at the field sites (7.9 to 8.0). Possible impacts of these high carbonate ion concentrations on foraminiferal isotopic composition are discussed below. The scatter in the calculated carbonate ion concentrations reflects the sensitivity of these calculations to small variations in alkalinity and DIC. The increase in culture water DIC through the course of each experiment resulted in 10 to 15 mmol/kg decreases in carbonate ion concentration. The dissolved inorganic carbon d13C values were similar in the 2001 and 2002 experiments, but d13C(DIC) showed significant variations during the course of each experiment (Fig. 4). The slow variation in d13C(DIC) was not recognized during the 2001 experiment because sample
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
3300
3300
3200
3200
Alkalinity DIC
2900
2800
6/1/2001
2900
9/1/2001
12/1/2001
DIC (μmol/kg)
Alkalinity (μeq/kg)
(a)
139
2800 3/1/2002 2900
(b)
2800 2900 2700
DIC (μmol/kg)
Alkalinity (μeq/kg)
3000
2800 6/1/2002
9/1/2002
12/1/2002
2600 3/1/2003
Date Fig. 2. Time series of the alkalinity (Alk, solid diamonds) and dissolved inorganic carbon (DIC, dotted diamonds) concentrations of the culture system water during the 2001 experiment (a) and 2002 experiment (b). Samples for Alk/DIC titrations were not collected until the second half of the 2001 experiment.
preservation problems early in this experiment meant that no reliable samples were collected prior to November 2001. However, the 2002 data clearly show a strong seasonal cycle (0.7‰, peak to trough), with enriched values during the summer and fall (June through September), and more depleted values during the winter (November through February). The seasonal pattern in culture system d13C(DIC) values reflects ongoing isotopic equilibration of the DIC with the CO2 in the air used to bubble the culture system. Air from outside the building was continuously bubbled through the culture system reservoir, to set the system’s dissolved O2 and CO2 partial pressures (Hintz et al. 2004). High-frequency (e.g. diurnal and air-mass driven) variations in the d13C of atmospheric CO2 were strongly damped by the large pool of DIC in the 1600 L reservoir, and the seasonal variations in the pCO2 of the outside air were small enough that changes in culture system carbonate chemistry through the course of each experiment were minimal (Figs 2 and 3; Hintz et al. (2004)). However, the several-week isotopic residence time of the DIC pool is short enough
that the DIC isotopic composition tracks the annual cycle of atmospheric d13C at continental sites in the eastern United States (Bakwin et al. 1998; see also the NOAA/ESRL Global Monitoring Division web site, http://www.esrl.noaa.gov/ gmd/ccgg/iadv/). Below, we account for the systematic changes in DIC d13C using the estimated growth history of the cultured foraminifera.
Foraminiferal abundances, and C and O isotopic compositions The faunal results of these experiments (initial and final abundances, by species) were discussed by Hintz et al. (2004). The experiments were terminated after culture periods ranging from 84 to 120 days in the 2001 experiment, and from 119 to 216 days in the 2002 experiment. Most culture chambers had experienced foraminiferal reproduction during the experiments, with final abundances of .75 mm individuals of one or more species a factor of ten higher than their initial values. Thus, by the end of each experiment, most culture
140
D. C. MC CORKLE ET AL.
Carbonate ion (μmol/kg)
(b)
240
8.3
230 220 8.2
Calculated CO32–
210
pH (total scale)
250
Calculated pH
200 6/1/2001
9/1/2001
12/1/2001
3/1/2002
250 240
8.2
230 220 8.1
210 200 6/1/2002
9/1/2002
12/1/2002
pH (total scale)
Carbonate ion (μmol/kg)
(a)
3/1/2003
Date Fig. 3. Time series of carbonate ion concentration (solid diamonds) and pH (dotted diamonds) for the 2001 and 2002 experiments ((a) and (b), respectively). Carbonate ion and pH were calculated from the alkalinity and DIC data using the program CO2SYS (Lewis & Wallace 1998) with the carbonate dissociation constants of Roy et al. (1993), the calcite solubility of Mucci (1983), and the assumption that the boron/salinity ratios were equal to the seawater value.
6/1/2001 1.0
8/1/2001
10/1/2001
12/1/2001
2/1/2002
10/1/2002
12/1/2002
2/1/2003
δ13C ‰
0.5
0.0
–0.5 2002 Experiment 2001 Experiment –1.0 6/1/2002
8/1/2002
Fig. 4. The d13C values of culture system DIC during the 2001 experiment (dotted diamonds, top axis) and 2002 experiment (black diamonds, bottom axis). The d13C(DIC) decrease from late summer into winter reflects the natural annual cycle in the d13C of CO2 in the atmosphere, which was used to aerate the culture medium reservoir.
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
chambers contained hundreds of individuals whose entire tests had been formed under known, controlled physical and chemical conditions. As a result of this successful foraminiferal reproduction, we did not need to use laser dissection methods (Hintz et al. 2006b) to separate the final chambers, grown during the culture experiments, from the pre-experimental, calcein-stained chambers of the specimens used to inoculate each culture chamber. Instead, C and O isotopic analyses were carried out on pooled whole specimens of culture chamber-born individuals (Table 1). Bulimina aculeata was the most abundant species in the 2001 culture chambers, and only this species was analysed in 2001. In 2002, both B. aculeata and Rosalina vilardeboana reproduced in culture and were abundant in most culture chambers at
141
the end of the experiment, and both species were analysed. The foraminiferal stable isotope data display several striking features (Fig. 5). These features, each of which is addressed in detail in the Discussion, include: 1. 2. 3. 4.
A systematic covariation between d13C and d18O values in both species, in both years; A roughly 1‰ d18O offset between 2001 B. aculeata and 2002 B. aculeata; Substantial interspecies d18O and d13C offsets between B. aculeata and R. vilardeboana in the 2002 data; Consistent isotopic offsets between large and small specimens, with larger specimens of both species having ‘heavier’ (more enriched)
Table 1a. Bulimina aculeata, 2001 experiment Culture chamber
d13C (DIC)*
Large specimens 13
d C 2 2 2 2 2 3 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 6 6 6 6 8 8 8 8 8 8
0.22 0.22 0.22 0.22 0.22 0.31 0.31 0.22 0.22 0.31 0.31 0.31 0.31 0.31 0.31 0.31 0.22 0.22 0.22 0.31 0.31 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22 0.22
20.59 20.45 20.44 20.42 20.39 20.39 20.46 20.41 20.28
20.55 20.47
18
d O
22.08 22.34 22.33 22.24 22.34 22.33 22.32 21.85 22.20
22.17 22.34
13 †
Dd C
20.90 20.76 20.75 20.73 20.70 20.70 20.77 20.72 20.59
20.86 20.78
Small specimens 18
Dd O
‡
13
d C
d18O
Dd13C
Dd18O
20.68 20.68 20.70 20.61 20.81
22.35 22.46 22.48 22.31 22.44
20.90 20.90 20.92 20.83 21.03
20.18 20.29 20.31 20.14 20.27
20.62 20.69
22.41 22.43
20.84 20.91
20.24 20.26
20.67 20.60 20.55
22.42 22.43 22.41
20.89 20.82 20.77
20.25 20.26 20.24
20.72 20.75 20.82 20.86 20.86 20.82 20.65 20.80 20.78 20.78 20.88 20.95
22.40 22.35 22.41 22.45 22.43 22.40 22.26 22.44 22.41 22.52 22.48 22.43
20.94 20.97 21.04 21.08 21.08 21.04 20.87 21.02 21.00 21.00 21.10 21.17
20.23 20.18 20.24 20.28 20.26 20.23 20.09 20.27 20.24 20.35 20.31 20.26
0.09 20.17 20.16 20.07 20.17 20.16 20.15 0.32 20.03
0.00 20.17
*d13C(DIC) is the weighted culture d13C; all isotopic values are ‰ (VPDB). † The Dd13C values are the difference between shell d13C and weighted culture d13C. ‡ The Dd18O values are the difference between shell d18O and the estimated d18O of equilibrium calcite, which was 22.17‰ in 2001 and 21.26 in 2002.
142
D. C. MC CORKLE ET AL.
Table 1b. Bulimina aculeata, 2002 experiment Culture chamber 5 5 5 5 5 6 5,6 5,6 7 7
Size range (mm)
d13C (DIC)
d13C
d18O
Dd13C
Dd18O
200 –300 300 –400 320 –400 .400 .400 200 –300 ,200 ,200 200 –300 200 –300
0.05 0.14 0.14 0.16 0.16 0.05 0.00 0.00 0.36 0.36
20.76 20.19 20.53 20.51 20.43 20.70 21.12 20.77 20.98 20.53
21.27 21.12 21.22 21.23 21.21 21.21 21.37 21.38 21.37 21.35
20.81 20.33 20.67 20.67 20.59 20.75 21.12 20.77 21.34 20.89
20.01 0.14 0.04 0.03 0.05 0.05 20.11 20.12 20.11 20.09
Table 1c. Rosalina vilardeboana, 2002 experiment Culture chamber 2 2,12 2,12 2,12 2,4 4 4 4 4 4,12 4,12 4,12 5 5 5 5 5 5 5 5 5,6 5,6 6 mixed
Size range (mm)
d13C (DIC)*
d13C
d18O
Dd13C
Dd18O
240 –380 200 –300 300 –400 .400 400 –500 300 –400 300 –400 .400 400 –500 200 –300 250 –300 .400 200 –300 300 –400 300 –400 .400 .400 .400 400 –500 400 –500 200 –300 200 –300 300 –400 300–.400
0.13 0.05 0.14 0.16 0.16 0.14 0.14 0.16 0.16 0.05 0.13 0.16 0.05 0.14 0.14 0.16 0.16 0.16 0.16 0.16 0.05 0.05 0.14 0.14
21.66 22.11 21.72 21.31 21.43 21.80 21.74 21.40 21.63 22.05 21.96 21.40 22.26 21.82 21.58 21.34 21.27 21.26 21.41 21.41 22.08 22.57 21.56 21.95
21.61 21.70 21.70 21.61 21.58 21.74 21.66 21.65 21.61 21.76 21.81 21.65 21.91 21.69 21.63 21.60 21.51 21.43 21.50 21.54 21.77 21.95 21.64 21.74
21.79 22.16 21.86 21.47 21.59 21.94 21.88 21.56 21.79 22.10 22.09 21.56 22.31 21.96 21.72 21.50 21.43 21.42 21.57 21.57 22.13 22.62 21.70 22.09
20.35 20.44 20.44 20.35 20.32 20.48 20.40 20.39 20.35 20.50 20.55 20.39 20.65 20.43 20.37 20.34 20.25 20.17 20.24 20.28 20.51 20.69 20.38 20.48
*The weighted culture d13C values for R. vilardeboana are based on growth data for B. aculeata.
5.
C and O isotopic compositions in both years; and Strongly depleted d13C values for both species, relative to the culture system d13C(DIC) values.
Discussion C and O isotopic correlation We observe a systematic covariation between d13C and d18O values in both foraminiferal species in
both years (Fig. 5). Carbonate standards spanning the full weight range of these foraminiferal samples show no sign of isotopic fractionation, and the foraminiferal data have d18O:d13C slopes of roughly 0.5, rather than the slope of 2 that would suggest isotopic fractionation of CO2 during extraction or analysis. Thus, we do not believe these isotopic trends reflect an analytical problem. Covariation between d13C and d18O values in benthic foraminifera is well known, and is generally attributed to ‘vital effects’ such as variations in
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
1981; Grossman 1984; Schmeidl et al. 2004). In the experimental data presented here, there is a strong correlation between shell size and C and O isotopic composition (discussed below), which suggests that the overall C and O isotopic covariation in the cultured foraminifera reflects ontogenetic changes in foraminiferal calcification.
(a) 0.0
δ13C (foram) ‰
large –0.5
–1.0
–1.5
Year to year d18O offset
small
2001
2002
B. aculeata –2.5
–2.0
–1.5
–1.0
(b) –1.0
–1.5 δ13C (foram) ‰
143
The roughly 1‰ d18O offset between 2001 B. aculeata and 2002 B. aculeata (Fig. 5a) reflects the increase in the d18O(water) of the culture system that resulted from replacing approximately 480 L of the 2001 culture system water with filtered Gulf Stream seawater prior to the 2002 experiment. Below, we explicitly account for interannual and within-experiment variations in the C and O isotopic composition of the culture medium by expressing the foraminiferal isotopic data in terms of differences from culture system DIC (for d13C) and from the equilibrium calcite d18O value calculated for the experimental temperature and d18O(water).
–2.0
Interspecies isotopic comparison –2.5 R. vilardeboana - 2002 –3.0
–2.5
–2.0
–1.5
–1.0
δ18O (foram) ‰ Large (>2 μg) – 2001 Small (<2 μg) – 2001 <200 μm – 2002 200–300 μm – 2002 300–400 μm–2002 >400 μm – 2002
Fig. 5. Cross-plots of the d13C values and d18O values of cultured benthic foraminifera. The Bulimina aculeata data (2001 and 2002 experiments (a)) and the Rosalina vilardeboana data (2002 experiment only, (b)) display a strong covariation between d13C and d18O values. In the 2001 B. aculeata data, larger specimens (.2 mg/shell, solid symbols) have higher C and O isotopic values than small specimens (,2 mg/shell, dotted symbols). Size fraction data from B. aculeata in the 2002 experiment show a similar size-linked trend. Size fraction data for R. vilardeboana in the 2002 experiment show stronger depletions in both 18O and 13C, and size-linked trends similar to B. aculeata.
kinetic fractionation during calcification, or variable incorporation of metabolic CO2 (Woodruff et al. 1980; Belanger et al. 1981; Graham et al.
We observe substantial interspecies d18O and d13C offsets between B. aculeata and R. vilardeboana in the 2002 data. By comparing benthic foraminiferal species co-cultured in culture chambers designed to minimize the influence of chemically distinct microenvironments, we can distinguish species-specific isotopic fractionation (‘vital effects’) from interspecies isotopic differences caused by microhabitat selectivity of field populations. The existence of species-specific vital effects has been inferred from the d18O differences between coexisting species, since microhabitat differences in temperature and d18O(water) are likely to be negligible (McCorkle et al. 1990; Rathburn et al. 1996; McCorkle et al. 1997; Schmeidl et al. 2004). The R. vilardeboana specimens show larger depletions in both 13C and 18O than co-cultured B. aculeata, and a stronger sizedependence in their isotopic values (below). The isotopic offsets between co-cultured B. aculeata and R. vilardeboana demonstrate that speciesdependent vital effects can exert a strong influence on benthic foraminiferal isotopic composition.
Ontogenetic trends in isotopic values Size-related isotopic trends are observed within each species (Fig. 5). Consistent isotopic offsets between large and small specimens are seen in both C and O isotopes, in both species, and in both years. In the 2001 experiment the large-small distinction was based on the visual identification
144
D. C. MC CORKLE ET AL.
of rough size classes, and the 2 mg weight cut-off reflects an arbitrary, after-the-fact division based on average shell weights. The 2002 specimens were individually measured and separated to provide isotopic data from four size fractions of B. aculeata and three size fractions of R. vilardeboana (Fig. 5). The 2002 B. aculeata data show more clearly the trend of increasing d13C and d18O with increasing size that was suggested by the 2001 B. aculeata large-small comparison. The data from R. vilardeboana suggest that the isotopic covariation occurs within size fractions, as well as between them, but there are not enough sizefraction B. aculeata data to determine whether the same is true for this species, as well. These size-linked changes in foraminiferal isotopic composition are too large to be explained by the observed temporal variations in the culture system during the experiments (temperature, d13 C(DIC), d18O(water), and carbonate chemistry, Figs 1–4). Instead, the cultured foraminiferal isotopic data suggest that there are significant ontogenetic variations in the relationships between environmental conditions and benthic foraminiferal shell chemistry. Ontogenetic variations in the Mg/ Ca of individual B. aculeata from these experiments have been reported by Hintz et al. (2006b). Previous field-based studies of size-dependent trends in benthic foraminiferal isotopic values have been inconclusive; some studies report no significant relationship between benthic foraminiferal test size and C and O isotopic composition (Grossman 1984; Corliss et al. 2002), while others do report size-linked isotopic trends (Dunbar & Wefer 1984; Schmeidl et al. 2004). An ontogenetic effect is most evident in the smaller size fractions studied by Schmeidl et al. (2004), and hints of this pattern also can be seen in the smallest size fraction analysed by Corliss et al. (2002). We note that even the ‘large’ specimens grown in these culture experiments (.2 mg/shell) are smaller and more lightly calcified than typical field conspecifics. The culture data confirm that ontogenetic effects influence benthic foraminiferal d13C and d18O values. However, the data also suggest that ontogenetic effects may not be a serious problem in palaeoceanographic studies of foraminiferal C and O isotopic composition, as long as the benthic foraminifera used in downcore reconstructions are picked from the larger size fractions, where isotopic values are relatively constant (Schmeidl et al. 2004).
Controls on the C and O isotopic composition of cultured B. aculeata Expected foraminiferal carbon isotopic composition. The d13C values of both foraminiferal
species were depleted relative to the d13C values of the dissolved inorganic carbon in the culture system, R. vilardeboana more strongly than B. aculeata. We focus the remaining discussion on B. aculeata because we have growth rate estimates, and isotopic analyses of field specimens, only for this species, and because B. aculeata isotopic composition has been discussed in core top calibration studies (McCorkle et al. 1997) and has been used in palaeoceanographic studies (e.g. Stott et al. 2002; Berelson & Stott 2003; Holsten et al. 2004). To account for the impact of the observed culture system d13C(DIC) cycle on the shell d13C of cultured foraminifera, we estimated the d13C(DIC) of the culture medium as a function of the time of shell growth (Fig. 6). To do so, we estimated shell growth as a function of time, and used these estimates together with the measured d13C(DIC) during the experiments to calculate a weighted average d13C(DIC) for shells of different final sizes. The start and end dates of each culture chamber in each experiment are shown in Figure 6a. We do not know when the foraminifera in each culture chamber reproduced, or their growth history during the experiments, so we assume that growth in each culture chamber continued until the date on which the culture chamber was harvested. Using data from earlier growth experiments with B. aculeata (not shown), we estimated the timing of chamber addition, and the mass added for each successive chamber. Earlier chambers are smaller, and are added more frequently; later chambers are larger, and are added less often. Using data for culture chamber 12 in the 2002 experiment as an example, the percentages reported in Figure 6a indicate the fraction of the final shell mass, for both large and small specimens, which we believe was deposited during each month prior to the harvesting of the culture chamber. In this example, 36% of the carbonate in small specimens was added in the final month of the experiment (January 2003), 28% and 36% in the preceding two months (December and November 2002, respectively), and none prior to November 2002. For large specimens, we estimate that 27%, 21% and 17% of the final mass was deposited in January 2003, December 2002, and November 2002, respectively, and that 36% of the final mass had been deposited prior to November 2002. The resulting estimates of the mass-weighted d13C(DIC) are þ0.04 for small specimens in culture chamber 12 in 2002, and þ0.16 for large specimens in the same culture chamber. The higher d13C value for the large specimens reflects the greater contribution of carbonate deposited earlier in the experiment, when the d13C of the chamber DIC was higher. Similar calculations were carried out for large and small size fractions
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
6/1/2001
(a)
9/1/2001
12/1/2001
145
3/1/2002
Microcosm ID #
2001
2002
δ13C
CC 2002-12
6/1/2002
‰
Large
<==36% =
17%
21%
27%
+0.16
Small
<== 0% =
36%
28%
36%
+0.04
9/1/2002
12/1/2002
3/1/2003
12/1/2002
3/1/2003
(b) 1.0
δ13C ‰
0.5
0.0
–0.5
–1.0
2002 Experiment 2001 Experiment 6/1/2002
9/1/2002
Fig. 6. The start and end times for each culture chamber experiment (a), and the time course of culture system d13C(DIC) together with a simple three-segment linear approximation of the time series (b), are used to estimate the growth-weighted d13C(DIC) for each culture chamber. These estimates rely on assumptions about ontogenetic changes in calcite addition (mass per chamber, and days to add each chamber, each as a function of chamber number), as discussed in the text. The percentages in (a) show examples of these mass fraction estimates, for large and small B. aculeata in culture chamber CC-2002– 12 (the bottom-most line in (a), with monthly tick marks). In this example, the growth-weighted d13C(DIC) values used as reference values for large and small B. aculeata are þ0.16 and þ0.04‰, respectively.
for each experiment, taking into account the harvest date of each culture chamber. The resulting weighted d13C(DIC) values were either þ0.22‰ or þ0.31‰ for all 2001 specimens, and ranged from 0.0‰ to þ0.16‰ for all 2002 specimens except those from culture chamber 7 (harvested in November), where the weighted d13C(DIC) value was þ0.36‰ (Table 1). Our foraminiferal growth calculations are based on estimates of chamber addition rates and of the mass addition per chamber. These rates are poorly
constrained, and are likely to respond to biological and ecological factors that may have differed between the initial B. aculeata growth experiments and the 2001 and 2002 culture chamber experiments described here. As a result of these assumptions, the fraction added vs. time estimates (percentages in Fig. 6a) are quite crude. Future studies could document the chamber addition vs. time during culture chamber experiments, and improve the foraminiferal mass vs. chamber number relationships for different species. However, the estimated
D. C. MC CORKLE ET AL.
corrections to the culture system d13C(DIC) values are all relatively small, and thus do not have a strong impact on the foraminifera : d13C(DIC) comparisons below. Expected foraminiferal oxygen isotopic composition. We calculate d18O offsets between foraminiferal calcite and the d18O values predicted for calcite in equilibrium with culture system d18O(water) at the average culture temperature (7 8C). The observed variation in the culture system d18O(water) values during each experiment (Fig. 1) is much smaller than the seasonal cycle in d13C(DIC) values (Figs 4 and 6b), and most of the 2002 d18O(water) shift occurred early in the experiment, before we expect significant foraminiferal calcification to have taken place (Fig. 1). We therefore used the average d18O(water) value for each experiment (d18O(water) ¼ 24.32‰ in 2001 and 23.39‰ in 2002) to calculate the equilibrium calcite d18O. Several expressions have been used to describe the difference between d18O(calcite) and d18O(water) (often abbreviated as (dc – dw)) as a function of temperature (e.g. Shackleton 1974; Erez & Luz 1983; Kim & O’Neil 1997; Bemis et al. 1998). The Shackleton (1974) benthic foraminiferal palaeotemperature equation is based on core-top Uvigerina data, the equations of Erez & Luz (1983) and Bemis et al. (1998) are based on laboratory cultured planktic foraminifera, and the equation of Kim & O’Neil (1997) is based on inorganic precipitation studies. In a study of the isotopic composition of Rose Bengal stained (live, or recently living) core-top benthic foraminifera, McCorkle et al. (1997) used a palaeotemperature equation based on earlier inorganic precipitation studies of O’Neil et al. (1969) and Friedman & O’Neil (1977): n 2 3 3 d18 O(calcite;SMOW) ¼ e((2:7810 =T )(2:89=10 )) o (d18 O(water) þ 1000) 1000
(1)
with d18 O(PDB) ¼ (0:97006 d18 O(SMOW) ) 29:94
(2)
At the culture temperature of this study, 7 8C, equations (1) and (2) predict an equilibrium (dc – dw) of 2.14‰, roughly midway between the prediction of the Erez & Luz (1983) equation (2.25‰) and the predictions of the Kim & O’Neil
(1997) and Bemis et al. (1998) equations (2.04‰ and 1.98‰, respectively); the 7 8C Shackleton (1974) prediction is considerably higher (2.48‰). In the discussion below, we use the McCorkle et al. (1997) equations (1 and 2, above), to facilitate comparison with our earlier Rose Bengal benthic foraminiferal d18O data. Carbon and oxygen isotopic offsets from culture medium. The Dd18O values (d18O offsets between foraminiferal tests and predicted equilibrium calcite) for cultured B. aculeata are near zero in both experiments (Fig. 7). The Dd13C values (d13C offsets between foraminiferal tests and d13C(DIC) of the culture medium) of cultured B. aculeata range from 20.3 to nearly 21.5‰, although most cultured Dd13C values fall between 20.5 and 21.2‰ (Fig. 8). Foraminifera from replicate culture chambers display different average Dd13C and Dd18O values (Figs 7 and 8). As was seen in the raw d13C and d18O data, both the carbon and oxygen isotopic offsets are greater for
0.5
Δδ18O (foram-eq.cal.) ‰
146
Bulimina aculeata
0.0
–0.5
2001 –1.0
2002
0.7 0.2 km km 0.6–0.8 km
Microcosm Field Large (>2 μg) Small (<2 μg) 2001 field specimens McCorkle et al. 1997
Fig. 7. The d18O offsets from equilibrium calcite of the cultured B. aculeata (diamonds), plotted by culture chamber, for the 2001 and 2002 experiments. The Dd18O values are negative when the foraminifera are depleted in 18O relative to the value predicted from culture system temperature and d18O(water). Small specimens (less than 2 mg per shell, dotted symbols) are more strongly depleted than large specimens (more than 2 mg per shell, solid symbols) from the same culture chamber. The Dd18O values of cultured B. aculeata are similar to the Dd18O values of live (CellTracker Green, this study (circles)) and Rose Bengal stained (McCorkle et al. (1997) (hexagons)) field specimens.
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
Δδ13C (foram-DIC) ‰
0.0
Bulimina aculeata
–0.5
–1.0
–1.5
2001
2002
0.7 0.2 0.6–0.8 km km km
Microcosm
Field
Large (>2 μg) Small (<2 μg) 2001 field specimens McCorkle et al. 1997 Fig. 8. The d13C offsets from culture system d13C(DIC) of the cultured B. aculeata (diamonds), plotted by culture chamber, for the 2001 and 2002 experiments. The d13C values of cultured B. aculeata are all lower than the d13C of the DIC in the reservoir water, by amounts ranging from 0.4 to more than 1.0‰. The isotopic offsets are consistently greater for smaller specimens (less than 2 mg per shell, dotted diamonds) than for large specimens (more than 2 mg per shell, solid diamonds) from the same culture chamber. The d13C offsets in the culture experiments approach the magnitude of Dd13C values for live (CellTracker Green, this study (circles)) and Rose Bengal stained (McCorkle et al. (1997) (hexagons)) field specimens. We believe these comparable overall offsets have different explanations: pore water microhabitat effects on the field specimens, and carbonate ion concentration effects on the cultured specimens.
smaller specimens than for large ones. The consistency of the relationship between isotopic composition and shell size (Fig. 5) suggests that the isotopic differences between culture chambers may simply reflect differing size distributions of specimens within each size class in each culture chamber, although we do not have the data needed to test this hypothesis. The Dd18O values for cultured B. aculeata are not significantly different from the d18O offsets observed in the live (CellTracker Green labeled) B. aculeata field samples from this study, and in the Rose Bengal stained specimens from the North Carolina margin sites of McCorkle et al. (1997) (Fig. 7). The field Dd18O values are calculated as the d18O offsets between live (stained) core-top specimens and the d18O values predicted for bottom water temperature and d18O(water). Likewise, the Dd13C values in the culture experiments are similar to the
147
Dd13C values for field specimens of B. aculeata, although the most strongly depleted field specimens have lower Dd13C values than all but one of the cultured B. aculeata analyses (Fig. 8). Here, the field Dd13C values give the d13C offsets between live core-top specimens and bottom water DIC. The agreement between culture and field B. aculeata Dd13C values is puzzling, given the very different environmental conditions experienced by field and cultured B. aculeata. We hypothesize that these comparable overall d13C offsets have two different causes: pore water microhabitat effects on the field specimens, and carbonate ion concentration or pH effects (Spero et al. 1997; Bijma et al. 1999), due to high culture medium alkalinity, on the cultured specimens. There is considerable evidence suggesting that pore water composition influences the carbon isotopic composition of infaunal benthic foraminifera. The d13C values of pore water dissolved inorganic carbon are lower than bottom water values as a result of the release of 13C-depleted CO2 by organic matter decomposition in surface sediments (Grossman 1984; McCorkle et al. 1985; Sayles & Curry 1988). Determinations of the vertical distribution of Rose Bengal stained benthic foraminifera within the top few cm of the sediments (e.g. Corliss 1985; Mackensen & Douglas 1989), and comparison of these distributions and associated benthic foraminiferal d13C values with pore water d13C profiles, demonstrated that infaunal species have lower d13C values than epibenthic species (Grossman 1984, 1987; McCorkle et al. 1990; Rathburn et al. 1996; McCorkle et al. 1997; Mackensen et al. 2000). The clear correspondence between benthic foraminiferal Dd13C values (the carbon isotopic difference between foraminifera and bottom water DIC), foraminiferal habitat depth, and pore water d13C in these studies suggests that the d13C offsets are a pore water microhabitat effect, with the shell chemistry of infaunal species reflecting the low d13C values of DIC in the pore water calcification environment. Core-top pore water d13C gradients can change in response to the decomposition rate of sediment organic matter, which is largely driven by the rain rate (flux) of particulate organic matter to the sea floor. This carbon flux effect on pore water d13C has been proposed to control variations in the carbon isotopic composition of shallow-dwelling (top 1– 2 cm) infaunal benthic foraminifera such as Uvigerina, Bulimina, and Bolivina species, including B. aculeata (Woodruff & Savin 1985; Zahn et al. 1986; McCorkle et al. 1990, 1997; Mackensen et al. 2000; Stott et al. 2002; Berelson & Stott 2003; Holsten et al. 2004; Mackensen & Licari 2004). We believe that this pore water
148
D. C. MC CORKLE ET AL.
microhabitat effect is the principal cause of the low d13C values in field B. aculeata (Fig. 8). Culture studies of planktic foraminifera have shown that their d18O and d13C values decrease as the carbonate ion concentration and/or pH of their growth medium increases (Spero et al. 1997; Lea et al. 1999; Bjima et al. 1999). The slopes of these isotopic responses were species-dependent, and were approximately three times steeper for d13C than for d18O in the two planktic species was 20.006‰/ studied (d(d13C)/d([CO22 3 ]) mmol/kg for Orbulina universa and approximately 20.013‰/mmol/kg for Globigerina bulloides, and d(d18O)/d([CO22 3 ]) was 20.002‰/mmol/kg for O. universa and approximately 20.0045‰/ mmol/kg for G. bulloides (Spero et al. 1997)). The low d18O values in high-carbonate ion planktic cultures have been proposed to reflect kinetic isotope effects (Spero et al. 1997), or a shift in the average d18O of DIC as the relative proportions of carbonate ion and bicarbonate ion vary in response to changing pH (Zeebe 1999). The low d13C values of planktic foraminifera from high-carbonate ion experiments have been proposed to reflect kinetic isotope effects, or more rapid conversion of metabolic CO2 to bicarbonate ion in high [CO22 3 ], high pH solutions, resulting in greater incorporation of 13C-depleted metabolic carbon into the foraminiferal tests (Spero et al. 1997; Bjima et al. 1999; Wolf-Gladrow et al. 1999; Zeebe et al. 1999). The carbonate ion concentration of the culture medium (230– 240 mmol/kg in 2001, 205 –225 mmol/kg in 2002) was substantially higher than the in-situ bottom water carbonate ion concentration at the sites where the field samples were collected (100 to 115 mmol/kg). If the d(d18O)/ d ([CO22 3 ]) slope for B. aculeata is similar to the O. universa slope observed by Spero and co-workers (1997) (20.002‰/mmol/kg), the þ100 to þ140 mmol/kg elevation of culture system carbonate ion relative to bottom water carbonate ion would lower the cultured B. aculeata d18O values by 0.2 to 0.3 ‰ relative to the offset between field specimens and calcite in equilibrium with bottom water. An analogous calculation for d13C (i.e. assuming that the d (d13C)/d ([CO22 3 ]) slope for B. aculeata is 20.006‰/mmol/kg) would predict a 0.6 to 0.85‰ decrease in the Dd13C values of cultured B. aculeata. If the Spero et al. (1997) slopes for G. bulloides are closer to the true B. aculeata values, we would expect the Dd18O and Dd13C decreases to be roughly twice as large (20.4 to 20.7‰ for oxygen, and 21.2 to 21.9‰ for carbon). The observed Dd13C values for cultured B. aculeata, 20.5 to 21.2‰, are consistent with the O. universa-based predictions and with the low end of the G. bulloides-based predictions, and the Dd18O values of small B. aculeata from the
2001 experiment are likewise consistent with the O. universa-based prediction (a d18O lowering of 0.2 to 0.3‰), and are not inconsistent with the low end of the G. bulloides-based d18O predictions. Because the culture chambers were designed to minimize the formation of low-d13C microhabitats, we believe that a carbonate ion concentration effect on benthic foraminiferal isotopic composition caused the low d13C values of cultured B. aculeata. However, the large-specimen B. aculeata data from the 2001 experiment, and all of the 2002 B. aculeata data, show little or no 18O depletion, with Dd18O values near zero. Thus the Dd13C values of the cultured B. aculeata are consistent with a carbonate ion or pH effect on the shell chemistry of this species, while the Dd18O values of the cultured B. aculeata do not require such an effect. Implications for interspecies offsets in benthic foraminiferal d18O values. Since the earliest determinations of the isotopic composition of benthic foraminifera, the d18O offset between Cibicidoides species and Uvigerina species has puzzled palaeoceanographers (Shackleton 1974). The d18O values of U. peregrina are typically about 0.64‰ higher than values from coexisting C. wuellerstorfi, although offsets ranging from 0.5 to 1.0‰ have been reported (e.g. Herguera et al. 1992). Studies of the isotopic composition of Rose Bengal stained benthic foraminifera have shown that while some other species have d18O values similar to Cibicidoides species (e.g. Elphidium excavatum), many more benthic foraminiferal species have d18O values similar to Uvigerina (McCorkle et al. 1990; Rathburn et al. 1996; McCorkle et al. 1997; Mackensen et al. 2000). Shackleton (1974) defined a benthic foraminiferal palaeothermometer by assuming that Uvigerina d18O values reflect equilibrium with bottom water d18O(water), and suggested that the lower Cibicidoides d18O values reflected a disequilibrium ‘vital effect’. The broad agreement between Uvigerina d18O values and the d18O:temperature relationship observed in cultured planktic foraminifera (Globigerinoides sacculifer)(Erez & Luz 1983) has also been used to suggest that Uvigerina d18O values reflect equilibrium precipitation (Zahn & Mix 1991). More recently, however, Bemis et al. (1998) showed that a d18O palaeothermometer calibrated using planktic foraminifera cultured between 15 8C and 25 8C (their Equation 1, for Orbulina universa grown at low light levels) fit the low-temperature d18O data for field Cibicidoides, and was likewise consistent with inorganic precipitation data from Kim & O’Neil (1997). This three-way agreement led Bemis et al. to conclude that Cibicidoides d18O values reflect equilibrium calcification. They suggested that the
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
(a)
Δδ13C ‰
0
–1
–1
0
1
0
1
(b)
Δδ13C ‰
0
–1
–1
Δδ18O ‰ McCorkle et al. 1997 Shackleton (1974)
Fig. 9. (a) Cross plot of cultured B. aculeata (both years) Dd13C and Dd18O values. The carbon isotopic offsets are calculated relative to culture system DIC, and are the same for both sets of points. The oxygen isotopic offsets are calculated relative to the equilibrium calcite d18O predictions of McCorkle et al. (1997) (solid diamonds) and Shackleton (1974) (open circles). Calculations relative to the equilibrium expression of Bemis et al. (1998) (not shown) would plot shifted to the right of the solid diamonds. The line shows the trend in isotopic offsets predicted by the planktic carbonate ion effect of Spero et al. (1997). The white hexagon in the upper right shows no offsets, the middle hexagon shows the predicted offsets for a 100 mmol/kg elevation of carbonate ion using the O. universa relationship, and the lower left hexagon shows the predicted offsets for the same elevation of carbonate ion using the G. bulloides relationship. (b) Cross plot of field B. aculeata Dd13C
149
higher d18O values of Uvigerina species reflect calcification in a pore water microhabitat with lower pH and lower [CO22 3 ] than bottom water, based on the evidence of a carbonate ion or pH effect on planktic foraminiferal isotopic composition (Spero et al. 1997; Lea et al. 1999; Bjima et al. 1999). An insightful reviewer pointed out that if we assume that the cultured benthic foraminifera from this study exhibit a carbonate ion effect similar to that seen in planktic foraminfera (Spero et al. 1997), then the culture and field B. aculeata isotopic data can be used to test these different isotopic temperature equations. To do so, we first calculate the oxygen isotopic offsets (Dd18O values) between B. aculeata and equilibrium calcite as calculated using the expression of McCorkle et al. (1997) (equations 1 and 2, above), and as calculated by Shackleton (1974). The foraminiferal carbon isotopic offsets (Dd13C values, calculated relative to culture system DIC, as discussed above) are then plotted against the Dd18O values (Fig. 9a). The two sets of points have the same Dd13C values, and the Dd18O offset between them reflects the different equilibrium predictions of McCorkle et al. (1997) (solid diamonds) and Shackleton (1974) (open circles). The line starting at 0,0 and trending to lower Dd18O and Dd13C values is predicted based on a 100 mmol/kg elevation of culture system carbonate ion concentration, using the O. universa relationship (middle point) and the G. bulloides relationship (lower left point in Fig. 9a). Use of a 7 8C equilibrium fractionation value between those of McCorkle et al. (1997) and Shackleton (1974) would cause the B. aculeata data to fall close to the line predicted by the Spero et al. (1997) carbonate ion relationship, and the carbon and oxygen data would both be consistent with the same carbonate ion effect. In contrast, use of the Bemis et al. (1998) equilibrium expression would cause the points to plot 0.15‰ Fig. 9. (Continued) and Dd18O values (this study, and McCorkle et al. 1997). The carbon isotopic offsets are calculated relative to bottom water DIC, and the oxygen isotopic offsets are calculated relative to calcite in equilibrium with bottom water using the expressions of McCorkle et al. (1997) (solid diamonds) and Shackleton (1974) (open circles). The vertical line shows the trend in isotopic offsets predicted by a pore water d13C effect only. The observed d13C depletions of pore water DIC in the 0 –0.5 cm depth interval range from 21 to 22.2‰ at the nearby sites discussed by McCorkle et al. (1997) (white hexagons, dotted line). Because pore water carbonate ion concentration is likely to be lower than bottom water values at these relatively shallow continental margin sites, a carbonate ion influence on shell composition would tend to cause Dd18O values greater than zero (ellipse).
150
D. C. MC CORKLE ET AL.
to the right of the McCorkle et al. (1997) points, in worse agreement with the carbonate ion effect-based prediction. The B. aculeata field data can be treated the same way (Fig. 9b). The calculated B. aculeata isotopic offsets (symbols) are compared with the prediction of a simple pore water influence on shell isotopic composition: lower foraminiferal Dd13C values with respect to bottom water DIC, but no change in Dd18O with respect to bottom water (the vertical line starting at zero). The oxygen isotopic offsets using the McCorkle et al. (1997) expression are slightly lower than this simple prediction, and the use of a d18O equilibrium expression intermediate between the McCorkle et al. (1997) expression and that of Shackleton (1974), as above, would make the agreement between the field data (symbols) and the prediction (vertical line) worse rather than better. In addition, if pore water carbonate chemistry also influences the isotopic composition of field specimens, then we actually might expect to see elevated Dd18O values in these field samples (ellipse in Fig. 9b). This is because pore water carbonate ion concentrations are almost certainly lower than bottom water concentrations at these relatively shallow, continental margin sites. For this to be the case – for the observed field Dd18O values to be higher than the simple equilibrium prediction line – would require an equilibrium d18O expression similar to that of Bemis et al. (1998). In summary, the cultured B. aculeata Dd13C and 18 Dd O values are consistent with no microhabitat d13C effect in culture, and a carbonate ion effect on shell isotopic composition. If the carbonate ion effect is similar to the planktic foraminifera effect (Spero et al. 1997), the most consistent results for both isotopes would be obtained using an equilibrium oxygen isotopic fractionation intermediate between the predictions of McCorkle et al. (1997) and Shackleton (1974). The isotopic offsets of the field specimens are consistent with a pore water d13C influence on shell d13C and no carbonate ion effect, and an equilibrium oxygen isotopic composition no higher than the McCorkle et al. (1997) prediction; consideration of likely trends in pore water carbonate chemistry suggests an even lower equilibrium oxygen isotopic composition, similar to the prediction of Bemis et al. (1998). Thus these culture and field data, considered in the context of the carbonate chemistry of the culture and field systems, do not yield a consistent estimate of the equilibrium fractionation factor for d18O in benthic foraminiferal calcite. Further culture experiments, including carbonate chemistry manipulations, may help resolve this discrepancy.
Conclusions The carbon and oxygen isotopic composition of benthic foraminifera grown in the laboratory under controlled, constant environmental conditions provide a valuable new perspective on the factors that influence benthic foraminiferal shell chemistry. Months-long culture experiments resulted in growth and calcification of several benthic foraminiferal species. Reproduction of two species during these experiments eliminated any contribution from preexisting (nonexperimental) carbonate, substantially simplifying the interpretation of the isotopic data. However, maintenance of culture system stability over the course of months continues to be challenging. Consistent interspecies differences in foraminiferal d13C and d18O values were observed in comparisons of B. aculeata and R. vilardeboana from the same culture chambers. Since the experiments were designed to minimize the possible influence of microhabitats on shell chemistry, these isotopic differences demonstrate the existence of speciesspecific, non-environmental vital effects on benthic foraminiferal shell chemistry. Systematic changes in test d13C and d18O values as a function of test size were observed in cultured B. aculeata and R. vilardeboana. These size-linked trends, with smaller specimens being more strongly depleted in both 13C and 18O than larger specimens, are too large to be explained by the observed changes in culture system chemistry or isotopic composition during the experiments. This suggests that ontogenetic effects on calcification influence the isotopic composition of benthic foraminifera, at least during the early stages of shell growth. Palaeoceanographic reconstructions typically make use of specimens from large size fractions, and foraminiferal shell mass is dominated by the final few chambers. As a result, the ontogenetic effects seen here may not have a large influence on palaeoceanographic proxy records, but these effects, and their possible impacts, clearly require additional study. Cultured B. aculeata show 0.5 to 1.5‰ depletions in 13C relative to the d13C of culture system dissolved inorganic carbon. These offsets are similar to the 13C depletions in field specimens relative to the d13C of bottom water DIC. This agreement between culture and field d13C offsets was unexpected, because the low d13C values in field B. aculeata are thought to reflect pore water microhabitat effects, which should be minimal in the culture chambers. We hypothesize that the relatively low d13C values in cultured B. aculeata reflect the elevated pH and [CO22 3 ] in the culture media, as has been observed in cultured planktic foraminifera. However, this hypothesis also
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA
implies that the d18O offsets between cultured B. aculeata and culture system water should be larger than the corresponding d18O offset between field B. aculeata and bottom water d18O. Instead, we observe no obvious difference between field and culture d18O offsets. This suggests that our understanding of some key factor influencing benthic foraminiferal d18O values – the field environment, the culture environment, or the impact of environment on shell chemistry – remains incomplete. This research was supported by NSF awards OCE– 9911654, OCE–0350794, and OCE –0437366. We thank the officers, crews, and science parties of the R/V Cape Henlopen and R/V Oceanus cruises. The lab work at WHOI was carried out by R. Belastock and C. Gramling. Oxygen isotopic analyses of field and culture chamber water samples were graciously provided by D. Schrag (Harvard University). JMB heartily thanks Bill Austin for the invitation to present a keynote talk on this project at the ‘Biogeochemical Controls on Palaeoceanographic Proxies’ meeting.
References B AKWIN , P. S., T ANS , P. P., W HITE , J. W. C. & A NDRES , R. J. 1998. Determination of the isotopic (13C/12C) discrimination by terrestrial biology from a global network of observations. Global Biogeochemical Cycles, 12, 555–562. B ELANGER , P. E., C URRY , W. B. & M ATTHEWS , R. K. 1981. Coretop evaluation of benthic foraminiferal isotopic ratios for paleoceanographic interpretations. Palaeogeography, Palaeoclimatology, Palaeoecology, 33, 205–220. B EMIS , B. E., S PERO , H. J., B IJMA , J. & L EA , D. W. 1998. Reevaluation of the oxygen isotopic composition of planktonic foraminifera; experimental results and revised paleotemperature equations. Paleoceanography, 13, 150– 160. B ERELSON , W. M. & S TOTT , L. D. 2003. Productivity and organic carbon rain to the California margin seafloor: modern and paleoceanographic perspectives. Paleoceanography, 18, 1002, doi:10.1029/ 2001PA000672. B ERNHARD , J. M. 2000. Distinguishing live from dead foraminifera: Methods review and proper applications. Micropaleontology, 46 (supplement 1), 38–46. B ERNHARD , J. M., B LANKS , J. K., H INTZ , C. J. & C HANDLER , G. T. 2004. Use of the fluorescent calcite marker calcein to label foraminiferal tests. Journal of Foraminiferal Research, 34, 96– 101. B IJMA , J., F ABER , W. W. & H EMLEBEN , C. 1990. Temperature and salinity limits for growth and survival of some planktonic foraminifers in laboratory cultures. Journal of Foraminiferal Research, 20, 95– 116. B IJMA , J., S PERO , H. J. & L EA , D. W. 1999. Reassessing foraminiferal stable isotope geochemistry: impact of the oceanic carbonate system (Experimental results). In: F ISCHER , G. & W EFER , G. (eds) Use of Proxies
151
in Paleoceanography: Examples from the South Atlantic. Springer-Verlag, Berlin, 489–512. B OYLE , E. A. 1988. Cadmium: chemical tracer of deepwater paleoceanography. Paleoceanography, 3, 471– 489. B OYLE , E. A. 1992. Cadmium and d13C paleochemical ocean distributions during the stage 2 glacial maximum. Annual Review of Earth and Planetary Sciences, 20, 245– 287. B OYLE , E. A. & R OSENTHAL , Y. 1996. Chemical hydrography of the South Atlantic during the last glacial maximum: d13C vs. Cd. In: W EFER , G., B ERGER , W. H., S IEDLER , G. & W EBB , D. (eds) The South Atlantic: Present and Past Circulation. SpringerVerlag, Berlin, 423– 443. B RADSHAW , A. L., B REWER , P. G., S HAFER , D. K. & W ILLIAMS , R. T. 1981. Measurements of total carbon dioxide and alkalinity by potentiometric titration in the GEOSECS program. Earth and Planetary Science Letters, 55, 99– 115. B REWER , P. G., B RADSHAW , A. L. & W ILLIAMS , R. T. 1986. Measurements of total carbon dioxide and alkalinity in the North Atlantic Ocean in 1981. In: R EICHLE , D. (ed.) The Global Carbon Cycle: Analysis of the Natural Cycle and Implications of Anthropogenic Alterations for the Next Century. SpringerVerlag, Berlin, 358– 381. C HANDLER , G. T., W ILLIAMS , D. F., S PERO , H. J. & X IAODONG , G. 1996. Sediment microhabitat effects on carbon stable isotopic signatures of microcosmcultured benthic foraminifera. Limnology and Oceanography, 41, 680– 688. C ORLISS , B. H. 1985. Microhabitats of benthic foraminifera within deep-sea sediments. Nature, 314, 435– 438. C ORLISS , B. H. & E MERSON , S. R. 1990. Distribution of Rose Bengal stained deep-sea benthic foraminifera from the Nova Scotian continental margin and Gulf of Maine. Deep-Sea Research, 37, 381– 400. C ORLISS , B. H., M C C ORKLE , D. C. & H IGDON , D. M. 2002. A time series of the carbon isotopic composition of deep-sea benthic foraminifera. Paleoceanography, 17, 1036, doi:10.1029/2001PA000664. C URRY , W. B., D UPLESSY , J.-C., L ABEYRIE , L. D. & S HACKLETON , N. J. 1988. Changes in the distribution of d13C of deep water SCO2 between the last glaciation and the Holocene. Paleoceanography, 3, 317– 341. DEPARTMENT OF ENERGY. 1994. Handbook of methods for the analysis of the various parameters of the carbon dioxide system in seawater; version 2. In: D ICKSON , A. G., G OYET , C. (eds) ORNL/CDIAC-74. D UNBAR , R. B. & W EFER , G. 1984. Stable isotope fractionation in benthic foraminifera from the Peruvian continental region. Marine Geology, 59, 215– 225. D UPLESSY , J. C., S HACKLETON , N. J., M ATTHEWS , R. K., P RELL , W., R UDDIMAN , W. F., C ARALP , M. & H ENDY , C. H. 1984. 13C record of benthic foraminifera in the Last Interglacial ocean: implications for the carbon cycle and the global deep-water circulation. Quaternary Research, 21, 225– 243. E REZ , J. & L UZ , B. 1983. Experimental paleotemperature equation for planktonic foraminifera. Geochimica et Cosmochimica Acta, 47, 1025–1031.
152
D. C. MC CORKLE ET AL.
F RIEDMAN , I. & O’N EIL , J. R. 1977. Compilation of stable isotope fractionation factors of geochemical interest (Geological Survey Professional Paper 440-KK). In: F LEISCHER , M. (ed.) Data of Geochemistry, Sixth Edition. U.S. Government Printing Office, Washington, DC, 1 –12. G RAHAM , D. W., C ORLISS , B. H., B ENDER , M. L. & K EIGWIN , L. D., J R . 1981. Carbon and oxygen isotopic disequilibria of recent deep-sea benthic foraminifera. Marine Micropaleontology, 6, 483– 497. G ROSSMAN , E. L. 1984. Carbon isotopic fractionation in live benthic foraminifera – comparison with inorganic precipitate studies. Geochimica et Cosmochimica Acta, 48, 1505–1512. G ROSSMAN , E. L. 1987. Stable isotopes in modern benthic foraminifera: a study of vital effect. Journal of Foraminiferal Research, 17, 48– 61. H AVACH , S. M., C HANDLER , G. T., W ILSON -F INELLI , A. & S HAW , T. J. 2001. Experimental determinations of trace element partition coefficients in cultured benthic foraminifera. Geochimica et Cosmochimica Acta, 65, 1277–1283. H ERGUERA , J. C., J ANSEN , E. & B ERGER , W. H. 1992. Evidence for a bathyal front at 2000 m depth in the glacial Pacific, based on a depth transect on Ontong Java Plateau. Paleoceanography, 7, 273– 288. H INTZ , C. J., C HANDLER , G. T., B ERNHARD , J. M., M C C ORKLE , D. C., H AVACH , S. M., B LANKS , J. K. & S HAW , T. J. 2004. A physicochemicallyconstrained seawater culturing system for production of viable, calcite-producing, paleoceanographicallyimportant benthic foraminifera. Limnology & Oceanography Methods, 2, 160–170. H INTZ , C. J., S HAW , T. J., C HANDLER , G. T., B ERNHARD , J. M., M C C ORKLE , D. C. & B LANKS , J. K. 2006a. Trace/minor element:calcium ratios in cultured benthic foraminifera, Part I: Inter-species and inter-individual variability. Geochimica et Cosmochimica Acta, 70, 1952– 1963, doi:10.1016/ j.gca.2005.12.018. H INTZ , C. J., S HAW , T. J., B ERNHARD , J. M., C HANDLER , G. T., M C C ORKLE , D. C. & B LANKS , J. K. 2006b. Trace/minor element:calcium ratios in cultured benthic foraminifera, Part II: Ontogenetic variation. Geochimica et Cosmochimica Acta, 70, 1964–1976, doi:10.1016/j.gca.2005.12.019. H O¨ NISCH , B., B IJMA , J., R USSELL , A. D., S PERO , H. J., P ALMER , M. R., Z EEBE , R. E. & E ISENHAUER , A. 2003. The influence of symbiont photosynthesis on the boron isotopic composition of foraminifera shells. Marine Micropaleontology, 49, 87– 96. H OLSTEN , J., S TOTT , L. & B ERELSON , W. 2004. Reconstructing benthic carbon oxidation rates using d13C of benthic foraminifers. Marine Micropaleontology, 53, 117– 132. K IM , S. T. & O’N EIL , J. R. 1997. Equilibrium and non-equilibrium oxygen isotope effects in synthetic carbonates. Geochimica et Cosmochimica Acta, 61, 3461–3476. L EA , D. & B OYLE , E. 1989. Barium content of benthic foraminifera controlled by bottom water composition. Nature, 338, 751 –753.
L EA , D. W. & B OYLE , E. A. 1990. Foraminiferal reconstruction of barium distributions in water masses of the glacial oceans. Paleoceanography, 5, 719– 742. L EA , D. W. & S PERO , H. J. 1992. Experimental determination of barium uptake in shells of the planktonic foraminifera Orbulina universa at 228C. Geochimica et Cosmochimica Acta, 56, 2673– 2680. L EA , D. W., M ARTIN , P. A., C HAN , D. A. & S PERO , H. J. 1995. Calcium uptake and calcification rate in the planktonic foraminifer Orbulina universa. Journal of Foraminiferal Research, 25, 14–23. L EA , D. W., M ASHIOTTA , T. A. & S PERO , H. J. 1999. Controls on magnesium and strontium uptake in planktonic foraminifera determined by live culturing. Geochimica et Cosmochimica Acta, 63, 2369–2379. L EAR , C. H., R OSENTHAL , Y. & S LOWEY , N. 2002. Benthic foraminiferal Mg/Ca-paleothermometry: A revised core-top calibration. Geochimica et Cosmochmica Acta, 66, 3375–3387. L EWIS , E. & W ALLACE , D. W. R. 1998. Program Developed for CO2 System Calculations. ORNL/ CDIAC-105. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U. S. Department of Energy, Oak Ridge, Tennessee. M ACKENSEN , A. & D OUGLAS , R. G. 1989. Down-core distribution of live and dead deep-water benthic foraminifera in box cores from the Weddell Sea and the California continental borderland. Deep-Sea Research, 36, 879–900. M ACKENSEN , A. & L ICARI , L. 2004. Carbon isotopes of live benthic foraminifera from the eastern South Atlantic Ocean: Sensitivity to organic matter rain rates and bottom water carbonate saturation state. In: W EFER , G., M ULITZA , S. & R ATHMEYER , V. (eds) The South Atlantic in the Late Quaternary: Reconstruction of Material Budget and Current Systems. SpringerVerlag, Berlin, 623–644. M ACKENSEN , A., H UBBERTEN , H.-W., B ICKERT , T., F ISCHER , G. & F UTTERER , D. K. 1993. d13C in benthic foraminiferal tests of Fontbotia wuellerstorfi (Schwager) relative to d13C of dissolved inorganic carbon in Southern Ocean deep water: implications for glacial ocean circulation models. Paleoceanography, 8, 587–610. M ACKENSEN , A., S CHUMACHER , S., R ADKE , J. & S CHMIDT , D. N. 2000. Microhabitat preferences and stable carbon isotopes of endobenthic foraminifera: clue to quantitative reconstruction of oceanic new production? Marine Micropaleontology, 40, 233–258. M ARCHITTO , T. M., C URRY , W. B. & O PPO , D. W. 2000. Zinc concentrations in benthic foraminifera reflect seawater chemistry. Paleoceanography, 15, 299–306. M ARTIN , P. A., L EA , D. W., R OSENTHAL , Y., S HACKLETON , N. J., S ARNTHEIN , M. & P APENFUSS , T. 2002. Quaternary deep sea temperature histories derived from benthic foraminiferal Mg/Ca. Earth and Planetary Science Letters, 198, 193– 209. M ASHIOTTA , T. A., L EA , D. W. & S PERO , H. J. 1997. Experimental determination of cadmium uptake in shells of the planktonic foraminifera Orbulina universa and Globigerina bulloides: implications for surface water paleoreconstructions. Geochimica et Cosmochimica Acta, 61, 4053– 4065.
d13C AND d18O IN CULTURED BENTHIC FORAMINIFERA M C C ORKLE , D. C. & K LINKHAMMER , G. P. 1991. Porewater cadmium geochemistry and the porewater cadmium:d13C relationship. Geochimica et Cosmochimica Acta, 55, 161–168. M C C ORKLE , D. C., E MERSON , S. R. & Q UAY , P. D. 1985. Stable carbon isotopes in marine porewaters. Earth and Planetary Science Letters, 74, 13– 26. M C C ORKLE , D. C., C ORLISS , B. N. & E MERSON , S. R. 1990. The influence of microhabitats on the carbon isotope composition of deep-sea benthic foraminifera. Paleoceanography, 5, 161– 185. M C C ORKLE , D. C., M ARTIN , P. A., L EA , D. W. & K LINKHAMMER , G. P. 1995. Evidence of a dissolution effect on benthic foraminiferal shell chemistry: d13C, Cd/Ca, Ba/Ca, and Sr/Ca results from the Ontong Java Plateau. Paleoceanography, 10, 699– 714. M C C ORKLE , D. C., C ORLISS , B. H. & F ARNHAM , C. A. 1997. Vertical distributions and Stable isotopic compositions of live (stained) benthic foraminifera from the North Carolina and California continental margins. Deep-Sea Research I, 44, 983– 1024. M UCCI , A. 1983. The solubility of calcite and aragonite in seawater at various salinities, temperatures, and one atmosphere total pressure, American Journal of Science, 283, 780–799. NOAA/ESRL Global Monitoring Division: http:// www.esrl.noaa.gov/gmd/ccgg/iadv/ N U¨ RNBERG , D., B IJMA , J. & H EMLEBEN , C. 1996. Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperatures. Geochimica et Cosmochimica Acta, 60, 803– 814. O’N EIL , J. R., C LAYTON , R. N. & M AYEDA , T. K. 1969. Oxygen isotope fractionation in divalent metal carbonates. Journal of Chemical Physics, 51, 5547– 5558. R ATHBURN , A. E., C ORLISS , B. H., T APPA , K. D. & L OHMANN , K. C. 1996. Comparisons of the ecology and stable isotopic compositions of living (stained) benthic foraminifera from the Sulu and South China Seas. Deep-Sea Research I, 43, 1617– 1646. R OSENTHAL , Y., B OYLE , E. A., L ABEYRIE , L. & O PPO , D. 1995. Glacial enrichments of authigenic Cd and U in subantarctic sediments; a climatic control on the elements’ oceanic budget? Paleoceanography, 10, 395–413. R OY , R. N., R OY , L. N., V OGEL , K. M., P ORTER M OORE , C., P EARSON , T., G OOD , C. E., M ILLERO , F. J. & C AMPBELL , D. M. 1993. The dissociation constants of carbonic acid in seawater at salinities 5 to 45 and temperatures 0 to 45 8C. Marine Chemistry, 44, 249–267. R USSELL , A. D., E MERSON , S. R., N ELSON , B. K., E REZ , J. & L EA , D. W. 1994. Uranium in foraminiferal calcite as a recorder of seawater uranium concentrations. Geochimica et Cosmochimica Acta, 58, 671 –681. R USSELL , A. D., H O¨ NISCH , B., S PERO , H. J. & L EA , D. W. 2004. Effects of seawater carbonate ion concentration and temperature on shell U, Mg, and Sr in cultured planktonic foraminifera. Geochimica et Cosmochimica Acta, 68, 4347–4361. S ANYAL , A., H EMMING , N. G., B ROECKER , W. S., L EA , D. W., S PERO , H. J. & H ANSON , G. N. 1996. Oceanic pH control on the boron isotopic composition of Foraminifera; evidence from culture experiments. Paleoceanography, 11, 513–517.
153
S AYLES , F. L. & C URRY , W. B. 1988. d13C, SCO2, and the metabolism of organic carbon in deep sea sediments. Geochimica et Cosmochimica Acta, 52, 2963– 2978. S CHMIEDL , G., P FEILSTICKER , M., H EMLEBEN , C. & M ACKENSEN , A. 2004. Environmental and biological effects on the stable isotope composition of recent deep-sea benthic foraminifera from the western Mediterranean Sea. Marine Micropaleontology, 51, 129– 152. S CHRAG , D. P., A DKINS , J. F., M C I NTYRE , K., A LEXANDER , J. L., H ODELL , D. A., C HARLES , C. D. & M C M ANUS , J. F. 2002. The oxygen isotopic composition of seawater during the Last Glacial Maximum. Quarternary Science Reviews, 21, 331– 342. S EGEV , E. & E REZ , J. 2006. Effect of Mg/Ca ratio in seawater on shell composition in shallow benthic foraminifera. Geochemistry, Geophysics, Geosystems, 7, Q02P09, doi:10.1029/2005GC000969. S HACKLETON , N. J. 1974. Attainment of isotopic equilibrium between ocean water and the benthonic foraminifera genus Uvigerina; isotopic changes in the ocean during the last glacial. In: L ABEYRIE , J. (ed./convenor) Colloques Internationaux du Centre National de la Recherche Scientifique, 219 – Les Methodes Quantitatives D’etude des Variations du Climat au Cours du Pleistocene, 203– 209. S HACKLETON , N. J. & O PDYKE , N. D. 1973. Oxygen isotope and palaeomagnetic stratigraphy of equatorial Pacific core V28-238: oxygen isotope temperatures and ice volumes on a 105 year and 106 year scale. Quaternary Research, 3, 39– 55. S PERO , H. J. & L EA , D. W. 1993. Intraspecific stable isotope variability in the planktic foraminifera Globigerinoides sacculifer: Results from laboratory experiments. Marine Micropaleontology, 22, 221– 234. S PERO , H. J., B IJMA , J., L EA , D. W. & B EMIS , B. E. 1997. Effect of seawater carbonate concentration on foraminiferal carbon and oxygen isotopes. Nature, 390, 497–500. S TOTT , L. D., B UNN , T., P ROKOPENKO , M., M AHN , C., G IESKES , J. & B ERNHARD , J. M. 2002. Does the oxidation of methane leave an isotopic fingerprint in the geologic record? Geochemistry, Geophysics, Geosystems, 3, 1012, doi:10.1029/2001GC000196. T OYOFUKU , T., K ITAZATO , H., K AWAHATA , H., T SUCHIYA , M. & N OHARA , M. 2000. Evaluation of Mg/Ca thermometry in foraminifera: Comparison of experimental results and measurements in nature. Paleoceanography, 15, 456– 464. W ILSON -F INELLI , A., C HANDLER , G. T. & S PERO , H. J. 1998. Stable isotope behavior in paleoceanographically important benthic foraminifera: results from microcosm culture experiments. Journal of Foraminiferal Research, 28, 312–320. W OLF -G LADROW , D. A., B IJMA , J. & Z EEBE , R. E. 1999. Model simulation of the carbonate chemistry in the microenvironment of symbiont bearing foraminifera. Marine Chemistry, 64, 181–198. W OODRUFF ,, F. & S AVIN , S. M. 1985. d13C values of Miocene Pacific benthic foraminifera: correlations with sea level and biological productivity. Geology, 13, 119– 122.
154
D. C. MC CORKLE ET AL.
W OODRUFF , F., S AVIN , S. M. & D OUGLAS , R. G. 1980. Biological fractionation of oxygen and carbon isotopes by recent benthic foraminifera. Marine Micropaleontology, 5, 3– 11. Z AHN , R. & M IX , A. C. 1991. Benthic foraminiferal d18O in the ocean’s temperature-salinity-density field: constraints on ice age thermohaline circulation. Paleoceanography, 6(1), 1– 20. Z AHN , R., W INN , K. & S ARNTHEIN , M. 1986. Benthic foraminiferal d13C and accumulation rates of organic
carbon: Uvigerina peregrina group and Cibicidoides wuellerstorfi. Paleoceanography, 1, 27–42. Z EEBE , R. E. 1999. An explanation of the effect of seawater carbonate concentration on foraminiferal oxygen isotopes. Geochimica et Cosmochimica Acta, 63, 2001– 2007. Z EEBE , R. E., B IJMA , J. & W OLF -G LADROW , D. A. 1999. A diffusion-reaction model of carbon isotope fractionation in foraminifera. Marine Chemistry, 64, 199–227.
Seasonal dynamics of coastal water masses in a Scottish fjord and their potential influence on benthic foraminiferal shell geochemistry ALIX G. CAGE & WILLIAM E. N. AUSTIN School of Geography & Geoscience, University of St Andrews, St Andrews, Fife KY16 9AL (e-mail:
[email protected]) Abstract: Pronounced seasonal heating of middle and high latitude shelf seas results in large temporal changes in seawater temperature that have the potential to be recorded by benthic foraminifera. Predicted oxygen isotope composition for calcite precipitating in equilibrium conditions with seawater suggest that pronounced ‘seasonal isotope effects’ may be encountered in the growth history of benthic foraminifera. Such ‘seasonal’ effects can be difficult to distinguish from so-called ‘vital effects’, where shell and equilibrium calcite values are offset by a constant difference in oxygen isotope values. Preliminary findings suggest that benthic foraminifera may have more than one phase of growth, for example Ammonia becarii calcifies in spring and late summer, potentially introducing apparent intra-annual and inter-annual temperature variations of .18C into palaeoclimatic reconstructions when mixed-season populations are sampled. We highlight the need to select species-specific palaeotemperature equations to establish reliable isotopic disequilibria and illustrate the importance of understanding the ‘seasonal isotope effect’ when considering disequilibrium effects in foraminifera which have grown in seasonallychanging environments.
Marginal marine environments, such as shelf-sea regions and fjords, are increasingly becoming the focus of palaeoclimatic research because the rapid sediment accumulation rates in these areas proffer the potential for high-resolution millennial- and centennial-scale palaeoenvironmental reconstructions (e.g. Mikalsen et al. 2001; Kristensen et al. 2004; Marret et al. 2004; Eirı´ksson et al. 2006). Additionally, these regions are typically sensitive to climatic processes often providing a link between terrestrial and deep-sea environments and thus capturing land-ocean interactions (e.g. Backhaus 1996, Scourse & Austin 2002). Shelf-sea foraminiferal assemblages are dominated by benthic foraminifera which live in or upon marine sediments (e.g. Alve & Nagy 1986, 1990; Conradsen et al. 1994; Mikalsen et al. 1999; Klitgaard-Kristensen et al. 2002; Murray 2003a; Murray et al. 2003; Scott et al. 2003; Husum & Hald 2004; Cage 2005). Indeed, planktonic foraminifera tend to be rare in most shelf sea sediments (e.g. Murray 1971). The calcium carbonate tests of foraminifera are generally wellpreserved and provide two key proxy pathways with respect to environmental parameters: (1) the biological pathway where there is a covariation between the abundance of individuals/assemblage composition and environmental variables; and (2) the chemical pathway where elements are incorporated into the shell material as the organism calcifies. One of the longest established chemical pathways widely used in palaeoceanographic research is the
stable oxygen isotopic composition of benthic foraminiferal tests, which are known to be influenced by environmental variables such as temperature and salinity (see Rohling & Cooke 1999 for a review). No previous studies have investigated isotopic composition in modern benthic foraminifera from NW Scottish coastal or fjordic environments. This study therefore aims to: (1) compare the temporal pattern of calculated equilibrium calcite at sites from NW Scotland (Loch Sunart, Fig. 1) to d18Oforam measured from different species of modern and live benthic foraminifera collected from the same sites; and (2) comment on potential seasonal effects in the stable oxygen isotopic composition of the tests.
Isotopic composition in benthic foraminifera Benthic foraminifera regularly calcify close to equilibrium with the oxygen isotopic composition of seawater, thus stable oxygen isotopic composition of foraminifera (d18Oforam) typically reflect the d18O of the seawater in which they calcified and the water temperature during calcification. The stable oxygen isotopic composition of seawater (d18Oseawater) in the coastal ocean is influenced by salinity and the regional salinity: d18O relationship, or mixing line, and can be defined as: d18 Oseawater ¼ m S c
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 155– 172. DOI: 10.1144/SP303.11 0305-8719/08/$15.00 # The Geological Society of London 2008.
(1)
156
A. G. CAGE & W. E. N. AUSTIN
14
13
12
11
10
9
8
7
6
5
59
1100 00
59
m m
58
58 Rockall Rockall
rrent
Anton AntonDhorn Dhorn Seamount Seamount
m 0000 m
tlan tic C u
56
20
56
Sco t. C o
ur
re nt
C tal as
nti a l t N. A
54
pe o l cS
C
55
t en urr
55
57
Loch Sunart
N. A
Latitude, ˚N
57
54
Porcupine Porcupine Bank Bank
m 0m 220
53 14
13
12
11
10
9
8
7
6
5
53
Longitude, ˚W Fig. 1. Location of Loch Sunart, a fjord on the NW coast of Scotland (highlighted rectangle). Also shown are the major surface ocean currents (arrows) to the west of Scotland and Ireland, together with a simplified regional bathymetry. Figure modified after Cage (2005).
where m is the slope of the salinity: d18O mixing line, determined empirically from regional water samples, S is the salinity of the seawater and c is the intercept of the mixing line, representing the isotopic composition of the freshwater end-member entering the coastal environment. Regional salinity: d18O mixing lines vary across the NW European shelf due to differences in the d18O value of precipitation (e.g. Darling 2004; IAEA/WMO 2004) and
varying regional freshwater inputs entering the coastal ocean (Austin et al. 2006). Stable oxygen isotopic fractionation in carbonate material is also heavily temperature dependent and since the seminal publications of Urey (1947) and McCrea (1950), much work has been undertaken to refine and improve the temperature: d18O relationship (e.g. Epstein et al. 1953; Craig 1965; O’Neil et al. 1969; Emrich et al. 1970; Kim &
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA
O’Neil 1997; Lynch-Stieglitz et al. 1999). An excellent historical review of the development of the palaeotemperature equation is provided in Zeebe & Wolf-Gladrow (2001). Recent palaeotemperature work carried out by Bemis et al. (1998) on experimentally cultured planktonic foraminifera yielded the following empirical relationship: T(8C) ¼ a þ b(dc dw )
(2)
where a and b are constants at 16.5 and 24.8 respectively; dc is the isotopic composition of calcite and dw is the isotopic composition of seawater. Such palaeotemperature equations can be used to calculate the theoretical isotopic composition of a carbonate (equilibrium calcite, d18OEq.calcite) calcifying at a given temperature and salinity. Using the Bemis et al. (1998) palaeotemperature equation, d18OEq.calcite can be defined as: d18 OEq:calcite ¼ [T 16:5 (4:8 (d18 Oseawater 0:27))]= 4:8 (3) where T is the observed bottom water temperature (8C). Since d18Oseawater measurements are determined relative to the standard Vienna Standard Mean Ocean Water (VSMOW) and d18Oforam is determined relative to the Vienna Peedee Belemnite (VPDB) standard, the d18Oseawater must be corrected by 20.27‰ to convert it to VPDB from the VSMOW scale (Hut 1987).
Vital effects in benthic foraminifera Before isotopic compositions from fossil benthic foraminiferal specimens are used in palaeoenvironmental reconstructions, it is important to determine whether the shell chemistry of ‘live’ individuals (i.e. those stained with rose Bengal; Bernhard 2000) of a particular species reflect the conditions of the surrounding seawater, i.e. does the foraminifera calcify in isotopic equilibrium with the surrounding environment? To test whether a marine organism calcifies in or close to isotopic equilibrium with the stable oxygen isotopic composition of the seawater it inhabits at a given temperature, the measured d18Oforam can be compared to the expected oxygen isotopic composition (d18OEq.calcite), given that the temperature and salinity of the sample site is known. This difference from equilibrium calcite is expressed by Dd18O and defined as: Dd18 O ¼ d18 Oforam d18 OEq:calcite
(4)
157
However, the degree of isotopic disequilibrium (Dd18O) is often dependent on the palaeotemperature equation used to calculate d18OEq.calcite as discussed below. Isotopic disequilibria in benthic foraminifera (and other calcium carbonate organisms) are generally referred to as ‘vital effects’ (e.g. Grossman 1987). These were first reported by Duplessy et al. (1970) who noted that although the isotopic curves of Pyrgo, Planulina wuellerstorfi and mixed species from a deep-sea stratigraphic record were parallel, they were offset by 1‰ or more. Since then, much attention has focused upon ‘vital effects’ in benthic foraminifera (e.g. Vinot-Bertouille & Duplessy 1973; Erez 1978; Woodruff et al. 1980; Graham et al. 1981; Vincent et al. 1981; Grossman 1984, 1987). The physiological processes responsible for vital effects fall into two main categories: (1) metabolic effects which are influenced by respiration and photosynthetic processes and have little impact on d18Oforam values (e.g. McConnaughey et al. 1997); and (2) kinetic isotopic fractionation which is thought to be caused by the preferential uptake of the lighter isotopes of carbon and oxygen during the hydration and hydroxylation of CO2 (e.g. McConnaughey 1989a, 1989b) and is often associated with calcification rates (e.g. Erez 1978; McConnaughey 1989a; Owen et al. 2002). The carbonate chemistry and pH of a foraminifera’s microenvironment or microhabitat also plays an important role in test isotopic composition (e.g. Spero et al. 1997; Bemis et al. 1998; Bijma et al. 1999; Wolf-Gladrow et al. 1999; Zeebe et al. 1999). The majority of vital effects studies in benthic foraminifera have focused upon deep-sea species since these are a key proxy in most palaeoceanographic reconstructions (e.g. Woodruff et al. 1980; Wilson-Finelli et al. 1998; Schmiedl et al. 2004). Cibicides wuellerstorfi and Uvigerina peregrina are often used: the former is thought to calcify close to isotopic equilibrium with the d13CDIC of seawater (e.g. Zahn et al. 1986) and the latter species calcifies close to seawater d18O (Shackleton 1974). More recently, however, these assumptions of isotopic equilibrium in deep sea foraminifera have been challenged (e.g. Mackensen et al. 1993). Palaeoceanographic research is increasingly being undertaken in shelf sea environments, taking advantage of the expanded sedimentary archives in order to generate high-resolution palaeoclimatic reconstructions (e.g. Eirı´ksson et al. 2006; Norgaard-Pedersen et al. 2006). Unfortunately, C. wuellerstorfi and U. peregrina are typically rare in shelf sea sediments (e.g. Alve & Nagy 1990; Conradsen et al. 1994; Bergsten et al. 1996; Murray 2003b; Scott et al. 2003; Husum & Hald 2004; Cage 2005). Therefore, in order to generate
158
A. G. CAGE & W. E. N. AUSTIN
palaeoenvironmental records from shelf sea environments there is a growing need to address the ecological controls on the d18Oforam of shelf sea benthic foraminifera (Erez 1978; Polyak et al. 2003; Bauch et al. 2004; Filipsson et al. 2004; Scourse et al. 2004; Fontanier et al. 2006).
Seasonal effect in stable oxygen isotopic compositions of benthic foraminifera Seasonal evolution of equilibrium calcite in coastal waters and Scottish fjordic environments This study focuses upon benthic foraminifera from the marginal marine, shelf-sea environment of Loch Sunart, a fjord on the NW coast of Scotland (Fig. 1). A fjordic water column essentially consists of three water masses: (1) a thin (c. 5– 10 m) low salinity, brackish surface water layer formed from the mixing of high freshwater fluvial inputs and surface runoff with basin waters; (2) an intermediate salinity layer separated from the brackish surface layer by a well-defined halocline; and (3) a high salinity deep basin water mass which enters the basin via exchange or flushing of the fjord by coastal waters during an important process in fjordic hydrography: deep water renewal events (DWRE, e.g. Gade & Edwards 1980; Gillibrand et al. 1995). The dynamics of fjordic hydrography can be complex (e.g. Farmer & Freeland 1983) and are not constant. Periods of high runoff typically result in a strongly stratified water column which can suppress DWREs, whereas lower freshwater inputs allow high salinity coastal water masses to slip over the sill, resulting in frequent DWREs and high basin water salinity (e.g. Gillibrand et al. 2005). As with most shelf-sea regions, fjords also experience seasonal stratification where strong surface heat flux during the summer months results in a well defined seasonal thermocline and as the heat flux diminishes towards the winter months, tidal energy, strong mixing and convective overturning overcomes the stratification process and the water column becomes fully mixed (Pingree & Griffiths 1978; Farmer & Freeland 1983; Cage 2005). There are strong spatial differences in the seasonal stratification of the NW European shelf sea areas reflecting the temperature difference between the surface waters and the bottom waters (Elliot et al. 1991). Austin et al. (2006) provided the first systematic view of the long-term averaged (modern) spatial and temporal pattern of oxygen isotopes in seawater, expressed according to
calculated equilibrium calcite (d18OEq.calcite) around the shelf seas of NW Europe. Using monthly maps of sea surface and bottom water temperatures, they showed that large seasonal differences in Dd18OEq.calcite occur in regions with a mixed water column where bottom water temperatures experience marked seasonality (Fig. 2). For example, BWTs in the mixed water column of the northern Celtic Sea vary by 6.5 8C, and this strong seasonality is reflected in the d18OEq.calcite values which vary from 1.65‰ in March to 0.3‰ in September. Conversely, in stratified regions, such as the southern Celtic Sea, the seasonal heating cycle in the bottom waters is very much damped relative to the surface waters, varying by only 2.7 8C and resulting in a Dd18OEq.calcite range of 0.57‰ (Fig. 3). Monthly d18OEq.calcite maps from the Loch Sunart area show a fairly strong seasonal equilibrium calcite profile with a Dd18OEq.calcite of 1.13‰ (values taken from Austin et al. 2006).
‘Seasonal isotope effect’ in benthic foraminifera Benthic foraminifera in both deep-sea environments and shelf sea environments typically exhibit a marked seasonality in their growth, probably as a response to seasonal pulses in food availability during spring/autumn phytoplankton blooms (e.g. deep-sea environments: Gooday 1988; Gooday & Rathburn 1999; Austin & Evans 2000; Gooday & Hughes 2002. Shelf-sea environments: Murray 1991; Alve & Murray 1994; Hannah & Rogerson 1997; Scott et al. 2003). Since benthic foraminifera typically live from a few months up to a few years (Lee et al. 1991), the seasonal variation in growth (and hence calcification) is likely to integrate a record of the seasonal temperature cycle (e.g. Allison & Austin 2003). Therefore, the evolution of a pronounced seasonal signal in d18OEq.calcite on some parts of the NW European shelf (Austin et al. 2006) has important implications for understanding stable isotope records obtained from marine calcareous-shelled organisms. The hypothesis of a ‘seasonal isotope effect’ in the d18O of benthic foraminifera was proposed by Austin & Scourse (1997) who noted offsets in the d18Oforam values of Quinqueloculina seminulum and Ammonia batavus in a Holocene sediment record from the Celtic Sea. They attributed this offset to species calcifying their tests at different stages of the temperature cycle: A. batavus typically calcifying during the summer months when the water column was strongly stratified and BWTs were cool, and Q. seminulum likely calcifying
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA
159
Δδ18OEq.calcite, ‰ less than –0.2 –0.2 – 0 0 – 0.2 0.2 – 0.4 0.4 – 0.6 0.6 – 0.8 0.8 – 1.0 1.0 – 1.2 1.2 – 1.4 1.4 – 1.6 1.6 – 1.8 1.8 – 2.0 2.0 – 2.2 2.2 – 2.4 2.4 – 2.6 2.6 – 2.8 2.8 – 3.0 3.0 or more
Fig. 2. D d18OEq.calcite record showing the difference between winter (February) and summer (August) d18OEq.calcite values (values are ‰ VPDB) for the NW European shelf. The data are positioned and averaged on a 37.1 km (20 nautical mile) grid (200 latitude by 300 longitude), covering an area from 478N to 628N and 128W to 98E. Equilibrium calcite values were calculated using: average monthly measurements of bottom water temperature (BWT: 45 000 XBT profiles from Elliot et al. 1991); mean bottom water salinity maps (BWS) for summer and winter (Lee & Ramster 1981), regional salinity mixing lines (see Austin et al. 2006) and the empirical palaeotemperature equation of Bemis et al. 1998). Black squares denote landmass. Taken from Austin et al. 2006.
over a longer time period, or at a different time of the year (i.e. late summer/autumn) when BWTs were warmer. This concept of a ‘seasonal isotope effect’ was further tested by Scourse et al. (2004) who used d18Oforam measurements from the tests of live and dead benthic foraminifera from transects spanning the Celtic Sea front to: (1) reconstruct BWTs using the palaeotemperature equation of O’Neil et al. (1969); and (2) compare these reconstructed BWTs with observed seasonal BWT cycles in the Celtic Sea in order to assess the timing of calcification for a particular species. They concluded that many benthic foraminiferal species calcified at different times during the summer months. For instance, Q. seminulum appeared to calcify during peak BWTs (September in the Celtic Sea) whereas A. batavus calcified during September in stratified localities but during spring or early summer in regions with a mixed water column. Cibicides lobatulus, which is common in shelf-sea assemblages, calcified
during peak late summer in the Celtic Sea with a large oxygen isotopic disequilibrium effect (Scourse et al. 2004).
Methods Surface sediment samples were collected from Loch Sunart, a sea loch (fjord) on the NW coast of Scotland, during cruises aboard the R. V. Clupea (April 1999) and the R. V. Envoy (July 2001 and June 2002). Bottom water temperature and salinity measurements were recorded using a STD Plus 646 conductivity, temperature and depth (CTD) probe at each sample site (Table 1). No data were collected on pH and carbonate ion concentration. Two stationary Anderaa RCM –7 current meters complete with temperature and conductivity sensors and a data logger were deployed in the inner basin (56.68428N, 25.62118W) between 27th June 2001 and the 18th June 2002, and the main
δ18OEq.calcite, ‰
160
A. G. CAGE & W. E. N. AUSTIN
2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0
Stable isotopic compositions are reported relative to Vienna Pee Dee Belemnite (VPDB) using the NBS-19 standard and the following equation: d18 O(‰) ¼ [(18 O=16 O)sample =(18 O=16 O)standard ) 1] 1000 (5)
Loch Sunart Celtic Sea - mixed Celtic Sea - stratified
J F M A M J J A S O N D Months Fig. 3. The seasonal evolution of calculated monthly averaged equilibrium calcite (d18OEq.calcite) for Loch Sunart (black line), a fjord in NW Scotland, and the Celtic Sea (mixed water column – grey line, stratified water column – dashed grey line).
basin (56.66608N, 25.84348W) between 7th April, 2001 until the 17th June 2002. These stationary moorings were suspended approximately 40 m above the sea bed by a configuration of buoys, well below the influence of the brackish surface layer (Gillibrand et al. 1995). A series of independent CTD casts were taken to calibrate the mooring instruments throughout the deployment (Cage 2005). The top 0–1 cm of the surface sediment (obtained via a Van der Veen grab sampler or a Sholkovitch corer) was preserved using ethanol and stained using rose Bengal (Murray & Bowser 2000). Modern benthic foraminifera were picked using a stereo light microscope and identified from a number of taxonomic reference sources (Murray 1971; Haynes 1973; Murray 1991; Murray 2000). Stable oxygen isotopic analyses were carried out on bulk samples (typically 20 specimens selected from the 125– 250 mm fraction) of: Cibicides lobatulus (epifaunal), Ammonia beccarii (shallow infaunal), Bulimina marginata (infaunal) and Nonionella turgida (deep infaunal; Fig. 4). Where possible, ‘live’ specimens, i.e. those stained with rose Bengal, were preferentially selected, otherwise pristine specimens with no visible signs of dissolution or test alteration were used. Foraminiferal shells were gently crushed and rinsed with methanol prior to analysis on either a Finnigan MAT 251 with Kiel device (Bremen University or Bjerknes Centre for Climate Change, Bergen) for large samples (c. 200 mg) or a ThermoFinnigan MAT 253 with Kiel device (Bjerknes Centre for Climate Change, Bergen) for low mass samples (,100 mg).
Analytical accuracy for the d18O measurements were: 0.035 + 0.015‰ for the Finnigan MAT 251 with Kiel device at Bremen University; +0.07‰ for the Finnigan MAT 251 with Kiel device and +0.08‰ for the ThermoFinnigan MAT 253 both at Bjerknes Centre for Climate Change, Bergen. The predicted bottom water temperature (BWT) for each benthic foraminiferal isotope sample was calculated using the regional salinity: d18O mixing line of Austin & Inall (2002), based on the regression of 22 samples (R2 ¼ 0.998; reported analytical uncertainity on d18Oseawater measurements ¼ +0.15‰), defined as: d18 Oseawater ¼ 0:18 S 6
(6)
and the measured d18O value from the sample (d18Oforam) and the palaeotemperature equation of Lynch-Stieglitz et al. (1999), defined as: T ¼ [(d18 Oseawater 0:27) d18 Oforam þ 3:38]=0:21
(7)
The Lynch-Stieglitz et al. (1999) palaeotemperature equation is used to calculate predicted BWT or d18OEq.calcite for the Loch Sunart benthic foraminifera rather than the Bemis et al. (1998) equation adopted by Austin et al. (2006). Although the two equations are similar, the Lynch-Stieglitz equation is calibrated to Cibicidoides data over BWTs comparable to those observed in Loch Sunart, i.e. around 7– 14 8C, whereas the Bemis et al. (1998) equation is empirically calibrated to Cibicidoides from much lower BWTs (,6 8C) and Bemis et al. (1998) show a significant deviation of d18OEq.calcite from empirical measured d18Oforam at BWTs greater than 10 8C. Based soley on the analytical uncertainties quoted above, i.e. d18Oseawater ¼ +0.15‰ and d18Oforam ¼ +0.08‰, then the uncertainty in predicted BWT is +18C.
Results Seasonal hydrography of Loch Sunart Mooring data from Loch Sunart show a strong seasonal heating cycle, with temperatures at 40 m
Table 1. Measured stable oxygen (d18Oforam) and carbon (d13Cforam) isotopic compositions of modern benthic foraminiferal samples taken from Loch Sunart surface sediments (0–1 cm). Bottom water temperature (BWT) and bottom water salinity (BWS) data are provided for each sample and these data are used to predict the d18O theoretical equilibrium calcite (d18OEq.calcite). Isotopic disequilibrium (Dd18O) is the difference between d18OEq.calcite and d18Oforam. Analytical accuracy on the d18O measurements was , +0.08‰ Species
Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides Cibicides
lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus lobatulus
Date of collection
Latitude N (decimal)
Longitude W (decimal)
Water depth, m
BWT, 8C
BWS
d18Oforam, ‰
d13Cforam, ‰
d18OEq.calcite, ‰
Dd18O, %
25 27a 27b 27c 44 45 72 76 112 155 173 182 183a 183b 183c 197 200 201 211
07/04/01 07/04/01 07/04/01 07/04/01 07/04/01 07/04/01 29/06/01 29/06/01 15/07/01 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02
56.661 56.662 56.662 56.662 56.676 56.699 56.687 56.693 56.681 56.678 56.702 56.671 56.663 56.663 56.663 56.674 56.683 56.683 56.683
26.012 25.992 25.992 25.992 25.820 25.773 25.584 25.686 25.650 25.628 25.726 26.003 25.997 25.997 25.997 25.866 25.813 25.810 25.634
67 65 65 65 90 70 30 12 45 27 31 46 76 76 76 111 55 68 93
7.88 7.88 7.88 7.88 8.13 7.54 10.87 10.91 10.87 11.00 10.67 11.03 12.14 12.14 12.14 10.41 10.58 10.44 10.69
34.16 34.16 34.16 34.16 33.84 33.65 32.82 33.49 33.11 32.94 33.73 33.86 34.05 34.05 34.05 33.67 33.70 33.77 33.33
1.42 1.04 1.34 1.42 1.12 1.16 1.33 1.36 1.44 1.24 1.25 1.42 1.51 1.48 1.42 1.33 1.52 1.50 1.42
21.21 21.41 21.37 21.61 21.38 21.29 21.44 20.93 20.50 21.06 21.20 21.26 21.61 21.44 21.43 21.30 21.23 21.41 21.58
1.60 1.60 1.60 1.60 1.49 1.58 0.73 0.85 0.79 0.73 0.94 0.89 0.69 0.69 0.69 0.98 0.95 1.00 0.86
20.18 20.56 20.26 20.18 20.37 20.42 0.60 0.51 0.65 0.51 0.31 0.53 0.82 0.79 0.73 0.35 0.57 0.50 0.56
25 30 34 41 44 110 112 162 180 181 182 183 198
07/04/01 07/04/01 07/04/01 07/04/01 07/04/01 15/07/01 15/07/01 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02
56.661 56.656 56.670 56.670 56.676 56.680 56.681 56.685 56.682 56.680 56.671 56.664 56.665
26.012 25.966 25.947 25.865 25.820 25.647 25.650 25.683 25.988 26.012 26.003 25.997 25.841
67 41 71 58 90 53 45 13 21 36 46 76 121
7.88 7.87 7.88 7.73 7.88 10.95 10.87 10.91 11.75 11.54 11.03 12.14 10.40
34.16 34.07 34.12 33.74 33.84 33.05 33.11 33.49 33.89 33.69 33.86 34.05 33.66
0.96 0.87 0.50 0.89 0.97 0.61 0.96 0.98 0.58 0.93 0.91 0.84 0.63
1.53 1.39 0.46 1.33 0.96 1.15 1.43 0.79 1.31 1.48 1.20 1.36 1.27
1.60 1.59 1.60 1.56 1.55 0.76 0.79 0.85 0.74 0.75 0.89 0.69 0.98
20.64 20.72 21.10 20.67 20.58 20.15 0.18 0.13 20.16 0.18 0.02 0.15 20.35
161
(Continued)
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA
Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii Ammonia beccarii
Sample station
162
Table 1. Continued Date of collection
Latitude N (decimal)
Longitude W (decimal)
Water depth, m
BWT, 8C
BWS
d18Oforam, ‰
198 206 207 Mooring
20/06/02 20/06/02 20/06/02 20/06/02
56.665 56.632 56.635 56.685
25.841 25.841 25.852 25.623
121 13 30 40
10.40 12.40 12.36 10.86
33.66 32.56 32.56 32.76
0.83 0.40 0.49 0.78
Nonionella turgida Nonionella turgida Nonionella turgida Nonionella turgida Nonionella turgida Nonionella turgida Nonionella turgida Nonionella turgida Nonionella turgida
27 44 72 183 197 200 201 202 211
07/04/01 07/04/01 29/06/01 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02
56.662 56.676 56.687 56.663 56.674 56.683 56.683 56.680 56.683
25.992 25.820 25.584 25.997 25.866 25.813 25.810 25.810 25.634
65 90 30 76 111 55 68 97 93
7.88 8.13 10.87 12.14 10.41 10.58 10.44 10.16 10.69
34.16 33.84 32.82 34.05 33.67 33.70 33.77 33.80 33.33
Bulimina marginata Bulimina marginata Bulimina marginata Bulimina marginata Bulimina marginata Bulimina marginata Bulimina marginata Bulimina marginata Bulimina marginata Bulimina marginata
27 44 76 173 182 183 197 200 201 202
07/04/01 07/04/01 29/06/01 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02 20/06/02
56.662 56.676 56.693 56.702 56.671 56.663 56.674 56.683 56.683 56.680
25.992 25.820 25.686 25.726 26.003 25.997 25.866 25.813 25.810 25.810
65 90 12 31 46 76 111 55 68 97
7.88 8.13 10.91 10.67 11.03 12.14 10.41 10.58 10.44 10.16
34.16 33.84 33.49 33.73 33.86 34.05 33.67 33.70 33.77 33.80
Cibicides Cibicides Cibicides Cibicides
lobatulus lobatulus lobatulus lobatulus
d13Cforam, ‰
d18OEq.calcite, ‰
Dd18O, %
1.51 0.50 0.99 0.37
0.98 0.37 0.38 0.73
20.16 0.03 0.12 0.05
1.31 1.36 1.15 1.40 1.12 1.45 1.33 1.38 1.09
21.79 21.79 21.82 22.04 21.96 21.75 21.88 22.01 21.62
1.60 1.49 0.73 0.69 0.98 0.95 1.00 1.06 0.86
20.29 20.13 0.42 0.71 0.14 0.50 0.33 0.32 0.23
1.47 1.75 1.52 1.67 1.67 1.62 1.67 1.89 1.79 1.65
20.84 20.52 20.56 20.07 20.08 20.68 20.89 20.49 20.81 20.99
1.60 1.49 0.85 0.94 0.89 0.69 0.98 0.95 1.00 1.06
20.13 0.26 0.67 0.73 0.78 0.93 0.69 0.94 0.79 0.59
A. G. CAGE & W. E. N. AUSTIN
Sample station
Species
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA
163
Fig. 4. Benthic foraminiferal species used in the study (a) Ammonia beccarii; (b) Cibicides lobatulus; (c) Bulimina marginata; and (d) Nonionella turgida.
water depth ranging from a minimum of 7.7 8C in March to 13.4 8C in October (Fig. 5). The inner and main basin temperature profiles are similar despite temporal differences in salinity, highlighting the strong thermal stratification in the coastal regions of NW Scotland. This seasonal heating cycle significantly impacts upon the d18OEq.calcite in Loch Sunart (Fig. 5), with the c. 5.7 8C BWT difference equating to a seasonal Dd18OEq.calcite of 1.32‰ (where Dd18OEq.calcite ¼ winter/February d18OEq.calcite 2 summer/August d18OEq.calcite). A long-term continuous record of temperature does not exist for Loch Sunart, however Gillibrand et al. (2005) show that basin water salinity and exchange in Loch Sunart responds to inter-annual variability in climatic forcing; notably rainfall amount and westerly airflow. Regional coastal water temperatures, including those from Loch Sunart, typically follow mainland air temperatures (Cage 2005). Monthly sea surface temperatures (SST) have been recorded at the Millport station (NW Scotland) since 1953 and superimposed upon a predictably strong seasonal heating cycle is inter-annual temperature variability, with mean annual temperatures ranging from 8.8 8C to 11.24 8C. Water temperature profiles from Loch Sunart show that temperatures at 40 m water depth are similar to water temperatures at deeper depths, thus BWTs are likely to record similar interannual variability to the Millport SSTs, albeit with a damped signal.
Oxygen isotopic disequilibria in Loch Sunart benthic foraminifera The oxygen isotopic disequilibria for benthic foraminifera from Loch Sunart are summarized in Figure 6. Live A. beccarii typically show enriched d18Oforam values with respect to d18OEq.calcite with
a mean enrichment of Dd18O ¼ 0.29 + 0.1‰ (n ¼ 19). Live B. marginata and N. turgida also have positive Dd18O values of 0.62 + 0.1‰ (n ¼ 10) and 0.25 + 0.1‰ (n ¼ 9) respectively. Live C. lobatulus show a negative Dd18O value of 20.24 + 0.29‰ for 2 samples, one collected in April and one collected in June. The number of samples included in the study was increased to include pristine dead C. lobatulus individuals or pristine individuals whose life status was not classified prior to analysis, after which the mean Dd18O shifted slightly towards a Dd18O of 20.21 + 0.94‰ (n ¼ 17). The measured d18Oforam for live specimens collected in April, when BWT is at a minimum and approximately 5 8C cooler than during the summer months, are consistently depleted with respect to predicted equilibrium calcite for all taxa. Once these April data are excluded, the mean Dd18O changes for each species (Fig. 6). The Dd18O for A. beccarii shifts to 0.57 + 0.04‰ (n ¼ 13), B. marginata has a mean Dd18O of 0.77 + 0.04‰ (n ¼ 8) whilst the mean Dd18O for N. turgida shifts to 0.38 + 0.07‰ (n ¼ 7). The Dd18O of C. lobatulus lies very close to isotopic equilibrium with a mean Dd18O of 0.003 + 0.17‰ (n ¼ 12 including 1 ‘live’ sample). When the measured d18Oforam values are translated to predicted BWTs, we observe that the predicted BWTs derived from the d18Oforam of A. beccarii and C. lobatulus collected ‘live’ in the cold BWTs of April are significantly higher than the observed BWT at the time of collection (Fig. 7), i.e. they have a large Dd18O with measured d18Oforam values lying well below the expected d18OEq.calcite (Fig. 6). This suggests that the measured d18Oforam values from these taxa are more representative of summer BWTs than of winter BWTs, and raises the question of the timing of calcification, i.e. the ‘seasonal isotope effect’.
164
A. G. CAGE & W. E. N. AUSTIN
(a)
16 14
Temperature, °C
12 10 8 6 4
Inner basin
Inner basin CTD
Main basin
Main basin CTD
2 0 1/3/01
30/4/01
29/6/01
28/8/01 27/10/01 26/12/01 24/2/02
25/4/02
24/6/02
25/4/02
24/6/02
25/4/02
24/6/02
Date
(b)
DWREs
36 35
Salinity units
34 33 32 31 30 29 1/3/01
30/4/01
29/6/01
28/8/01 27/10/01 26/12/01 24/2/02 Date
(c)
Theoretical δ18OEq.CaCO3 ‰
2.0
1.5
1.0
0.5
0.0
–0.5 1/3/01
30/4/01
29/6/01
28/8/01 27/10/01 26/12/01 24/2/02 Date
Fig. 5. Hourly temperature (a) and salinity (b) data from the inner basin mooring (black line) and main basin mooring (grey line) from Loch Sunart over a 15 month period (April 2001–June 2002). Data courtesy of Dr P. Gillibrand collected using Anderaa RCM– 7 current metres equipped with data loggers and temperature and conductivity sensors. The mooring data were calibrated using CTD casts taken throughout the collection period. Deep water renewal events (DWREs) are indicated by black arrows. The theoretical range of equilibrium d18O calcite (d18O Eq.CaCO3; part c) in Loch Sunart has been calculated from the mooring data using the Austin & Inall (2002) regional salinity: d18O mixing line and the palaeotemperature equation of Lynch-Steiglitz et al. (1999). The effect of the seasonal heating cycle on d18 OEq:CaCO3 can be clearly observed, particularly in the main basin where salinity is fairly stable in comparison to the inner basin.
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA
165
Δδ18Ο, ‰ –1.5
Cibicides lobatulus
Ammonia beccarii
Bulimina marginata
Nonionella turgida
–1.0
–0.5
all summer
0.0
0.5
1.0
1.5
Cibicides lobatulus Ammonia beccarii Bulimina marginata Nonionella turgida
all summer
all summer
all summer
Fig. 6. Oxygen isotopic disequilibria in benthic foraminifera from Loch Sunart calculated using the palaeotemperature equation of Lynch-Stieglitz et al. (1999), the regional salinity:d18O mixing line of Austin & Inall (2002) and measured d18O from the benthic foraminiferal tests. Samples were collected from the uppermost 1 cm of sediment. Black symbols denote samples collected in both winter and summer: grey symbols show summer data only. Mean (large symbols) deviations of Dd18O away from equilibrium calcite for ‘live’ Ammonia beccarii (triangles), Bulimina marginata (diamonds) and Nonionella turgida (circles). Cibicides lobatulus data (squares) include 2 ‘live’ samples and samples which have no classified life status (i.e. could be dead or live). Small symbols show individual d18O measurements. The data has been plotted against the relative depth habitat within the sediment as suggested by measured d13C for each species (Cage 2005).
Discussion Seasonal isotope effect on d18O of shelf sea benthic foraminifera Discrepancies between the observed BWTs recorded during the collection of the mostly ‘live’ benthic foraminiferal specimens and predicted BWTs (calculated using equations 6 and 7) point to the possibility that a ‘seasonal isotope effect’ influences the d18Oforam values of benthic foraminifera in Loch Sunart (Fig. 7). In particular, the similarity of the predicted BWT for samples collected in April to the observed BWT of samples collected in June/July suggests that ‘live’ tests collected in the cool BWTs of April actually calcified during the warmer BWTs of the previous summer (Fig. 7). For example, the measured d18Oforam of C. lobatulus collected from a BWT of 7.88 8C in April, 2001, yields a predicted BWT of 13.45 8C and is similar to the
maximum temperature of 13.4 8C recorded at a water depth of 40 m in Loch Sunart during September, 2001 (Fig. 5). Such differences are unlikely to be accounted for by propagating the analytical uncertainties of the methods used to calculate BWT, which are approximately+1 8C. This suggests that C. lobatulus from Loch Sunart typically calcifies from summer (June) to the late summer peak of the BWT cycle (October/ November). This agrees well with the proposed calcification period for this species in the Celtic Sea (Scourse et al. 2004) and also with the calcification period of benthic foraminifera from Norwegian fjords, which tend to calcify around November (pers. comm., Dr. G. Mikalesen 2004). Predicted BWTs from live Ammonia beccarii collected in April are also higher than the predicted BWTs for live A. beccarii collected in June/July, suggesting calcification during the warmer BWTs of the previous late summer. However, the measured d18Oforam of live A. beccarii individuals captured in June/July produce predicted
166
A. G. CAGE & W. E. N. AUSTIN
(a) 14 (ii)
Predicted temperature, °C
(i)
A. beccarii ‘Live’ C. lobatulus ‘U’ C. lobatulus x=y
12
(iii)
10
(iv) 8
6
6
8
10 12 Observed temperature, °C
14
Observed / predicted temperature, °C
(b) 14 12 ΔT = 0.3 °C 10 ΔT = 1.1 °C 8 Ammonia beccarii Cibicides lobatulus
6 4 Aug
Oct
Dec
Feb Month
Apr
Jun
Aug
Fig. 7. (a) Observed bottom water temperatures (BWTs) and calculated BWTs (using the equations from Lynch-Steiglitz et al. 1999 and Austin & Inall 2002) for ‘live’ Ammonia beccarii (triangles) and ‘live’ Cibicides lobatulus (black squares) and C. lobatulus speciments with an unspecified life status (i.e. possibly collected ‘live’; grey squares). The theoretical oxygen isotopic equilibrium of foraminiferal calcite (d18OEq.calcite) is represented by the dashed x ¼ y line. C. lobatulus data are distinguished as: (i) April collected C. lobatulus, where predicted BWT . observed BWT; and (ii) June/July collected C. lobatulus, where the predicted BWT the observed BWT. A. beccarii data are distinguished as: (iii) April collected A. beccarii, where predicted BWT . observed BWT; and (iv) June/July collected A. beccarii, where the predicted BWT the observed BWT; (b) Typical temperature profile for BWT from Loch Sunart (black line). Mean predicted BWTs for A. beccarii and C. lobatulus are shown together with minimum and maximum reconstructions. Differences in predicted BWT, based upon a comparison of April and June/July collections of ‘live’ specimens, are represented by DT. Differences between the observed and predicted BWTs are potentially accounted for by the ‘seasonal isotope effect’ (dashed horizontal lines indicate likely time of major phase of calcification).
BWTs which are cooler than the observed BWT at the time of their collection, suggesting that individuals of A. beccarii collected in June/July likely calcified the majority of their calcite test in the cooler BWTs of spring/early summer. These two calcification periods may reflect the spring onset of
seasonal stratification and winter mixing of Loch Sunart (Cage 2005). Scourse et al. (2004) also reported that A. batavus from the Celtic Sea sites with a mixed water column calcified in spring/ early summer, whereas A. batavus from stratified sites calcified during the peak late summer BWTs.
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA
Oxygen isotopic disequilbria in Cibicides lobatulus The incorporation of oxygen isotopes into biogenic carbonate is complex and as Wefer et al. (1999) point out, the d18O value of a carbonate is a function of temperature, ice volume, selective growth in space and time, ‘vital effects’ and post-depositional dissolution. The carbonate chemistry and pH of a benthic foraminifera’s microenvironment or microhabitat also plays an important role in test isotopic composition (e.g. Spero et al. 1997; Bemis et al. 1998; Bijma et al. 1999; Wolf-Gladrow et al. 1999), with a 0.2 increase in pH resulting in a d18O shift of approximately 20.22‰ (Zeebe 1999). Data presented in this study suggest that C. lobatulus collected during the summer months from Loch Sunart typically calcify close to the oxygen isotopic composition of seawater (Fig. 6). This is interesting as the Cibicidoides group typically calcify close to the stable carbon isotopic composition of dissolved inorganic calcite in seawater (e.g. Wilson-Finelli et al. 1998; Tachikawa & Elderfield 2002), whereas a number of studies report a tendency for Cibicidoides species to calcify in isotopic disequilibrium with respect to theoretical equilibrium calcite, exhibiting consistently negative Dd18O values of up to 21‰ (e.g. Belanger et al. 1981; Graham et al. 1981; Vilks & Deonarine 1988; McCorkle et al. 1990; Rathburn
167
et al. 1996; McCorkle et al. 1997; Scott et al. 1998; Wilson-Finelli et al. 1998; Schmiedl et al. 2004; Scourse et al. 2004; Fontanier et al. 2006). Conversely, Bemis et al. (1998) demonstrated that deep-sea core-top Cibicidoides from the Atlantic, Pacific and Indian Oceans and the Arabian Sea and Gulf of Mexico appeared to calcify close to d18OEq.calcite, and Lynch-Stieglitz et al. (1999) showed Cibicidoides specimens calcified close to isotopic equilibrium at temperatures of 5–27 8C. Species-specific palaeotemperature equations and potential carbonate ion effects. The disparity between the low Dd18O reported for C. lobatulus in this study and the typically high Dd18O reported from other studies appears largely to be due to the choice of palaeotemperature equation used to calculate d18OEq.calcite (Fig. 8). Studies reporting a small Dd18O for deep-dwelling infaunal species, such as Uvigerina peregrina and a large Dd18O for epifaunal species such as Cibicidoides (e.g. Grossman 1984; McCorkle et al. 1997; Schmiedl et al. 2004; Scourse et al. 2004; Fontanier et al. 2006) typically use the O’Neil et al. (1969) palaeotemperature equation (or a close derivative) defined as; T(8C) ¼ 16:9 4:38(d18 Oforam d18 Owater ) þ 0:1(d18 Oforam d18 Owater )2
(8)
8
Difference between observed BWT and predicted BWT, °C
O‘Neil et al. (1969) 6
Bemis et al. (1998)
4
Lynch-Stieglitz et al. (1999) Equilibrium calcite
2 0 –2 –4 –6 Ammonia beccarii
Cibicides lobatulus
Fig. 8. Mean differences (large symbols) between observed temperature and predicted temperature for ‘live’ or unspecified life status, Cibicides lobatulus (black) and Ammonia beccarii (grey) using the palaeotemperature equations of O’Neil et al. (1969; circle), Bemis et al. (1998; triangle) and Lynch-Stieglitz et al. (1999; square). Small symbols show individual measurements.
168
A. G. CAGE & W. E. N. AUSTIN
O’Neil et al. (1969) used synthetic carbonates precipitated at temperatures of 0–500 8C to derive this palaeotemperature equation (largely in water super-saturated with respect to CO22 3 ). Shackleton (1974) modified the O’Neil et al. (1969) equation and calibrated it to the d18Oforam of the deep infaunal species Uvigerina peregrina. The palaeotemperature equation derived by Bemis et al. (1998) on the other hand was obtained through laboratory and field studies on planktonic foraminifera such as Orbulina universa (using filtered seawater). They reported that core-top (deep-sea) Cibicidoides collected from BWTs of ,6 8C agreed well with the empirical palaeotemperature equation derived for O. universa cultured under ‘low-light’ irradiance conditions (seen in eq. 3), whereas the data from the deep infaunal species Uvigerina peregrina was in closer agreement with the palaeotemperature equations of Shackleton (1974) and Erez & Luz (1983). Lynch-Steiglitz et al. (1999) provide the most recent regression palaeotemperature equation from modern Cibicidoides and Planulina calibrated for BWTs between 5–27 8C and show that it is similar to Bemis et al. (1998) and to the laboratory ‘fractionation’ derived palaeotemperature equations of Kim & O’Neil (1997). The temperature range of calibration is important: Cibicidoides data from Bemis et al. (1998) appear to deviate from theoretical equilibrium calcite when bottom water temperatures are greater than c. 6 8C and Bemis et al. (table 1, 1998) highlight offsets in palaeotemperature equations over a given temperature range. Therefore, species-specific palaeotemperature equations which have been calibrated over a temperature range similar to the BWTs experienced at the study site should be used. The difference between the O’Neil et al. (1969) palaeotemperature equation and the Bemis et al. (1998) or Lynch-Stieglitz et al. (1999) palaeotemperature equations, and their respective calculations of isotopic equilibria for the infaunal and epifaunal species, could also be due to a carbonate ion effect within the sediment porewater profile. This appears plausible, because stable carbon isotopic compositions from benthic foraminifera show that infaunal taxa calcify in different porewater chemistry than epifaunal taxa (e.g. McCorkle et al. 1997). Since a 0.2 increase in pH (directly linked to the carbonate ion effect) results in a d18O depletion of c. 0.22‰ (Zeebe 1999), the decrease in the pH of porewaters with sediment depth (e.g. Jahnke & Jahnke 2004; Zhu et al. 2006) is likely to be reflected in the stable oxygen isotopic compositions of benthic foraminiferal tests (Bemis et al. 1998), and one would expect the d18Oforam of infaunal taxa to be enriched with respect to the d18Oforam of epifaunal benthic foraminifera. The d18O data from the infaunal benthic foraminifera (A. beccarii, B. marginata and N. turgida)
from Loch Sunart do show enriched d18O values with respect to the epifaunal taxon, C. lobatulus. An estimate of how d18Oforam varies with the sediment pH profile is not possible as pH data is not currently available for Loch Sunart. However, when foraminiferal d18O values are compared from taxa collected live at the same station, the average Dd18Oinfaunal-epifaunal differences correspond to pH changes of 2.2 for A. beccarii-C. lobatulus and 1.9 for N. turgida-C. lobatulus (Table 1). Data on pH collected from Long Island Sound, an estuary on the NE coast of America, shows that water column pH varies from 8.5 to c. 7.7 from the surface to 28 m water depth (Hanson & Donaghay 1998) and pH can vary by 1 –2 pH units in sediment porewaters with the pH minimum occurring at around 1 cm below the sediment surface (Zhu et al. 2006). Therefore, it is not unrealistic that the differences between these infaunal and epifaunal taxa are being driven by porewater pH. Some species of infaunal benthic foraminifera, such as Buliminids are known to migrate within the sediment to optimize on certain living conditions (e.g. Schmeidel et al. 2004 and Jorissen 1999). How the movement of individual benthic foraminifera within the sediment profile influences the shell geochemical composition with respect to changing porewater chemistry during migration is largely unknown. In addition, the carbonate-ion effect, together with factors such as the ‘seasonal isotope effect’ and ‘vital effects’, further complicate our understanding of in the d18O signal in benthic foraminifera. Disentangling microhabitat and ‘seasonal effects’ could be achieved through benthic foraminiferal culturing experiments (e.g. Hintz et al. 2004) and regular monitoring of field sites.
Conclusions Stable oxygen isotope data measured from four benthic foraminiferal taxa collected live during April (cool BWT) and June/July (warm BWT) point to a possible ‘seasonal isotope effect’ in test geochemistry, as recently discussed by Scourse et al. (2004). Predicted BWTs calculated using the palaeotemperature equation of Lynch-Steiglitz et al. (1999) and the measured d18O of Cibicides lobatulus collected from multiple sites in Loch Sunart suggest this taxon typically calcifies in the warmer BWTs of summer/late summer. By contrast, the d18O of live-collected Ammonia beccarii points to two phases of recruitment, and hence two major phases of calcification, one during the late summer and one during the spring. These distinct phases of growth and calcification may arise from the recruitment of new cohorts following reproduction events triggered by food availability during the spring and autumn seasonal
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA
phytoplankton blooms (e.g. Murray 1983; Gooday 1988; Gooday & Rathburn 1999; Scott et al. 2003). These data highlight the greater awareness that is currently emerging on the incremental nature of growth in benthic foraminifera, potentially providing a means to resolve growth and calcification during different parts of the annual cycle (e.g. Allison & Austin 2003; Scourse et al. 2004). However, a number of outstanding issues remain to be resolved with respect to the interpretation of the effects of seasonal dynamics upon d18O of benthic foraminifera. In particular, further research is required in order to separate ‘vital effects’, seawater carbonate chemistry and pH from ‘seasonal isotope effects’ upon shell chemistry. The d18Oforam deviations away from d18OEq.calcite reported here are most likely to be influenced by a ‘seasonal isotope effect’ rather than simply an unknown ‘vital effect’, since the latter should produce a constant offset in d18O irrespective of the time of sample collection, rather than the variable Dd18O observed in this study. We thank Rune Søra˚s and Ulysses Ninneman, University of Bergen, and Monika Segl, University of Bremen, for analytical support in conducting the stable isotope analyses and Dr Phil Gillibrand and the Fisheries Research Services, Aberdeen for the mooring data. We acknowledge the support of the Universities of Bergen (EU Marie Curie support to AGC), Bremen (EU Paleostudies support to WENA) and St Andrews. This work is a contribution to the EU Millennium project. We would like to thank the reviewers, Dr Hilary Kennedy and Dr Mike Rogerson for their constructive comments on an earlier draft of the manuscript.
References A LLISON , N. & A USTIN , W. E. N. 2003. The potential of ion microprobe analysis in detecting geochemical variations across individual foraminifera tests. Geochemistry Geophysics Geosystems, 4, 8403. A LVE , E. & M URRAY , J. W. 1994. Ecology and taphonomy of benthic foraminifera in a temperate mesotidal inlet. Journal of Foraminiferal Research, 24, 18–27. A LVE , E. & N AGY , J. 1986. Estuarine foraminiferal distribution in Sandebukta, a branch of the Oslo Fjord. Journal of Foraminiferal Research, 16, 261– 284. A LVE , E. & N AGY , J. 1990. Main features of foraminiferal distribution reflecting estuarine hydrography in Oslo Fjord. Marine Micropaleontology, 16, 181– 206. A USTIN , W. E. N. & E VANS , J. R. 2000. NE Atlantic benthic foraminifera: modern distribution patterns and palaeoecological significance. Journal of the Geological Society, 157, 679– 691. A USTIN , W. E. N. & I NALL , M. E. 2002. Deep-water renewal in a Scottish fjord: temperature, salinity and oxygen isotopes. Polar Research, 21, 251– 257. A USTIN , W. E. N. & S COURSE , J. D. 1997. Evolution of seasonal stratification in the Celtic Sea during the
169
Holocene. Journal of the Geological Society, 154, 249– 256. A USTIN , W. E. N., C AGE , A. G. & S COURSE , J. D. 2006. Mid-latitude shelf seas: a NW European perspective on the seasonal dynamics of temperature, salinity and oxygen isotopes. Holocene, 16, 937 –947. B ACKHAUS , J. O. 1996. Climate-sensitivity of European marginal seas, derived from the interpretation of modelling studies. Journal of Marine Systems, 7, 361–382. B AUCH , H. A., E RLENKEUSER , H., B AUCH , D., M UELLER -L UPP , T. & T ALDENKOVA , E. 2004. Stable oxygen and carbon isotopes in modern benthic foraminifera from the Laptev Sea shelf: implications for reconstructing proglacial and profluvial environments in the Arctic. Marine Micropaleontology, 51, 285– 300. B ELANGER , P. E., C URRY , W. B. & M ATTHEWS , R. K. 1981. Core-top evaluation of benthic foraminiferal isotopic-ratios for paleo-oceanographic interpretations. Palaeogeography Palaeoclimatology Palaeoecology, 33, 205– 220. B EMIS , B. E., S PERO , H. J., B IJMA , J. & L EA , D. W. 1998. Re-evaluation of the oxygen isotopic composition of planktonic foraminifera: experimental results and revised paleotemperature equations. Paleoceanography, 13, 150– 160. B ERGSTEN , H., N ORDBERG , K. & M ALMGREN , B. 1996. Recent benthic foraminifera as tracers of water masses along a transect in the Skagerrak, north-eastern North Sea. Journal of Sea Research, 35, 111– 121. B ERNHARD , J. M. 2000. Distinguishing live from dead foraminifera: Methods review and proper applications. Micropaleontology, 46, 38–46. B IJMA , J., S PERO , H. J. & L EA , D. W. 1999. Reassessing foraminiferal stable isotope geochemistry: impact of the oceanic carbonate system (experimental results). In: F ISCHER , G. & W EFER , G. (eds) Use of Proxies in Paleoceanography. Springer, 489–512. C AGE , A. G. 2005. The modern and late Holocene marine environments of Loch Sunart, N.W. Scotland. Ph.D thesis, University of St Andrews. C ONRADSEN , K., B ERGSTEN , H., K NUDSEN , K. L., N ORDBERG , K. & S EIDENKRANTZ , M. S. 1994. Recent benthic foraminiferal distribution in the Kattegat and the Skagerrak, Scandinavia. In: S EJRUP , H. P. & K NUDSEN , K. L. (eds) Late Cenozoic Benthic Foraminifera: Taxonomy, Ecology and Stratigraphy, Cushman Foundation for Foraminiferal Research. Special Publication, 32, 53– 68. C RAIG , H. 1965. Measurement of oxygen isotope paleotemperatures. In: T ONGIORGI , E. (ed.) Stable Isotopes in Oceanographic Studies and Paleotemperatures. Laboratory of Nuclear Geology, National Research Council, Pisa, 161– 182. D ARLING , W. G. 2004. Hydrological factors in the interpretation of stable isotopic proxy data present and past: a European perspective. Quaternary Science Reviews, 23, 743– 770. D UPLESSY , J. C., L ALOU , C. & V INOT , A. C. 1970. Differential isotopic fractionations in benthic foraminifera and paleotemperatures reassessed. Science, 168, 250–251. E IRIKSSON , J., B ARTELS -J ONSDOTTIR , H. B., C AGE , A. G. ET AL . 2006. Variability of the North Atlantic
170
A. G. CAGE & W. E. N. AUSTIN
Current during the last 2000 years based on shelf bottom water and sea surface temperatures along an open ocean/shallow marine transect in western Europe. Holocene, 16, 1017–1029. E LLIOT , A. J., C LARKE , T. & L I , Z. 1991. Monthly distributions of surface and bottom temperatures in the Northwest European shelf seas. Continental Shelf Research, 11, 453 –466. E MRICH , K., E HHALT , D. H. & V OGEL , J. C. 1970. Carbon isotope fractionation during the precipitation of calcium carbonate. Earth and Planetary Science Letters, 8, 363–371. E PSTEIN , S., B UCHSBAUM , R., L OWENSTAM , H. A. & U REY , H. C. 1953. Revised carbonate-water isotopic temperature scale. Geological Society American Bulletin, 64, 1315–1326. E REZ , J. 1978. Vital effect on stable-isotope composition seen in foraminifera and coral skeletons. Nature, 273, 199– 202. E REZ , J. & L UZ , B. 1983. Experimental paleotemperature equation for planktonic foraminifera. Geochemica et Cosmochimica Acta, 47, 1025–1031. F ARMER , D. M. & F REELAND , H. J. 1983. The physical oceanography of fjords. Progress in Oceanography, 12, 147 –219. F ILIPSSON , H. L., N ORDBERG , K. & G USTAFSSON , M. 2004. Seasonal study of delta O-18 and delta C-13 in living (stained) benthic foraminifera from two Swedish fjords. Marine Micropaleontology, 53, 159– 172. F ONTANIER , C., M ACKENSEN , A., J ORISSEN , F. J., A NSCHUTZ , P., L ICARI , L. & G RIVEAUD , C. 2006. Stable oxygen and carbon isotopes of live benthic foraminifera from the Bay of Biscay: microhabitat impact and seasonal variability. Marine Micropaleontology, 58, 159 –183. G ADE , H. G. & E DWARDS , A. 1980. Deep-water renewal in fjords. In: F REELAND , H. J., F ARMER , D. M. & L EVINGS , C. D. (eds) Fjord Oceanography. Plenum Press, New York, 435 –489. G ILLIBRAND , P. A., C AGE , A. G. & A USTIN , W. E. N. 2005. A preliminary investigation of basin water response to climate forcing in a Scottish fjord: evaluating the influence of the NAO. Continental Shelf Research, 25, 571 –587. G ILLIBRAND , P. A., T URRELL , W. R. & E LLIOTT , A. J. 1995. Deep-water renewal in the upper basin of Loch Sunart, a Scottish fjord. Journal of Physical Oceanography, 25, 1488– 1503. G OODAY , A. J. 1988. A response by benthic Foraminifera to the deposition of phytodetritus in the deep sea. Nature, 332, 71– 73. G OODAY , A. J. & H UGHES , J. A. 2002. Foraminifera associated with phytodetritus deposits at a bathyal site in the northern Rockall Trough (NE Atlantic): seasonal contrasts and a comparison of stained and dead assemblages. Marine Micropaleontology, 46, 83–110. G OODAY , A. J. & R ATHBURN , A. E. 1999. Temporal variability in living deep-sea benthic foraminifera: a review. Earth-Science Reviews, 46, 187–212. G RAHAM , D. W., C ORLISS , B. H., B ENDER , M. L. & K EIGWIN , L. D. 1981. Carbon and oxygen isotopic disequilibria of recent deep-sea benthic foraminifera. Marine Micropaleontology, 6, 483– 497.
G ROSSMAN , E. L. 1984. Stable isotope fractionation in live benthic foraminifera from the Southern-California Borderland. Palaeogeography Palaeoclimatology Palaeoecology, 47, 301– 327. G ROSSMAN , E. L. 1987. Stable isotopes on modern benthic foraminifera: a study of vital effect. Journal of Foraminiferal Research, 17, 48–61. H ANNAH , F. & R OGERSON , A. 1997. The temporal and spatial distribution of foraminiferans in marine benthic sediments of the Clyde Sea area, Scotland. Estuarine Coastal and Shelf Science, 44, 377–383. H ANSON , A. K. , J R . & D ONAGHAY , P. L. 1998. Micro- to fine-scale chemical gradients and layers in stratified coastal waters. Oceanography, 11, 10– 17. H AYNES , J. R. 1973. Cardigan Bay recent foraminifera (Cruises of the R. V. Antur, 1962–1964). Bulletin of the British Museum (Natural History) Zoology, Supplement 4, 1– 245. H INTZ , C. J., C HANDLER , G. T., B ERNHARD , J. M., M C C ORKLE , D. C., H AVACH , S. M., B LANKS , J. K. & S HAW , T. J. 2004. A physicochemically constrained seawater culturing system for production of benthic foraminifera. Limnology and Oceanography-Methods, 2, 160–170. H USUM , K. & H ALD , M. 2004. Modern foraminiferal distribution in the subarctic Malangenfjord and adjoining shelf, northern Norway. Journal of Foraminiferal Research, 34, 34–48. H UT , G. 1987. Consultants group meeting in stable isotope reference samples for geochemical and hydrological investigations, Report to Director General, International Atomic Energy Agency, 16– 18. IAEA/WMO. 2004. Global Network of Isotopes in Precipitation. The GNIP Database. Website: http://isohis.iaea.org. J AHNKE , R. A. & J AHNKE , D. B. 2004. Calcium carbonate dissolution in deep sea sediments: reconciling microelectrode, pore water and benthic flux chamber results. Geochimica et Cosmochimica Acta, 68, 47–59. J ORISSEN , F. J. 1999. Benthic foraminiferal microhabitats below the sediment-water interface. In: S EN G UPTA , B. K. (eds) Modern Foraminifera. Kluwer, 161– 179. K IM , S.-T. & O’N EIL , J. R. 1997. Equilibrium and nonequilibrium oxygen isotope effects in synthetic carbonates. Geochimica et Cosmochimica Acta, 61, 3461– 3475. K LITGAARD -K RISTENSEN , D., S EJRUP , H. P. & H AFLIDASON , H. 2002. Distribution of recent calcareous benthic foraminifera in the northern North Sea and relation to the environment. Polar Research, 21, 275–282. K RISTENSEN , D. K., S EJRUP , H. P., H AFLIDASON , H., B ERSTAD , I. M. & M IKALSEN , G. 2004. Eight-hundred-year temperature variability from the Norwegian continental margin and the North Atlantic thermohaline circulation. Paleoceanography, 19, art. no.-PA2007. L EE , A. J. & R AMSTER , J. W. 1981. Atlas of the Seas Around the British Isles., MAFF, Lowestoft. L EE , J. J., F ABER , W. W., A NDERSON , O. R. & P AWLOWSKI , J. 1991. Life cycles of foraminifera. In: L EE , J. J. & A NDERSON , O. R. (eds) Biology of Foraminifera. Academic Press Limited, London, 285– 335.
SEASONAL ISOTOPE EFFECT IN FORAMINIFERA L YNCH -S TIEGLITZ , J., C URRY , W. B. & S LOWEY , N. 1999. A geostrophic transport estimate for the Florida Current from the oxygen isotope composition of benthic foraminifera. Paleoceanography, 14, 360–373. M ACKENSEN , A., H UBBERTON , H. W., B ICKERT , T., F ISHER , G. & F UTTERER , D. K. 1993. The d13C in benthic foraminiferal tests of Fontbotia –Wuellerstorfi (Schwager) relative to the d13C of dissolved inorganic carbon in southern-ocean-deep-water-implications for glacial ocean circulation models. Paleoceanography, 8, 587– 610. M ARRET , F., S COURSE , J. & A USTIN , W. 2004. Holocene shelf-sea seasonal stratification dynamics: a dinoflagellate cyst record from the Celtic Sea, NW European shelf. The Holocene, 14, 689–696. M C C ONNAUGHEY , T. 1989a. C-13 and O-18 isotopic disequilibrium in biological carbonates. 1. Patterns. Geochimica et Cosmochimica Acta, 53, 151–162. M C C ONNAUGHEY , T. 1989b. C-13 and O-18 isotopic disequilibrium in biological carbonates. 2. Invitro simulation of kinetic isotope effects. Geochimica et Cosmochimica Acta, 53, 163–171. M C C ONNAUGHEY , T. A., B URDETT , J., W HELAN , J. F. & P AULL , C. K. 1997. Carbon isotopes in biological carbonates: respiration and photosynthesis. Geochimica et Cosmochimica Acta, 61, 611–622. M C C ORKLE , D. C., C ORLISS , B. H. & F ARNHAM , C. A. 1997. Vertical distributions and stable isotopic compositions of live (stained) benthic foraminifera from the North Carolina and California continental margins. Deep Sea Research Part I: Oceanographic Research Papers, 44, 983–1024. M C C ORKLE , D. C., K EIGWIN , L. D., C ORLISS , B. H. & E MERSON , S. R. 1990. The influence of microhabitats on the carbon isotopic composition of deep-sea benthic foraminifera. Paleoceanography, 5, 161–185. M C C REA , J. M. 1950. On the isotopic chemistry of carbonates and a paleotemperature scale. Journal of Chemical Physics, 18, 849– 857. M IKALSEN , G., F LATEBO , T. & S EJRUP , H. P. 1999. Distributions and ecology of benthic foraminiferal faunas in fjord basins of western Norway. PhD Thesis, University of Bergen. M IKALSEN , G., S EJRUP , H. P. & A ARSETH , I. 2001. LateHolocene changes in ocean circulation and climate: foraminiferal and isotopic evidence from Sulafjord, western Norway. The Holocene, 11, 437–446. M URRAY , J. W. 1971. An Atlas of British Recent Foraminiferids. Heinemann Educational Books, London. M URRAY , J. W. 1983. Population-dynamics of benthic foraminifera – results from the Exe Estuary, England. Journal of Foraminiferal Research, 13, 1–12. M URRAY , J. W. 1991. Ecology and Palaeoecology of Benthic Foraminifera. Longman, Harlow. M URRAY , J. W. 2000. Revised taxonomy, an atlas of British recent foraminiferids. Journal of Micropalaeontology, 19, 44. M URRAY , J. W. 2003a. Foraminiferal assemblage formation in depositional sinks on the continental shelf west of Scotland. Journal of Foraminiferal Research, 33, 101–121.
171
M URRAY , J. W. 2003b. An illustrated guide to the benthic foraminifera of the Hebridean Shelf, west of Scotland, with notes on their mode of life. Palaeontologica Electronica, 5, 1– 31. M URRAY , J. W. & B OWSER , S. S. 2000. Mortality, protoplasm decay rate, and reliability of staining techniques to recognize ‘living’ foraminifera: a review. Journal of Foraminiferal Research, 30, 66–70. M URRAY , J. W., A LVE , E. & C UNDY , A. 2003. The origin of modern agglutinated foraminiferal assemblages: evidence from a stratified fjord. Estuarine, Coastal and Shelf Science, 58, 677 –697. N ORGAARD -P EDERSEN , N., A USTIN , W. E. N., H OWE , J. A. & S HIMMIELD , T. 2006. The Holocene record of Loch Etive, western Scotland: influence of catchment and relative sea level changes. Marine Geology, 228, 55– 71. O’N EIL , J. R., C LAYTON , R. N. & M AYEDA , T. 1969. Oxygen isotope fractionation in divalent metal carbonates. Journal of Chemical Physics, 51, 5547– 5558. O WEN , R., K ENNEDY , H. & R ICHARDSON , C. 2002. Isotopic partitioning between scallop shell calcite and seawater: effect of shell growth rate. Geochimica et Cosmochimica Acta, 66, 1727–1737. P INGREE , R. D. & G RIFFITHS , D. K. 1978. Tidal fronts on the shelf seas around the British Isles. Journal of Geophysical Research, 89, 4615– 4622. P OLYAK , L., S TANOVOY , V. & L UBINSKI , D. J. 2003. Stable isotopes in benthic foraminiferal calcite from a river – influenced Arctic marine environment, Kara and Pechora Seas. Paleoceanography, 18, 1003. R ATHBURN , A. E., C ORLISS , B. H., T APPA , K. D. & L OHMANN , K. C. 1996. Comparisons of the ecology and stable isotopic compositions of living (stained) benthic foraminifera from the Sulu and South China Seas. Deep Sea Research Part I: Oceanographic Research Papers, 43, 1617–1646. R OHLING , E. J. & C OOKE , S. 1999. Stable oxygen and carbon isotopes in foraminiferal carbonate shells. In: G UPTA , B. K. S. (ed.) Modern Foraminifera. Kluwer Academic Publishers, 239–258. S CHMIEDL , G., P FEILSTICKER , M., H EMLEBEN , C. & M ACKENSEN , A. 2004. Environmental and biological effects on the stable isotope composition of recent deep-sea benthic foraminifera, from the western Mediterranean Sea. Marine Micropaleontology, 51, 129– 152. S COTT , G., T HOMPSON , L., H ITCHIN , R. & S COURSE , J. 1998. Observations on selected salt-marsh and shallow-marine species of agglutinating foraminifera: grain size and mineralogical selectivity. Journal of Foraminiferal Research, 28, 261– 267. S COTT , G. A., S COURSE , J. D. & A USTIN , W. E. N. 2003. The distribution of benthic foraminifera in the Celtic Sea: the significance of seasonal stratification. Journal of Foraminiferal Research, 33, 32–61. S COURSE , J. D. & A USTIN , W. E. N. 2002. Quaternary shelf sea palaeoceanography: recent developments in Europe. Marine Geology, 191, 87– 94. S COURSE , J. D., K ENNEDY , H., S COTT , G. A. & A USTIN , W. E. N. 2004. Stable isotopic analyses of modern benthic foraminifera from seasonally stratified shelf
172
A. G. CAGE & W. E. N. AUSTIN
seas: disequilibria and the ‘seasonal effect’. Holocene, 14, 747 –758. S HACKLETON , N. J. 1974. Attainment of isotopic equilibrium between ocean water and the benthonic foraminifera genus Uvigerina: isotopic changes in the ocean during the last glacial. Centre National de la Recherche Scientifique Colleagues Internationeaux, 219, 203–209. S PERO , H. J., B IJMA , J., L EA , D. W. & B EMIS , B. E. 1997. Effect of seawater carbonate concentration on foraminiferal carbon and oxygen isotopes. Nature, 390, 497–500. T ACHIKAWA , K. & E LDERFIELD , H. 2002. Microhabitat effects on Cd/Ca and delta C-13 of benthic foraminifera. Earth and Planetary Science Letters, 202, 607–624. U REY , H. C. 1947. The thermodynamic properties of isotopic substances. Journal of the Chemical Society, 562– 581. V ILKS , G. & D EONARINE , B. 1988. Labrador Shelf benthic foraminifera and stable oxygen isotopes of Cibicides lobatulus related to the Labrador Current. Canadian Journal of Earth Sciences [Journal Canadien des Sciences de la Terre], 25, 1240–1255. V INCENT , E., K ILLINGLEY , J. S. & B ERGER , W. H. 1981. Stable isotope composition of benthic foraminifera from the equatorial Pacific. Nature, 289, 639– 643. V INOT -B ERTOUILLE , A. & D UPLESSY , J. C. 1973. Individual isotopic fractionation of carbon and oxygen in benthic foraminifera. Earth and Planetary Science Letters, 18, 247– 252. W EFER , G., B ERGER , W. H., B IJMA , J. & F ISCHER , G. 1999. Clues to Ocean History: a brief overview of proxies. In: F ISCHER , G. & W EFER , G. (eds) Use of
Proxies in Palaeoceanography: Examples from the South Atlantic. Springer-Verlag, Berlin, 1– 68. W ILSON -F INELLI , A., C HANDLER , G. T. & S PERO , H. J. 1998. Stable isotope behavior in paleoceanographically important benthic foraminifera: results from microcosm culture experiments. Journal of Foraminiferal Research, 28, 312– 320. W OLF -G LADROW , D. A., B IJMA , J. & Z EEBE , R. E. 1999. Model simulation of the carbonate chemistry in the microenvironment of symbiont bearing foraminifera. Marine Chemistry, 64, 181 –198. W OODRUFF , F., S AVIN , S. M. & D OUGLAS , R. G. 1980. Biological fractionation of oxygen and carbon isotopes by Recent benthic foraminifera. Marine Micropaleontology, 5, 3 –11. Z AHN , R., W INN , K. & S ARNTHEIN , M. 1986. Benthic foraminiferal d13C and accumulation rates of organic carbon: Uvigerina peregrina group and Cibicidoides wuellerstorfi. Paleoceanography, 1, 27– 42. Z EEBE , R. E. 1999. An explanation of the effect of seawater carbonate concentration on foraminiferal oxygen isotopes. Geochimica et Cosmochimica Acta, 63, 2001– 2007. Z EEBE , R. E. & W OLF -G LADROW , D. A. 2001. CO2 in Seawater: Equilibrium, Kinetics, Isotopes. Elsevier Oceanography Series, 65. Elsevier. Z EEBE , R. E., W OLF -G LADROW , D. A. & J ANSEN , H. 1999. On the time required to establish chemical and isotopic equilibrium in the carbon dioxide system in seawater. Marine Chemistry, 65, 135 –153. Z HU , Q. Z., A LLER , R. C. & F AN , Y. Z. 2006. Twodimensional pH distributions and dynamics in bioturbated marine sediments. Geochimica et Cosmochimica Acta, 70, 4933– 4949.
Isotopic variability in the intertidal acorn barnacle Semibalanus balanoides: a potentially novel sea-level proxy indicator K. F. CRAVEN1, M. I. BIRD1, W. E. N. AUSTIN1 & J. WYNN1,2 1
School of Geography and Geosciences, University of St. Andrews, St. Andrews, Fife, KY16 9AJ, Scotland, UK (e-mail:
[email protected]) 2
Current address: Department of Geology, University of South Florida, 4202 E Fowler Ave, SCA 528, Tampa, FL 33620, USA
Abstract: We report variations in the d13C and d18O values of barnacle skeletal carbonate as well as the d13C and d15N value of tissue from specimens of the acorn barnacle S. balanoides, as a function of elevation within its living range on the Isle of May, Scotland. Individuals were sampled over a 3.50 m range at 0.25 m intervals (1.00–4.50 m above ordnance datum). Carbonate d18O values (2.44 + 0.13‰ [1s], n ¼ 45) and tissue d15N values (8.17 + 0.42‰, n ¼ 15) do not vary systematically with elevation. The d13C value of shell carbonate increases with elevation by c. 1‰ (total range: 20.77‰ to þ0.63‰), but the variability between samples at the same elevation suggests that this relationship will be of limited use in constraining palaeo-elevation. By contrast, tissue d13C values show systematic variation with elevation, increasing by c. 8‰ (total range: 219.36‰ to 28.77‰) with increasing elevation. These results suggest that there is potential to use the tissue d13C values to determine the elevation of a Fixed Biological Indicator (FBI) such as S. balanoides within its living range. If this is also true of the organic matrix of the carbonate skeleton, and if this organic matrix is preserved in Holocene FBIs, then the potential exists to use carbon isotopes to more precisely constrain the palaeo-elevation of FBIs within their living range and hence palaeo sea-level. The small range of carbonate d18O values suggests that oxygen isotopes in FBIs can be used to constrain water d18O values if an independent estimate of temperature is available, or temperature if an independent estimate of salinity is available.
Sea-level reconstructions have played a significant role in shaping our understanding of both palaeogeography and palaeoclimate through the Quaternary. While the major changes in global sea-level since the Last Glacial Maximum are relatively well known (Fairbridge 1961; Shepard 1964; Fleming et al. 1998; Lambeck 2002), the fine structure of sea-level change regionally and globally is less well-known. A more detailed understanding of the trajectory of sea-level change in the Holocene in particular, is a pre-requisite for predicting the likely course of sea-level change in the future, but many sea-level proxies lack the resolution to unambiguously identify changes in sea-level at the decimetre scale. Submerged wave-cut notches (Blanchon et al. 2002), evidence for the catastrophic drainage of large volumes of meltwater from ice-dammed lakes into the ocean (Clarke et al. 2001) and buried barrier complexes (Stapor & Stone 2004) all suggest that millennial/centennial oscillations in sea-level must have occurred in the Holocene. Some interpretations of regional Holocene sea-level
curves and evidence from polar ice-caps support the hypothesis that eustatic sea-level during the mid to late Holocene may have oscillated by 0.2 to 0.4 m, either due to climate-induced variability in the rates of freshwater delivery to, or withdrawal from, the alpine and polar ice-caps or steric effects on sea-level associated with warming and cooling of the global ocean (Goodwin 1998; Baker & Haworth 2000a, b; Baker et al. 2001a, b; Siddall et al. 2003; Rohling et al. 2003; Stone et al. 2003). While the authors of Working Group 1 of the IPCC (2001) consider that fluctuations in post-mid-Holocene eustatic sea-level are unlikely to have exceeded 0.3–0.5 m, they make no statement as to the likelihood of such changes having occurred. Some of the most reliable indicators of past sealevel are the range of carbonate secreting inter-tidal organisms that have been used to develop Holocene sea-level curves in many parts of the world. These include: oysters (Tjia et al. 1983, 1984; Yim & Huang 2002; Maeda et al. 2004; Larcombe & Carter 1998; Nunn et al. 2002; Baker & Haworth
From: AUSTIN , W. E. N. & JAMES , R. H. (eds) Biogeochemical Controls on Palaeoceanographic Environmental Proxies. Geological Society, London, Special Publications, 303, 173– 185. DOI: 10.1144/SP303.12 0305-8719/08/$15.00 # The Geological Society of London 2008.
174
K. F. CRAVEN ET AL.
1997, 2000a, b; Baker et al. 2003), tubeworms (Baker & Haworth 1997, 2000a, b; Baker et al. 2003), reef-forming gastropods (Laborel 1986; Antonioli et al. 1999) and barnacles (Baker & Haworth 1997; Flood & Frankel 1989). These Fixed Biological Indicators (FBIs) grow within defined limits in the intertidal zone and therefore can be related to mean sea-level. In addition, their calcareous parts remain fixed in position upon death and thus can be confidently related to palaeo sea-level at the time of their growth, with the time of growth being determinable by radiocarbon or uranium series dating techniques (e.g. Maeda et al. 2004). Under exceptional circumstances, when the FBI has a very narrow vertical range and/or tidal amplitude is very small, it is possible to estimate the elevation of a former mean sea-level to +0.08 m (Antonioli et al. 1999), or +0.25 m (Lambeck 2002). However in most instances, FBIs occupy a comparatively wide interval within the intertidal zone and/or a large tidal range expands the vertical interval over which an FBI can survive. This commonly increases the uncertainty in relating a fossil FBI with a palaeo sea-level to a metre or more, usually sufficient to obscure any fine structure that might be present in a regional sea-level curve. The development of a chemical proxy that could enable the more accurate determination of the elevation of any FBI species within its vertical growth range would greatly reduce the uncertainty associated with the estimation of palaeo-sea level from FBIs. Some evidence that the development of such a proxy may be possible is provided by Pilkey & Harriss (1966) who found that shell trace element compositions (Mg, Sr and Mn) of barnacles and oysters varied predictably between species, depths and location but with elevation exerting more control on trace element composition than other factors. In this study, we test the possibility that the isotopic composition of intertidal FBIs at open coastal locations is primarily determined by immersion time, using samples of the acorn barnacle Semibalanus balanoides collected from bottom to top of its 3.5 m living-range on the Isle of May, Firth of Forth, Scotland.
Study area and samples Ecology and physiology of Semibalanus balanoides S. balanoides is an intertidal acorn barnacle with a living-range generally extending between low and high water marks. It is a northern hemisphere
species confined to regions where the minimum monthly mean temperature is less than 7.2 8C (Stubbings 1975) and is the dominant barnacle on the east coasts of Scotland and England. The upper limit of the organism coincides closely to high water neap tides in sheltered areas, but the vertical range increases with increasing exposure (Stubbings 1975), possibly due to swell and spray reaching higher elevations. The barnacle is composed of 6 calcareous plates that overlap to form its shell. Two movable opercular valves, formed from calcite, close the apical opening. The carina-rostral axis is the long axis of this aperture and can be used as a measure of size for individuals. Yearly recruits tend to settle between April and June and maximum growth of the shell occurs during the spring, with growth reduction in the summer and negligible growth in the winter (Bourget & Crisp 1975). Calcium carbonate (CaCO3) is precipitated as calcite in S. balanoides with no aragonite formation, and direct experiment has indicated that deposition of the calcareous shell occurs only during submersion (Bourget & Crisp 1975). Calcifying marine organisms gain their calcium ions from the surrounding seawater (Erez 1978) while carbon is derived both from dissolved inorganic carbon (DIC) in the water (McConnaughey 1989a, b), and metabolic carbon (Tanaka et al. 1986). S. balanoides is capable of remaining dry for several days at a time, and during emersion aerial respiration can occur (Grainger & Newell 1965; Davenport & Irwin 2003). Barnacle calcite is precipitated out of oxygen-isotopic equilibrium with sea water, with d18O higher by around 1.3‰ (Killingley & Newman 1983). The magnitude of fractionation is not known for carbon isotopes.
Site description The Isle of May is situated c. 8 km off the east coast of Scotland at the mouth of the Firth of Forth (Fig. 1). This location was chosen because of its open coastal aspect, well-mixed surface layer and minimal surface freshwater runoff due to the small area of the island. A south-facing rock section at the southern end of the island was chosen for barnacle removal (Fig. 2). The sampling site (Lady’s Beds; OSGB map grid reference: NT 662 986) is located in a sheltered bay, protected from all but south-easterly swells. The swell at Lady’s Beds was estimated at 0.1 m, compared to 0.3 m outside the bay, on the day of sampling. The barnacle zone extended to 3.62 m above low tide on 8th August 2005, with a spring tidal range of 3.92 m recorded on this date.
ISOTOPIC VARIABILITY IN THE INTERTIDAL ACORN BARNACLE
175
Fig. 1. Location of the Isle of May and places mentioned in the text.
Experimental methods Sample collection Barnacles were collected over 2 days at the beginning of August 2005. Individual specimens of S. balanoides were removed from 15 sampling levels at 0.25 m intervals up the rockface. Estimated barnacle coverage of the rock varied from 20 –85% at different sampling levels. Due to the absence of a local benchmark, the elevation at the base of the transect was determined by recording the difference in height between sea-level at low tide and the lowest sampling level. Air pressure recorded for the day suggests that actual tide height would not deviate greatly from tidal predictions for the day. Barnacles were collected from within a 10 10 cm area surrounding the marker of the sampling level (i.e. +5 cm). Samples were frozen at 220 8C, upon return from the island. In addition, 25 individuals, ranging from the smallest to the largest, were collected from sampling level 8 (midway up the section; immersion time 13:09 hrs/day) to determine whether size (age) affected the isotopic composition of individuals.
Water samples were taken from the site at high tide in order to determine the stable oxygen isotopic composition of local seawater. Water was sampled from depths of 0 m, 1 m and 3 m with temperature and salinity being recorded immediately upon collection.
Isotopic analysis Three individuals from each sampling level were selected for isotopic analysis, with the smallest (youngest) individuals being selected on the basis of aperture diameter (Stubbings 1975), with a mean aperture carina-rostral axis distance of 1.99 + 0.39 mm (1s). All 45 barnacles selected in this manner are believed to have settled on the rock face during the year of collection, thus reducing the potential for age-related differences in isotopic composition. All samples selected for analysis were dried at 60 8C for 20 hrs. For carbonate analysis, organic tissue was initially removed using forceps/dissection probe and samples were immersed in 1 ml of 30% H2O2 overnight to oxidize the remaining organic tissue
176
K. F. CRAVEN ET AL.
Fig. 2. Photograph taken from the inter-tidal sampling section at Lady’s Beds, Isle of May, looking north, shortly before low tide on 8th August 2005. Note the distinct upper limit of the barnacle zone, seen as a horizontal band; for scale, the exposed (emersed) barnacle zone is approximately 3.6 m high.
(Pilkey & Harriss 1966). Any remaining H2O2 was removed by rinsing, and the samples placed in an oven at 60 8C until dry. Barnacle shells were ground to a powder using a mortar and pestle, and 250– 350 mg of each sample was measured into clean vials for isotopic analysis. Samples were sealed and flushed with He gas in a Gasbench coupled to a Finnegan Deltaþ XL mass spectrometer. Five drops of 100% phosphoric acid (H3PO4) were added to each vial, and samples left for 24 hrs at 25 8C before analysis for d13C and d18O by continuous flow mass spectrometry. Carrara marble standards were analysed in parallel with the samples and the precision of analysis was 0.02‰ for both d13C and d18O (standard deviation of 16 standards with similar mass to samples). The average deviation from the mean for replicate samples analysed in different runs was +0.11‰ for d13C and +0.08‰ for d18O (n ¼ 14). Due to the small size of the barnacles, three individuals from each level were combined for analysis of d13C and d15N in organic tissue matter. Samples were roughly broken using a mortar and pestle prior to immersion in 0.5 ml of sulphurous acid for 24 hrs to remove carbonates. A further 0.5 ml of acid was added and the samples left for another 24 hrs. The acid was removed and the samples were rinsed with deionized water, freeze-dried and then ground using a mortar and pestle. Samples of 4 mg from each level were weighed into tin
capsules for analysis by elemental analyser coupled to a Finnegan Deltaþ XL mass spectrometer operated in continuous flow mode. Peak jumping and He dilution of the CO2 peak enabled the determination of d13C and d15N on the same aliquot with an uncertainty of + 0.10‰ and +0.20‰ for d13C and d15N, respectively. All results are reported as per mil (‰) deviations from the accepted values for the international standards VSMOW (d18Owater), VPDB (d13C/ d18O) and AIR (d15N).
Surface water temperature, salinity and oxygen isotopes Temperature and salinity data from the Firth of Forth were available for the period October 1998 to September 2003, collected at intervals of approximately three months by researchers from ‘Marlab’, Aberdeen. Average monthly mean temperatures, based on a 37.1 km (20 nautical mile) grid (200 latitude by 300 longitude) are also available from Austin et al. (2006). The oxygen isotope composition of seawater (d18Owater) was calculated using a regional salinity: d18Owater relationship established for Scottish coastal waters (method in Austin & Inall 2002; Austin et al. 2006). Incorporation of d18Owater into the calcite shell of an organism (d18Ocalcite) is
ISOTOPIC VARIABILITY IN THE INTERTIDAL ACORN BARNACLE
temperature dependent (McCrea 1950). In order to calculate the predicted or ‘equilibrium’ isotopic composition of a calcareous organism (d18OEq.calcite) growing in seawater of known isotopic composition and temperature, we employ the temperature: d18O relationship of Bemis et al. (1998) in this study as this relationship was developed for temperatures similar to those in the field area: T(W C) ¼ 16:5 4:8(d18 Ocalcite d18 Owater ) However, since oxygen isotopes in seawater and calcite are measured relative to different standards, we convert from the seawater (VSMOW) to the calcite (VPDB) scale, by applying a simple correction (Hut 1987): d18 Owater(VPDB) ¼ d18 Owater(VSMOW) 0:27
Immersion times The immersion time of an organism is a function of local tidal range and does not change in a linear manner with change in height above chart datum. Thus tidal immersion times for each sampling level were estimated based on best fit to tidal data for the months of June and July 2005 by expressing tidal curves for the region as a mathematical function related to elevation. Total immersion time at each level was calculated for these two months and immersion times are expressed as the average time spent submerged over 24 hrs during this period (Fig. 3, Table 1). Immersion durations were calculated for each period between high tide and the point of emersion (i.e. barnacle exposed to air) using the equation: t ¼ ([t2 t1 ]=180W ) cos1 ([2h (h1 þ h2 )]= [h1 h2 ]) þ t1 where: h ¼ given elevation of sampling; h1 ¼ elevation of high tide; h2 ¼ elevation of low tide; t ¼ time of emersion; t1 ¼ date and time of high tide; t2 ¼ date and time of low tide. and for each period between the point of immersion (i.e. barnacle submerged) and the subsequent high tide using the equation: t0 ¼([t3 t2 ]=180W ) sin1 ([2h0 h2 h3 ]= [h3 h2 ]) þ ([t2 þ t3 ]=2) 0
where: h ¼ given elevation of sampling level; h2 ¼ elevation of low tide; h3 ¼ elevation of high tide; t0 ¼ time of immersion; t2 ¼ time of low tide; t3 ¼ time of high tide.
177
Sine curve immersion Line immersion 18:00
12:00
6:00
1
1.5 2 2.5 3 3.5 4 Elevation above datum (m)
4.5
Fig. 3. Summary figure illustrating the average daily immersion time of the 15 sampling levels in hours/day estimating tidal immersion time as both linear and sine curves. Note the difference between the two possible calculation methods, with a tendency for the linear calculation to overestimate immersion time at low elevations and to underestimate it at high elevations. Sampling Level 15 is 1.00 m above datum while Sampling Level 1 is 4.50 m above datum. The reference datum used to define the base of the sampling section is 2.9 m below ordnance datum.
Table 1. Relationship between sampling level, height above datum and immersion time (‘sine calculation’). The reference datum used to calculate immersion times was 2.9 m below ordnance datum (see text for details) Sampling Level 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Height above datum (immersion time hrs/24hrs) 4.50 m (2:39) 4.25 m (4:26) 4.00 m (6:20) 3.75 m (8:00) 3.50 m (9:23) 3.25 m (10:40) 3.00 m (11:54) 2.75 m (13:09) 2.50 m (14:27) 2.25 m (15:53) 2.00 m (17:37) 1.75 m (19:17) 1.50 m (20:58) 1.25 m (22:11) 1.00 m (23:07)
Results Water chemistry Water chemistry on the day of sampling varied only slightly over the 3 m depth interval that was
178
K. F. CRAVEN ET AL.
sampled. Temperature decreased from 13.8 to 13.4 8C and pH from 8.15 to 8.08 with increasing depth, while total dissolved solids and conductivity increased slightly with depth (30.3 to 33.1 ppt and 60.3 to 66.1 mS respectively). The increases in both of these variables suggest an increase in salinity with depth, possibly associated with a slightly fresher and more buoyant surface layer. However, while water isotope (d18Owater) values ranged from 0.22‰ to 0.54‰ (mean ¼ 0.33‰), they did not co-vary with conductivity.
Carbonate isotopic composition Forty-five individuals were analysed to explore the effects of sample elevation on isotopic composition (Appendix Table A1). Over the entire range of elevation, shell d18O ranged from 2.19‰ to 2.75‰ and d13C ranged from 21.09‰ to 0.94‰ (Fig. 4a, b). A further 25 individual barnacle specimens from sampling level eight were analysed to explore the effects of size on isotopic composition (Appendix Table A1). Shell d18O ranged from 2.31‰ to 2.81‰, and d13C ranged from 20.14‰ to 0.91‰ (Fig. 4c, d).
Fig. 4. Regression analyses for stable isotope data of barnacles sampled from the Isle of May including 95% prediction and confidence intervals and the maximum error associated with the analyses. (a) d13C of shell carbonate relative to immersion time; (b) d18O of shell carbonate relative to immersion time; (c) d13C of shell carbonate from sampling level eight (immersion time 13:09 hrs) relative to aperture diameter; (d) d18O of shell carbonate from sampling level eight relative to aperture diameter; (e) d13C of organic tissue relative to immersion time; (f) d15N of organic tissue relative to immersion time.
ISOTOPIC VARIABILITY IN THE INTERTIDAL ACORN BARNACLE
179
Table 2. Statistical results from regression analyses of stable isotope data. The predictor is the parameter measured. A p-value ,0.05 is statistically significant. ‘Equation of line’ refers to the linear regression. R2 is a measure of how much variation within the data is accounted for by the variation in the predictor Analysis d18O Elevation d18O Size d13C Elevation d13C Size d13C Organic tissue d15N Organic tissue
Predictor
T44-Value
F44-Value
Immersion time Diameter of shell aperture Immersion time Diameter of shell aperture Immersion time
0.31 20.04
Immersion time
R2 (adj)
P-Value
Equation of line
0.10 0.00
0.755 0.971
y ¼ 0.024x þ 2.423 y ¼ 0.001x þ 2.512
5.62 23.29
31.63 10.80
,0.001 0.003
y ¼ 1.224x 2 0.515 y ¼ 20.131x þ 0.749
41.0% 29.0%
4.46
19.86
0.001
y ¼ 10.670x 2 17.750
57.4%
20.59
0.35
0.562
y ¼ 20.254x þ 8.311
0.0%
The d18O values of shell carbonate from sampling level eight display no significant variation with size (Table 2); mean d18O is 2.51 + 0.12‰ (1s). In addition, the d18O value of shell carbonate varied little between the base of the section and the top of the section, with a mean d18O value over all immersion times of 2.44 + 0.13‰ (1s) and no significant correlation between d18O value and the immersion time (see Table 2). The d13C values for carbonate shells from level eight do vary as a function of size with large individuals having lower d13C values than smaller individuals, however, there is a significant amount of variation in carbonate d13C values, especially in smaller individuals, that cannot be attributed to size alone. Predicted d18OEq.calcite values vary greatly for the months for which temperature data (1997– 2003) from the Firth of Forth are available; ranging from 1.94‰ in March (6.7 8C), to 1.00‰ in June (11.0 8C) and 0.61‰ in September (13.1 8C). If the mean June 2005 d18OEq.calcite value is indicative of mean summer values, then shells would be expected to exhibit a d18Ocalcite signature of 1.00 + 0.27‰. As such, the difference between predicted and observed values (Dd18O) is þ1.44‰, which is similar to the isotope disequilibrium of 1.3‰ observed by Killingley & Newman (1983). Further information of the timing and rate of barnacle growth and calcification is required to fully constrain this ‘disequilibrium effect’. Despite a mean within-level standard deviation of 0.31‰ for triplicate analyses, it is apparent that the d13C of shell carbonate ranges from 20.77‰ to þ0.63‰, tending to increase as immersion time increases, and this relationship is significant (Fig. 4a). There is no correlation between d13C and d18O of shell carbonate (Pearson’s coefficient ¼ 0.058; p ¼ 0.705).
0.0% 0.0%
Stable isotopes in organic tissue The d13C values in the dissected tissue samples range widely from around 210‰ at the base to 218‰ at the top of the section (Figure 4e), a considerably greater range than observed in shell carbonate d13C. There is a significant relationship between tissue d13C and immersion time (Table 2). The nitrogen isotope composition of barnacle organic tissues varies erratically along the section, ranging between 7.6‰ and 8.8‰ (Fig. 4f), with no significant relationship evident between immersion time and d15N value (Table 2). This suggests that the variation observed in carbon isotope ratios is not caused by a difference in diet.
Discussion Oxygen isotopes (d18O) in barnacle shells Measured shell carbonate d18O values do not vary significantly with immersion times and values close to 2.44‰ throughout the profile suggest that relatively constant physical environmental factors control the isotope composition of the shells. Barnacles only calcify while submerged, and obtain the oxygen atoms necessary for calcification from the ambient, well-mixed surface water layer. There were slight variations in the water chemistry with depth at the time of sampling, but the relatively constant carbonate d18O values suggest that depth-related differences in water composition have little effect on shell chemistry. The barnacles sampled also exhibited relatively stable d18O values across all sizes. Again, this suggests a strong seasonal bias upon the incorporation of the d18O signal into the shell (e.g. Austin et al. 2006), which is consistent with the observation that the specimens sampled represent a single growth cohort.
180
K. F. CRAVEN ET AL.
Small, non-systematic variations between samples do occur and these may be related to differences in genotype (Weiner & Dove 2003), to small deviations from the 1.3‰ disequilibrium enrichment factor typical of barnacles (Killingley & Newman 1983), or to slight differences in the season and rate of calcification between individual specimens.
Carbon isotopes (d13C) in barnacle shells and tissues There are significant relationships between the d13C values of both shell carbonate and organic tissues with immersion time. There are a large number of explanations for d13C variations in the carbonate skeletons of organisms in nature, however few of the available explanations would produce the small, systematic change of c. 1‰ observed over the 3.5 m of sample profile. Only kinetic effects, metabolic effects, varying pH or the local influx of 13C-depleted waters are likely to have been able to significantly influence shell isotope composition over the spatial scale sampled. An influx of 13 C-depleted meteoric water derived from adjacent terrestrial areas should lead to covariance between d13C and d18O in the shell carbonate. This is not observed and hence can be discounted as the mechanism behind the observed relationship. The same is true for pH. Although it has been suggested that a decrease in pH can result in higher d13C values for calcifying organisms (Keatings et al. 2002), an increase in d18O would also be expected. In this study, pH was observed to decrease by 0.07 from 8.15 to 8.08 between 0 m and 3 m depth. This decreasing pH could be expected to yield increasing d13C values for individuals further down the shore (i.e. with increased submersion times) as is observed. However, the lack of covariance between d13C and d18O in shell carbonate suggests that pH is not the dominating factor governing carbon isotopic fractionation of S. balanoides. It is possible that kinetic effects determine shell d13C values. However, the kinetic (McConnaughey 1989a, b) and carbonate (Adkins et al. 2003) models for skeletal carbonate accretion in marine organisms both predict simultaneous depletions in d13C and d18O. This is not observed and since both kinetic and carbonate effects are thought to dominate isotope fractionation in carbonates (McConnaughey 1989a), the absence of a strong correlation between the two isotopes in the samples from this study suggests that neither of these processes is responsible for the observed trends. Shanahan et al. (2005) investigated isotopic variability in gastropods in near constant
environments and reported variations in d13C without corresponding changes in d18O, similar to the trends observed in this study. The shallowest-dwelling species had the lowest d13C values and Shanahan et al. (2005) believed this to be due to an increased incorporation of atmospheric CO2 in their shell. Increased uptake of atmospheric, as opposed to seawater, CO2 may lead to a reduction in d13C values due to the gas ratio differences between CO2 and O2 in both these environments (McConnaughey et al. 1997). The atmospheric CO2/O2 ratio is c. 0.0017, while in seawater it is about 30 times higher at 0.005. As a result terrestrial organisms absorb less environmental CO2 while obtaining O2 for respiration. It is hypothesized that air-breathing calcifying organisms incorporate more metabolic carbon into their skeletons to compensate for this shortfall (McConnaughey et al. 1997). S. balanoides, on emersion, empties seawater from its shell and fills its mantle cavity with atmospheric air through a pneumostome. This air is used for respiration (Grainger & Newell 1965; Davenport & Irwin 2003). The air bubble is replenished periodically for the first two hours of exposure before the pneumostome is closed completely to prevent desiccation and at that point hypoxic conditions begin to develop within the shell (Davenport & Irwin 2003). Upon re-immersion, the gas bubble is expelled and replaced with water. Analyses of the gas bubble emitted have revealed varying O2 concentrations related to exposure times, but an absence of CO2 (Grainger & Newell 1965). It is believed that CO2 is absent due to its increased solubility relative to O2; thus, instead of being expelled, CO2 is absorbed into the tissues of the organism. This process could explain the comparatively large changes in tissue d13C of up to 8‰ that are related to changes in immersion time. Metabolic carbon is isotopically lighter than molecular carbon due to the preferential uptake of 12C from food and release through respiration (Michner & Schell 1994). Although there is controversy over exactly how much metabolic carbon is incorporated in skeletal carbon across numerous taxa (e.g. Shanahan et al. 2005), there is much evidence for its presence (Erez 1978). One estimate places metabolic carbon comprising about 56% of barnacle shell carbonate (Tanaka et al. 1986) and the same researchers predict that if 50% of the shell is derived from metabolic carbon, d13C values could vary between 0.3 and 7.5‰ lower than ambient DIC. McConnaughey (2003) argues that metabolic carbon makes up a small proportion of overall skeletal carbon and predicts that respired carbon can only affect overall skeletal carbon by 1–2‰.
ISOTOPIC VARIABILITY IN THE INTERTIDAL ACORN BARNACLE
However, without the influence of kinetic effects, there is nothing to mask the inclusion of isotopically lighter metabolic CO2. In this experiment, a variation of about 1‰ is observed up the shoreface; in keeping with predictions of a metabolic carbon origin for part of the skeletal carbonate and consistent with the large systematic variations observed in tissue d13C values. Although metabolic fractionation does initially form CO2 that is isotopically depleted in both 13C and 18O, this CO2 is produced in the mitochondria of the cells. Metabolic CO2 must then be transported from these organelles to the site of calcification. During transport, oxygen atoms may exchange with water atoms within cells, catalysed by the enzyme carbonic anhydrase (McConnaughey 1989a). Thus the d18O of the metabolic CO2 would remain in equilibrium with water and no deviation in d18O would be observed; again, in keeping with our observations.
Sea-level prediction It is clear from the results of this study that, for barnacles at least, there is no potential for using the d18O value of shell carbonate as a predictor of elevation within living range in the intertidal zone, and hence oxygen-isotopes will be of no use as a proxy for refining estimates of palaeo sealevel. Indeed, the consistency in the d18O composition between specimens of varying size and position on the shoreface implies that palaeotemperature estimates could be derived from fossil barnacles if an independent estimate of salinity were available. Although large within level variation does exist for the d13C data presented here, there is potential for using d13C as a predictor of the elevation of a specimen within its living range in the intertidal zone in future studies. While carbonate d13C values vary around 1‰ from the base of the section to the top this is unlikely to be a useful tool for determining elevation due to the range of d13C that exists in samples from the same elevation. The larger range in organic tissue d13C of 8‰ offers some support for the assertion that more accurate estimations of elevation could be made through a reduction in associated error. Unfortunately, in this study, the combining of tissue samples from individual barnacles at the different sampling levels prior to analysis, as well as the small sample size, makes it difficult to estimate a meaningful error based on the 95% PI. Nevertheless, the large range of values observed in organic tissues in this study and the coherent variations observed up the shoreface do suggest that, with further research, it may be possible to develop
181
robust estimates of immersion time from FBIs preserved in the geological record. While the organic body-parts of FBIs will not survive long beyond the death of the organism, the carbonate matrix of FBIs contains 1– 3% organic matrix (Marin & Luquet 2005) and the organic matrix of biogenic carbonates has been shown to survive in harsh terrestrial environments for other organisms (Bird et al. 2003). Therefore, if the integrity of this organic matter is retained in the organic matrix of fixed intertidal carbonates, improved estimates of palaeo sea-level could be achieved.
Conclusion Although many factors may potentially contribute to variations in the d18O and d13C of the shell of S. balanoides, this study suggests that the main cause of variation in the carbon isotope ratio is differences in the CO2/O2 ratio between the air and seawater. This difference is expressed in S. balanoides through the uptake of atmospheric air by the barnacle on emersion and results in an increase of isotopically lighter metabolic carbon, leading to higher d13C values with decreasing exposure to the atmosphere. Both kinetic and carbonate models were dismissed as explanations of the observed trends due to a lack of covariance between oxygen and carbon isotope ratios. Our stable isotope results for S. balanoides suggest that some isotope proxies have the potential to yield information on elevation and hence may be of use in more tightly constraining past sea-level variations, although carbonate d18O and tissue d15N values are not sensitive to elevation with living range. Shell carbonate d13C exhibits an enrichment of c. 1‰ with increasing immersion time but this variation is unlikely to be large enough to be useful in constraining elevation due to the magnitude of variability amongst samples collected from the same elevation. In contrast, d13C values of barnacle tissues in these samples exhibit a much larger range of 8‰, and this variation appears to be strongly correlated with immersion time. If the strong sensitivity of tissue organic d13C value to immersion time is inherited by carbonate matrix organic matter, then the potential exists, with further work, to provide an estimate of palaeo sea-level from FBIs, potentially further refined by trace element variations (Pilkey & Harriss 1966). The authors are grateful to Scottish Natural Heritage for allowing samples to be collected from the Isle of May; ‘Marlab’ Aberdeen for providing oceanographic data, and C. Marr for her invaluable help with determining immersion times. We also thank K. Rogers and C. Elder.
182
K. F. CRAVEN ET AL.
Appendix Table A1. Isotope table: stable isotope data used in text Sample
d13Ccarbonate
d18Ocarbonate
Immersion time (hrsday21)
Elevation (m)
Elevation IOM05E01 20.79 2.48 02:39 4.50 IOM05E02 21.09 2.53 – – IOM05E03 20.42 2.60 – – IOM05E04 20.77 2.30 04:26 4.25 IOM05E05 0.16 2.46 – – IOM05E06 20.28 2.37 – – IOM05E07 20.47 2.23 06:20 4.00 IOM05E08 0.10 2.30 – – IOM05E09 20.95 2.42 – – IOM05E10 0.47 2.19 08:00 3.75 IOM05E11 0.22 2.23 – – IOM05E12 0.01 2.44 – – IOM05E13 20.04 2.41 09:23 3.50 IOM05E14 0.02 2.34 – – IOM05E15 0.33 2.59 – – IOM05E16 20.35 2.58 10:40 3.25 IOM05E17 20.26 2.51 – – IOM05E18 0.15 2.37 – – IOM05E19 0.56 2.63 11:54 3.00 IOM05E20 0.60 2.59 – – IOM05E21 0.38 2.38 – – IOM05E22 0.59 2.61 13:09 2.75 IOM05E23 0.69 2.75 – – IOM05E24 0.61 2.47 – – IOM05E25 0.32 2.31 14:27 2.50 IOM05E26 20.44 2.30 – – IOM05E27 0.42 2.36 – – IOM05E28 0.38 2.34 15:53 2.25 IOM05E29 20.05 2.50 – – IOM05E30 0.34 2.41 – – IOM05E31 0.66 2.38 17:37 2.00 IOM05E32 0.41 2.62 – – IOM05E33 0.55 2.33 – – IOM05E34 0.47 2.26 19:17 1.75 IOM05E35 20.02 2.44 – – IOM05E36 0.45 2.56 – – IOM05E37 0.29 2.52 20:58 1.50 IOM05E38 20.39 2.54 – – IOM05E39 0.94 2.61 – – IOM05E40 0.70 2.37 22:11 1.25 IOM05E41 0.86 2.47 – – IOM05E42 0.19 2.56 – – IOM05E43 0.55 2.38 23:07 1.00 IOM05E44 0.64 2.40 – – IOM05E45 0.70 2.19 – – --------------------------------------------------------------------------------------------------------------------------------------Size IOM05S01 0.19 2.40 13:09 2.75 IOM05S02 0.66 2.51 – – IOM05S03 0.74 2.81 – – IOM05S04 0.59 2.59 – – IOM05S05 0.91 2.59 – – IOM05S06 0.45 2.41 – – IOM05S07 0.39 2.49 – – IOM05S08 0.59 2.78 – – (Continued)
ISOTOPIC VARIABILITY IN THE INTERTIDAL ACORN BARNACLE
183
Table A1. Continued IOM05S09 0.28 2.31 – – IOM05S10 0.53 2.46 – – IOM05S11 0.66 2.49 – – IOM05S12 0.33 2.43 – – IOM05S13 0.34 2.53 – – IOM05S14 0.63 2.42 – – IOM05S15 0.06 2.38 – – IOM05S16 0.27 2.35 – – IOM05S17 0.63 2.57 – – IOM05S18 0.17 2.48 – – IOM05S19 0.24 2.43 – – IOM05S20 0.03 2.54 – – IOM05S21 0.27 2.61 – – IOM05S22 0.06 2.68 – – IOM05S23 20.14 2.54 – – IOM05S24 20.01 2.56 – – IOM05S25 0.43 2.38 – – --------------------------------------------------------------------------------------------------------------------------------------d13Corganic d15Norganic Organics IOM05O01 219.36 8.12 02:39 4.50 IOM05O02 218.11 8.46 04:26 4.25 IOM05O03 210.89 7.71 06:20 4.00 IOM05O04 217.92 8.02 08:00 3.75 IOM05O05 211.24 7.82 09:23 3.50 IOM05O06 211.31 8.76 10:40 3.25 IOM05O07 210.71 7.78 11:54 3.00 IOM05O08 213.88 8.57 13:09 2.75 IOM05O09 28.77 8.76 14:27 2.50 IOM05O10 28.68 8.72 15:53 2.25 IOM05O11 29.53 8.51 17:37 2.00 IOM05O12 28.79 8.17 19:17 1.75 IOM05O13 210.12 7.75 20:58 1.50 IOM05O14 29.31 7.77 22:11 1.25 IOM05O15 28.77 7.63 23:07 1.00
References A DKINS , J. F., B OYLE , E. A., C URRY , W. B. & L UTRINGER , A. 2003. Stable isotopes in deep-sea corals and a new mechanism for “vital effects”. Geochimica et Cosmochimica Acta, 67, 1129– 1143. A NTONIOLI , F., C HEMELLO , R., I MPROTA , S. & R IGGIO , S. 1999. Dendropoma lower intertidal reef formations and their palaeoclimatological significance, NW Sicily. Marine Geology, 161, 155 –170. A USTIN , W. E. N. & I NALL , M. E. 2002. Deep-water renewal in a Scottish fjord: temperature, salinity and oxygen isotopes. Polar Research, 21, 251– 258. A USTIN , W. E. N., C AGE , A. G. & S COURSE , J. D. 2006. Mid-latitude shelf seas: a NW European perspective on the seasonal dynamics of temperature, salinity and oxygen isotopes. The Holocene, 16, 936– 947. B AKER , R. G. V. & H AWORTH , R. J. 1997. Further evidence from relic shellcrust sequences for a late Holocene higher sea level for eastern Australia. Marine Geology, 141, 1 –9. B AKER , R. G. V. & H AWORTH , R. J. 2000a. Smooth or oscillating late Holocene sea-level curve? Evidence from cross-regional statistical regression of fixed biological indicators. Marine Geology, 163, 353– 365.
B AKER , R. G. V. & H AWORTH , R. J. 2000b. Smooth or oscillating late Holocene sea-level curve? Evidence from the palaeo-zoology of fixed biological indicators in east Australia and beyond. Marine Geology, 163, 367– 386. B AKER , R. G. V., H AWORTH , R. J. & F LOOD , P. G. 2001a. Warmer or cooler marine late Holocene palaeoenvironments?: Interpreting southeast Australian and Brazilian sea level curves using changes in fixed biological indicators and their d18O content. Palaeogeography, palaeoclimatology, Palaeoecology, 168, 249– 272. B AKER , R. G. V., H AWORTH , R. J. & F LOOD , P. G. 2001b. Intertidal fixed sea level indicators of former Holocene sea-levels in Australia: a summary of sites and a review of methods and models. Quaternary International, 83–85, 257– 273. B AKER , R. G. V., D AVIS , A. M., A ITCHISON , J. C., F LOOD , P. G., M ORTON , B. S. & H AWORTH , R. J. 2003. Comment on ‘Mid-Holocene higher sea level indicators from the south China coast.’ by W. W.-S. Yim and G. Huang [Mar. Geol. 182 (2002) 225–230]: a regional perspective. Marine Geology, 196, 91–98. B EMIS , B. E., S PERO , H. J., B IJMA , J. & L EA , D. W. 1998. Re-evaluation of the oxygen isotopic
184
K. F. CRAVEN ET AL.
composition of planktonic foraminifera: experimental results and revised paleotemperature equations. Paleoceanography, 13, 150–160. B IRD , M. I., T URNEY , C. S. M., F IFIELD , L. K., S MITH , M. A., M ILLER , G. H., R OBERTS , R. G. & M AGEE , J. W. 2003. Radiocarbon dating of organic- and carbonate-carbon in Genyornis and Dromaius eggshell using stepped combustion and stepped acidification, Quaternary Science Reviews, 22, 1805–1812. B LANCHON , P., J ONES , C. & F ORD , D. C. 2002. Discovery of a submerged relic reef and shoreline off Grand Cayman: further support for an early Holocene jump in sea level. Sedimentary Geology, 147, 253– 270. B OURGET , E. & C RISP , D. J. 1975. Factors affecting deposition of the shell in Balanus balanoides. Journal of the Marine Biological Association of the United Kingdom, 55, 231– 249. C LARK , P. U., M ARSHALL , S., C LARKE , G., L ICCIARDI , J. & T ELLER , J. 2001. Freshwater forcing of abrupt climate change during the last glaciation. Science, 293, 283–287. D AVENPORT , J. & I RWIN , S. 2003. Hypoxic life of intertidal acorn barnacles. Marine Biology, 143, 555–563. E REZ , J. 1978. Vital effect on stable-isotope composition seen in foraminifera and coral skeletons. Nature, 273, 119– 202. F AIRBRIDGE , R. W. 1961. Eustatic change in sea level. In: A HRENS , L. S., P RESS , F. & R AUKAWA , K. (eds) Physics and Chemistry of the Earth 4. Pergamon, New York, 99–185. F LEMING , K., J OHNSTON , P., D AN Z WARTZ , D., Y OKOYAMA , Y., L AMBECK , K. & C HAPPELL , J. 1998. Refining the eustatic sea-level curve since the Last Glacial Maximum using far- and intermediatefield sites. Earth and Planetary Science Letters, 163, 327– 342. F LOOD , P. G. & F RANKEL , E. 1989. Late Holocene higher sea level indicators from eastern Australia. Marine Geology, 90, 193–195. G OODWIN , I. E. 1998. Did changes in Antarctic ice volume influence late Holocene sea-level lowering? Quaternary Science Reviews, 17, 319– 332. G RAINGER , F. & N EWELL , G. E. 1965. Aerial respiration in Balanus balanoides. Journal of the Marine Biological Association of the United Kingdom, 45, 469– 479. H UT , G. 1987. Consultants group meeting on stable isotope reference samples for geochemical and hydrological investigations, 16–18 September 1985, Vienna. Report to Director General, International Atomic Energy Agency. IPCC 2001. Third assessment report – Climate Change 2001. Inter-governmental Panel on Climate Change, Geneva, Switzerland. K EATINGS , K. W., H EATON , T. H. E. & H OLMES , J. A. 2002. Carbon and oxygen isotope fractionation in non-marine ostracods: results from a “natural culture” environment. Geochimica et Cosmochimica Acta, 66, 1701–1711. K ILLINGLEY , J. S. & N EWMAN , W. A. 1983.18O fractionation in barnacle calcite: a barnacle paleotemperature equation. Journal of Marine Research, 40, 893 –902.
L ABOREL , J. 1986. Vermetid gastropods as sea level indicators. In: VON DE P LASSCHE , O. (ed.) Sea Level Research. Geo Books, Norwich, 167 –189. L ARCOMBE , P. & C ARTER , R. M. 1998. Holocene bay sedimentation and sea-levels, Magnetic Island. In: J OHNSON , D. P. & H ENDERSON , R. A. (eds) Short Geological Field Trips in the Townsville-Charters Towers Region. Geological Society of Australia, Canberra, ACT, 72– 79. L AMBECK , K. 2002. Sea-Level change from midHolocene to recent time: an Australian example with global implications. In: M ITROVICA , J. X. & VERMEERSEN , B. (eds) Glacial Isostatic Adjustment and the Earth System. American Geophysical Union, Washington, DC, 3350. M AEDA , Y., S IRINGAN , F., O MURA , A., B ERDIN , R., H OSONO , Y., A TSUMI , S. & N AKAMURA , T. 2004. Higher-than-present Holocene mean sea levels in Ilocos, Palawan and Samar, Philippines. Quaternary International, 115–116, 15– 26. M ARIN , F. & L UQUET , G. 2005. Molluscan biomineralization; the proteinaceous constituents of Pinna nobilis L. Material Science and Engineering, C25, 105–111. M C C ONNAUGHEY , T. 1989a. 13C and 18O isotopic disequilibrium in biological carbonates: I. Patterns. Geochimica et Cosmochimica Acta, 53, 151– 162. M C C ONNAUGHEY , T. 1989b. 13C and 18O isotopic disequilibrium in biological carbonates: II. In vitro simulation of kinetic isotope effects. Geochimica et Cosmochimica Acta, 53, 163–171. M C C ONNAUGHEY , T. 2003. Sub-equilibrium oxygen-18 and carbon-13 levels in biological carbonates: carbonate and kinetic models. Coral Reefs. 22, 316–327. M C C ONNAUGHEY , T. A., B URDETT , J., W HELAN , J. F. & P AULL , C. K. 1997. Carbon isotopes in biological carbonates: Respiration and photosynthesis. Geochimica et Cosmochimica Acta, 61, 611– 622. M C C REA , J. M. 1950. On the isotopic chemistry of carbonates and a palaeotemperature scale. The Journal of Chemical Physics, 18, 849 –857. M ICHNER , R. H. & S CHELL , D. M. 1994. Stable isotope ratios as tracers in marine aquatic food webs. In: L AJTHA , K. & M ICHNER , R. H. (eds) Stable Isotopes in Ecology and Environmental Science. Blackwell, Oxford. N UNN , P. D., O LLIER , C., H OPE , G., R ODDA , P., O MURA , A. & P ELTIER , W. R. 2002. Late Quaternary sea-level and tectonic changes in northeast Fiji. Marine Geology, 187, 299 –311. P ILKEY , O. H. & H ARRISS , R. C. 1966. The effects of intertidal environment on the composition of calcareous skeletal material. Limnology and Oceanography, 11, 381–385. R OHLING , E. J., S IDDALL , M., S MEED , D. A. & H EMLEBEN , C. 2003. Holocene climate variability – a sea-level perspective. IMAGES/HOLOCENE Working Group Workshop, Eikum Hotel, Hafslo, Sognefjord, Norway, 27–29 August 2003. S HANAHAN , T. M., P IGATI , J. S., D ETTMAN , D. L. & Q UADE , J. 2005. Isotopic variability in the aragonite shells of freshwater gastropods living in springs with nearly constant temperature and isotopic composition. Geochimica et Cosmochimica Acta, 69, 3949–3966.
ISOTOPIC VARIABILITY IN THE INTERTIDAL ACORN BARNACLE S HEPARD , F. P. 1964. Sea level changes in the past 6000 years: possible archeological significance. Science, 143, 574– 576. S IDDALL , M., R OHLING , E. J., A LMOGI -L ABIN , A., H EMLEBEN , C., M EISCHNER , D., S CHMELZER , I. & S MEED , D. A. 2003. Sea-level fluctuations during the last glacial cycle. Nature, 423, 853 –858. S TAPOR , F. W. J R ., & S TONE , G. W. 2004. A new depositional model for the buried 4000 yr BP New Orleans barrier: implications for sealevel fluctuations and onshore transport for a nearshore shelf source. Marine Geology, 204, 215– 234. S TONE , J. O., B ALCO , G., S UGDEN , D. E., C AFFEE , M. W., S ASS , L. C. III. C OWDERY , S. G. & S IDDOWAY , C. 2003. Late Holocene Deglaciation of Marie Byrd Land, West Antarctica. Science, 299, 99–102.
185
S TUBBINGS , H. G. 1975. Balanus balanoides. Liverpool University Press, Liverpool, 174. T ANAKA , N., M ONAGHAN , M. C. & R YE , D. M. 1986. Contribution of metabolic carbon to mollusc and barnacle shell carbonate. Nature, 320, 520–523. T JIA , H. D., F UJII , S. & K IGOSHI , K. 1983. Holocene shorelines of Tioman Island in the South China Sea. Geologie en Mijnbouw, 62, 599 –604. W EINER , S. & D OVE , P. M. 2003. An overview of biomineralisation processes and the problem of the vital effect. In: D OVE , P. M., D EYOREO , J. J. & W EINER , S. (eds) Biomineralization. Reviews in Mineralogy and Geochemistry, 54, 1 –29. Y IM , W. W.-S. & H UANG , G. 2002. Middle Holocene higher sea-level indicators from the south China Coast. Marine Geology, 182, 235– 230.
Index Note: Page numbers in italics refer to Figures. Page numbers in bold refer to Tables. Acarinina bullbrooki 63, 67, 68 acorn barnacle see Semibalanus balanoides Acropora spp., chemistry 39, 49– 51 Alabaminella weddellensis 108 alkalinity, relation to stable isotopes in foraminifera 137–138, 139 alkenones 15– 16, 77 37 UK (alkenone unsaturation index) 15 Allogromia laticollaris 109 Ammonia spp. A. batavus 158, 166 A. beccarii 109, 161, 163, 165, 166, 167 Antarctica, first ice 10 aragonite biological controls on 36 in corals 37, 87, 88, 90 Arborammina hilaryae 110 Archohelia 95 ARENA 114 Aschemonella ramuliformis 109 Atlantic Ocean (North), foraminifera diversity 102, 103, 104 atomic emission spectrometry (AES) 22 Bathysiphon capillare 108 benthic foraminifera distribution in deep ocean 110–111 biogeochemical cycles 112 phylogeny 112 –113 diversity 97 ecology 97–98 effect of water depth on large species 98–99 small species 99– 101 factors affecting d13C 122 carbonate ion 126–129 isotope disequilibrium 123–124 methane 124–126 microhabitat 122–123 nutrients 121 ontogeny 126 organic matter flux 124 d18O effect of salinity 155–156 effect of temperature 156– 157 vital effects 157 –158 d18O seasonality in Scottish fjord 155 culturing experiments 135– 136 interspecies variation 143 ontogenetic trends 143– 144 yearly d18O offset 143 trophic biology d13C labelled food experiments 101, 104–106, 107 diets 106– 108 bacterivores 109– 110 herbivores 108 omnivores 108– 109 stercomata-bearing 109 suspension feeders 110
Benthic Foraminiferal Accumulation Rate (BFAR) 111 biogenic minerals 3 biogeochemical cycling 112 biological proxies coccolithophores 6, 7 corals 8 –9 diatoms 6, 8 foraminifera 5– 6 molluscs 9 biological pump 5 biomineral, defined 3 –4 biomineralization biological controls 35– 36 bones 42, 44 cell level 36–37 coccolithophores 37–42 corals 37 foraminifera 42– 44 genetic level 44 defined 3 experimental observations 34– 35 factors affecting 33–34 modelling the vital effects of boron in foraminifera biological controls 49–51 chemistry 48 fractionation 46– 48 isotopes 48– 49 process 4 protistan promoted 4 –5 black corals 9 Bolivina spp. B. pacifica 104, 108 B. spathulata 108 B. spissa 108 bones and biomineralization 42, 44 boron in foraminifera fractionation 46–48 biological controls 49–51 inorganic chemistry 48 isotopes 48– 49 Bulimina spp. B. aculeata 104 B. exilis 109 B. marginata 162, 163, 165 B. mexicana 123, 124 d13C of cultured foraminifera 135–136 factors affecting d13C in foraminifera 122 carbonate ions 126– 129 methane 124– 126 microhabitat 122– 123 nutrients 121 ontogeny 126 organic matter flux 124 relation to micro-environment 45, 51–52 size 73 in S. balanoides 175–181
188 d13C (Continued ) use in food experiments for foraminifera 101, 104–106, 107 14 C in corals 17, 18 d44Ca in CaCO3 35 in corals 17, 39, 40, 41, 42 in foraminiferal 42– 44 in molluscs 19 calcification 48, 74 calcite dissolution in foraminifera 60 factors affecting precipitation 33–34 gametogenic 75, 76, 78 overgrowths in foraminifera 60–61 recrystallization in foraminifera 61 calcium carbonate biological precipitation 4–5 coccolithophores and marine production 6 export to deep ocean 5 physical precipitation 4 Cambrian explosion 5 Cape Verde Abyssal Plain, foraminifera diversity 101 carbon cycle (oceanic) 5 carbon dioxide relation to isotopes in foraminifera 45–46, 53 sequestration 5 solubility in sea water 79 carbonate compensation depth (CCD), relation to foraminifera diversity 101 carbonate ion concentration effect on d13C in foraminifera 126– 129 effect on stable isotopes in shells 138, 140 carbonate pumps 5 Cassidulina spp. C. carinata 108 C. neoteretis 125 C. reniforme 125 Cd/Ca, foraminifera 22 cells, effect on biomineralization 36– 37 Chilostomella spp. C. oolina 109 C. ovoidea 104, 109, 110 Cibicides spp. C. lobatulus 159 Loch Sunart 161–162, 163, 165, 166, 167 C. wuellerstorfi 157 Cibicidoides spp. C. pachyderma 123, 124 C. wuellerstorfi 41 clay, role in foraminifera test preservation 59, 69 CLIMAP 3, 4 coccolithophores as biological proxy 6, 7 biomineralization 37– 38, 42 as geochemical proxies 14–16 corals Ca and Mg ion concentrations 39–41 biomineralization 37 as geochemical proxies 17–18 skeletogenesis 87– 88 skeleton composition studies 88– 95 Cyclammina cancellata 104
INDEX dark/light (day/night) cycles, relation to foraminifera chemistry 51– 52 deep water renewal events (DWRE) 158 deer bones, biomineralization 42 diagenesis foraminifera tests 59, 60– 62 problems of 19–21 diatoms as geochemical proxies 16 DIC see dissolved inorganic carbon dinoflagellates, impact on microenvironment 77 dissolution, problems of 21 dissolved inorganic carbon (DIC) and isotopes in foraminifera 48, 137– 138, 139 DNA, studies in foraminifera 25– 26 DWRE see deep water renewal events Early Eocene Climatic Optimum (EECO) 10 see also Palaeocene/Eocene Thermal Maximum (PETM) El Nin˜o (ENSO) 17 Eocene (early) climatic optimum (EECO) 10 see also Palaeocene/Eocene Thermal Maximum (PETM) Epistominella exigua 108 Eponides pusillus 108 equilibrium constant, for CaCO3 precipitation 33 ESONET 114 Favia 88, 93 fixed biological indicator (FBI) 174 see also Semibalanus balanoides Fontbotia wuellerstorfi 123, 124, 125 foraminifera Ca and Mg ion concentration 39–41 biomineralization 42–44 as used in CLIMAP 3, 4 as geochemical proxies Mg/Ca ratio 12– 14 oxygen isotope record 9– 12 impact on chemical micro-environment 51– 53 Mg/Ca ratio 54–55 pH record 53– 54 species Acarinina bullbrooki 63, 67, 68 Acropora spp. 49– 51 Alabaminella weddellensis 108 Allogromia laticollaris 109 Ammonia spp. A. batavus 158, 166 A. beccarii 109, 161, 163, 165, 166, 167 Arborammina hilaryae 110 Aschemonella ramuliformis 109 Bathysiphon capillare 108 Bolivina spp. B. pacifica 104, 108 B. spathulata 108 B. spissa 108 Bulimina spp. B. aculeata 104 stable isotope culture experiment 135–151 B. exilis 109
INDEX B. marginata 162, 163, 165 B. mexicana 123, 124 Cassidulina spp. C. carinata 108 C. neoteretis 125 C. reniforme 125 Chilostomella spp. C. oolina 109 C. ovoidea 104, 109, 110 Cibicides spp. C. lobatulus 159 Loch Sunart 161–162, 163, 165, 166, 167 C. wuellerstorfi 157 Cibicidoides spp. C. pachyderma 123, 124 C. wuellerstorfi 41 Cyclammina cancellata 104 Epistominella exigua 108 Eponides pusillus 108 Fontbotia wuellerstorfi 123, 124, 125 Fursenkoina mexicana 123, 124 Gaudryina siphonifera 41 Globigerina bulloides 47 chemistry 39 controls on growth 78, 79, 80, 81 Globigerinatheka index 63, 66 Globigerinoides spp. 60 G. conglobatus 39 G. ruber chemistry 40 controls on growth 75, 76, 79 G. sacculifer chemistry 40, 41, 49– 51 controls on growth 75, 76, 79 Globobulimina spp. G. affinis 104, 123, 124 G. pacifica 108 Globocassidulina subglobosa 108 Globopelorhiza sublittoralis 110 Globoquadrina conglomerata 20 Globorotalia spp. 60 G. hirsuta 39 G. inflata chemistry 39 controls on growth 79 G. menardii chemistry 39, 40 controls on growth 78, 79, 80 G. truncatulinoides 80 test composition 40, 76 test size 79 test structure 75, 77 G. tumida, controls on growth 79 Globoturbototalita spp. 63, 65, 68 Hantkenina australis 62 Hastigerina pelagica 40 Lobatula lobatula 123, 124 Neogloboquadrina spp. N. dutertrei 39 N. pachyderma 127 controls on growth 79, 80 DNA studies 25–26
189
Orbulina universa 20, 47, 75, 77, 79, 168 boron chemistry 49–51 Mg/Ca ratio 41 74, 78 Planulina wuellerstorfi 157 Porites cylindrica, boron chemistry 49– 51 Pulleniatina obliquiloculata 41 Pyrgo 157 Quinqueloculina seminulum 158, 159 Rosalina vilardeboana stable isotope culture experiment 135– 144 Rhizammina algaeformis 109 Saccorhiza ramosa 110 Sphaeroidinella dehiscens 41 Subbotina linaperta 63, 64, 68 Syringammina spp. S. fragilis 110 S. corbicula 109 Texularia kattegatensis 104, 108, 110 Thurammina albicans 108 Uvigerina ssp. 41 U. akitaensis 104, 108, 126 U. mediterranea 126 U. peregrina 108, 157, 168 d13C 124, 126 see also benthic foraminifera also planktonic foraminifera Forth, Firth of see Semibalanus balanoides experiment Fursenkoina mexicana 123, 124 gametogenic calcite, in foraminifera 75, 76, 78 gas chromatography-mass spectroscopy (GCMS) 25 gas hydrates, d13C in foraminifera 124 –126 Gaudryina siphonifera 41 Ge/Si ratio, diatoms 16 genetics biomineralization 44 use in foraminifera research 114 geochemical proxies coccolithophores 15– 16 corals 17– 18 diatoms 16– 17 foraminifera 9– 15 molluscs 19 problems of measurement calibration 21 contaminants 19 element abundance 22 isotope analysis 22–25 organic matter 25 preservation 19– 21 germanium (Ge) in diatoms 16– 17 glacial-interglacial cycles, effect on foraminifera size 79–81 Globigerina bulloides 47 chemistry 39 controls on growth 78, 79, 80, 81 Globigerinatheka index 63, 66 Globigerinoides spp. 60 G. conglobatus 39 G. ruber chemistry 40 controls on growth 75, 76, 79 G. sacculifer chemistry 40, 41, 49–51 controls on growth 75, 76, 79
190 Globobulimina spp. G. affinis 104, 123, 124 G. pacifica 108 Globobulimina affinis 104, 123, 124 Globocassidulina subglobosa 108 Globopelorhiza sublittoralis 110 Globoquadrina conglomerata 20 Globorotalia spp. 60 G. hirsuta 39 G. inflata chemistry 39 controls on growth 79 G. menardii chemistry 39, 40 controls on growth 78, 79, 80 G. truncatulinoides 80 test composition 40, 41, 76 test size 79 test structure 75, 77 G. tumida, controls on growth 79 Globoturbototalita spp. 63, 65, 68 Goniastrea, skeletal composition 88– 90 greenhouse climate, early Eocene 10 Ha˚kon Mosby Mud Volcano 125 Hampden Beach, New Zealand 59 foraminifera test texture study 63, 64, 65, 66, 67, 69 Hantkenina australis 62 Hastigerina pelagica 40 heterococcoliths 6 Hexacoralla see corals holococcoliths 6 hydrocorals 9 icehouse climate, first Antarctic ice 10 Indian Ocean, foraminifera diversity 103 inductively coupled plasma-mass spectrometry (ICP-MS) 22 intertidal organisms as sea-level indicators see Semibalanus balanoides isotope analysis methods 22– 25 precision 24 isotope dilution thermal ionization mass spectrometry (ID-TIMS) 22 Kerguelen Plateau 59, 69 foraminifera test texture study 63, 64, 65, 66, 67, 68 Kolmogorov scale 51 Komokiacea 99, 100 Labrador Sea ODP site 59 foraminifera test texture study 62, 68, 69 laser ablation (LA) ICP-MS 22 Last Glacial Maximum (LGM), sea surface temperature 3, 4 Late Palaeocene Thermal Maximum (LPTM) 11–12 light/dark cycles, relation to foraminifera chemistry 51–52 lithium, isotope ratio 24 Lobatula lobatula 123, 124 lunar cycle, effect on foraminifera growth 78
INDEX Madeira Abyssal Plain, foraminifera diversity 99 magnesium (Mg) biological controls on 36, 37, 38 in corals and foraminifera 37, 39, 40, 41, 42– 44 in fixed biological indicator species 174 in foraminifera 46 in seawater 35, 36 effect on CaCO3 precipitation 33– 34 Mg/Ca ratio see Mg/Ca ratio manganese (Mn) in fixed biological indicator species 174 oxide coatings 19 Mn/Ca ratio in foraminifera 20 Maud Rise 59, 69 foraminifera test texture study 62–63, 64, 66, 68 May, Isle of 174– 175 see also Semibalanus balanoides experiment MC-ICP-MS 25 Merulina 88, 90 metal binding capacity, foraminifera 6 methane, effect on d13C in foraminifera 121, 124–126 methane release 12 Mg/Ca ratio in corals 85 in foraminifera 20, 46, 54–55, 74, 76, 78 intralab precision of measurement 22 thermometry corals 17 foraminifera 12–15, 20 molluscs 19 micropalaeontology 5 Milankovitch cycles 10 Mn/Ca ratio in foraminifera 20 modelling boron in foraminifera biological controls 49–51 chemistry 48 fractionation 46–48 isotopes 48– 49 molluscs as geochemical proxies 19 d15N coral 93 diatoms 16–17 nanoSIMS 25 neodymium (Nd) isotopes, in coral 18 Neogloboquadrina spp. N. dutertrei 39 N. pachyderma 127 controls on growth 79, 80 DNA studies 25– 26 neomorphism see recrystallization New Zealand, Hampden Beach 59 foraminifera test texture study 63, 64, 65, 66, 67, 68 night/day cycles, relation to foraminifer a chemistry 51–52 Nonionella turgida 162, 163, 165 North Atlantic Deep Water (NADW) 18 nucleation, CaCO3 precipitation 34 d18O of cultured foraminifera 135– 136 O and C isotope correlation 142 –143 ontogenetic trends 143–144
INDEX in corals 88–169 factors affecting in corals 17 in diatoms 16– 17 in foraminifera 9– 12, 45 in molluscs 19 first measured 9 in S. balanoides 174– 181 Ocean Drilling Program (ODP) sites see Kerguelen Plateau; Labrador Sea; Maud Rise ontogeny, effect on d13C in foraminifera 126 optical emission spectrometry (OES) 22 Orbulina universa 20, 47, 75, 77, 79, 168 boron chemistry 49–51 Mg/Ca ratio 41, 74, 78 organic matter effect on d13C in foraminifera 124 export to deep ocean 5 Pachythecalia 92, 93, 94, 95 Pacific Ocean, foraminifera diversity 103, 106 Palaeocene, (Late) thermal maximum (LPTM) 10– 11 Palaeocene/Eocene Thermal Maximum (PETM) 121 see also Eocene, (early) climatic optimum (EECO) palynology 5 pH effect on stable isotopes in shells 138, 140, 180 record in foraminifera 46, 53– 54 phylogeny, foraminifera 112–113 planktonic foraminifera calcification and growth 74 pH record compared with benthic 53– 54 test arrangement 59–60 test diagenesis 59 dissolution 60 overgrowth 60– 61 recrystallization 61 test preservation 61 test size and composition 73–74 test textural study 62–70 test trace elements 74–77 effect of environment 78– 79 effect of glacial-interglacial cycle 79– 81 effect on growth rate 77– 78 effect of test size 78 photosynthesis 48 Planulina wuellerstorfi 157 Pocillopora sp., chemistry 39 Porcupine Abyssal Plain, foraminifera diversity 99 Porcupine Seabight, foraminifera diversity 101 Porites cylindrica, boron chemistry 49– 51 pressure, effect on CaCO3 precipitation 33 primary organic membrane (POM) 60, 74 protistans, role in biomineralization 4– 5 proxies, defined 1 proxy methods 3 proxy records biological coccolithophores 6, 7 corals 8–9 diatoms 6, 8 foraminifera 5– 6 molluscs 9
191
geochemical methods of measurement 19–25 useful species coccolithophores 15–16 corals 17–18 diatoms 16– 17 foraminifera 9– 15 molluscs 19 proxy variables 1, 3 Pulleniatina obliquiloculata 41 Pyrgo 157 Quinqueloculina seminulum 158, 159 reductive cleaning methods 19 respiration 48 Retiophyllia 92 Rhizammina algaeformis 109 Rosalina vilardeboana stable isotope culture experiment 135– 144 Saccorhiza ramosa 110 Sagami Bay (Japan), foraminifera diversity 104, 106 salinity, relation to stable isotopes in foraminifera 137, 138, 155– 156 satellite telemetry, use in foraminifera studies 114 Scleractinia see corals sclerochronology 5 Scotland see Semibalanus balanoides experiment; Sunart, Loch sea surface temperature see temperature sea-level change Holocene 173 use of fixed biological indicator 174 use of S. balanoides as indicator 175 –181 seasonality foraminifera in Scottish fjord 155–157 fjord hydrography 158–159 seawater chemistry 34– 35, 48, 53 secondary ion mass spectrometry (SIMS) 22, 25 Semibalanus balanoides ecology 174 experiments on isotope composition 175– 181 d30Si in diatoms 16– 17 silica in diatoms 8 physical precipitation 4 silicate weathering 5 skeletal evolution, Cambrian explosion 5 soft corals 9 Sphaeroidinella dehiscens 41 stable isotopes see boron; d13C; d44Ca; d15N; d18O; d30Si stony corals 8 stromatolites 4 strontium (Sr) in fixed biological indicator species 174 in foraminifera 74, 76, 78, 79– 80 Sr/Ca ratio coccolithophores 15–16 corals 17–18, 95 foraminifera 20 interlab precision of measurement 22 molluscs 19
192 Subbotina linaperta 63, 64, 68 Sunart, Loch hydrography 158– 159 d18O in foraminifera experiment 159–169 Syringammina spp. S. corbicula 109 S. fragilis 110 temperature effect on CaCO3 precipitation 33, 36 effect on foraminifera growth 79, 80 Last Glacial Maximum 3, 4 relation to d18O in foraminifera 156 –157 shelf seas of NW Europe 158 Texularia kattegatensis 104, 110 Thurammina albicans 108 trace elements in fixed biological indicator species 174 in foraminifera 73, 74–77 effect of environment 78–79
INDEX effect of glacial-interglacial cycle 79– 81 effect on growth rate 77–78 effect of test size 78 role in foraminifera test neomorphism 61 Triassic, coral composition 88, 90– 91 U/Ca ratio, in corals 17, 18 U37 K (alkenone unsaturation index) 15 Uvigerina ssp. 41 U. akitaensis 104, 108, 126 U. mediterranea 126 U. peregrina 108, 157, 168 d13C 124, 126 Weddell Sea, foraminifera diversity 103, 105 X-ray absorption spectroscopy (XAS) 25 zinc (Zn) in diatoms 16