The Use of Ostracoda in the Palaeoenvironmental Reconstruction of the Gulf - Facies Analysis and Morphological Variation

THE USE OF OSTRACODA

IN THE PALAEOENVIRONMENTAL RECONSTRUCTION

OF THE GULF OF CARPENTARIA, AUSTRALIA,

FROM THE LAST INTERGLACIAL TO PRESENT

A thesis submitted in fulfilment of the requirements for the award of the

degree

DOCTOR OF PHILOSOPHY

from the

UNIVERSITY OF WOLLONGONG

JESSICA MARIE REEVES, BSc (Hons).

SCHOOL OF EARTH AND ENVIRONMENTAL SCIENCES

2004

CERTIFICATION

I, Jessica M. Reeves, declare that this thesis, submitted in fulfilment of the requirements for the award of Doctor of Philosophy, in the School of Earth and Environmental Sciences, University of Wollongong, is wholly my own work unless otherwise referenced or acknowledged. The document has not been submitted for qualifications at any other academic institution.

Jessica M. Reeves

03 May 2004

Table of Contents List of figures…………………………………………………………………………………….v List of tables…………………………………………………………………………………….vii Abstract………………………………………………………………………………………….vii Acknowledgements………………………………………………………………………………ix

INTRODUCTION………………………………………………………………………………1 Thesis outline……………………………………………………………………………………. 3

1. THE GULF OF CARPENTARIA………………………………………………………… 7

1.1 Location of the study area………………………………………………………………….7 1.2 Geological history of the Gulf of Carpentaria……………………………………………...8 1.3 Physiography of the modern basin………………………………………………………...13 1.4 Modern sedimentation in the gulf………………………………………………………….20 1.5 Meteorological context…………………………………………………………………….23 1.6 The modern environment………………………………………………………………….26 1.7 Summary…………………………………………………………………………………..28

2. THE STAT OF KNOWLEDGE…………………………………………………………….29

2.1 Pleistocene sea-level history………………………………………………………………29 2.2 The importance of the Gulf of Carpentaria………………………………………………..31 2.3 Previous studies of the gulf region………………………………………………………...32 2.3.1 Environments of the Gulf of Carpentaria – past and present………………………..32 2.3.2 Sea-level reconstructions of the gulf region…………………………………………37 2.3.3 Palaeoclimatic framework…………………………………………………………...38 2.4 Present coring and research program………………………………………………………42 2.5 Direction and purpose of this study………………………………………………………..43

3. MATERIALS AND METHODS……………………………………………………………..47

3.1 Background to the Gulf of Carpentaria project……………………………………………..47 3.2 Coring methodology and preliminary investigations……………………………………….48 3.3 Sampling method employed………………………………………………………………...49 3.4 Sediment particle-size analysis technique…………………………………………………..50 3.5 Micropalaeontological sample preparation………………………………………………….51 3.6 Microscopic analysis of sediment samples………………………………………………….52 3.7 XRD analysis of sediment samples…………………………………………………………52 3.8 Ostracod analysis methodology……………………………………………………………..52 3.9 SEM analysis methods………………………………………………………………………55 3.10 Stable-isotope analysis of ostracod valves…………………………………………………55 3.11 Dating methods…………………………………………………………………………….56 3.11.1 Radiocarbon dating…………………………………………………………………..56 3.11.2 Amino acid racemisation dating……………………………………………………..57 3.11.3 Luminescence dating………………………………………………………………...58

4. SEDIMENT ANALYSES……………………………………………………………………...59

4.1 Sedimentary parameters investigated from the core material……………………………….59 4.1.1 Particle-size……………………………………………………………………………..59 4.1.2 Colour…………………………………………………………………………………..59 4.1.3 Water content……………………………………………………………………………59 4.1.4 Mineralogy………………………………………………………………………………60 4.1.5 Microfauna………………………………………………………………………………62 4.2 Depositional facies defined from the gulf cores……………………………………………...63 4.4 Sampled sites…………………………………………………………………………………64 4.5 Sediment description of core MD-32…………………………………………………………69 4.6 Results of preliminary observation of the physical parameters………………………………71 4.6.1 Particle-size analysis……………………………………………………………………..71 4.6.2 Water content…………………………………………………………………………….73 4.7 Microscopic observation of the sediment of core MD-32……………………………………74 4.8 Discussion and interpretation of the sediment of core MD-32……………………………….80 4.9 Comparison with other cores…………………………………………………………………98 4.9.1 Sediments of core MD-33………………………………………………………………..101 4.9.2 Sediments of core MD-31………………………………………………………………..109 4.9.3 Sediments of core MD-30………………………………………………………………..117 4.9.4 Sediments of core MD-29………………………………………………………………..123 4.9.5 Sediments of core MD-28………………………………………………………………..129 4.10 Palaeoenvironments of the gulf, as observed through the core sediment…………………..135

5. OSTRACOD ANALYSES…………………………………………………………………..146

5.1 Introduction to ostracods………………………………………………………………….146 5.1.1 Overview of ostracods………………………………………………………………...146 5.1.2 Ontogeny………………………………………………………………………………147 5.1.3 Valve structure and taxonomy………………………………………………………...150 5.2 Ostracods as palaeoenvironmental indicators……………………………………………...152 5.2.1 Ecology………………………………………………………………………………...152 5.2.2 Marine environment……………………………………………………………………155 5.2.3 Marginal marine environment………………………………………………………….157 5.2.4 Non-marine environment……………………………………………………………….159 5.2.5 Environmentally cued polymorphism…………………………………………………..164 5.2.6 Depositional environment………………………………………………………………167 5.3 Modern ostracod fauna………………………………………………………………………172 5.3.1 Ostracods from surface samples of the Gulf of Carpentaria……………………………172 5.3.2 Previous regional studies………………………………………………………………..179 5.4 Results of the ostracod analysis of core MD-32……………………………………………...198 5.4.1 Ostracod fauna present in core samples………………………………………………..198 5.4.2 Ostracod assemblages from core MD-32……………………………………………….205 5.5 Cluster analysis………………………………………………………………………………223 5.5.1 R-mode cluster analysis………………………………………………………………….224 5.5.2 Q-mode cluster analysis………………………………………………………………….226 5.6 Discussion and interpretation of the ostracod observations………………………………….235

6. STABLE-ISOTOPE ANALYSES…………………………………………………………….252

6.1 Introduction to stable-isotopes………………………………………………………………252 6.2 Oxygen stable-isotope composition…………………………………………………………253 6.2.1 Relationship between δ180 of water and δ180 of inorganically precipitated carbonates……………………………………………………………………….253 6.2.2 δ180 of meteoric water…………………………………………………………………..255 6.2.3 δ180 of seawater…………………………………………………………………………256 6.2.4 δ180 of marginal marine water…………………………………………………………..257 6.2.5 δ180 of lake water……………………………………………………………………….257 6.3 Stable carbon isotope composition…………………………………………………………..260 6.3.1 Relationship between δ13C of water and δ13C of precipitated carbonates………………260 6.3.2 δ13C of plants…………………………………………………………………………….262 6.3.3 δ13C of seawater…………………………………………………………………………263 6.3.4 δ13C of marginal marine water…………………………………………………………..264 6.3.5 δ13C of non-marine water………………………………………………………………..264 6.4 Ostracods and stable-isotope analysis……………………………………………………….264 6.5 Vital effects………………………………………………………………………………….266 6.6 δ18O and δ13C of ostracod valves as palaeoenvironmental indicators………………………268 6.6.1 Marine environment…………………………………………………………………….268 6.6.2 Marginal marine environment………………………………………………………….269 6.6.3 Non-marine environment……………………………………………………………….270 6.7 Covariance of δ18O and δ13C………………………………………………………………...272 6.8 δ18O of the modern Gulf of Carpentaria……………………………………………………..277 6.9 Stable-isotope analyses from core MD-32…………………………………………………..283 6.10 Results………………………………………………………………………………………284 6.11 Discussion and interpretation……………………………………………………………….301 6.12 Other palaeoenvironmental indicators………………………………………………………317 6.12.1 Pollen analyses…………………………………………………………………………317 6.12.2 Other geochemical analyses……………………………………………………………323

7. SYNTHESIS…………………………………………………………………………………..335

7.1 Sea level……………………………………………………………………………………335 7.2 Climate and Hydrology…………………………………………………………………….353 7.3 Conclusion………………………………………………………………………………… 364

REFERENCES…………………………………………………………………………………..366

APPENDICES……………………………………………………………………………………395

Appendix 1a. Core log of core MD-28…………………………………………………………397 Appendix 1b. Core log of core MD-29………………………………………………………….399 Appendix 1c. Core log of core MD-30………………………………………………………….401 Appendix 1d. Core log of core MD-31………………………………………………………….403 Appendix 1e. Core log of core MD-33………………………………………………………….405 Appendix 2a. Sedimentary parameters of core MD-31…………………………………………407 Appendix 2b. Sedimentary parameters of core MD-32…………………………………………415 Appendix 2c. Sedimentary parameters of core MD-33…………………………………………423 Appendix 3a. Quartz grains of unit 7, MD-32…………………………………………………..427 Appendix 3b. Quartz grains of unit 4, MD-32…………………………………………………..427 Appendix 4a. Results of EDX analyses of material from core MD-32…………………………429 Appendix 4b. Results of XRD analyses of material of cores MD-28, 29, 30, 32………………429 Appendix 5a. 14C AMS dates from the Gulf of Carpentaria cores………………………………431 Appendix 5b. Marine reservoir correction for NE Australia……………………………………433 Appendix 6. Taxonomic list of ostracods described from core MD-32………………………...435 Appendix 7. Ostracod plates…………………………………………………………………….450 Appendix 8. Ostracod genus distribution……………………………………………………….470 Appendix 9. Stabel isotope results from core MD-32………………………………………….473

List of figures

Chapter 1

1.1 Locality map of the Gulf of Carpentaria region. 5 1.2a Location of basins within the Gulf of Carpentaria region. 9 1.2b Outline of the basin stratigraphy and lithology of the Gulf of Carpentaria. 9 1.3 Interpretive line drawing of seismic line 17 across the Gulf of Carpentaria. 11 1.4 Geomorphic units of the Gulf of Carpentaria, Australia. 15 1.5 Geomorphic units of the southern lowlands of New Guinea. 15 1.6 Major surficial sedimentary zones of the Gulf of Carpentaria. 19 1.7 Map showing the location of the Indo-Pacific Warm Pool. 21 1.8 Schematic of the Walker Circulation. 21 1.9 Dominant seasonal circulation patterns across Australia. 22 1.10a Mean monthly temperature around the Gulf of Carpentaria. 25 1.10b Mean monthly precipitation around the Gulf of Carpentaria. 25

Chapter 2

2.1 Relative sea-level curve, showing Marine Isotope Stages. 30 2.2 Location of sites used for comparison with the Gulf of Carpentaria record. 35 2.3 Map showing the core localities. 41

Chapter 3

3 Schematic of the core slicing method employed. 50

Chapter 4

4.1 Core log, physical and geophysical data from core MD-32. 67 4.2 Sedimentary log and interpretation of core MD-32. 81 4.3 Comparative log between cores MD-32 and GC-2. 93 4.4 Sedimentary log and interpretation of core MD-33. 99 4.5 Sedimentary log and interpretation of core MD-31. 107 4.6 Sedimentary log and interpretation of core MD-30. 115 4.7 Sedimentary log and interpretation of core MD-29. 121 4.8 Sedimentary log and interpretation of core MD-28. 127 4.9 Comparative log between all cores. 133

Chapter 5

5.1 Valve morphology of a cytheroidean ostracod. 149 5.2 Schematic of ostracod diversity versus water salinity. 153 5.3 Aquatic environments inhabited by ostracods. 153 5.4 Relationship between ostracod abundance and diversity. 160 5.5 Schematic of non-marine solute evolution. 160 5.6 Examples of morphological variation of ostracods. 163 5.7 Schematic of ostracod population age structure. 168 5.8 Variation in preservation of Cyprideis australiensis 171 5.9 Location of previous ostracod studies in the Gulf of Carpentaria 175 5.10 Map showing the location of previous ostracod studies in the region. 177 5.11 Map showing the ostracodal zoogeographic provinces of the Indo-Pacific region. 178 5.12 Morphological variation of Leptocythere valves from core MD-32. 193 5.13 Ostracod genus distribution through core MD-32. 203 5.14 Morphological variation of Cyprideis sieve pores through core MD-32. 221 5.15 R-mode cluster analysis of ostracods from core MD-32. 230 5.16 Q-mode cluster analysis of ostracods from core MD-32. 232 5.17 Pie-diagrams of the distribution of ecological associations through core MD-32 as determined from ostracod faunal analysis. 233

Chapter 6

6.1 Global meteoric water line. 256 6.2 δ13C values of some relevant carbon reservoirs. 261 6.3 Range of δ13C values of plants. 261 6.4 Long-term monthly mean δ18O values in precipitation. 275 6.5 δ18O of ostracods from core MD-32. 279 6.6 δ13C of ostracods from core MD-32. 281 6.7a Stable oxygen and carbon isotopic values of ostracods of unit 6 of core MD-32. 285 6.7b Plot of the covariance between δ13C and δ18O through unit 6. 285 6.8a Stable oxygen and carbon isotopic values of ostracods of unit 5 of core MD-32. 289 6.8b Plot of the covariance between δ13C and δ18O through unit 5. 289 6.9a Stable oxygen and carbon isotopic values of ostracods of unit 3 of core MD-32. 293 6.9b Plot of the covariance between δ13C and δ18O through unit 3. 293

6.10a Stable oxygen and carbon isotopic values of ostracods of unit 2 of core MD-32. 295 6.10b Plot of the covariance between δ13C and δ18O through unit 2. 295 6.11a Stable oxygen and carbon isotopic values of ostracods of unit 1 of core MD-32. 299

6.11b Plot of the covariance between δ13C and δ18O through unit 1. 299 6.12 Comparative plots of the Mg/Ca, Sr/Ca and ∆Sr of ostracods from cores GC-2 and GC-10A and the stable carbon and oxygen isotopic records of core MD-32. 327

Chapter 7

7.1 Sea-level curve showing an approximation of the height of the Arafura Sill. 337 7.2 Schematic maps of the palaeoshorelines of the Gulf of Carpentaria and past extent of Lake Carpentaria, through the last glacial cycle. 342 7.3 Comparison of palaeoclimatic reconstructions from the Gulf of Carpentaria with other regional records. 351

List of tables

Table 1 Location of the sediment cores from the Gulf of Carpentaria. 42 Table 2 Varying hydrology and extent of Lake Carpentaria. 142 Table 3 Visual preservation index key. 171 Table 4 Recorded salinity ranges of a selection of non-marine ostracods found in Australia. 190 Table 5 Ostracod faunal comparison of core MD-32 with the surrounding region. 199 Table 6 Ecological ranges of ostracod species found in core MD-32. 231

Abstract

Throughout the majority of the last glacial cycle, the Gulf of Carpentaria, the large body of seawater that extends as a broad tongue into the north of Australia, was an enclosed lake. It would have been possible to walk around the perimeter of the lake from Australia to New Guinea. Aboriginal oral tradition recalls that during some periods, the lake teamed with freshwater fish and waterfowl. At other times, it was shallow and segmented into a series of saline swamps and pools, or even dry and subaerially exposed. Marine waters transgressed the lake margins during periods of high sea level, forming a shallow marine embayment through to open marine conditions. A large brackish lake remained as the waters again receded. The present open shallow marine conditions were emplaced within the last eight thousand years.

Fluctuations in the extent and nature of this waterbody through the last glacial cycle are preserved in the sediments of the gulf. The sedimentological interpretation of core material retrieved from the gulf provides a framework of palaeoenvironmental change in the region, in particular the extent of the lake basin, timing of marine influence and evidence of channel activity. Through the detailed analysis of ostracod faunal assemblages of the core sediment and comparison with modern species distribution, inferences are drawn about the ecology of the gulf basin at the time of valve formation. Morphological variation and preservation of the valves also provide information regarding the changing conditions of the waterbody and post-depositional effects. The geochemistry of the ostracod valves permits insight to variations in climate, particularly temperature and effective precipitation change. Comparison and correlation are made with the global sea-level curve and regional climatic records.

The implications for the range of environments evident in the gulf region through this period include the presence of warm shallow water for latent heat transport and the generation of cyclones, strengthened monsoon conditions and the mixing of Indian and Pacific Ocean waters through Torres Strait at high sea-level times, and greater continentality, reduced rainfall and altered oceanic currents during sea-level lowstands. The impact of the changing conditions, and shorelines, would have been particularly significant for the human inhabitants who have been present in this region for at least the last fifty thousand years. The nature and extent of the waterbody through the period has enormous implications for people, both as a resource for food and a potential land-bridge between Australia and New Guinea.

Acknowledgments

I would firstly like to thank my supervisor; Prof. Allan Chivas, for giving me the opportunity to be part of the Gulf of Carpentaria project. This thesis was financially supported by a "University of Wollongong Matching Scholarship" and the grants awarded to Prof. Chivas by the Australian Research Council (A39600498 and DP0208605). I acknowledge these sources for their contributions to both my work and the Gulf of Carpentaria project as a whole.

I owe a great debt of gratitude to my 'GoC' comrades, Dr Adriana Garcia, Sabine Holt and Martine Couapel, who endured the cold room with songs and laughter and have been a constant support to me, both professionally and personally. For frequent and lengthy discussions, I am indebted to Prof. Patrick De Deckker and Assoc. Prof. Brian Jones. I thank you both for your interest and support of my work. Thanks to Dr Sander van der Kaars for discussions and for allowing me the use of his excellent pollen record. I would also like to thank Prof. Dave Horne and the late Dr Ken McKenzie for their assistance in my taxonomy. For technical support, I acknowledge the assistance of Sue Wang, David Wheeler, David Carrie, Nick Mackie, John Reid and John Marthick.

To my partner Richard Goodman, I thank you for your patience, understanding and constant support (not to mention proof-reading, editorial, financial support, etc.). Thanks also to Alison Vaughan for proofreading, constructive criticism and all-round niceness. I would like to acknowledge all of the regulars at Flor for their support and interest over the years, even if they didn't understand what I was doing.

Lastly, I would like to thank Prof. Jim Bowler "for getting me into this mess in the first place". His belief in understanding place to understand people has lead me down this twisted track.

The range of life experience I have had through the period of my candidature are somewhat analogous to a sea-level curve and are as much a testament to being a PhD student as the thesis itself. In this alone I feel that I have learnt and achieved a great deal.

Introduction

This investigation examines in detail the sequence of environmental change observed in the Gulf of Carpentaria, northern Australia, through the last glacial cycle. The work identifies two periods of marine transgression and numerous minor incursions of the basin throughout this period, which provide independent evidence of the extent and timing of sea-level fluctuations. As the gulf is located on a relatively stable portion of the continental plate and away from major ice- sheets, the region is relatively free of tectonic and isostatic influence, that may hamper sea-level determination of other records.

The thesis study presented here highlights the use of Ostracoda in palaeoenvironmental studies. The ubiquitous nature of this microfauna in aquatic environments renders it an ideal proxy to use for the reconstruction of transgressive marine/non-marine settings. Identification of subfossil ostracodal assemblages in comparison with those in the modern environment allows insight to be gained about past conditions. Special attention has been made of the morphological variation of the carapace of some ostracod taxa in response to changes in the environment.

In conjunction with geochemical analysis of the valves, inferences may be drawn about the salinity and solute composition of the host waters. As ostracod valves precipitate their calcitic carapaces in a very short time from constituents taken directly from the ambient water, the chemistry of the valves is a valuable indicator

vii of conditions at the time of valve formation. This in turn is utilised to derive information about the hydrology and climate of the environment at the time of deposition.

The Gulf of Carpentaria is a vast basin with a very shallow gradient over much of its extent. The sedimentological and microfaunal records clearly identify the influence of previous marine connections on the basin. The heights of the sills that separate the gulf from the open oceans are particularly significant for determining the extent of the marine highstands through the last glacial cycle.

The influence of past river activity may also be determined and compared with the better established terrestrial records. Owing to the shallow nature of the basin, short-term fluctuations may in some instances obscure the broader picture of regional climatic change. As such this may be used as an indicator of the stability of the system. The intention of this study is to present an overview of the changing environment of the Gulf of Carpentaria, from the Last Interglacial to present, through utilisation of ostracod assemblage data, morphology and shell geochemistry as extracted from sediment cores from the gulf.

Thesis outline

With an understanding of the sedimentation processes, and by determining the assemblages and stable-isotope shell chemistry of ostracods throughout core

MD-32, this thesis study extends the knowledge of the palaeoenvironments witnessed in the Gulf of Carpentaria beyond those previously investigated, through the last glacial cycle. A comparison of this record with others from coastal and continental Australia and through Southeast Asia broadens the understanding of climatic evolution within the region. The principal scientific objectives of this study of the Gulf of Carpentaria include:

• To formulate a palaeoenvironmental reconstruction of the Gulf of

Carpentaria through the last glacial cycle, including an independent sea-

level change history;

• To elucidate an understanding of the hydrologic conditions of Lake

Carpentaria in relation to climatic fluctuation and thus provide a link

between marine and terrestrial records within the region;

• To broaden the understanding of the environmental tolerances of ostracods

within a large basinal system as it fluctuates between marine and non-

marine conditions.

This thesis is arranged as follows:

• Chapter 1 outlines the geological history, physiography, current

geomorphology and climate of the Gulf of Carpentaria region.

• Chapter 2 provides an introduction to the Gulf of Carpentaria and previous

work on the Late Pleistocene of the region. This review presents the

background framework to the interpretations of the palaeoenvironmental

3 and climatic reconstructions presented herein. An overview of the

knowledge of ostracods in the region and ostracod shell geochemistry is

also presented. The current project is outlined in this context.

• Chapter 3 provides an overview of the materials and methodologies

employed in this study.

• Chapter 4 details the sedimentological framework established for the cores

retrieved for the present study. From this, interpretations regarding the

timing and extent of marine and non-marine environments are

extrapolated.

• Chapter 5 presents the results of the ostracodal faunal analysis of core

MD-32. Inferences regarding environment of deposition are drawn from

comparisons with the modern gulf region and knowledge of the ecology of

the species present.

• Chapter 6 presents the results of the stable-isotope analyses of ostracod

valves from the core material and the implications for the various

environmental and climatic settings encountered. These data are used in

conjunction with previous geochemical data obtained from ostracods from

the gulf and the pollen record of the present cores, to provide further

information on particularly the lacustrine phase of the basin.

• Chapter 7 devises a synthesis of the palaeoenvironments of the gulf in

comparison with previously established regional data. Inferences are

drawn using a combination of the above-mentioned data.

Figure 1.1. Locality map of the Gulf of Carpentaria region, showing the bathymetry of the gulf.

1. The Gulf of Carpentaria

An overview of the modern environment

The physical setting of the Gulf of Carpentaria is presented, including locality, geology, physiography, modern sedimentation processes and meteorology. An understanding of the modern environment is necessary before interpretation of past settings can be realised.

1.1 Location of the study area

The Gulf of Carpentaria is an epicontinental sea located between Australia and

New Guinea (Fig. 1.1). It is bordered to the north by the south coast of West

Papua, Indonesia and Papua New Guinea and to the south by the north coast of

Australia from Arnhem Land, Northern Territory in the west around to Cape York

Peninsula, Queensland in the east. This broad, shallow embayment reaches a maximum depth of 70 m toward the eastern margin. Below 50 m, the bottom is essentially flat with a gradient of about 1:13,000 (Edgar et al., 2003).

The Arafura Sill is a sedimentary feature that separates the gulf from the Indian

Ocean to the west, 53 m below present sea level. This sill, spans 100 km in both

N-S and E-W directions, and has less than 2 m of relief (Torgersen et al., 1983).

The flat topography is most likely due to tidal movement planing the surface.

Beyond the sill, the Arafura Sea reaches depths of 40-80 m, descending to

120-200 m beyond an ill-defined shelf before dropping rapidly into the >3000 m trough that divides the Sahul Shelf of northwestern Australia from Timor

(Galloway and Löffler, 1972).

Torres Strait is a shallow platform extending from Papua New Guinea to the tip of

Cape York Peninsula and separating the Gulf of Carpentaria from the Pacific

Ocean. The Strait is generally between 15 m and 25 m deep and is bounded on the eastern side by a sill only 12 m deep. To the east of the sill, the shelf edge is around -70 to -80 m, before dropping to depths of >4000 m in the Coral Sea. The

Strait comprises a discontinuous chain of granitic, volcanic and coral islands

(Haddon et al., 1894, Woodroffe et al., 2000). Coral reef growth is prolific throughout.

1.2 Geological history of the Gulf of Carpentaria

The Australian continental plate includes the island of New Guinea at its northern margin. The underlying sedimentary units of the Gulf of Carpentaria form a series of stacked intracratonic basins and depressions on a basement of largely

Precambrian and Palaeozoic metamorphosed sedimentary rocks and igneous intrusions (Smart et al., 1980; Blake et al., 1984; Burgess, 1984; Thomas et al.,

1990; Passmore et al., 1993a,b) (Fig. 1.2 a,b). The earliest of the basins, the

Bamanga Basin, is an asymmetric sag basin of possible Palaeozoic age (Passmore et al., 1993a). The basin extends from the northern half of the gulf, beyond the

Australian/New Guinea boundary.

1. The Gulf of Carpentaria

Figure 1.2a. Location of basins within the Gulf of Carpentaria region (after Passmore et al., 1993).

Figure 1.2b. Outline of the basin stratigraphy and lithology for the Gulf of Carpentaria (after Passmore et al., 1993) 9

The Carpentaria Basin (Jurassic and Cretaceous) underlies most of the gulf and

Cape York Peninsula and is separated from the Eromanga, Laura, Papuan and

Morehead Basins by passive basement features. The basin formed as a down-warp in the final stage of craton formation in response to the

Australian-Antarctic breakaway from Gondwana (Smart et al., 1980). The structure of the basin is that of a simple, oval, north-trending depression. The approximately 1200 m of sedimentation consists of sand deposition in the early

Jurassic, overlain by the marine mudstones of the Cretaceous transgression

(Smart et al., 1980).

The Karumba Basin (Neogene and Pleistocene) is superimposed on the

Carpentaria Basin, extending into the Northern Territory and the south of New

Guinea. The formation of the basin corresponds to the northerly drift of the

Australian plate after severance from Antarctica (Doutch, 1972). The 300 m sedimentary sequence of the basin is described in Edgar et al. (2003), interpreted from seismic data, petroleum exploration well data and seafloor samples. This sequence comprises terrestrial sedimentation in a temperate climate in the

Cainozoic, followed by an expansion of carbonate deposition to the south of the basin during the Miocene, replaced by terrigenous clastics from the newly formed

Central Ranges of New Guinea in a wetter, more tropical climate through the Late

Miocene.

1. The Gulf of Carpentaria

Figure 1.3. Interpretive line drawing of seismic line 17 across the Gulf of Carpentaria. Channel widths have been enlarged to enable representation (after Edgar et al., 2003)

Edgar et al. (2003) present a high-resolution seismic survey of the region that reveals at least 14, perhaps 17 basin-wide transgressive-regressive cycles, since about the Miocene (Fig. 1.3). They divide the sediment record into 2 broad units; a lower unit of uniform sediments and an upper stratified unit, marked by numerous sediment-filled channels. The erosional surfaces have been preserved by the low-energy of the inflowing marine waters, due to the vast surface area and low gradient of the gulf basin. The uppermost 10 reflectors indicate abundant channels, incising the floor of the gulf to depths of 40 m. Channelling over the

Arafura Sill is particularly intense and stratification is poorly preserved as a result.

The fundamental structure of the Carpentaria Basin was largely developed in the

Early Jurassic as an intracratonic sag; the major fault zones are inherited from the basement (Thomas et al., 1990). The formation of the Karumba Basin was initiated in the Late Cretaceous. The present form developed in the Late

Cainozoic by uplifting of the basin margins and the modification of the northwest- plunging Mesozoic Staaten River Embayment, forming the down-warped

Gilbert-Mitchell Trough in the east of the basin (Doutch et al., 1972). Edgar et al.

(2003) describe the gulf as a half-graben, faulted on the western margin and hinged on the east. Tectonic activity during the Pliocene and Pleistocene elevated many regions of the Great Dividing Range, which forms a spine through Cape

York Peninsula (Doutch, 1972). Smart et al. (1980), note minor faulting during the Holocene on Mornington and Sweers Islands. The rapid northward movement of the Australian Plate, and convergence with the Pacific Plate, resulting in an arc-continent collision, have dominated the tectonics of New Guinea since the

1. The Gulf of Carpentaria

Eocene (Hill and Hall, 2003). Tectonic uplift in the highlands of New Guinea is still in progress.

The majority of the islands of Indonesia were formed during the late Miocene, with most of New Guinea emerged from the sea by the Pliocene. During this time, identical genera of flora and fauna occurred in both Australia and New

Guinea (Doutch, 1972). The water barrier dividing the two occurred for the first time since the Carboniferous during the Pleistocene. A land connection between the two "countries", at least across Torres Strait has been present for much of the

Late Pleistocene.

1.3 Physiography of the modern basin

The area of northern Australia that surrounds much of the Gulf of Carpentaria comprises former Tertiary erosion surfaces with deep weathering profiles and shallow stony soils. Relief is relatively low; mostly below 500 m. Galloway and

Löffler (1972) have divided the region into twelve major geomorphic regions exhibiting distinctive landforms (Fig. 1.4). The uplands are generally of

Precambrian and Palaeozoic age with strong structural control (Smart et al.,

1980). Open tropical eucalypt forest, with pockets of rainforest and heath are characteristic of these units. The lowlands are low-gradient depositional, lateritic plains, with a characteristic dense, black soil with savannah-type vegetation

(Nix and Kalma, 1972).

1. The Gulf of Carpentaria

Major streams and rivers with anastomosing channels crosscut the plains and migrate annually. These form large alluvial fans and prograding deltas during the high-flow wet season. There are currently 37 streams which inflow to the gulf, although many of these are only active in the wet season. In the rivers with annual flow, sand is transported only during monsoonal flow, whereas mud is redistributed in the tidal portion of the channels during the dry season (Jones et al., 1993). The change from river to tidal channel occurs at the high tide limit.

The coastal plain region, around 5-30 km wide, forms a near continuous band around the gulf over which the coast migrates between wet and dry seasons.

Mangrove swamps, bare saline tidal flats and winding tidal channels characterise these heavy clay plains (Galloway and Löffler, 1972). Beach ridges, parallel to the modern coast, mark former shorelines and are common in the south and east of the gulf. In addition, shelly chenier ridges are particularly common in the southeast corner of the gulf (Phipps, 1980; Rhodes et al., 1980). These formed during periods of progradation followed by reduced fluvial sediment input

(Rhodes, 1982; Chappell and Thom, 1986). Older beach ridges, described by

Rhodes (1982) as Pleistocene, are located behind the extent of the chenier ridges, on the coastal plain. On the west coast of the gulf and Groote Eylandt there are siliceous sand dunes, indicative of increased aeolian activity during their formation (Lees et al., 1990; Shulmeister and Lees, 1992).

New Guinea forms the northern boundary of the gulf. The central region of New

Guinea is steeply mountainous, with sharp-crested ridges and deeply incised valleys. South of the central mountain chain lie the southern lowlands. The lowlands may be divided into the older and more recent alluvial plains, and the fringing coastal plains (Fig. 1.5). Low ridges, undulating plains, broad floodplains, swamps and lakes characterise the lowlands region. The older alluvial plains are typically above the flood-level of modern streams, but may still become inundated during the wet season. The soils of these plains are strongly weathered sandy loams, commonly containing laterite concretions (Galloway and

Löffler, 1972). The recent alluvial plains occur in wide bands along modern rivers, comprising cut-off meanders and sand bars, backed by lower plains and swamps. Soils here are sticky alluvial clays. The recent alluvial plains extend in places to the northern banks of the modern Gulf of Carpentaria. Strongly alkaline clays form the soil of the coastal plain in areas free from fluvial influence.

Bryomal reefs and platforms have recently been identified by sonar mapping in the south of the gulf, offshore of Mornington Island (Geoscience Australia, 2003).

The reefs are at depths of 25-30 m, with the highest reaching -18 m below present sea level (bpsl). These reefs are thought to have formed during the last interglacial (~120-70 ka BP), during an extended period when sea was lower than present and the gulf was an embayment open only to the west via the Arafura Sill, with Torres Strait being a land bridge. The reefs appear to be mostly relict, though live coral specimens have been recovered. The surfaces of the reefs appear to have been subaerially exposed during sea-level lowstands.

1. The Gulf of Carpentaria

Figure 1.6. Major surficial sedimentary zones of the Gulf of Carpentaria (after Jones, 1987).

1.4 Modern sedimentation in the gulf

Modern sedimentation in the gulf has been examined by Jones (1987), Jones and

Torgersen (1988) and Somers and Long (1994) and may be divided into three categories (Fig. 1.6). The nearshore zone on the east and west coasts, to a water depth of 20 m is a region of active sedimentation, comprising deltaic and nearshore sand and offshore prodelta sandy mud (Somers and Long, 1994).

Deposition is primarily due to the activation of the numerous rivers feeding into the gulf during monsoon periods. Expansive sand sheets and bars have developed around the mouths of the rivers, whilst supratidal mudflats extend behind beach ridges. An offshore relict sand of medium- to coarse-grained iron-stained quartz characterises the southern half of the gulf (Jones, 1987). The remainder of the sediment comprises Holocene sandy mud and muddy sand, with variable components of biogenic carbonate detritus (Yassini et al., 1993). Ooids, which are formed in water shallower than 5 m, have been found in water depths of greater than 25 m (Phipps, 1980; Jones, 1987). Seagrasses in the high tidal- current zones of the gulf and Torres Strait trap a large amount of fine-grained sediment, resulting in carbonate mud accumulation (Harris, 1994).

1. The Gulf of Carpentaria

Figure 1.7. Map showing the location of the Indo-Pacific Warm Pool (IPWP) and the major currents in the region (after Martinez et al., 1997).

Figure 1.8. Locality map showing Darwin and Papeete, from which the SOI is measured, and a schematic of the Walker Circulation.

Figure 1.9. Dominant seasonal circulation patterns across Australia (after Harrison, 1993).

1. The Gulf of Carpentaria

1.5 Meteorological context

The Gulf of Carpentaria is located adjacent to the Indo-Pacific Warm Pool

(IPWP) (Fig. 1.7). The IPWP is defined by the region of waters with an average annual sea-surface temperature in excess of 28oC (Tomczak and Godfrey, 1994).

The Indonesian throughflow, whereby the westerly flowing equatorial currents move water from the Pacific Ocean to the Indian Ocean, occurs within the IPWP.

The maximum throughflow occurs during August, the southern hemisphere winter. Part of the IPWP is recycled back into the South Pacific via the East

Australian current, and into the North Pacific via the Kuroshio Current, which are fed by both the north and south equatorial currents (Hirst and Godfrey, 1993).

The South Java, South Equatorial and Leeuwin Currents transport water south and west respectively, from the throughflow to the Indian Ocean (Godfrey and

Ridgway, 1985). The high atmospheric temperatures in the equatorial region induce high evaporation of oceanic waters and subsequent precipitation, resulting in high sea-surface temperatures, yet low sea-surface salinity. The IPWP is responsible for the largest transfer of heat and moisture between the surface and the atmosphere and is implicated in the generation of the El Niño - Southern

Oscillation (ENSO) phenomenon.

The dominant air currents responsible for driving Australia's weather patterns are the Hadley Cell and the Walker Circulation (Fig. 1.8). The Hadley Cell circulation is caused by air rising at the equator and sinking at 35oS to produce the dominant surface easterly winds across Australia, known as trade winds. The

Walker Circulation, which is a longitudinal circulation cell centred over the equator, overprints the Hadley Cell. Warm sea-surface temperatures on the west

of the Pacific in contrast with cold upwelling off the coast of Peru, cause more evaporation and latent heat input to the atmosphere in the western Pacific, producing the Walker Circulation. The difference between these two extremes

(measured as the mean sea-level pressure at Papeete minus the mean sea-level pressure at Darwin) is quantified as the Southern Oscillation Index (Troup, 1965;

Allan et al., 1991). The ENSO phenomenon is the strongest natural interannual climatic fluctuation (Philander, 1990). During an ‘El Niño’ event, summer monsoon rainfall in northern Australia is decreased, as the trough is weak and displaced towards the equator. Tropical cyclone activity is low in the northeast of

Australia, but strong in the north and northwest (Evans and Allan, 1992). The reverse is true of a ‘La Niña’ event, whereby the monsoon trough is stronger and rainfall heavier, with fewer cyclones to the north and west (Suppiah, 1993).

The annual climate in the gulf region is dominated by the migration of the

Intertropical Convergence Zone (ITCZ) (Fig. 1.9). During the southern hemisphere summer, the northwest monsoon is generated as it crosses the sea from the Asian high-pressure belt toward the ITCZ. The two intersect over northern Australia around January where the moisture-laden air rises causing heavy rains. South of the ITCZ, precipitation is decreased. During winter the situation is reversed. The ITCZ migrates north, forming the Southeast Asian monsoon, leaving the north of Australia dry. The southern region of the gulf however, may receive some minor rainfall from the influence of the southeast trade winds that dominate the winter precipitation across the south and east of the continent.

1. The Gulf of Carpentaria

Nhulunbuy Cape York

40 40 30 30 C) C) o 20 o 20 T ( 10 T ( 10 0 0 JFMAMJJASOND J FMAMJ J ASOND

Borroloola Karumba

40 40 30 30 C) C) o o 20 20 T ( T ( 10 10 0 0 JFMAMJJASOND J FMAMJ J ASOND

Figure 1.10a. Mean monthly maximum and minimum temperatures for several stations around the Gulf of Carpentaria (Commonwealth Bureau of Meteorology, 2004).

Nhulunbuy Cape York

300 400 200 300 200 100 100 rainfall (mm) 0 rainfall (mm) 0 J FMAMJ J ASOND J FMAMJ J ASOND

Borroloola Karumba

250 300 200 150 200 100 100 50

rainfall (mm) 0 rainfall (mm) 0 JFMAMJJASOND J FMAMJ J ASOND

Figure 1.10b. Mean monthly precipitation for several stations around the Gulf of Carpentaria (Commonwealth Bureau of Meteorology, 2004).

1.6 The modern environment

The Gulf of Carpentaria fills a broad, shallow basin, spanning 9-18oS from the tropical climate of the north, to the arid Australian interior. The maximum extent of the modern gulf covers a surface area of approximately 500 000 km2 with a catchment area of 1 200 000 km2 (Torgersen et al., 1983) (Fig. 1.1). Such a vast expanse covers a wide range of climatic conditions.

Mean air temperature measured at stations around the gulf is reasonably uniform

~33/21oC (maximum/minimum) (Fig. 1.10). During summer, stratification of the water column occurs in the north of the gulf. This is due largely to the seasonal water temperature difference of 30oC in summer and 24oC in winter, rather than salinity variation, which is relatively uniform at depth (Hill, 1994). Turbidity is low, excepting within the outflow zones of the major rivers, particularly during monsoon discharge (Somers and Long, 1994). A slight salinity gradient is apparent over the extent of the gulf, ranging from 34.8 ‰ at Albatross Bay in the northeast to 36.2 ‰ at Karumba, in the southeast (Rothlisberg et al., 1994).

Rainfall is dominated by the summer monsoon, varying from a two- to three- month dry season in the north, to a six- to eight-month dry season in the south.

Mean annual rainfall is highest in the south of New Guinea, exceeding

2500 mm.a-1, brought by the northwestern monsoon. The precipitation gradient decreases steeply across the gulf, registering only 500 mm.a-1 in the southwest of the basin (Fig. 1.10). This is contributed to predominantly by the southeast trade winds, with some contribution from irregular cyclone activity. Mean annual run-off from Australian drainage (641 000 km2, AWRC, 1987) is 95 615 GL,

1. The Gulf of Carpentaria

representing around 24.7% of the total annual run-off for Australia. Evaporation rates are in the order of 1650 mm.a-1 (Newell, 1973). River discharge into the gulf from Australia is largely seasonal, displaying an extreme ratio of high to low flow. An example is given for the Gilbert and Einasleigh Rivers in the southeast of the gulf, which show a 1150-fold difference between mean monthly maximum and minimum flow (Jones et al., 1993). The sediment load carried by these rivers is as little as two orders of magnitude lower than that delivered by New Guinea waters (Woolfe et al., 1998). A small area of the southern coastal plain of Papua

New Guinea is included within the catchment; however, at present it contributes only a minor amount to the total discharge into the gulf (Torgersen et al., 1983).

The tidal range of the gulf reaches up to 4 m, with circulation in a slow clockwise pattern, and tides vary from semi-diurnal to diurnal (Church and Forbes, 1981).

A pronounced seiche effect is noted, due to the low gradient of the basin

(Woodroffe and Chappell; 1993). The tidal currents across Torres Strait are strong, but are dissimilar and out of phase in the Coral Sea and the Gulf of

Carpentaria (Bode and Mason, 1994). Because of the low gradient, tidal waters extend tens of kilometres inland around much of the gulf. Strong, dry southeasterly winds dominate in the winter, whilst weaker, moist north east to northwest winds occur during the summer monsoon and generate long-period waves (Woolfe et al., 1998).

The greatest biomass within the gulf is toward the southeast, northeast and west, margins. The biota present is dominated by small, opportunistic species, predominantly surface-feeders with broad tolerance limits (Burford et al., 1995).

Light rather than nutrient availability is the limiting factor in phytogrowth, particularly in the coastal zones where turbidity is high. Little land-derived carbon is transported beyond 10-20 km of the coastal fringe

(Rothlisberg et al., 1994).

1.7 Summary

With consideration of the modern setting of the Gulf of Carpentaria outlined above, comparisons may be made with previous conditions and inferences drawn about past environments. Changes in the physiography, with respect to variations in sea level and shifts in climatic regimes may then be elucidated.

2. The State of Knowledge

Background to the environmental and climatic change in the Gulf of Carpentaria region through the last glacial cycle

Changes to the environmental settings of the tropics, particularly during times of sea-level minima, potentially have a great impact on global climate. The Gulf of Carpentaria provides a key site for recording these changes through the last glacial cycle.

2.1 Pleistocene sea-level history

The Pleistocene period, spanning around the past 2 Ma, is dominated by cycles of sea-level change, associated with the expansion and contraction of the world's major ice sheets. Intrinsic to this oscillation are significant changes in global climate, caused by astronomically driven changes in insolation (Imbrie et al.,

1993). The astronomical theory of Milankovitch (1941) describes the periodicity of three components of the Earth's orbit around the Sun:

• Eccentricity maps the shape of the Earth's orbit around the Sun (~100 ka);

• Obliquity refers to the tilt of the Earth's spin axis (~41 ka);

• Precession of the Earth's equinoxes (~23 ka).

Quaternary climate variation is thus related to the distribution of solar radiation, rather than the amount of solar radiation reaching the Earth. These variations in climate have lead to the glaciations and deglaciations of the world's ice sheets, resulting in sea-level change.

Figure 2.1. Relative sea level (showing confidence limits) as determined by Waelbroeck et al., 2002. The numbers above the curve refer to the SPECMAP Marine Isotope Stages (MIS) (Martinson et al., 1987), the extent of which are indicated by the bands; the darker bands representing cooler, or glacial periods.

A proxy for ice volume change is the δ18O of foraminifers in the marine record, measured as a deviation from the mean δ18O of modern ocean water. This forms the basis for the subdivision of the Quaternary into the Marine Isotope Stages

(MIS), which mark the time between successive events (Fig. 2.1). This method, pioneered by Emiliani (1955) and refined by Shackleton (1977), Pisias et al.

(1984) and Martinson et al. (1987), numbers the periods backwards from the present, with odd numbers referring to interglacials (warm, high sea level events) and even numbers referring to glacials. These may be further divided into substages. As the stages correspond broadly with climatic changes, they may be utilised as a temporal reference to both marine and terrestrial records. Orbital forcing affects sea-surface and air temperature, effective precipitation and the amount and distribution of land and sea ice as well as the gas composition of the atmosphere.

2. The State of Knowledge

The most recent glacial cycle extends back to around 130 ka BP. The spatial and temporal fluctuation in sea level is produced by both the variation in ice volume, predominantly the northern hemisphere and Antarctic ice sheets, and the Earth's response to the redistribution of the surface load (Nakada and Lambeck, 1987).

Throughout the last glacial cycle, sea level has varied over a range of about

130 m. Records of sea-level change have been obtained predominantly through benthic and planktic foraminiferal δ18O records (e.g. Shackleton and Opdyke,

1973; Shackleton et al., 1983; Shackleton, 1987) and coral records from Barbados

(e.g. Fairbanks, 1989; Bard et al., 1990; Gallup et al., 1994) and the Huon

Peninsula, Papua New Guinea (e.g. Bloom, 1974; Chappell, 1974; Chappell et al.,

1996). Waelbroeck et al. (2002) present a composite sea-level curve of high- resolution deep marine δ18O records and coral records, taking into consideration changes in δ18O with respect to ice volume, temperature and local fractionation effects (Fig. 2.1). This curve is utilised in the present study as a basis for comparison with the Gulf of Carpentaria record.

2.2 The importance of the Gulf of Carpentaria

The Gulf of Carpentaria presents a myriad of opportunities for the palaeogeographer. The shallow nature of this large basin allows identification of the local extent of the major global transgressive and regressive marine phases, with the sills that separate the Gulf from the open ocean acting as height markers.

Being located within the relatively stable portion of the Australian continental plate, the record of sea-level change in the gulf is more readily deciphered than areas, such as the Huon Peninsula of Papua New Guinea (Fig. 2.2), which have

experienced significant tectonic activity. In addition, being the site of a junction of the Pacific and Indian Oceans, at least at times of high sea level, a record of ocean mixing may be obtained.

The proximity of the gulf to the Indo-Pacific Warm Pool, where the greatest exchange of temperature and moisture between the ocean and the atmosphere occur, also implies that the climatic fluctuations witnessed during the last glacial cycle, may be registered in the sediments of the basin.

Today the presence of warm shallow pools of water over the Sunda-Sahul shelf is essential for the supply of latent heat. During low sea level times, the exposure of these shelves would have had a significant effect on the supply of atmospheric moisture in the region (Nix and Kalma, 1972). An understanding of the timing and extent of the varying hydrology of the gulf provides a unique insight into the climatic regime of the tropics through the last glacial cycle.

2.3 Previous studies of the gulf region

2.3.1 Environments of the Gulf of Carpentaria - past and present

Early work by Phipps (1966, 1970, 1980) recovered around 100 sediment cores of up to 4 m length from a series of north-south transects spanning the gulf. Based on the interpretation of marine sediments overlying non-marine strata, Phipps

(1970) suggested the former presence of a closed basin. Nix and Kalma (1972) further developed this idea and predicted the occurrence of a shallow brackish lake. Exploration for bauxite in the gulf prompted shallow coring 250 km west of

Weipa in 1970 and 1972 by Canadian Superior Mining (Australia) Pty. Ltd.

2. The State of Knowledge

A review of these data by Smart (1977) noted pedogenic over-printing, indicating subaerial exposure in shallow cores from the gulf, and with uncalibrated

14C-dating proposed the development of a non-marine facies prior to 11 ka BP.

A review of the Torres Strait region, focussing on the comparison between the landmasses to the north and south, i.e. New Guinea and Australia is presented in the publication "Bridge and Barrier", edited by D. Walker (1972). Included in the volume is a geological history (Doutch, 1972), description of the geomorphology

(Galloway and Löffler, 1972) and review of climate and its control of biogeography (Nix and Kalma, 1972), with an attempt made to reconstruct past climates (Webster and Streten, 1972). Differences in the modern environments of

New Guinea and Australia are highlighted and inferences drawn about conditions during the LGM, when the greatest area of land was exposed in the region.

The most comprehensive investigation of the Gulf of Carpentaria to date was undertaken in 1982 in a joint study headed by Tom Torgersen (then at the

Australian National University) and Mal Jones (Queensland Geological Survey).

During this project, 1600 km of seismic line and 35 shallow cores were collected.

Torgersen et al. (1983) composed a bathymetric map of the Gulf of Carpentaria from the Royal Australian Navy sounding data. From this they identified closed contours at -53 m bpsl, which they named Lake Carpentaria (Fig. 2.3). In addition, they identified a series of shoreline features and discuss the features of the Arafura Sill, including the existence of prior incised channels, identified from seismic profiles.

The coring campaign identified both lake sediments and the most recent marine transgression (Torgersen et al., 1985, 1988). A lithological key was defined, based on the observation of 35 cores. The authors suggest the presence of a marine to brackish waterbody prior to ~36 ka BP, which dried, forming a thin soil.

A fresh to brackish waterbody then returned to the basin between 36-12 ka BP, before the most recent marine transgression. Full marine conditions, with an open connection across Torres Strait, are considered to have occurred after 8 ka BP.

The extent of the lacustrine unit varied between the present -58 and -67 m contours. The cores beyond these contours show evidence of subaerial exposure before the marine transgression. Fine laminae of authigenic calcite occur in the cores from the deeper part of the basin dating to around 26-23 ka BP, which may be associated with periodic anoxic conditions. These findings were confirmed by a series of geochemical analyses on ostracod valves taken from the cores

(De Deckker et al., 1988; McCulloch et al., 1989).

Palaeoenvironmental reconstructions have concentrated on the two cores from the deepest part of the gulf, GC-2 and GC-10A, both of which are around 2 m in length and represent at least the last 40 ka (Torgersen et al., 1988) (Fig. 2.3).

Material from these cores became the focus of the application of then new geochemical techniques, utilising in particular the Sr/Ca, Mg/Ca (De Deckker et al., 1988), and 87Sr/87Sr ratios (McCulloch et al., 1989) of ostracod valves and authigenic carbonates. Later studies on these cores include the work of

De Deckker and Corrège (1991) and De Deckker (2001) on aeolian activity within the gulf region throughout the last 30 ka, the trace metal concentration of marine and lacustrine sediments by Norman and De Deckker (1990) and boron

2. The State of Knowledge

fractionation between water and modern biogenic carbonate by Vengosh et al.

(1991). Prior to the most recent coring expedition (Chivas et al., 2001), this body of work represented the understanding of the latest Quaternary and Holocene history of the Gulf of Carpentaria.

Figure 2.2. Location of sites used for comparison with the Gulf of Carpentaria.

Records of past fluvial activity in the gulf have been determined by Nanson et al.

(1991) from coring the Gilbert and Einasleigh river alluvial fans (Fig. 2.2). They suggest the increased intensity of fluvial sediment transport occurred during the

Last Interglacial, MIS 3 and the Holocene (Nanson et al., 1991). The early and late part of the glacial period is typified by the development of both calcrete and ferricrete within yellow-mottled clays in the soil profiles of the floodplain deposits (Nanson et al., 1991). This surface is equivalent to the Holroyd surface described by Doutch (1976). It should be noted that no major incision of the

rivers of the gulf would have occurred during the low sea level, Lake Carpentaria phase, as the onshore and offshore gradients are equivalent (Jones et al., 2003).

The Holocene environmental record of the gulf has been established primarily from the coastal plain and the dunefields to the west to the gulf, as much of the basin was inundated with seawater during this time. Episodes of increased sediment influx, particularly in the early Holocene and increased aridity throughout the late Holocene are characteristic, as noted by chenier ridge formation and periodic dune activity (Rhodes, 1982; Chappell and Thom, 1986;

Lees, 1992; Shulmeister and Lees, 1992). Shulmeister and Lees (1995) have presented a vegetation record from Groote Eylandt that supports an increase in local aridity during the late Holocene, with amelioration of climate around 1 ka

BP to the present.

A major ecological study of the modern gulf was undertaken in 1990 and 1991 by the CSIRO Division of Fisheries (Hill, 1994; Long and Poiner, 1994; and the following eleven papers). Sampling by trawling, dredging and grab sample and water quality testing was performed at over 100 stations. These works provide a thorough investigation of current ecosystems of the gulf, with the focus being on the modern prawn industry. Information on the species present within the gulf, including over 680 animal species, the environment of both the seabed and the water column and the relationship between animal and environment was obtained.

A two-dimensional survey of modern sedimentation within the gulf formed part of this study and is complemented by the work of Jones (1987).

2. The State of Knowledge

Other work has focussed on the river systems, particularly in the south of the gulf, such as the Gilbert (Nanson et al., 1991, in review; Jones et al., 1993, 2003) and the McArthur Rivers (Woodroffe and Chappell, 1993). A recent 1:2,500,000 bathymetric map of the Gulf of Carpentaria and the Arafura Sea has been compiled by Grim and Edgar (1998). This bathymetry has been utilised in the present study.

2.3.2 Sea-level reconstructions of the gulf region

Previous sea-level records from the Gulf of Carpentaria region have focussed on the Late Pleistocene, particularly the last marine transgression. There is a general consensus that this occurred around 10 ka BP (e.g. Smart, 1977; Torgersen et al.,

1983; Jones and Torgersen, 1988; Chivas et al., 2001). Phipps (1970) also describes an earlier marine section from a series of short cores collected in the gulf, dated from around 19.6 to 16.7 ka BP and a non-marine section sediment dated 9.2 to 6.5 ka BP, however his assignment of facies is based on questionable microfaunal associations (Smart, 1977; Torgersen et al., 1983) and is not supported by any of the other reconstructions.

The basal unit of the cores investigated by Torgersen and others suggest a marine influence in the gulf prior to ~35 ka BP, as indicated by the sea-level curve of

Chappell (1983). This is identified in the microfaunal and geochemical records

(Torgersen et al., 1985; De Deckker et al., 1988; McCulloch et al., 1989). These authors suggest that the Fly River, which was thought to have flowed into the gulf prior to diversion to the Coral Sea as a result of the Oriomo Uplift (Blake and

Ollier, 1969), may have flushed the marine waters out of the gulf through a

channel across the Arafura Sill. An alternative explanation is that the sea level at the time of deposition of this unit was around the height of the sill, thus only allowing a minor and short-lived marine connection (McCulloch et al., 1989).

Other sea-level information includes reconstructions from the Arafura Sea

(e.g. Fairbridge, 1953; Jongsma, 1970) and the Joseph Bonaparte Gulf (e.g. van

Andel et al., 1967; Yokoyama et al., 2000, 2001; Clarke et al., 2001). These latter records are taken from deeper waters than those of the gulf and are better utilised to ascertain the sea-level minima of the Last Glacial Maximum (LGM), estimated to be around -125 m, from 23 to 19 ka BP (Yokoyama et al., 2000, 2001).

The post-glacial sea-level maximum is considered to have reached +0.5 m apsl

(above present sea level) on the peninsulas to the north of the gulf and +2.0 m at

Karumba (Nakada and Lambeck, 1989). Chappell and Thom (1986) estimate a rapid rise in sea level until 8 ka BP (12 m/1000 a) followed by a slightly slower rise 8-6.5 ka BP (4.5 m/1000 a), then stabilisation between 6.5-6 ka, before slowly falling (0.2 m/1000a).

2.3.3 Palaeoclimatic framework

Records of climate through the Late Pleistocene of Australia primarily reflect changes in available moisture (Kershaw and Nanson, 1993). Kershaw et al.

(2003) have found that throughout the northern part of the country, which is dominated by the effect of the summer monsoon, climate change and variability can be shown to follow Milankovitch frequencies. As such, there is a general

2. The State of Knowledge

positive relationship between global sea level, and temperature change and precipitation regimens in Australia. A further 30 ka periodicity has also been noted by the authors in several records, which they relate to changes in intensity of ENSO variability (Kershaw et al., 2003).

The most widely referenced records of climate change in the north of Australia come from pollen records from the Atherton Tablelands, NE Queensland

(e.g. Kershaw 1978, 1983, 1986) (Fig. 2.2). These stretch back to the Last

Interglacial. It should be noted however, that this region falls under an anomalous climatic regime from the rest of the region, receiving a significant amount of rainfall from the southeast trade winds as well as the Australian monsoon. In addition, influence of anthropogenic activity from ~40 ka BP is considered to have altered the vegetation record. As such, its use as a comparison to the gulf is limited. Generalisations that can be made however indicate wet conditions, similar to today, through MIS 5 and 1 and drier conditions through MIS 2,

(Kershaw, 1986; Kershaw and Nix, 1990).

Other records that give evidence of prior monsoon activity and strength of pluvial episodes in the north of Australia include those of Lake Eyre, central Australia

(e.g. Magee et al., 1995; Magee and Miller, 1998; Croke et al., 1996, 1999;

Johnson et al., 1999), pollen records from offshore NW Western Australia

(van der Kaars et al., 2000; van der Kaars and De Deckker, 2002) (Fig. 2.2) and fluvial records from northern Australia (Nanson et al., 1991, 1992, in review).

These records all generally support wetter conditions through MIS 5, although the peak in effective precipitation is considered to be between 110-100 ka BP

(Nanson et al., 1991; Croke et al., 1999; van der Kaars and De Deckker, 2002).

There is little terrestrial evidence recorded for MIS 4, but drier conditions are indicated at Lake Eyre and in the offshore records (Magee et al., 1995; van der

Kaars et al., 2000). MIS 3 is considered to be a wet period, although not to the same degree as MIS 5 (Johnson et al., 1999; Nanson et al., in review; van der

Kaars et al., 2000). The driest period in most records is centred on the LGM.

During MIS 2, dune mobilisation is active through much of the country and dust entrainment is dominant (e.g. Wasson, 1990; Hesse and McTainsh, 2003). Most of the northern Australian fluvial records indicate low flow, with higher flow events during this period (Nanson et al., 1998, Nott and Price, 1999). An amelioration of conditions is evident in most records through the Early Holocene, with some indicating slightly drier conditions in the Late Holocene (Magee et al.,

1995; Nanson et al., 1992).

2. The State of Knowledge

Figure 2.3. Map showing the location of the cores presented in this study (MD-) and in earlier work by Torgersen and others (GC-)(e.g. Torgersen et al., 1983) referred to herein.

2.4 Present coring and research program

During 1997 six sediment cores were collected from the gulf using a giant piston-corer deployed from the Marion Dufresne. The cruise was a joint

Australian/French/USA expedition, performed as an add-on leg that travelled from Cairns to the northern and central Gulf of Carpentaria then to the Gulf of

Papua. A high-resolution seismic survey was also conducted to aid the selection of core sites. The six cores, from 4.2 m to 14.8 m in length, were collected from water depths ranging from the near bathymetric centre of the gulf to shallower areas to the northwest (Table 1, Fig. 2.3). The strategy for the selection of the core locations was to provide a transect in current water depths of 60-70 m that would intersect older lacustrine sediments, hence enabling an estimation of past water-depths of the Lake Carpentaria (Chivas et al., 2001). The longest of these cores, MD972132 (and hereafter referred to in the abbreviated form, MD-32), which was obtained from the eastern-central section of the gulf at a water depth of

64 m, forms the main focus of this study.

Table 1. Location of sediment cores from the Gulf of Carpentaria.

Core Latitude (oS) Longitude (oE) Water Depth (m) Core Length (m) MD972128 11o11.48' 139o57.53' 62 4.19 MD972129 10o47.36' 138o43.20' 60 6.24 MD972130 12o16.01' 138o44.92' 60 8.23 MD972131 12o03.96' 138o44.98' 59 13.6 MD972132 12o18.79' 139o58.73' 64 14.84 MD972133 12o23.55' 140o20.32' 68 6.68

2. The State of Knowledge

General description of the cores and further details on core extraction and sampling methods are outlined in Chivas et al. (2001). These authors record basal dates from core MD-32 to be around 125 ka, obtained by both thermal- and optically-stimulated luminescence techniques. The dates were taken from a barren quartzose unit, with evidence of subaerial exposure, overlain by a shallow marine facies. This sequence suggests the sea-level rise associated with the Last

Interglacial (Chivas et al., 2001). Non-marine facies, corresponding to the Lake

Carpentaria-phase have been identified in each of the cores. Pollen, microfaunal and geochemical (δ13C, C/N) analyses, performed on the top 1.5 m of the cores, have identified the most recent marine transgression, dated to around 9.7 ka BP by

AMS 14C methods (Chivas et al., 2001).

2.5 Direction and purpose of this study

This thesis study presents the results of analyses of sedimentology, ostracod species assemblages, morphological variation and shell chemistry from the sediments of core MD-32. Palaeoenvironmental reconstructions were based primarily on identification of ostracods from the core and their associated ecological requirements in the modern environment.

Ostracods are aquatic micro-crustaceans found in marine, estuarine and terrestrial waters. Their calcitic shells are commonly preserved in sediments. Identification of ostracodal assemblages, indicative of ecological facies, allows the palaeoenvironmental reconstruction of areas that have experienced changing aquatic conditions. Ostracods are regarded as sensitive indicators of such environmental parameters as salinity, solute composition, water depth and

temperature, variable oceanographic conditions, transportation and preservation

(e.g. Delorme, 1969; Forester, 1986: Benson, 1988; Carbonel et al., 1988; Cronin,

1988; De Deckker, 1988; Neale, 1988; Swanson and van der Lingen, 1994;

Holmes and Chivas (eds.), 2002 and papers therein). Whereas some taxa are found only in very specific conditions, others have a broad tolerance range and thus may be identified from a range of environments.

The ecological interpretations of the ostracod assemblages obtained from this study are largely derived from the literature on the identification of modern environmental settings for such taxa. Early work by Brady (1880) on material collected during the HMS Challenger expedition of 1874 from Torres Strait and

Cape York Peninsula, identified 46 ostracod species. A study by Yassini et al

(1993), on the ostracod fauna in surface sediment samples from the central and southeastern regions of the Gulf of Carpentaria reported 82 species of ostracods.

The marine ostracod fauna from the waters around the north and northwest of

Australia have been the focus of taxonomic studies by Labutis (1977), Hartmann

(1978, 1979, 1981), Howe and McKenzie (1989) and Behrens (1991). Other relevant works in the southeast Asian region include those by Kingma (1948),

Whatley and Zhao (1987, 1988), Mostafawi (1992) and Dewi (1997, 2000).

Studies of non-marine ostracod fauna from the Australian region include

McKenzie (1966, 1978) and De Deckker (1981a,b, c, 1982a, 1983a,b, 1988).

Recent works by Yokoyama et al. (2000), Clarke et al. (2001) and De Deckker and Yokoyama (in review) have used facies analysis of microfauna within core material from the Bonaparte Gulf, northwest Western Australia, to identify

2. The State of Knowledge

sea-level variation associated with the Last Glacial Maximum and the Holocene high stand, respectively. Comparison of ostracod assemblages found within the

Gulf of Carpentaria cores with those stated above allows inferences to be made about the environment in which the ostracods were living when the sediments were deposited.

More detailed palaeoenvironmental information may be obtained through analyses of the chemistry of the ostracod shell. Oxygen and carbon stable-isotope ratios of shell calcite are primarily controlled by the isotopic composition of the host waters and (in the case of oxygen) the temperature at the time of precipitation.

Durazzi (1977) pioneered the analysis of stable-isotopes of ostracods from marine environments, however the method has more frequently been applied to those of non-marine settings. The range of values obtained from these analyses is usually far greater in non-marine environments, the marine environment being buffered from minor climatic variations. Increasingly, ostracod valves are being used in stable-isotope studies to reconstruct palaeoenvironments in both the marine and non-marine realms (e.g. Lister, 1988; Lister et al., 1991; Chivas et al., 1993;

Holmes et al., 1997; Ingram, 1998; Hammarlund, 1999; von Grafenstein et al.,

1999; Schwalb et al., 1999; Mischke, 2001).

This thesis study traces oxygen and carbon stable-isotope ratios through the changing environments experienced in the Gulf of Carpentaria. When used in conjunction with trace element ratios (Mg/Ca and Sr/Ca) of ostracod valves from previous studies in the gulf (De Deckker et al., 1988), and supporting information

from the pollen record, further inferences can be made regarding the hydrology and climate variations recorded in the core sediment.

From the combined evidence referred to above, the palaeoenvironments of the

Gulf of Carpentaria may be elucidated and inferences drawn about past sea level and climatic regimes of the region. The core material is shown to intersect the basin of the past Lake Carpentaria and includes sediments expressing evidence of subaerial exposure. The sequence provides a record of sea level, hydrologic and climatic change through the last glacial cycle within the Gulf of Carpentaria. The implications of an enclosed lake in the area include a land bridge from Australia to New Guinea, both providing a corridor for migrations of people and animals, and an increased continental effect on climatic conditions in the region.

3. Materials and Methods

Objectives and methodological approach undertaken in this study

A combination of techniques has been employed in this study to elucidate changes in the environmental and climatic regimes of the Gulf of Carpentaria region through the last glacial cycle. Sedimentological, micropalaeontological and geochemical methods utilised are outlined below.

3.1 Background to the Gulf of Carpentaria project

This thesis study forms part of a large multidisciplinary project, whose overall aim is to understand the Late Pleistocene evolution of the Gulf of Carpentaria region. The group, headed by Prof Allan Chivas, is based primarily at the

University of Wollongong (UoW), Australia. All researchers involved are working from core material, representing over 50 m of sediment, collected from the gulf in 1997. Preliminary findings and an introduction to the project have been published by Chivas et al. (2001). Members of the team and their primary interest in the project include:

• Jessica Reeves (UoW) ostracods, stable-isotope geochemistry;

• Dr Adriana García (UoW) charophytes, foraminifera;

• Sabine Holt (UoW) foraminifera;

• Martine Couapel (UoW) nannoplankton, OSL-dating;

• Dr Sander van der Kaars (Monash University) palynology;

• Dr Dioni Cendon (UoW) trace element geochemistry;

• Dr Brian Jones (UoW) sedimentology.

As is the nature of group investigations, it is inevitable that some of the results of this project were obtained through discussions with other group members. As such, these results form part of the intellectual property of the group and will not be separately referenced. Where a particular group member has contributed an 47

independent piece of evidence, that information will be referred to in this thesis in the form of ‘pers. comm.’, unless previously published elsewhere.

3.2 Coring methodology and preliminary investigations

Six cores were obtained from the Gulf of Carpentaria during the 1997 expedition

(Fig. 2.3). Prof Chivas, A. Garcia and M. Couapel took part in the coring campaign. With the aid of seismic data, the core locations were chosen to intersect the previous Lake Carpentaria sediments. This strategy was to allow direct estimation of past lake levels, by noting the exposure horizons in the cores.

Although almost 10 tonnes of lead weights were used to impale the piston-corer, core lengths of only 4.2 – 14.8 m were extracted (compared to 40-60 m in open marine oozes commonly recovered using the same system) (Table 2). In most cases, the corer ceased penetration in calcretised and/or ferruginous sediments.

The cores from the outer lake margins display extensive pedogenic over-printing.

Even core MD-33, near the depocentre of the gulf in a current water depth of

68 m, shows evidence of subaerial exposure of the lake floor.

On board the ship, the retrieved cores were sectioned into 1.5 m lengths and passed through a multi-sensor track (MST) scanner, recording magnetic susceptibility, P-wave velocity and γ–ray density measurements. The cores were then split longitudinally for digital photography, digital and Munsell colour measurement and preliminary core description. The information on colour includes reflectance and chromaticity as determined using a Minolta

Spectrophotometer CM-2002. The results of these investigations have been published elsewhere (Chivas et al., 2001). The cores were then wrapped in

3. Materials and Methods

cling-film and stored within thick plastic bags within PVC cylinders at 4oC. Five of the six cores were landed at Weipa, flown to Sydney and then driven to the

University of Wollongong. Where possible, the cores were stored and transported under cool-room conditions. The sixth core, and the main focus of this study,

MD-32, was returned to France. This core was also stored initially under cool- room conditions at CEREGE, Aix-en-Provence.

3.3 Sampling method employed

One half of each of the split cores has been kept as an archive. The other half was sliced through at 1 cm depth-intervals, weighed and stored flat in lidded glass

Petri dishes and sealed with plastic tape. The outer 3 mm of each slice was discarded, to minimise contamination from the PVC tubing. Every fifth 1 cm sample was further dissected for a variety of purposes namely micropalaeontology, palynology, sedimentology (particle-size, X-ray diffraction

(XRD) mineralogy, water, organic matter and CaCO3 content), organic geochemistry and nannopalaeontology (Fig. 3.1). Approximately one third of each sampled slice was kept as an archive and stored in the abovementioned method. A. Garcia, M. Couapel, J. Reeves, S. Holt and S. Pendu performed the dissection of the five cores stored in Wollongong under cool-room conditions during 1998-9. The sixth core was similarly dissected at CEREGE during

January-February 2000 by A. Garcia, M. Couapel and Prof Chivas, however each length was removed from the cool-room and dissected at room temperature before being returned to cool-room conditions. Segmentation of individual core lengths was completed within a day. In total, 5358 individually stored segments were obtained from 53.68 linear metres of core from six coring sites. Some segments

were extracted whole as the presence of large concretions prevented slicing. Of these, 1030 have been further subdivided for each of the previously listed purposes.

Figure 3. Method for sub-sampling the 1 cm slices of core material from the Gulf of Carpentaria. Only the micropalaeontology “circle” and sedimentology section were utilised in this study.

3.4 Sediment particle-size analysis technique

Sediment particle-size analysis was undertaken by A. Switzer in the Faculty of

Engineering, University of Wollongong, utilising a Malvern Mastersizer 2600, employing the method outlined in Chivas et al. (2001). Analyses were performed at 5 cm intervals on all of the MD cores, representing 1030 samples. The results are presented as both mean particle size and clay/silt/sand percentages, where clay

<2 μm, silt 2-63 μm, fine sand >63 μm. As no pre-treatment was performed on the samples, the coarse fraction is considered to be composed of quartz, other minor lithic fragments and biogenic material able to withstand the agitation.

3. Materials and Methods

3.5 Micropalaeontological sample preparation

The sample preparation for micropalaeontological analyses was performed by

J. Reeves, S. Holt, A. Garcia and S. Pendu for cores MD-28, -29, -30, -31 and -33 and A. Chivas, A. Garcia and M. Couapel for core MD-32 The sediment subsamples, weighing approximately 8 g when wet, were oven dried at 40-60oC overnight in 250 ml glass jars. The attendant weight loss was noted and used to calculate the water content of the samples. After drying, the jar was filled with de-ionised water and the sediment left to disaggregate for at least 15 h. The slurry was wet-sieved with de-ionised water through a 63-µm-nylon mesh, mounted in plastic sieve-support rings. The coarse fraction was oven-dried at 40-50oC, weighed and stored in glass screw-capped bottles. For samples weighing greater than 25 mg, aliquots of material were prepared for analysis using a microsplitter and re-weighed.

At all times, care was taken not to contaminate the sample material. Techniques commonly used to accelerate the disaggregation of sediment include the use of the reagents hydrogen peroxide or sodium hypochlorite, supplemented by vacuum roasting or plasma ashing, to remove organic matter. These methods were not employed here. Studies have shown each of these techniques to have an effect on the shell chemistry (Land, 1989). Although this may only be minor in the case of isotopic composition of the shell, the effect on trace-element composition may be both greater and non-systematic (e.g. Griffiths and Holmes, 2000). Hence de-ionised water was the only product used in the cleaning process.

3.6 Microscopic analysis of sediment samples

Microscopic investigation of the >63 μm fraction of cores MD-31, 32 and 33 was performed systematically every 10 cm by the author. One aliquot from each sample depth was observed using a Zeiss SV-8 (research) stereomicroscope, with a nominal magnification range of x8-64. Sedimentological and microfossil characteristics, prominent biota, lithic fragments and evidence of alteration were noted.

3.7 XRD analysis of sediment samples

X-ray diffraction (XRD) analyses were carried out on concretions, rock fragments and other items of sedimentological interest retrieved from the samples during microscopic investigation. The samples were mounted as acetone smears on sample holders designed for a Philips automatic sample changer. These were then

X-rayed at 40 kV, 30 mA using a Philips diffractometer, incorporating a graphite crystal monochromator, using Cu Kα radiation. The scanning speed was

2o 2θmin-1 with a scanning angle of 4-70o 2θ. Data processing was accomplished via Difftech 122 automation and Difftech traces v. 4.0. The patterns were then searched via μPDSM (Fein-Marquart and Associates) using the JCPDS database and interpreted by B. Jones. The analyses were carried out at the University of

Wollongong by D. Carrie.

3.8 Ostracod analysis methodology

The author performed a more detailed micropalaeontological inventory of the samples of core MD-32. All ostracod valves present in an individual aliquot

3. Materials and Methods

(<25 mg) from the >63 μm fraction of each sample were identified and counted.

Where possible, the ratio of adults to juveniles of each species present and the state of preservation of the sample was estimated to establish biocoenosis/thanatocoenosis. In samples with more than one aliquot, the other fractions were examined to ensure that the aliquot chosen for detailed study was representative of the whole sample. Rare species were also noted.

In samples exhibiting a broad diversity (i.e. more than five species present) or low abundance, all adult and juvenile valves were hand picked and stored in cleaned plastic micropalaeontological slides for further examination. Only valves that were sufficiently whole to allow identification were picked. In samples greatly dominated by one or two taxa (e.g. Cyprideis australiensis, Leptocythere hartmanni, Venericythere darwini etc., in some cases greater than 1000 valves) only a few representative valves of the dominant species and all valves of the less prevalent species were picked and stored. Absolute and relative abundances were calculated and presented as for a standardised sample size of 1 g dry sediment.

Sample aliquots taken at 10 cm intervals from cores MD-31 and -33 were also investigated under the microscope. Observations of ostracod species present and their abundance, other biota and lithic material were noted to allow a comparison with core MD-32. No material was extracted from these samples.

Species identification is based on the taxonomic literature including van

Morkhoven (1962), Labutis (1977), Hartmann (1978), Neale (1979), De Deckker

(1981a,b, c,), Whatley and Zhao (1987, 1988), Howe and McKenzie (1989),

Mostafawi (1992), Yassini et al. (1993) and Dewi (1997). The specimen collection is currently stored in the School of Earth and Environmental Sciences,

University of Wollongong.

Ostracods of 72 species from 52 genera were identified in this study. Cluster analysis of the ostracod data collected enabled comparison of the composition of ostracod assemblages (R-mode) and their distribution through the core material of

MD-32 (Q-mode) using the program CORRMAT/PROG (Jones and Facer, 1981).

A reconnaissance examination of previously retrieved material included

99 surface and core-top samples collected by Phipps in 1966 from throughout the gulf, 9 samples from the central and northern part of the gulf collected by The

Australian National University and the Queensland Geological Survey expedition in 1982, and 63 surficial samples from the southeastern margin of the gulf including offshore Karumba and nearby rivers and inlets, collected by B. Jones

(1988) and with the author and S. Holt (1999). This enabled the allocation of facies names to the core assemblages. Ecological tolerance limits of the non-marine taxa present in the core samples were ascribed to the fossil species from other comparable sites in Australia (De Deckker 1981 a, b, c, 1983b, 1988;

McKenzie, 1966, 1978).

3. Materials and Methods

3.9 SEM analysis methods

Scanning electron microscope (SEM) images were captured of each species of ostracod present for identification and recording purposes using a Leica

Cambridge 440 SEM in the Faculty of Engineering, University of Wollongong by the author. In addition, Leptocythere hartmanni and Cyprideis australiensis valves from a range of sample depths were examined, to observe morphological variation of the valves in response to changing environmental conditions. Images of a range of quartz grains extracted from quartz-rich units of core MD-32 were also obtained. The surface texture of the quartz grain is indicative of the mode and degree of transportation, i.e. aeolian, fluvial. Other lithic fragments, such as concretions, were analysed for their chemical composition using the energy dispersive X-ray (EDX) facility of the SEM.

3.10 Stable-isotope analysis of ostracod valves

The species for stable-isotope analysis were chosen based on their prevalence and relative abundance throughout the core. Valves were selected according to their size (weight), maturity and preservation state. A total of 306 ostracod samples from 114 core samples have been analysed for coupled δ18O and δ13C ratios.

Where present, valves were extracted from the samples used for facies identification. In areas of particular interest, further samples were taken at 5 cm or 1 cm intervals. Single valves of Cyprideis australiensis, Neocytheretta spp. (N. adunca, N. vandijki, Alocopocythere goujoni), 2 valves of Venericythere darwini and 3-6 valves (weight dependent) of Ilyocypris australiensis were selected. All valves were individually ultra-sonically cleaned for no more than three half-

seconds in de-ionised water, oven dried at 40oC and checked for adhering particles or diagenic alteration using a binocular microscope. All analyses presented here were performed on well-preserved adults, where possible, or A-1 valves.

Analyses were undertaken in Prof. Chivas' geochemistry laboratory at the

University of Wollongong by the author with the assistance of S. Wang and

D. Wheeler. Samples were reacted with 103% phosphoric acid at 90oC in a

Multiprep online individual-carbonate preparation line connected to a Micromass

Prism III stable-isotope mass-spectrometer. The δ18O and δ13C values are reported in delta per mil (δ, ‰) notation relative to the Vienna Pee Dee Belemnite

(V-PDB) marine-carbonate standard for carbon and oxygen by assigning a δ18O value of -2.20‰ and a δ13C value of +1.95‰ to the international standard; NBS-

19. The precision is recorded to have been better than 0.1‰ for both δ18O and

δ13C.

3.11 Dating methods

3.11.1 Radiocarbon dating

A variety of dating methods have been applied to core MD-32. Accelerator mass spectrometer (AMS) 14C methods have been applied to eleven small whole mollusc samples, from the top 1.5 m of this core. The analyses were carried out at the Australian Nuclear Science and Technology Organisation (ANSTO) AMS centre1. Shells chosen for analysis were well-preserved, clean and did not display clear evidence of reworking. A further 42 AMS14C dates have been obtained

1 The funding for the AMS dates was kindly provided by AINSE grants 98/155R, 01/032 and 02/025, awarded to Prof Chivas.

3. Materials and Methods

from mollusc and ostracod samples from the other MD cores. Conventional ages were determined on the basis of the Libby half-life of 5568 years and related to the NBS oxalic acid standard for adaptation to the international radiocarbon scale.

Where possible, results have been corrected for δ13C content. Calibrated 14C ages

(an approximate calendar timescale, pre-1950 AD) were calculated using the relationships provided by CALIB4.4 (Stuiver and Reimer, 1993), for samples with conventional ages less than 20 ka. For marine samples a reservoir correction of

450±31 years has been applied, which is considered to be the average for NE

Australian waters (Reimer, 2003). Calibrated samples are denoted, for example,

"cal. ka BP". No correction has been applied to the lacustrine samples owing to the lack of information regarding the lacustrine 14C reservoir values for the Gulf of Carpentaria (Chivas et al., 2001).

3.11.2 Amino acid racemisation dating

Amino acid racemisation (AAR) analyses (total acid hydrolysate) were undertaken on fossil marine mussel shells (Anadara (Tegillarca) granosa and

Bassina sp.) from three levels in core MD-32, by C. Murray-Wallace of the

University of Wollongong, following the methods documented in Murray-

Wallace (1993) and Chivas et al. (2001). The uncertainties and assumptions made in determining the preliminary numeric ages are outlined in Appendix D of

Chivas et al., 2001.

3.11.3 Luminescence dating

To extend the dating of the core beyond radiocarbon limits and to provide a reference point for the AAR chronology, a combination of optically-stimulated luminescence (OSL) and thermoluminescence (TL) techniques have been employed. TL dates have been obtained from two quartz rich units of core

MD-32; from 5.81-5.84 m and 14.82-14.83 m, by D. Price, at the University of

Wollongong. OSL dates have also been obtained from the basal unit, by

D. Banerjee, also at the University of Wollongong. The methods, uncertainties and assumptions in determining the ages obtained are discussed in Chivas et al.

(2001).

4. Sediment Analyses

Physical examination and interpretation of the sedimentary parameters of the Gulf of Carpentaria cores

The sedimentology of core material provides information on the environment at the time of deposition and as such, forms the framework of any palaeoenvironmental analysis. The parameters investigated in the Gulf of Carpentaria cores, include particle-size, colour, water content, mineralogy and microfauna. A combination of these factors allows a preliminary reconstruction of environments experienced in the gulf region through the last glacial cycle, including the timing and extent of the palaeolake.

4.1 Sedimentary parameters investigated from the core material

4.1.1 Particle-size

The particle-size of sediment is a physical measure of its primary constituents.

This is indicative of the source material, energy of environment, degree of sorting and mode of transportation and deposition. Most of the sediments analysed from the Gulf of Carpentaria cores are within the clay to fine-sand sediment range

(< 100 µm). Fine particle-size, such as clay, is indicative of low-energy environments and is deposited out of suspension. Coarser material, such a sand-sized fragments, is indicative of higher energy environments.

4.1.2 Colour

The colour of the sediment is primarily a reflection of the mineralogy. It may also be indicative of post-deposition effects, such as oxidation, which produces an orange colour in iron-rich environments. Dark-coloured sediment is commonly due to the sediment's being enriched in organic matter. Within the gulf cores, the predominant colours range from grey to green. The green sediment is generally

associated with marine environments and the grey with non-marine environments, although this is an over-simplification.

4.1.3 Water content

The water content of sediment is associated with the holding capacity and pore-size within the matrix, and hence may show a strong correlation with the particle-size. It is also an indication of the degree of compaction of the sediment.

In the uppermost sediments, the water content gives an indication of the saturation state of the sediment.

4.1.4 Mineralogy

4.1.4.1 Quartz

Quartz forms a dominant component of several of the units of the Gulf of

Carpentaria cores. Investigation of the quartz grains, including the size, colour and lustre, roundness and sphericity, give an indication of the mode of transportation and depositional environment. The degree of sorting of the various grain-types is also an important indicator of the energy of the system.

4.1.4.2 Calcite

Authigenic calcite is a common component of lacustrine environments, forming when the water is supersaturated with respect to calcium carbonate. The occurrence of concretions in the core indicates a change in the chemistry of the host water. This may be caused by altered Eh/pH conditions, following a dilution of the waterbody by an influx of water with subsequent evaporation and

4. Sediment Analyses

concentration, or decomposition of organic matter leading to increased pCO2 of the water. Concretions may form within a soil profile, caused by the mobilisation of carbonate by meteoric or groundwaters and reprecipitation within pore spaces in the sediment. Carbonate concretions found within the gulf sediment are generally calcitic, although barite and siderite have also been identified.

4.1.4.3 Gypsum

The presence of gypsum in the sediment is indicative of evaporative, saline conditions. Primary gypsum is precipitated from supersaturated aqueous solutions, usually identified by prismatic crystals. Secondary gypsum is formed displacively within the sediment at the capillary fringe, identified by pyramidal, discoidal or lenticular crystals (e.g. Magee, 1991). Within the gulf, primary gypsum is more common, preserved as discrete laminae. Secondary gypsum may also form from oxidation of pyrite upon exposure of the core, although this is not considered to be a major effect here. Both pyrite and gypsum occur discretely.

4.1.4.5 Pyrite

Pyrite is formed from reduction of sulphates in anoxic, organic-rich environments at the sediment-water interface. Within the gulf sediment, pyrite is most commonly found as framboidal aggregates. Where pyrite is found infilling whole ostracod carapaces, this suggests rapid burial of the organism following death.

4.1.4.6 Iron-oxide cement

Iron-oxide cement and staining forms from the oxidation of iron-rich material, such as pyrite. It is highly soluble, so requires rapid sedimentation or may be

formed in situ due to prolonged exposure in a dry environment. Iron-oxide mottling of swelling and shrinking clays is common in regions with seasonal or other periodic exposure of the sediment.

4.1.5 Microfauna

Microfauna form a dominant component of sedimentary units, particularly in the environments of the gulf. The groups considered here are Foraminiferida,

Ostracoda, Mollusca and Charophyta. The most common foraminiferal taxon present here is Ammonia, which is common to shallow and marginal marine and brackish non-marine environments. There is current debate over the taxonomy of the various species of Ammonia (Hayward et al., 2004). As the systematics of foraminifers is not considered to make a significant contribution to this thesis, only the genus name will be referred to herein. The other foraminiferal taxa present are referred to by sub-order; rotaliids and miliolids are common in marginal to open marine environments, textulariids are more common in marginal marine and intertidal environments. Special mention is made of planktic foraminifers, as these are indicative of open marine conditions. The molluscs present are referred to here simply as marine or non-marine forms and the family names are given of some of the more common taxa. Charophytes are indicative of lacustrine waters, ranging from saline to fresh. Ostracods may be found in most aquatic environments and are classified here in terms of their gross ecological affinities. Species identification will be discussed in detail in the following chapter.

4. Sediment Analyses

4.2 Depositional facies defined from the gulf cores

The main depositional facies identified in the sediment cores from the Gulf of

Carpentaria are outlined below:

Open marine: well-preserved, diverse and abundant microfauna including planktonic forms and pteropods. Both Arafura Sill and Torres Strait open.

Shallow marine: usually well-preserved, abundant and diverse benthic microfauna, bryozoan and echinoid fragments may also be present. The presence of bairdiid ostracods indicates marine salinity. Arafura Sill is open, but Torres

Strait may be closed.

Marginal marine: lower diversity of microfauna, miliolids and rotaliids common and robust ostracods, scaphopods, broken shell and quartz common. A shallow

(<5 m) open connection across the Arafura Sill is suggested.

Lagoon: low diversity, but high abundance of microfauna, bivalve molluscs may also be common. Quartz is rare, whereas pyrite and organic matter common. A low-energy environment with sea level around the height of the sill is inferred.

Tidal channel: low diversity and low abundance of microfauna. Miliolid and textulariid foraminifers and phytal-dwelling ostracods common. Broken shell material may be abundant, similarly quartz or pyrite. A higher energy environment with channel connection across the sill is indicated.

Mudflat: rare microfauna, those present may be dwarfed and sugary-textured.

Gypseous laminae are common. Pyrite and organic matter may be common, especially alternating with the gypseous laminae. Sediment is generally fine-grained. The mudflats may be supratidal, with some channel connection to the sea during king tides.

Saline lake: low diversity and high abundance of microfauna, particularly the ostracod Cyprideis and the foraminifer Ammonia. Bivalved molluscs and some charophytes may be present. Quartz is rare and pyrite may be abundant. This facies is representative of a closed lake basin, with sea level below sill height.

Oligohaline - fresh lake: a greater diversity, although lower abundance of non- marine ostracods. Charophytes and non-marine gastropods and bivalved molluscs may be present. Ammonia may also be present, although tests are commonly dwarfed. The lake basin is possibly open.

Fluvial: microfauna are rare and those present show evidence of reworking.

Quartz is dominant. Mica and other terrestrial material may also be transported.

Iron-oxide is common in the fluvial facies and may either be pedogenic or transported within the matrix of the sediment.

4.4 Sampled sites

The six cores obtained from the Gulf of Carpentaria were collected in water depths of 59 m to 68 m, near the current bathymetric centre of the gulf, to shallower areas toward the northwest (Fig. 2.3; Table 2). Please note that no corrections have been made for compaction, hence the depths represented are uncorrected values, measured directly from the core.

As outlined previously (Section 3.2), each of the cores underwent geophysical investigation and preliminary description onboard the Marion Dufresne

(Fig. 4.1, App. 1a-e). More detailed observations were made in the laboratory where the cores were sliced into 1 cm thick vertical intervals. The method of slicing employed allowed a far greater surface area of the core to be inspected.

4. Sediment Analyses

Characteristics noted include the colour and texture of the sediment, the occurrence of laminations and the presence of features such as shells, concretions, evidence of bioturbation and any other features visible to the naked eye.

Particle-size analysis and water content of the sediment were also determined for each of the core samples at 5 cm intervals.

Cores MD-31, -32, and -33 were selected for further investigation, as MD-31 and

-32 were the longest two cores extracted and core MD-33 was taken from near the deepest part of the gulf. Cores MD-28 and -29 both displayed extensive pedogenesis, deeming them less suitable for geochemical studies and unlikely to fully preserve fossil material. Core MD-30 was set aside for future studies, pending the outcome from the first three cores investigated. Samples at 10 cm intervals were prepared using the method described in the previous chapter for micropalaeontological investigation and the sand-sized fraction (>63 µm) was viewed under a binocular microscope. Observations made include description of the quartz grains and other lithic fragments, noting the presence of pyrite and iron-oxide staining, description of the biogenic material present including foraminifers, ostracods, charophytes, molluscs and other shells and fragments, and the general state of preservation of the sample (App. 2 a-c.). Representative samples were selected from each of the observed units of cores MD-28, -29 and

-30 and viewed microscopically to allow comparison with the other cores.

After preliminary microscopic observation, core MD-32 that shows the most abundant and well-preserved microfossil material, was selected as being most suitable for more detailed study. This was chosen as the "key core" from which

depositional units were defined and used for comparison with the other cores.

Samples were once again investigated under the microscope for qualitative and quantitative ostracod analyses. The results of these studies are outlined in

Chapter 5. Representative valves from each ostracod species present and a range of quartz grains were extracted for SEM imaging (App. 3). In addition, a selection of lithic material, including concretions and crystals was removed for identification and analysis using the EDX-facility of the SEM (App. 4a).

Concretions and other fragments from several of the cores were also examined utilising X-ray diffraction analysis (App. 4b).

The results of the sedimentary analysis of core MD-32, including geophysical, colour, physical and initial microscopic investigations, are presented and discussed here (Fig. 4.1, 4.2). Preliminary observations of the other cores are presented below (App. 1a-c, Fig. 4.4-4.8). A comparison between each of the cores, showing the extent of the units, is shown in Fig. 4.9.

4. Sediment Analyses

4.5 Sediment description of core MD-32

Initial observation of whole lengths (1.5 m each) of core MD-32 was made with the naked eye before slicing. The sediment represented (Fig. 4.2) comprises largely unbedded silty-clay, with few obvious sedimentary structures developed.

The colouring of the sediment is predominantly grey, with some mottling from

5.05 m to 8.05 m, most developed between 5.60 m to 5.05 m, and orange staining at the base of the core and at 3.73 m to 5.05 m. The upper 0.40 m of the core is distinct, pale green. Slicing of the core revealed further features including the texture of the sediment, macrofossils, concretions and more detailed colour variation, allowing the preliminary division of the core into sedimentary units.

The lowermost section of the core (14.84-13.95 m) shows some yellow staining of the otherwise greenish-grey silty-clay and fine sand. The sediment from 13.95 m to 9.30 m comprises a dark grey silty-clay, grading to a more greenish-grey clayey-silt, with scattered shell material throughout. Small, lightweight, pale-grey vesicular pebbles, resembling pumice, were extracted from the core between

11.36 m and 11.30 m. From 11.01 m to 10.93 m are a series of millimetre-thick coarse, pale and slightly irregular crystalline laminae. A shell layer, with broken and bleached material is visible at 10.10-10.00 m.

A thin unit of gritty sediment marks a transition to darker sediment (9.30-8.88 m), capped by a shell layer consisting of small bivalved molluscs. Dark grey silty-clay extends to 8.08 m, interrupted by thin pale grey crystalline laminae, each approximately 1-2 mm thick, between 8.80-8.60 m and 8.30-8.10 m.

From 8.08 m to 6.55 m the sediment is light greenish-grey colour, with increasing degrees of orange mottling. Irregular light-grey dry, silty laminae, around 2 mm thick, occur in the core at 7.57-7.40 m and 7.29-7.21 m, with darker laminae at

7.19-7.15 m and 7.43 m.

A gradational change in colour to a darker greenish-grey with prominent orange mottling is observable in the overlying sediment (6.55-5.68 m). Several fine-grained, rounded carbonate concretions have been extracted from this unit, varying in size from 0.5-2 cm. Between 5.68 m and 5.05 m the dark grey silty-clay is heavily mottled and numerous concretions are present, many of which are orange. Scattered bleached shell material is also present. Above 5.05 m, the sediment returns to a lighter greenish-grey colour, with some orange staining to

3.73 m, and contains abundant shell material. Evidence of sediment disturbance, most likely due to minor bioturbation, is present from 4.20 m to 3.73 m.

A clear sediment change occurs at 3.73 m, to a very dark grey, fine silty-clay alternating with lenses of darker silt-sized material between 2.77-2.55 m, and irregular lighter grey laminae around the base of the unit to 3.35 m. From 2.49 m to 1.03 m the sediment consists of greenish-grey grading to mid grey, clayey-silt.

Pale, 1-mm-thick, incomplete laminae are present at the base of the unit and a layer of small bivalved mollusc shells is visible (1.47-1.35 m). The sediment from 1.86 m is sticky and crumbly, becoming darker and more crumbly above

1.03 m, with even darker marks suggestive of organic-rich matter to 0.50 m. A piece of polystyrene was placed in the core (0.47-0.40 m) to replace material that was lost on recovery. The sample immediately above the gap resembles the

4. Sediment Analyses

underlying unit before an abrupt change to wet, greenish-grey bioclastic ooze, comprising the upper 0.38 m of the core.

4.6 Results of preliminary observation of the physical parameters

4.6.1 Particle-size analysis

The particle-size analysis (Fig. 4.1) of MD-32 sediment has revealed fairly uniform grain-size throughout most of the core, the mean being 9.82 µm. There is an inverse relationship between the percentage of clay and silt fractions, with the sand fraction representing less than 10% through the majority of samples in the core. Where possible, particle-size analysis was performed at 5 cm intervals through core MD-32, as outlined in Section 3.4.

The basal samples, from 14.84 m to 13.95 m, have the coarsest sediment in the core, averaging 25.06 µm. Discrete peaks in excess of 63 µm mean particle-size occur at 14.82 m, 14.75 m and 13.95 m and are represented by 30.56, 19.00 and

25.54% coarse fraction respectively. This corresponds with an increase in the quartz content of the sediment. Clay and silt are represented in around equal proportions.

From 13.90 m to 9.30 m there is a gradual coarsening-upwards of the sediment.

The mean particle-size through this unit ranges from around 5 µm at the base to around 18 µm toward the top. The clay fraction is dominant in the basal half of this unit, representing around 60% of each sample. This clay percentage declines to around equal proportions with the silt-sized fraction from around 11.60 m to

10.60 m. Silt becomes the dominant fraction in the upper samples, with the fine sand fraction representing as much as 10%. Peaks in the coarse fraction largely determine peaks in the mean particle-size through this unit. These correspond to increases in biogenic material in these samples, rather than a coarsening of the grain-size per se.

The shell layer at 9.30 m to 8.95 m is noted by an above-average particle-size of

13.49 µm, again due to biogenic material rather than quartz or other rock fragments. The top and bottom samples of the shell layer have the coarsest particle-size, corresponding to the increased amount of shell material in these samples. The clay and silt-sized fractions are represented in near-equal proportions.

Through much of the remainder of the core, the particle-size is strikingly uniform, with a mean of 6.19 µm from 8.90 m to 0.45 m, with minor peaks of >12 µm at

8.45 m, 6.90 m and 1.10 m. From 8.90 m to 5.70 m, clay is the dominant particle-size fraction, representing around 60% of the sediment, with the silt- fraction making up most of the remainder. The fine sand-fraction, where present, rarely represents more than 5% of the sediment. For a brief interval between

5.60 m and 4.70 m, the silt-fraction dominates, representing 65-50% of the sample. This is primarily due to an increase in fine-grained quartz material and small carbonate concretions found in these samples. Above this section the clay-sized fraction regains dominance, representing 70% of the sediment up to

1.60 m. Silt is the other major component, with the sand fraction contributing less than 5% where present. Again an increase in biogenic material in the sediment

4. Sediment Analyses

above 1.50 m has contributed to an increase in the silt-sized fraction, which comprises around 60% of the samples to 0.40 m.

The uppermost sediment, from 0.40 m to core top, is comparatively coarse; with a mean particle-size of 20.46 µm. Again, this is noted by an increase in the sand-sized fraction and correlates with an increase in biogenic material as well as quartz and other lithic fragments.

4.6.2 Water content

The water content of the sediment of core MD-32 was established every 10 cm, where possible, by comparing the wet and dry weights of the sediment used for microscopic analysis. On first observation of the graphs (Fig. 4.1), the water content of the core shows some correlation with the particle-size measurements and indeed the sedimentary units described above. The lowest values of around

20% occur in the basal unit of the core, corresponding to the coarser grained sediment. Water content increases abruptly to 40% at 13.90 m, decreasing gradually to 25% at 9.60 m, following the grain-size trend. Across the shell layer the water content increases to 45% at 8.95 m, and back down to 35% by 8.50 m, with excursions of around 20% at 8.65 m and 8.55 m. The water content steadily decreases to 27% by 5.65 m, in the region of mottled sediment, with a single low value of 17% at 7.40 m corresponding to a silty lamina. The trend increases again to 35% at 5.45 m, remaining constant until 3.70 m, where there is a jump to 45% and noticeable sediment change. Between 3.70 m and 2.65 m, the trend is increasingly negative reaching 35% at 2.0 m, with values around 30% at 2.45 m,

2.30 m and 2.05 m again corresponding to silty, dry sediment lenses. Up to

1.00 m, the water content of the sediment remains around 40%, and then decreases gradually through the sticky, crumbly clays to 25% at 0.40 m. An abrupt increase to over 50% water occurs in this upper unit, decreasing to 40% at the top of the core. Although the grain-size of this uppermost unit is comparatively coarse, the sediment was still wet and 'soupy' on recovery. The relatively high water content throughout the core is indicative of only minor compaction of the sediment.

4.7 Microscopic observation of the sediment of core MD-32

The sedimentological parameters observed under microscope from the coarse fraction (>63 µm) samples2 taken at 10 cm intervals from core MD-32 have been tabulated and are represented in App. 2b. The samples from the base of core MD-

32, (14.8-13.9 m), are barren of microfauna, but contain abundant quartz. There are two types of quartz present, one clear, glassy and angular and the other frosted, subrounded and showing iron-oxide staining, particularly in the lower samples (App. 3a). The samples are moderately well-sorted. Both are fine-grained, commonly <100 µm. The silt and clay fractions showed orange, iron-oxide staining upon sieving. In some samples, hard flakes and chunks of this material remained in the coarse fraction. Mica is present, and more abundant in the upper samples. Muscovite is the dominant species, although some minor biotite and chlorite occur towards the top of the unit. Other minor rock fragments are present in increasing quantities upwards through these samples, including

2 In this Section, a 'sample' refers to the >63 µm fraction of dry sediment, utilised for microscopic examination.

4. Sediment Analyses

small fragments of pumice (identified by EDX, App. 4a) between 14.05 m and

13.9 m.

From 13.9 m, the microfauna are abundant in the samples, including a variety of ostracod genera, all foraminiferal suborders, (Rotaliina, Miliolina and

Textulariina, although in low abundance), bivalve and gastropod molluscs, scaphopods, echinoid spines and fish teeth. The sediment of samples at 13.9 m and 13.8 m has a large coarse fraction in comparison to the rest of the unit. Above

13.5 m, the assemblage broadens to include more abundant foraminifers, including benthic and planktic species and pteropods. The microfossils are generally well-preserved, with some broken fragments of larger shell material.

The samples at both 13.2 m and 13.1 m however, contain larger foraminifers with irregular growth habits. Quartz is present, representing up to 50% of most samples, although with decreasing iron-oxide staining and fewer glassy grains.

Flakes of mica are rare. Pyrite becomes increasingly abundant as framboidal aggregates and replacing organic material. Small quantities of glauconite are present and pumice fragments occur in the samples between 13.4 m and 13.3 m.

Between 13.0 m and 12.0 m, the faunal assemblage comprises a lower diversity of ostracods and large miliolid and rotaliid foraminifers, in high abundance. Other faunal material includes numerous small bivalves, fish fragments and broken shell material. Fragments of pteropods are also present in the lower samples. Quartz is increasingly common and is largely well-sorted, subrounded and frosted in appearance, representing as much as 80% of some samples. Pyrite is present, infilling microfossils at the top of the unit, but decreasing in abundance upwards

through the unit. Glauconite and mica are present in small amounts in most samples.

Above 11.9 m the microfaunal assemblage increases in diversity and abundance, with a wide variety of both ostracods and foraminifers, including textulariids, present. Pteropods and planktic foraminifers are absent. Quartz representation is decreased through the samples, but remains well-sorted. Dark-brown pyrite is common and appears to have formed as a replacement of organic material and as cement between or infilling shell material.

Between 11.2 m and 10.7 m, the coarse fraction samples are much smaller and fine-grained with fewer microfossils and rare quartz. Dwarfed forms of Ammonia, exhibiting a sugary texture, dominate the foraminiferal assemblage. Pyrite is abundant, as framboidal aggregates and as a replacement of organic material.

Small pyramidal gypsum crystals are also present. Although both pyrite and gypsum appear in the same sieved sample, representing 1 cm of core sediment, within the intact core they occur as discrete laminae.

Above 10.6 m the preservation of the foraminiferal tests is improved and the ostracod assemblage is highly diverse. Fragments of broken shell material and echinoid spines are also common, with a layer of abundant bivalves and other shell and fish fragments around 10.1-10.0 m. Quartz occurs in low abundance

(<20%) and pyrite is present, infilling microfauna and as framboidal aggregates, not exceeding 100 µm. This basic assemblage continues until 9.4 m, although the preservation and abundance of the microfossils gradually declines, with many

4. Sediment Analyses

showing evidence of reworking. Small carbonate concretions are also commonly present.

A dramatic change occurs at 9.3 m. Almost exclusively the ostracod, Cyprideis and the foraminifer, Ammonia replace the diverse microfaunal assemblage in great abundance. Broken shell and fish fragments, many of which are black and shiny, are also common. This unit continues to 9.0 m with increasing shell material, including blackened bivalve and gastropod molluscs, which EDX and XRD analyses reveal to be pyritic (App. 4a,b). The samples are dominated by biogenic material and contain abundant pyrite and organic matter. Very little quartz is present.

The samples between 8.9 m and 8.1 m are fine-grained with flakes of pyritic material and pyritised organic fragments, common prismatic gypsum crystals and minor quartz. Microfossils are rare and where present comprise mainly dwarfed, sugary Ammonia and isolated ostracod valves. The small coarse fractions continue through to 6.7 m. Planktic foraminifers occur within some of the samples, however ostracods are extremely rare, except in the sample at 6.7 m that hosts Cyprideis valves and fragments of charophyte oospores. Quartz and mica flakes are present in small quantities in most of the samples as are small gypsum crystals. Flakes and fragments of iron-oxide-rich material increasingly replace pyrite.

From 6.6 m to 5.7 m the samples are much larger and comprise a large percentage of quartz and concretions. The quartz grains are on average smaller than those in

the basal unit of the core, and are generally more rounded, showing a higher degree of sphericity. Most grains are frosted, commonly yellow-stained. There are two types of concretions present in the samples; between 6.6 m and 6.2 m the concretions are fine-grained, pinkish toned and around 200 µm across, whereas between 6.1 m and 5.7 m they are larger, up to 500 μm diameter, and white and yellow stained often with hollow centres. XRD analysis reveals them both to be principally calcite, with siderite in the former and minor iron-oxide in the latter

(App. 4b). Mica flakes are also present and oxidised pyrite and gypsum occur rarely in the lower samples. Ammonia remains the dominant microfossil, with both large and small tests being preserved, but generally display minor dissolution. Other isolated larger rotaliids occur between 6.3 m and 6.1 m. These are abraded, indicating reworking.

In contrast to the previous samples, quartz is virtually absent from 5.6 to 3.8 m.

Ostracods dominate the samples, generally with only two species represented.

The ostracods commonly show evidence of dissolution, including opaque and chalky valves, particularly from 5.6 m to 4.8 m, and some orange staining from the surrounding iron-oxide-rich fine-fraction sediment. Other fauna present include foraminifers, predominantly Ammonia, and both whole and broken bivalve and gastropod molluscs and fish fragments. Shell material is bleached where present above 5.0 m and is particularly abundant between 5.0 m and 4.5 m.

Pyrite is present in small amounts up to 5.0 m and above 4.0 m, commonly infilling microfossils. Flakes and small clumps of ferruginous clay occur throughout.

4. Sediment Analyses

Above 3.7 m there is little evidence of iron-oxide in the sediment. The samples are rich in pyrite to 2.7 m, as small framboidal aggregates and infilling microfossils, and contain little quartz. The microfauna are numerous, once again dominated by Ammonia and Cyprideis that show good preservation. Between

2.2 m and 1.7 m the samples are very fine-grained and whilst ostracods are abundant and well-preserved, the foraminifers present are in low numbers; the

Ammonia are small and frosted in appearance, and there are rare reworked rotaliids. Some broken shell material is also present along with some small fragments of iron-oxide-rich material. The thin silty laminae observable in the un-cut core from 3.73-3.35 m and 2.77-2.55 m are not identifiable in the coarse fraction of the sediment matrix under the microscope.

The size and preservation of the samples increases markedly from 1.6 m to 1.3 m.

Well-preserved bivalved molluscs form a shell layer observable in the cut surface of the core sediment. Fragments of charophytes, fish and small gastropods are also present. Quartz and pyrite are absent. The size of the samples decreases from 1.2 m and whilst ostracods are less abundant, the variety of species present increases. Much of the shell material present, including planospiral molluscs, is broken. Fragments of organic matter, however, have been preserved. One sample, at 0.9 m, stands out as containing planktic foraminifers within the assemblage, whilst the ostracods remain diverse and numerous and charophyte fragments are abundant. Much of the sample has a 'sugary' quality to it and small, gypseous crystals are present. From 0.8 m to 0.6 m the samples are larger, containing abundant broken shell material and some small white concretions, however the overall assemblage, without the planktic foraminifers, resembles that

seen below 0.9 m with some evidence of gypsum and abundant pyrite above

0.7 m.

From 0.5 m, the coarse fraction samples are much larger, with increasing amounts of quartz, small concretions and large amounts of shell fragments and broken ostracods. Both whole charophytes and fragments are preserved and Ammonia is well represented, gradually joined by other foraminifers initially miliolids. The samples above 0.3 m show a dramatic change in the assemblage to include a vast diversity and abundance of foraminifers, including rotaliids and textulariids, and ostracods. Fragments of gastropods, echinoid spines, scaphopods and oysters are present, as are small carbonaceous pellets. Pteropods and planktic foraminifers have not been identified in these samples. This unit is present to the top of the core.

4.8 Discussion and interpretation of the sediment of core MD-32

Utilising the combined observational and physical evidence above, the sediment of core MD-32 may be sub-divided into a series of differentiable units, numbered from base to core top (7→1), and based upon the environments in which they were deposited (Fig. 4.2). Subunits are utilised where there is a noted variation within the broad environmental setting. These units have been dated by a variety of methods, as outlined in Section 3.11 and tabulated in App. 5. The dates have been included in the following discussion to assist in the interpretation of the core material and comparison between each of the collected cores. Comparisons may also be made with previously published reconstructions, such as Torgersen et al.

(1985, 1988).

4. Sediment Analyses

The basal unit 7 (14.84-13.95 m), being high in quartz, consequently having low water content, and devoid of microfauna, is indicative of a dynamic environment.

The variety and sorting of the quartz grains suggests extensive transportation, and the discrete peaks in the very fine sand fraction particle-size through this unit imply periodic relative increases in coarser material. The samples show orange staining of the finer fractions and coating of some of the quartz grains as opposed to mottling of the sediment, suggesting that the iron-oxide has been transported as part of the sediment, rather than forming as a product of pedogenesis. The orange staining is most pronounced in the lowermost samples, indicating either closer proximity to the source material of progressive stripping of the landscape. The variety and abundance of mica in the upper samples of the unit imply a lower energy environment. In combination these factors support dry conditions, followed by increased fluvial activity and deepening water upward through the unit. This sequence resembles the modern prograding delta packages that extend from the major rivers of the gulf, such as the Gilbert River (Jones et al., 1993).

Dates of 123±16 ka and 125±12 ka have been obtained from the base of this unit by TL and OSL methods respectively (Chivas et al., 2001). A further OSL date of

126±6.25 ka was established for the top of this unit (Martine Couapel, pers. comm., 2003). These dates indicate rapid deposition of the basal unit of the core, most likely due to fluvial activity, around the time of the marine transgression following the last interglacial, (Marine Isotope Stage (MIS) 6/5e, with a duration of 130-116 ka; Martinson et al., 1987; Stirling et al., 1998). There is no direct evidence of a marine incursion at this time, such as marine microfauna, however increasingly wet conditions with rising sea level may be postulated.

The marine transgression is made obvious in the sediment of unit 6, by the abundant marine microfauna. The homogeneous appearance of the unprocessed core sediment throughout unit 6 (13.95-9.30 m), showing little variation in colour or texture, indicate a broadly similar deposition environment, with no evidence of subaerial exposure. The grain-size analysis notes a coarsening upwards of the sediment, and concurrent decline in water content, indicating an increased sedimentation rate through the unit. When viewed microscopically, the microfaunal assemblages are seen to vary significantly, and are clearly differentiable as subunits.

The lowermost subunit 6f (13.9-13.1 m) is defined on the basis of the diverse array of well-preserved foraminifers and other shallow marine material including echinoid spines. The relatively large coarse fraction of the lowermost samples of this unit suggests an initial rapid incursion of marine material concurrent with the breaching of the Arafura Sill. The introduction of planktic foraminifers and pteropods above 13.5 m indicates influence of open marine waters. As pteropods in particular are very fragile, their preservation indicates a lower energy depositional environment. The broken shell material, quartz and minor mica, suggesting some transportation of terrestrial and coastal matter to this part of the gulf by fluvial activity, however the occurrence of pyrite is indicative of a reducing, anoxic, low-energy environment and is supportive of a low sedimentation rate. Most of the pyrite occurs in these samples as microscopic aggregates, commonly > 100 µm. The increasing abundance of pyrite in the samples is most likely due to either an increase in organic matter or a decrease in energy of the environment, perhaps due to deepening water. A date of 108±16 ka

4. Sediment Analyses

BP has been obtained from 13.18-13.19 m by AAR analysis of an articulated

Anadara (tegillarca) granosa, a common mangrove mollusc species (Chivas et al., 2001). This initial marine subunit represents an established shallow marine environment open to at least the Arafura Sea. The presence of planktic foraminifers, and pteropods suggest that Torres Strait may have been breached at this stage also.

A decrease in the diversity of microfauna and prevalence of more robust forms, particularly foraminifers, as well as an increase in the quantity of fish fragments and shell material, especially small bivalves, in subunit 6e (13.0-12.0 m) indicate a restriction of marine conditions and a more dynamic environment. There is abundant well-sorted subrounded quartz, however overall the samples are still predominantly clayey-silts, suggesting a more marginal depositional environment, with an increased influence of fluvial activity. The decrease in pyrite through the subunit may be due to a decrease in organic material or increased energy of the environment. An AAR date has been obtained from an articulated Bassina sp. mollusc at 12.38-12.40 m of 122±18 ka BP (Chivas et al., 2001). Although this and the former date suggest an age reversal, the two dates fall within their respective error margins. In addition, as both dates were obtained from articulated, shallow burrowing molluscs, the degree of post-mortem reworking is considered to be minimal they should be broadly contemporaneous with the surrounding sediment.

Between 11.9 m and 11.3 m (subunit 6d), the microfaunal assemblage broadens and the amount of quartz decreases supporting deepening water and more open

marine influence, distal from fluvial sources. The coarser grain-size in these samples reflects the higher microfossil content. However, neither planktic foraminifers nor pteropods are present suggesting that open marine conditions were not restored to the same extent as in subunit 6f. This may otherwise be an indication of decreased sedimentation rates at this time. The increase in the presence of pyrite, both as infilling microfossils and in the form of small crystal aggregates again suggests either the presence of more organic matter or deeper waters.

A clear restriction of environment is noted in subunit 6c (11.2-10.7 m).

Compared with the diverse species present in the underlying subunit, very few microfauna are preserved in these samples and only rare quartz occurs. The dwarfed and sugary appearance of the foraminifers present indicates a restricted and hypersaline environment. Bi-pyramidal gypsum, such as is present in these samples, forms displacively within sediment. The preserved gypseous laminae, which alternate with dark organic and pyrite-rich laminae, suggest periodic inflow and stagnation and evaporative concentration of the waterbody. A regression of sea level to around the height of the Arafura Sill is implicated, along with reduced fluvial activity.

A renewed connection to the marine environment may be noted from 10.6 m to

10.2 m (subunit 6b) with better representation of the coarse fraction, including well-preserved foraminifers and a great diversity of ostracods species. A significant amount of broken shell material and some quartz is also present, particularly in the lower samples of this subunit suggesting a shallow marine

4. Sediment Analyses

setting of moderate energy and sedimentation rate. The sedimentology and microfaunal assemblage of this subunit most closely resembles that of the core-top samples of unit 1. A renewed connection to at least the Arafura Sea in a dynamic environment is suggested.

The shell layer (10.1-10.0 m) largely comprises reworked bivalves and other shell and fish fragments and represents a concentration deposit, which may indicate a shoreline feature. A further regression of seawater is implicated. The relatively coarse particle-size of this layer is most likely a function of the large percentage of biogenic material in an otherwise silty-clay matrix. Above this layer (subunit

6a, 10.1-9.4 m) the preservation of shell material within the samples declines up sequence, implying transportation of material and the gradual decrease in sea level and restriction of marine waters. Foraminifers appear to be affected by this change in environment before the ostracods; the diversity decreases, with more robust species remaining through this subunit. The location of the sea level at this time is inferred to have been around the height of the Arafura Sill, with channel connection to the centre of the basin, rather than an open embayment.

Although very few dates have been obtained from unit 6, the sedimentary sequence, varying through open marine to a more restricted embayment, is indicative of the sea-level fluctuations through marine isotope stage 5 (128-74 ka

BP, Martinson et al., 1987). It should also be noted that at present the sedimentation rate in the locality of core MD-32 is very low. As such, periods of high sea level may not be well represented in the core sediment.

Unit 5 (9.30-8.88 m) represents a distinctive change in environment. The absence of foraminifers, other than the near-ubiquitous genus Ammonia, in conjunction with dominance of the ostracod Cyprideis is indicative of an environment separated from fully marine waters. This faunal coupling has previously been described as the 'Lake Carpentaria facies' with these two dominant species being present in the brackish lacustrine material of unit IV and II in the cores examined by Torgersen et al. (1985, 1988). The abundance of these two microfossils, along with plentiful shell and fish material, identifies this to have been a productive environment, the species existing within it thriving. Pyrite has replaced much of the shell material giving it a black speckled appearance, indicating an organic-rich, anoxic substrate. The shell layer present at 9.0 m may represent a shoreline feature, indicating the extent of the brackish lake or lagoon at this time.

The samples of unit 4 (8.88-5.68 m) are indicative of a very shallow saline environment, supporting very few microfauna. Clay is the dominant particle-size, representing at least 60% of the sediment matrix. The first two subunits are represented by very small coarse fractions that become larger in the remaining two subunits with more abundant quartz, indicating an increase an energy of the environment through the unit. The diminutive grain-size of unit 4d (8.9-8.1 m), which contains very little microfauna, with only rare Ammonia present, suggests an inhospitable environment, possibly with ephemeral waters. The fragments of pyritic organic material indicate anoxic conditions, perhaps to due to covering by algal mats. The presence of prismatic gypsum occurring as irregular fine crystalline laminae at regular intervals between 8.8-8.6 m and 8.3-8.1 m suggests an evaporating environment with episodic sediment influx, such as a fluctuating

4. Sediment Analyses

shoreline of a shallow, brackish lake or lagoon. These stubby prismatic crystals are precipitated from supersaturated aqueous solutions (Magee, 1991). The preservation of the laminae suggests deposition in a sheltered environment, away from vigorous wave or current action. This sequence resembles the broad flat supratidal zones of the gulf that extend beyond the high tide mark during the dry season today, periodically flooded during the wet and during spring tides (Jones et al., 1993). Intermittent connection to the open ocean is suggested, most likely via channels across the Arafura Sill.

The samples of subunit 4c (8.0-7.0 m) are also predominantly silty clays, but show evidence of mild pedogenesis, the sediment being mottled with iron-oxide, indicating drier conditions. The absence of ostracods, and the presence of rare small Ammonia and planktic foraminifers, favours transportation in suspension by either aeolian or alluvial means, rather than in situ deposition. The small size and light weight of these tests allows transportation over extended distances and the fine-grained sediments provide a soft bottom for deposition and preservation.

Such taxa have been described up to 100 km inland in tidal rivers of the Northern

Territory (Wang and Chappell, 2001). The microfauna present have a sugary texture indicating slight abrasion and possible post-depositional recrystallisation.

Laminations of small prismatic gypsum crystals are apparent, particularly in the lower samples of this unit, identifying episodic drying. Pyrite is present in most samples in small quantities, but is commonly oxidised, indicating some degree of subaerial exposure after formation. A shallow saline waterbody with fluctuating margins is postulated, although more restricted than the underlying subunit, with

cycles of flooding followed by evaporation and drying. Comparisons may be made with the modern Lake Eyre playa (e.g. Magee et al., 1995).

From 6.9 m to 6.7 m (subunit 4b) both the amount of quartz and the abundance of microfauna, including Ammonia with abraded tests showing evidence of reworking, increase in the samples suggesting a more energetic environment.

This is supported by the increase in the sand-sized fraction through this subunit with the quartz grains quite rounded and well-sorted, showing some pitting, supporting extensive transportation. The presence of ostracods, predominantly

Cyprideis but including some true marine forms, suggest transportation from a permanent water source and implicate a minor marine incursion at this time, most likely through a channel incised into the Arafura Sill.

The orange-mottled samples with calcareous concretions in unit 4a (6.6-5.7 m) are indicative of subaerial exposure and pedogenesis. The increase in quartz content, the grains being well-sorted and more rounded than those in unit 7, again indicate reworking and transportation of sediment around exposed margins of the basin.

The scarcity of microfauna is not surprising in this unit, given the prevalence of concretions. The rare fossils that occur are poorly preserved. The marine foraminifers present are all heavily abraded, suggesting they were also reworked from the surrounding exposed margins. The Ammonia and Cyprideis on the other hand show evidence of dissolution, implying that they grew in situ and were later chemically altered, most probably by the influx of meteoric waters. A preliminary

TL date of >64.3±7 ka BP was obtained from a quartz sample taken at

5.84-5.81 m (D. Price, pers. comm.). This date and the evidence of subaerial

4. Sediment Analyses

exposure indicate low sea level at the time of deposition, contemporaneous with marine isotope stage 4 (74-59 ka BP, Martinson et al., 1987). There is evidence for renewed fluvial activity in the Gilbert River, in the southeast of the gulf.

Similar units of quartz-rich material with iron-oxide and carbonate concretions have been described from the fluvial sections and have been dated to around

55-40 ka BP (Nanson et al., 1991; in review).

A more hospitable aquatic environment is implicated through unit 3

(5.68-3.73 m). The sediment of sub unit 3b (5.6-4.8 m) is largely dark grey in colour with some orange mottling, representing cracks in swelling and shrinking clays. The replacement in the coarse fraction of quartz by ostracods indicates an established waterbody and decreased energy within the environment. Renewed fluvial activity or a brief marine incursion is implicated. The vast number of ostracod valves present, predominantly Cyprideis, suggests brackish conditions.

The abundance of chalky, broken shell material and evidence of dissolution of many of the ostracod valves further implicates the later influence of meteoric water permeating the sediment. This may indicate a seasonal drying or hiatus in sedimentation and pedogenesis, subsequent to the deposition of this subunit. A date of 72±11 ka BP has been established via AAR methods from a depth of

5.00-5.02 m (Chivas et al., 2001). This date may represent the time deposition of this unit, however it was obtained from a disarticulated Bassina mollusc, hence further verification of the age of the sediment may be necessary. As it stands, the date is broadly contemporaneous with that of the underlying unit, hence suggests rapid deposition.

Subunit 3a (4.7-3.8 m) is only subtly differentiable from unit 3b. Although there is a change in the colour of the sediment from 5.05 m, from dark grey to a lighter greenish grey, the shift in microfaunal assemblage, to estuarine species, is obvious at 4.7 m, accompanied by a slight shift in particle-size, with an increase in the clay fraction to comprise around 70% of the sediment. A change in water chemistry, to more marine composition waters in a lagoonal setting, is implicated. Abundant ostracods and Ammonia, which are generally well-preserved, and broken shell and fish fragments all support a productive environment.

In contrast to the lower subunit, orange staining of the fine fraction of the sediment is diffuse and the microfauna show no evidence of dissolution, implying that the iron-rich clays have been washed in, rather than formed by in situ pedogenic mottling. This suggests that the edges of the basin were exposed and finer material was transported to settle at this deeper region with the marine incursion. This sequence is comparable to the lowest obtained unit (V) of

Torgersen et al. (1985, 1988), dating prior to 36 ka BP (Fig. 4.3). These authors identified the influence of marine waters, followed by mild pedogenesis.

Deposition during the sea-level highstand of marine isotope stage 3 (59-30 ka BP,

Martinson et al., 1987) is supported. The sea level at this stage was approximately that of the modern sill height, between -44 and -55 m bpsl

(e.g. Waelbroeck et al., 2002) (Fig. 2.1). The subaerial weathering subsequent to the deposition of this unit suggests that channelling across the Arafura Sill was active, draining the basin. Jones and Torgersen (1988) identify a channel at -75 m bpsl to be active at this time.

4. Sediment Analyses

Established lacustrine conditions characterise unit 2 (3.73-0.38 m). Subunit 2c

(3.7-2.8 m) is distinguished from the unit below by the absence of iron-oxide staining. The sediment is also noticeably more fine-grained; this may in part be attributable to the decrease in abundance of fauna present. The return of the 'Lake

Carpentaria' facies and the lack of iron-oxide both suggest more stable and extensive lacustrine conditions at this time. The increase in pyrite and the dark colour of the sediment are indicative of a low-energy environment, with a high organic content. This may be compared with unit II of Torgersen et al. (1985,

1988) (Fig. 4.3). The irregular light grey laminae at 3.7-3.55 m and again around

2.5 m (subunit 2b), although not observable in the microscopic samples, are comparable to unit III of the Torgersen et al. cores (Fig. 4.3). These were found to be composed of authigenic calcite crystals (20 µm or less), preserved in anoxic sediments and represent deposition in an environment of fluctuating lake levels

(Torgersen et al., 1985, 1988).

Under the microscope, the overlying unit 2b (2.7-1.1 m) shows little difference to unit 2c, with the same microfaunal assemblage, although more abundant, and little quartz present. The slight change in sediment colour, from dark grey to greenish grey towards the top of the unit, may be due to a decrease in organic matter or a change in the lake water chemistry. The absence of pyrite in these samples supports this. The samples between 2.2 m and 1.7 m, which include the presence of some reworked marine foraminifers and small flakes and fragments of ferruginous material suggest a brief contraction of the lake, exposing the margins.

The abundance and preservation of the non-marine material in these samples remains high, implying that this region of the lake was little affected by this

restriction. The increase in grain-size toward the top of this unit is most likely attributable to the rich microfauna, especially small bivalves and planospiral molluscs, in these samples.

The high abundance of well-preserved shell material from 1.6-1.4 m suggests a productive environment. A series of dates have been determined from this shell layer. A date of 18.40 cal. ka BP was obtained from a freshwater planospiral mollusc fragment at 1.50-1.51 m by the 14C AMS method. Three further 14C

AMS dates obtained from molluscs at 1.45-1.46 m, 1.20-1.21 m and 1.00-1.01 m are considered to be contemporaneous within the given error margins, at around

17.18 cal. ka BP. This indicates very rapid deposition of this shell-rich unit, under shallow water conditions. A comparable shell-rich layer has been described from the Bonaparte Gulf cores of Yokoyama et al. (2000, 2001), De Deckker and

Yokoyama (in review). The authors interpret this layer to have been rapidly deposited under shallow water conditions following the LGM lowstand. This unit from the Carpentaria cores may similarly be indicative of shallow water, but implicates an increase in meteoric water, owing to the occurrence of freshwater forms.

The sticky, dark grey crumbly clays that characterise unit 2a (1.0-0.4 m) exhibit an increasing diversity of freshwater ostracods, charophytes and molluscs.

Fragments of organic material are commonly preserved in the sediment, adding to its dark colouring. Within the sample at 0.9 m, the presence of planktic foraminifers implies some marine influence, although this is unlikely given the numerous non-marine ostracods present. Aeolian transport is again suggested,

4. Sediment Analyses

perhaps with increased dust storm activity, and may explain the number of broken valves in some of the samples in this unit. The sugary nature of some of the valves and presence of gypsum crystals also suggest a change in the chemistry of the lake waters at this time, implicating increased seasonality. The abundance of pyrite above 0.7 m may be due to preservation of anoxic conditions or an increase in organic matter. Two 14C AMS dates of 17.25 cal. ka BP and 14.63 cal. ka BP were obtained from mollusc samples at 0.75-0.76 m and 0.70-0.71 m respectively.

The former is most likely a relict mollusc reworked from the underlying shell layer. A more prominent influence of marine material is implicated at 0.5 m by the increase in coarse fraction material, quartz and shell fragments, with rare miliolid foraminifers. Well-preserved non-marine ostracods are still dominant.

Noted as pale green ooze in the core sediment from 0.38 m, unit 1 (0.38-0 m) exhibits a clearly marine assemblage present to the core top, with a wide diversity and abundance of microfaunal material. Although the initial samples contain a large amount of broken material, the preservation and size of the microfaunal samples increases in the top 10 cm of the core. Dates of 13.38 cal. ka BP and

12.35 cal. ka BP (14C AMS) were obtained from a non-marine mollusc at

0.40-0.41 m and a marine mollusc at 0.35-0.36 m respectively. This places the initial marine transgression around 12.4 cal. ka BP in this region of the gulf.

Three further 14C AMS dates were determined for the uppermost unit; 1.32 cal. ka

BP, 10.40 cal. ka BP and 10.44 cal. ka BP from 0.2 m, 0.1 m and the core top sample respectively. As the molluscs used for these dates were well-preserved, it is suggested that the material used to obtain the two older dates might have been displaced from underlying soft sediment during the penetration of the coring

device. This transition from the underlying lacustrine material to this uppermost unit has been dated from the Torgersen cores to around 9.8-7.2 ka BP (Torgersen et al., 1988), which is broadly comparable to the modern dates, when calibrated3.

Holocene sedimentation is largely absent from core MD-32. This is may be due to the failure of recovery of the uppermost sediments. It should also be noted that sedimentation rates in this region of the gulf at present are very low, hence the

Holocene may be poorly represented. In addition, cyclones are common features of the modern environment, reworking the uppermost sedimentary unit.

4.9 Comparison with other cores

Utilising the unit key developed for core MD-32, comparisons of contemporaneous deposition in the other MD cores are outlined below (Fig. 4.9).

The units represent the lateral equivalent of those defined for MD-32, hence may be represented by different sedimentological components.

3 It should be noted that these dates were obtained using standard (i.e. non-AMS) 14C methods on bulk samples weighing in the order of 5-10 g, hence the accuracy is not expected to be as precise as modern techniques permit.

4. Sediment Analyses

4.9.1 Sediments of core MD-33

Core MD-33 (6.58 m; Fig. 4.4) was extracted from the deepest part of the modern gulf (12o23.55'S, 140o20.32'E), at a water depth of 68 m. The basal unit (6.5-6.4 m) comprises hard, dry clayey-silt, dark ochre in colour. The coarse fraction samples contain abundant iron-oxide-stained fine-grained quartz. The quartz is generally well-rounded and well-sorted, suggesting extensive transportation.

Most of the fine-fraction material is iron-oxide-rich and was likely to have been transported in a fluvial episode from surrounding exposed margins of the basin, formed during an extended period of drier conditions. Alternatively, this may represent in situ pedogenesis, in a no-lake environment and may explain the cessation of the penetration of the core at this depth.

The sediment from 6.3 m to 4.7 m is grey in colour with some iron-oxide stained quartz grains, gypsum and large concretions. A cemented section extending from

5.8-5.7 m was extracted whole from the core. The carbonate concretions are considered to be authigenic, forming subaqueously due to an increase in the carbonate saturation of the water. This was caused by altered Eh/pH conditions, either following a dilution of the waterbody by an influx of water with subsequent evaporation and concentration, or decomposition of organic matter leading to increased pCO2 of the water.

Above the cemented section, silt is the dominant particle-size fraction, representing around 50-60% of the sediment. The coarse fraction represents around 15% of the sediment where smaller concretions are present and around 5% through the rest of the unit. Quartz remains abundant. Marine microfauna are

101

present within the samples above 5.5 m. Although there is a wide variety of fauna present, including planktic foraminifers and larger rotaliids and ostracods, all of the specimens have a sugary texture and many show minor abrasion, indicating transportation and recrystallisation. The fragile nature of some of the tests implies a low energy environment. It is suggested that this material was washed into the deepest part of the basin during a minor marine incursion, most likely via channel connection across the Arafura Sill. This sequence is considered contemporaneous with unit 4 of core MD-32. The higher degree of cementation and greater abundance of microfauna in core MD-33 are probably both due to this core being near the depocentre of the basin. This is also supported by the thinner sequence represented here, indicating a more established waterbody and lower sedimentation rates at this locality in comparison to the more marginal MD-32.

Iron-oxide staining is evident in the core from 4.6 m to 2.2 m on both the fine fraction and coating quartz grains. The microfaunal assemblage is less diverse although still implicating a marine influence through 4.6 m to 3.6 m. Neither pyrite nor concretions are present. Many of the microfossils have a bleached, rather than sugary appearance implying in situ deposition, with subsequent leaching by meteoric water. These samples represent a continuation of the underlying unit as an enclosed embayment, more distal from open marine influence. A standing body of water of essentially marine composition with later drying and mild pedogenesis is indicated.

Above 3.5 m the microfossils present include the Lake Carpentaria assemblage of

Ammonia and Cyprideis. Specimens are rare and very poorly preserved, most

4. Sediment Analyses

being bleached. The particle size of the sediment is finer, with the clay fraction dominant. Permanent brackish water in a non-marine environment at the time of deposition of this unit is indicated, with later drying and exposure. Numerous marginal marine fauna, some with evidence of reworking, replace the non-marine fauna from 2.5 m to 2.2 m indicating a renewed influx of marine waters. Quartz is rare in these samples and iron-oxide staining is less prominent, however carbonate concretions are present. This sequence is compatible with unit 3 of MD-32, suggesting that a shallow, non-marine waterbody extended across the deeper regions of the basin, before contraction of the lake followed by a short-lived marine incursion, diluting the water and precipitating calcium carbonate.

A shell layer occurs in the core at 2.1 m, with an uncalibrated 14C AMS date of

40±0.6 ka BP. In addition to numerous gastropod shells, calcareous rock fragments and large reworked rotaliids are present, commonly with green staining giving a green tinge to the sediment. Although not seen in core MD-32,

Torgersen et al. (1985, 1988) have described similar material in unit IV of their cores, dated to around 35-26 ka BP, which is broadly comparable to the date for

MD-33, given the difference in methodology (Fig. 4.3).

From above the shell layer to 0.25 m, the core sediment comprises dark grey silty clay. Although not obvious in the sieved samples, thin pale grey laminations alternate with the darker clays from 1.8 m to 1.6 m implying fluctuating lake levels. These are comparable to the authigenic calcite laminations identified as unit III in the Torgersen cores (Torgersen et al., 1985, 1988). The microfaunal assemblage consists of abundant Ammonia and Cyprideis. An increasing diversity

103

of freshwater ostracods is apparent above 1.4 m. Another shell layer, dominated by reworked bivalves, at 0.77-78 m has 14C AMS dates of 18.82 cal. ka BP and

17.45 cal. ka BP, taken from Cyprideis valves and Corbulidae molluscs, respectively. This is most likely equivalent to, but preceding, the shell layer at

1.5 m in core MD-32. A more dynamic environment is suggested for this core.

The top 30 cm of this unit consist of darker, sticky and crumbly clays, as in unit

2a of MD-32. The diminished extent of the non-marine units in MD-33, in comparison to core MD-32, is probably due to decreased sedimentation in the deepest part of the lake. A mollusc at 0.3 m has been dated to 410 cal. a BP (14C

AMS). This young date suggests that the mollusc from which it was obtained was not in situ and may represent a burrowing form. Alternatively the upper sediment from the lacustrine unit may have been reworked with some of the incoming marine material.

Reworked marine material, including diverse foraminifers and broken echinoid spines, is present in the core from 0.45 m. However, the full marine transgression, as recorded in core MD-32, is obvious in core MD-33 above 0.3 m.

Open shallow marine fauna are abundant in the upper samples of the core. Dates of 290 cal. a BP and 480 cal. a BP a from 0.2-0.21 m, 790 cal. a BP from

0.1-0.11 m and 2.19 cal. ka BP and 2.57±0.05 ka from the core top sample were obtained via 14C AMS methods. This reversal indicates some kind of reworking, perhaps by cyclone action, of the uppermost samples.

Core MD-33 was retrieved from close proximity to core GC-2 (12o31'S,

140o21'E), collected by the Torgersen team (Torgersen et al., 1985). Although

4. Sediment Analyses

core GC-2 only extends to 2.2 m, there are clear correlations that extend between core GC-2 and the upper units of core MD-33 (Fig. 4.3). In addition to stratigraphy, more recent work has identified variations in the relative abundance of the sand- and silt-sized fractions of core GC-2 (De Deckker and Corrège, 1991;

De Deckker, 2001). These have been attributed to increased aeolian activity during a ~2600 yr cyclic pattern of aridity. Similar peaks at comparable depths have been identified from the particle-size analysis of core MD-33. The absence of these peaks from the other MD cores suggests that these signals have only been recorded in the deepest part of the lake, where sedimentation rates are lower and an increase in particle size during an arid phase is most obvious.

105

4. Sediment Analyses

107

4. Sediment Analyses

4.9.2 Sediments of core MD-31

Core MD-31 (13.61 m; Fig. 4.5) was retrieved from the central section of the gulf

(12o03.96'S, 138o44.98'E) at a water depth of 59 m. The basal section of the core

(13.6-13.4 m) comprises very dark grey, sticky sediment. The coarse fraction contains abundant well-sorted and rounded quartz and calcareous concretions, representative of a marginal mudflat environment, perhaps with fluvial input. The overlying sediment from 13.3 m to 12.0 m is composed of silty-clay. Here the coarse fraction is generally very small, containing minor mica, pyrite framboidal aggregates and organic matter with dwarfed Ammonia. Above 12.7 m planktic foraminifers are also present in the samples although, along with Ammonia, they have a sugary appearance. The microfauna are transported either by aeolian or low energy fluvial means, rather than being deposited in situ and subsequently recrystallised. These samples are comparable with those of MD-32 unit 4, having been deposited in a mudflat environment with periodic flooding and evaporation.

Lighter green-grey sediment forms the matrix from 12.0 m to 10.7 m, with many fine dry silty laminae. The particle-size analysis of this unit shows coarsening upwards through the sequence, with silt becoming the dominant fraction. The samples of this unit are highly variable in size and contain rare quartz. Whilst many of the finer-grained samples are virtually barren, the coarser samples contain abundant and diverse marine microfauna, fragments of fish, insects and shell, including pteropods from 11.8 m. This is indicative of a fluctuating marginal marine environment, most likely receiving pulses of terrigenous input.

109

A shell layer composed mainly of articulated bivalves occurs in the core between

10.68-10.57 m. This shell layer marks the cessation of the laminae. The overlying sediment is dark grey in colour and clay-dominant and the diversity of marine species present gradually decreases upward. The dominant fauna present include large rotaliid and miliolid foraminifers (showing evidence of reworking), broken shell material and echinoid spines, indicative of a moderate-energy marginal marine environment, perhaps tidally influenced.

The sequence is reminiscent of the fluctuating marine environment of unit 6 in

MD-32, although only the last two subunits (6a and b) appear to be represented here. The basal unit of MD-31 may thus represent the lateral extension of subunit

6c at the margins of the lagoon. True, open marine conditions are not apparent in this region of the core.

Above 8.9 m the sediment is smooth grey-green silty-clay with some calcareous concretions, a progressive upward increase in clay content, and only very small coarse fractions. The microfauna indicate a further restriction of marine waters, with rare estuarine foraminifers and ostracods and abundant pyrite. This unit may represent the lateral equivalent of the brackish lagoon (unit 5) of core MD-32.

From 7.8 m iron-oxide staining is evident in the fine fraction of the core sediment up to 1.9 m, and quartz dominates the coarse fraction of most samples through to

0.6 m. As such the interpretation of the sedimentary sequence and comparison with the facies of core MD-32 is difficult to define. The abundance of quartz in the samples may be due to this core being located near the margin of the lake.

4. Sediment Analyses

The accumulation of quartz may be due to either aeolian dune processes or fluvial activity. As the quartz is within a matrix of silty sediment and many of the grains have a glassy, subangular texture, an alluvial origin is favoured here. The bathymetry of the gulf also indicates the presence of a channel in this region, as outlined by the -60 m (bpsl) contour (Fig. 1.1).

Between 7.8 m and 7.1 m the coarse fraction is composed predominantly of quartz with some oxidised pyrite. Thin pale gypseous grey lenses are present in the core

(7.58-7.37 m) indicating periodic wetting and subsequent drying. The assemblage comprises sugary Ammonia and some planktic foraminifers that may have been transported some distance. Rare bleached whole ostracod carapaces occur. At

7.0 m the sample consists almost entirely of recalcified and cemented Ammonia tests. The cemented sample may represent a brief expansion, perhaps due to a minor marine incursion, and subsequent evaporation of the waterbody, such as through subunit 4d of core MD-32.

Quartz remains a prominent feature of the samples above 6.9 m along with iron-oxide-rich concretions and only rare microfauna, with those present showing extensive iron-oxide staining. The samples between 5.8 m and 4.8 m are barren of microfauna. The sequence may be correlated with core MD-32. The sediment may be considered contemporary with the mudflat environments of unit 4, without permanent water here in core MD-31, but periodic input of quartz-rich material, most likely indicating fluvial activity in the area. A period of non-deposition and pedogenesis may also be inferred.

111

Above 4.8 m, although the percentage of quartz in the coarse fractions of samples remains high, rare bleached microfauna enter the assemblage, predominantly

Ammonia, Elphididae and Cyprideis. Fragments of charophytes and planospiral molluscs also occur from 4.1 m. These taxa indicate proximity to permanent water at the time of deposition, but their poor state of preservation suggests transportation and pedogenic overprinting of the sediment. The presence of small carbonate concretions within the samples from 4.5 m further supports this. This material in MD-31 may represent an outflow channel from the depocentre of the basin at the time of subunit 3b of core MD-32. From 4.1 m reworked rotaliids also join the assemblage. These appear to have travelled some distance, rather than being deposited in situ and may have been reworked from the surrounding sediment.

The same basic assemblage continues through to 1.9 m, however the latter samples have a sugary texture and the microfossils present are bleached and chalky in appearance. These non-marine samples with evidence of subaerial exposure may be equivalent to unit 3b of MD-32. The unit is more thickly represented here, owing to the abundance of quartz and the more marginal proximity of the environment. Two uncalibrated 14C AMS dates of 46.0±2 ka BP and 45.8±1.7 ka BP have been obtained from this shell material at 3.20-3.21 m and 3.30-3.31 m respectively. These dates are at the upper limit of reliable radiocarbon capabilities, so may in fact represent minimum ages. A further

14C AMS date of 40.0±0.5 ka BP taken from a mollusc at 2.10-2.11 m, near the top of unit, supports the former dates and indicates relatively rapid sedimentation

4. Sediment Analyses

of this unit. These shells however, may have been reworked from previous shoreline deposits.

The iron-oxide staining of the sediment ceases at 1.9 m, the sediment being predominantly dark grey silty clay. Between 1.9 m and 1.4 m large rotaliids and small concretions are common in the samples with some heavily reworked planospiral molluscs and Cyprideis. Quartz remains a dominant feature. This unit most likely represents a lateral extension of the estuarine facies of unit 3a in core

MD-32 and the marine incursion between 2.5 m and 2.2 m in MD-33. The reworked non-marine material and abundant quartz again supports this as being a marginal marine environment and suggests that this minor marine incursion was concentrated in the deeper parts of the basin. The lack of iron-oxide staining in this sediment supports the continual presence of water in this region.

A reversal of the above assemblage is present in the samples from 1.30 m to

0.60 m, with non-marine material showing good preservation and the rotaliids being reworked. The sediment is essentially fine-grained with coarse-fraction

'peaks' in the particle-size curve corresponding to concretions, broken shell material and abundant quartz. There is no evidence of pedogenesis, however the reworked rotaliids and the quartz imply an energetic environment, most likely near the lake margin, or influenced by channels entering the lake. This is comparable to the unit IIB described by Torgersen et al. (1985, 1988), which the authors found in the cores taken from around the margins of the lake. They interpret this to represent either an ancient shoreline or reworked deposit. This unit represents the maximum extent of Lake Carpentaria and is the lateral

113

equivalent of unit 2 of MD-32. Two 14C AMS dates of 12.24 cal. ka BP and

12.25 cal. ka BP were obtained from non-marine molluscs within shell layers at

0.70-0.71 m and 0.65-0.66 m respectively. There is some evidence of minor reworked marine material, including echinoid spines, ooids and some marine foraminifers in these shell layers. They may represent concentration deposits near the lake margin at this time.

The transition to greenish-grey sediment, marking the most recent marine transgression (unit 1), is clearly delineated in core MD-31 on a diagonal line extending from 0.68-0.59 m. The first well-preserved marine material is evidenced in the samples at 0.6 m, with reworked non-marine material. Broken shell, echinoid spines, quartz, glauconite and ooids are also present. Above this transitional material the percentage of quartz decreases within the sediment, concretions are absent and the marine microfossil assemblage increases in abundance and diversity, including planktic forms. Several 14C AMS dates have been obtained from this unit: both 7.37 cal. ka BP and 1.41 cal. ka BP from

0.55-0.56 m, 1.84 cal. ka BP from 0.30-0.31 m, 10.85 cal. ka BP from

0.10-0.11 m, 3.62 cal. ka BP from 0.05-0.06 m and 7.32 cal. ka BP and 7.79 cal. ka BP from core top samples. These dates indicate a disturbance of the soupy sediment caused either during the retrieval of the core or overturning during cyclone activity. The shell from which the younger date at 0.55-0.56 m was obtained showed some evidence of recalcification and that at 0.1 m suggested possible reworking, most likely from the shell layer at 0.7 m. As such it is difficult to ascertain the age of the first marine contact evidenced in this core and estimate the sedimentation rate.

4. Sediment Analyses

115

4. Sediment Analyses

4.9.3 Sediments of core MD-30

Core MD-30 (8.23 m; Fig. 4.6) was extracted from around 30 km south of core

MD-31 and west of MD-32 (12o16.01’S, 138o44.92’E) at a water depth of 60 m.

Some clear comparisons can be made between this core and MD-31. The basal units of the core comprise dark grey clays with scattered shell matter. The lowermost 20 cm of the core is a dark grey, silty-clay with a very sticky consistency. The microfaunal samples are dominated by estuarine and some reworked shallow marine foraminiferal and ostracodal species with broken bivalve shells and echinoid spines. Pyrite framboidal aggregates are common in most samples and quartz is rare. A layer of bleached bivalved mollusc shells overlies a thin layer of small concretions between 7.7 m and 7.5 m. Above the shell layer the microfaunal assemblage becomes gradually more restricted. An orange band of hard, coarse, dry sediment with small concretions occurs at

7.15-7.04 m. The isolation of this band and the lack of evidence of pedogenesis below suggest that this unit was blown or washed in from surrounding exposed margins. The fine dark grey, clay-rich sediment recommences from 7.0 m to

5.7 m. Although not obvious in the core material, the sieved coarse fraction reveals some oxidised pyrite and fragments of iron-oxide-rich clays, although the fine fraction shows no iron-oxide staining. The estuarine fauna are well- preserved, but some larger rotaliids are reworked. These samples resemble a condensed sequence of the lowermost unit in core MD-31 (i.e. subunit 6b and 6a) however the true open marine samples are absent from this core, suggesting that core MD-30 is more marginal at this stage. The shell layer at 7.60 m most likely corresponds to that in core MD-31 at 10.57 m and core MD-32 at 10.10 m,

117

representing a retreating concentration deposit. The regression of marine conditions and restriction to the deeper part of the basin are evident in the presence of iron-oxide in this core, indicating exposed margins at this time.

Between 5.7 m and 3.5 m (unit 4), the sediment shows significant iron-oxide staining and abundant quartz. The mean particle-size remains fine, gradually coarsening upwards through the unit. The lower samples of this unit contain large rotaliids, showing very poor preservation and abundant broken and bleached shell material. The quartz present is of a uniform size, but ranges from subrounded to subangular, many grains showing a frosty, pitted texture indicating extensive reworking either by wind or water. Above 4.85 m pale grey gypseous lenses and laminae are common. This region of the core is largely devoid of microfauna, excepting a few isolated Ammonia and other small marine foraminiferal tests, which are sugary in appearance. Periodic flooding, evaporation and exposure with possible aeolian activity are implicated in this unit. The equivalent samples in core

MD-31 occur between 7.8 m and 4.8 m, although the inputs to MD-31 are both more extreme and erratic.

The sediment extending from 3.5 m to 2.9 m is darker greenish-grey in colour.

Black rootlets in the core sections indicate increased organic matter and bioturbation. The degree of iron-oxide staining is less pronounced in this section of the core, suggesting permanent water and a more extensive waterbody. From

2.9 m to 2.0 m the sediment is increasingly yellow, dry and crumbly, many with large concretions, particularly at 2.9 m to 2.7 m. The percentage of clay is larger, although the concretions skew the mean particle-size curve. XRD-analysis of

4. Sediment Analyses

concretions from 2.9 m indicates these to be barite, which is commonly associated with organic matter. Calcite concretions were identified higher up in the core at

2.4 m. Fine-grained clear quartz is abundant in the coarse fraction. Rare microfauna present include the Lake Carpentaria facies, but with very poor preservation. The equivalent marginal lacustrine environment, with evaporation, subaerial exposure and subsequent pedogenesis, is seen in core MD-31 from

4.8 m to 1.9 m and may be considered the lateral equivalent of subunit 3b of

MD-32. A depositional hiatus subsequent to this is postulated.

From 2.0 m to 0.7 m, the sediment is represented by dark grey smooth silty-clay, coarsening-up through the sequence, with scattered concretions and shell material.

The absence of iron-oxide staining indicates continuous water coverage and deposition. Like MD-31, quartz remains a significant component of the coarse fraction of samples and numerous large rotaliids of varying preservation levels, most likely transported, occur in the lower samples. Two dates have been obtained at 1.5 m and 0.9 m, of 13.94 cal. ka BP and 12.76 cal. ka BP respectively, suggesting rapid deposition. Non-marine microfauna are present in the samples, but are mostly reworked up to 1.05 m. This unit is comparable to the marginal unit IIB of Torgersen et al. (1985, 1988) as also seen in core MD-31.

From this point in the core, the samples show much better preservation of the lacustrine material, including charophytes and gastropods, and some reworked miliolids. These samples represent the duration of the maximum extent of Lake

Carpentaria represented in this core, equivalent to subunit 2a of MD-32.

Continued channel influence, such as reworked marine material, is reflected,

119

suggesting this core was marginal to the main lake. A shell layer, centred about

0.75 m has returned 14C dates of 12.46 cal. ka BP and 12.39 cal. ka BP from

0.80-0.81 m and 0.70-0.71 m, also suggesting rapid deposition rates at this time.

The date of the shell layer is consistent with a similar feature of core MD-31.

A gradational transition zone (0.71-0.67 m) is observable in the core slices between the dark grey clays and overlying soft pale green sediment with abundant broken shell, reworked lacustrine material and quartz, with marine material such as foraminifers, broken echinoid spines, quartz, glauconite and ooids. The marine assemblage, as in unit 1 of core MD-32, appears to have been rapidly established in this core. This may be due to the proximity of the cores to channels that may have crossed the Arafura Sill. Several 14C AMS dates have been obtained from this uppermost unit; 4.36 cal. ka BP from 0.65-0.66 cm, 10.53 cal. ka BP from

0.6-0.61 m, 5.10 cal. ka BP from 0.30-0.31 m, 1.16 cal. ka BP from 0.05-0.06 m and a core top date of 8.73 cal. ka BP. Again there appears to be some disturbance to the uppermost samples, rendering sedimentation rates and conditions for the uppermost unit difficult to establish.

4. Sediment Analyses

121

4. Sediment Analyses

4.9.4 Sediments of core MD-29

Core MD-29 (6.24 m; Fig. 4.7) was extracted from the northwest section of the gulf (10o47.36’S, 138o43.20’E) at a water depth of 60 m. Although the sediment is largely silty-clay throughout, there are discrete peaks in the coarse fractions.

Most of the core shows some degree of iron-oxide staining, with extensive quartz and rare microfauna in the coarse fraction samples. The basal 20 cm of the core comprise dark grey sticky clays. Although not observed when cutting the core, upon sieving the fine fraction of these samples was seen to consist of iron-oxide-rich clays. Minor quartz is also present. The contained microfauna include dwarfed Ammonia, and planktic foraminifers with a sugary appearance, such as those in unit 4c in MD-32. A comparable depositional environment of ephemeral mudflats is inferred for these samples.

From 6.0 m to 4.5 m a greenish-grey highly consolidated silty-clay with extensive iron-oxide staining in the sediment matrix. Very fine-grained, clear white-grey thin gypseous lenses, forming incomplete laminae, indicating periodic inflow and evaporation, occur between 5.8 m and 5.0 m. In addition, evidence of organic-rich matter including rootlets, are observed in this region of the core. In the coarse fraction these samples contain abundant quartz, most of which is pitted and frosty. Peaks in the mean particle-size curve correspond to the increase in quartz. Rare Ammonia are the only microfossils present up to 4.5 m and are dwarfed and sugary. A mudflat environment with episodic inflow is suggested and may be contemporaneous with unit 4 of MD-32.

123

Between 4.5 m and 3.5 m the sediment is a darker grey colour, with orange mottling. The microfaunal assemblage expands to include non-marine ostracods, although most of the valves are bleached. More established lacustrine conditions are interpreted, with later subaerial exposure and pedogenesis. This is a more marginal equivalent of subunit 3b of MD-32, with more coarse material being transported into the region of core MD-29, comparable to MD-30 and MD-31.

From 3.5 m to 2.3 m the sediment comprises greenish grey grading to dark grey silty-clay with occasional calcite concretions, but no evidence of iron-oxide staining. The lowermost samples contain microfauna indicative of a freshwater environment. Quartz is present and shell material is reworked with a pinkish tinge; it is particularly abundant around 3.0 m. Fragments of charophytes and non-marine molluscs are common. In the upper samples, quartz occurs in greater abundance and Ammonia represents the only faunal material, once again reworked and commonly with overgrowths of calcium carbonate. Concretions increase in abundance toward the top of the unit. This sediment represents the establishment of lacustrine conditions proximal to the core and may be the lateral equivalent of unit 2c of MD-32. Dating of this unit in core MD-29 is necessary to confirm this.

The position of core MD-29 may identify the maximum extent of the lake at this time, although it was most likely marginal at this stage, perhaps connected via an outflow channel. Within this core there is no evidence of the marine influence indicated in unit 3a of MD-32 and observed in cores MD-31 and -30. This may have been a more local effect, caused by a minor marine incursion through a channel in that vicinity.

4. Sediment Analyses

Evidence of iron-oxide staining returns to the core above 2.3 m, first as spots within the core slices and then as incomplete laminae (2.2-2.0 m). Carbonate concretions are common in the lower samples of this unit. The dark grey-green clays extend to 0.86 m with yellow-orange mottling, representing in situ pedogenesis. Numerous fine red and light grey silty laminae are present in the core between 1.4-1.1 m. Quartz is abundant throughout these samples and microfossils are rare. Pyrite occurs in some of the samples, although it is oxidised. Ammonia is common and occurs in different sizes and as well-preserved tests, however the samples between 1.4 m and 0.9 m are barren of microfauna.

This unit represents a marginal non-marine environment without a permanent water supply and the beginnings of the development of a soil profile, implicating a contraction of the lake at this time, as suggested by the equivalent unit 2b in

MD-32.

More substantial lacustrine conditions are apparent from a thin, dark grey sticky clay unit from 0.86 m to 0.45 m, with the best-preserved non-marine microfauna at the base of the unit. Samples from the upper half of the unit comprise abundant quartz, reworked Ammonia and calcite concretions, indicative of a non-marine environment, shallowing upwards. This represents the maximum extent of the most recent Lake Carpentaria phase in core MD-29, and may be correlated with subunit 2a in MD-32. The latter lacustral expansion in core MD-29 may have been short-lived.

Reworked marine fauna, including broken shell material, appear in the assemblage from 0.6 m. Although the sediment change to the familiar pale green

125

ooze occurs at 0.45 m, full modern marine conditions are not apparent below

0.2 m. Above this, foraminifers and ostracods are diverse and abundant, quartz is common and glauconite, echinoid spines and bivalves are present. Dates of

10.52 cal. ka BP from 0.20 m and 410 cal. a BP from a core top sample have been obtained by 14C AMS means. The temporal gap in these dates may represent a hiatus in marine deposition after the initial transgression, or may indicate disturbance of uppermost samples.

4. Sediment Analyses

127

4. Sediment Analyses

4.9.5 Sediments of core MD-28

Core MD-28 (4.19 m; Fig. 4.8) was extracted from the northern region of the gulf

(11o11.48’S, 139o57.53’E) at a water depth of 62 m; around 300 km north of core

MD-33. Initial observation of the core reveals the most extensive iron-oxide staining of all of the cores collected, throughout all but the top 60 cm and the basal 10 cm. This core is also notable in that the average particle size is much coarser than in the other cores investigated. The sediment is dominated by silt, with a significant percentage of very fine sand in the lower half of the core, with the clay and very fine sand each representing as much as 40% in the top 2 m of the core.

The lowermost samples contain abundant mica including biotite, muscovite and chlorite, small rock fragments and angular quartz grains. Irregular laminae of alternating grey and dark yellow-green clayey silt are present from 4.05 m to

3.71 m. The overlying sediment is greenish-grey sandy-silt with extensive iron-oxide staining. Finer-grained pale lenses and scattered concretions are common from 2.9 m to 1.9 m. The coarse fractions within these samples contain an upward decreasing percentage of mica and increased quartz. The quartz shows poor sorting compared with that seen in other cores. Some larger broken fragments of shelly material occur within the upper samples of this unit and represent the first evidence of fauna within this core. This sequence resembles a prograding delta package, with the basal samples being most proximal to the source material. Rapid deposition is implied. A comparable unit may be that of the base of core MD-32, unit 7, although the two are not considered to be contemporaneous.

129

From 1.9 m the clay fraction increases to form alternating silty-clay and clayey-silt units with a significant fine sand-sized component to 1.05 m. The sediment is a darker greenish-grey colour with well developed mottling to 0.6 m.

Throughout this section of the core quartz remains a dominant feature of the coarse fraction. The samples above 1.7 m comprise largely fine-grained, very clear quartz with better sorting. White carbonate concretions are also common.

The Lake Carpentaria facies is present in samples from 1.7 m, although most samples contain poorly preserved, bleached fauna, indicating an extension of the lake to this region, with later exposure. These sediments may be considered the lateral equivalent to subunit 2c of core MD-32 and represent establishment of non-marine conditions, although marginal to the main lake.

Samples from 1.15 m contain reworked Ammonia and other rotaliids, indicative of a contraction of the lake and increased reworking of marine material, such as seen in subunit 2b of MD-32. This unit is most closely related to the quartz-dominated sediment of MD-30 and -31 and unit IIB of Torgersen et al. (1985, 1988). A shell layer, dominated by small, articulated bivalves, occurs at the top of this unit. A series of 14C AMS dates have been obtained from this material: 17.12 cal. ka BP,

17.90 cal. ka BP and 17.20 cal. ka BP from 0.78-0.77 m, 0.75-0.76 m and

0.70-0.71 m respectively, contemporaneous with the shell layer of subunit 2b in core MD-32. The sediment directly above this unit (0.65-0.56 m) is dark grey crumbly silty-clay and includes abundant Ammonia and several reworked marine foraminifers, although no ostracods are preserved. This may be equivalent to the influence of marine material at 0.9 m in core MD-32.

4. Sediment Analyses

The pale grey marine sediment is first apparent at 0.56 m, although the full modern marine assemblage is not present until the top 0.3 m of the core. The intervening samples display a gradual mixing of the two marine and non-marine end-members, with a large percentage of reworked microfaunal material and broken shell. Quartz is still a dominant component in these samples. Dates from this initial marine material have been obtained via 14C AMS of 10.13 cal. ka BP at

0.50-0.51 m and 10.66 cal. ka BP at 0.35-0.36 m, which indicate penecontemporaneous deposition.

The uppermost 0.3 m comprise well-preserved marine fauna, which are both diverse and abundant, some ooids, glauconite and minor quartz. 14C AMS dates of 2.64 cal. ka BP and 2.21 cal. ka BP have been established from mollusc and ostracods samples respectively, at 0.15-0.16 m and a further date of 350 cal. a BP was recovered from the core top sample. This succession seems to represent the most intact recent marine section of all of the cores and gives the best indication of the impact of initially the Arafura Sill and later Torres Strait being breached.

There does seem to be a hiatus between the deposition of the shell layer and the marine transgression. This may be due to exposure or decreased deposition during this interval.

131

4. Sediment Analyses

133

4. Sediment Analyses

4.10 Palaeoenvironments of the gulf, as observed through the core sediment

By identification and comparison of the sedimentological units within the cores, depositional facies and post-depositional affects may be inferred. As outlined above, the units were defined for core MD-32 and extrapolated to the other cores

(Fig. 4.9). The facies reconstructions are referenced to the Marine Isotope Stage

(MIS) chronology (Fig. 2.1). From this, a model of basin evolution, through the last glacial cycle, may be outlined.

The basal unit 7 (14.84-13.95 m) of core MD-32 appears to be unique to this core and may represent the oldest of all material extracted. The extensive quartz and iron-oxide material are indicative of dry surrounds, due to exposed and poorly vegetated gulf margins. The peaks in the coarser fraction sediment are indicative of periodic flooding events, associated with fluvial activity. The dating of this unit to around 125 ka BP suggests rapid deposition during the initial highstand of the Last Interglacial (substage 5e). Although there is no direct evidence of a marine incursion, rising sea levels are suggested by increased fluvial activity.

A marine environment is clearly evident through unit 6, best observed in core

MD-32 (13.9-9.4 m). At least three periods of relatively high sea level may be noted through this unit, separated by periods of a more restricted character. This sequence follows the sea-level fluctuations throughout Marine Isotope Stage

(MIS) 5 (128-74 ka BP).

Open shallow marine conditions were rapidly established in the vicinity of core

MD-32, identified as sub-unit 6f (13.9-13.1 m), following the transgression of the

135

Arafura Sill by seawater around 125 ka BP. The occurrence of in situ planktic microfauna suggests that the Torres Strait may also have been breached at this stage. A sea level similar to that of the gulf today is supported.

A restriction of marine conditions to a large embayment, open only to the Arafura

Sea is identified through unit 6e (core MD-32, 13.0-12.0 m). A regression of sea level to about 10 m above sill height, perhaps around -45 m bpsl, is suggested, in association with the sea-level regression of MIS 5d (116-103 ka BP). Fluvial activity in the vicinity is increased, which is supported by Nanson et al., 1991; in review) indicating that the Australian monsoon was most probably active at this stage.

A return to open shallow marine conditions, although more distal from fluvial sources, is apparent through subunit 6d (MD-32, 11.9-11.3 m), implicating sea-level rise. The lack of planktic forms suggest conditions were not restored to those witnessed in subunit 6f. A rise in sea level from that of the underlying subunit is supported, to around -20 m bpsl, as in the highstand of MIS 5c

(103-93 ka BP). Torres Strait is likely to have remained above water at this stage.

The lowstand of MIS 5b (93-85 ka BP) is represented by a lagoonal environment through subunit 6c, implicating a further regression of sea level to about the height of the Arafura Sill. Periodic marine inflow reached the deeper regions of the basin (MD-32, 11.2-10.7 m) whereas only margin mudflat deposition takes place around the current -60 m contour (MD-31, 13.6-12.0 m). This lowstand is even more restricted than that of subunit 6e, indicating either comparably lower

4. Sediment Analyses

sea level or a higher sill. There is no evidence of high-energy fluvial activity, suggesting conditions were drier during this time and that the Australian monsoon may not have been active.

Renewed influence of marine waters, associated with the MIS 5a (85-74 ka BP) highstand, is evidenced through subunit 6b, approaching a sea level similar to that of subunit 6d. Whereas open shallow marine taxa are present throughout this subunit in core MD-32 (10.6-10.2 m), pulses of terrigenous silts are intercalated with the marine sediments in core MD-31 (12.0-10.7 m), indicating episodic fluvial influence in this more marginal region. MD-30 (8.2-7.6 m) is also marginal, but the channel activity of MD-31 is not seen here. Sedimentation rates are considerably higher in these more marginal regions. There is no definitive evidence of a renewed contact across Torres Strait.

A shell layer, present in three cores (MD-32, 10.1-10.0 m; MD-31,

10.68-10.57 m; MD-30, 7.6-7.5 m) may be identified as a concentration deposit formed during marine regression. This feature has not been dated in any of the cores, but a similar depositional environment may be inferred. High-energy conditions, with ever decreasing marine influence, are evident above this layer

(subunit 6a MD-32, 10.1-9.4 m; MD-31, 10.7-9.0 m; MD-30, 7.6-5.8 m). The sea is considered to have retreated to below the level of the Arafura Sill during this time, associated with the regression during the MIS 5a/4 (around 74 ka BP) transition, but channel connection may have been maintained. There is also evidence of exposure around the margins of the basin, suggesting that the remaining waterbody had retreated to within the modern -60 m contour.

137

The brackish lagoon of unit 5 (9.3-9.0 m) is only clearly recorded in core MD-32, where a bloom in the 'Lake Carpentaria' facies identifies it. A separation from marine waters, retreating below the level of the sill, is inferred. There is some evidence of the occurrence of this unit in core MD-31 (8.9-7.9 m), marginal to the lagoon proper, perhaps without permanent water. Fluvial activity does not appear to be significant at this time. Again, this material has not been dated directly, but relatively rapid deposition during the marine regression of early MIS 4 (around

74-72 ka BP) is implicated.

Unit 4 is present in all of the cores to varying degrees (MD-32, 8.9-5.6 m;

MD-33, 6.5-3.6 m; MD-31, 7.8-4.1 m; MD-30, 5.7-3.5 m; MD-29, 6.2-4.5 m), excepting MD-28. The overall environment is representative of mudflats with cycles of inundation, evaporation and exposure. The abundance of quartz and the reworked microfaunal material in cores MD-30 and 31 implicate some channel influence through this unit. The absence of ostracods in core MD-32 in subunit

4d (8.9-8.1 m) and 4c (8.0-7.0 m) suggests the lack of a permanent waterbody. In support of this, iron-oxide is a common feature unit 4, with extensive mottling in core MD-31, -30 and -29, and more diffusely in MD-32 and -33. This indicates subaerial exposure at least around the basin margins and a fluctuating water table.

Short-lived basin-wide exposure may have occurred subsequent to the deposition of the unit. It should be noted that highly oxidising conditions are characteristic of this humid, seasonally dry region today.

4. Sediment Analyses

Briefly wetter conditions are evident through subunit 4b (MD-32, 6.9-6.7 m), with an established waterbody extending to at least the -63 m contour. A marine influence is evident throughout this subunit. Evidence of fluvial activity is present in subunit 4a (MD-32, 6.6-5.7 m), with abundant quartz and iron-oxide and carbonate concretions. The presence of the two types of concretions, at discrete intervals in the same subunit, suggests climatic variation or fluctuating groundwater (Nanson et al., 1991). This unit has been identified from the fluvial sequences of the Gilbert and Einasleigh Rivers, dating to around 55-40 ka BP

(Nanson et al., in review). The quartz-rich unit near the centre of the gulf may be older than this. Only one date has been established for this unit, being >64.7 ka

BP at 5.8 m in core MD-32, suggesting deposition during the marine lowstand of

MIS 4 (74-59 ka). This is supported by the generally drier conditions throughout the unit. An outflow channel through the Arafura Sill is interpreted to have been active at this time, thus draining the basin. Such a channel has been identified in a seismic profile taken across the channel, extending down to around -75 m bpsl

(Jones and Torgersen, 1988).

The establishment of a short-lived perennial waterbody is evident through unit 3, with a depositional hiatus between the two subunits. Unit 3b represents a brackish waterbody, extending to at least the -59 m bathymetric contour of the modern gulf

(MD-32, 5.6-4.8 m; MD-33, 3.5-2.6 m; MD-29, 4.5-3.5 m). The return of water to the basin may be due to renewed river activity and closure of the sill channel, temporarily filling the basin to this level around the time of increasing sea level during MIS 4/3 (around 60 ka). There is evidence of channel activity in the west

(cores MD-31, 4.1-2.0 m and MD-30, 3.5-2.1 m) and north (core MD-28,

139

4.2-1.8 m) of the basin. Dates have been obtained from shell material in core

MD-31 of ~46 ka BP at 3.3-3.2 m and 40 ka BP at 2.1 m. Contraction of the lake, with minor pedogenesis is then implicated in all of the cores, with perhaps the exception of MD-28.

Unit 3a represents a minor marine incursion identified in cores MD-32 (4.7-3.8 m) and MD-33 (2.5-2.2 m). This may be a local effect of channels in the vicinity of

MD-30 (2.0-1.7 m) and MD-31 (1.9-1.4 m), transporting marine material to the depocentre of the basin and associated with the higher sea level of MIS 3 (around

50 ka BP). Cores MD-32 and MD-33 indicate establishment of estuarine conditions at this time, indicating connection to the sea. Sea level around the height of the Arafura Sill is implicated at this time, infilling the -75 m channel.

Dry conditions at least surrounding the basin, perhaps with minor pedogenesis in the deeper regions following deposition are also implicated. Marine material is absent from the more northerly cores, MD-28 and MD-29.

This estuarine unit has been identified in the basal sediments of the Torgersen cores (V). These authors interpret this unit to have been subaerially exposed, subsequent to deposition. For this to be possible, Torgersen et al (1988) have suggested that the Fly River, Papua New Guinea, which was thought to have flowed into the Carpentaria Basin (Blake and Ollier, 1969, 1971), may have diluted marine waters in the basin to establish the estuarine conditions, diverting to its present course prior to drying of the basin. The authors also suggest that the

-75 m channel across the sill drained the basin, after sea level had dropped back below sill height, around 36 ka BP (Torgersen and Jones, 1988). The recent cores

4. Sediment Analyses

show that the estuarine conditions appear only to have been short-lived and established in the deepest parts of the basin, whilst the margins were exposed, supporting the existence of the outflow channel across the sill after the marine incursion. There is no evidence in the recent cores to support inflowing of the

Fly River.

The establishment of fluctuating lacustrine conditions, extending at least to the

-59 m modern bathymetric contour with exposed lake margins is evident through unit 2 (MD-32, 3.7-2.8; MD-33, 2.1-0.5; MD-31, 1.3-0.6 m; MD-30, 1.6-0.7 m;

MD-29, 3.5-1.4 m; MD-28, 1.9-0.6 m). A lake of this extent would have had a surface area greater than 100 000 km2 and a volume of almost 300 km3 requiring an evaporation/precipitation ratio around 57% of that of today (Table 2). There is no evidence of marine connection, indicating that sea level is below the sill height during the waning stages of MIS 3 and throughout MIS 2 (45-12 ka BP).

A shell-rich layer has been identified at the base of this unit in core MD-33

(2.1-2.0 m), dated to around 40 ka BP. This layer in the Torgersen cores, referred to as unit IV, was estimated to be dated around 35-26 ka BP (Torgersen et al.,

1988), which is keeping with the new dates. The closure of the basin to the

Arafura Sill, and infilling by active, low-energy rivers is implicated through this unit. The initial lacustrine material is concentrated in the deeper section of the basin, around the modern -63 m contour (MD-32, 3.7-2.7 m; MD-33, 2.0-1.6 m).

There is evidence of lacustrine material in core MD-29 (3.5-2.3 m) and MD-28

(1.7-1.2 m) that may be contemporaneous, however the poor preservation of material and abundant quartz suggest these localities to be marginal to the lake. A lake of this size (around 35 000km2 and 48 km3) would require approximately

141

43% of the effective precipitation witnessed in the present gulf (Table 2). The lake may have extended to the modern -59 m contour, before contracting and exposing these more marginal areas. Authigenic calcite laminae are evident in the deeper part of the basin (MD-32, 3.73-3.35 m and 2.77-2.55m; MD-33, 1.8-1.6 m), which

Torgersen et al. (1985, 1988) have tentatively dated to around 29-23 ka BP.

Brackish conditions within a confined lake are inferred, with periodic anoxia.

Table 2 Varying hydrology and extent of Lake Carpentaria (after Grim and Edgar, 1998; Bowler, 1986).

1 2 3 4 5 Contour SA Volume Av. Depth Ac/Al F (E/P)el %mod (m) (km2) (106m3) (m) E/P6 53.1 189750 918070 4.24 5.91 0.15 1.73 63 59 107630 292450 ~3 10.42 0.1 1.94 57 60 92930 227800 2.46 12.06 0.1 2.11 52 63 34630 48350 ~2 32.37 0.05 2.57 43 65 9690 10070 1.04 115.73 0.025 3.87 28

1 SA = surface area; estimates adapted from Grim and Edgar, 1998. 2 Av. Depth = average depth 3 Ac = area of catchment, Al = area of lake 4 F = run-off coefficient (from Bowler, 1986) 5 (E/P)el = F(Ac/Al-1)+1, which is a comparison the present evaporation/precipitation ratio with that required to maintain lake-full state at the given depth. 6 %modE/P is an estimate of the percentage of the modern evaporation/precipitation ratio of previous extents of the lake

A temporary contraction of the lake is witnessed in cores MD-32 (2.2-1.7 m) and is suggested to be around the time of the Last Glacial Maximum (23-19 cal. ka

BP), during which time the sea was at the lowest level (~-125 m) covered by these cores (Yokoyama et al., 2000, 2001). The absence of this unit from MD-30 and

MD-31 is thought to be due to localised channel influence in this region. Some fluvial activity must have continued through this period in order to maintain the lake. Cores MD-28, MD-30, MD-31 and MD-32 all give evidence of a minor

4. Sediment Analyses

influence of reworked marine material at this stage, most likely eroded from subaerially exposed sediment.

A feature common to cores MD-28 (0.78-0.70 m), MD-32 (1.6-1.3 m) and MD-33

(0.77-0.75 m) is a shell layer, dominated by articulated bivalved molluscs that returned dates of 17.9-17.0 cal. ka BP. A highly productive environment is suggested in each of these cores, associated with an influx of water and expansion of the lake.

The uppermost lacustrine subunit is present in each of the cores, to varying degrees. Within the deeper region of the lake, freshwater conditions are evidenced (MD-32, 1.1-0.5 m; MD-33, 0.6-0.3m). The more marginal cores contain abundant quartz and the Lake Carpentaria facies (MD-31, 1.3-0.6 m; MD-

30, 1.05-0.75; MD-29, 0.85-0.45; MD-28, 0.7-0.56 m). Whereas the central basin is considered to have held permanent water, the lake margins appear here to have fluctuated about the modern -59 m contour. The greater aerial extent of the lake through subunit 2a and the overall fresher conditions suggest an increase in the inflow to the lake, at least seasonally, and implicate more effective precipitation which may be associated with the most recent onset/expansion of the Australian monsoon. An outflow across the Arafura Sill is also considered to have been active at this stage, allowing more saline waters to be "flushed out". Jones and

Torgersen (1988) have identified a channel cutting down to -62 m across the sill with only a thin veneer of sedimentary infill.

143

Unit II of Torgersen et al. (1985, 1988) comprises a dark grey firm mud with abundant non-marine microfauna and corresponds to unit 2 described above, particularly the cores in the deeper part of the basin. The authors interpreted the development of a permanent, but fluctuating lake between -57 m and -63 m bpsl, commencing near 26 ka BP. Unit IIB, described by Torgersen et al. (1985, 1988), comprising quartz-rich sediment with reworked shell material and carbonate concretions, with varying degrees of subaerial exposure, may be better compared with the sediments described as unit 2 for cores MD-28, -29, -30 and -31, being marginal to the main lake.

The first clear evidence of marine material is most commonly as reworked shell hash and may be dated to around 12.2 cal. ka BP. This is identified in cores

MD-33 (0.45 m) and MD-31 (0.65 m) as a concentration deposit and core MD-32

(0.38 m), MD-33 (0.45 m), MD-29 (0.65 m) and MD-28 (0.56 m) as scattered shell material. The initial transgression of the Arafura Sill appears to have occurred as a relatively rapid event. In some cores, there is minor reworked marine material above this shell layer, which may either have been brought in from the surrounding margins, or represent earlier contact, perhaps via channel connection across the sill.

The uppermost unit, 1 (MD-32, 0.3 m, MD-33, 0.3 m; MD-31, 0.6 m; MD-30,

0.65 m; MD-29, 0.25 m; MD-28, 0.3 m), is perhaps the most distinctive in each of the cores, seen as pale green shelly ooze. The wet nature of this sediment has meant that the most recent material in several of the cores appears either to have been reworked, disturbed during coring, or may not have been fully recovered.

4. Sediment Analyses

As such a complete Holocene sequence is difficult to ascertain from this material.

Permanent connection to marine waters has been dated to around 10.5 cal. ka BP.

The establishment of true open marine conditions incorporating flooding of Torres

Strait, thus allowing circulation in the gulf of waters of the Indian and Pacific

Oceans, cannot as yet be constrained. A date of at least 8 cal. ka BP, if not older, seems most probable given the available evidence.

Torgersen et al. (1985, 1988) interpret the most recent marine transgression to be two-fold. They suggest a permanent connection with the sea across the Arafura

Sill, establishing an estuarine environment that may be responsible for the -53 m seismic shoreline, to have occurred by around 12 ka BP, as evidenced in the top of their unit II. They also consider their unit I to represent fully marine conditions, with Torres Strait being breached, around 8.5 ka BP (Torgersen et al., 1988). The evidence for an intermediary estuarine environment has not been clearly defined from the more recent cores, but there is some evidence for mixing of non-marine and marine material. The dates are broadly consistent between the two records.

The sedimentological interpretation of the gulf cores provides a framework of palaeoenvironmental change in the region. In particular, the extent of the lake basin, timing of marine influence and evidence of channel activity at various intervals through the last glacial cycle.

145

5. Ostracod Analyses

The use of Ostracoda in the palaeoenvironmental reconstruction of the gulf - facies analysis and morphological variation

Having established the broad context of environmental change in the Gulf of Carpentaria through the last glacial cycle, analysis of ostracod facies associations, with respect to ecology, allows further information to be drawn regarding the conditions in the waterbody at the time of shell formation. Observation of morphological variation of the valves enables inferences to be made about changes to the water chemistry through this period.

5.1 Introduction to ostracods

5.1.1 Overview of ostracods

Ostracods are small bivalved crustaceans, common in most aquatic environments.

They are an incredibly diverse group represented by somewhere in the order of 33

000 extant and fossil species of 4 500 genera and subgenera (E. Kempf, pers. comm. cited in Horne et al., 2002)4. Their occurrence in the fossil record extends from the Ordovician (around 500 Ma) (Martens, 1998). Ostracods are present in both marine and non-marine environments, ranging from the deep ocean to freshwater lakes and including temporary pools, springs, rivers, estuaries, swamps and even some semi-terrestrial habitats. In addition, their calcitic valves preserve a "snapshot' of the ambient water conditions at the time of shell formation.

Ostracods provide an eminently suitable proxy for the study of past environments as they are generally well-represented in most aquatic habitats, are well-preserved in the fossil record and are sensitive to a broad range of ecological variables, with

4 In excess of 65 000 living and fossil species of 5 000 genera and subgenera, including synonomies have been described in the "Cologne (now Kempf) Database of Ostracoda" (e.g. Kempf 1980, 1986).

5. Ostracod Analyses

most taxa living along gradients defined by these parameters.

Palaeoenvironmental reconstruction may be ascertained by a combination of techniques: a) extrapolating modern ecological constraints to fossil assemblages, b) morphological analysis of the carapaces with respect to physical and chemical parameters both during the organisms' life time and post-deposition, and c) chemical analysis of the ostracod valves.

5.1.2 Ontogeny

As adults, ostracods are usually less than 3 mm in length. The body of the animal, resembling a shrimp, is entirely encased within a bivalved carapace, hinged along the dorsal margin. These two valves are composed of low-magnesium calcite

(Kesling, 1951), formed from components taken directly from the host water

(Turpen and Angell, 1971). As with other arthropods, ostracods moult their carapaces with growth. There is some argument as to whether re-absorption of calcium of the moulted carapace is possible, particularly in Ca2+-depleted waters

(Peypouquet et al., 1988). The moult stages, known as instars, are numbered in descending order, with A being the adult stage, A-1 the penultimate, A-2, etc.

The ontogeny of most ostracods consists of eight to nine instars, comprising one adult and seven to eight juvenile stages. The first instar is hatched from the egg as a free-swimming nauplius larva, already with a carapace and three pairs of limbs.

With each successive stage, further complexity is developed. Growth occurs shortly after each moult and before the full calcification of the carapace. The reproductive organs are only present in the final, adult stage. No further growth or moulting occurs after sexual maturity has been reached.

147

The majority of ostracods reproduce sexually, although parthenogenesis occurs in some non-marine species (Horne, 1983; Horne et al., 1998). For most species, fully developed eggs are deposited within the sediment or on host-plants either singly or in clusters. Brood care of the first few instars is known for some taxa including Cyprideis and Xestoleberis. The life span of ostracods species varies from a few months to a couple of years. Longer life cycles are known from examples in colder climates, such as Darwinula stevensoni, which lives up to four years within cold-water lakes in Finland (Martens, 1998). The life cycle of ostracod taxa is also varied and depends largely on the organisms' preferred habitat and seasonality. Whereas some species produce a single generation per year in optimal conditions, others with a shorter lifespan may produce four or five generations over the warmer or wetter months. The eggs of some species can withstand desiccation and/or freezing, before hatching under more favourable conditions. In addition, the eggs are easily transportable by wind, flowing waters, current action, migratory birds, or within the guts of fish or birds, facilitating colonisation of newly formed waterbodies.

Most ostracod taxa are benthic-dwellers, living both on the surface and within sediment; however, some nektic and pelagic species are also known. The majority of ostracods are detritivores or herbivores with little feeding specialisation (Griffiths and Holmes, 2000). Larger ostracods in terminal lakes have been known to become predators, feeding on copepods and chironomid larvae (De Deckker, 2002). Predation of ostracods is most commonly by fish

(Mbahinzireki et al., 1991) and aquatic birds (De Deckker, 1977).

5. Ostracod Analyses

Figure 5.1. Valve morphology of a cytheroidean ostracod (Podocopida, Cytheroidea, Hemicytheridae), male (after Horne et al., 2002).

149

5.1.3 Valve structure and taxonomy

The class Ostracoda (Bowmann and Abele, 1982) is divided into two subclasses,

Myodocopa and Podocopa. All of the myodocopans are marine taxa; however, as many have poorly calcified valves, they are rarely represented in the fossil record.

The podocopans comprise the three orders, Platycopida (marine), Podocopina

(ubiquitous) and Palaeocopida (extremely rare marine, not found in this study).

The podocopids are by far the best-represented order in this study.

Shortly after death, the soft parts of an ostracod usually decompose, leaving only the calcareous carapace. The valves are usually the only parts preserved in the fossil record, hence taxonomic classification in this thesis is based on the structure of the valves present. Diagnostic characteristics of the ostracod carapace are shown in Fig. 5.1. The epidermis covering the body of the ostracod secretes the calcareous material that forms the carapace (Harding, 1964). Valves have both an inner and outer lamellae, however usually only the periphery of the inner lamella is calcified and thus preserved. The two valves that make up the carapace may be described as left and right. Commonly in Podocopa, one valve is larger and overlaps the smaller valve. The two valves are joined by a dorsal hinge structure that varies between taxa from simple to complex and may incorporate interlocking grooves and ridges or teeth and sockets. This hinge structure is an important diagnostic characteristic of ostracod families and genera in marine ostracods. The valves are opened and closed by adductor muscles that run through the central body of the ostracod. Distinctive scars are formed where the muscles attach to the inner side of the valves. Other minor scars include the frontal scars, associated with the mandibular muscles, and the mandibular scars, which are the points of

5. Ostracod Analyses

attachment of chitinous rods. The pattern of these scars may also be used for taxonomy at the suborder, family and subfamily levels.

The outer lamella is perforated by a series of so-called 'normal' pore canals, allowing contact between the soft parts of the ostracod and the surrounding outside environment. There are three basic types of pores: simple pores with protruding sensory hairs (sensilla), sieve pores with a mesh-like structure and sensilla, and exocrine pores without sensilla. The opening of the pore may occur either flush with the surface of the valve, recessed or at the end of protrusions. A second set of pore canals, referred to as marginal pore canals, extends from the line of concrescence where the inner and outer lamellae meet, through the marginal zone. The pore canals may be simple, branching, straight or curved.

The arrangement and type of pores present is a useful tool at the generic level of classification.

Ostracod valves vary widely in both overall shape and ornamentation. These features are used to distinguish ostracods at the genus and species levels. The overall shape of an ostracod carapace may be spheroidal or elongated, inflated or compressed, biconvex lensoidal or posteriorly inflated and laterally ovate, tabular or reniform. A more detailed view may reveal a caudal process (posterior projection), alae (wing-like projection) or sulcus (inward depression). Sexual dimorphism is pronounced in some species, generally with male carapaces being longer and females broader. Carapaces can be smooth or incorporate a variety of features including pitting, reticulations, spines, tubercles and nodes. Surface

151

ornamentation of some taxa may vary in expression in relation to environmental conditions.

5.2 Ostracods as palaeoenvironmental indicators

5.2.1 Ecology

A range of variables determines the distribution of ostracods. Physical factors include habitat type, stability and energy of the environment, water depth, temperature and turbidity, substrate type, nutrient level, macrophyte cover, competition and predation. Chemical parameters include salinity, solute composition, alkalinity and dissolved oxygen content of the host water. Ostracod species exhibit preference as well as a tolerance range for each of the above parameters. Broadly speaking, the diversity of ostracods is much greater in stable environments, such as the open sea, whereas more marginal environments are often characterised by a great abundance of just one or two species (Neale, 1988)

(Fig. 5.2). Generalisations may be made regarding the habitat preferences of common ostracod groups. Smith and Horne (2002) have provided an overview of ecological faunal associations; a brief discussion of taxa relevant to the present study is outlined below. The faunal associations discussed are grouped as marine, marginal marine and non-marine as represented in Fig. 5.3.

5. Ostracod Analyses

153

5. Ostracod Analyses

5.2.2 Marine environment

As previously mentioned, Myodocopa are predominantly marine taxa and poorly preserved in the fossil record. Although the subclass is present in all water depths

(above the calcite compensation depth – CCD) and exhibits a variety of life habits, the superfamily Cladocopoidea are nektobenthic ostracods. The deeper water forms occur on fine sand or muddy bottom sediment and some shallow water species have been found to live interstitially in coarser sediment (Smith and

Horne, 2002). Of the Podocopa, the Platycopida are exclusively marine-dwellers and the Podocopida are found in marine, marginal and non-marine environs.

Platycopids are benthic filter-feeders, found from shelf to outer estuarine environments and are most common in warm shallow carbonate-rich waters.

The Podocopids are ubiquitous, however habitat specialisation is discernable at the superfamily and family level. The Bairdoidea are only found in marine settings and commonly dominate the littoral zone; warm, shallow carbonate environments are preferred. They are benthic detritus feeders and poor swimmers.

Both Bairdiidae and Bythocytheridae are also represented in the shelf regions. Of the suborder Cypridocopina, both Pontocypridoidea and Cypridoidea are present in this study. The former are considered to be mostly marine or brackish in habitat preference whilst the latter are dominant in non-marine settings. All are essential benthic dwellers, although have the ability to swim.

The vast majority of ostracod taxa in marine environments today are of the superfamily Cytheroidea. All representatives are benthic crawlers or burrowers and unable to swim. Whatley (1988), in his review of coastline and continental

155

shelf ostracods, assigned familial associations to these settings. Families within phytal littoral substrates include Xestoleberididae, Paradoxostomidae and

Loxoconchidae, with many species adapted to living on plants. Sandy and muddy shores are characterised by Cytheruridae, Cytherideidae and Hemicytheridae, although these environs are not as productive as the phytal zone. The fauna of shelf settings are diverse and typically include the Trachyleberidae,

Pectocytheridae and Leptocytheridae, also the Hemicytheridae, Cytherideidae and

Cytheruridae. The inner shelf is differentiated from the outer shelf by the predominance of Loxoconchidae in the former and Krithidae in the latter.

Within the marine environment, ostracod distribution is largely determined by water depth, substrate type and the dynamics of the environment. These characteristics are directly reflected in the nature of the substrate on which (within which) ostracods live. The vast majority of marine ostracods live in the littoral and shallow zones. Common valve features include heavy ornamentation, complex hinge structures, eye tubercules or spots and branching marginal pore canals (van Morkhoven, 1962). Smoother shelled ostracods are found either in the deep ocean or live interstitially within sandy sediment, particularly those with tapering carapaces. Oblong or elongate forms are commonly benthic, whilst rounded forms are better adapted to swimming. Typically, shallow marine ostracods have short lifecycles, in the order of months to one year. Several species are known to produce more than one generation over the summer

(Horne et al., 2002)

5. Ostracod Analyses

5.2.3 Marginal marine environment

The majority of ostracods living in marginal marine environments are podocopids.

These may be divided into three groups based on their ecological affinities (after

Smith and Horne, 2002):

1. Taxa that are essentially marine in character, but may tolerate reduced

and/or fluctuating salinities include the Cytherocopina (e.g. Semicytherura

and Paradoxostoma) and Cypridocopina (e.g. Propontocypris). These are

commonly found in sandflat environments (Penney, 1987).

2. Non-marine taxa that can tolerate increased salinities, such as the

Darwinuloidea (e.g. Darwinula stevensoni) and some Cypridoidea. These

are more common in stable waters further from marine influence.

3. Taxa adapted to brackish waters, the cytheroideans Cyprideis,

Leptocythere, Loxoconcha and Venericythere. Each of these also includes

representatives in either the marine or non-marine realms. This is the most

dominant group and often achieves very high population densities in

brackish environments.

Within marginal environments the solute composition may also determine which ostracod species are present. Marine species are more likely to survive in Na+ and

Cl- dominated waters whereas non-marine ostracods are generally adapted to Ca2+

- 5 and HCO3 dominated water . Forester and Brouwers (1985) have noted that the co-occurrence of marine and non-marine ostracods is more common in marginal environments with warm and dry, rather than cool and wet climates. They attribute this to similar solute composition in marginal and non-marine environments under evaporative conditions.

5 The exception being in Australia, where most non-marine waters are Na+-Cl- composition. 157

Marginal marine settings may be defined here as extending from lowest tide level on the marine side to highest astronomical tide level at the landward extent and include those areas influenced by marine waters via salt spray and groundwater.

Salinity is the major control of ostracod distribution within marginal marine environments, varying from freshwater (<3‰), through brackish (3-30‰) to marine (30-40‰) and hyperhaline (>40‰) in evaporative environments (Neale,

1988). In addition to the range of salinity, the variability (stenohaline or euryhaline) and the solute composition need also be considered. Exposure at low tide represents a further stress in such settings. Although species diversity is generally low, the abundance within the marginal environment can be very great.

Those species that flourish are adapted to varying salinity, temperature and dissolved oxygen levels over a range of temporal scales. Common taxa such as

Cyprideis, Leptocythere and Loxoconcha exhibit a combination of hyper- and hypo-osmotic regulation to cope with such stresses (Aladin, 1993). Whilst burrowing, carapace closure and reduction in respiration rate can assist in withstanding short-term exposure at low tide, it is as yet unclear whether these taxa are capable of withstanding longer-term desiccation (Boomer and

Eisenhauer, 2002). The identification of marginal marine taxa has been utilised in numerous sea-level studies (e.g. Cronin, 1983; Carbonel and Hoibian, 1988;

Boomer, 1998).

5. Ostracod Analyses

5.2.4 Non-marine environment

Ostracods are commonly a dominant calcitic component of non-marine settings.

As such, they have been the focus of many detailed studies. An excellent summary of non-marine ostracods and their application to palaeoclimatic studies has been presented by Griffiths and Holmes (2000). All non-marine ostracods are podocopids. The majority of taxa are Cypridoidea, although there are some

Cytheroidean families, including Limnocytheridae and Cytherideidae that are present in non-marine environments. These ostracods are generally smooth- shelled with a simple hinge structure and no caudal process, eyespot or conspicuous branching marginal pore canals. Those that are rounded are often nektic, whilst elongate taxa are predominantly benthic. Both sexual and parthenogenic reproduction, or a combination of both, are known for non-marine ostracods.

Within lacustrine settings, the distribution of taxa occurs along gradients defined by water temperature, salinity, solute composition, nutrient availability, dissolved oxygen concentration and water energy. The seasonality of each of these variables must be considered. The local and regional hydrology of the environment also has an effect in terms of groundwater recharge or discharge, the ratio of precipitation to evaporation, water-rock interaction and whether the basin is hydrologically open or closed. In Australian freshwater lakes, the diversity of species increases with the stability of the environment, such as deeper or more permanent water, and the availability of particular ecological niches (De Deckker,

1982a).

159

Figure 5.4. Generalised relationship between ostracod species diversity and abundance with respect to calcite saturation of the host water (after De Deckker and Forester 1988).

Figure 5.5. A modification of the Eugster-Jones-Hardie model of solute evolution, specifically for Na-Cl dominant waters (after Radke et al., 2002).

5. Ostracod Analyses

The ionic concentration and composition of the host water are the most important characteristics in determining ostracod distribution in non-marine environments

(Smith and Horne, 2002). Most ostracods are sensitive to salinity and exhibit specific tolerance ranges (e.g. De Deckker, 1981b,c,d 1983; Neale, 1988; Radke et al., 2002). The number of species present is also partly controlled by salinity, with very low numbers in dilute waters, rising to the point where waters are saturated with respect to calcium carbonate, the calcite branchpoint (De Deckker and Forester, 1988). Although the diversity of species decreases beyond this point, the abundance is thought to increase, such that in hypersaline waters blooms of monospecific assemblages may occur (Fig. 5.4).

The calcite branchpoint refers to the concentration-level at which calcite precipitates within natural waters (Fig. 5.5). Mineral precipitation is the principal means of solute fractionation and evaporation is the principal cause. Beyond this point, inland waters under evaporative processes follow two major pathways of water chemistry evolution (Hardie and Eugster, 1970; Eugster and Jones, 1979;

- Radke et al., 2002). The resultant waters are either enriched in HCO3 and depleted in Ca2+ and Mg2+, or vice versa. Many Australian inland waters generally follow a different chemical pathway, being dominated by Na+ and Cl- ions, which lead to the eventual precipitation of gypsum and halite (De Deckker,

1988). In the simplest cases, for example where there is a direct connection between seawater and lake water, such as by groundwater, the dominance of ions is similar to that of seawater, whereby Na+>Mg2+>K+Ca2+ and

- 2- 3- 2- Cl >SO4 HCO +CO3 (Bayly and Williams, 1966). Deviations from this are determined by water-rock interaction, mineral dissolution and cation-exchange

161

reactions within the catchment; and sulphate reduction, mixing, mineral precipitation, brine reflux and recycling of soluble components within the lakes

(Radke et al., 2002). The relationship between ostracods' salinity and ionic tolerance in non-marine environments has been the focus of several authors, the most notable being Delorme (1969, 1971), Delorme et al. (1977), Forester (1983,

1986), Smith (1993) and Curry (1999).

In line with other proxy studies, such as palynology and the study of diatoms, the development of transfer functions have become more prominent in the study of ostracods (e.g. Smith, 1993, Fritz et al., 1994; Smith et al., 1997; Mourgiart et al.,

1998; Dean and Schwalb, 2000). This method provides a more holistic approach to the understanding of ostracod ecology by monitoring a range of variables, hence being of great value to palaeoenvironmental studies. The first of this form of training set has recently been established for Australian lakes (Radke et al.,

2003). The identification of ostracods by De Deckker (1975) within athalassic saline lakes in southeastern Australia, previously surveyed by Bayly and Williams

(1966) and Bayly (1970), represent the first ecological data for ostracods in this country. Since then, De Deckker (1981b,c,d 1983b, 1988) in particular has carried out a great deal of ecological work, thus the salinity ranges for most

Australian taxa are well constrained.

5. Ostracod Analyses

Figure 5.6. Examples of morphological variation of ostracods.

a. Schematic representation of the hypothetical variation of external ornamentation (from fine to coarse reticulation) within a species under a sliding scale of carbonate saturation levels (after Carbonel, 1988). The white portion represents the raised reticulation above the valve surface (shown in grey).

b. Loci of nodes observable on the carapace of Cyprideis torosa of the Northern Hemisphere. This feature is most commonly developed at low salinities (after Kilenyi, 1971).

c. Examples of sieve-pore morphology of Cyprideis torosa (after Rosenfeld and Vesper, 1977).

163

5.2.5 Environmentally cued polymorphism

In addition to the population structure of aquatic environments, morphological features of ostracods such as surface reticulation, carapace size and thickness, noding and sieve-pore shape may vary in response to fluctuating environmental conditions (Fig. 5.6). Numerous studies of taxa in waters transitional between fresh and saline have revealed an intraspecific gradational change in ornamentation (Fig. 5.6a) (Carbonel, 1988; Carbonel and Hoibian, 1988;

Peypouquet et al., 1988). The variation refers to the degree of expression of ornamentation, the locus of which is genetically predetermined (Carbonel, 1988).

This in turn corresponds to the carbonate equilibrium at the sediment-water interface, controlled by the biological consumption of organic matter; a relationship termed the aggradation-degradation "theory" of Peypouquet et al.

(1980). This is a two-phase process. The first phase produces dissolving

2+ - conditions, undersaturated with respect to Ca and HCO3 and resulting in degraded, smooth morphs. The second phase results in precipitating conditions

2+ - supersaturated with respect to Ca and HCO3 and aggraded morphs with heavy reticulation (Peypouquet et al., 1988). These situations occur particularly in marginal environments where there is an increase in organic matter from fluvial discharge or marine upwelling, or where there is strong seasonality. Such observations have not as yet been checked under controlled conditions.

In his studies of littoral ostracods of Australia, Hartmann (1978, 1979, 1980,

1981) noted variation in the surface ornamentation of at least three species from different locations. He attributed this primarily to variations in temperature of the hostwater (Hartmann, 1982). Temperature has a strong influence on both calcium

5. Ostracod Analyses

carbonate saturation and precipitation. As such, the rate and degree of calcification of ostracods valves is shown to be temperature dependent (Roca and

Wansard, 1997).

Bodergat et al. (1993) carried out geochemical analyses on various pseudomorphs of Leptocythere psammophila from a range of estuarine environments. They found that whilst the carapace morphology was notably different from each locality, the seasonal variation in chemistry was stronger than that among different environments. They relate the different shell chemistry to the temporal variability of the solute composition and salinity, determined by variations in seasonal inputs and hydrology.

Both McKenzie (1971) and Martens (1985) have examined the relationship between carapace size and salinity and have found some correlations, although these are specific to certain taxa. A detailed study of Limnocythere inopinata clones from saline lakes in Austria by Yin et al. (1999) related conductivity, and hence salinity of the water, to carapace length. This species was also used to display the relationship between Sr/Ca-inferred salinity and carapace roundness through a Holocene sequence from Kajemarum Oasis, NE Nigeria (Roberts et al.,

2002).

Although no systematic studies have as yet been carried out, field observations on variations in valve thickness suggest that the same general principal may be applied, with more heavily calcified valves being precipitated in carbonate-rich environments and poorly calcified valves in depleted environments

165

(De Deckker, 2002). Under stressed conditions, such as high temperature or low oxygen level, smaller adult valves are also likely to be formed by organisms.

The occurrences of nodes, projections that may be present on the external carapace of some ostracods, are another morphological response to environmental variability (Fig. 5.6b). These features, commonly found on the ostracod Cyprideis torosa, have been the subject of many studies (e.g. Kilenyi, 1971; Vesper, 1975; van Harten, 1996, 2000, Keyser, 2001). The location of the nodes on the carapace is genetically determined, however the number and degree of expression is variable. The noded forms are a feature of low salinity conditions, commonly

<8 ‰. Whereas salinity appears to be the primary factor controlling nodosity, culturing experiments reflect that this alone is too simplistic (e.g. van Harten,

- 1996). A second parameter, most likely the pH, HCO3 or Ca content of the host water, also exerts an influence (van Harten, 2000; Keyser, 2001).

Sieve pores, which occur only in cytheroidean podocopans, show a variety of shapes (Fig. 5.6c). These have been attributed to variations in the salinity of the host waters (Rosenfeld and Vesper, 1977) although this is based on observation, rather than empirical testing or culturing experiments. Once again, the ostracod

Cyprideis torosa has been the main focus of study, although the phenomenon is known for other taxa. Of the three sieve-pore-types identified, Rosenfeld and

Vesper (1977) found round forms dominant in freshwater and irregular shapes dominant in hypersaline conditions. The elongate form appears to be intermediary to these two end-members. Rather than salinity per se, De Deckker

(2002) relates the variation in pore shape to a reflection of the 'readiness' with

5. Ostracod Analyses

which the ostracod secretes its carapace; more irregular forms of sieve-pore shape indicative of more stressed conditions such as poor calcite saturation in highly saline estuarine waters.

5.2.6 Depositional environment

Palaeoecological information can be extracted from ostracod valves and assemblages regarding both ambient conditions at the time of valve formation and post-depositional effects. Within a species, the population age structure preserved may be determined by measuring the length to width ratio of the valves present in the assemblage. In an ideal situation there should be 18 valves deposited for every individual adult ostracod, or one adult to every 8 juvenile instars, and sexual dimorphism may be apparent. This is rarely the case in reality (Fig. 5.7). The absence of juvenile stages is commonly due to the effects of current activity, breakage or dissolution (in waters poorly saturated with respect to CaCO3) of the earlier and more fragile instars. A low-energy environment may be indicated by a broad distribution of instars being preserved, whereas higher energy is indicated by the dominance of adult valves (e.g. Brouwers, 1988). In addition, the absence of adult valves may represent a change in conditions during the life cycle of the ostracod, such as a decrease in temperature or dissolved oxygen concentration

(De Deckker, 2002). Dynamic environments may also be indicated by allochthonous elements within the assemblage, transported by water, wind or with the aid of agents such as waterfowl (Boomer and Eisenhauer, 2002). As such, population age structure may be indicative of palaeoenvironments (e.g. Whatley,

1988; Ruiz et al., 2003). Careful note should also be taken, particularly in fossil samples, of mixed assemblages. This may reflect mixed provenance, reworking

167

of the sediment or a change in environment in the time represented by the sediment interval being examined (De Deckker, 2002).

Figure 5.7. Schematic diagram of ostracod population age structure and associated environmental conditions (after Whatley, 1988 and De Deckker, 2002). M and F refer to male and female valves, respectively. A-1, -2, -3… indicates the juvenile instar, whereby A-1 refers to the penultimate to adult, etc.

5. Ostracod Analyses

The preservation of the valves themselves may give some clues to the post-depositional conditions of the environment. De Deckker (2002) provides a summary of these effects. When living, a chitinous sheath protects the ostracod carapace. Shortly after death, microbes consume the sheath and the two valves separate. As such, whole carapaces are rarely preserved in the fossil record.

Where present, this may be indicative of either rapid deposition or burrowing of the ostracod in the sediment prior to death (De Deckker, 1988). The microbial activity leaves a trail on the calcitic shell. The trails represent areas of weakness in the valve and leave it susceptible to breakage. Hence, the presence of broken shells within an assemblage may be indicative of a higher-energy environment or bacterial activity in a low-energy environment (De Deckker, 1988).

Dissolution is a common effect in eutrophic environments, with ostracod valves being deposited in organic-rich sediment upon death. The first sites of dissolution within the carapace are the sieve-pores, as the shell is thinner at these points

(Fig. 5.8). If bacterial activity is prominent in an anoxic environment, microscopic pyrite crystals form, adhering to the calcite valves and in some cases, particularly for interstitial ostracods, infilling the carapace (Oertli, 1971). If this pyrite is later exposed and oxidised, further dissolution of the valve may occur.

This is identifiable as a whitish or chalky appearance to the valves. Experiments carried out by Danielopol and Handl (1990) on well-preserved valves of

Cytherissa lacustris found that milky valves could be produced when placed in a solution of pH 6.5 and dark coloured valves could be produced within weeks of exposure to organogenic and sulphidic sediment. Minor dissolution, by way of modification of the carapace reticulation, may also occur. This is due to

169

fluctuations of the host water carbonate saturation at or near the sediment-water interface (Swanson, 1995). The degree of dissolution, based on observation of ostracod valves under stereomicroscope, has been described in an index by

Swanson and van der Lingen (1994). The "carapace corrosion scale", ranges from

0 (transparent valve, near perfect preservation) to 6 (valve chalky, collapses during manipulation or when wet) and has been applied to the taxa observed in this current study (Table 3).

Diagenesis of ostracod valves may take place within the sediment. Such affects should be observable on the ostracod valves, as a grainy or chalky texture when viewed with the aid of a binocular microscope. Recrystallisation of ostracod valves, leading to a sugary appearance, is also possible in sediment influenced by a secondary water source that may be enriched with calcium carbonate.

Alternatively, a calcareous coating may be formed around the entire carapace or encasing the individual valves (De Deckker, 2002).

5. Ostracod Analyses

Table 3. Visual preservation index (VPI) as applied to ostracod valves from the core material from the Gulf of Carpentaria. The index refers to the degree of dissolution observable under a stereomicroscope (after Swanson and van der Lingen, 1994).

Figure 5.8. Variation in the valve preservation of Cyprideis australiensis from core MD-32. The ostracods are presented from the most well- preserved valve to that showing the greatest dissolution (A, VPI - 0; B, VPI - 2; C, VPI - 4; D, VPI – 5) as seen under SEM. Note that the sieve-pore regions are both the first and the most adversely effected by dissolution of the carapace. (Valves taken from sample depths: A - 9.1 m, B -1.5 m, C - 6.7 m, D - 5.6 m).

171

5.3 Modern ostracod fauna

Publications on modern ostracod taxonomy and assemblage data from the Gulf of

Carpentaria and surrounding regions have been reviewed (Fig. 5.9, 5.10).

Comparisons between these papers and material from the present study have been made to determine species identification, enable facies delineation and palaeoenvironmental reconstruction of the core assemblages, and postulate affinities between the fauna of the Gulf of Carpentaria and other zoogeographic provinces.

5.3.1 Ostracods from surface samples of the Gulf of Carpentaria

The most comprehensive study of modern ostracods of the Gulf of Carpentaria was conducted by Yassini, Jones and Jones (1993), in which they identified 82 species. These authors delineated three faunal assemblages based on Q-mode cluster analysis: a shallow open marine, a lower tidal and foreshore, and an upper tidal assemblage.

The open shallow marine assemblage includes 74 species, although many of these are only present in low abundance. The dominant species are Neonesidea australis, Paranesidea sp., Neocytheretta spongiosa, Venericythere papuensis and

Pterygocythereis velivola. Other species present include Cytherella semitalis,

Cytherelloidea darwinensis darwinensis, Cytherelloidea malaccaensis and

Labutisella quadrata. This assemblage is common to the main body of the gulf.

5. Ostracod Analyses

The lower tidal and nearshore assemblage has been largely identified from samples taken from Accident Inlet and Karumba, located in the southeastern corner of the gulf (Fig. 5.9). The Accident Inlet samples are largely sandy-muds and contain 30 species, dominated by Phlyctenophora zealandica and

Neomonoceratina bataviana, with Paracytheroma caudata, Loxoconcha judithae,

Alocopocythere goujoni, Actinocythereis tetrica and Neocytheretta adunca present. The samples from Karumba were taken from a more sandy coastal setting. Only 13 species are represented here, dominated by Loxoconcha judithae,

Paracytheroma caudata, Phlyctenophora zealandica and Neomonoceratina bataviana, with minor Actinocythereis tetrica and Neomonoceratina porocostata.

The third assemblage is common to samples taken from the upper tidal limit at

Fitzmaurice Creek, 21 km inland of Accident Inlet, and to two samples taken from the deepest part of the modern gulf. This correlation is based on the dominance of

Cyprideis australiensis, a common marginal marine ostracod. As the sedimentation rate is very low in regions of the gulf deeper than around 20 m, the presence of this species in deeper water most probably represents a relict deposit

(Yassini et al., 1993). This is supported by the presence of abundant reworked valves of the non-marine ostracod Ilyocypris australiensis. The well-preserved ostracods in the two deeper samples are common to the open marine assemblage outlined above.

173

5. Ostracod Analyses

175

5. Ostracod Analyses

Figure 5.10. Map showing the location of previous ostracod investigations referred to in this thesis. 1. Brady, 1880. 2. Hartmann, 1978. 3. Hartmann, 1979. 4. Hartmann, 1980. 5. Hartmann, 1981. 6. Howe and McKenzie, 1989. 7. Yokoyama et al., 200, 2001; Clarke et al., 2001. 8. Labutis, 1977. 9. Behrens, 1991. 10. Yassini and Jones, 1995. 11. Kingma, 1948. 12. Dewi, 1997. 13. Dewi, 2000. 14. Whatley and Zhao, 1987, 1988. 15. Mostofawi, 1992. 16. Zhao and Whatley, 1989. 17. Hanai et al., 1977, 1980. 18. McKenzie, 1966. 19. De Deckker, 1981b. 20. De Deckker. 1981d.

177

Figure 5.11. Map showing the ostracodal zoogeographic provinces of the Indo-Pacific region (after Titterton and Whatley, 1988).

5. Ostracod Analyses

5.3.2 Previous regional studies

5.3.2.1 Marine ostracods

Titterton and Whatley (1988) reviewed the entire body of work published on

Indo-Pacific and Southern Ocean, Tertiary to Recent, shallow marine platycopid and polycopid ostracods, current to 1985. From the 2599 species recorded in 285 publications, they established 13 zoogeographic provinces (Fig. 5.11). The

Australian province (incorporating the mainland and including Tasmania) is separated from the East Indian province (comprising most of Southeast Asia including West Papua) to the northwest, the Southwestern Pacific province

(including Papua New Guinea) to the northeast, the New Zealand province to the southeast and the Southern Ocean province to the south. The Australian province has the largest number of recorded species, showing greatest affinity with the adjacent provinces and moderate affinity with the Indian Ocean, via Eastern

African and the Bengalian Provinces. The presence of the Tethys during the early

Tertiary allowed for the rapid dispersal of marine ostracods through the region

(McKenzie, 1967). The establishment of provinciality has occurred since the closure of the Tethys. The modern ostracod distribution is largely determined by modern ocean currents (Titterton and Whatley, 1988).

The locations of previous studies of modern ostracods in the region are outlined in the following text are presented in Fig. 5.10. The earliest data on ostracods in the northern Australian region were recorded by Brady (1880), following the

Challenger expedition during 1874. During this campaign samples were taken from offshore of Cape York Peninsula in Torres Strait (Station 185, Lat. 11o35'S,

Long. 144o3'E, depth approx. 282.5 m) and near Booby Island (Station 187, Lat.

179

10o36'S, Long. 141o55'E, depth approx. 11-14.5 m). From these samples 46 species were identified, including 12 from both Torres Strait and Booby Island, north of Queensland, and a further five from Humboldt Bay, Papua New Guinea and the Arafura Sea combined (Brady, 1880). These have been reviewed by Puri and Hulings (1976). Species common with those found in the Gulf of Carpentaria cores include Polycope favus, Phlyctenophora zealandica, (Neocytheretta)6 adunca, (Alocopocythere) goujoni, (Venericythere) papuensis, (Pistocythereis) cribriformis, (Pterygocythereis) velivola, (Bishopina) spinulosa and (Foveoleberis cypraeoides). Earlier work had been carried out by Brady in Indonesian waters from the 'Les Fonds de la Mer' cruise (e.g. Brady, 1868). The species Cytherella semitalis, (Phlyctenophora zealandica), (Pontocypris) attenuata,

(Alocopocythere) goujoni, (Neocytheretta) spongiosa, (Venericythere) darwini,

(Pistocythereis) euplectella, (Pistocythereis) cribriformis, (Bishopina) spinulosa,

(Pseudopsammocythere) reniformis, (Cytherura) bataviana and (Foveoleberis) cypraeoides were described in these papers.

A more systematic survey of the nearshore region around much of Australia was investigated by Hartmann (1978, 1979, 1980, 1981). This series of works has been subdivided on the basis of location: 102 species described from the tropical and subtropical West Australian coast (Hartmann, 1978); 78 species from the west and southwest coast, from Perth to Eucla (Hartmann, 1979); 107 species from the southern and southeast coast, from Ceduna in the west to Lakes Entrance in the east (Hartmann, 1980); 75 species from the subtropical to tropical east coast, from

6Where a taxon has been renamed since the listed publication, the current name in usage is represented in parentheses; for example Cythere adunca (Brady, 1880) is referred to here as (Neocytheretta) adunca.

5. Ostracod Analyses

Eden in the South to Heron Island in the north (Hartmann, 1981). The ostracods recorded from the west coast tropical region show the closest affinity to those of the Gulf of Carpentaria. Common or related species include Paranesidea onslowensis, (Paracytheroma) mangrovicola, Tanella gracilis, Alocopocythere

(goujoni), Caudites cf. javana, Caudites exmouthensis and Phlyctenohora cf. zealandica. The majority of these ostracods are also found in the tropical

Indo-Pacific region. South of Shark Bay, the authors note a change in ostracod fauna, with the presence of permanent brackish water conditions noted by the appearance of Cyprideis australiensis (Hartmann-Schröder and Hartmann, 1977).

This is also a common species in the brackish lagoons and inlets in the southern regions of the west Australian coast. It is of interest to note that the species

Leptocythere hartmanni was recorded in all of the abovementioned studies, most commonly as a constituent of the brackish water biotope. A second species

Leptocythere keijia, which resembles more heavily calcified Leptocythere hartmanni morphs, was also found in the western sample sites.

Howe and McKenzie (1989) identified 130 species from Darwin, Northern

Territory, and Port Hedland and Hamelin Pool Shark Bay, Western Australia. All of the samples were taken from the nearshore environment. The Darwin assemblage is rich and diverse, with abundant Cytherella semitalis (mostly juveniles), Neomonoceratina porocostata, Xestoleberis darwinensis, Actinoleberis arafurae, Mackenziartia bentleyi, Callistocythere cf. insolita, Keijia nordaustraliae, Labutisella darwinensis, Venericythere cf. darwini and Caudites cf. javana. The Port Hedland assemblage shows less diversity7, dominated by

7 Please note that in this instance and throughout the text the term 'diversity' refers to simple species diversity, not a diversity index. 181

Xestoleberis broomensis, Xestoleberis porthedlandensis and Xestoleberis cf. cauticola, also with abundant Paranesidea onslowensis, Neomonoceratina porocostata and Australimoosella liebaui. The Hamelin Pool, Shark Bay, assemblage is smaller, most probably due to the high salinity of this waterbody, ranging from 60-65‰. Species present include Paracypris occidenslevis,

Paracytheroma hamlinensis, Xestoleberis cf. exmouthensis, Xestoleberis cf. portaugustensis, Loxoconcha georgei and Leptocythere cf. lacustris, although no adults of this last species were recovered.

Two recent studies have been undertaken on core material from the Joseph

Bonaparte Gulf, off northwestern Australia. The first utilised faunal assemblages to distinguish facies variation in response to sea-level change around the Last

Glacial Maximum (Yokoyama et al., 2000; 2001). Open marine (pteropods, diverse foraminifers and ostracods), shallow marine (ostracod genera including

Argilloecia, Callistocythere, Loxoconcha, Pterygocythereis and Uloberis

(Foveoleberis), and no pteropods), marginal marine (Neocytheretta and

Xestoleberis) and brackish (ostracods Cyprideis, Leptocythere and Neocytheretta and large numbers of the foraminifer Ammonia beccarii) environments were recognised. The second study looked at material spanning the most recent post-glacial transgression and sea-level highstand (Clarke et al., 2001). The biota found may be divided into six broad environments; intertidal, lagoonal, estuarine

(characterised by the dominance of Phlyctenophora bentleyi and the absence of bairdiids), strandline, shelf (inshore – Neocytheretta spongiosa and Neonesidea australis, open marine – McKenziarta bentleyi, Labutisella cf. darwinensis,

Labutisella quadrata, Keijia nordaustraliae and Praemunita sp.) and riverine.

5. Ostracod Analyses

The ostracods identified show a strong correlation with those found in the Gulf of

Carpentaria cores, with 18 common species including Foveoleberis cypraeoides,

Loxoconcha judithae, Neocytheretta spp., Neomonoceratina spp. and

Labutisella spp.

Although not published, Labutis' thesis (1977) made a detailed study of the cytheracean fauna of the Great Barrier Reef, southeast Queensland. A total of 45 samples were studied from about Gladstone Harbour, Wistari Reef and Heron

Island lagoons and in two transects from Curtis Island to the Capricorn Reefs.

Sample depths did not exceed 50 m. The samples are rich and diverse and show a high degree of endemism at species level. At least 21 genera show affinities with the Indo-Pacific region and 18 genera are cosmopolitan in their distribution, although many taxa are unique at species level. Taxa common to the Gulf of

Carpentaria include Neomonoceratina spp., Neocytheretta spp., Callistocythere spp., Tanella gracilis, Xestoleberis spp., Caudites spp. and 17 species of pectocytherinids.

The ostracods from Lizard Island, of the northern Great Barrier Reef, were studied by Behrens (1991). Samples were taken at depths of 0.5-16 m from a range of environments including mangroves, sand flats, mud flats, sea-grasses, dead coral fragments and algae. The sampling was undertaken over a seven-month period to ascertain the seasonality of the ostracod population. Indeed, Microcytherura aestuaricola was observed in large numbers in the sandy eulittoral zone in the winter, whereas Paracytheroma aestaphila was the dominant species in the same environment in summer. Six species of Loxoconcha were identified, mostly from

183

dead coral and algae samples. Cyprideis australiensis was also recovered in significant numbers from brackish waters among mangroves in muddy sediment.

The ostracods of estuarine and continental shelf setting of southeastern Australia, from Port Stephens in the north to Bass Strait in the south, have been reviewed by

Yassini and Jones (1995). They describe 241 species of mostly temperate biogeography, and provide a generalised distribution pattern within the environments studied. Taxa common to the Gulf of Carpentaria include

Neonesidea australis, Phlyctenophora zealandica, Alocopocythere (goujoni),

Cytheropteron wrighti, Leptocythere hartmanni, Leptocythere lacustris and

Cyprideis australiensis.

An important contribution to the knowledge of ostracods in the Southeast Asian region was published by Kingma (1948). The work includes samples taken near the coast of Aceh, North Sumatra, the South Kendeng region and core samples from Bodjonegoro, East Java and material collected from the Java Sea by the

Snellius Expedition, 1929-30. A total of 97 species were identified; 41 from the

Bodjonegoro core, 37 from Aceh, 31 from Kendeng and 19 from the Java Sea. Of the sample sites, the greatest degree of commonality occurs between the samples of Aceh and Kendeng. The fauna from Bodjonegoro is almost entirely unique.

Each of the Kendeng, Aceh and Java Sea samples share common species with the

Gulf of Carpentaria, including Cytherella semitalis, (Phlyctenophora) zealandica,

(Neocytheretta) vandijki, (Venericythere) papuensis, (Pistocythereis) cribriformis,

Caudites javana, Tanella gracilis, Javanella kendengensis, Neomonoceratina

(bataviana) and (Foveoleberis cypraeoides). Zhao and Whatley (1989a) revised

5. Ostracod Analyses

31 of the new species described by Kingma (1948). The material of the Snellius

Expedition was revisited by Keij (1953) who described the fauna as relatively poor both in diversity and abundance. Hanai et al., (1980) compiled a checklist of all known Ostracoda of Southeast Asia. Within this list are 19 marine species common to the Gulf of Carpentaria.

In her thesis on the Ostracoda of the Java Sea, west of Bawean Island, Indonesia,

Dewi (1997) describes 113 species. She delineated three faunal assemblages based on Q-mode cluster analysis: a nearshore assemblage, close to the coasts of

Java and Bawean Islands, is dominated by Pistocythereis cribriformis and

Phlyctenophora orientalis; a shallow marine assemblage, present in water depths

<63 m, with abundant Cytherella semitalis, Cytherelloidea cingulata, Hemikrithe orientalis and Neomonoceratina bataviana; and a deeper water assemblage, on the north of the study area dominated by Borneocythere paucipunctata, Neocytheretta vandijki, Alataconcha pterogona, Foveoleberis cypraeoides and Actinocythereis scutigera.

The distribution of ostracods in the Java Sea off Selatan, southern Borneo, has also been studied by Dewi (2000). A total of 99 species were identified from 40 surface sediment samples, taken in a range of water depths from 11-42 m. Only dead ostracods were recovered. The greater concentration of ostracods occurs in water depths of 20-40 m. The shallower areas are dominated by high sediment discharge from rivers. Common species that are widely distributed throughout the study area include Foveoleberis cypraeoides, Loxoconcha paiki,

185

Cornucoquimba sp., Phlyctenophora orientalis, Neomonoceratina bataviana,

Mutilis sp. and Pistocythereis cribriformis.

Whatley and Zhao (1987, 1988) have identified 129 species from 18 bottom samples taken from the Malacca Straits from depths of 20-100 m. They argue that the diversity of the fauna present is largely a function of the substrate, with the greatest variety of species being found in the medium to coarse sands rich in organic matter. They have defined two broad assemblages: a shallow water fauna

(<50 m) comprising Foveoleberis cypraeoides, Hemikrithe peterseni, Keijia labyrinthica, Keijella multisulcus, Cytherelloidea leroyi, Neocytheretta snelli and

Neocytheretta spongiosa; and a deeper water assemblage (>50 m) including

Abrocythereis guangdongensis, Bythoceratina paiki, Bythocytheropteron alatum,

Bradleya andamanae, Cytheropteron rhombiforme, Cytheropteron parasinense,

Cytheropteron malaccaensis, Keijella paucipunctata, Ruggiera indopacifica and

Stigmatocythere parakingmai. Several species are common throughout this area and beyond, such as Actinocythereis scutigera, Alocopocythere kendengensis,

Cytherelloidea leroyi, (Venericythere) papuensis, Lankacythere coralloides,

Neocytheretta adunca, Neomonoceratina indonesiana and Pistocythereis bradyi.

The ostracods of the Sunda Shelf, taken from 44 surface samples along a transect between the south of Malaysia and the west of Borneo, have been described by

Mostafawi (1992). From this, 116 species were identified. Again, substrate type has been determined to be the major controlling factor in the distribution of ostracods, with organic-rich medium to coarse sands the preferred environment for most taxa. Quartz-dominated, carbonate poor sands were less populous.

5. Ostracod Analyses

Based on faunal composition, two broad assemblages have been defined: a shallow water assemblage comprising Atjehella kingmai, Neomonoceratina delicata, Hemicytheridea cf. oculosa and Tanella gracilis; and a deeper water assemblage including Bythocytheropteron carinatum, Abrocythereis malaysiana,

Heinzmalzina rhombiformis and Propontocypris rostrate. Most of the ostracod species present in these samples show no clear depth preference within the study area. Such species include Foveoleberis cypraeoides, Actinocythereis scutigera,

Pistocythereis cribriformis, Parakrithella pseudornata and Neocytheretta spongiosa.

Further work on the Malay Peninsula has been undertaken by Zhao and Whatley

(1989b). They have described ostracods from shallow waters (<20m) from the

Sedili Bay and Jason Bay regions. A total of 101 species were recorded, however only 50 of these were live at the time of collection. The majority of the live specimens were obtained from the shelf. Another two samples were retrieved from the freshwater Sedili River and a further four from the brackish Sedili

Estuary. The combined dead assemblage of the three sampling sites constitutes a total of 98 species. Biotopes are defined on the basis of the living assemblage.

From this study the principal component of ostracod distribution is salinity. The river biotope is restricted to the species Darwinula stevensoni and ?Cytherissa sp.

(juv.), at the most upstream sampling station, 30 km inland. Species of the estuarine biotope were present from the river mouth to about 8 km upstream. The living species comprise Sinocythere superba, Hemicytheridea reticulata,

Neocytherideis sp. and Keijella multisulcus. Of the common open-sea species,

Parakrithella pseudornata and Tanella gracilis are widespread and along with

187

Cushmanidea subjaponica, span the intertidal zone. Hemicytheridea reticulata,

Hemicytheridea wangi, Javanella kendengensis and Xestoleberis cf. hanaii are confined to waters shallower that 14 m, whereas Atjehella semiplicata and

Phlyctenophora orientalis occur in depths of 10-20 m. The dead fauna is far more diverse, particularly in the open sea environment. The dominant species are

Hemicytheridea reticulata, Neomonoceratina bataviana, Neomonoceratina delicata and Lankacythere coralloides. Abundant species include Actinocythereis scutigera, Alocopocythere kendengensis, Keijella jankeiji, Keijella papuensis,

Parakrithella pseudornata, Phlyctenophora orientalis, Pistocythereis bradyi,

Stigmatocythere rosemani and Tanella gracilis. In deeper waters Cytherella semitalis, Cytherelloidea cingulata, Cytherelloidea leroyi and Cytherelloidea malaccaensis are also common.

The faunal checklists of Hanai et al. (1977, 1980) indicate that faunal associations with ostracods from the Gulf of Carpentaria extend across the South China Sea and to a lesser degree, Japan. Species common to these areas include Cytherella semitalis, Cytherelloidea malaccaensis, Neocytheretta spp., (Venericythere) darwini, (Venericythere) papuensis, (Pistocythereis) bradyi, (Pistocythereis) euplectella, Argilloecia hanaii, Copytus posterosulcus, Javanella kendengensis,

Neomonoceratina bataviana and (Foveoleberis cypraeoides).

Faunal affinities with the Gulf of Carpentaria also extend across the Indian Ocean.

Hussain and Mohan (2001), in a study of the Adyar River Estuary in the Bay of

Bengal, India, describe 26 species from 15 surface sediment and bottom water samples. Of these, Tanella gracilis and Caudites javana were the most

5. Ostracod Analyses

widespread and abundant. Other species found also in the present study include

Phlyctenophora (orientalis) and Neomonoceratina porocostata, common in shallow marine and brackish water samples. Upstream of the estuary, the genera

Cyprideis, Cypridopsis and Cyprinotus were observed to be abundant.

5.3.2.2 Non-marine ostracods

To date there have been no taxonomic studies of non-marine ostracods in the region surrounding the Gulf of Carpentaria. The majority of work on non-marine ostracods of Australia has focussed on saline lakes of the south and southeast of the continent (e.g. De Deckker and Geddes, 1980; De Deckker, 1981b,c; Geddes et al., 1981; Radke et al., 2003). Genera common to those found in the present study, Cyprideis, Ilyocypris, Leptocythere, Limnocythere and Cyprinotus, have been described from athalassic saline lakes by De Deckker (1981d). Victor and

Fernando (1982) consider the genera Cypretta, Ilyocypris, Limnocythere and

Darwinula as cosmopolitan throughout the Southeast Asian region.

McKenzie (1966) has presented the only systematic record of freshwater ostracods from northern Australia, based on plankton samples collected from shallow ponds in the Northern Territory and the Kimberley, Western Australia.

He described 17 species of 12 genera including Ilyocypris australiensis present in half of the ostracod-bearing samples, common Newnhamia fenestrata, Cyprinotus kimberleyensis, Heterocypris sydneia, and Hemicypris cf. fossulata, and four species of Cypretta. These fauna exhibit affinities with those of eastern Australia,

189

Indonesia and South Africa. De Deckker (1982b) has also described the species

Cypretta yapinga from a billabong along the Magela Creek, Northern Territory.

Ecological notes for the non-marine ostracods found in the core samples of this present study of the Gulf of Carpentaria are available from previously published literature such as De Deckker (1981b, c, d). Details of these fauna are presented below. The salinity range of a selection of ostracods that are comparable with those found in Lake Carpentaria is presented in Table 4.

Table 4. Recorded salinity ranges of a selection of non-marine ostracods found in Australia. (Information from De Deckker, 1981a,b,c,d, 1982, 1983a, 1988). Solid bars indicate salinity range, circles denote single observations.

5. Ostracod Analyses

Cyprideis australiensis

The genus Cyprideis is typical of permanent saline lakes and intertidal saltmarshes. It is commonly the dominant taxa in brackish water environments, particularly in lagoons, in which it may be found as monospecific 'blooms'.

Cyprideis australiensis is the local representative of the cosmopolitan genus. It may be considered similar in habit to the northern hemisphere species, Cyprideis torosa, about which much has been written (Mezquita et al., 2000). Cyprideis torosa is both eurythermal and euryhaline. Although most abundant in the

2-16.5‰ salinity range, it is known to tolerate salinities of up to 150‰

(De Deckker, 1983b). As previously mentioned, noded forms may be present in fresh and oligohaline waters. The versatility of this organism extends to anoxic and sulphidic conditions, whereby it switches to anaerobic respiration (Aladin,

1993; Jahn et al., 1996). Cyprideis occurs only in permanent waters, as the eggs cannot withstand desiccation (De Deckker, 1983b). It is a benthic taxon and most commonly found in soft sediment and shallow waters (Heip, 1976).

Leptocythere hartmanni

Leptocythere hartmanni is a common species in lagoonal and other marginal environments. It is able to tolerate wide fluctuations in salinity (10-45‰), although requires permanent water to reproduce (Yassini and Jones, 1995). No males of this species have been recorded (De Deckker, 1981b). Leptocythere hartmanni is found throughout Australia. Morphological variation of the valves, ranging from finely reticulate to heavy ornamentation has been described for this taxon (McKenzie, 1967; Hartmann, 1978, 1979, 1980, 1981).

191

Leptocythere lacustris

Leptocythere lacustris is a benthic ostracod, restricted to coastal saline lakes.

Typical of the genus, Leptocythere lacustris requires permanent water to reproduce, as the eggs are not able to withstand desiccation (De Deckker, 1981b).

This species has been collected from four sites near Robe, South Australia, with salinities of 19-28‰ and 2.8‰ (De Deckker, 1981b). Juveniles of a comparable species have been identified from Hamelin Pool, Shark Bay, with salinities of

60-65‰ (Howe and McKenzie, 1989). There has been some morphological variation described in the valves, from almost smooth to coarsely reticulate

(De Deckker, 1981b).

These two forms of Leptocythere (Leptocythere hartmanni and Leptocythere lacustris) have not previously been described from the same locality.

Differentiation between the two is equivocal, particularly when the range of morphological variation described for each taxon is considered. This is further demonstrated in the gradation of morphologies of the Leptocythere valves retrieved from the present study (Fig. 5.12), suggesting that the two previously assigned species may be better considered to be morphological end-members of the one species. Previous authors note the similarity of the females of the two taxa, found in different locations, and note that the carapace morphology appears to be the main difference (De Deckker, 1981b; Howe and McKenzie, 1989). As no males of Leptocythere hartmanni forms have yet been found, a definitive comparison of the soft parts of the two has not been made. The name

Leptocythere hartmanni lacustris will be utilised hereafter in reference to the finely reticulate morphotype, most common in lacustrine environments.

5. Ostracod Analyses

193

5. Ostracod Analyses

Limnocythere sp.

Species of the genus Limnocythere are most commonly associated with permanent freshwaters (<3‰) in Australia, but have also been found in brackish conditions.

One species, Limnocythere milta (synonymous with the northern hemisphere species, Limnocythere inopinata), was recorded in a small lake near Colac, western Victoria, with a salinity of 15.4‰ (De Deckker, 1981d). Limnocythere is often characteristic of the littoral zone, with a preference for still waters (Griffiths and Holmes, 2000). Common species recorded in Australia include Limnocythere dorsosicula and Limnocythere mowbrayensis. The genus is cosmopolitan (Victor and Fernando, 1982). Morphological variation of Limnocythere valves has been noted by Carbonel and Peypouquet (1983). They associate heavier reticulation with decreased salinity and increased dissolved calcium ions, although this is yet to be experimentally verified.

Ilyocypris australiensis

Ilyocypris australiensis is a common species in temporary pools, shallow and slightly saline lakes and flowing waters. The species is capable of swimming, however it is essentially a benthic dweller. Ilyocypris australiensis occurs predominantly in waters within the salinity range of 4-7‰, although has been recovered from Lake Kariah, in Victoria, with a salinity of 10.4‰ (De Deckker,

1981b). It has neither been recorded in deep nor permanent freshwater lakes.

Morphological variation of the species, with regards outline and ornamentation has been noted, but cannot yet be attributed to an ecological variable (De Deckker,

1982a). Ilyocypris australiensis has been collected across Australia and is also

195

recorded in North Africa, Asia and Southern Europe. It may be synonymous with the species Ilyocypris bradyi, described throughout Europe.

Cyprinotus cingalensis

Species of Cyprinotus are a common feature of temporary pools, with salinities of up to 5‰ (De Deckker, 1983b). It has also been noted inhabiting streams throughout Southeast Asia (Victor et al., 1981). Cyprinotus cingalensis is found predominantly in freshwater settings in Australia. It is known also from Japan, where it shows a preference for brackish waters (Okubo, 1974).

Candonocypris novaezelandiae

Candonocypris novaezelandiae is found in both permanent freshwater and semi- permanent water-bodies. It is common in eutrophic waters, commonly in high numbers in black organic muds and decaying vegetal matter, particularly near lake shores (De Deckker, 1981d). Whereas both sexes are present in permanent waters, only females have been collected from more temporary waters, such as farm dams (De Deckker, 1983b). Although juveniles of the species are good swimmers, adults are strictly benthic. Candonocypris novaezelandiae has been described from Australia, New Zealand and Japan (De Deckker, 1983b).

Cypretta sp.

Species of the genus Cypretta are common to permanent freshwater and semi-permanent water-bodies. They also tolerate slowly flowing waters (De

Deckker, 1983b). Cypretta baylyi has been described from temporary pools in

Western Australia and the Northern Territory (De Deckker, 1983b; McKenzie,

5. Ostracod Analyses

1966). Although Cypretta baylyi is predominantly benthic, other species of the genus are good swimmers.

Zonocypretta sp.

The genus Zonocypretta has been described for the species Zonocypretta kalimna from Western Australia. This species was recovered from a pond that was fresh at the time of collection (De Deckker, 1981c).

Darwinula sp.

Species of Darwinula preferentially inhabit permanent freshwater environments, although have been found in brackish conditions (Griffiths and Holmes, 2000).

They are cosmopolitan in distribution, found living on every continent except

Antarctica. The species Darwinula stevensoni is the most common. It is characteristically eurythermal, with a preference for shallow, clear waters. All

Darwinula reproduce parthenogenetically (Smith and Horne, 2002).

It is of interest to note that none of the six endemic halobiont ostracod genera

(Australocypris, Diacypris, Mytilocypris, Platycypris, Reticypris and

Trigonocypris), so characteristic of Australian saline lakes, have been found in the core material of the present study. This may be due to these taxa having no direct marine or freshwater ancestry. Conversely, the two halobiont species with marine ancestry, Cyprideis australiensis and Leptocythere hartmanni lacustris, are common in the lacustrine sediment of the Gulf of Carpentaria. These species require permanent water to reproduce. It has also been noted by De Deckker

(1983b) that species diversity of ostracods in saline lakes is greater near the coast

197

than inland. This is most likely due to the more regular water supply of coastal lakes, being filled with water on an annual basis compared with the sporadic nature with which inland lakes may receive water.

5.4 Results of the ostracod analysis of core MD-32

5.4.1 Ostracod fauna present in core samples

Ostracod faunas from 149 sediment samples, weighing approximately 8.0 g each when wet and taken at 10 cm intervals from core MD-32, were examined8. From this study, 29,734 ostracod valves, representing 72 species of ostracods from 52 genera, were identified. Of these, 14 remain in open nomenclature, 8 having been figured before without full taxonomic description. A further 10 species are annotated 'cf.', indicating the taxon is comparable with, but possibly not identical to a previously described species.

A complete systematic species list comprising all taxa found in core MD-32 is given in App. 6. SEM images of 70 taxa are presented in plates 1-5 of App. 7.

The classification of the species present is based on previously published literature, as outlined in chapter 3. Formal descriptions of the taxa present have been omitted from this thesis, as most species are known to science. Brief descriptions of those taxa that were identified with reservation or have been left in open nomenclature and previously undescribed have been included. Most of these examples exist in too few numbers to be positively identified.

8 In samples whereby the sieved and dried >63 µm fraction exceeded 25 mg, the material was randomly split and an appropriate fraction (up to 20 mg) examined and taken to be representative of the whole sample.

5. Ostracod Analyses

Table 5 Ostracod faunal comparison of core MD-32 with the surrounding region.

199

5. Ostracod Analyses

The species abundance and diversity varies markedly throughout the core, indicating distinct depositional environments. Six species are present in only one sample and 20 species are represented by no more than two valves per sample.

Ostracods were not observed in 32 of the samples, however 48 samples comprise greater than 10 species and 9 samples comprise in excess of 100 valves per gram of dry sediment, dominated by just one or two species.

Many of the species identified are found in the modern Gulf of Carpentaria and surrounding estuaries. Yassini et al. (1993) record 44 species that are in common with those of the core samples. The ostracod fauna display tropical-subtropical affinities. Taxa common to core MD-32 that have been described from the abovementioned regional studies are outlined in Table 5. Of the 72 species identified, 38% are common to the Arafura Sea and northern Australian waters

(Brady, 1880; Hartmann, 1978; Howe and McKenzie, 1989), 36% to Western

Australia and the Joseph Bonaparte Gulf (Hartmann, 1979; Howe and McKenzie,

1989; Clarke et al., 2001) and 13 % to the east coast of Australia (Hartmann,

1981; Yassini and Jones, 1995). Links with Indonesian waters are strong, with approximately 43% of identified species reported including 29 from the Java Sea

(Kingma, 1948; Dewi, 1997; 2000), 23 from the Sunda Shelf and Singapore platform (Mostafawi, 1992), 17 species from the Malacca Strait (Whatley and

Zhao, 1987; 1988) and 17 from the southeast Malay Peninsula (Zhao and

Whatley, 1989b). Affinities also exist with fauna from the South China Sea, the north Indian Ocean and the southwestern Pacific (e.g. Zhao et al., 1985; Zhao and

Wang, 1988; Ikeya and Cronin, 1993; Bonaduce et al., 1976, 1980; Benson, 1964;

Hornibrook, 1952; Swanson, 1979). Although not identified to species level, 5 of

201

the non-marine ostracods found in the core samples have also been recovered from freshwater sites from northwestern Australia (McKenzie, 1966). Two species of the genus Hemikrithe, characteristic of the tropical waters of the Java and South China Seas, have been identified for the first time outside of this immediate region.

5. Ostracod Analyses

203

5. Ostracod Analyses

5.4.2 Ostracod assemblages from core MD-32

Ostracod assemblages have been assigned to commonly co-existing species throughout core MD-32, identified from samples taken at 10 cm intervals9. In the following text the abundance of ostracods is described as dominant (>20%), abundant (>10%), common (>5%), present (<5%, present in all samples of the assemblage), rare (present in only some samples), very rare (present in few samples), with respect to the relative standardised number of valves per gram of dry sediment. Species that are statistically insignificant, but indicative of the depositional environment, have been included and will be referenced accordingly.

All taxa mentioned are as adult forms; juveniles are noted where present.

Figure 5.13 provides a plot of the ostracods noted throughout the core material, the details for which are tabulated in App. 8. The graphs indicate the number of valves per gram of dry sediment extracted from the core samples; adult and juvenile valves are not distinguished. Mostly the ostracods are presented at genus level. Exceptions to this are Leptocythere hartmanni and L. hartmanni lacustris, and Venericythere darwini and V. papuensis which have been considered as individual species as they are abundant in some samples and occur in distinct environments. In addition, some genera have been grouped; Neocytheretta, which includes both Neocytheretta and the similar genus, Alocopocythere; Polycope, which includes Eupolycope and Pontocypris, which includes Pontocypria. Note that genera represented by less than two valves or occurring in less than two samples, and samples barren of ostracods have not been included.

9 A 'sample' refers to the >63μm fraction of dry sediment, utilised for micropalaeontological examination 205

Special mention has been made of the morphological variations noted through the core, particularly of the Leptocythere taxa (Fig. 5.12). In some samples, observation of the sieve pores of a selection of Cyprideis valves has been made under SEM (Fig. 5.14). Typically observations were made on three randomly chosen adult valves from each sample, with all clearly visible sieve pores counted on each valve considered. These data provide further information enabling a more detailed palaeoenvironmental diagnosis.

As outlined in the previous chapter, the basal unit of core MD-32 (14.8-14.0 m) contains no microfauna, but abundant quartz. The first occurrence of ostracods within the core samples is at 13.9 m. The sample is dominated by two species of the genus Cytherella: C. semitalis and C. cf. hemipuncta, comprising 41% and

19% of the assemblage respectively. At least five instars, from adult to A-4 of

C. semitalis may be identified. Both Praemunita broomensis and

Neomonoceratina bataviana are common species, as is the genus Neocytheretta, represented by three species, N. adunca, N. spongiosa and N. vandijki. A broad diversity of other species occurs as minor constituents. These include Hemikrithe sp., Bishopina spinulosa, Cytherelloidea malaccaensis and Phlyctenophora zealandica present, and rare Stigmatocythere indica, Labutisella quadrata and

Callistocythere warnei. All of the ostracod valves show a high degree of preservation.

The ostracodal assemblage outlined above is common to samples from 13.9 m to

13.1 m. The first two samples of this unit are large, with a coarse fraction weight approaching 2 g and species abundance of 62 and 120 valves per gram of

5. Ostracod Analyses

sediment respectively. The rest of the samples have an average coarse fraction weight of 0.3 g and abundance of 14 valves per gram of dry sediment. Cytherella semitalis remains the dominant species, representing 30% of the fauna on average, throughout this unit. At least 4 instars, including adults, are present. Both

Cytherella cf. hemipuncta and Neomonoceratina bataviana are common, joined by Hemikrithe sp.1, which is present as both adult and A-1 valves from 13.8 m.

Other species present include Praemunita broomensis, Phlyctenophora zealandica, Parakrithella pseudornata from 13.7 m and Venericythere papuensis, although more commonly as juvenile forms, from 13.5 m. The genus

Neocytheretta is represented in most samples, but each species only rarely;

Neocytheretta adunca occurs in all samples in low percentages. Other rare species of note include Praemunita sp. and Pseudopsammocythere cf. reniformis, both present in five of the samples of this unit, and Mediocytherideis (Sylvestra) jellineki, which occurs in samples at 13.2 and 13.3 m. Broken mollusc shell is a prominent element of this unit and pyrite becomes increasingly common, however the preservation of most of the ostracods is very good.

A shift in character and fauna occurs in the samples from 13.0 m to 12.0 m. The average sample size is reduced to 0.2 g, with samples at 12.7 and 12.2 m weighing less than 0.1 g. The average number of valves per gram of dry sediment however is increased to 33, ostracods representing a greater proportion of the coarse fraction. Less prolific samples occur at 12.9 m to 12.7 m and 12.4 m, these averaging less than 12 ostracod valves per gram of dry sediment. Loxoconcha judithae is by far the most dominant species, representing an average of 61% of the assemblage and including at least five instars, where present in these samples

207

from 12.6 m. Common species include Venericythere papuensis, both adult and

A-1 valves, Parakrithella pseudornata and Praemunita broomensis, often as whole carapaces. The latter two taxa are more abundant in the lower samples.

Prominent species in the previous assemblage, including Cytherella semitalis and

Neomonoceratina bataviana are also common at the base of the unit, many showing signs of reworking. Similarly, Hemikrithe sp.1, Phlyctenophora zealandica and Mediocytherideis (Sylvestra) jellineki are only present in the lowermost samples. Pseudopsammocythere cf. reniformis is present throughout.

Other species that occur sporadically, but are worthy of comment include

Praemunita sp., Neocytheretta adunca and Copytus posterosulcus. Within this unit, the sample at 12.9 m is rather enigmatic. Although only small and containing fewer than 8 valves per gram of dry sediment, the assemblage is conspicuous for both the absence of the dominant species, Loxoconcha judithae and the presence of a number of rare species, including Eupolycope sp.,

Argilloecia cf. lunata, Stigmatocythere indica, Pistocythereis bradyi and

Microcytherura cf. punctaella, which are not present in the other samples of the unit. Generally, these samples show a high degree of preservation of ostracods, with the exceptions outlined above. Whilst the occurrence of Loxoconcha judithae marks a change in the assemblage, many of the species characteristic of the previous assemblage are present, especially in the lower samples, and are gradually phased out.

Perhaps best described as a subset of the above unit, the samples from 11.9 m to

11.5 m share many common species with the previous assemblage. The samples of this subunit contain a larger coarse fraction, weighing on average 0.4 g,

5. Ostracod Analyses

including over 40 valves per gram of dry sediment. Although the basic assemblage of the previous unit continues here, many new species are added.

Xestoleberis darwinensis is common and Cytheropteron wrighti and

Pterygocythereis velivola are present from 11.9 m, joined by Foveoleberis cypraeoides at 11.8 m. From 11.6 m, the genus Neocytheretta becomes more prominent, represented by a number of species including N. adunca, N, vandijki and the related Alocopocythere goujoni. Other rare species include

Callistocythere warnei and Paijenborchella solitaria. Although Loxoconcha judithae remains a dominant component of the assemblage, its relative abundance decreases through this subunit. All of the valves present are well-preserved.

A shift in assemblage is indicated by the following two samples, with the occurrence of Neonesidea australis and Paranesidea onslowensis, representing

10% and 7% of the ostracod fauna respectively at 11.4 m. Although a continuation of the above assemblage, the sample at 11.3 m is very diverse and contains abundant fauna, with 205 valves per gram of dry sediment. The assemblage at this level is dominated by Xestoleberis darwinensis, Neonesidea australis and Paranesidea onslowensis; the latter two being represented by four instars including adult. A large number of broken ostracods are present in the sample at 11.4 m, whilst those at 11.3 m show better preservation.

From 11.2 m to 10.7 m the samples are smaller and ostracods less abundant, weighing 0.07 g and containing 13 valves per gram of dry sediment on average.

Only three single ostracod valves were extracted from the sample at 10.9 m. The basic assemblage of these samples comprises abundant Neonesidea australis,

209

Cytheropteron wrighti, Xestoleberis darwinensis and Pseudopsammocythere cf. reniformis. Multiple instars of Neonesidea australis are present in most samples.

Common species in the assemblage include Cytherella semitalis and Praemunita broomensis, with Praemunita sp., Loxoconcha judithae, Neomonoceratina bataviana, Paijenborchella solitaria and Neocytheretta spp. also present. Several other species occur only rarely or very rarely and will not be listed here. There are comparatively few adults in these samples and many of the foraminifer,

Ammonia present are dwarfed and have a frosty appearance. The ostracods present are in better condition, although some have pyrite infilling whole carapaces, such as at 11.2 m. Neonesidea australis valves taken from 11.1 m show algal borings as viewed under SEM.

Although the assemblage remains broadly the same from 10.6 m to 10.2 m, the preservation and abundance of the sample is improved. The samples themselves are larger, with an average of 0.5 g of dry sediment in the coarse fraction, containing 65 valves per gram. The species Paranesidea onslowensis and

Pterygocythereis velivola are common in the samples, in addition to those listed for the above unit. Other species present include Neocytheretta adunca,

Parakrithella pseudornata, Callistocythere warnei, Cytherelloidea malaccaensis and Foveoleberis cypraeoides. Once again there are many additional species that occur only rarely or very rarely in the assemblage such as Pistocythereis cribriformis, Pistocythereis bradyi, Labutisella quadrata and Labutisella darwini.

On the whole, the samples show a much better level of preservation than those of the unit described above. Up to five instars of both Neonesidea australis and

Paranesidea onslowensis are present. SEM images obtained of Neonesidea

5. Ostracod Analyses

australis reveal that hairs have been preserved on many of the valves. Species of the genus Praemunita are commonly present as whole carapaces.

The samples from 10.1 m to 9.4 m also display a similar assemblage, although with dwindling representatives. The sample sizes are relatively large, weighing on average 0.5 g, however the abundance of ostracods has dropped significantly, represented by a mean of 24 valves per gram of dry sediment. Both Paranesidea onslowensis and Neonesidea australis are common, exhibiting various instars, from adult to at least A-2. The genera Neocytheretta and Praemunita are also common, represented by several species. Other species present include

Pterygocythereis velivola, Pseudopsammocythere cf. reniformis, Cytheropteron wrighti and Xestoleberis darwini. A further 23 species occur only rarely in one to three samples. Both the degree of preservation and the diversity of species present decrease through this subunit. Valves with evidence of bleaching are common and indicate possible reworking from 10.1 m to 9.7 m.

A dramatic change in faunal composition may be noted at 9.3 m. The assemblage is dominated by Cyprideis australiensis, although mostly juvenile valves are present in this sample. Leptocythere hartmanni is the other major constituent of the assemblage, representing 33%; many of the valves present are heavily calcified and show coarse reticulations. Other species present in this sample include Callistocythere warnei, Praemunita broomensis, Cytherois sp. and

Ilyocypris australiensis. Broken shell material is abundant, mostly blackened and glossy with the effect of pyritisation.

211

The samples from 9.3 m to 9.0 m form a clearly distinct unit. These are relatively large samples, averaging 0.93 g and 78 valves per gram of dry sediment.

The dominance of Cyprideis australiensis continues from 9.2-9.0 m, representing at least 99% of the assemblage. All instars of Cyprideis australiensis are present and show good preservation, although most of the valves are quite small and some have been pyritised. Leptocythere hartmanni and Darwinula sp. occur very rarely. Broken shell material is common throughout this unit, particularly at

9.0 m, where a shell layer is identified on the surface of the core. The shell hash comprises abundant bivalves and gastropods, mostly bleached and 'pinkish' in colour with blackened specks of pyrite coating.

From the point of view of ostracodal assemblages, the samples from 8.9 m to

5.7 m may be regarded as one broad unit, subdivided on the basis of sample size and species abundance. The meagre fauna is considered similar throughout. The sample sizes are all very small; from 8.9 m to 8.1 m the average sample is 0.01 g, containing only 1-4 ostracod valves per sample where present. Cyprideis australiensis is commonly present, mostly as juveniles with evidence of dissolution. Poorly preserved Leptocythere hartmanni occur in three of these samples. Both 8.3 m and 8.2 m contain single valves of Neomonoceratina bataviana, showing evidence of reworking. The sample at 8.6 m is barren of ostracods. Pyrite is common to these samples and gypsum occurs rarely.

Dwarfed Ammonia foraminifers are present in most samples and frosted in appearance, indicating recrystallisation.

5. Ostracod Analyses

From 7.9 m to 7.0 m, most of the samples are barren of ostracods. Rare juvenile

Cyprideis australiensis occur at 7.6 m, 7.5 m and 7.1 m and very rare

Leptocythere hartmanni are present at 7.7 m and 7.6 m, all of which are poorly preserved. The microfauna that may be found are frosty in appearance and gypsum is common in the samples.

More diverse and abundant ostracods are found in the samples at 6.9 m and 6.7 m, containing 19 and 60 valves respectively. The assemblage at 6.9 m comprises frosted Leptocythere hartmanni, Hemikrithe sp. and rare juveniles of Cyprideis australiensis and Cytherella semitalis. At 6.7 m, Cyprideis australiensis is dominant, with minor Leptocythere hartmanni, Neomonoceratina bataviana and

Cytherella semitalis. Charophyte fragments are also present in this sample. The intermediary sample at 6.8 m is very small and comprises only 3 broken Cyprideis australiensis and one very heavily calcified Leptocythere hartmanni valve.

The samples from 6.6 m to 5.7 m are more considerable in size, largely due to the presence of quartz and carbonate concretions. Sample weights average 0.16 g, although the abundance of ostracods is very low, with most samples barren of microfauna. The entire ostracod constituent of this subunit comprises two broken

Cyprideis australiensis valves at 6.4 m, three reworked and heavily calcified valves of Neomonoceratina bataviana at 5.9 m, and one of each Ilyocypris australiensis and Leptocythere hartmanni, both reworked and frosted, at 5.7 m.

Rare foraminifers are present in some of these samples, but they too show poor preservation.

213

In contrast to the above unit, the samples from 5.6 m to 4.8 m are large, weighing on average 0.37 g, and contain abundant ostracods. The sample at 5.6 m is the most densely represented, with 291 valves per gram of dry sediment. Cyprideis australiensis is again the dominant species, typically representing 61% of the assemblage. All instars are present in most samples, although the adults small in size and are commonly bleached and chalky in appearance.

A selection of these Cyprideis valves was viewed under SEM to allow observation of the sieve pore shape (Fig. 5.14). The valves from 5.6 m to 5.4 m exhibited a high percentage of elongate pores and significant etching. The valves from 5.3 m to 4.9 m however, show dominantly rounded sieve pores, heavier calcification and better preservation. A combination of rounded, elongate and some irregular pores were observed in one valve from 4.8 m. The two morphotypes of Leptocythere are present in the assemblage. L. hartmanni is dominant and generally occurs as large and heavily calcified valves (Fig. 5.12). As viewed under SEM, the adult valves at 4.9 m appear smaller and more finely reticulate. The second form,

L. hartmanni lacustris, is present from 5.3-4.9 m and occurs as well-preserved valves with fine reticulation. The various morphologies of this species through the core are shown in Fig. 5.12.

A shift in the faunal assemblage is evident from 4.7 m to 3.8 m. These samples are dominated by Venericythere darwini and Leptocythere hartmanni. The relative abundance of these two species shifts from 30:70% between 4.7 m and

4.5 m to 70:30% between 4.1 m and 3.8 m, the intermediary samples showing equal abundance. In the first two samples, Venericythere darwini occurs only as

5. Ostracod Analyses

juveniles. These samples also contain some poorly preserved juvenile Cyprideis australiensis valves. From 4.5 m, a life assemblage of Venericythere darwini is present. The preservation of ostracods in most samples is good, although there is some evidence of dissolution at 4.3 m and 3.8 m.

The Leptocythere hartmanni valves present are small and very heavily calcified, particularly in the samples from 4.2 m to 4.0 m (Fig. 5.12). Leptocythere hartmanni lacustris has been identified as a minor constituent of the samples from

4.1 m to 3.8 m. These too are dwarfed and more coarsely reticulated, rendering differentiation between the two morphotypes difficult.

Valves of Neomonoceratina bataviana and Phlyctenophora zealandica have rarely been recovered from 3.9 m and 3.8 m respectively. Compared to the previous unit, the samples are small, averaging just 0.04 g. The abundance of ostracods is increased to around 110 valves per gram of dry sediment. Although iron-oxide staining of the sediment is still present to 4.0 m, the ostracod valves are generally well-preserved, with many samples containing numerous whole carapaces.

The samples from 3.7 m to 0.4 m show broad similarities, but may be divided into three subgroups on the basis of the faunal assemblage and abundance of ostracods.

The first subgroup may be defined for the samples from 3.7 m to 2.8 m. Sample sizes are relatively small, weighing around 0.03 g, with an average of 14 valves per gram of dry sediment. Cyprideis australiensis returns as the dominant species, representing over 90% of the assemblage. Most of the valves present are

215

juveniles; adults present at 3.4 m are small in size and show evidence of minor dissolution. Sieve pore observation of a Cyprideis valve at 3.5 m, reveals a high percentage of elongate and some irregular pores (Fig. 5.14).

Common to most of the samples is Ilyocypris australiensis, with most valves well-preserved adult or A-1 instars. Other species occur only rarely through this subunit and include Leptocythere hartmanni lacustris at 3.7 m and 3.5 m, and

L. hartmanni and Venericythere darwini at 3.3 m (Fig. 5.12).

In comparison to the previous subunit, the samples from 2.7 m to 1.2 m are considerably more abundant, bearing an average of over 50 valves per gram of dry sediment. The samples are a similar small weight, excepting at 1.5-1.4 m, where a shell layer is clearly evident. These two samples each weigh in excess of 0.2 g.

The broad assemblage is also similar to the previously described subunit. The dominance of Cyprideis australiensis is clearly evident in these samples, representing an average 80% of the assemblage. Most samples are composed of well-preserved valves of all instars. Adults are large and well calcified.

Exceptions to this include the first sample of the subunit, at 2.7 m with frosted valves and the samples at 2.6 m and 1.2 m, which are very small and contain only juvenile valves.

Sieve pore observation of Cyprideis australiensis valves through these samples reveals a dominance of rounded pores and a significant percentage of elongate pores, with rare irregular pores at 2.7 and 2.4 m (Fig. 5.14). Leptocythere hartmanni lacustris is abundant in most samples, although absent from

5. Ostracod Analyses

2.6 and 2.3 m (Fig. 5.12). The valves present are all full-sized and many display heavy reticulation to 1.8 m.

Ilyocypris australiensis is a common species, occurring in all but three samples

(at 2.6 m, 1.3 m and 1.2 m) generally as well-preserved adult or A-1 valves.

Some valves show evidence of minor recalcification within the vestibulum. Two other 'rogue' ostracods were identified in this subunit. These are four valves of

Rhombobythere alata at 2.5 m and a single reworked valve of Pterygocythereis velivola at 1.7 m. These appear to be isolated occurrences and not characteristic of the unit. In addition, a single valve of Zonocypretta sp. and one of

Candonocypris cf. novaezelandiae were identified at 1.9 m and 1.8 m respectively. Broken fragments of similar ostracods are present in some of the samples above these, however they could not be positively identified. Generally, the samples within this subunit show a high degree of preservation.

The subunit from 1.1 m to 0.4 m is distinguished by the diversity of species present. The samples are heavier than those of the previous subunit, weighing

0.17 g on average, however the ostracods are not as abundant, returning around 20 valves per gram of dry sediment. At 1.1 m, 0.8 m, 0.5 m and 0.4 m Ilyocypris australiensis is the dominant species and comprises well-preserved adult and A-1 valves. Cyprideis australiensis is the other abundant species, dominating the samples at 1.0 m, 0.9 m, 0.7 m and 0.6 m, although mostly juvenile valves are present, many with calcareous infilling. Where present, adult valves are large and well calcified. These species combined represent almost 90% of the assemblage of this subunit.

217

Where present, Leptocythere hartmanni lacustris represents around 10% of the assemblage, commonly as finely reticulated valves (Fig. 5.12). This species is absent from samples at 1.1 m, 0.5 m and 0.4 m. The rest of the assemblage is comprised of a number of species in low abundance. These include Zonocypretta sp., Cyprinotus cf. cingalensis and Limnocythere sp., which are rare and occur in at least half of the samples, and Microcytherura cf. punctaella, Candonocypris cf. novaezelandiae, Cypretta sp. and Darwinula sp., which are very rare in isolated samples. There are several broken valves in some samples that cannot be positively identified.

A clearly distinguishable change in assemblage occurs from 0.3 m and extends to the top of the core. The average sample weight is much larger, exceeding 2 g, containing around 45 ostracod valves per gram of sediment, with the exception of the sample at 0.2 m. The samples are very large in comparison to those of the rest of the core. The assemblage contains a broad variety of ostracods, many present in only very low numbers. The dominant species is Paranesidea onslowensis, which represents on average 18% of the assemblage. Most samples contain instars from adult to A-4. Other common species include Neonesidea australis, with several instars, Neomonoceratina bataviana and Venericythere papuensis.

The genus Neocytheretta is also significantly represented in the uppermost unit by the species N. adunca, N. spinobifurcata, N. spongiosa, N. cornea and

Alocopocythere goujoni, listed in order of abundance. Species that comprise greater than 2% of the assemblage, where present include Cytherella semitalis,

Xestoleberis darwinensis, Pseudopsammocythere cf. reniformis, Cytherelloidea

5. Ostracod Analyses

malaccaensis, Paijenborchella solitaria and Cytheropteron wrighti. Other species of note include Hemikrithe sp., represented by only two valves in the uppermost sample and Polycope favus, Polycope sp. and Eupolycope sp., which are present as only single valves in the sample at 0.1 m.

Several species from the lower unit have been found in the assemblage. These include adult valves of Ilyocypris australiensis present in all of the samples, and isolated occurrences of Cyprideis australiensis, Candonocypris cf. novaezelandiae, and Cypretta sp., although all show signs of reworking.

219

5. Ostracod Analyses

221

5. Ostracod Analyses

5.5 Cluster analysis

Cluster analysis has been performed on the above data to compare the composition of ostracod assemblages (R-mode) and their distribution through the core material of MD-32 (Q-mode) using the program CORRMAT/PROG (Jones and Facer, 1981). A data matrix was constructed, comprising the number of valves per gram of dry sediment of each ostracod genus within each core sample, taken a 10 cm intervals. Adults and juveniles have not been differentiated. To simplify the matrix, the same restrictions with regards genus abundance and representation as were defined for the genus plot (Fig. 5.15, section 5.6.2 for explanation) have been applied here. Similarly, samples barren of ostracods (i.e. samples in the depth ranges 5.7-6.6 m, 7.0-8.1 m and 14.0-14.8 m) have been eliminated from the analysis. In total the matrix comprises 37 genera (or groupings) from 117 samples. A further restriction was imposed on the data set, whereby the value 0.01 was appointed for cases in which a genus was not identified in a sample, but was found in similar assemblages at other depths, and the value 0 was assigned for cases in which a genus was not identified in a sample and was not found in similar assemblages at other depths. This data forcing was imposed to compensate for the large number of "zeros" in the data set. The analysis was performed by calculating the similarity matrix based on relative species abundance for each depth sample. The measurement of the similarities is determined using the cosine-theta similarity coefficient for the Q-mode analysis and a Pearson product moment correlation coefficient for the R-mode analysis.

223

5.5.1 R-mode cluster analysis

R-mode cluster analysis groups fauna found to commonly co-exist within the depth samples. In this study, six bio-groups have been identified (Fig. 5.15), numbered in order of the closest correlation between the groups. The first group comprises four closely correlated genera: Cytherella, Cytherelloidea, Hemikrithe and Phlyctenophora. These taxa are abundant in shallow, phytal, open marine environments. They are also characteristic of tropical climates, Hemikrithe being generally regarded as endemic to Southeast Asia. Whilst these types are common also to the open sections of the modern gulf, they do not comprise a dominant constituent of the assemblages.

The second group consists of 13 genera. The dominant taxa include the bairdiids

Paranesidea and Neonesidea, Pectocytherinids Labutisella and Keijia, the phytal taxon Foveoleberis and the Neocytheretta species. Most of the fauna present are common to open marine environments from the inner to middle shelf. This is the largest grouping represented in this cluster analysis and is indicative of the diversity of fauna in shallow tropical seas. This group is common to the open waters of the modern gulf.

The third grouping comprises eight genera. The dominant taxa in this set are

Xestoleberis, Cytheropteron and Praemunita, all of which are common to more marginal environments with abundant algae and sea-grasses. There is some correlation between this and the previous grouping, with the genera

Pterygocythereis, Xestoleberis, Neonesidea, Neocytheretta, Cytheropteron and

Javanella showing similarities above the seventieth percentile. This subset is

5. Ostracod Analyses

characteristic of the inner shelf environment and is common in such settings in the modern gulf.

The fourth grouping comprises four genera. This cluster is perhaps skewed by the dominance of Loxoconcha, which occurs in large numbers only with the taxa common to this group: Venericythere papuensis, Parakrithella and Caudites.

These other ostracods show greater affinities with other groups: Venericythere papuensis and Caudites to group 3 and Parakrithella to group 2. Loxoconcha is most commonly found in shallow marine, algal environments. It is often the dominant taxa in sandy substrates, such as the channel mouth regions of the modern gulf.

The fifth grouping comprises 6 taxa common to non-marine environments. This group may be further divided into two subgroups. The first includes Leptocythere hartmanni lacustris, Ilyocypris and Cyprideis, which occur in oligohaline to saline waters. The grouping has been identified in the tidal Fitzmaurice Creek, 21 km from the mouth and as part of a relict deposit near the centre of the modern gulf

(Yassini et al., 1993). The second contains the taxa Limnocythere, Cyprinotus and

Zonocypretta, which occur in fresh and low salinity water (<5 ‰). The latter grouping has no known modern analogue in the gulf region.

The final grouping consists of only two taxa, Venericythere darwini and

Leptocythere hartmanni. These taxa are common to shallow and restricted marine waters and may occur in large 'blooms'. The two taxa cluster in isolation from the other ostracods. This is primarily due to their dominance in the samples from 4.7

225

m to 3.8 m within the core. Both of these species have been found in other assemblages, although this does not show in the cluster analysis. These taxa are common to the shallower regions of the modern gulf.

5.5.2 Q-mode cluster analysis

Q-mode cluster analysis correlates depth samples with similar faunal assemblages, with the inference that they were deposited under similar conditions. In this study, six broad groupings have been identified (Fig. 5.16). These may be correlated with the bio-groups defined by the R-mode analysis.

The first group (labelled "2" in Fig. 5.16) clusters two sections of the core.

Included within the group are the uppermost samples, from core top to 0.3 m and samples from 9.4 m to 10.1 m (with the exception of 9.7 m), 10.4 m and 10.9 m.

The bairdiids Paranesidea and Neonesidea, with Neocytheretta present

(bio-group 2), dominate the assemblage common to these samples. This association indicates that conditions at the top of the core were previously witnessed at the deposition of the sediment around 10 m in depth and not at any other time in the recovered material.

The second group (labelled "3" in Fig. 5.16) is somewhat 'inter-fingered' with the previous cluster and includes samples at depths of 9.7 m, 10.2 m, 10.3 m, 10.5 m to 10.8 m and 11.0 m to 11.3 m. The distinction between this and the first cluster is based on the dominance of Xestoleberis, with Neonesidea and Cytheropteron also present in significant abundance. This assemblage is most closely related to bio-group 3.

5. Ostracod Analyses

The third group (labelled "4" in Fig. 5.16) produces a much tighter cluster than either of the above groups. The samples included encompass the depths of 11.4 m to 13.0 m although exclude 12.9 m. The dominance of Loxoconcha defines this grouping, with Parakrithella, Venericythere papuensis and Praemunita, as per bio-group four. This cluster is clearly distinct from samples from the rest of the core.

The fourth group (labelled "1" in Fig. 5.16) is similarly well constrained and distinct. Samples included within this cluster span the depths 12.9 m to 13.9 m although exclude 13.0 m. Cytherella dominates these samples. A sub-group within the cluster has been differentiated for the lower samples within this unit, based on the abundance of the other taxa Cytherelloidea, Hemikrithe and

Phlyctenophora. Together, these are defined as bio-group one.

The fifth grouping (labelled "5" in Fig. 5.16) includes almost half of the samples within the cluster analysis, covering depths of 0.4 m to 3.7 m, 4.8 m to 5.6 m,

6.7 m, 6.8 m and 8.2 m to 9.3 m. The first four samples, from depths of 0.4 m,

0.5 m, 0.8 m and 1.1 m, have been differentiated from the rest of the cluster.

These samples all contain freshwater ostracods, identified as bio-group 5b. The majority of the rest of the cluster is very tightly constrained, with 34 samples displaying more than 99% similarity. This is primarily due to the dominance of

Cyprideis, with Leptocythere hartmanni lacustris and Ilyocypris common to many samples. This assemblage characterises bio-group 5a. Several samples show a

227

more disparate association within the cluster. This is most likely due to isolated valves of species not common to the basic assemblage.

The sixth grouping (labelled "6" in Fig. 5.16) is again tightly constrained and spans the depth samples of 3.8 m to 4.7 m. It contains the samples with the common bio-group 6, dominated by the taxa Venericythere darwini and

Leptocythere hartmanni. This cluster is clearly differentiated from samples from the rest of the core.

Although not included in the cluster analysis, the samples from 5.7 m to 6.6 m,

7.0 m to 8.1 m and 14.0 m to 14.8 m could be considered as a single cluster.

These samples are all barren of ostracods.

5. Ostracod Analyses

Figure 5.15. R-mode cluster analysis of ostracods from core MD-32.

229

Figure 5.16. Q-mode cluster analysis of ostracods from core MD-32.

5. Ostracod Analyses

Table 6. Ecological ranges of ostracod species found in core MD-32, inferred from modern surface samples and previously described distributions (see sections 5.3.1 and 5.3.2). Thin lines indicate the taxa to be present, thick lines, abundant.

231

5. Ostracod Analyses

233

5. Ostracod Analyses

5.6 Discussion and interpretation of the ostracod observations

The assemblages of co-existing ostracods that occur throughout core MD-32 are distinct and correlate largely with variations in the recorded sedimentology of the samples. As such, the same facies subdivisions as were outlined in Section 4.8 will be utilised here. The sequence of assemblage variation may be related to changes in the ecology and climatology of the region. Ecological comparisons are made with common species identified in the surrounding region, as shown in

Table 6. Figure 5.17 depicts the distribution of ecological associations ascribed to each of the units and subunits determined for the core as defined by the ostracod species present. Palaeoenvironmental reconstruction may be extrapolated from the combined information outlined above.

As previously discussed, the basal unit of core MD-32 (14.84-13.95 m) is barren of microfauna, the coarse fraction comprising abundant quartz and mica within a matrix of iron-oxide-rich clays. The lack of ostracods indicates an environment of too high energy, exposed or under-saturated with respect to calcium carbonate either to support the life of ostracods or preserve the valves. The unit is considered to have been deposited by periodic flooding within a restricted basin with exposed margins. This is associated with increased fluvial activity, coincident with rising sea level within a restricted basin, with exposed margins.

The decrease in iron-oxide-staining of the sediment suggests an amelioration of conditions toward the top of the unit.

The ostracodal assemblages throughout unit 6 (13.95-9.3 m) represent a fluctuating marine environment. The first evidence of ostracods within the

235

recovered core material occurs at a depth of 13.9 m. The dominant species,

Cytherella semitalis, is represented by at least five instars indicating a life assemblage. The high degree of preservation of the other species present also suggests that the material was living at this locality, rather than being transported.

The basic assemblage of this sample continues through subunit 6f (13.9-13.1 m) and clusters together to form a group unique from the rest of the core. The principal taxa; Cytherella semitalis, Cytherelloidea malaccaensis, Phlyctenophora zealandica and Hemikrithe spp., are characteristic of open shallow marine sea-grass environments. There are strong affinities with similar environments in

Southeast Asia. The association resembles assemblages described from the Java

Sea west of Bawean Island (Dewi, 1997), the Malacca Straits (Whatley and Zhao,

1987, 1988) and The Malay Peninsula (Zhao and Whatley, 1989b).

Taxa common in the modern nearby shallow Pacific waters, such as

Neocytheretta species are poorly represented and the pectocytherinids are absent.

The number of both ostracodal and foraminiferal species broadens through the subunit, with the establishment of stable shallow marine conditions. Based on this evidence, it is might be suggested that only the Arafura Sill and not Torres Strait was breached at this time. However, the presence of planktic forams and pteropods do indicate a greater influence of open marine waters.

The subunit 6e (13.0-11.7 m) is clearly noted by the shift in the ostracodal assemblage to the dominance of Loxoconcha judithae. From 12.6 m, this species is represented by at least five instars, indicating a biocoenosis. The prominent taxa of the assemblage, including Venericythere papuensis, Parakrithella

5. Ostracod Analyses

pseudornata and Praemunita spp. form a tight cluster, distinct from the other assemblages of the core. The coarse fraction sample size is smaller for 6e than the previous subunit and the diversity of species is decreased, however the abundance of ostracods is increased, indicating an environment well suited to the dominant taxa. Representatives of the genus Loxoconcha are most commonly found in marginal marine and inner shelf waters and tidal channels with abundant algal matter and a sandy substrate. Indeed, there is an increase in quartz through this subunit. This assemblage suggests a decline in sea level or restriction of the marine conditions. The sedimentological evidence supports an increase in energy of the environment, in part due to closer proximity to the coast, and increased fluvial activity.

The anomalous assemblage toward the base of this subunit at 12.9 m, with diverse open shallow marine fauna, including planktic foraminifera and no Loxoconcha judithae, implies a brief resurgence of marine waters at this time. It is most likely that the material within this sample has been transported as many of the microfauna have broken valves. The presence of poorly preserved valves of the species such as Cytherella semitalis and Neomonoceratina bataviana however, common to subunit 6f, may have been reworked from the underlying sediment.

Although the genus Loxoconcha is cosmopolitan, the species Loxoconcha judithae has thus far only been described from northern Australian waters (Howe and

McKenzie, 1989; Yassini et al., 1993). This is a prolific and opportunistic genus, showing a high degree of endemicity at species level. Both Parakrithella pseudornata and Venericythere papuensis have been identified throughout

237

Southeast Asian shallow marine waters (e.g. Whatley and Zhao, 1987, 1988; Zhao and Whatley, 1989b; Mostofawi, 1992; Dewi, 1997).

A gradual change in the composition and the character of the samples occurs from the underlying unit and through subunit 6d (11.7-11.3 m). The coarse fraction percentage has increased, largely due to an improved quantity of microfauna. The basic assemblage of the samples is the same, with new fauna introduced.

Although Loxoconcha judithae dominates to 11.5 m, the other species, including

Xestoleberis darwinensis, Pterygocythereis velivola and Neocytheretta spp., suggest a more hospitable environment in a shallow marine setting. These species are common to the waters around northern Australia, having been described from the nearshore of the modern Gulf of Carpentaria (Yassini et al., 1993), Darwin

Harbour (Howe and McKenzie, 1989) and the Joseph Bonaparte Gulf

(Clarke et al., 2001). The higher degree of preservation, along with a decrease in quartz content and an increase in the amount of pyrite in the samples, particularly infilling carapaces, support a lower energy environment, more distal from fluvial influence.

The introduction of bairdiids to the assemblage from 11.4 m indicates a stronger marine influence. The first sample contains numerous broken valves, suggesting a higher energy environment or possible transportation. By 11.3 m, the taxa are well established, with both Paranesidea onslowensis and Neonesidea australis represented by numerous instars. As bairdiids require full marine conditions to reproduce, a deepening of water is evident. Both Neonesidea australis and

Paranesidea onslowensis have been recorded in the modern Gulf of Carpentaria,

5. Ostracod Analyses

the latter only in the sample taken from nearest the Arafura Sill (Yassini et al.,

1993). Whereas Paranesidea onslowensis has only been described from northwestern Australia (Hartmann, 1978; Howe and McKenzie, 1989),

Neonesidea australis is known also from the east coast (Yassini and Jones, 1995).

The abundance and diversity of the assemblage, dominated by Xestoleberis darwinensis, suggests a productive, phytal, open shallow marine environment.

Based on ostracodal evidence, it is also possible that the Torres Strait was breached by this stage, or at least sea level was well above sill height. The planktic foraminifer and pteropods that were present in subunit 6f have not been identified from these samples, suggesting that sea level was not restored to its former highstand.

A restriction of marine conditions is noted in the samples from 11.2 m to 10.7 m, comprising subunit 6c. Although the assemblage is diverse, the abundance of ostracods, and indeed the size of the coarse fraction are reduced. The fauna represents a progression of the underlying subunit, dominated by Neonesidea australis, Cytheropteron wrighti and Xestoleberis darwinensis. Most of the species represented by this subunit are endemic to Australia. Loxoconcha and

Xestoleberis, in particular, are characteristic taxa of phytal, shallow marine environments. It is noteworthy that Paranesidea is absent from these samples.

This genus requires normal marine salinity to subsist, suggesting that this environment may be restricted or influenced by meteoric waters and hence unsuitable for this taxon. The low percentage of adult valves and the abundance of dwarfed and frosted tests of the foraminifer Ammonia, suggest inhospitable or altered conditions. The numerous pyrite-filled valves and algal borings found on

239

the valves of some Neonesidea australis indicate a low energy environment with abundant organic matter and dissolution. The precipitation and preservation of gypseous laminae within the sediment further support this to be a restricted and low energy, evaporitic environment. The inter-layering of gypsum with predominantly silty-clay sediment suggests periodic flooding of marine waters in a restricted lagoonal environment, with subsequent stagnation and drying.

The improved diversity, abundance and preservation of ostracods indicate more favourable conditions through the deposition of subunit 6b (10.6-10.2 m). The basic assemblage remains the same as that of the underlying subunit, dominated by Xestoleberis darwinensis, Neonesidea australis and Cytheropteron wrighti.

The presence of taxa such as Paranesidea onslowensis, Cytherella semitalis,

Pterygocythereis velivola and Foveoleberis cypraeoides suggest a return to a more open shallow marine environment. The occurrence of several instars of the bairdiids, and the observation of hairs preserved on the valves of Neonesidea australis support this to be an in situ assemblage. The fauna contains both local and cosmopolitan elements. Taxa such as Labutisella quadrata and Labutisella darwinensis and Callistocythere warnei are endemic to tropical Australia, having been described from the Great Barrier Reef to the Kimberley (Labutis, 1977;

Howe and McKenzie, 1989; Yassini et al., 1993; Clarke et al., 2001). Other species including Neocytheretta adunca, Parakrithella pseudornata,

Cytherelloidea malaccaensis and Pistocythereis spp. are better known from the shallow seas of Southeast Asia (Dewi, 1997; Whatley and Zhao, 1987, 1988;

Zhao and Whatley, 1989b; Mostofawi, 1992).

5. Ostracod Analyses

Subunit 6a (10.1-9.3 m) may be differentiated on the basis of the dominant ostracod taxa and the degree of diversity and preservation of the microfauna present. Species such as Paranesidea onslowensis, Neocytheretta spp. and

Praemunita spp., which were present in the underlying assemblage, are dominant in these samples, indicating the influence of more open marine conditions.

Conversely, shallow and marginal marine fauna are less abundant. Such an assemblage is found in the samples at the top of the core, unit 1 (0.3-0 m) and in the sample described from nearest the Arafura Sill in the modern gulf

(Yassini et al., 1993). Both Paranesidea onslowensis and Neonesidea australis exhibit three instars, a reduction from the underlying subunit, perhaps caused by transportation of these valves.

A shell layer marks the start of this subunit, which may indicate a shoreline feature. The abundance of broken shell material and quartz throughout, would suggest this to be a more dynamic environment. Compared with the underlying subunit, the samples show a declining level of diversity and degree of preservation. Although the taxa present are characteristic of the middle or inner shelf environment, both bleached and reworked valves are common, suggesting a recession of marine waters with surges transporting material, rather than in situ deposition. In addition, the reduction in the foraminiferal assemblage, with only robust taxa present, also supports the notion of this being a regressive environment. Those tests present are white and frosty in appearance indicating transportation and/or recrystallisation.

241

Unit 5 (9.3-8.95 m) is clearly delineated from the surrounding facies based on the dominance of two ostracods, Cyprideis australiensis and Leptocythere hartmanni and the foraminifer Ammonia. All of these taxa are known to be derived from marine waters, although adapted to marginal and restricted, even lacustrine waters of Na-Cl solute composition (Cann and De Deckker, 1981; Anadón, 1992). The first sample, at 9.3 m, comprises mostly juvenile Cyprideis australiensis and valves of other species, including Callistocythere warnei, Praemunita broomensis and Cytherois sp., all of which may occur in marginal environments. At this time, marine waters must have been clearly cut-off from the main body of the Gulf of

Carpentaria, leaving a stranded waterbody of essentially marine composition. A life assemblage of Cyprideis australiensis is present throughout the rest of the unit, best described as a 'bloom'. Although the presence of Cyprideis australiensis is indicative of marginal or non-marine waters, it is a euryhaline species, so gives little indication of the salinity of the water body. The presence of isolated valves of Ilyocypris australiensis and Darwinula sp. suggest at least proximal or seasonally oligohaline water. The valves of Leptocythere hartmanni are heavily calcified through this unit, suggesting that the water is supersaturated with respect to calcite, however the small size and lightweight of the Cyprideis valves, suggests a carbonate-limited environment. Higher temperature may have prompted more rapid calcification, producing the crude reticulation. Other boundary conditions include the salinity and oxygen content of the water, either of which may have an effect on the life habits of the ostracods and calcification of the carapace.

5. Ostracod Analyses

The large amount of broken shell material in the unit, especially at 9.0 m suggests a shoreline, indicative of the extent of the lake at the time. Small bivalves, gastropods and fish fragments are common, most likely as a concentrated deposit due to transportation. The blackening and shiny nature of the shells is due to pyritisation, suggesting a high amount of organic matter in a reducing, anoxic environment after the death of the organisms. Cyprideis australiensis is known to withstand anoxic conditions (Aladin, 1993; Jahn, et al., 1996; Gamenick et al.,

1996). The environment is evocative of a swamp or brackish lagoon, cut-off from marine influence.

The ostracod assemblages through unit 4 are very poor with few species represented, and many of the samples are even barren of microfauna. Much of the environmental interpretation of these samples has been extrapolated from the sedimentology. Subunit 4d (8.9-8.1m) is characterised by fine-grained sediment and gypseous laminae, indicating a low-energy non-marine environment with episodic flooding and subsequent evaporation. The rare ostracods that are found are predominantly juvenile Cyprideis australiensis valves with signs of abrasion.

No adults have been recovered. This suggests that winnowing has segregated the instars. For this species to be living in this environment, some water must have been maintained, as it is not able to withstand desiccation. Transportation to this depositional environment is favoured. Both Leptocythere hartmanni and

Neomonoceratina bataviana are common taxa in tidal channels. The poor preservation of the rare valves of these two species found in these samples also indicates transportation. The foraminifer Ammonia is common to saline, non- marine settings (Cann and De Deckker, 1981). The frosty and dwarfed tests found

243

in these samples suggest an inhospitable environment and probable post mortem recrystallisation. The abundance of pyrite favours reducing conditions, whereas gypsum indicates oxidising conditions, hence alternation of Eh conditions is interpreted. An organic-rich mudflat environment with periodic flooding and evaporation is inferred.

Subunit 4c (8.0-7.0 m) is segregated on the basis of iron-oxide mottling of the sediment, caused by subaerial exposure and fluctuations in groundwater. This depositional environment is essentially the same as the underlying subunit, however post-depositional exposure indicates drier conditions. Pyrite is common in the samples, generally oxidised, indicating exposure after formation. Most of the samples are barren of ostracods, implying either an even less hospitable environment without permanent water or dissolution of valves during pedogenesis. Those valves that are present are juvenile, frosty and poorly preserved, again suggesting transportation and recrystallisation. Foraminifers, including Ammonia and planktic species, are more common in these samples.

They are thought to have been transported, as the tests show minor abrasion; their lightweight and spherical shape allows them to be carried in suspension over long distances. A continuation of the under-lying unit, though more remote from permanent water is interpreted.

The more diverse and abundant taxa of subunit 4b (6.9-6.7 m) are indicative of permanent water in the environment at the time of deposition. An incursion of waters from the Indian Ocean is indicated in the lowermost samples by the presence of the shallow marine fauna Cytherella semitalis and Hemikrithe sp.,

5. Ostracod Analyses

which are common to subunit 6f. The assemblage at 6.8 m is somewhat diminished, suggesting a regression of marine waters, with only marginal taxa remaining. The heavy calcification of Leptocythere hartmanni implies a high concentration of calcium carbonate within the host water. By 6.7 m, the sample is dominated by Cyprideis australiensis, exhibiting several instars and accompanied by Leptocythere hartmanni and charophyte oogonia fragments, revealing a likely establishment of a permanent non-marine waterbody. The presence of

Neomonoceratina bataviana and Cytherella semitalis valves does suggest some continued channel influence.

The samples from 6.6 m to 5.7 m are considered to be a subunit of unit 4 based on their diminutive fauna, indicating inhospitable conditions for ostracod habitation.

The few ostracods valves that have been found are poorly preserved and all show evidence of reworking or dissolution. The fauna are common to non-marine or marginal settings. The presence of Ilyocypris australiensis at 5.7 m indicates proximity to low salinity water, as this is essentially an oligohaline species. The foraminiferal assemblage is better represented, with reworked tests of robust rotaliids, indicating transportation; from 6.3 m to 6.1 m. Ammonia is abundant and present in a variety of sizes, suggesting an in situ population. This foraminifer is able to withstand a wide range of salinities, moderately energetic and ephemeral waters for short periods. Samples of this subunit have a large coarse-fraction component, dominated by quartz and carbonate concretions. This is considered to have been deposited in a high-energy environment, most probably alluvial, with subsequent subaerial exposure and pedogenesis. The dissolution of the ostracod valves and Ammonia is most likely due to the penetration of meteoric

245

waters through the sediment, with subsequent precipitation of carbonate concretions.

The vast abundance of Cyprideis australiensis of all instars in the samples of subunit 3b (5.6-4.8 m) indicates the presence of a permanent non-marine waterbody. The dominance of Cyprideis australiensis in saline waters of Na-Cl solute composition, suggest that the waterbody derives from a marine origin. The small size of the adult valves may also be indicative of a saline environment. A life assemblage of this species is present, although the adults in the first few samples show evidence of dissolution. This may be a seasonal phenomenon, with minor etching of the adult valves caused by meteoric water.

Saline conditions, becoming fresher above 5.3 m, are suggested by the sieve pore morphology of Cyprideis. The heavily calcified valves of Leptocythere hartmanni in the lower samples also suggest more concentrated conditions, freshening up through the subunit. This is supported by the dominance of

Leptocythere hartmanni lacustris in the higher samples. The shell layer at 5.0 m, with abundant bivalved molluscs and gastropods, indicates a lake shoreline feature. The chalky nature of the shells suggests later exposure. A short-lived brackish waterbody, with later restriction and minor pedogenesis, with permeation of meteoric water is suggested.

Subunit 3a (4.7-3.8 m) is differentiated from 3b on the basis of the dominant taxa.

Venericythere darwini is abundant and well-preserved, with all instars present in most samples, indicating a life assemblage. Adults are absent from the base of the

5. Ostracod Analyses

subunit, but the presence of poorly preserved Cyprideis australiensis valves suggests that these valves may have been transported. Venericythere darwini is a marginal marine species, common to intertidal settings in Southeast Asia. Its presence indicates a renewed contact with marine waters in a low energy environment, such as a lagoon. Small and heavily calcified valves, suggesting a restrictive environment, represent the other prominent ostracod, Leptocythere hartmanni. Whole carapaces of both Venericythere darwini and Leptocythere hartmanni are common, suggesting rapid burial. The species Leptocythere hartmanni lacustris occurs in the upper samples of the subunit, but as dwarfed and coarsely reticulated valves. This may represent the upper salinity tolerance limit for this taxon. The presence of other species, namely Neomonoceratina bataviana and Phlyctenophora zealandica, in the upper samples may indicate an increased channel influence in this restricted environment toward the top of the unit.

The re-establishment of lacustrine conditions is evident through unit 2. Cyprideis australiensis resumes dominance in the assemblage through subunit 2c

(3.7-2.8 m), indicating permanent water. As with subunit 3b, mostly juvenile valves are represented in the lower samples. Adults where present showing evidence of dissolution, which may indicate increased seasonality. The sieve pore analysis and small valve size suggests saline conditions. Whole carapaces infilled with pyrite are common, suggesting rapid burial of the ostracods after death. The abundance of pyrite is representative of low energy, organic-rich conditions.

The introduction of Ilyocypris australiensis as a significant component of the assemblage indicates fresher conditions than those seen before in the core. The

247

absence of many juvenile instars of Ilyocypris australiensis may be a factor of preservation, or indicate something of the life habits of the species. As the species is known to inhabit streams and temporary pools, transportation of adult valves to more permanent water in wetter seasons is suggested. Unfortunately, Cyprideis australiensis and Ilyocypris australiensis have not been found to coexist in the modern environment to allow comparison. The sample at 3.3 m, with both

Leptocythere hartmanni and Venericythere darwini valves suggests a brief reconnection to estuarine water.

Continued lacustrine conditions are evident through subunit 2b (2.7-1.2 m). The increase in the abundance of ostracods and the life assemblage of Cyprideis australiensis with well-calcified valves suggest stable conditions and perhaps a deeper, more extensive lake at this time. The dominance of rounded sieve pores in the Cyprideis valves of these samples indicates fresher water. This is supported by the abundance of other taxa, such as Leptocythere hartmanni lacustris and

Ilyocypris australiensis. Again, the lack of juvenile Ilyocypris australiensis valves and the evidence of minor recalcification in the vestibulum of some of the adult valves indicates transportation of this species from temporary pools to a more established waterbody. The degree of reticulation in the valves of

Leptocythere hartmanni lacustris, from heavily to finely reticulated up through the subunit, may be caused by a decrease in the solute concentration of the waterbody, due to increased freshwater inflow. A deeper lake and possibly cooler conditions, with less evaporation are postulated.

5. Ostracod Analyses

The presence of the ostracods Zonocypretta sp. and Candonocypris cf. novaezealandiae in subunit 2b imply proximity to fresh water. As Cyprideis australiensis is not characteristic of freshwater environments and there are no noded forms (a phenomenon of the Northern Hemisphere species Cyprideis torosa in freshwater environments), it is suggested that these isolated valves of other taxa were transported to the core site in the lake, perhaps from an adjacent fresh pool.

The presence of the so-called 'rogue' ostracods, Rhombobythere alatum and

Pterygocythereis velivola at 2.5 m and 1.7 m respectively, suggests minor influence of estuarine waters. Both of these species have been identified from tidal channels of the modern gulf (Yassini et al., 1993). The presence of reworked rotaliids in the samples from 2.2 m to 1.7 m supports this. This may be concurrent with a contraction of the lake.

The absence of pyrite in these samples could be indicative of a decrease in the organic content of the lake at this level, for example in deeper water, or an increase in the energy of the environment, perhaps by increased surface water.

The abundance of shell material in the assemblage from 1.5-1.1 m, including freshwater bivalves, planospiral molluscs, fish fragments and charophytes also indicates an increase in the energy of the environment and freshwater conditions.

This shell-rich unit has similar radiocarbon dates throughout. An increase in meteoric water input and transportation of material is inferred.

A change in the character of the samples may be noted in subunit 2a (1.1-0.4 m).

The two species, Cyprideis australiensis and Ilyocypris australiensis dominate the assemblage alternately. A periodic variation in conditions is suggested, which

249

may be indicative of drier/wetter conditions respectively. At this stage, the sampling resolution does not allow definition of the periodicity concerned. The diversity and increased abundance of freshwater taxa; including finely reticulate

Leptocythere hartmanni lacustris, Limnocythere, Cypretta and Cyprinotus and charophyte oogonia, indicate this to be the freshest environment recovered in the core material. An increase in alkalinity is also implied (Radke et al., 2003). This

+ 2+ - - grouping of taxa has been described from the Na -Mg -Cl -HCO3 -rich, monsoon deposited waters of the Kimberleys in north-western Australia (Williams and

Buckney, 1976; McKenzie, 1966).

Other ostracod species, such as Microcytherura cf. punctaella and the presence of rare planktic foraminiferal tests, indicate the continued influence of estuarine waters. These may have been transported some distance in suspension before being deposited in the lake. A decrease in the overall abundance of ostracods may be due either to a more variable climate, or may indicate deeper water in this region of the lake at the time of deposition. A shift in the solute content of the host-water, away from Na+, Cl- dominance to a higher alkalinity to chlorine ratio is also inferred by the change in ostracod assemblage (e.g. Radke et al., 2003).

The incursion of marine waters into the lake is apparent in core MD-32 from

0.38 m upward. The samples of the uppermost unit contain a large amount of broken shell material, indicating dynamic conditions. The fauna is diverse and open shallow marine in ecology, with such taxa as bairdiids and the planktic ostracod Polycope spp. Neither planktic foraminifers nor pteropods have been recovered. Planktic foraminifers are very rare in the modern Gulf of Carpentaria,

5. Ostracod Analyses

due to the shallow nature of the basin. Also, the energy of the environment may be too great to preserve transported tests. Many of the taxa, such as Cytherella semitalis, Hemikrithe and Paranesidea onslowensis are Southeast Asian in character. Ostracods such as the Labutisella spp. and Praemunita spp., which are common to the Pacific Ocean, have not been identified in these samples.

This assemblage is most like subunit 6a (10.1-9.4 m); the last identifiably marine unit before sea level dropped below the Arafura Sill. This suggests that the most recent breaching of Torres Strait is not clearly represented in core MD-32. It is likely that the uppermost material of the core was lost on recovery, or is poorly represented due to low sedimentation rates in this region of the gulf in the modern environment. In addition to the marine fauna, valves of Cyprideis australiensis and Ilyocypris australiensis have been identified, obviously reworked from the underlying unit. This phenomenon has been documented by De Deckker et al.

(1988) in cores GC-2 and GC-10A, retrieved from the deepest section of the modern gulf and represents relict material.

Through the detailed analysis of ostracod assemblages of the core and comparison with modern species distribution, inferences have been drawn about the ecology of the gulf basin at the time of valve formation. Morphological variation of the valves also provides information regarding the changing conditions of the waterbody. The preservation of the valves, including dissolution and pyritisation, and the population age structure are indicative of post-depositional effects.

251

6. Stable-Isotope Analyses

δ18O and δ13C values of ostracod valves from core MD-32

Having established a framework of palaeoenvironmental and ecological constraints through sedimentological and ostracod faunal analyses, presented earlier, the geochemistry of the ostracod valves can provide further insight into the ambient conditions at the time of shell formation. The δ18O and δ13C values, in comparison with the palynology and, for the upper lacustrine unit, trace-element and strontium isotope geochemistry, may elucidate variations in temperature, effective precipitation and primary productivity recorded in the core sediment.

6.1 Introduction to stable-isotopes

The stable-isotope analysis of carbonates measures the ratios of 18O to 16O and 13C to 12C within a sample. In nature, oxygen is composed of 99.763% 16O and

0.1995% 18O and carbon is accounted for by approximately 98.89% 12C and

1.11% 13C (Hoefs, 1997). Isotopic fractionation occurs during most natural processes, chiefly because weaker bonds that exist between lighter isotopes break more readily than those of heavier isotopes.

The ratio of two isotopes within a carbonate sample is expressed in the delta (δ) notation, in units of per mil (‰):

3 1. δ(‰) = [(Rsample/Rstandard)-1] x 10 where R = 18O/16O or 13C/12C. With respect to carbonates, the international standard to which samples are compared is V-PDB (Vienna Pee Dee Belemnite) for both δ18O and δ13C. The original reference standard, a belemnite from the Pee

Dee Formation from South Carolina, USA, is now exhausted and has been replaced by other standards. The δ18O and δ13C of V-PDB is defined relative to the isotopic reference material NBS-19 (TS-limestone), with values of -2.2‰ and

6. Stable-Isotope Analyses

+1.95‰ respectively, by the International Atomic Energy Agency (IAEA) in

Vienna.

For the δ18O value of water, the international standard V-SMOW (Vienna

Standard Mean Ocean Water) is commonly used. For calcite, conversion between

V-SMOW and V-PDB is given by:

18 18 2. δ OV-SMOW = 1.03091 δ OV-PDB + 30.91 as modified by Coplen et al. (1983).

The fractionation factor (α) is defined as the ratio of any two isotopes of an element in one chemical compound (x) divided by the corresponding ratio for another chemical compound (y):

3. αxy = Rx/Ry

This ratio in turn varies as a function of temperature and may be expressed in the form of an isotopic distribution coefficient:

3 6 -2 4. 10 ln αxy = B (10 T ) ± A where A and B are constants and T is temperature in Kelvins.

6.2 Oxygen stable-isotope composition

6.2.1 Relationship between δ18O of water and δ18O of inorganically precipitated carbonates

The oxygen isotopic composition of inorganic carbonate precipitated from water is determined primarily by the isotopic composition and temperature of the host water. For low-Mg calcite formed in thermodynamic equilibrium with water, the

253

temperature dependence on the oxygen-isotope fractionation between water and carbonate has the relationship:

o 2 5. T( C) = 16.9 - 4.2(δc-δw) + 0.13(δc-δw)

18 18 where δc = δ Ocarbonate and δw = δ Owater, from which the carbonate was precipitated under laboratory conditions at 25oC, as defined by Craig (1965), modified after Epstein et al. (1953). This may also be expressed in terms of the thermodynamic equilibrium fractionation factor between calcite and water:

3 6 -2 o 6. 10 ln αcalcite-water = 2.78 (10 T ) - 2.89 (0-500 C) as determined by O'Neil et al. (1969). This corresponds to a shift of around

-0.24‰ in δ18O values for carbonate precipitates for every 1oC increase in water temperature.

More recently, a relationship has been identified between pH and δ18O of biogenic

2- calcite. Spero et al. (1997) found a correlation between increasing CO3 concentrations, and lower δ18O values of cultured foraminiferal calcite. Zeebe

(1999) relates the mechanism of the fractionation to be due to the relative

- 2- proportions of HCO3 and CO3 in the host water. Dissolved organic calcite

- occurs predominantly as H2CO3 at low pH, HCO3 at intermediate values and

2- 18 2- CO3 at high pH. The δ O of CO3 has been determined by Usdowski et al.

- (1991) to be lower than that of HCO3 under equilibrium, owing to a smaller

2- fractionation factor between CO3 and water. Essentially, the oxygen isotopic fractionation between water and DIC decreases with increasing pH.

6. Stable-Isotope Analyses

6.2.2 δ18O of meteoric water

The isotopic composition of atmospheric precipitation is geographically and temporally specific, controlled by a series of fractionation stages, including evaporation from the source, condensation and precipitation. After a worldwide survey of monthly δ2H and δ18O values of precipitation, initiated by the IAEA and the World Meteorological Organisation (WMO) in 1961, a series of empirical relationships were observed between the isotopic composition of precipitation and a range of geographical effects (Dansgaard, 1964; Rozanski et al., 1993). It was shown that the isotopic ratio of precipitation at any one locality through time depends on the source of the moisture and the air-mass trajectory. This takes into consideration those factors identified by Dansgaard (1964) of latitude, altitude, continentality, surface air temperature and seasonality. A further factor, referred to as the 'amount effect', describes an inverse relationship between the amount and the isotopic composition of precipitation (Dansgaard, 1964; Rozanski et al.,

1993). This is most significant in areas of monsoonal activity. In regions that experience both summer and winter precipitation, without monsoonal deluge, there may be an observed seasonal offset in the δ18O values, with heavier values in the warmer months.

The δ18O values of meteoric water typically vary from -2‰ (vs. V-SMOW), near the equator to as low as -50‰ (vs. V-SMOW) near the poles (Hays and

Grossman, 1991). A generalised representation of the isotopic composition of precipitation may be described by the linear equation:

7. δ2H = 8 (δ18O) + 10 of Craig (1961) referred to as the Global Meteoric Water Line (Fig. 6.1).

255

Figure 6.1. Global meteoric water line, as defined by δD=8δ18O+10 (Craig, 1961; Hoefs, 1997).

6.2.3 δ18O of seawater

The global ice volume primarily determines the δ18O of deep seawater.

Throughout the last glacial cycle, δ18O of ocean water has only varied in the order of 0±1‰ (v. V-SMOW). Measurement of the δ18O of calcitic tests of benthic foraminifers is a well-established proxy for ice volume (e.g. Shackleton and

Opdyke, 1973; Imbrie et al., 1984, Labeyrie et al., 1987). However, the effect of variations in bottom water temperature on the δ18O of benthic foraminifers needs also to be considered, and can be tested with the aid of other proxies such as

Mg/Ca ratios of marine carbonates (Fairbanks and Matthews, 1978; Chappell and

Shackleton, 1986; Corrège and De Deckker, 1997). Other factors such as the salinity and pH variations may also have an effect on the resultant δ18O recorded in the shell calcite.

6. Stable-Isotope Analyses

6.2.4 δ18O of marginal marine water

The δ18O of water within marginal marine settings is dictated by the mixing of seawater and continental waters, from both surface and groundwater sources, and hence is a refection of salinity more so than temperature. The effects of tides, wind regimes, river and groundwater discharge and surface air temperature need to be considered. The simplest variation in δ18O values would be that of a two end-member mixing model along a salinity gradient with marine and freshwater

(35‰ and 0‰ salinity respectively) representing the two extremes. This is shown to be a linear relationship in estuarine systems with short residence times, such as the Sacramento - San Joaquin River discharge into San Francisco Bay

(Ingram, et al., 1996a,b). Complications arise in restricted settings, such as lagoons, whereby the oxygen isotopic composition of the water is also influenced by the degree of evaporation and concentration of solutes, relative humidity of the atmosphere and the isotopic composition of the water vapour (Swart et al., 1989;

Hendry and Kalin, 1997; Anadón et al., 2002; Swart and Price, 2002). The relationship between δ18O and salinity in these restricted environments is thus non-linear.

6.2.5 δ18O of lake water

Within lakes, local effects specific to the hydrology of the basin influence the

δ18O values of the water. The hydrologic balance is defined by the amount of inflow versus outflow of the basin. Within closed basins (i.e. those with no outflow) the primary variables are the relative degrees of influence of inflowing water and evaporation on the waterbody. Evaporation preferentially removes lighter isotopes, leaving the water enriched in heavier isotopes. The magnitude of

257

the evaporation effect is controlled by the surface air temperature, relative humidity, wind strength and solar radiation (Kelts and Talbot, 1990). This is largely determined by the ratio of the surface area of the lake to that of the catchment, collectively referred to as the 'catchment effect' (Gat and Lister, 1995).

Lakes with large drainage basins in comparison to the lake size are considered to be most suitable for reflecting the δ18O of precipitation (von Grafenstein, 2002).

For open basins (i.e. with surface outflow) both the amount of inflow and the residence time are the critical factors that determine the hydrological balance. For these lakes, temperature has a greater control on the isotopic composition of the water than does evaporation. Where environmental conditions are stable for some time, the lakewater may attain isotopic equilibrium with the atmospheric water vapour. This is not the case for lakes with short residence times or strong seasonality.

The δ18O of the lake water generally becomes more negative with increasing volume, particularly in hydrologically closed basins. This is because river discharge and meteoric input are characteristically isotopically lighter.

Ground-water inflow may be either heavy or light depending on the source and isotopic evolution of the water before entering the basin. As lakes discharge, the effect on the δ18O of the water remaining in the lake is a function of the rate of out-flow. Changes to the hydrologic balance of the lake cause disruptions to the composition of the lake water. Although the transient condition is reflected in the initial response, lasting up to decades, there is a lag between climate forcing and the resultant steady-state δ18O of the water (Benson and Paillet, 2002). This in

6. Stable-Isotope Analyses

turn is dependent on the volume and hydrology of the basin at the time of the perturbation.

For shallow lakes, the effective precipitation and the degree of evaporation are of paramount importance. Variations in air temperature also have a marked effect, with warmer air temperatures resulting in more positive δ18O values of the water

(Schlesinger et al., 1988). It must be noted however, that in settings with strong seasonality, lake waters that are isotopically enriched at the end of the dry season may be overwhelmed by a sudden addition of rainwater. This can cause salts to re-dissolve, leaving the water saline, but relatively isotopically light (e.g. Chivas et al., 1993).

Deep, thermally stratified lakes have the potential to record changes in air temperature. Whereas the surface water may become seasonally enriched with both evaporation and isotopically heavy summer rainfall, the hypolimnion is exempt from these effects and has the potential to reflect long-term air temperature trends in precipitated carbonate (von Grafenstein, 2002). The contribution of groundwater to the overall hydrological balance is again specific to each lake basin. Lakes that are predominantly fed by groundwater reflect the isotopic composition of the water in the recharge zone of the aquifer (Smith et al.,

1997). Those lakes that have partial groundwater input may be identifiable by episodic recharge events, with different isotopic signatures to that of local precipitation in the lake basin.

259

6.3 Stable carbon isotope composition

6.3.1 Relationship between δ13C of water and δ13C of precipitated carbonates

The δ13C of inorganic carbonate precipitated from water depends primarily on the

δ13C of the dissolved inorganic carbon (DIC) of the water, which in turn is a function of the dissolved CO2 in equilibrium with the atmospheric CO2. Calcite is enriched in 13C by 1.0±0.2‰ compared with the hostwater DIC, within the temperature range 10-40oC (Romanek et al., 1992). Other factors that influence the carbon budget of a waterbody include dissolved organic carbon, particulate organic carbon (e.g. plant remains, algae, bacteria), and particulate inorganic carbon, such as authigenic calcite and dolomite. The δ13C values of some important carbon reservoirs are presented in Fig. 6.2.

The dissolved CO2 of the water approaches equilibrium with the atmospheric CO2 reservoir through exchange across the water surface. The δ13C composition of atmospheric CO2 is in the order of -7 to -9‰ and the fractionation factor between bicarbonate and water is around +8 to +10‰ (Friedman and O'Neil, 1977). Hence lake water DIC in equilibrium with the atmosphere has a δ13C of approximately

+1 to +3‰. This equilibrium is only achieved in waterbodies with long residence times.

6. Stable-Isotope Analyses

Figure 6.2. δ13C values of some relevant carbon reservoirs (after Hoefs, 1997).

Figure 6.3. Range of δ13C values of C3, C4, CAM and aquatic plants (after Hoefs, 1997 and McKenzie, 1985).

261

6.3.2 δ13C of plants

The productivity of a waterbody has a dominant influence over the resultant δ13C of the DIC of the water. The process of photosynthesis preferentially fixes the lighter isotope of carbon (12C), leaving the DIC of the water relatively enriched in

13C. The various photosynthetic pathways of plants exhibit different fractionation factors. The distribution of δ13C of various vegetation types is depicted in Figure

6.3. Plants of the Calvin-Benson, or C3 group, such as trees, most shrubs and herbs, and cool season grasses, have a range of δ13C of around -23 to -33‰

(vs. V-PDB), with an average of -26‰. The other major plant group, the

Hatch-Slack or C4 plants, which include tropical grasses, temperate grasses and some herbs and shrubs, are characterized by δ13C ratios of -9 to -16‰

(vs. V-PDB) with an average of -13‰. A third group of plants that exhibit

Crassulacean acid metabolism, known as the CAM plants, has intermediate δ13C, spanning the range of both C3 and C4 plants. This group includes the succulents, such as cacti and yucca. Aquatic plants are most commonly in the range of -20 to

-30‰ (McKenzie, 1985). Therefore it may be difficult to distinguish between C3 and aquatic plant influence, based on the δ13C values.

In stable water-bodies with minimal mixing, stratification of the carbon budget may occur in the water column. As such, the surface waters may be enriched with

13C, due to the preferential removal of 12C through photosynthesis (Kroopknik,

1985). Conversely, bottom-waters may be depleted with respect to 13C, owing to the degradation of organic material. The increased production of CO2 at the sediment-water interface due to biological activity results in decreased pH and

DIC undersaturated with respect to bicarbonate.

6. Stable-Isotope Analyses

6.3.3 δ13C of seawater

The ocean stores approximately 90% of the Earth's active carbon, around 95% of which is in the form of DIC (Hedges and Keil, 1995). The global average of ocean δ13C is around 0‰ (vs. V-PDB), with the majority of surface water around

+1.5 ± 0.8‰ (vs. V-PDB) (Kroopnik, 1985). The most stable δ13C of surface water is within the equatorial Atlantic and Indian Oceans, measuring +1.9‰ and

+1.3‰ (vs. V-PDB) respectively. The equatorial Pacific surface water δ13C varies between +1.5 and +2.0‰ (vs. V-PDB). Variations in δ13C are largely due to productivity. Generally surface waters are isotopically heavy, due to high productivity, however upwelling events bring 13C-depleted waters to the surface.

6.3.4 δ13C of marginal marine water

In marginal marine settings, the δ13C of the DIC of the water is reflective of the relative contribution of marine and non-marine waters, the dominant plant types and CO2 derived from organic matter degradation (Anadón et al., 2002). The total dissolved carbon δ13C of fluvial input is in the order of -6 to -9‰ (vs. V-PDB).

This includes δ13C values of -25 to -30‰ (vs. V-PDB), derived from particulate organic matter. As with oxygen, in an ideal two end-member setting, δ13C would be seen to vary linearly in accordance with salinity, but this is only the case if the

- dissolved HCO3 content of the end-members is the same (Sackett and Moore,

1966).

263

6.3.5 δ13C of non-marine water

Within lakes, the δ13C is driven by the photosynthetic pathways of the dominant plant material of the lake and the surrounding catchment, the decay and recycling of aquatic organic matter and the CO2 exchange rates with the atmosphere.

Bacterial processes and carbonate dissolution may also affect the overall carbon budget of the lake (Kelts and Talbot, 1990). The contribution of river and groundwater to the lake, usually depleted in δ13C, influences the carbon isotopic composition of the water. In deep, stratified lakes, more negative δ13C is transferred from the surface to deep waters by sinking and degradation of organic carbon. However, in shallow lakes, the water column is generally well mixed hence the δ13C of the DIC of the water should be similar throughout. In highly reducing environments, enrichment in δ13C may occur as a result of methanogenesis, which involves bacterial methane production from CO2 reduction in anoxic environments. This imparts a strong fractionation effect, producing methane with very negative δ13C, leaving the residual DIC heavily isotopically-enriched in carbon (Curry et al., 1997). Methanogenesis is common in organic-rich lakes, wetlands and marshes. A sharp gradient in the δ13C profile across the sediment-water interface may be developed.

6.4 Ostracods and stable-isotope analysis

Minerals such as calcite and aragonite are formed within the water column, by either inorganic or biogenic precipitation, upon supersaturation of the waterbody with calcium carbonate. Bulk carbonate samples represent a combination of both authigenic and detrital particles, hence are regarded as less effective for detailed palaeoenvironmental work. Although endogenic inorganic carbonates have been

6. Stable-Isotope Analyses

used in such reconstructions, there are still difficulties in isolating primary authigenic carbonate from detrital or diagenetically altered material

(von Grafenstein et al., 1999). The use of biogenic material can provide a more reliable proxy for determining the ambient conditions at the time of shell formation.

Ostracods have recently gained favour as palaeoenvironmental indicators as they are generally regarded to form their low-Mg calcite valves from material taken directly from the host water over a short period of time (commonly <24 hours)

(Turpen and Angell, 1971; Chivas et al., 1993; Roca and Wansard, 1997). Hence the chemistry of the shells provides a "snap-shot" into the ambient water conditions. There are ostracods in most aquatic environments and their microscopic size and bivalved-carapace, usually resulting in two open valves upon death, make them readily recoverable from the bulk sediment and relatively easy to clean. The assemblage of ostracod species present in a sample, the morphology of the valves and their ontogeny are indicative of the ecology and energy of the environment at the time of deposition (De Deckker, 1988). In addition, the preservation of the ostracod valve can reveal important information regarding post-depositional effects that may impact on the shell geochemistry (Swanson and van der Lingen, 1994; De Deckker, 2002). When these factors are combined, the potential of ostracods as palaeoenvironmental indicators may be realised. Caution must be applied although that ostracods within a sample represent an in situ deposit and not material transported to the sampling site (see Section 5.4.6).

Knowledge of the autoecology of the ostracods, that is when and where they calcify, also greatly aids geochemical interpretations.

265

6.5 Vital effects

The ostracod shell carbonate is precipitated out of isotopic equilibrium with the host water (McConnaughey, 1989). With respect to δ18O, ostracod calcite shows a positive offset from equilibrium, which thus far appears to be genus specific

(e.g. Xia et al., 1997a; von Grafenstein et al., 1992, 1999). By culturing ostracods under controlled conditions, this fractionation can be quantified.

The positive offset displayed by ostracods is in contrast to other biogenic material formed in isotopic disequilibrium with the host water, such as molluscs and foraminifers. A possible explanation for this offset suggested by von Grafenstein et al. (1999), is the storage of calcite within the outer lamella, prior to moulting, which is then recycled in the new valve. Chivas et al. (2002) however, dispute this on the grounds that it is at odds with the widely accepted radiocalcium tracer experiments of Turpen and Angell (1971), who show that no storage (at least of

Ca) occurs between moults. Other possibilities include the rate of reaction, whereby the duration of carbonate-water isotopic exchange governs the extent of isotopic equilibrium achieved, with ostracods that calcify more rapidly exhibiting a more positive δ18O offset (McConnaughey, 1989; Chivas et al., 2002).

The effect of pH on the fractionation of oxygen isotopes between water and carbonate has been noted in planktic foraminifers (Spero et al., 1997) and postulated for inorganic carbonates (Kim and O'Neil, 1997; Zeebe and Zheng,

1999). Chivas et al. (2002) found a correlation between increased pH and lower vital offsets in their culturing experiments of Australocypris robusta, although this may also be due to temperature fractionation effects near the species torpidity

6. Stable-Isotope Analyses

limit. Isotopic studies of ostracods from spring-fed ponds in southern England by

Keatings et al. (2002) support the notion of a relationship between equilibrium fractionation and the rate and pH of the reaction. They further suggest that the pH at the site of calcification, in the inner lamella of the ostracod, could be constant for a given species, and could be different to that of the ambient water.

A range of observations has been made to test the rigour of stable-isotope fractionation within a given species. Chivas et al. (2002) analysed poorly calcified valves and found no relationship between δ18O and degree of calcification or valve size. Von Grafenstein et al. (1999) observed some offset in the δ18O between adults and juveniles of the same taxa, however this was attributed to the seasonal moulting patterns; the degree of fractionation remained constant for a given genus. No appreciable difference has been noted in the stable-isotope ratios between either male and female or left and right valves of ostracods of the same species formed under the same conditions, or for different parts or fragments of an individual valve (Heaton et al., 1995; Bridgwater et al.,

1999).

Vital effects of fractionation seem also to be apparent for δ13C in non-marine ostracods, but are more difficult to quantify than those of δ18O. There appears to be no co-variance between δ18O and δ13C offsets in individual species

(von Grafenstein et al., 1999). The effects of microhabitat, such as whether the species is benthic or phytal, water depth of habitation and the influence of diet

(herbivore, carnivore or detritivore), are of particular relevance to the δ13C of ostracod calcite. A significant difference in the δ13C values of juvenile and adult

267

valves of Candona rawsoni was noted by Xia et al. (1997b), which they suggest may be due to characteristics in the life habit of this species.

6.6 δ18O and δ13C of ostracod valves as palaeoenvironmental indicators

6.6.1 Marine environment

Within the marine environment, both benthic and planktic foraminifers have classically been utilised for stable-isotope analyses to determine past oceanographic and climatic change as reflected in bottom and surface waters respectively. The δ18O in shell carbonate is a function of both the δ18O of the host water and the temperature at the time of calcification. Within planktic species a factor of sea-surface salinity is also incorporated. The isotopic ratio of carbon preserved in the shell is related to the δ13C of the water, which in turn is determined largely by the degree of productivity and organic matter decay. The

δ18O of deep ocean water, as determined by benthic foraminiferal analysis, is seen to vary over only around 2‰ (vs. V-PDB) through the last glacial cycle, with δ13C varying in the order of 1-5‰ (vs. V-PDB) (e.g. Martinson et al., 1987; Shackleton and Pisias, 1985).

The use of ostracods for stable-isotope analyses in the marine environment has lagged behind that of foraminifers. Some comparative studies have been made between the stable-isotope ratios of foraminifers and the trace element composition, namely Mg/Ca, of ostracods with respect to water temperature and salinity change (e.g. Dwyer et al., 1995; Corrège and De Deckker, 1997). Didié

(2001) presented the first systematic stable-isotope study on deep marine

6. Stable-Isotope Analyses

ostracods. Comparisons between the stable-isotope ratios obtained from ostracods and foraminifers from the same sediment core samples allow identification of trends for infaunal and epifaunal species of both taxa (Didié and Bauch, 2002).

In particular, the δ13C values are shown for foraminifers to be more negative the deeper in the sediment the organism lives. This is attributed to the decreasing

δ13C of the pore water with depth caused by the oxidation of 12C-enriched organic matter (Woodruff et al., 1980). This was not observed in the ostracodal analyses, perhaps due to vital effects (Didié and Bauch, 2002). The authors also note a positive offset in the δ18O of ostracods with respect to their foraminiferal counterparts. As yet there is no published data that show experimentally the relationships between environmental parameters and stable oxygen and carbon isotopes in marine ostracods.

6.6.2 Marginal marine environment

Within marginal marine environments, the stable-isotope ratios of the host water chiefly determine that of the shell carbonate. This in turn is governed by the degree of mixing between the two end-members of marine and terrestrial waters in an open system, and the extent of evaporation in restricted water bodies.

As such the carbonate chemistry is indicative of the salinity at the time of shell formation and the degree of input of the various aqueous sources. Variation in both the δ18O and δ13C is generally much greater than that of deep marine environments and may span more than 10 ‰.

Stable-isotope analysis in marginal marine environments has most commonly been performed on molluscs (e.g. Ingram et al., 1996b; Klein et al., 1996, 1997;

269

Hendry and Kalin, 1997). The use of ostracods has gained popularity more recently. Due to the number of variables in these environments, it can be difficult to identify the effect that governs shell chemistry. Most successful reconstructions have utilised multi-proxy studies, including ecology of the ostracod species present, trace element (Mg/Ca, Sr/Ca) and stable-isotope

(δ18O, δ13C) analyses of ostracod valves (e.g. Ingram et al., 1998; Hammarlund,

1999; Mazzini et al., 1999; Anadón et al., 2002). When considering these factors in combination, the causes of salinity change (e.g. variations in the evaporation/precipitation ratio, sea-level change or mixing of waters of different sources) may be discerned. Variations in shell chemistry due to temperature are commonly masked by the larger variations induced by salinity changes.

6.6.3 Non-marine environment

The hydrology of the basin in question primarily defines the principal effects that may be recorded in the stable-isotope geochemistry of the host water, and hence in the ostracod valve, in a non-marine environment. The ecology (e.g. whether benthic or nektic, phytal, littoral dwelling) and life cycle (i.e. season of spawning or moulting) of the ostracod species concerned determines the scale of environmental variations recorded in the microhabitats. The degree of seasonality may also have a marked effect on the spread of isotopic results obtained from valves within a single sample (e.g. Lamb et al., 1999; Schwalb et al., 2002).

Ostracods commonly form one of the dominant biogenic carbonate components in non-marine aquatic bodies. It is not surprising then that these settings represent the largest body of work in which ostracods have been utilised for stable-isotope

6. Stable-Isotope Analyses

analysis. The first published work was by Fritz et al. (1975). They investigated the δ18O and δ13C of both ostracods and molluscs in a core spanning ~15 ka from

Lake Erie, North America. The authors note correlations between δ18O and climate improvements as well as changes to drainage and a link between δ13C and changes in aquatic vegetation.

Ostracod stable-isotope analyses have been used to interpret past changes in air temperature (e.g. Lister, 1988, 1989; von Grafenstein et al., 1992, 1994, 1996,

1997; Hammarlund, 1999), effective precipitation (Curtis and Hodell, 1993;

Anadón et al., 1994; Hodell et al., 1995; Bridgwater et al., 1999, Mischke, 2001), relative humidity (Curry et al., 1997), hydrology (Lister 1988; Lister et al., 1991;

Schwalb et al., 1995; Xia et al., 1997c; Schwalb and Dean, 1998; Curtis et al.,

1999; Ricketts et al., 2001), atmospheric circulation (Schwalb et al., 1995, 1999;

Smith et al., 1997, 2002) vegetation (Schwalb et al., 1995; Holmes et al., 1997;

Bridgwater et al., 1999), dissolved oxygen levels (Cerling et al., 1993; Curry et al., 1997) and anthropogenic effects (Whitmore et al., 1996; Curtis et al., 1998;

Filippi et al., 1999).

Stable-isotope studies of non-marine ostracods are commonly coupled with assemblage associations, trace element analyses (particularly Mg/Ca and Sr/Ca) sedimentology and other proxies (such as pollen and diatom assemblage data) to differentiate between the effects of these variables (e.g. Gasse et al., 1987; Chivas et al., 1993; Palacios-Fest et al., 1993; Holmes et al., 1997; Smith et al., 1997,

2002; Lamb et al., 1999; Mischke, 2001; Griffiths et al., 2002).

271

6.7 Covariance of δ18O and δ13C

The degree of covariance between δ18O and δ13C of ostracod calcite gives an indication of the hydrology of the waterbody. Classically, a correlation coefficient of ≥ 0.7 is considered to represent a closed basin (Talbot and Kelts,

1990). Conversely, a poor correlation is typical of an open basin (Talbot, 1990).

The residence time of the water is the primary determinate for the isotopic response of the waterbody (Fontes et al., 1996). Long residence times result in evaporative enrichment of δ18O and increased δ13C due to exchange between DIC if the water and atmospheric CO2 (Kelts and Talbot, 1990) or increased primary productivity (e.g. Anadon et al., 1994). Short residence times result in lighter isotopic values corresponding to those of precipitation (δ18O) and the DIC of inflowing water (δ13C), which is generally enriched in 12C derived from dissolved plant material (Kelts and Talbot, 1990).

Covariance of δ18O and δ13C is not a definitive means of determining a basin's hydrologic history. There are examples for which open basins have been known to co-vary (Drummond et al., 1995) and closed basins that show poor covariance

(Chivas et al., 1993; Schwalb et al., 1999). It should be remembered that even closed waterbodies are usually connected to groundwater systems, which may act to either recharge or discharge within the basin. There are also examples of negative covariance, particularly in marshes with shallow water and dense vegetation, resulting in evaporatively enriched δ18O, yet depleted in δ13C owing to the biogenic CO2 (Fontes et al., 1996). In addition, Li and Ku (1997) state that covariance may not be observed in hyper-alkaline lakes, as the δ13C is insensitive to changes in lake volume under these circumstances. They also estimate that a

6. Stable-Isotope Analyses

basin needs to have been hydrologically closed for at least 5000 years to behave in a predictable covariant manner; even then, stable lake volume and productivity may result in poor covariance.

273

6. Stable-Isotope Analyses

275

6. Stable-Isotope Analyses

6.8 δ18O of the modern Gulf of Carpentaria

The catchment of the Gulf of Carpentaria spans approximately 10o latitude. There is a steep precipitation gradient from 2500 mm.a-1 in the south of New Guinea to

500 mm.a-1 in the south of the gulf. Monthly records of the oxygen isotopic ratio of precipitation have been taken from stations in the area since 1962 (Fig. 6.4).

The closest IAEA/WMO network station at Darwin gives a long-term arithmetic mean for δ18O of precipitation of -3.5‰ (vs. V-SMOW), with a weighted mean of

-5.0‰. Around 80% of annual precipitation falls during the wet season

(December to March) with an average isotopic value of -5.8‰. The comparatively little precipitation that falls through the dry season is significantly less negatively δ18O values, averaging -1.3‰. The low value for July corresponds to only two records taken in 1978 and 1999 of -6.9‰ on 7mm precipitation and -

4.5‰ on 1 mm precipitation respectively. The monsoon effect, showing an inverse relationship between the δ18O and the amount of precipitation is dominant.

Several other stations in the region may bear some relevance to the isotopic ratio of precipitation in the gulf. Records from Brisbane, on the east coast of Australia, show less variation than those of Darwin, ranging between -5.0‰ in the July and

-2.4‰ in September, with a weighted annual mean of -4.4‰. Here, the majority of rainfall occurs in summer, although not under deluge conditions, with an average δ18O of -3.4‰. The range of values is less negative than those of Darwin, and the amount effect is not apparent, however there appears to be a loose correlation with temperature. In contrast, the records from Alice Springs, in central Australia, show a wide variation from +1‰ in spring to -6.5‰ in late summer, with a weighted annual mean of -6.4‰. The amount of precipitation is

277

much lower than at the other stations considered here, falling mostly in summer.

The continentality effect is dominant here.

Records from the north of the gulf include stations in Madang, Papua New Guinea and Jayapura, West Papua. The monsoon-driven amount effect is dominant in both of these records, with the most negative δ18O results correlating with maximum precipitation. Isotopic ratios for the two stations range from -10.5 to

-4.0‰, with a weighted annual mean of -7.7‰ for Mandang and -6.5 to -3.8‰, with a mean of -5.8‰ Jayapura.

Under the current climatic regime, isotopic ratios of precipitation, similar to those recorded for Darwin, may be expected in the gulf throughout times when the sea level was close to that of present. During times of lower sea level, with increased continentality, higher δ18O values of precipitation may be expected across the gulf region, particularly if the Australian monsoon was not active.

6. Stable-Isotope Analyses

279

6. Stable-Isotope Analyses

281

6. Stable-Isotope Analyses

6.9 Stable-isotope analyses from core MD-32

A total of 306 analyses for both stable carbon and oxygen isotopic ratios were performed on ostracod valves. Samples were taken from 114 depth intervals; where possible, at least three samples were taken from each depth. Analyses were made on well-preserved, fully calcified valves. Single valves were analysed, where the valve-weight permitted. A visual rating (on a scale of 1-5) of the degree of preservation was made of each valve and has been noted

(VPI, App. 9). Adult valves were utilised preferentially, although penultimate instars were also used in some cases. Left, right, male, and female valves were analysed, however no systematic difference was noted between each of the valve types.

There is no single species that occurs in all of the sedimentary units of core

MD-32. Cyprideis australiensis was the most widely utilised species for isotopic analysis, as it occurs through much of the core, commonly as the dominant species. Only one valve per sample was required for analysis. In many samples it was recognised in biocoenotic deposits, hence the geochemistry of the valves is considered suitable to reflect the ambient conditions at the time of formation. The genus Neocytheretta was utilised through sedimentary unit 1 (0-0.38 m) and unit 6

(9.4-13.9 m). Species of this genus (N. adunca, N. spongiosa, N. vandijki and the closely related species Alocopocythere goujoni) were present in most of the depth intervals of these units, commonly with three or more instars present. The majority of samples utilised single valves. Throughout subunit 3a, Venericythere darwini was used, again for its abundance and biocoenotic distribution in these depth intervals. Two valves of this taxon were required per sample. The species

283

Ilyocypris australiensis was used at various depth intervals throughout unit 2, especially in subunit 2a where in some samples it was the dominant species represented. The light weight of this taxon necessitated 3-6 valves to be analysed together. Some Cyprideis australiensis valves have also been analysed at the same depths to allow for comparison between the two taxa. No results were obtained from either unit 4 or 7 as the samples were either barren, or did not contain suitable material for isotopic analysis.

Of the 306 samples analysed, 42 did not produce sufficient gas upon dissolution to provide reliable results. This was for the most part for sample weights of less than

30 μg. Some samples that weighed greater than 40 μg, also recorded low transducer readings, which may indicate that the valves were not fully dissolved before analysis. A further three samples were discounted as they produced poor internal precision in the results. This may be due to contamination of the samples.

The results considered successful are presented in Figs. 6.5, 6.6, and App. 9.

6.10 Results

In keeping with the ostracod assemblage data, the stable-isotope results will be discussed in terms of the unit and subunits, as defined by the sedimentary analysis of core MD-32, in order of deposition. All of the results are expressed relative to the V-PDB standard. The number of valves available per sample and the lack of known vital effects for the taxa consider limit the quantification of the interpretation. However relationships may be described between the relative isotopic results presented.

6. Stable-Isotope Analyses

285

6. Stable-Isotope Analyses

As mentioned above, no isotopic analyses were obtained from unit 7, as it is barren of microfauna. The isotopic results for unit 6 are presented in Fig. 6.7.

Throughout unit 6, the δ18O results for Neocytheretta are relatively uniform, compared with the rest of the core. The highest δ18O values for the unit are found in the stratigraphically lowest samples, within subunit 6f, decreasing from -0.5‰ near the base to -2.0‰. The remainder of the unit gives values ranging from

-3.0‰ to -1.4‰, with an average of -2.1‰ and showing a slight negative trend.

On closer inspection some other trends may be determined from the δ18O record, within the span of variation for a given depth. These broadly follow the sedimentological subunits, in that subunits 6d and 6b display slightly more negative isotopic ratios, on average, than subunits 6e, 6c and 6a. The δ13C values for the same suite of samples produced a greater range of values, spanning from

-6.8 ‰ to -3.1‰. Trends in the δ13C also broadly follow the sedimentological subunits. Subunit 6f shows a positive trend, from -5.8‰ to -4.6‰. Although there is a hiatus of results in the lower section of subunit 6e, owing to the lack of suitable sampling material, the top section of the subunit shows a negative trend, from -4.6‰ to -6.2‰. A positive trend is again apparent through subunit 6d, spanning a range of -6.3‰ to -3.4‰. The values display an overall negative trend for the rest of unit 6, from averages of -4.3‰ in subunit 6c to -5.0‰ in subunit 6a.

Again, a broad range of values is represented, from -6.8‰ to

-3.1‰, with as much as 2.7‰ variation within the one depth interval. It should be noted that a variety of species of Neocytheretta were used for analysis through unit 6 (Fig. 6.7). No constant offset, nor vital effect could be determined from the

δ18O results. The species N. adunca however, did commonly have less negative

δ13C values, compared with the other taxa.

287

Unit 5 covers only 35 cm of the length of the core, however the isotopic ratios obtained from samples within the unit are extremely broad spread (Fig. 6.8).

Initially samples were taken from this unit at 10 cm intervals. As the results obtained appeared so varied, the unit was further sub-sampled at 5 cm, then 1 cm intervals to identify the spread of the isotopic values. The range of values span

16.1‰ (-7.8 to +8.3‰) and 10.4‰ (-12.9 to -2.5‰) with averages of +0.86‰ and

-6.42‰ for δ18O and δ13C respectively. Variations of up to 12.8‰ for δ18O were obtained from a single depth (9.0 m). Although not quite as scattered, the δ13C values for the same depth span 7.1‰. A broad, and somewhat erratic, negative trend may be made out for both δ18O and δ13C, however there is no covariance of the two isotopic ratios (Fig. 6.8b). Valves of Cyprideis were used throughout unit 5.

6. Stable-Isotope Analyses

289

6. Stable-Isotope Analyses

Unfortunately there were no results obtained from unit 4. The few ostracod valves that were extracted from the sediment samples were poorly calcified and considered unsuitable for isotopic analysis.

Analyses through unit 3 were based on two ostracod taxa, Cyprideis for subunit

3b and Venericythere for subunit 3a (Fig. 6.9). The δ18O values obtained were similar for both taxa, ranging from -3.3‰ to -1.0‰, with the average results for subunit 3b (-1.0‰), marginally more negative than those for 3a (-0.4‰). These values are intermediate to those obtained for unit 5. A more distinct difference is noted in the δ13C ratios, with subunit 3b clustering between -6.0‰ to -3.8‰ and subunit 3a displaying more negative results from -6.7‰ to -5.6‰, with one sample at 3.8 m of -4.3‰.

291

6. Stable-Isotope Analyses

293

6. Stable-Isotope Analyses

295

6. Stable-Isotope Analyses

The isotopic ratios produced from samples of unit 2 again display a broad range of results (Fig. 6.10). For both oxygen and carbon the samples of Ilyocypris valves are generally offset to the right (i.e. isotopically more positive) than those of

Cyprideis valves, although there are some exceptions. Only three successful results were obtained from subunit 2c. The δ18O values of these samples are similar, with the Cyprideis valve (0.0‰) slightly higher than the two Ilyocypris valves

(-0.6‰ and -0.7‰). The δ13C values are more widely spread with Cyprideis the lowest (-5.0‰) and Ilyocypris valves giving results of -3.2‰ and -1.7‰.

Through subunit 2b, most of the Cyprideis δ18O values cluster in the range of

-2.4‰ to 0.0‰, with some positive results as high as +5.0‰. The major positive excursions occur at depths of 2.1 m to 1.8 m, 1.45 m to 1.30 m and an outlier at

1.65 m. Samples with the most abundant Cyprideis valves, such as those from the shell layer at 1.5 m and 1.6 m, produce a range of δ18O results from -2.2‰ to

+0.1‰. The average for δ18O for Cyprideis throughout subunit 2b is 0.0‰, as although there are more negative results, the positive results show a broader spread and hence tip the mean. Reliable results have been obtained from only a few Ilyocypris samples in subunit 2b. Samples at 2.5 m and 2.1 m produced similar results to Cyprideis valves taken from the same depths. The Ilyocypris samples at 1.7 m, however recorded values near the Cyprideis mean for the subunit, of +0.6‰ and +0.7‰ and one significantly higher result of +8.5‰.

297

The δ13C patterns of subunit 2b are more straightforward than the oxygen isotopic ratios of the same samples. The results from the Cyprideis valves show a fairly consistent spread throughout, spanning -5.1‰ to -2.2‰. A slight positive trend may be noted. The Ilyocypris valves display a consistently positive offset from the

Cyprideis samples, and range from -1.7‰ to -1.1‰.

The samples from subunit 2a comprise a larger percentage of Ilyocypris valves.

These show a very wide spread of δ18O ratios, reaching +10.0‰ and spanning greater than 6‰ at a given depth. The lowest δ18O value obtained from an

Ilyocypris valve was -1.8‰ at 1.0 m. Analyses of Cyprideis valves show a similar range of values to those of subunit 2b. These values are markedly lower than those derived from Ilyocypris in all except the samples at 1.0 m. One

Neocytheretta valve from a depth of 0.7 m recorded a δ18O value of -2.0‰. A similar relationship between the two dominant taxa is reflected in the δ13C ratios of this subunit. Whereas Cyprideis valve results were consistent with the underlying subunit, Ilyocypris valves are again significantly higher, ranging from

-4.4‰ to -0.4‰. The one Neocytheretta valve δ13C value of -5.1‰ was the most negative for the subunit.

6. Stable-Isotope Analyses

299

6. Stable-Isotope Analyses

The uppermost unit of the core, unit 1, sees a return of Neocytheretta as the dominant species suitable for analysis. Isotopic ratios obtained from these valves are of a similar magnitude to those from unit 6 (Fig. 6.11). The δ18O results cluster around -2.0‰, spanning -2.4‰ to -1.5‰, with one sample from the core top recording 0.0‰. Again the δ13C values are more widely spread, ranging from

-2.5‰ to -5.1‰ in a negative trend, with the sample at the top an outlier at -1.9‰.

One Cyprideis valve from 30 cm was analysed, with values of -1.2‰ and -8.6‰ for δ18O and δ13C respectively.

6.11 Discussion and interpretation

The stable isotopic data for core MD-32 may be subdivided into at least three sections:

• A lower section (13.9 m to 9.4 m), with fairly consistent δ18O and variable

δ13C,

• A middle section (9.3 m to 8.95 m), with a very wide spread for both δ18O

and δ13C particularly toward more negative values,

• An upper section (5.6 m to core top), which shows a spread of δ13C and a

wide range of δ18O, particularly to positive values.

These correlate broadly with the palaeoenvironmental changes identified in both the sedimentology and ostracod assemblage data, presented in previous chapters.

It should be noted that all of the ostracod species used for analysis are benthic-dwellers and are not known for burrowing, hence the isotopic composition of the valves may be generally considered to be reflective of ambient water conditions at the sediment-water interface during valve formation.

301

The species of Neocytheretta utilised for analysis through unit 6 of the core are common to marine environments, ranging from the open ocean (Neocytheretta vandijki) to inter-tidal channels (Alocopocythere goujoni), although are most abundant in shallow inner-shelf settings. The isotopic ratios obtained from these valves correspond to the fluctuating marine sequence through unit 6 and as such primarily reflect the sea level, relative to the core location, at the time of deposition. The average δ18O of the unit is -2.1‰. This value most likely includes a component of the vital offset between the δ18O of the bottom seawater at the time of calcification and that of the shell carbonate for the taxon Neocytheretta.

However, the global signature for biogenic carbonates precipitated from seawater at a nominal 25oC is also around -2‰.

A study of Quaternary sea-level change of the Tyrrhenian coast near Orbetello,

Tuscany by Mazzini et al. (1999) noted variations in δ13C of -10.35‰ to -4.08‰ and δ18O of -3.45‰ to +0.93‰ of ostracod calcite. This they attributed to variations in salinity for the former, i.e. mixing between marine and continental waters, and δ18O of the water and temperature for the latter. They also identify a

δ18O value of -2‰ to indicate a dominant seawater influence in such a setting, suggesting that marginal basin isotopic ratios reflect the global ocean signature, imprinted with local effects. The influence of seawater within the varying marginal marine setting is considered to be the dominant influence on isotopic variation within the Gulf of Carpentaria also.

It should also be noted that the degree of variation in isotopic ratios observed from the gulf record is far greater than that of the global marine isotope record. This is

6. Stable-Isotope Analyses

due primarily to the shallow nature and low gradients of the basin. Evaporation and inflowing river water both have marked effects on the resultant isotopic composition of the water and hence ostracod calcite. Immerhauser et al (2003) made similar observations from Late Carboniferous shallow marine carbonate platforms in the Cantabrian Mountains of northern Spain.

Throughout unit 6, isotopic variations of the ostracod calcite between subunits may be attributable primarily to the degree of marine influence (i.e. open marine or shallow estuary) and nature of the waterbody at the time of deposition. This in turn determines the effectiveness of other parameters including variation in the composition of the water due to mixing (e.g. within the tidal zone), influence of inflowing rivers, evaporation, temperature, vapour exchange with the atmosphere and primary productivity.

The negative trend of δ18O isotopic ratios through subunit 6f, from -0.5‰ to

-2.0‰, approaching the mean for the other samples of unit 6, is in keeping with deepening water and an increasing marine influence. This is supported by the introduction of planktic foraminifers and pteropods to the microfaunal assemblage from 13.5 m. The steady increasing trend of the δ13C suggests perhaps an increase in primary productivity and distance from inflowing rivers. The dominant ostracod taxon through this subunit, Cytherella, is commonly associated with sea-grass beds. As with the oxygen isotopes, the carbon isotopes also approach the mean for unit 6 at the top of subunit 6f. A gradual incursion of seawater, from an estuarine embayment to more established open shallow marine conditions, is implicated.

303

Only a small number of Neocytheretta valves suitable for analysis were extracted from subunit 6e, hence the paucity of isotopic data. However, the first sample, at

12.1 m, appears to follow on from the underlying samples in both δ18O and δ13C values. The samples from 11.9 m and 11.75 m show a slight negative trend for

δ18O and significantly more negative values for δ13C. This may indicate the influence of terrestrial waters, depleted in both 18O and 13C, and increased organic debris. The dominant ostracod taxon here is Loxoconcha, common in phytal, inter-tidal environments and estuaries. An increase in energy of the environment is noted by both the increase in the coarse fraction of the sediment and the decrease in foraminiferal diversity, favouring more robust taxa. This is consistent with a restriction of marine conditions, with possible channel influence.

Subunit 6d is characterised by decreasing mean δ18O values and increasing mean

δ13C values. This suggests deepening marine water, further from the influence of fluvial input, with increasing productivity, which is supported by the ostracod assemblage data. The dominant species, including Loxoconcha and Xestoleberis, are indicative of a nearshore phytal environment, deepening to open shallow marine from 11.4 m, as evidenced by the abundance of bairdiids from this level.

Each of the depth samples shows a spread in data, being as great as 0.8‰ and

2.0‰ for δ18O and δ13C respectively. At least part of the variation in δ13C may be due to the life habits and microenvironment of N. adunca, which is offset to the right of the other taxa. This species is commonly associated with phytal environments; the water of which would be enriched in 13C, as the lighter 12C would be preferentially utilised by the aquatic plants. The greater δ18O variance

6. Stable-Isotope Analyses

may represent increased seasonality, attributable to an active monsoon regime.

Both the amount effect on the δ18O of the precipitation, and hence river water input, and the warmer summer temperatures lead to more negative δ18O recorded by the ostracod valves.

Few samples were available from subunit 6c. Those analysed gave relatively high ranges for both δ18O and δ13C, in comparison to the rest of unit 6. Although the

δ18O results are representative of marine-composition water, the higher values may be indicative of increased effect of evaporation, such as in a restricted basin, or decrease in temperature. Unfortunately there were no depth samples for which multiple analyses were made; hence shorter-term variation is unavailable for this section of the core. The δ13C results suggest either shallow water with increased atmospheric vapour exchange or increased primary productivity. Li and Ku (1997) note that organic matter breakdown in restricted saline environments may be suppressed, resulting in higher than expected δ13C. The ostracod assemblage and sedimentological data support both of these notions. The taxa present are common to nearshore or restricted environments. Many of the valves are in-filled with pyrite, suggesting abundant organic matter during the life of the organisms, with reducing conditions within the sediment upon death. Gypseous laminae are present through this subunit, suggesting periodic inundation and subsequent evaporation in a restricted basin.

The better preservation of ostracods through subunit 6b rendered more valves suitable for analysis. The δ18O values here are the most negative found in unit 6, and represent a return to more established marine conditions, although a spread of

305

results as great as 1.2‰ may be noted within the one depth. Increased fluvial activity, possibly due to enhanced monsoon seasonally bringing depleted waters to the gulf, is suggested. Similarly the δ13C values become increasingly negative through the subunit and range as much as 1.7‰, supporting deepening waters and fluvial input. Again the use of N. adunca may act to skew the plot to less negative values then may otherwise be expected. The taxa within this subunit are indicative of open shallow marine conditions. Although the preservation of most of the ostracod valves is very good, the presence of some broken shell material suggests an increase in the energy of the environment.

The uppermost subdivision of unit 6 shows a reversal of the δ18O isotopic trend of subunit 6b to become heavier, indicating either cooler conditions or increased evaporation. A further restriction of marine waters may be postulated. A spread of results is also acknowledged, in keeping with that of the underlying subunit. The isotopic values are negative however, in support of a marine origin for the taxa analysed. The δ13C results continue in a negative trend, although show greater variability. This may again be due to the microhabitat preference of the different organisms being analysed, or be accounted for by a variations in productivity or variable fluvial inputs throughout the time represented in single depth samples.

The microfaunal assemblage for subunit 6a, being dominated by a shell hash layer at 10.0 m, supports a regression in sea level. There remains an open-marine influence, although this may be a result of storm surges and transportation, as many of the valves show signs of reworking. The diversity and level of preservation decrease through the sub-sample.

6. Stable-Isotope Analyses

The ostracod species used for analysis through unit 5, Cyprideis australiensis, is common in restricted lagoon and saline lacustrine environments rich in organic detritus. It is characteristic of waters dominated by Na-Cl over a wide salinity range and is unable to withstand desiccation. Although the exact ontology of this species has not as yet been studied, it may be compared with the northern hemisphere species, Cyprideis torosa. Heip (1976) observed this species in a brackish lake in Belgium, to produce one generation per year, reaching maturity in spring. Over-wintering was noted for some nauplius larvae, with temperature although to be the limiting condition. In contrast, Mezquita et al. (2000) observed two generations per year, in spring and autumn, to reach adulthood within a coastal saline march in the western Mediterranean. The species thus appears opportunistic, reproducing whenever optimum conditions are apparent.

The extreme variation in both δ18O and δ13C values of unit 5 suggest an environment decidedly different from the underlying marine unit 6. An enclosed body of water, subject to marked seasonal effects is suggested. There appears to be no relationship between δ18O and δ13C variations through the unit, with valves depleted in 18O whether enriched or depleted in 13C, or vice versa. This implies that the basin was not under steady-state conditions. There is some relationship between positive δ18O values and valves that show evidence of bleaching, but this is not universal. These valves were perhaps formed in highly evaporated waters, which may have been subjected to leaching after deposition. There appears to be a correlation between pyrite and depleted isotopic ratios. Although pyritisation is also a post-depositional effect, the link may indicate the valves were formed in a

307

highly organic-rich, reducing environment. A shallow, swamp-like waterbody is suggested.

The surface-area to volume ratio of the waterbody has been identified to determine the degree of response of a waterbody to local environmental variations

(Lister, 1988). In this example of unit 5, one may interpret a high ratio, owing to the large isotopic variations herein. If temperature alone were responsible for the range in δ18O, then a temperature variation of as great as 70oC may be anticipated.

This is obviously not the case. The evaporation/precipitation ratio is the major contributing factor to the variation in δ18O, and probably masks any temperature effect.

Primary productivity and degradation of organic matter are the principal controls on δ13C, as indicated by the consistently negative values. The amount of both reworked shell material and pyrite in the samples of unit 5 suggest seasonal influxes of water, with later stagnation and evaporation. The solute composition of the water, as determined by the abundance of Cyprideis, is essentially that of seawater, Na+-Cl- dominated. This taxon is known to tolerate a wide range of both salinity and temperature. The spread of isotopic values is indicative of calcification under a range of environmental conditions, and suggests that the ostracods reached maturity at different times of the year. The minor occurrence of other taxa, such as Ilyocypris and Darwinula, support periodic influxes of freshwater.

6. Stable-Isotope Analyses

Other authors have observed wide spreads in stable-isotope ratios. Heaton et al.

(1995) noted variations in δ18O and δ13C values among single-centimetre samples for lake sediment cores from Jamaica and Mexico to be as great as the expected variation throughout the last glacial cycle. This they attributed to seasonal, or more likely inter-annual variations in temperature and/or isotopic composition of the water. Dettman and Lohmann (1993) found variations as great as 9.8‰ for

δ18O within growth bands of individual bivalved molluscs, corresponding to seasonal temperature and water δ18O fluctuations. Marked seasonality is considered responsible for the 13‰ variations in δ18O from ostracod valves extracted from a sediment core within Raymond Basin, Illinois (Curry et al.,

1997). The authors attribute the most positive δ18O values to calcification during cool and dry conditions and the lowest to pluvial, humid and warm conditions.

They also note δ13C values as high as 16.5‰, which they attribute to methanogenesis. This effect has not been observed in the results from the Gulf of

Carpentaria material. Large intra-sample variations were observed by Lamb et al.

(1999) from Sidi Ali, a semi-arid lake in Morocco. The complete range of isotopic values attained from this material was -6‰ to +12.5‰ for δ18O, with as much as

12.5‰ variation within one horizon, and -4‰ to +2‰ for δ13C. They consider the isotopes to reflect an alternation of winter and summer precipitation and suggest that the insensitivity of the data is attributable to the lake being an open system with short residence time, not permitting steady-state conditions to be attained.

Certainly unit 5 of the MD-32 core is indicative of an environment in flux, however there is no positive evidence to suggest that the basin was open at this stage. Enhanced monsoon conditions are implicated for the δ18O variation of ostracod valves from the Late Holocene record of Lake Patzcuaro, Michoacán

309

central Mexico (Bridgwater et al., 1999). The Australian monsoon must be considered to be active during the deposition of unit 5, even during retreating sea level. From the combined evidence presented above, a confined shallow body of water of essentially marine composition, undergoing seasonal inundation and evaporation, is postulated for the depositional environment of unit 5.

The lack of suitable valves for analysis leaves an unfortunate gap in the stable- isotope record through unit 4. The insight that may be gained from the sedimentological and microfaunal suggest that the environment at the time of deposition of unit 4 was largely that of a mudflat with episodes of flooding, evaporation and exposure. Periodic influences of both estuarine and freshwater, and transportation of biogenic material to this region of the basin with inflow, are implicated.

Unit 3 marks a return to more established conditions. The return of Cyprideis as the dominant taxa, represented by a life assemblage indicates permanent water during the deposition of subunit 3b. The δ18O values show a steady negative trend to results, which suggests a freshening or deepening of water. The spread of δ18O values across as much as 3‰ indicates marked environmental variation. This could be due to a temperature variation in the order of 12oC, although seasonal precipitation and active river input may also make a contribution, with the lower

δ18O values representing both warmer and wetter conditions. The δ13C values form a fairly tight cluster between -6.0‰ and -3.8‰, with a slight negative trend.

This supports the notion of increased inflow and deepening of waters through the subunit, in a relatively stable setting and may also indicate increased decay of

6. Stable-Isotope Analyses

organic matter. The sediment record identifies a significant amount of pyrite, indicative of an organic-rich, reducing, low-energy environment. The microfaunal evidence shows an increasing percentage of finely reticulate Leptocythere hartmanni valves through the subunit at the expense of the more aggraded morphs, characteristic of lowering pH and increasing pCO2 of the host-water.

There is a shell layer, or concentration deposit noted in the core at 5.0 m. Above this there is a slight positive shift in both δ18O and δ13C. This may be due to decrease in fluvial input, contraction of the lake and increase in salinity. There is evidence of drier conditions at this stage in the core, with mottling of the sediment indicating subaerial exposure and fluctuations in the watertable.

For subunit 3a, the species Venericythere darwini has been utilised for stable-isotope analysis, owing to its dominance in the assemblage and the absence of Cyprideis. Venericythere darwini is common to restricted estuarine environments. The faunal shift indicates a renewed contact with a marine source.

Only one sample has been taken at each depth interval through subunit, however owing to the weight of Venericythere, two valves were used per sample. At the one depth (4.4 m) where there are two samples, a range of 1.8‰ for δ18O and

0.3‰ for δ13C is noted, which is comparable to the spread of result obtained within a single sample for the underlying subunit.

An isotopic shift occurs with the change of subunit, which may in part be due to the different taxa being analysed, but also a change in environment. The δ18O record is more positive than that of subunit 3b, averaging around 0.0‰, with an

311

outlier of -2.8‰ at 4.0 m. This indicates either an increase in salinity, or more effective evaporation. A decrease in temperature may also produce higher δ18O values, but owing to the faunal assemblage shift, contact with marine waters in a restricted, lagoon-type environment is favoured.

The δ13C values of subunit 3a are negatively offset from those of subunit 3b, forming a relatively tight cluster around -6.0‰. The uppermost sample is an exception to this, with a value of -4.3‰. The more negative isotopic ratios may be due to decreased productivity in the water column, or increased organic degradation at the sediment water interface. The occurrence of aggraded morphs of Leptocythere hartmanni through the subunit; indicating supersaturation with respect to calcium carbonate, favour the former explanation. The values are similar to those obtained for Neocytheretta valves at the transition between subunit 6e and 6d and at the top 6a. A similar, restricted marine-influenced environment is suggested for subunit 3a.

Throughout unit 2, analyses were performed on both Cyprideis australiensis and

Ilyocypris australis valves. The genus Ilyocypris is characteristic of temporary ponds and flowing streams. In Australia, Ilyocypris australiensis occurs predominantly in the salinity range of 4-7‰, although may tolerate up to 10‰

(De Deckker, 1982a). In contrast, Cyprideis australiensis is capable of tolerating a much broader salinity range, but prefers standing waters and cannot withstand desiccation of the waterbody. The two species have not been found to coexist in any modern samples and hence environmental interpretation is difficult. The occurrence of the two taxa in the same depth is most likely due to the Ilyocypris

6. Stable-Isotope Analyses

valves being transported via streams. Whereas Cyprideis is represented by a number of instars at most depths, Ilyocypris occurs only as adult or A-1 valves in these samples.

Only three results were obtained from subunit 2c, and all of these from the same depth interval (3.5 m). Other attempted analyses of samples from this subunit were not considered reliable and hence, have not been included. Analyses were made on both Cyprideis and Ilyocypris valves. The δ18O values are tightly constrained, with the results from Cyprideis valves marginally higher. The mean value of -0.4‰ is consistent with unit 3. The δ13C ratios are more widely spread, with Cyprideis producing the lowest results, comparable in particular to subunit

3b. The heavier values obtained for Ilyocypris may represent a difference in the microhabitat of calcification, being less 13C-depleted, or indeed valve formation at a different time of the year to Cyprideis. A seasonal variation is not supported by the δ18O results. The taxa suggest an enclosed body of non-marine water, although the stable-isotope results indicate conditions not far removed from seawater, at least at the base of this subunit. Brackish conditions are suggested.

Initial observations of the stable-isotope record through unit 2b reveal a broad spread of results, particularly for δ18O, and isotopic enrichment compared with other units. On closer inspections, the majority of the δ18O results form a cluster between -2.4‰ and 0.0‰. The most consistent results are from samples between

1.6-1.5 m. These coincide with a concentration deposit, with abundant non-marine molluscs and charophytes, and suggest an increase of freshwater into the lake at this time. The range of δ18O for these samples spans -2.1‰ to 0.0‰. This would

313

correspond to a seasonal temperature variation in the order of 8oC. Some valves however, show positive excursions up to +5.0‰, particularly around 2.0 m and

1.4 m. The valves show a higher degree of preservation; hence they are probably in situ. It is more likely that they were formed during periodically more evaporated conditions from 18O-enriched water.

The δ13C results of the same samples reveal a more consistent pattern, suggesting relatively stable conditions. The values obtained for Cyprideis valves, spanning

-5.1‰ to -2.2‰ are comparable, but slightly higher to those of subunit 3b. The spread may reflect a periodic effect of freshwater input or variable productivity.

The results from Ilyocypris form a separate grouping from those of Cyprideis, varying between -1.7‰ and -1.1‰. The consistently higher values suggest formation under different, possibly more highly productive waters, or shallow pools with greater atmospheric vapour exchange.

Ilyocypris valves were analysed from a number of depth samples through subunit

2b and show a positive isotopic offset of both δ18O and δ13C in comparison to

Cyprideis valves in most samples. This is not thought to be a vital fractionation effect alone, as the amount of offset is not consistent and the degree is far greater than that observed in other taxa. Most recorded δ18O offset of ostracod calcite from equilibrium are in the order of +0.5‰ to +2‰ (e.g. Xia et al., 1997a; von Grafenstein et al., 1999; Chivas et al., 2002). Valve formation of the two species under different environmental settings is strongly suggested.

6. Stable-Isotope Analyses

Analyses of Cyprideis valves through subunit 2a show a continuation in the trends of both δ18O and δ13C of the underlying subunit. The majority of analyses however, were performed on Ilyocypris valves. The results are very broadly spread, spanning as great as 10‰ and 4‰ for δ18O and δ13C respectively, within a given depth. These suggest that the valves were formed under fluctuating of conditions. Between 3 and 6 valves are utilised in each Ilyocypris sample, and hence the result represents a combined average of the individual valves. It is not only the range of results, but also the positive δ18O values in particular, that require explanation. The genus Ilyocypris is commonly associated with fresh to oligohaline conditions; Ilyocypris australiensis being found in water

<10‰ salinity. The highly enriched δ18O values of Ilyocypris from subunit 2a suggest formation in evaporated waters.

A possible scenario could be if the carapaces were formed in pools of water that were filled with freshwater (i.e. monsoonal precipitation) and then evaporated. If there were no concentration of salts in the confined pool, then the valves could precipitate from solute fresh, yet isotopically enriched water. These valves could then be transported to the depositional site. The predominance of adult Ilyocypris valves supports this, although the valves show no obvious signs of reworking.

A second explanation could be that the valves were originally formed in fresh to oligohaline pools and then recalcified in more enriched waters. There is some evidence of minor recalcification under the vestibulum of some valves, but this is not consistent with the enriched values. Certainly there is evidence of an

2+ - increasing number of freshwater taxa, most commonly found in Ca -HCO3

315

dominated waters, found through subunit 2a, suggesting at least a periodic freshwater influence, if not established fresh conditions.

The one red dot in Fig. 6.10 refers to an analysis of a Neocytheretta valve. The isotopic values obtained, being -2.0‰ and -5.1‰ for δ18O and δ13C respectively are comparable with the average results obtained through unit 6. This indicates a marine origin for this valve, which is to be expected, suggesting transportation to the lake.

Unit 1 identifies a return to marine conditions comparable to those of unit 6. The isotopic values, generally being around -2.4‰ to -1.5‰ for δ18O and -5.8‰ to

-2.0‰ for δ13C, are most similar to those of subunit 6a. There is a slight negative trend in both isotopic ratios, with the exception of the uppermost sample, which is significantly heavier. This valve may have precipitated within a more restricted setting. As with unit 6, a variety of species of Neocytheretta were utilised. Again there appears to be a slight offset noticeable in the δ13C ratios of N. adunca in comparison to the other taxa, supporting a microhabitat difference for this species.

One Cyprideis valve was analysed from 0.3 m. The δ13C values in particular are removed from the trend of the Neocytheretta valves, and indeed are quite negative

(-8.6‰) in comparison to other δ13C obtained from Cyprideis valves through the rest of the core. Such low values were obtained only from unit 5 of the core.

Calcification in 13C-depleted waters and transportation may be inferred. Both the isotopic and microfaunal analyses suggest that the marine material in unit 1 was largely transported as part of the most recent marine incursion, rather than being deposited in established open marine conditions.

6. Stable-Isotope Analyses

6.12 Other palaeoenvironmental indicators

For further information regarding variations in aquatic conditions and climatic change, other proxies may be considered. The pollen record has been established for core MD-32 and may be used in comparison to the sedimentological, ostracod faunal and geochemical data presented above, throughout this core. In addition, trace element and strontium isotope analyses of ostracod valves from the

Torgersen cores can also be extrapolated to compare with the stable carbon and oxygen isotopic data, particularly through the Lake Carpentaria phase of the basin.

With these combined data, it is possible to begin to differentiate between the effects of temperature and precipitation variations and marine influence on the basin.

6.12.1 Pollen analyses

The pollen preserved in sediment gives an indication of the regional vegetation and hence environmental and climatic conditions at the time of deposition. Pollen content is dependent on the density of vegetation, transportation agents, sedimentation rate and preservation. The palynology of the entire core MD-32

(at 5 cm intervals) has been established by Dr Sander van der Kaars. He has kindly permitted his interpretation of the pollen record to be presented below. In addition, a preliminary investigation of the top 1.5 m of each of the other cores,

MD-28, -29, -30, -31 and -33, has been published (Chivas et al., 2001).

The pollen record is broadly in agreement with the sedimentological and microfaunal records presented above, with the subdivision of the core into basal non-marine (14.8-13.93 m) lower marine (13.93-10.08 m), terrestrial

317

(10.08-5.62 m), lacustrine (5.62-0.38 m) and upper marine (0.38-0.0 m) units. The lowermost unit is dominated by grasses, with few trees represented and is indicative of dry climatic conditions. Significant biomass burning is noted. The presence of Botryococcus implies an aquatic environment, such as a restricted lake. An increase in pteridophytes and sedges toward the top of the unit signify the onset of wetter conditions contemporaneous with rising sea level. This unit represents the driest environment encountered in core MD-32. There is no firm evidence for an established body of water at the location of core MD-32 from either the sedimentological or microfaunal record, which suggests episodic alluvial transportation of material in an otherwise terrestrial environment, with a restricted lake near the depocentre of the basin.

The establishment of mangroves along the coast from 13.93 m signifies the marine transgression. On land, freshwater swamps rich in sedges and grasses, and with Azolla occur. Woodland taxa are well represented, dominated by Callitris with some rainforest cover. Eucalyptus is present, but to a lesser degree than in the modern environment. A warm and humid, tropical environment, similar to that of southern New Guinea today, is postulated.

From around 11.38 m to 10.68 m, there is a noted decline in mangrove cover, which may be due to a minor sea-level retreat. The abundance of mangroves then increases briefly, before petering out at 10.08 m. Rainforest vegetation increased through this period, with significant taxa including Anarcardiaceae,

Macaranga-type, Araucariaceae, Podocarpus and ferns. The richest pollen

6. Stable-Isotope Analyses

concentration of the marine facies is at the base, decreasing through the unit; biomass burning is also reduced.

The pollen record through the marine unit follows the broad trends outlined in the sediment and microfaunal records. There is also a relationship between the vegetation present and the δ13C ratios. Higher δ13C values are generally associated with increased mangroves, supporting increased primary productivity. The lower

δ13C values may be correlated with declining sea level and increased influence of trees, particularly rainforest vegetation, through fluvial input.

Restriction of the waterbody from the open ocean is apparent in the pollen record from 10.08 m. This is first noted by the disappearance of mangroves. A decline and change in composition of rainforest taxa also occurs, with the co-dominance of Araucariaceae and Podocarpus. Other taxa present include Eucalyptus woodland, sedges, grasses and Restionaceae and with Azolla associated with freshwater swamps. This assemblage suggests warm and possibly drier or more seasonal conditions. In the microfaunal record, there is a continued influence of marine material although to 9.3 m, however much of the material is reworked and may have been transported, rather than deposited in situ.

The establishment of a swamp-like environment, rich in Typha follows

(9.63-8.88 m), with some sedges, grasses and Azolla. A reduction in rainforest cover also occurs. There is a dramatic increase in pollen concentration through this unit and charcoal is also more abundant. Drier or more seasonal conditions are inferred, which is supported by the very broad variation in the stable-isotope

319

evidence. The development of the Typha swamp is concurrent with the bloom in the euryhaline non-marine ostracod, Cyprideis.

The presence of a large swamp in an otherwise terrestrial environment continues through to 5.62 m, with a change in dominant vegetation first to grasses

(8.88-6.48 m) and then to sedges (6.48-5.62 m). Pollen concentration remains high, although decreases through the unit. The algal taxa are poorly represented, with only minor Botryococcus to 6.48 m, suggesting that a permanent waterbody was not present at the core site. This is supported by the paucity of microfauna and occurrence of gypsum laminae at this level in the core. On land, a cover of woodland, including Acacia, Callitris, Casuarinaceae, Dodonaea and Eucalpytus, is concurrent with the grass-expansion, implying wetter conditions. These taxa are again reduced during the sedge-expansion suggesting a return to drier conditions.

The sediment comprises barren, iron-oxide stained quartz, with abundant calcareous concretions indicating possible alluvial deposition, followed by subaerial exposure, which may have an effect on the pollen preservation.

The establishment of lacustrine conditions is indicated by the increased values of

Azolla, Botryococcus and minor Coelastrum and Pediastrum from 5.62 m to

4.38 m. Typha swamp vegetation, sedges and grasses surround the lake. Wetter conditions are inferred from the presence of a lake at this time. The microfaunal evidence is supportive of a lacustrine environment, however brackish rather than fresh conditions are postulated.

6. Stable-Isotope Analyses

In the interval 4.68 m to 4.33 m, there is a change in taxa to a dominance of sedges and grasses. A restriction in lacustrine conditions is inferred. There is a strong increase in the relative charcoal values and a very low pollen count, indicating a high level disturbance. The taxa Myrtaceae, Asteraceae and an unidentified Leguminosae are well represented. This may be concurrent with the change in pollen taxa and increased disturbance including evidence of burning identified from the Atherton Tableland (Kershaw, 1986) and Banda Sea

(van der Kaars et al., 2000). The sedimentological and microfaunal evidence suggest drier conditions, a hiatus in deposition and subaerial exposure. There is some indication of a marine incursion in the microfaunal record subsequent to the hiatus, however there is no evidence of this in the pollen record.

The return to a lacustrine environment is indicated by the increase in algae from

4.33 m of firstly Botryococcus, then Coelastrum and Pediastrum from 3.73 m.

The freshest conditions are apparent toward the top of the sequence. Although there is no clear climatic indicator, a relatively wet climate is postulated. Initially surrounding vegetation is marked by a dominance of Typha, which is gradually replaced by grasses (3.78 m) and then sedges (3.48 m). There is an increase in woodland cover in the interval 3.48 m to 2.78 m, consisting mainly of Callitris,

Casuarinaceae and Eucalyptus. A reduction in woodland taxa occurs from 2.78 m; with Casuarinaceae becoming dominant from 1.98 m. Typha increases in abundance from 1.48 m, accompanied by an increase in biomass burning. At this same level, Pediastrum is increased, indicating an increase in freshwater in the lake. A shell-rich unit (1.6-1.3 m) and increase in freshwater ostracod taxa in the overlying unit supports this.

321

The lacustrine unit, or lateral equivalents, have been identified in the pollen records of the uppermost 1.5 m of the other five cores in this study (namely

MD-28, -29, -30, -31 and -33) and have been presented by Chivas et al. (2001).

An abundance of local swamp and grassland taxa, and variable representation of

Botryococcus, Typha and Pediastrum identify this non-marine unit. The extent of lake may be determined by the expression of lacustrine pollen taxa in the various cores, in support of the sedimentological evidence. The lake appears to have first existed in the deepest part of the basin, extending to the vicinity of MD-31 and

MD-33, near the present -60 m contour. The unit is identified by Botryococcus,

Coelastrum and Pediastrum in MD-33, surrounded by sedges, grassland and

Chenopods, indicating saline soil around the margins. This is supported by both the sedimentological and microfaunal data. An expansion of the lake and freshening of lake water is indicated in the overlying sediments. Pediastrum joins the algal assemblage in MD-33 and MD-31. Well-preserved, non-marine ostracod taxa are present in both of these cores at this stage. There is also an expansion of woodland taxa, dominated by Casuarina, although sedges and grasses remain dominant. The uppermost lacustrine material represents the greatest extent of the lake and the freshest conditions. All cores show evidence of surface water ranging from shallow and swampy at the extremities (MD-29) to deeper and freshwater near the depocentre, toward the east of the basin (MD-33). This is paralleled in the microfaunal assemblages. Open grassland with sedges and Typha remain the dominant terrestrial taxa.

The return to marine conditions, as evidenced in the sedimentological and microfaunal records, is apparent in the pollen record from 0.38 m. Here it appears

6. Stable-Isotope Analyses

with the establishment of mangroves in the coastal fringe and increased representation of Podocarpus, Callitris, Eucalyptus, Restionaceae and ferns. The expansion of woodland and rainforest cover is suggestive of the onset of wetter conditions. This transition to marine conditions is evidenced in the pollen record of all of the cores investigated. Transported woodland, lowland and montane taxa, mangroves and pteridophytes as well as the dominant swamp and grassland vegetation are common throughout.

It is demonstrated that the pollen record, although established independently, is in broad agreement with sedimentological, microfaunal and stable-isotope data as outlined above. Differences that do occur may be interpreted in terms of scale and timing of depostion of the respective records.

6.12.2 Other geochemical analyses

Geochemical analyses have previously been undertaken on core material taken from the Gulf of Carpentaria. Both trace element (De Deckker et al., 1988) and strontium isotope (McCulloch, et al., 1989) analyses have been performed on ostracod valves from cores GC-2 and GC-10A, collected by Torgersen and others

(as outlined in Chapter 4). Single valves of Cyprideis were used in the majority of the analyses. The close correlation between cores GC-2 and particularly MD-33 has allowed comparison between the Torgersen cores and the material in the present study (Fig. 6.12).

Trace elements are known to co-precipitate with ostracod shell calcite

(e.g. Kesling, 1951; Sohn, 1958; Cadot et al., 1972; Cadot and Kaesler, 1977).

323

With the understanding that ostracod carapaces are formed directly from ions taken from the host water (Turpen and Angell, 1971), it follows that the trace-element concentration of the valve is related to that of the host water. Chivas et al. (1983; 1985; 1986a,b) demonstrated this relationship for the elements magnesium and strontium in a series of controlled experiments. In addition, the authors found clear and positive temperature dependence for Mg uptake. The partitioning of trace elements into the ostracod valve was shown to been genus specific. Distribution coefficients for Cyprideis are as follows;

-1 KD[Mg] = (Mg/Ca)shell(Mg/Ca)water = 0.00458 ± 0.00072

(at 25oC, De Deckker et al., 1988)

-1 KD[Sr] = (Sr/Ca)shell(Sr/Ca)water = 0.475 ± 0.057

(within the temperature range 10-25oC, De Deckker et al., 1988)

De Deckker et al. (1999) have also postulated thermodependences of trace-element partitioning for Cyprideis as follows;

o KD[Mg] = -0.000514 + 0.00019 x T( C)

(within the Mg/Ca range 5-20)

o KD[Sr] = 0.223 + 0.0086 x T( C) although the latter is tentative at present. The above relationships are based on equilibrium conditions. For waters with substantially elevated trace-element ratios, particularly in the case of Mg/Ca, the valves will precipitate in out-of-equilibrium conditions in order to produce low-Mg calcite valves and hence may not follow these distribution coefficients. In addition, these authors demonstrate that there is no direct connection between the salinity of the hostwater and the trace element (either Mg or Sr) composition of the shell, unless the water is saturated with respect to calcium.

6. Stable-Isotope Analyses

The global average for the Sr/Ca molar ratio of seawater is 0.0086 ± 0.0004. At present the Sr/Ca of the Gulf of Carpentaria has not been determined. As the

Sr/Ca of seawater remains relatively constant, the Sr/Ca of ostracod valves should also be constant. In the example of Cyprideis, valve ratios of Sr/Ca approximately

0.0041 are indicative of marine influenced water, or at least waters with a similar

Sr/Ca ratio to seawater (De Deckker et al., 1988). The average value for Mg/Ca of seawater is 5 ± 0.8, corresponding to a Cyprideis valve ratio of around 0.0229

(at 25oC). Deviations from these values of Mg/Ca and Sr/Ca indicate a movement away from marine conditions, or larger temperature differences. In addition, hostwater Mg/Ca ratios of around 1, extrapolated from valve ratios, are indicative of freshwater conditions (De Deckker et al., 1988). As soon as waters become slightly saline (>3‰), the Mg/Ca deviates from unity (e.g. Eugster and Hardie,

1978). It should also be noted that freshwater can have Mg/Ca ratios other than 1

(De Deckker et al., 1988).

In contrast to the trace-element relationships, strontium isotopic ratios of biogenic calcite are unaffected by chemical fractionation and precipitation processes. As such the 87Sr/86Sr value of the ostracod valve should be directly equivalent to that of the host water at the time of shell formation. The strontium isotopic ratio of modern seawater is 0.70912 (DePaolo and Ingram, 1985), and may be considered to be constant over the time period investigated in this study. Continental waters return variable 87Sr/86Sr depending on their solute sources, with older bedrock such as in northern Australia generally producing higher ratios. The results of

McCulloch et al. (1989) express the 87Sr/86Sr values of ostracod valves measured as deviations from the seawater value as follows:

325

87 86 87 86 -1 4 ΔSr = {[( Sr/ Srmeas)( Sr/ Sr) ]-1}10

Essentially, waters of non-marine origin exhibit a different 87Sr/86Sr ratio from that of seawater. In response to transitional lacustrine-marine conditions, such as those outlined in the Gulf of Carpentaria, the strontium isotopic ratios recorded by the ostracods display clearly identifiable changes relative to seawater.

The trace element and strontium isotope records obtained from cores GC-2 and

-10A have been established by De Deckker et al. (1988) and McCulloch et al.

(1989) respectively, and are presented in Figure 6.12. As the two cores have different sedimentation rates, the depths from GC-10A have been adjusted for better comparison with GC-2. The inferred water trace-element ratios are determined from the distribution coefficients outlined above. The facies equivalents in the MD-cores have been outlined previously (Section 4.5) and enable comparison between the trace-element and strontium isotopic ratios of the

GC-cores with the oxygen and carbon stable-isotope ratios of the MD-cores

(Fig. 6.12).

6. Stable-Isotope Analyses

327

6. Stable-Isotope Analyses

These records present the majority of the lacustrine phase of the Carpentaria region and include the most recent marine transgression. The trace element results indicate that the site of core GC-2 was almost always more dilute than that of

GC-10, which the authors attribute to either increased freshwater inflow or rainfall to the former (De Deckker et al., 1988).

In the basal units V and IV, the Mg/Ca ratios indicate fresh to slightly saline waters. The Sr/Ca ratios however are close to seawater values, indicating a marine source for the lake water, which has been subsequently diluted. Within the top and base of unit IV, the range in values of Mg/Ca is narrowly spread (within 0.005) with a more broad distribution (up to 0.015) in the intervening samples. This may indicate either fluctuating lake levels, with shallow water and hence greater temperature effect in the middle of the unit, and/or more rapid deposition at the top and bottom of the unit. A change in water composition, perhaps due to the precipitation of CaCO3, has also been postulated accompanying the shallower conditions, as Sr/Ca ratios follow the trend of Mg/Ca through this unit.

The strontium isotopic results from unit V and IV of core GC-2 increases from

ΔSr = 6.1 (2.15 m) to ΔSr = 7.9 (1.89 m). McCulloch et al. (1989) put forward two hypotheses for these relatively low values. The first suggests drainage of the

Fly River into the Carpentaria Basin, rather than the present course into the Coral

Sea. This is based on the assumption that rivers draining the younger landmass of

Papua New Guinea would have lower 87Sr/86Sr ratios than those of the older, weathered Australian landmass. There is no evidence in the pollen record to

329

suggest influence of New Guinean waters at this time hence this argument seems unlikely. The second possibility suggests a minor incursion of marine water during or just prior to the deposition of unit V, followed by influence of

Australian continental waters. Both the ostracod assemblage data and the oxygen and carbon isotopic ratios for the equivalent facies in core MD-32 support this notion. Torgersen et al (1988) relate this to the sea-level high recorded from the

Huon Peninsula coral reef terraces of MIS 3 (~40 ka BP) (Chappell, 1983a;

Chappell and Shackleton, 1986). The improved evidence from the most recent gulf cores and the refined sea-level curves of Chappell et al (1996) and

Waelbroeck et al (2002) now associate this highstand incursion to the gulf to be around 52 ka BP.

Through unit III there is a marked enrichment in both Mg and Sr as a result of precipitation of substantial authigenic calcite during periods of anoxic conditions, most likely in shallow water. A two-fold increase in Sr/Ca and three-fold increase in Mg/Ca, some of which may be due to temperature increase, have been noted in the samples of this unit.

No strontium isotope ratios have been obtained from unit III, however analyses from the base of unit II have ΔSr = 14.8 (at 1.4 m), which mirrors the increase in trace-element ratios. These heavier values suggest input of Australian continental waters with high isotopic values. The enrichment of Sr/Ca for the same interval suggest an increase in salinity, or at least Sr-content of the lake water, exceeding that of the more dilute inflow.

6. Stable-Isotope Analyses

Through most of unit II there is little variation in the Sr/Ca ratios from the underlying unit. This implies no enrichment in Sr either directly or relative to Ca and hence no further substantial calcite precipitation. There is a steady decline in

Sr/Ca from around 1.1 m (GC-2), which may be due to a dilution of the lake water. McCulloch et al. (1989) suggest minor authigenic precipitation of aragonite to be the cause, however there is no positive evidence for this in the sedimentary record. The Mg/Ca ratios remain enriched after the calcite precipitation of the underlying unit, becoming increasingly heavier through much of unit II. Around

1.1 m in core GC-2, there is a broader spread of results as great as 0.015, with some significantly depleted in comparison to the rest of the unit, recording raios of as low as 0.005. This may implicate the inflow of more dilute waters or temperature variation, suggesting greater seasonality.

In tandem with the Sr/Ca ratios, the strontium isotopic values remain relatively constant through much of unit II, ranging from ΔSr = 12.2 to 14.7. Similar values were obtained from ostracods at an equivalent depth in core GC-10A. A well-established lacustrine environment, with a stable drainage pattern and little climatic variation has been inferred from the ΔSr results through this unit

(McCulloch et al., 1989).

It is difficult to compare the lacustrine phase in cores GC-2 and MD-32, as the sedimentation rate in the latter is significantly greater. Nevertheless there do appear to be some trends that may be extrapolated across the two cores between the Mg/Ca of GC-2 and the δ18O of MD-32. There is a broad positive correlation between the two parameters, suggesting variations in concentration and dilution of

331

solutes through the unit. Also the spread in δ18O results supports the notion of increased seasonality as evidenced in the Mg/Ca ratios. Perhaps most interesting is the peak in δ18O values around 1.7 m in MD-32, which correlates with the trough in Mg/Ca and in little change in Sr/Ca results at around 1.0 m in GC-2, suggesting a decrease in temperature. This may represent the height of the Last

Glacial Maximum in the gulf region. Evidence of a contraction of lake extent, including reworked marine material and iron-oxide stained flakes present in the sediment of core MD-32 from 2.2-1.7 m support this.

Above 80 cm (core GC-2), dated to around 14 ka BP (uncalibrated), both Mg/Ca and Sr/Ca show a broad range in values, as great as 0.020 and 0.006 respectively, with an overall negative trend. As both ratios are affected, a dilution of the lake water is supported. The strontium isotope results are the highest obtained for these cores of ΔSr = 17.2-18.0 at 0.56 m. This represents an estimation of the 87Sr/86Sr of the inflow to the lake at the time to be >0.71039 (McCulloch et al., 1989). The high ΔSr value is indicative of increased run-off from the Australian continent, which may be correlated with more pluvial conditions following the LGM.

De Deckker et al. (1988) suggested a reconnection to the sea during this interval in core GC-10A, based on the trace element ratios and the presence of some open marine fauna, however the results are not conclusive. The fauna may have been reworked from the surrounding exposed margins of the basin. The strontium isotope data give no evidence of marine waters in the lake at this time.

The oxygen isotope ratios of MD-32 (upper unit 2b and 2a) in particular, also note this trend in increased variability, which may be attributable to increased

6. Stable-Isotope Analyses

seasonality and freshwater influence around this time. This is supported in the microfaunal evidence first by a shell-rich layer spanning around 30 cm, with abundant bivalved molluscs, and also by the increase in freshwater ostracods above this layer. The onset and strengthening of the Australian monsoon may be implicated through this period.

The upper unit of both cores represents the marine transgression. Trace-element analyses of Cyprideis valves produced a very broad range of results, suggesting either that these valves were reworked from former deposits and not in situ or that they were formed under different conditions in a highly fluctuating, marginal environment. Results of Sr/Ca of open marine taxa from above 0.45 m in core

GC-2 produced consistent values around that of ostracod valves formed in seawater. Strontium isotopic ratios at 0.25 m (GC-2) show a dramatic decrease from the results obtained from the underlying unit, indicating the influence of marine waters. It should be noted however, that the ratios range from

ΔSr = 0.8-7.0, suggesting that a mixing of waters, rather than true marine conditions was in evidence at the time (McCulloch et al., 1989). Open marine Sr ratios have been obtained from valves sampled in the upper 2 cm (GC-2). The ostracod assemblage, sedimentological and stable isotope data for the unit support this interpretation.

It is thus shown with a sound understanding of the ostracod facies changes and post-depositional effects, within the framework of the sedimentology, the geochemistry of the ostracod valves can provide further information regarding the waterbody at the time of valve formation. Although there are limitations in the

333

data presented, when used in comparison with other proxies, such as the pollen record and previous geochemical data, inferences regarding variations in climate, particularly temperature and effective precipitation change may be extrapolated.

7. Synthesis

Palaeoenvironmental reconstruction of the Gulf of Carpentaria from the Last Interglacial to Present

The sedimentary record preserved in the core material extracted from the Gulf of Carpentaria presents an environmental history spanning the last glacial cycle. From the sedimentological, micropalaeontological and geochemical analyses performed through this study, conclusions about the sea-level record and climatic change through this period can be drawn. Comparisons may then be made between this and other regional records.

7.1 Sea level

The sequence of marine and non-marine deposition recorded in the Gulf of

Carpentaria cores provides independent sea-level data for the region. As the gulf is located in a tectonically relatively stable portion of the continental plate and far removed from glacial activity, tectonic and isostatic influences are minor. Owing to the large, shallow embayment of the gulf, the effect of subsidence due to water loading is minimal; hence deciphering broad sea-level variation is comparatively straightforward.

The main basin of the gulf, at a maximum modern depth of -70 m, is isolated from the Pacific Ocean by Torres Strait (-12 m bpsl) and the Indian Ocean by the

Arafura Sill (-53 m bpsl). These depths provide the key parameters for past sea-level estimations. Fossil assemblages of open marine fauna, such as are common in the deeper waters of the gulf today, are suggestive of both the Arafura

Sill and Torres Strait being open at the time of deposition. The presence of well- preserved shallow marine fauna in the core material is indicative of sea level above at least the Arafura Sill height, with the gulf being a large embayment open only to the Indian Ocean.

335

7. Synthesis

337

7. Synthesis

During sea-level regression, when the level of the sea was about or just below the height of the sill, incision of channels provided a tidal connection between the ocean and the basin, with established estuarine fauna. The absence of marine or estuarine fauna and presence of non-marine taxa identifies the basin as being restricted from the ocean and implies a sea level well below that of the sill height.

It should be noted that the sill is a sedimentary feature, subject to construction and erosion and as such its height is not fixed. Throughout the period under investigation, the sill is considered to have been between -60 to -50 m bpsl. At least two prior channels, incised to -75 m and -65 m bpsl and later infilled, have been identified from seismic sections across the sill (Jones and Torgersen, 1988).

As outlined above, the data retrieved from the Gulf of Carpentaria do not provide a complete range of sea-level variability through the last glacial cycle, however trends within the broad context of transgression and regression across the Arafura

Sill and Torres Strait are discernable. These may be compared with global sea- level reconstructions for this period, such as that of Waelbroeck et al. (2002).

These authors have constructed a composite relative sea-level curve based on the data derived from δ18O values of benthic foraminifers from the East Pacific core,

V19-30 (Shackleton et al., 1983; Shackleton and Pisias, 1985) and the North

Atlantic cores NA 87-22 (0-75 ka BP) (Vidal et al., 1997) and NA 87-25 (75-140 ka BP) (Cortijo et al., 1994, 1999), correlated to the SPECMAP benthic stack of

Martinson et al. (1987). Corrections have been made for temperature and local effects. Additional data have been contributed from U-Th dated coral and other records (e.g. Bard et al., 1990a,b; Stein et al., 1993; Zhu et al., 1993; Gallup et al.,

1994; Stirling et al., 1995; Chappell et al., 1996; Hanebuth et al., 2000;

339

Yokoyama et al., 2000). The marine isotope stages (MIS) are used as a temporal reference. The system of units defined for core MD-32 may be compared with the global sea-level curve (Fig. 7.1). A schematic of the changing shorelines and environments is represented in Fig. 7.2.

7. Synthesis

341

7. Synthesis

343

7. Synthesis

The lowermost unit 7, dated to around 125 ka BP, with abundant quartz and iron-oxide staining, was most likely deposited rapidly, immediately prior to the marine transgression of the last interglacial. The complete lack of marine microfauna favours a terrestrial origin for this material, by fluvial activity, rather than sheetwash flooding. As such the level of the sea is considered not to have breached the sill at this stage. There is evidence for at least two sea-level peaks, at around 135 ka BP and 120 ka BP, separated by a lowstand (Zazo, 1999) during

MIS 5e. The intervening drop in sea level may have been as low as -60 m, based on evidence from the Huon Peninsula (Esat et al., 1999) and Barbados (Gallup et al., 2002). It is possible that unit 7 was deposited during this lowstand period.

The presence of marine material is evident throughout unit 6. At least three periods of relatively high sea level are apparent (subunits 6f, 6d and 6b), separated by deposition in more restricted conditions. Although dates have not been directly established for this unit in the core, the variation in sea level, as noted by the faunal assemblage shifts, may be compared with the sea-level curve through

MIS 5. Subunit 6f is characterised by open shallow marine fauna. The presence of planktic forms and pteropods, deposited in situ, indicate transgression of the

Torres Strait at this stage, suggesting sea level to be greater than -10 m bpsl.

Maximum sea level in the gulf may have been at least +3 m, extrapolated from the surrounding shoreline ridges (Rhodes et al., 1980, 1982) and a coral reef (Nott,

1996) from the south of the gulf, dating to the last interglacial. This is consistent with the sea-level curve, which records heights of up to 6 m greater than modern sea level for MIS 5e.

345

The regression in sea level of MIS 5d may be noted through subunit 6e by the marginal marine fauna, indicating continued connection to the Arafura Sea. The gulf at this stage is represented by a large embayment, closed on the eastern margin. This is consistent with the estimations of Waelbroeck et al. (2002) of around -45 m bpsl for this lowstand.

A return to open shallow marine conditions is evident in subunit 6d, although not to the extent of subunit 6f. There is no positive evidence of Torres Strait having been breached at this stage, indicating that the sea level remained below about

-15 m. Findings from the Huon Peninsula identify the MIS 5c highstand to have been around -20 m bpsl (e.g. Chappell et al., 1996; Lambeck et al., 2002).

Restriction of marine waters is evident through subunit 6c, by the paucity of marine fauna and occurrence of gypseous laminae. This suggests an enclosed lagoon-type environment, indicating that sea level had dropped to around or just below the level of the Arafura Sill. This subunit is identified with MIS 5b and is in support of sea level of this substage being lower than that of 5d, at around

-50 m, as illustrated by Waelbroeck et al. (2002). It is also possible that the sill was higher at this stage, thus further restricting the marine influence.

The final highstand of the last interglacial is represented in subunit 6b. A similar microfaunal assemblage to that of subunit 6d and the core top samples is present.

The open shallow marine fauna attest to an open connection with the Arafura Sea.

Although planktic forms are absent from this subunit in core MD-32, there are fauna present which are common to the modern Pacific and northern Australian

7. Synthesis

waters, suggesting that there may have been some minor connection across Torres

Strait at this stage. A sea level similar to that of unit 6d, of at least -20 m or perhaps slightly higher is suggested. This is again consistent with the Huon

Peninsula data (Chappell et al., 1996).

Subunit 6a shows a gradual regression of marine waters, associated with the MIS

5a/4 transition around 74 ka BP. The concentration deposit at the base of the unit in core MD-32 may be an indication of a shoreline feature, representing the sea-level extent at the time of deposition. Many of the taxa above this layer show evidence of dissolution indicating exposure or reworking, supporting a sea-level retreat below the height of the sill.

A confined brackish waterbody, isolated from the open sea is evident through unit

5. The height of sea level below that of the sill and without channel influence is implicated. The abundance of microfauna and the presence of a shell layer suggest rapid deposition of this unit, most likely at the beginning of MIS 4 lowstand.

The core sediment of unit 4 shows no established connection with marine water and hence direct correlation with the sea-level curve is not possible, excepting to say that the level must have remained below that of the sill height. This is consistent with deposition through the lowstand of MIS 4 (74-59 ka BP). It is likely however that a palaeochannel across the sill, such as that identified from seismic profiles at a depth of -75 m bpsl (Jones and Torgersen, 1988), was active at this time. There is some evidence of a marine material in subunits 4d and 4c,

347

although it is considered to have been transported rather than deposited in situ.

The more robust material is most probably reworked from the surrounding exposed margin. The spherical Ammonia and planktic forams may have been transported in suspension some distance during king tides. Such taxa have been reported from 105 km inland along the South Alligator River, a tidal river of the

Northern Territory (Wang and Chappell, 2001). A brief marine incursion via channels across the sill is implicated in subunits 4b, concentrated in the deepest part of the basin. This may be associated with rising sea levels of the MIS 4/3 transition around 60 ka BP.

The return of permanent brackish water to the basin, isolated from open marine sources, is evident through subunit 3b. A bloom of fauna of low diversity, similar to that observed in unit 5, identifies the subunit and a comparable depositional environment is supported. Although marine influence is implicated in the initial infilling of the basin, sea level is again considered to have dropped below sill height. A hiatus in deposition with minor pedogenesis is evident following this subunit.

Subunit 3a is differentiated from 3b on the basis of the dominant fauna. Estuarine fauna are present, indicating a permanent, confined connection to the sea. The marine influence is greater than that of subunit 4b and a higher sea level is implicated. The sea-level curve of Waelbroeck et al. (2002) indicates higher sea level at the start of MIS 3, around 59 ka BP (i.e. subunit 4b and 4a). In contrast, data from the Huon Peninsula (Chappell et al., 1996) and from δ18O of sediment from the Red Sea (Siddall et al., 2003) suggest that sea level around 52 ka BP may

7. Synthesis

have been higher, reaching around -50 m bpsl. The data from the gulf support the latter interpretation.

Fluctuations in sea level in the order of 10-15 m occur over 6-7 ka intervals during the period 60-30 ka BP, associated with short-term melting of the Laurentide and

Greenland ice sheets known as Heinrich events recorded in the North Atlantic

(Bond et al., 1993; van Kreveld et al., 2000). These cycles have been identified in the Huon Peninsula record (Chappell, 2002) and may also be interpreted from the episodic marine influence on the otherwise enclosed gulf basin during this period.

Prior to the most recent marine transgression, subunit 3a represents the last instance of connection to the sea recorded in the gulf cores. The non-marine sediment of unit 2, with basal dates around 40 ka BP, is in agreement with the sea level being below the height of the sill. Again, other than being separated from the sea, the gulf sediments of this time provide no independent evidence for sea-level variation. The sea-level curve oscillates around -60 to -80 m bpsl

(Waelbroeck et al., 2002) through the remainder of MIS 3, dropping as low as

-125 m (Yokoyama et al., 2000, 2001) during the last glacial maximum

(23-19 cal. ka BP), before rapidly increasing to reach -60 m by around

12 cal. ka BP.

The first marine material associated with the most recent transgression is evident in the gulf core sediment toward the top of subunit 2a and has been dated to around 12 cal. ka BP, coincident with the MIS 2/1 transition. The initial marine material is reworked, most likely transported by channel or storm activity across

349

the sill. Material transitional between the marine and non-marine environment is in evidence in most cores, dating to between 11.8-10.5 cal. ka BP. This is contemporaneous with the Younger Dryas sea-level still-stand (Fleming et al.,

1998), which may indicate a pause in transgression around the height of the sill at this time. Well-preserved marine material, considered to have been deposited in situ, dates from 10.4 cal. ka BP. At this stage the Arafura Sill was clearly breached. Connection across Torres Strait occurred sometime after, perhaps around 8 cal. ka BP, although this has not been dated directly. These dates are in close correlation with the sea-level curve for MIS 1 (12-0 ka) (Waelbroeck et al.,

2002).

The Holocene record is not well constrained for the gulf core material. The uppermost unit of the cores is predominantly soupy material that has suffered poor recovery in some of the cores. In addition, sedimentation rates in the modern gulf environment are very low beyond the reach of the fluvial deltas. Chappell and

Thom (1986) have summarised the regional Holocene for northern Australia to have occurred in four phases: (i) prior to 8 ka BP, rising sea level at 12 m.ka-1; (ii)

8-6.5 ka BP, rising sea level at 4.5 m.ka-1; (iii) 6.5-6 ka BP, sea-level stabilisation;

(iv) 6.0 ka BP-present, stable or slowly falling sea level, typically 0.2 m.ka-1.

Local sea-level data indicate the Holocene optimum to have reached +2 m in the southern region of the gulf, such as at Karumba and +0.5-1.0 m in the north, around 6-5 ka BP (Nakada and Lambeck, 1989). Gradual recession to the present level continued through the remainder of the Holocene.

7. Synthesis

351

7. Synthesis

7.2 Climate and Hydrology

Beyond the direct impact of sea level fluctuations on the basin, the sediments of the gulf preserve a record of climatic variation. This may be determined from the hydrology of the basin, including the impact of the sea, the nature of the lacustrine phases being open or closed, the extent of the lake, periods of exposure and pedogenesis and the influence of river activity. Variations in temperature and effective precipitation throughout the last glacial cycle are inferred from the gulf cores and compared with records from the surrounding region (Figs. 7.3). Again the marine isotope stages are used as a temporal reference.

The oldest dated material returned from the gulf cores (unit 7) deposited around

125 ka BP, indicates dry climatic conditions in the basin. A small lake, confined to near the depocentre of the basin, with exposed and poorly vegetated margins may be postulated. The pollen record identifies this as the driest period encountered in the core material. Episodic alluvial activity and amelioration of climate are evident, associated with the rise in sea level. Prior to the onset of the

Last Interglacial, conditions during the sea-level lowstand of MIS 6 are considered to have been drier and cooler than present. The gulf region is expected to have been largely exposed at this stage. As such, channels crossing the exposed margins would have transported oxidised material. Today the tropical conditions of the gulf are highly oxidising. This is in evidence in the Gilbert floodplain, where heavily ferricreted surfaces may be shown to be only Late

Pleistocene in age (Jones et al., 1993).

353

Through MIS 5 (128-74 ka BP) marine conditions dominate the gulf material.

The fluctuations in sea level are observable in unit 6, as outlined above. In fact the sediments of the gulf primarily record the marine water depth and proximity to the shoreline through this period. The ostracod assemblage shifts and the isotopic record support this variation. A warm, humid, tropical environment may be inferred from the pollen record through MIS 5e-d, becoming drier or more seasonal through MIS 5a-c. Increased seasonality is also noted in the isotope record. Variation in fluvial activity is evident from the sedimentology, with apparent peaks during MIS 5d and 5a, which are indicative of both the evaporation/precipitation regime and the influence of the Australian monsoon.

The monsoon is recorded to have been active during this period at Lake Eyre, central Australia. Today this is a major ephemerally flooded playa, with a largely monsoon-dominated catchment. The lake is thought to have approached full conditions (i.e. up to 25 m deep) through much of the period 130-90 ka BP, indicating enhanced monsoon conditions (Magee et al., 1995; Croke et al., 1996).

The timing of peak effective precipitation is not clearly determined from the Lake

Eyre records, however the earlier stage 5 is considered to have been wetter

(Nanson et al., 1992; Kershaw and Nanson, 1993; Croke et al., 1996; Magee and

Miller, 1998).

Pollen data from the lake records of the Atherton Tablelands (e.g. Kershaw, 1986;

Kershaw et al., 1991), in comparison with modern bioclimatic data sets (Kershaw and Nix, 1990) suggest that effective precipitation was greater in later MIS 5, due to lower temperatures. This is supported by pollen evidence from the Banda Sea,

7. Synthesis

which reveal a peak in the precipitation signal of the Australian taxa during

MIS 5a (van der Kaars et al., 2000). Whereas the temperature record through this period broadly follows the stages defined by the sea-level curve, with higher temperature associated with higher sea level, effective precipitation appears out of phase (e.g. van der Kaars et al., 2000). The generally wetter signature of MIS 5 is associated in the first instance with the submergence of large portions of the

Sunda and Sahul shelves through this period, providing broad shallow warm seas for enhanced latent heat transport to the atmosphere.

The evidence for fluvial activity through unit 6 is controlled primarily by the position of the cores in relation to the shoreline. During the lowstand of MIS 5d

(subunit 6e of core MD-32), rivers are considered active, identified in both the micropalaeontological and isotopic records. A shallow, restricted, low-energy environment represents the lowstand of MIS 5b, in comparison, and a decrease in surface water may be postulated. Episodic fluvial input is apparent through

MIS 5a, represented by pulses of terrigenous silt through subunits 5b and 5a, within core MD-32, although these may be associated with the more marginal proximity of the core to the shoreline at this time.

An extensive sand body has been identified by Nanson et al. (1991; in review) underlying the Gilbert fan delta in the southeast of the gulf, dating older than

85 ka BP, and most likely around 120 ka BP. This is representative of a major fluvial period that has not been replicated in the region since. Other positive evidence for enhanced fluvial activity during MIS 5 is derived from the Lake Eyre catchment. Nanson et al. (1992) have dated fluvial deposits of the middle reaches

355

of the Cooper and Diamantina rivers, which are monsoon-fed and outflow to Lake

Eyre, to a mean age of 109 ± 4 ka BP. Similar results of fluvial activity between

110-100 ka BP have been obtained from Neales River, to the west of Lake Eyre

(Croke et al., 1996). These dates are in broad support of the increased fluvial influence described from the gulf cores.

A swamp-like environment remains after sea level has receded to below the sill height in unit 5, around 72 ka BP, as noted in the sedimentary, micropalaeontological and pollen records. The extent of the waterbody is confined to the -63 m contour. The basic composition of these waters remains dominated by Na-Cl, although the influence of freshwater is indicated by the presence of isolated oligohaline ostracods within the assemblage. A high degree of variability is noted by the extreme range of values in the isotopic record of this unit. Periods of flooding and evaporation may be inferred within a very shallow waterbody. There is some evidence of a short-lived lacustral event around this time at Lake Eyre (Magee et al., 1995), with channel filling and minor fluvial activity (Croke et al., 1996) although the monsoon is considered to be waning.

Drier climatic conditions may be discerned from the sediments of unit 4 deposited through the lowstand of MIS 4. The overall environment of the gulf is representative of mudflats with cycles of inundation and evaporation. Periodic channel connection across the Arafura Sill is implicated through the unit.

Subaerial exposure and pedogenesis is evident, with a permanent saline waterbody restricted to near the depocentre of the basin. This may be in the form of a groundwater dominant playa at this stage.

7. Synthesis

There has been comparatively little attention focussed on the terrestrial record through MIS 4. The Lake Eyre record indicates deflation, base-level lowering and aridity during this period, all of which suggest a weakened monsoon (Croke et al.,

1999). There is some evidence for ephemeral fluvial activity, although aeolian transportation is considered dominant (Nanson et al., 1998). Conditions in the

Atherton Tablelands are thought to be drier (Kershaw, 1978). This is supported by pollen records from the Banda Sea, which indicate that both northern

Australian and eastern Indonesia were drier and cooler during this period, with increased burning (van der Kaars et al., 2000).

An expansion of the waterbody in the gulf is apparent in subunit 4b, which may be coincident with a brief rise in sea level. Short-lived wetter conditions are also noted at this depth in the pollen record by the expansion of grasses. The quartz-rich, concretion-bearing subunit 4a is indicative of fluvial activity in the region; wet episodes in a generally drier environment. The presence of both iron-oxide rich and carbonate concretions in the same unit support changing climatic conditions through the formation of the sub-unit. An increase in sedges and decrease in overall pollen count through subunit 4a may be a function of preservation, rather than directly reflecting climate.

Variations in the hydrologic balance of the basin are apparent through unit 3. The establishment of a permanent brackish waterbody, perhaps due to the fluvial input and extending to the -59 m contour is indicated within the core material

(subunit 3b), before desiccation and pedogenic over-printing across the entire basin. The subsequent sea-level highstand enabled a tidal connection across the

357

sill, forming an estuarine waterbody extending to the -63 m contour (subunit 3a).

Infilling of the channel across the sill associated with this minor transgression is implied. The impact of minor marine incursions and regression is dominant through this period in the gulf, as the sea oscillated about the height of the sill.

Climatic parameters are less evident. However, the pollen record suggests wetter conditions during the brackish lake phase, before a high level of disturbance and increased charcoal, indicating exposure and dry conditions. A return to relatively wetter conditions is associated with the estuarine influence.

The regional evidence for climatic conditions through MIS 3 has produced some conflicting results. Sedimentological data from Lake Eyre indicate a major deflation episode dating around 60-50 ka BP, which marked the transition from the surface-water-dominated to groundwater-dominated basin, which is present today (Magee and Miller, 1998). Wetter conditions are then evident during

50-35 ka BP within the lake, although there is little evidence for fluvial activity within the catchment at this stage (Magee and Miller, 1998). Nanson et al. (1998) have dated material from stranded high beaches around the margins of the lake to around 55-40 ka BP that are in agreement with dates from Lake Frome (Gardner et al., 1987; Bowler and Magee, 1988) and suggest limited and episodic fluvial activity from the north (e.g. Nanson et al., 1991, 1992).

Evidence for the Australian monsoon being most effective, although highly variable, in the period 65-45 ka BP has been obtained from carbon isotopic analysis of fossil emu eggshell from Lake Eyre (Johnson et al., 1999). These data implicate the increase of C4 grasses, which are indicative of summer precipitation

7. Synthesis

or a monsoonal regime, through this period. Lakes Amadeus (Chen et al., 1990,

1991) and Lewis (Chen et al., 1995), in central Australia, also indicate wetter conditions, most likely monsoon-driven. The Atherton Tablelands pollen records also identified slightly wetter conditions from 50-40 ka BP (Kershaw, 1986).

Anthropogenic impact, particularly regarding the use of fire, has been suggested for the vegetation change in some of the records of this period (e.g. Kershaw,

1986; Johnson et al., 1999). A change in the pollen record is also noted from this time in a deep-sea core from northwestern Western Australia, however this has been attributed to a shift in boundary conditions of precipitation, as there is no evidence of increased charcoal (van der Kaars and De Deckker, 2002). A shift from a more effective monsoon in the north during early MIS 3 to more effective precipitation in the south through late MIS 3 is suggested. The gulf record supports this.

The waterbody, known as Lake Carpentaria, remained a permanent feature of the basin until the most recent marine transgression, spanning around 40-12 ka BP.

The margins of the lake were exposed, with rivers draining the continent extending up to hundreds of kilometres across the flats to reach the lake. An overall freshening-up and expansion of the lake is evident through MIS 3 and 2.

The lake may be considered closed through much of this period. Fluvial influence is implicated, as is more effective precipitation. The lake expands through unit 2 from around the -63 m contour to at least the -59 m contour. Effective precipitation rates through this time may have varied between 45-60% compared with modern conditions.

359

There is some evidence of a contraction of the lake back to around the -63 m contour and higher salinity water about the time of the LGM, however permanent water was maintained in the deeper region of the basin. Geochemical data in particular suggest that cooler temperatures and seasonal fluctuations in the water balance are indicated through this period. A greater temperature gradient between the inland and the tropics of Australia is implicated. In addition, a change in the meteorological regime, to dominance of the southeast tradewinds at the expense of the monsoon is strongly suggested. Both the shift in precipitation and the increased effectiveness of the westerly winds may explain the spread of results in the isotopic records.

Lake Eyre was dry through much of the period 35-10 ka BP (Magee and Miller,

1998). Playa deflation and the deposition of a thick halite crust are associated with aridity during the LGM (Magee et al., 1995). Conflicting results obtained by

Nanson et al. (1998) from TL dating of dunes around 26 and 22 ka BP suggest high lake levels during the end of MIS 3 and MIS 2, which may be attributable to short-lived, high flow events in northern Australia. Ephemeral fluvial deposits in the Neales River, inflowing to the west of Lake Eyre, were deposited between

30-16 ka (Croke et al., 1996). The sediment from the lake basin does not support this, although deflation may have removed the evidence (Nanson et al., 1998).

Evidence of fluvial activity in the north of Australia generally supports drier conditions through the LGM (Nanson et al., 1991; Nanson et al., 1992). A change in the river regimes to fan formation, braiding and reduction in activity is evident in northeast Queensland (Thomas et al., 2001). Evidence from plunge pools from

7. Synthesis

the north of Australia however, suggests major palaeofloods dating to 30-22 ka

BP (Nott and Price, 1999), which implicates some pluvial events, rather than monsoon-proper.

Low temperatures and reduced summer precipitation, i.e. no monsoon conditions are indicated from the carbon isotopic studies of the Lake Eyre basin during

28-15 ka BP (Johnson et al., 1999). Mean annual temperature of the Lake Eyre region for the period 45-16 ka BP has been estimated via AAR of emu eggshells to be at least 9oC cooler (Miller et al., 1997). Glacial evidence for cooling in New

Guinea is in the order of 5-6oC, with resultant decline in the altitude of the snowline in the order of 1000 m (Löffler, 1972; Hope, 1983).

In contrast to the terrestrial evidence, estimates of the sea surface temperature at the time of the LGM indicate only a minor drop in the tropical oceans

(e.g. CLIMAP, 1981; Barrows et al., 2000). The IPWP is thought to have existed through the LGM, although its extent may have been reduced. A decrease in temperature of no more than 2oC (Thunell et al., 1994) and increase in sea-surface salinity by about 1‰ (Martinez et al., 1997) has been recorded for this region.

Wang (1999) suggest that the discrepancy between the terrestrial and marine temperature records for the tropics may be due to increased seasonality in the marginal seas, which constitute a significant portion of the IPWP. The record of the contracted lake in the Carpentaria Basin supports this.

Perhaps the most significant regional feature of the sea-level lowstand of the

LGM is the increase in exposed landmass of the Sunda and Sahul shelves,

361

reducing the percentage of shallow warm waterbodies in the region and altering the pattern of oceanic circulation (Webster and Streten, 1978). This has implications for evaporation and latent heat transport, hence drier atmospheric conditions, weakened monsoon and decrease in cyclone activity and cloud cover.

In addition, diurnal temperature extremes are suggested to have been greater

(De Deckker et al., 2003).

A significant change to wetter conditions is evident from the gulf record around

14 ka BP, with the expansion of the lake and freshwater conditions becoming established. The lake at this stage may have expanded toward the -53 m contour, indicating run-off to have been around 60% of present values. A shift in solute composition with at least periodic input of more bicarbonate-dominant waters is associated with this change, as noted by the shift in ostracod fauna. More humid conditions and the onset of the monsoon are implicated. Supporting evidence for monsoon onset around this time is found in the northwestern coast of Australia

(van der Kaars and De Deckker, 2002), the Kimberley (Wywroll and Miller,

2001) and Indonesia (Grindrod et al., 1999). An increase in effective precipitation has also been noted from the Atherton Tablelands (Wywroll and Miller, 2001), although the pollen record indicates this to be a dry period, with conditions not improving until after 11 ka BP (Kershaw, 1983). De Deckker et al. (2003) suggest this recent onset to be associated with the flooding of the IPWP region with post-glacial sea-level rise. Recent evidence from combined δ18O and Mg/Ca analyses of foraminifers from the Makassar Strait indicate a rise in temperature of

o 3-4 C in the IPWP, synchronous with the global increase in atmospheric CO2 and

7. Synthesis

Antarctic warming, but prior to the melting of the northern hemisphere ice sheets

(Visser et al., 2003).

As outlined above, the most recent marine transgression was a stepped process.

The first evidence of marine material in the Gulf of Carpentaria is apparent from around 12 cal. ka BP throughout most of the basin. The Younger Dryas

(11-10 cal. ka BP) is not easily recognisable from the gulf material as this is coincident with the first evidence of the marine transgression. The establishment of a permanent connection to the sea across the Arafura Sill is considered to have occurred by 10.4 cal ka BP and Torres Strait around 8 cal. ka BP. Wetter conditions are associated with the rise in sea level. Subsequent to the mid-Holocene peak, sea level retreated to around the present position in the gulf.

The current climatic pattern was established shortly after this.

Although the Holocene record is not well represented in the gulf cores, climatic evidence may be gleaned from the region. Conditions throughout much of

Australia and Indonesia are generally considered to have been wetter in the early

Holocene as evidenced by fluvial (Nanson et al., 1991; Thomas et al., 2001), lacustrine (Bowler et al., 1976; Harrison, 1993; Magee et al., 1995) and pollen records (van der Kaars et al., 2001; van der Kaars and De Deckker, 2002). The monsoon is considered to be active through the Holocene, although there are conflicting views as to the date of onset (Markgraf et al., 1992; Johnson et al.,

1999). The climatic conditions of the gulf region today were emplaced during the mid-late Holocene (e.g. Shulmeister, 1992; Shulmeister and Lees, 1992).

363

7.3 Conclusion

The sequence of environmental change in the Gulf of Carpentaria, through the last glacial cycle has been presented in this study. In doing so, an independent sea-level record has been established for this region of northern Australia. This low-lying tropical basin is shown to have been a lake for a large percentage of the time period under investigation and hence provides a link between terrestrial and marine records. The extent and nature of the varying waterbody of the gulf has been postulated. Although the sediments of the gulf primarily record the influence of transgressive/regressive marine water, due to the broad and shallow nature of the basin, shorter-term variations may obscure some of the larger-scale change. The combination of proxy data employed provides a reliable reconstruction of the changing hydrology and climatic influences of the gulf, enabling comparison with other regional records.

The use of ostracodal facies analysis has provided a greater level of detail to the understanding of environmental change in the aquatic environments of the gulf than sedimentary analysis alone would permit. Further information on the energy of the environment and changes in the carbonate saturation of the water have been determined from the population dynamics of the ostracod assemblages and morphological variation of some key species, through the core sediment.

The stable isotope analyses have provided further insight into the composition of the host-water at the time of ostracod carapace formation. The influence of marine and fluvial water sources on the basin, and changes in the sensitivity of the record have been documented through the stable isotope and pollen records. This has

7. Synthesis

been related to climatic changes through the period such as the influence of the

Australian monsoon. Particular attention has been drawn to the major lacustrine phase of the gulf record. The comparison of the recent stable isotope data with previously established trace element and strontium isotopes records document hydrologic, climatic and environmental change in the basin through the period

40-12 ka BP. Influences of salinity and temperature change through this time have been postulated.

From the combined evidence presented above, the Gulf of Carpentaria has been shown to be a key area in understanding environmental and climatic change through the last glacial cycle in northern Australia. The location adjacent to the

Indo-Pacific Warm Pool has implications in particular for the forces which have driven the tropics, and hence the oceanic-atmospheric interaction of the region through this period. The use of ostracods has proven an invaluable tool for identifying the significant variations witnessed the Carpentaria Basin from the

Last Interglacial to present.

365

References

Aladin, N.V., 1993. Salinity tolerance, morphology and physiology of the osmoregulatory organ in Ostracoda with special reference to Ostracoda from the Aral Sea. In: McKenzie, K.G. and Jones, P.J. (Eds.) Ostracoda in the earth and life sciences. Balkema, Rotterdam, pp. 387-403. Allan, R.J., Nicholls, N., Jones, P.D. and Butterworth, I.J., 1991. A further extension of the Tahiti-Darwin SOI, early ENSO events and Darwin pressure. Journal of Climate 4, 743-749. Anadón, P., 1992. Composition of inland waters with marine-like fauna and inferences for a Miocene lake in Spain. Palaeogeography, Palaeoclimatology, Palaeoecology 99, 1-8. Anadón, P., Gliozzi, E. and Mazzini, I., 2002. Paleoenvironmental reconstruction of marginal marine environments from combined palaeoecological and geochemical analysis of ostracods. In J.A. Holmes and A.R. Chivas (Eds.) The Ostracoda: Applications in Quaternary Research. American Geophysical Union, Washington, D.C. pp. 227-247. Anadón, P., Utrilla, R. and Julià, R., 1994. Palaeoenvironmental reconstruction of a Pleistocene lacustrine sequence from faunal assemblages and ostracode shell geochemistry, Baza Basin, Spain. Palaeogeography, Palaeoclimatology, Palaeoecology 111, 191-205. Australian Water Resources Commission (1987) Review of Australia's Water Resources and Water Use. Volume 2: Water use data set. Department of Primary Industries and Energy, Canberra Bard, E., Hamelin, B. and Fairbanks, R.G., 1990a. U-Th ages obtained by mass spectrometry in corals from Barbados: sea level during the past 130000 years. Nature 346, 456-458. Bard, E., Hamelin, B., Fairbanks, R.G., Zindler, A., 1990b. Calibration of the 14C timescale over the past 30000 years using mass spectrometric U-Th ages from Barbados corals. Nature 345, 405-410. Barrows, T.T., Juggins, S., De Deckker, P., Thiede, J. and Martinez, J.I., 2000. Sea-surface temperatures of the southwest Pacific Ocean during the Last Glacial Maximum. Paleoceanography 15(1), 95-109. Bayly, I.A.E., 1970. Further studies on some saline lakes of south-east Australia. Australian Journal of Marine and Freshwater Research 21, 117-129. Bayly, I.A.E. and Williams, W.D., 1966. Chemical and biological studies on some saline lakes of south-east Australia. Australian Journal of Marine and Freshwater Research 17, 117-228. Behrens, P., 1991. Ostracoda (Crustacea) from Lizard Island, northern Great Barrier Reef, Australia. I. Families: Cytherellidae, Loxoconchidae, Cytherideidae, Cytheruridae, Paracytherideidae, Pectocytheridae, Krithidae, Cytheromatidae, Bythocytheridae, Cytheridae. Helogoländer Meeresuntersuch 45, 107-142. Benson, L. and Paillet, F., 2002. HIBAL: a hydrologic-isotopic-balance model for application to paleolake systems. Quaternary Science Reviews 21, 1521-1539. Benson, R.H., 1964. Recent marine podocopid and platycopid ostracodes of the Pacific. Publicazione della Stazione Zoologica di Napoli 33 (S), 387-420.

Benson, R.H., 1988. Ostracods and palaeoceanography. In: De Deckker, P., Colin, J.-P., Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp. 1-26. Blake, D.H. and Ollier, C.D., 1969. Geomorphological evidence of Quaternary tectonics in southwestern Papua. Revue de Geomorphologie Dynamique 19, 28-32. Blake, T., Burgess, I.R., Stewart, O.C. and Low, E., 1984. Duyken-1 well completion report. Permit NT/P30, Carpentaria Basin, Northern Territory, Australia. Canada Northwest Australia Oil N.L. (held as Northern Territory Geological Survey, unpublished Report PR85/31B; and as Geoscience Australia unpublished Report 84/577). Bloom, A. L., Broecker, W. S., Chappell, J. M. A., Matthews, R. K., and Mesolella, J. J., 1974. Quaternary sea-level fluctuation on a tectonic coast: New 230Th/234U dates from the Huon Peninsula, New Guinea. Quaternary Research, 6,185-205 Brady, G.S., 1880. Report on the Ostracoda. In: Report on the scientific results of the voyage of the H.M.S. Challenger during the years 1873-76; Zoology 1(3), 1-184. Bode, L. and Mason, 1994. Tidal modelling in Torres Strait and the Gulf of Papua. In: Bellwood, O., Choat, H. and Saxena, N. (Eds.), Recent advances in Marine Science and Technology '94. PACON International and James Cook University of North Queensland, Townsville, pp. 55-65. Bodergat, A.M., Carbonnel, G., Rio, M. and Keyser, D., 1993. Chemical composition of Leptocythere psammophila (Crustacea: Ostracoda) as influenced by winter metabolism and summer supplies. Marine Biology 117, 53-62. Bonaduce, G., Masoli, M. and Pugliese, N., 1976. Benthic ostracod from the Gulf of Aqaba (Red Sea). Publicazione della Stazione Zoologica di Napoli 40, 372-428. Bonaduce, G., Masoli, M., Mincheli, G. and Pugliese, N., 1980. Some new benthic ostracod species from the Gulf of Aqaba (Red Sea). Bullettino della Societa Paleontologica Italiana 19(1), 143-178. Bond, G., Broecker, W.S., Johnsen, S. McManus, J., Labeyrie, L., Jouzel, J. and Bonani, G., 1993. Correlations between climate records from North Atlantic sediments and Greenland ice. Nature 365, 143-147. Boomer, I, 1998. The relationship between meiofauna (Ostracoda, Foraminifera) and tidal levels in modern intertidal environments of North Norfolk: A tool for palaeoenvironmental reconstruction, Bulletin of the Geological Society of Norfolk, 46, 17-29. Boomer, I. and Eisenhauer, G., 2002. Ostracod faunas as palaeoenvironmental indicators in marginal marine environments. In: Holmes, J.A. and Chivas, A.R. (Eds.) The Ostracoda: Applications in Quaternary Research. AGU Mon. Ser., pp 135-150. Bowler, J.M., 1986. Quaternary landform evolution. In: D. Jeans (Ed.), A Geography of Australia. University of Sydney Press, Sydney, pp. 117-147. Bowler, J.M., Hope, G.S., Jennings, J.N., Singh, G. and Walker, D., 1976. Late Quaternary climates of Australia and New Guinea. Quaternary Research 6, 359-394.

367

Bowler, J.M. and Magee, J.W., 1988. Lake Frome: stratigraphy, facies analysis, and evolution of a major playa. In: SLEADS Conference, 1988. Research School of Earth Sciences, Australian National University, Canberra, pp. 4-6. Bowman, T.E. and Abele, L.G., 1982. Classification of the Recent Crustacea. In: Abele, L.G. (Ed.), The Biology of Crustacea. Volume 1, Systematics, the fossil record and biogeography. Academic Press, New York, London, pp. 1-27. Brady, G.S., 1868. Ostracoda. In: De Folin, L. and Perier, L. (Eds.) Les Fonds de la Mer. Volume 1, Savy, Paris, pp. 1-316. Brady, G.S., 1880. Report on the Ostracoda. In: Report on the scientific results of the voyage of the H.M.S. Challenger during the years 1873-76; Zoology 1(3), 1-184. Bridgwater, N.D., Heaton, T.H.E. and O'Hara, S.L., 1999. A late Holocene palaeolimnological record from central Mexico, based on faunal and stable-isotope analysis of ostracod shells. Journal of Paleolimnology 22, 383-397. Brouwers, E.M., 1988. Sediment transport detected from the analysis of ostracod population structure: an example from the Alaska continental shelf. In: De Deckker, P., Colin, J.-P. and Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp. 231-244. Burford, M.A., Rothlisberg, P.C. and Wang, Y.C., 1995. Spatial and temporal distribution of tropical phytoplankton species and biomass in Gulf of Carpentaria, Australia. Marine Ecology Progress Series 118, 255-266. Burgess, I.R., 1984. Carpentaria Basin: A regional analysis with reference to hydrocarbon potential. APEA (Australian Petroleum Exploration Association) Journal 24 (1), 7-18. Cadot, H.M., van Schmus, W.R. and Kaesler, R.L., 1972. Magnesium in calcite of marine Ostracoda. Bulletin of the Geological Society of America 83, 3519-3522. Cadot, H.M. and Kaesler, R.L., 1977. Magnesium content of calcite in carapaces of benthic marine Ostracoda. University of Kansas Paleontological Contributions 87, 1-23. Cann, J.H. and De Deckker, 1981. Fossil Quaternary and living Foraminifera from athalassic (non-marine saline lakes, southern Australia. Journal of Paleontology 55, 660-670. Carbonel, P., 1988. Ostracods and the transition between fresh and saline waters. In: De Deckker, P., Colin, J.-P. and Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp.157-174. Carbonel, P., Colin, J.-P., Danielpol, D.L., Loffler, H. and Neustrueva, I., 1988. Paleoecology of limnic ostracodes: A review of some major topics. Palaeogeography, Palaeoclimatology, Palaeoecology 62, 413-461. Carbonel, P. and Hoibian, T., 1988. The impact of organic matter on ostracods from an equatorial deltaic area, the Mahakam Delta – Southeastern Kalimantan. In: Hanai, T., Ikeya, N. and Ishizaki, K. (Eds.) Evolutionary Biology of Ostracoda, its fundamentals and applications. Kodansha, Tokyo, pp. 353-366.

Carbonel, P. and Peypouquet, J.P., 1983. Ostracoda as indicators of ionic concentrations and dynamic variations: methodology (Lake Bogoria, Kenya). In: Maddocks, R.F. (Ed.) Applications of Ostracoda. University of Houston, Houston, pp. 264-276. Cerling, T.E., Bowman, J.R. and O'Neil, J.R., 1993. An isotopic study of a fluvial- lacustrine sequence: the Plio-Pleistocene Koobi Fora sequence, East Africa. Palaeogeography, Palaeoclimatology, Palaeoecology 63, 335-356. Chappell, J., 1974. Geology of coral terraces, Huon Peninsula, New Guinea: A study of Quaternary tectonic movements and sea-level changes. Geological Society of America Bulletin 85, 553-570. Chappell, J. 1983a. A revised sea-level record for the last 300,000 years from Papua New Guinea. Search 14 (3-4), 99-101. Chappell, J., 1983b. Evidence for smoothly falling sea level relative to north Queensland, Australia, during the past 6,000 yr. Nature 302, 406-408. Chappell, J., 2002. Sea level changes forced ice breakouts in the Last Glacial cycle: new results from coral terraces. Quaternary Science Reviews 21, 1229-1240. Chappell, J. and Shackleton, N.J., 1986. Oxygen isotopes and sea level. Nature 324, 137-140. Chappell, J. and Thom, B.G., 1986. Coastal morphodynamics in North Australia: review and prospect. Australian Geographical Studies 24, 110-127. Chappell, J., Omura, A., Ezat, T., McCulloch, M., Pandolfi, J., Ota, Y. and Pillans, B., 1996. Reconciliation of late Quaternary sea levels derived from coral terraces at Huon Pensinsula with deep-sea oxygen isotope records. Earth and Planetary Science Letters 141, 227-236. Chen, X.Y., Bowler, J.M. and Magee, J.W., 1991. Aeolian landscapes in central Australia: gypsiferous and quartz dune environments from Lake Amadeus. Sedimentology 38, 519-538. Chen, X.Y., Chappell, J.M.A. and Murray, A.S., 1995. High (ground) water levels and dune development in central Australia: TL dates from gypsum and quartz dunes from around Lake Lewis (Napperby), Northern Territory. Geomorphology 11, 311-322. Chen, X.Y., Prescott, J.R. and Hutton, J.T., 1990. Thermoluminescence dating on gypseous dunes of Lake Amadeus, central Australia. Australian Journal of Earth Sciences 37, 93-101. Chivas, A.R., De Deckker, P., Cali, J.A., Chapman, A., Kiss, E. and Shelley, J.M.G., 1993. Coupled stable-isotope and trace-element measurements of lacustrine carbonates as palaeoclimatic indicators. In: P.K. Swart, K.C. Lohmann, J.A. McKenzie and S. Savin (Eds.) Climatic Change in Continental Isotope Records. American Geophysical Union, Geophysical Monograph 78, pp. 113-121. Chivas, A.R., De Deckker, P. and Shelley, J.M.G., 1983. Magnesium, strontium and barium partitioning in non-marine ostracode shells and their use in palaeoenvironmental reconstructions - a preliminary study. In: R.F. Maddocks (Ed.) Applications of Ostracoda. The University of Houston, Houston, pp. 238-249. Chivas, A.R., De Deckker, P. and Shelley, J.M.G., 1985. Strontium content of ostracods indicates lacustrine palaeosalinity. Nature 316: 251-253.

369

Chivas, A.R., De Deckker, P. and Shelley, J.M.G., 1986a. Magnesium and strontium in non-marine ostracod shells as indicators of palaeosalinity and palaeotemperature. Hydrobiologia 143, 135-142. Chivas, A.R., De Deckker, P. and Shelley, J.M.G., 1986b. Magnesium content of non-marine ostracod shells: a new palaeosalinometer and palaeothermometer. Palaeogeography, Palaeoclimatology, Palaeoecology 54, 43-61. Chivas, A.R., De Deckker, P., Wang, S.X. and Cali, J.A., 2002. Oxygen-isotope systematics of the nektic ostracod Australocypris robusta. In J.A. Holmes and A.R. Chivas (Eds.) The Ostracoda: Applications in Quaternary Research. American Geophysical Union, Washington, D.C., pp. 301-313. Chivas, A.R., Garcia, A., van der Kaars, S., Couapel, M.J.J., Holt, S., Reeves, J.M., Wheeler, D.J., Switzer, A.D., Murray-Wallace, C.M., Banerjee, D., Price, D.M., Wang, S.X., Pearson, G., Edgar, N.T., Beaufort, L., De Deckker, P., Lawson, E., Cecil, C.B., 2001. Sea-level and environmental changes since the Last Interglacial in the Gulf of Carpentaria, Australia: An overview. Quaternary International 83-85, 19-46. Church, J.A. and Forbes, A.M.G.,1981. Non-linear model of tides on the Gulf of Carpentaria. Australian Journal of Marine and Freshwater Research 32, 685-697. Clarke, J.D.A., Bone, Y., Cann, J.H., Davies, M., MacPhail, M.K. and Wells, F., 2001. Post-glacial biota from the inner part of southwest Joseph Bonaparte Gulf. Australian Journal of Earth Sciences 48, 63-79. CLIMAP Project members, 1981. Seasonal reconstructions of the earth's surface at the Last Glacial Maximum. Geological Society of America, Map Chart Series MC-36, 18pp., 9 maps. Commonwealth Bureau of Meteorology, 2004. Retrieved 23/04/2004. http://www.bom.gov.au/climate/map/climate_avgs/clim_avg1.shtml Coplen, T.B., Kendall, C. and Hopple, J., 1983. Comparison of stable isotope reference samples. Nature 302, 236-238. Corrège, T. and De Deckker, P., 1997. Faunal and geochemical evidence for changes in intermediate water temperature and salinity in the western Coral Sea (northeast Australia) during the Late Quaternary. Palaeogeography, Palaeoclimatology, Palaeoecology 131, 183-205. Cortijo, E., Duplessy, J.C., Labeyrie, L., Leclaire, H., Duprat, J. and van Weering, T.C.E., 1994. Eemian cooling in the Norwegian Sea and North Atlantic Ocean preceding continental ice-sheet growth. Nature 371, 446-449. Cortijo, E., Lehman, S., Keigwin, L.D., Chapman, M., Paillard, D. and Labeyrie, L., 1999. Changes in meridional temperature and salinity gradients in the North Atlantic Ocean (32o to 72oN) during the last interglacial period. Paleoceanography 14, 23-33. Craig, H., 1961. Isotopic variations in meteoric waters. Science 133, 1702-1703. Craig, H., 1965. The measurement of oxygen isotope paleotemperatures. In: Tongiorgi, E. (Ed.) Stable isotopes in oceanographic studies and paleotemperatures. Pisa, Consiglio Nazionale delle Richerche Laboratorio di Geologia Nucleare, pp. 161-182. Croke, J., Magee, J. and Price, D., 1996. Major episodes of Quaternary activity in the lower Neales River, northwest of Lake Eyre, central Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 124, 1-15.

Croke, J.C., Magee, J.W. and Wallensky, E.P., 1999. The role of the Australian Monsoon in the western catchment of Lake Eyre, central Australia, during the Last Interglacial. Quaternary International 57/58, 71-80. Cronin, T.M., 1983. Ostracods and sea-level changes: Case studies from the Quaternary of North and South Carolina, U.S. Atlantic Coast. In: Maddocks, R.F. (Ed.) Application of Ostracoda. University of Houston, Houston, pp. 77-88. Cronin, T.M., 1988. Ostracods and sea-level changes: case study from the Quaternary of North and South Carolina, U.S. Atlantic coast. In: De Deckker, P., Colin, J.-P., Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp. 77-88. Curry, B.B., 1999. An environmental tolerance index for ostracodes as indicators of physical and chemical factors in aquatic habitats. Palaeogeography, Palaeoclimatology, Palaeoecology 148, 51-63. Curry, B.B., Anderson, T.F. and Lohmann, K.C., 1997. Unusual carbon and oxygen isotopic ratios of ostracodal calcite from last interglacial (Sangamon episode) lacustrine sediment in Raymond Basin, Illinois, USA. Journal of Paleolimnology 17, 421-435. Curtis, J.H., Brenner, M., Hodell, D.A., Balser, R.A., Islebe, G.A. and Hooghiemstra, H., 1998. A multi-proxy study of Holocene environmental change in the Maya Lowlands of Peten, Guatemala. Journal of Paleolimnology 19, 139-159. Curtis, J.H., Brenner, M. and Hodell, D.A., 1999. Climate change in the Lake Valencia Basin, Venezuela, ~12600 yr BP to present. The Holocene 9(5), 609-619. Curtis, J.A. and Hodell, D.A., 1993. An isotopic and trace element study of ostracods from Lake Miragoane, Haiti: A 10, 500 year record of paleosalinity and paleotemperature changes in the Caribbean. In: P.K. Swart, K.C. Lohmann, J.A. McKenzie and S. Savin (Eds.) Climatic Change in Continental Isotope Records. American Geophysical Union, Geophysical Monograph 78, pp. 135-152. Danielopol, D.L. and Handl, M., 1990. The carapace preservation of Cytherissa lacustris; its possible usage for paleological reconstructions. In: Danielopol, P., Carbonel, P. and Colin, J.-P. (Eds.) Cytherissa (Ostracoda), the Drosophila of paleolimnology (Scientific results of a multidisciplinary project on recent and fossil Ostracoda), Bull. Inst. Geol. Bassin d'Aquitaine 47-48, 239-246. Dansgaard, W., 1964. Stable isotopes in precipitation. Tellus 16, 436-468 Dean, W.E. and Schwalb, A., 2000. Holocene environmental and climatic change in the Northern Great Plains as recorded in the geochemistry of sediments in Pickerel Lake, South Dakota. Quaternary International 67, 5-20. De Deckker, P., 1975. Determination of an ostracod collection in relation to Australocypris (Cyprididae). Australian Journal of Marine and Freshwater Research 26, 423-424. De Deckker, P., 1977. The distribution of the "giant" ostracods (Family: Cyprididae Baird, 1845) endemic to Australia. In: Löffler, H. and Danielopol, D. (Eds.) Aspects of the Ecology and Zoogeography of Recent and Fossil Ostracoda. Dr. W. Junk, The Hague, pp. 285-294

371

De Deckker, P., 1981a. Taxonomy and ecological notes of some ostracods from Australian inland waters. Transactions of the Royal Society of South Australia 105(3), 91-138. De Deckker, P., 1981b. Ostracoda from Australian inland waters – notes on taxonomy and ecology. Proceedings of the Royal Society of Victoria 93, 43-85. De Deckker, P., 1981c. Taxonomic notes on some Australian ostracods with descriptions of new species. Zoologica Scripta 10, 37-55. De Deckker, P., 1981d. Ostracods of athalassic lakes. Hydrobiologia 81, 131-144. De Deckker, P., 1982a. Holocene ostracods, other invertebrates and fish remains from cores of four maar lakes in southeastern Australia. Proceedings of the Royal Society of Victoria 94 (4), 183-220. De Deckker, P., 1982b. On Cypretta yaginga De Deckker sp.nov. Stereo-Atlas of Ostracod Shells 9(23), 133-140 De Deckker, P., 1983a. Terrestrial ostracods in Australia. In: Lowry, J.K. (Ed.), Papers from the Conference on Evolution of Crustacea. Australian Museum Memoir 18, 87-100. De Deckker, P., 1983b. Notes on the ecology and distribution of non-marine ostracods in Australia. Hydrobiologia 106, 223-234. De Deckker, P., 1988. An account of techniques using ostracodes in palaeolimnology in Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 62, 463-475. De Deckker, P., 2001. Late Quaternary cyclic aridity in tropical Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 170, 1-9. De Deckker, P., 2002. Ostracod Palaeoecology. In: Holmes, J.A. and Chivas, A.R. (Eds.) The Ostracoda: Applications in Quaternary Research. AGU Mon. Ser., pp. 121-134. De Deckker, P., Chivas, A.R. and Shelley, J.M.G., 1999. Uptake of Mg and Sr in the euryhaline ostracod Cyprideis determined from in vitro experiments. Palaeogeography, Palaeoclimatology, Palaeoecology 148, 105-116. De Deckker, P., Chivas, A.R., Shelley, J.M.G., Torgersen, T., 1988. Ostracod shell chemistry: A new palaeoenvironmental indicator applied to a regressive/transgressive record from the Gulf of Carpentaria, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 66, 231-241. De Deckker, P. and Corrège, T., 1991. Late Pleistocene record of cyclic eolian activity from tropical Australia suggesting the Younger Dryas is not an unusual climatic event. Geology 19, 602-605. De Deckker, P. and Forester, R.M., 1988. The use of ostracods to reconstruct continental palaeoenvironmental records. In: De Deckker, P., Colin, J.-P. and Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp. 175-199. De Deckker, P. and Geddes, M.C., 1980. Seasonal fauna of ephemeral saline lakes near the Coorong Lagoon, South Australia. Australian Journal of Marine and Freshwater Research 31, 677-699. De Deckker, P. Tapper, N.J. and van der Kaars, S., 2003. The status of the Indo- Pacific Warm Pool and adjacent land at the Last Glacial Maximum. Global and Planetary Science 35(1-2), 25-35. De Deckker, P. and Yokoyama, Y., in review. Micropalaeontological evidence for Late Quaternary sea-level changes in the Bonaparte Gulf, Australia. Australian Journal of Earth Sciences.

Delorme, L.D., 1969. Ostracodes as Quaternary palaeoecological indicators. Canadian Journal of Earth Sciences 6, 1471-1476. Delorme, L.D., 1971. Paleoecological determinations using freshwater ostracodes. In: Oertli, H.J. Paléoécologie des Ostracodes. Bull. Centre Rech. Pau- SNPA, 5 suppl, pp. 341-347. Delorme L.D., Zoltau, S.C. and Kalas, L.L., 1977. Freshwater shelled invertebrate indicators of paleoclimate in northwestern Canada during late glacial times. Canadian Journal of Earth Sciences 14, 2029-2046. DePaolo, D.J. and Ingram, B.L., 1985. High-resolution stratigraphy with strontium isotopes. Science 227, 938-941. Dettman, D.L. and Lohmann, K.C., 1993. Seasonal change in Paleogene surface water δ18O: fresh-water bivalves of western North America. In: P.K. Swart, K.C. Lohmann, J.A. McKenzie and S. Savin (Eds.) Climatic Change in Continental Isotope Records. American Geophysical Union, Geophysical Monograph 78, pp. 153-164. Dewi, K.T., 1997. Ostracoda from the Java Sea, west of Bawean Island, Indonesia. Marine Geological Institute, special publication No. 4. Ministry of Mines and Energy, Directorate General of Geology and Mineral Resources, Bandung. Dewi, K.T., 2000. Distribution of Recent Ostracoda Off Tg. Selatan, South Kalimantan, Indonesia. Bulletin of the Marine Geology, 15(1), 1-18. Didié, C., 2001. Late Quaternary climate variations recorded in North Atlantic deep-sea benthic ostracodes. Polarforschung Meeresforschung, 390. Didié, C. and Bauch, H.A., 2002. Implications of Upper Quaternary stable isotope records of marine ostracodes and benthic foraminifers for paleoecological and paleoceanographical investigations. In J.A. Holmes and A.R. Chivas (Eds.) The Ostracoda: Applications in Quaternary Research. American Geophysical Union, Washington, D.C. pp. 279-299. Doutch, H.F., Smart, J., Grimes, K.G., Needham, R.S. and Simpson, C.J., 1972. Progress report on the geology of the central Carpentaria Basin. Bureau of Mineral Resources, Australia, Record 1972/64 (unpubl.). Doutch, H.F., 1972. The paleogeography of the northern Australia and New Guinea and its relevance to the Torres Strait area. In: Walker, D. (Ed.), Bridge and Barrier, The Natural and Cultural History of Torres Strait. Research School of Pacific Studies, Department of Biogeography and Geomorphology, The Australian National University, Canberra, ACT, Publication BG/3, pp. 1-10. Drummond, C.N., Patterson, W.N. and Walker, J.C.G., 1995. Climatic forcing of carbon-oxygen isotopic covariance in temperate-region marl lakes. Geology 23, 1031-1034. Durazzi, J.T., 1977. Stable isotopes in the ostracod shell: a preliminary study. Geochimica et Cosmochima Acta 41, 1168-1170. Dwyer, G.S., Cronin, T.M., Baker, P.A., Raymo, M.E., Buzas, J.S. and Corrège, T., 1995. North Atlantic deepwater temperature change during Late Pliocene and Late Quaternary climatic cycles. Science 270, 1347-1351.

373

Edgar, N.T., Cecil, C.B., Mattick, R.E., Chivas, A.R., De Deckker, P. and Djajadihardja, Y.S., 2003. A modern analogue for tectonic, eustatic, and climatic processes in cratonic basins: Gulf of Carpentaria, northern Australia. In: C.B. Cecil and N.T. Edgar (Eds.), Climate Controls on Stratigraphy: SEPM Special Publication 77, pp. 193-205. Emiliani, C., 1955. Pleistocene temperatures. Journal of Geology 63, 538-578. Epstein, S., Buchsbaum, R., Lowenstam, H.A. and Urey, H.C., 1953. Revised carbonate-water isotopic temperature scale. Bulletin of the Geological Society of America 64, 1315. Esat, T.M., McCulloch, M.T., Chappell, J., Pillans, B. and Omura, A., 1999. Rapid fluctuations in sea level recorded at Huon Peninsula during the penultimate deglaciation. Science 283, 197-201. Eugster, H.P. and Hardie, L.A., 1978. Saline lakes. In: A. Lerman (Ed.) Lakes, Chemistry, Geology, Physics. Springer, New York, 237-293. Eugster, H.P. and Jones, B.F., 1979. Behaviour of major solutes during closed- basin evolution. American Journal of Science 272, 609-631. Evans, J.L. and Allan, R.J., 1992. El Niño /Southern Oscillation modification to the structure of the monsoon and tropical cyclone activity in the Australasian region. International Journal of Climatology 12, 611-623. Fairbanks, R.G., 1989. A 17, 000-year glacio-eustatic sea level record: influence of glacial melting rates on the younger dryas event and deep-ocean circulation. Nature 342, 637-642. Fairbanks, R.G. and Matthews, R.K., 1978. The marine oxygen isotope record in Pleistocene coral, Barbados, West Indies. Quaternary Research 10, 181-196. Fairbridge, R.W., 1953. The Sahul Shelf - northern Australia - its structural and geological relationships. Journal of the Royal Society of Western Australia 37, 1-33. Filippi, M.L., Lambert, P., Hunziker, J., Kübler, B. and Bernasconi, S., 1999. Climatic and anthropogenic influence on the stable isotope record from bulk carbonates and ostracodes in Lake Neuchâtel, Switzerland, during the last two millennia. Journal of Paleolimnology 21, 19-34. Fleming, K., Johnston, P., Zwartz, D., Yokoyama, Y., Lambeck, K. and Chappell, J., 1998. Refining the eustatic sea-level curve since the Last Glacial Maximum using far- and intermediate-field sites. Earth and Planetary Science Letters 163, 327-342. Fontes, J.C., Gasse, F., Gilbert, E., 1996. Holocene environmental changes in Lake Bangong basin (Western Tibet), Part 1: Chronology and stable isotopes of carbonates of a Holocene lacustrine core. Palaeogeography, Palaeoclimatology, Palaeoecology 120, 20-47. Forester, R.M., 1983. Relationship of two lacustrine ostracods species to solute composition and salinity: Implications for paleohydrology. Geology 11, 435-438. Forester, R.M., 1986. Determination of the dissolved anion composition of ancient lakes from fossil ostracodes. Geology 14, 796-798. Forester, R.M. and Brouwers, E.M., 1985. Hydrochemical parameters governing the occurrence of estuarine and marginal estuarine ostracodes: an example from south-central Alaska. Journal of Paleontology 59, 344-369.

Friedman, I. and O'Neil, J.R., 1977. Compilation of stable isotope fractionation factors of geochemical interest. US Geological Survey Professional Paper (440-KK). Fritz, P., Anderson, T.W. and Lewis, C.F.M., 1975. Late-Quaternary climatic trends and history of Lake Erie from stable isotope studies. Science 190, 267-269. Fritz S.C., Engstrom, D.R. and Haskell, B.J., 1994. Little Ice Age aridity in the North American Great Plains: A high-resolution reconstruction of salinity fluctuations from Devils Lake, North Dakota, USA. The Holocene 4, 69-73. Galloway, R.W. and Löffler, E., 1972. Aspects of geomorphology and soils in the Torres Strait region. In: Walker, D. (Ed.), Bridge and Barrier, The Natural and Cultural History of Torres Strait. Research School of Pacific Studies, Department of Biogeography and Geomorphology, The Australian National University, Canberra, ACT, Publ. BG/3, pp. 11-28. Gallup, C.D., Cheng, H., Taylor, F.W. and Edwards, R.L., 2002. Direct determination of the timing of sea level change during Termination II. Science 295, 310-313. Gallup, C.D., Edwards, R.L and Johnson, R.G., 1994. The timing of high sea levels over the past 200,000 years. Science 263, 796-800. Gamenick, I., Jahn, A., Vopel, K. and Giere, O., 1996. Hypoxia and sulphide as structuring factors in a macrozoobenthic community on the Baltic Sea shore: colonisation studies and tolerance experiments. Marine Ecology Progress Series 144: 73-85. Gardner, G.J., Mortlock, A.J., Price, D.M., Readhead, M.L. and Wasson, R.J., 1987. Thermoluminescence and radiocarbon dating of Australian desert dunes. Australian Journal of Earth Sciences 34, 343-357. Gasse, F., Fontes, J.C., Plaziat, J.C., Carbonel, P., Kaczmarska, I., De Deckker, P., Soulie-Marsché, I., Callot, Y. and Dupeuble, P.A., 1987. Biological remains, geochemistry and stable isotopes for the reconstruction of environmental and hydrological changes in the Holocene lakes from North Sahara. Palaeogeography, Palaeoclimatology, Palaeoecology 60, 1-46. Gat, J.R. and Lister, G.S., 1995. The "catchment effect" on the isotopic composition of lake waters; its importance in palaeolimnological interpretations. In: B. Frenzel, B. Stauffer and M.M. Weiss (Eds.) Problems of Stable Isotopes in Tree Rings, Lake Sediments and Peat Bogs as Climatic Evidence for the Holocene. Gustav Fischer Verlag, Stuttgart/New York, pp. 1-16. Geddes, M.C., De Deckker, P., Williams, W.D., Morton, D.W. and Topping, M., 1981. On the chemistry and biota of some saline lakes in Western Australia. Hydrobiologia 82, 201-222. Geoscience Australia, 2003. Reefs found in murky Carpentaria depths. AusGeo news 72, 4-5. Gillespie, R. 1977. Sydney University Natural Radiocarbon Measurements IV. Radiocarbon 19, 101-110. Gillespie, R. and Polach, H. A., 1979. The suitability of marine shells for radiocarbon dating of Australian prehistory, In: R. Berger and H.E. Suess (Eds.) Radiocarbon dating, International 14C Conference 9th, Proceedings. Berkeley, University of California Press, pp 404-421.

375

Godfrey, J.S. and Ridgway, K.R., 1985. The large-scale environment of the poleward-flowing Leeuwin Current, Western Australia: Longshore steric height gradients, wind stresses and geostrophic flow. Journal of Physical Oceanography 15, 481-495. Griffiths, H.I. and Holmes, J.A., 2000. Non-marine ostracods and Quaternary palaeoenvironments. Technical Guide 8, Quaternary Research Association, London; 188p. Griffiths, S.J., Street-Perrott, F.A., Holmes, J.A., Leng, M.J. and Tzedakis, C., 2002. Chemical and isotopic composition of modern water bodies in the Lake Kopias Basin, central Greece: analogues for the interpretation of the lacustrine sedimentary sequence. Sedimentary Geology 148 (1-2), 79-103. Grim, M.S. and Edgar, N.T., 1998. Bathymetric map of the Gulf of Carpentaria and the Arafura Sea, 1:2,500,000. US Geological Survey, Geologic Investigations Series, map I-2550. Grindrod, J., Moss, P. and van der Kaars, S., 1999. Late Quaternary cycles of mangrove development and decline on the north Australian continental shelf. Journal of Quaternary Science 14(5), 465-470. Haddon, A.C., Sollas, W.J. and Cole, G.A.J., 1894. On the geology of the Torres Straits. Transactions of the Royal Irish Academy 30, 411-476. Hammarlund, D., 1999. Ostracod stable isotope records from a deglacial isolation sequence in southern Sweden. Boreas 28(4), 564-574. Hanai, T., Ikeya, N., Ishizaki, K. and Yajima, M, 1977. Checklist of Ostracoda from Japan and its adjacent seas. University of Tokyo Press Bulletin 12, 1-102. Hanai, T., Ikeya, N. and Yakima, M., 1980. Checklist of Ostracoda from Southeast Asia. University of Tokyo Press Bulletin 17, 1-242. Hanebuth, T., Stattegger, K. and Grootes, P.M., 2000. Rapid flooding of the Sunda Shelf: a late-glacial sea-level record. Science 288, 1033-1035. Hardie, L.A. and Eugster, H.P., 1970. The evolution of closed-basin brines. Mineralogical Society of America Special Publication 3, 273-290. Harding, J.P., 1964. Crustacean cuticle with reference to the ostracod carapace. Publicazioni della Stazione Zoologica di Napoli 33(S), 9-31. Harris, P.T., 1994. Comparison of tropical, carbonate and temperate, siliciclastic tidally dominated sedimentary deposits: Examples from the Australian continental shelf. Australian Journal of Earth Sciences 41, 241-254. Harrison, S.P., 1993. Late Quaternary lake-level changes and climates of Australia. Quaternary Science Reviews 12, 211-231. Hartmann, G., 1978. Die Ostracoden der Ordnung Podocopida G.W. Müller, 1894 der tropisch-subtropischen Westküste Australiens (zweischen Derby im Norden und Perth im Süden). In: G. Hartmann and G. Hartmann-Schröder (Eds.) Zur Kenntnis des Eulitorals der australishchen Küsten unter besonderer Berücksichtigung der Polychaeten und Ostracoden, Tl. 1; Mitt. Hamburg. Zool. Mus. Inst. 75, 64-219. Hartmann, G., 1979. Die Ostracoden der Ordnung Podocopida G.W. Müller, 1894 der warm-temperierten (antiborealen) West- und Südwestküste Australiens (wischen Perth im Norden und Eucla im Süden). In: G. Hartmann and G. Hartmann-Schröder (Eds.) Zur Kenntnis des Eulitorals der australishchen Küsten unter besonderer Berücksichtigung der Polychaeten und Ostracoden, Tl. 3; Mitt. Hamburg. Zool. Mus. Inst. 76, 219-301.

Hartmann, G., 1980. Die Ostracoden der Ordnung Podocopida G.W. Müller, 1894 der warmtemperierten und subtropisch-tropischen Küstenabschnitte der Süd- und Südostküste Australiens (zwischen Ceduna im Westen und Lakes Entrance im Osten). In: G. Hartmann and G. Hartmann-Schröder (Eds.) Zur Kenntnis des Eulitorals der australishchen Küsten unter besonderer Berücksichtigung der Polychaeten und Ostracoden, Tl. 5; Mitt. Hamburg. Zool. Mus. Inst. 77, 2111-204. Hartmann, G., 1981. Die Ostracoden der Ordnung Podocopida G.W. Müller, 1894 der subtropisch-tropischen Ostküste Australiens (zwishen Eden im Süden und Heron-Island im Norden). In: G. Hartmann and G. Hartmann- Schröder (Eds.) Zur Kenntnis des Eulitorals der australishchen Küsten unter besonderer Berücksichtigung der Polychaeten und Ostracoden, Tl. 7; Mitt. Hamburg. Zool. Mus. Inst. 78, 97-149. Hartmann, G., 1982. Variation in surface ornament of the valves of three ostracod species from Australia. In Bate, R.H.., Robinson, E. and Sheppard, L.M. (Eds.) Fossil and Recent Ostracods. Ellis Horwood, Chichester, pp. 365-380. Hartmann-Schröder, G. and Hartmann, G., 1977. Preliminary notes on the distribution of Polychaeta and Ostracoda of the Australian coast. In Löffler, H. and Danielopol, D. (Eds.) Aspects of the Ecology and Zoogeography of Recent and Fossil Ostracods. Dr. W. Junk, The Hague, pp. 187-196. Hays, P.D. and Grossman, E.L., 1991. Oxygen isotopes in meteoric calcite cements as indicators of continental paleoclimate. Geology 19, 441-444. Heaton, T.H.E., Holmes, J.A. and Bridgwater, N.D., 1995. Carbon and oxygen isotope variations among lacustrine ostracods: implications for palaeoclimatic studies. The Holocene 5(4), 428-434. Hedges, J.I . and Keil, R.G., 1995. Sedimentary organic matter preservation: An assessment and speculative synthesis. Marine Chemistry 49: 81-115. Heip, C., 1976. The life-cycle of Cyprideis torosa (Crustacea, Ostracoda). Oecologia 24, 229-245. Hendry, J.P. and Kalin, R.M., 1997. Are oxygen and carbon isotopes of mollusc shells reliable indicators in marginal marine environments? A case study from the Middle Jurassic of England. Journal of the Geological Society 154, 321-333. Hesse, P.P. and McTainsh, G.H., 2003. Australian dust deposits: modern processes and the Quaternary record. Quaternary Science Reviews 22, 2007-2035. Hill, B.J., 1994. Preface. Ecology of the Gulf of Carpentaria. Australian Journal of Marine and Freshwater Research 45(3), iii-vii (plus pp. 265-454). Hill, K.C. and Hall, R., 2003. Mesozoic-Cenozoic evolution of Australia's New Guinea margin in a west Pacific context. Geological Society of Australia Special Publication 22 and Geological Society of America Special Paper 372, 265-290. Hirst, A.C. and Godfrey, J.S., 1993. The role of the Indonesian throughflow in a global ocean GCM. Journal of Physical Oceanography 23, 1057-1085. Hodell, D.A., Curtis, J.H. and Brenner, M., 1995. Possible role of climate in the collapse of Classic Maya civilization. Nature 375, 391-394. Hoefs, J., 1997. Stable Isotope Geochemistry. 4th Edition. Springer, Berlin, 201pp.

377

Holmes. J.A., Street-Perrott, F.A., Allen, M.J., Fothergill, P.A., Harkness, D.D., Kroon, D. and Perrott, R.A., 1997. Holocene palaeolimnology of Kajemarum Oasis, Northern Nigeria: an isotopic study of ostracodes, bulk carbonate and organic carbon. Journal of the Geological Society, London 154, 311-319. Holmes, J.A. and Chivas, A.R. (Eds.), 2002. The Ostracoda: Applications in Quaternary Research. Geophysical Monograph 131, American Geophysical Union, Washington DC, pp. 313. Hope, G., 1983. Montane Papua New Guinea 9000-5000 BP. In: J.M.A. Chappell and A. Grinrod (Eds.), Proceedings of the 1st CLIMANZ Conference, Howman's Gap, Victoria, February 1981. Department of Biogeography and Geomorphology, ANU, pp.97-98. Horne, D.J., 1983. Life-cycles of podocopid Ostracoda – a review with particular reference to marine and brackish-water species. In: Maddocks, R. (Ed.) Application of Ostracoda. University of Houston, Houston, pp. 581-590. Horne, D.J., Cohen, A. and Martens, K., 2002. Taxonomy, morphology and biology of Quaternary and living Ostracoda. In: Holmes, J.A. and Chivas, A.R. (Eds.) The Ostracoda: Applications in Quaternary Research. Geophysical Monograph Series, pp 5-36. Horne, D.J., Danielopol, D.L. and Martens, K., 1998. Reproductive behaviour. In: Martens, K. (Ed.), Sex and parthenogenesis: evolutionary ecology of reproductive modes in non-marine ostracods. Backhuys, Leiden, pp. 157-195. Hornibrook, N. De B., 1952. Tertiary and Recent marine Ostracoda of New Zealand. New Zealand Geological Survey Paleontological Bulletin 17, 1-242. Howe, H.V. and McKenzie, K.G., 1989. Recent marine Ostracoda (Crustacea) from Darwin and North-Western Australia. Northern Territory Museum Arts and Science Monograph Series 3, 1-50. Hussain, S.M. and Mohan, S.P., 2001. Distribution of recent benthic Ostracoda in the Adyar river estuary, east coast of India. Indian Journal of Marine Sciences 30, 53-56. Ikeya, N. and Cronin, T.M., 1993. Quantitative analysis of Ostracoda and water masses around Japan: Application to Pliocene and Pleistocene paleoceanography. Micropaleontology 39(3), 263-281. Imbrie, J., Berger, A., Boyle, E.A., Clemens, S.C., Duffy, A., Howard, W.R., Kukla, G., Kutzbach, J., Martinson, D.G., McIntyre, A., Mix, A.C., Molfino, B., Morley, J.J., Peterson, L.C., Pisias, N.G., Prell, W.L., Raymo, M.E., Shackleton, N.J. and Toggweiler, J.R., 1993. On the strucure and origin of major glaciation cycles, 2. The 100,000-year cycle. Paleoceanography 8(6): 699-735. Imbrie, J., Hays, J.D., Martinson, D.G., McIntyre, A., Mix, A.C., Morley, J.J., Pisias, N.G., Prell, W.L. and Shackleton, N.J., 1984. The orbital theory of Pleistocene climate: Support from a revised chronology of the marine δ18O record. In: A.L. Berger et al., (Eds.) Milankovitch and Climate. Vol. 1. D. Riedel, Hingham, MA, pp. 269-305. Immerhauser, A., Della Porta, G., Kenter, J.A.M. and Bahamonde, J.R., 2003. An alternative model for positive shifts in shallow-marine carbonate δ13C and δ18O. Sedimentology 50, 953-959.

Ingram, B.L., Conrad, M.E. and Ingle, J.C., 1996a. Stable isotope and salinity systematics in estuarine waters and carbonates: San Francisco Bay. Geochimica et Cosmochimica Acta 60, 455-467. Ingram, B.L., Ingle, J.C. and Conrad, M.E., 1996b. Stable isotope record of late Holocene salinity and river discharge in San Francisco Bay, California. Earth and Planetary Science Letters 141, 237-247. Ingram, B.L., De Deckker, P., Chivas, A.R., Conrad, M.E. and Byrne, A.R., 1998. Stable isotopes, Sr/Ca and Mg/Ca in biogenic carbonates from Petaluma Marsh, northern California, USA. Geochimica et Cosmochimica Acta 62, 3229-3237. Ingram, C., 1998. Palaeoecology and geochemistry of shallow marine Ostracoda from the Sand Hole Formation, Inner Silver Pit, southern North Sea. Quaternary Science Reviews 17, 913-929. Jahn, A., Gamenick, I. and Theede, H., 1996. Physiological adaptations of Cyprideis torosa (Crustacea, Ostracoda) to hydrogen sulphide. Marine Ecology Progress Series 142, 215-223. Johnson, B.J., Miller, G.H., Fogel, M.L., Magee, J.W., Gagan, M.K. and Chivas, A.R., 1999. 65,000 years of vegetation change in central Australia and the Australian summer monsoon. Science 284, 1150-1152. Jones, B.G. and Facer, R.A., 1981. CORMATT/PROG, a programme to create and test correlation coefficient matrix from data with missing values. Computers and Geoscience 8, 191-198. Jones, B.G., Martin, G.R. and Senapati, N., 1993. Riverine-tidal interactions in the monsoonal Gilbert River fan delta, northern Australia. Sedimentary Geology 83, 319-337. Jones, B.G., Woodroffe, C.D. and Martin, G.R., 2003. Deltas in the Gulf of Carpentaria, Australia: forms, processes and products. In: F.H. Sidi, P. Imbert, H. Darman and H.W. Posamentier (Eds.). Tropical Deltas of Southeast Asia - Sedimentology, Stratigraphy and Petroleum Geology. SEPM Special Publication 76, pp. 21-43. Jones, M.R., 1987. Surficial sediments of the western Gulf of Carpentaria, Australia. Australian Journal of Marine and Freshwater Research 38, 151-167. Jones, M.R., and Torgersen, T., 1988. Late Quaternary evolution of Lake Carpentaria on the Australian-New Guinea continental shelf. Australian Journal of Earth Sciences 35, 313-324. Jongsma, D., 1970. Eustatic sea level changes in the Arafura Sea. Nature 228, 150-151. Keatings, K.W., Heaton, T.H.E. and Holmes, J.A., 2002. Carbon and oxygen isotope fractionation in non-marine ostracods: results from a 'natural culture' environment. Geochimica et Cosmochimica Acta 66(10), 1701-1711. Keij, A.J., 1953. Preliminary note on the Recent Ostracoda of the Snellius Expedition. Koninkl. Neder. Akad. van Wetenschappen Amsterdam. Proc. Ser. B 56(2), 155-68. Kelts, K. and Talbot, M., 1990. Lacustrine carbonates as geochemical archives of environmental change and biotic/abiotic interactions. In: M.M.Tilzer and C. Serrya, C. (Eds.) Large Lakes. Springer-Verlag, Berlin, pp. 288-315. Kempf, E., 1980. Index and Bibliography of nonmarine Ostracoda. 1, Index A, Supplement 1. Geol. Univ. Koeln Sonderveroeffentlichungen 35, 1-188.

379

Kempf, E., 1986. Index and Bibliography of marine Ostracoda. 1, Index A, Supplement 1. Geol. Univ. Koeln Sonderveroeffentlichungen 50, 1-762. Kershaw, A.P., 1978. Record of last interglacial-glacial cycle from northeastern Queensland. Nature 272, 159-161. Kershaw, A.P., 1983. The vegetation record of northeastern Australia, 15-10 ka. In: J.M.A. Chappell and A. Grinrod (Eds.), Proceedings of the 1st CLIMANZ Conference, Howman's Gap, Victoria, February 1981. Department of Biogeography and Geomorphology, ANU, pp. 69-70. Kershaw, A.P., 1986. Climatic change and Aboriginal burning in north-east Australia during the last two glacial/interglacial cycles. Nature 322, 47-49. Kershaw, A.P., Baird, J.G., D'Costa, D.M., Edney, P.A., Peterson, J.A. and Strickland, K.M., 1991. A comparison of long Quaternary pollen records from the Atherton Tableland and Western Plain volcanic provinces. In: M.A.J. Williams, P. De Deckker and Kershaw, A.P. (Eds.), The Cainozoic in Australia: a Re-appraisal of the Evidence. Geological Society of Australia Special Publication 18, 288-301. Kershaw, A.P. and Nanson, G.C., 1993. The last full glacial cycle in the Australian region. Global and Planetary Change 7, 1-9. Kershaw, A.P. and Nix, H.A., 1990. The use of bioclimatic envelopes for estimation of quantitative palaeoclimatic values. In: D.H. Donnelly and R.J. Wasson (Eds.), CLIMANZ 3: Symposium Proceedings, Melbourne, 1987. CSIRO Division of Water Resources, Canberra, pp. 78-85. Kershaw, A.P., van der Kaars, S. and Moss, P.T., 2003. Late Quaternary Milankovitch-scale climatic change and variability and its impact on monsoonal Australasia. Marine Geology 201, 81-95. Kesling, R.V., 1951. The morphology of the ostracod molt stages. Illinois Biological Monograph 21, 1-126. Keyser, D., 2001. Cytological aspects in the noding process of Cyprideis torosa. 14th International Symposium on Ostracoda, Shizuoka, Japan, August1-4. Program and Abstracts, p38. Kilenyi, T.I., 1971. Some basic questions in the paleoecology of ostracods. In: Oertli, H.J. Paléoécologie des Ostracodes. Bull. Centre Rech. Pau-SNPA, 5 suppl, pp. 31-44. Kim, S.T. and O'Neil, J.R., 1997. Equilibrium and non-equilibrium oxygen isotope effects in synthetic carbonates. Geochimica et Cosmochimica Acta 61, 3461-3475. Kingma, J.T., 1948. Contributions to the knowledge of the Young Cainozoic Ostracoda from the Malayan region. Keminck Printers, Utrecht,118pp. Klein, R.T., Lohman, K.C. and Thayer, C.W., 1996. Bivalve skeletons record sea- surface temperature and δ18O via Mg/Ca and 18O/16O ratios. Geology 24, 415-419. Klein, R.T., Lohman, K.C. and Kennedy, G.L., 1997. Elemental and stable isotopic proxies of paleotemperature and paleosalinity: Climate reconstruction on the marginal northeast Pacific ca. 80 ka. Geology 25, 363-366. Kroopnik, P.M., 1985. The distribution of CO2 in the world oceans. Deep-sea Research 32(1), 57-84. Labeyrie, L.D., Duplessy, J.C. and Blanc, P.L., 1987. Variations in mode of formation and temperature of oceanic deep waters over the past 125, 000 years. Nature 327, 477-482.

Labutis, V.R., 1977. Cytheracean Ostracoda from the Great Barrier Reef. – MSc. Thesis, Macquarie University, Sydney, Australia. Lamb, H., Roberts, N., Leng, M., Barker, P., Benkaddour, A. and van der Kaars, S., 1999. Lake evolution in a semi-arid montane environment: response to catchment change and hydroclimatic variation. Journal of Paleolimnology 21, 325-343. Lambeck, K. Yokoyama, Y. and Purcell, T., 2002. Into and out of the Last Glacial Maximum: sea-level change during oxygen isotope stages 3 and 2. Quaternary Science Reviews 21 (1-3), 343-360. Land, L.S., 1989. The carbon and oxygen isotopic chemistry of surficial Holocene shallow water carbonate sediment and Quaternary limestone and dolomite. In: Fritz, P. and Fontes, J. Ch. (Eds.) Handbook of Environmental Isotope Geochemistry, Vol. 3. The Marine Environment, A. Elsevier 191-217; Amsterdam. Lees, B.G., 1992. Geomorphological evidence for Late Holocene climatic change in northern Australia. Australian Geographer 23, 1-10. Lees, B.G., Yanchou, L. and Head, M.J., 1990. Reconnaissance thermoluminescence dating of northern Australian coastal dune systems. Quaternary Research 34, 169-185. Li, H.-C. and Ku, T.-L., 1997. δ13C-δ18O covariance as a paleohydrological indicator for closed-basin lakes. Palaeogeography, Palaeoclimatology, Palaeoecology 133, 69-80. Lister, G.S., 1988. Stable isotopes from lacustrine ostracoda as tracers for continental palaeoenvironments. In: De Deckker, P., Colin, J.-P., Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp. 201-218. Lister, G.S., 1989. Reconstruction of palaeo air temperature changes from oxygen isotopic records in Lake Zürich: the significance of seasonality. Ecologae Geologica Helvetiae 82(1), 219-234. Lister, G.S., Kelts, K., Zao, C.K., Yu, J.-Q. and Niessen, F., 1991. Lake Qinghai, China: closed-basin lake levels and the oxygen isotope record for ostracoda since the latest Pleistocene. Palaeogeography, Palaeoclimatology, Palaeoecology 84, 141-162. Löffler, E., 1972. Pleistocene glaciation in Papua and New Guinea. Zeitschrift für Geomorphologie Supplement Band 13, 32-58. Long, B.G. and Poiner, I.R., 1994. Infaunal benthic community structure and function in the Gulf of Carpentaria, northern Australia. Australian Journal of Marine and Freshwater Research 45, 293-316. McConnaughey,T., 1989. 13C and 18O isotopic disequilibrium in biological carbonates: II. In vitro simulation of kinetic isotope effects. Geochimica et Cosmochimica Acta 53, 163-171. McCormac, F.G., Reimer, P.J., Hogg, A.G., Higham, T.F.G., Baillie, M.G.L., Palmer, J., Stuiver, M., 2002. Radiocarbon 44, 641-651.

McCormac, F.G., Hogg, A.G., Higham, T. F. G., Lynch-Stieglitz, J., Broecker, W.S., Baillie, M.G.L., Palmer, J.G., Xiong, L., Pilcher, J.R., Brown, D., and Hoper, S.T., 1998. Temporal variation in the interhemispheric 14C offset. Geophysical Research Letters 25(9),1321-1324.

381

McCulloch, M.T., De Deckker, P. and Chivas, A.R., 1989. Strontium isotope variations in single ostracod valves from the Gulf of Carpentaria, Australia: A palaeoenvironmental indicator. Geochimica et Cosmochimica Acta 53, 1703-1720. McKenzie, J.A., 1985. Carbon isotopes and productivity in the lacustrine and marine environment. In Stumm, W. (Ed.) Chemical processes in lakes. Wiley, New York, pp. 99-118. McKenzie, K.G., 1966. Freshwater Ostracoda from north-western Australia. Australian Journal of Marine and Freshwater Research 17, 259-279. McKenzie, K.G., 1967. The distribution of Caenozoic marine Ostracoda from the Gulf of Mexico to Australasia. In: Adams, C.G. and Ager, D.V. (Eds.) Aspects of Tethyan Biogeography Systematics Association Publication (7), pp. 219-238. McKenzie, K.G., 1971. Entomostraca of Aldabra, with special reference to the genus Heterocypris (Crustacea, Ostracoda). Philosophical Transactions of the Royal Society of London (Series B) 260, 257-297. McKenzie, K.G., 1978. Ostracoda (Crustacea : Podocopida) from southern Australian salt lakes, with the description of Reticypris new genus. Transactions of the Royal Society of South Australia 102 (7), 175-190. Magee, J.W., 1991. Late Quaternary lacustrine, groundwater, aeolian and pedogenic gypsum in the Prungle Lakes, southeastern Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 84, 3-42. Magee, J.W., Bowler, J.M., Miller, G.H. and Williams, D.L.G., 1995. Stratigraphy, sedimentology, chronology and palaeohydrology of Quaternary lacustrine deposits at Madigan Gulf, Lake Eyre, South Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 113, 3-42. Magee, J.W. and Miller, G.H., 1998. Lake Eyre palaeohydrology from 60 ka to the present: beach ridges and glacial maximum aridity. Palaeogeography, Palaeoclimatology, Palaeoecology 144, 307-329. Markgraf, V., Dodson, J.R., Kershaw, A.P., McGlone, M.S. and Nicholls, N., 1992. Evolution of late Pleistocene and Holocene climates in the circum- South Pacific land areas. Climate Dynamics 6, 193-211. Martens, K., 1985. Salinity tolerance of Mytilocypris henricae (Chapman)(Crustacea, Ostracoda). Hydrobiologia 124, 81-83. Martens, K., 1998. Diversity and endemicity of Recent non-marine ostracods (Crustacea, Ostracoda) from Africa and South America: a faunal comparison. Verh. Internat. Verein. Limnol. 26, 2093-2097. Martinez, J.I., De Deckker, P. and Chivas, A.R., 1997. New estimates for salinity changes in the Western Pacific Warm Pool during the Last Glacial Maximum: oxygen-isotope evidence. Marine Micropaleontology 32, 311-340. Martinson, D.G., Pisias, N.G., Hays, J.D., Imbrie, J., Moore, T.C., and Shackleton, N.J. 1987. Age Dating and the orbital Theory of the Ice Ages: Development of a High-Resolution 0 to 300,000-Year Chronostratigraphy. Quaternary Research 27, 1-29. Mazzini, I., Anadón, P., Barbieri, M., Castorina, F., Ferreli, L., Gliozzi, E., Mola, M. and Vittori, E., 1999. Late Quaternary sea-level changes along the Thyrrenian coast near Orbetello (Tuscany, Central Italy): Palaeoenvironmental reconstruction using ostracods. Marine Micropaleontology 37, 289-311.

Mbahinzireki, G., Uiblein, F. and Winkler, H., 1991. Microhabitat selection of ostracods in relation to predation and food. Hydrobiologia 222, 115-119. Mezquita, F., Olmos, V. and Oltra, R., 2000. Population ecology of Cyprideis torosa (Jones, 1850) in a hypersaline environment of the western Mediterranean (Santa Pola, Alacant) (Crustacea: Ostracoda). Ophelia 53(2), 119-130. Milankovitch, M., 1941. Canon of Insolation and the Ice-Age Problem. Royal Serbian Academy Special Publication 132. Israel Program for Scientific Translations, Jerusalem, 1969. Miller, G.H., Magee, J.W. and Jull, A.J.T., 1997. Low latitude glacial cooling in the Southern Hemisphere from amino acids in emu eggshells. Nature 385, 214-244. Mischke, S., 2001. Mid and Late Holocene palaeoenvironment of the lakes of the Eastern Juyanze and Sogo Nur in NW China, based on ostracod species assemblages and shell chemistry. Berliner Geowissenschaftliche Abhandlungen, E 35. Mostafawi, N., 1992. Rezente Ostracoden aus dem Sunda-Schelf, zwischen der malaiischen Halbinsel und Borneo. Senckenbergiana Lethaea 72, 129-168. Mourguiart, P., Corrège, T., Wirrmann, D., Argollo, J., Montenegro, M.E., Pourchet, M. and Carbonel, P., 1998. Holocene palaeohydrology of Lake Titicaca estimated from an ostracod-based transfer function. Palaeogeography, Palaeoclimatology, Palaeoecology 143, 51-72. Murray-Wallace, C.V., 1993. A review of the application of the amino acid racemisation reaction to archaeological dating. The Artefact 16, 19-26. Nakada, M and Lambeck, K., 1987. Glacial rebound and relative sea-level variations: a new appraisal. Geophysical Journal of the Royal Astronomical Society 90, 171-224. Nakada, M. and Lambeck, K., 1989. Pleistocene and Holocene sea-level change in the Australian region and mantle rheology. Geophysical Journal of the Royal Astronomical Society 96, 497-517. Nanson, G.C., Price, D.M., Short, S.A., Young, R.W. and Jones, B.G., 1991. Comparative uranium-thorium and thermoluminescence dating of weathered Quaternary Alluvium in the tropics of northern Australia. Quaternary Research 35, 347-366. Nanson, G.C., Price, D.M. and Short, S.A., 1992. Wetting and drying of australia over the past 300 ka. Geology 20, 791-794. Nanson, G.C., Callen, R.A. and Price, D.M., 1998. Hydroclimatic interpretation of Quaternary shorelines on South Australian playas. Palaeogeography, Palaeoclimatology, Palaeoecology 144, 281-305. Nanson, G.C., Jones, B.G., Price, D.M. and Pietsch, T.J., in review. Rapid alluvial induration and its effect on channel morphology in an Australian tropical river. Geomorphology.

Neale, J.W. 1979 On the genus Cyprinotus and its interpretation. In: Krstić, N. (Ed.) Proceedings of the VII International Symposium on Ostracodes – taxonomy, biostratigraphy and distribution of ostracodes, pp. 77-85. Geological Society of Serbia, Belgrade.

383

Neale, J.W., 1988. Ostracods and palaeosalinity reconstructions. In: De Deckker, P., Colin, J.-P. and Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp. 125-155. Newell, B.S., 1973. CSIRO Division of Fisheries and Oceanography Technical Paper 35. Nix, H.A. and Kalma, J.D., 1972. Climate as a dominant control in the biogeography of northern Australia and New Guinea. In: Walker, D. (Ed.), Bridge and Barrier, The Natural and Cultural History of Torres Strait. Research School of Pacific Studies, Department of Biogeography and Geomorphology, The Australian National University, Canberra, ACT, Publ. BG/3, pp. 61-91. Norman, M.D. and De Deckker, P., 1990. Trace metals in lacustrine and marine sediments: a case study from the Gulf of Carpentaria, northern Australia. Chemical Geology 82, 299-318. Nott, J., 1996. Late Pleistocene and Holocene sea-level highstands in Northern Australia. Journal of Coastal Research 14 (4), 907-910. Nott, J. and Price, D., 1999. Plunge pools and paleoprecipitation. Geology 22, 1047-1050. Oertli, H.J., 1971. The aspects of ostracod faunas. A possible new tool in petroleum sedimentology. In: Oertli, H.J. Paléoécologie des Ostracodes. Bulletin du Centre de Recherches, Pau-SNPA, 5 suppl, pp. 137-151. Okubo, I., 1974. Two species of the genus Cyprinotus (Ostracoda, Cyprididae) from Japan. Researches on Crustacea 6, 58-70. O'Neil, J.R., Clayton, R.N. and Mayeda, T.K., 1969. Oxygen isotope fractionation in -divalent metal carbonates. Journal of Chemical Physics 51, 5547-5558. Palacios-Fest, M.R., Cohen, A.S., Ruiz, J. and Blank, B., 1993. Comparative paleoclimatic interpretations from nonmarine ostracodes using faunal assemblages, trace elements shell chemistry and stable isotope data. In: P.K. Swart, K.C. Lohmann, J.A. McKenzie and S. Savin (Eds.) Climatic Change in Continental Isotope Records. American Geophysical Union, Geophysical Monograph 78, pp. 179-190. Passmore, V.L., Williamson, P.E., Gray, A.R.G. and Wellman, P., 1993a. The Bamanga Basin – a new exploration target. In: Carman, G.J., Carman, Z. (Eds.), Petroleum Exploration and Development in Papua New Guinea. Proceedings of the Second Papua New Guinea Petroleum Convention, Port Moresby, 31 May – 2 June 1993, pp. 233-240. Passmore, V.L., Williamson, P.E., Maung, T.U. and Gray, A.R.G., 1993b. The Gulf of Carpentaria – a new basin and new exploration targets. APEA (Australian Petroleum Exploration Association) Journal 33, 297-314. Penney, D.N., 1987. Application of Ostracoda to sea-level studies. Boreas 16, 237-247. Peypouquet, J.-P., Carbonel, P., Ducasse, O., Tolderer-Farmer, M. and Lete, C., 1988. Environmentally cued polymorphism of ostracods. In: Hanai, T., Ikeya, N. and Ishizaki, K. (Eds.) Evolutionary Biology of Ostracoda, its fundamentals and applications. Kodansha, Tokyo, pp. 1003-1018. Peypouquet, J.-P., Ducasse, O., Gayet, J. and Pratviel, L., 1980. Agradation et dégradation des tests d'ostracodes. Intérêt pour la connaissance de l'évolution paléohydrologique des domaines margino-littoraux carbonatés. "Cristallisation-Déformation-Dissolution des carbonates". Réunion spéciale III, 17-18, University Bordeaux, Bordeaux, pp. 357-369.

Philander, S.G.H., 1990. El Niño, La Niña and the Southern Oscillation. Academic, San Diego. Phipps, C.V.G., 1966. Gulf of Carpentaria (Northern Australia). In: Fairbridge, R.W. (Ed.), The Encyclopedia of Oceanography (Encyclopedia of Earth Sciences, Vol. 1). Reinhold, New York, 316-424. Phipps, C.V.G., 1970. Dating of eustatic events from cores taken in the Gulf of Carpentaria and samples from the New South Wales continental shelf. Australian Journal of Science 32, 329-331. Phipps, C.V.G., 1980. The Carpentaria plains. In: Henderson, R.A. and Stephenson, P.J. (Eds.) The Geology and Geophysics of Northeastern Australia. Geological Society of Australia, Queensland Division, pp. 382- 386, 388-390. Pisias, N.G., Martinson, N.G., Moore, T.C., Jr., Shackleton, N.J., Prell, W., Hays, J. and Bowden, G., 1984. High resolution stratigraphic correlation of benthic oxygen isotopic records spanning the last 300,000 years. Marine Geology 56, 119-136. Puri, H.S. and Hulings, N.C., 1976. Designation of lectotypes of some ostracods from the Challenger Expedition. Bull. Br. Mus. Nat. Hist. (Zool). 29(5), pp. 249-315. Radke, L.C., Howard, K.W.F. and Gell, P.A., 2002. Chemical diversity in southeastern Australian saline lakes I: geochemical causes. Marine and Freshwater Research 53, 941-959. Radke, L.C., Juggins, S., Halse, S.A., De Deckker, P. and Finston, T., 2003. Chemical diversity in south-eastern Australian saline lakes II: biotic implications. Marine and Freshwater Research 54, 895-912. Reimer, P., 2003. Marine reservoir correction. Retrieved 21/02/2003. http://radiocarbon.pa.qub.ac.uk//marine/refs/116.html Rhodes, E.G., Polach, H.A., Thom, B.G. and Wilson, S.R., 1980. Age structure of Holocene coastal sediments, Gulf of Carpentaria, Australia. Radiocarbon 22, 718-727. Rhodes, E.G., 1982. Depositional model for a chenier plain, Gulf of Carpentaria, Australia. Journal of Sedimentology 29, 201-221. Ricketts, R.D., Johnson, T.C., Brown, E.T., Rasmussen, K.A. and Romanovsky, V.V., 2001. The Holocene paleolimnology of Lake Issyk-Kul, Kyrgyzstan: trace element and stable isotope composition of ostracodes. Palaeogeography, Palaeoclimatology, Palaeoecology 176, 1-21. Roberts, J.H., Holmes, J.A. and Swan, A.R.H., 2002. Ecophenotypy in Limnocythere inopinata (Ostracoda) from the late Holocene of Kajemarum Oasis (north-eastern Nigeria). Palaeogeography, Palaeoclimatology, Palaeoecology 185, 41-52. Roca, J.R. and Wansard, G., 1997. Temperature influence on development and calcification of Herpetocypris brevicaudata Kaufmann, 1900 (Crustacea: Ostracoda) under experimental conditions: Hydrobiologia 347, 91-95. Romanek, C.S., Grossman, E.L. and Morse, J.W., 1992. Carbon isotopic fractionation in synthetic aragonite and calcite: Effects of temperature and precipitation rate. Geochimica et Cosmochimica Acta 56, 419-430. Rosenfeld, A. and Vesper, B., 1977. The variability of the sieve pores in recent and fossil species of Cyprideis torosa (Jones, 1850), as an indicator for salinity and palaeosalinity. In: Löffler, H. and Danielopol, D. (Eds.)

385

Aspects of ecology and zoogeography of recent and fossil Ostracoda. Dr. W. Junk b.v. Publishers, The Hague, pp. 55-67. Rothlisberg, P.C., Pollard, P.C., Nichols, P.D., Moriarty, D.J.W., Forbes, A.M.G., Jackson, C.J. and Vaudrey, D., 1994. Phytoplankton community structure and productivity in relation to the hydrological regime of the Gulf of Carpentaria, Australia in summer. Australian Journal of Marine and Freshwater Research 45, 265-282. Rozanski, K., Araguas-Araguas, L. and Gonfiantini, R., 1993. Isotopic patterns in modern global precipitation. In: P.K. Swart, K.C. Lohmann, J. McKenzie and S. Savin (Eds.) Climate Change in Continental Isotope Records. American Geophysical Union, Geophysical Monograph 78, pp. 1-36. Ruiz, F., González-Regaldo, J.M., Muñoz, J.G., Pendón, A., Rodríguez-Ramírez, L., Cáceres, J. and Rodríguez, V., 2003. Population age structure techniques and ostracods: Applications in coastal hydrodynamics and palaeoenvironmental analysis. Palaeogeography, Palaeoclimatology, Palaeoecology 199, 51-69. Sackett, W.M. and Moore, W.S., 1966. Isotopic variations of dissolved inorganic carbon. Chemical Geology 1, 317-322. Schlesinger, W.H., Marion, G.M. and Fonteyn, P.J., 1988. Stable isotope ratios and dynamics of caliche in desert soils. In: P.W. Rundel, J.R. Ehleringer and K.A. Nagy (Eds.) Stable Isotopes in Ecological Research. Springer- Verlag, Berlin, pp. 308-317. Schwalb, A., Burns, S.J., Cusminsky, G., Kelts, K. and Markgraf, V., 2002. Assemblage diversity and isotopic signals of modern ostracodes and host waters from Patagonia, Argentina. Palaeogeography, Palaeoclimatology, Palaeoecology 187, 323-339. Schwalb, A., Burns, S.J. and Kelts, K., 1999. Holocene environments from stable isotope stratigraphy of ostracods and authigenic carbonate in Chilean Altiplano Lakes. Palaeogeography, Palaeoclimatology, Palaeoecology 148, 153-168. Schwalb, A. and Dean, W.E., 1998. Stable isotopes and sediments from Pickerel Lake, South Dakota, USA: a 12ky record of environmental changes. Journal of Paleolimnology 20, 15-30. Schwalb, A., Locke, S.M. and Dean, W.E., 1995. Ostracode δ18O and δ13C evidence of Holocene environmental changes in the sediments of two Minnesota lakes. Journal of Paleolimnology 14, 281-296. Shackleton, N.J., 1977. The oxygen isotope stratigraphic record of the Late Pleistocene. Philosophical Transactions of the Royal Society of London Series B 280, 169-182. Shackleton, N.J., 1987. Oxygen isotopes, ice volume and sea-level. Quaternary Science Reviews 6, 183-190.

Shackleton, N.J., Imbrie, J. and Hall, M.A., 1983. Oxygen and carbon isotope record of East Pacific core V 19-30: implications for deep water in the late Pleistocene North Atlantic. Earth and Planetary Science Letters 65, 233-244. Shackleton, N.J. and Opdyke, N.D., 1973. Oxygen isotope and paleomagnetic stratigraphy of equatorial Pacific core V28-238; oxygen isotope

temperatures and ice volumes on a 105 and 106 year scale. Quaternary Research 3, 39-55. Shackleton, N.J. and Pisias, N.G., 1985. Atmospheric carbon dioxide, orbital forcing, and climate. In: E.T. Sundquiat and W.S. Broecker (Eds.) The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present. American Geophysical Union, Washington D.C., pp. 303-317. Shulmeister, J., 1992. A Holocene record from lowland tropical Australia. The Holocene 2, 107-16. Shulmeister, J. and Lees, B.G. 1992. Morphology and chronostratigraphy of a coastal dunefield; Groote Eylandt, northern Australia. Geomorphology 5, 521-534. Shulmeister, J. and Lees, B.G., 1995. Pollen evidence from tropical Australia for the onset of an ENSO-dominated climate at c. 4000 BP. The Holocene 5, 10-18. Siddall, M., Rohling, E.J., Almogi-Labin, A., Hemleben, C., Meischner, D., Schmetzer, I. and Smeed, D.A., 2003. Sea-level fluctuations during the last glacial cycle. Nature 423, 853-858. Smart, J., 1977. Late Quaternary sea-level changes, Gulf of Carpentaria, Australia Geology 5, 755-759. Smart, J., Grimes, K.G., Doutch, H.F. and Pinchin, J., 1980. The Carpentaria and Karumba Basins, North Queensland, Australia, Bureau of Mineral Resources, Geology and Geophysics, Bulletin 202. Smith, A.J., 1993. Lacustrine ostracodes as hydrochemical indicators in lakes of the north-central United States. Journal of Paleolimnology 8, 121-134. Smith A.J., Donovan, J.J., Ito, E. and Engstrom, D.R., 1997. Ground-water processes controlling a prairie lake's response to middle Holocene drought. Geology 25 (5), 391-394. Smith, A.J., Donovan, J., Ito, E., Engstrom, D.R. and Panek, V.A., 2002. Climate- driven hydrologic transients in lake sediment records: multiproxy record of mid-Holocene drought. Quaternary Science Reviews 21, 625-646. Smith, A. and Horne, D.J., 2002. Ecology of marine, marginal-marine and non- marine ostracods. In: Holmes, J.A. and Chivas, A.R. (Eds.) The Ostracoda: Applications in Quaternary Research. AGU Mon. Ser., pp. 37-64. Sohn, I.G., 1958. Chemical constituents of ostracodes: some applications to paleontology and paleoecology. Journal of Palaeontology 32, 730-736. Somers, I.F. and Long, B.G., 1994. Sediments and hydrology of the Gulf of Carpentaria, Australia. Australian Journal of Marine and Freshwater Research 45, 293-316. Spero, H.G., Bijma, J., Lea, D.W. and Bemis, B.E., 1997. Effect of seawater carbonate concentration on foraminiferal carbon and oxygen isotope. Nature 390, 497-500.

Stein, M., Wasserburg, G.J., Aharon, P., Chen, J.H., Zhu, Z.R., Bloom, A. and Chappell, J., 1993. TIMS U-series dating and stable isotopes of the last interglacial event in Papua New Guinea. Geochimica et Cosmochmica Acta 57, 2541-2554. Stirling, C.H., Esat, T.M., McCulloch, M.T. and Lambeck, K., 1995. High- precision U-series dating of corals from Western Australia and

387

implications for the timing and duration of the Last Interglacial. Earth and Planetary Science Letters 135, 115-130. Stirling, C.H., Esat, T.M., Lambeck, K. and McCulloch, M.T., 1998. Timing and duration of the last interglacial: evidence for a restricted interval of widespread coral reef growth. Earth and Planetary Science Letters 160, 745-762. Stuiver, M, Pearson, G W, and Braziunas, T F, 1986. Radiocarbon age calibration of marine samples back to 9000 cal yr BP. Radiocarbon 28:980-1021. Stuiver M and Polach H A, 1977, Discussion: Reporting of 14C data. Radiocarbon 19:355-363. Stuiver, M. and Reimer, P.J., 1993. Extended 14C data base and revised CALIB 3.0 14C age calibration program. Radiocarbon 35, 215-230. Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B., McCormac, F.G., v.d. Plicht, J., and Spurk, M.,1998. INTCAL98 Radiocarbon age calibration 24,000-0 cal BP. Radiocarbon 40:1041-1083. Suppiah, R., 1993. ENSO phenomenon and the 30-50 day variability in the Australian summer monsoon rainfall. International Journal of Climatology 13, 837-851. Swanson, K.M., 1979. The marine fauna of New Zealand: Ostracods of the Otago Shelf. New Zealand Oceanographic Institute Memoir 78, 4-53. Swanson, K.M., 1995. Strategies for, and evolutionary implications of, carapace conservation in some living, benthic, marine Ostracoda. Senckenbergiana Lethaea 75(1/2), 193-213. Swanson, K.M. and van der Lingen, G.J., 1994. Podocopid ostracod dissolution – description of a new palaeoenvironmental tool, with examples from the Tasman Sea. In: van der Lingen, G.J. and Muir, R.J. (Eds.) Evolution of the Tasman Sea Basin. Proceedings of the Tasman Sea Conference, Christchurch, New Zealand, 27-30 November, 1992. Balkema, Rotterdam, pp. 245-260. Swart, P.K. and Price, K., 2002. Origin of salinity variations in Florida Bay. Limnology and Oceanography 47, 1234-1241. Swart, P.K., Sternberg, L.D.S.L., Steinen, R. and Harrison, S., 1989. Controls on the oxygen and hydrogen composition of the waters of Florida Bay, USA. Chemical Geology 79, 113-123. Talbot, M.R., 1990. A review of the palaeohydrological interpretation of carbon and oxygen isotopic ratios in primary lacustrine carbonates. Chemical Geology 80, 261-279. Talbot and Kelts, 1990. Palaeolimnological signatures from carbon and oxygen isotopic ratios in carbonates from organic-rich lacustrine sediments. In: B. Katz (Ed.) Lacustrine Basin Exploration: Case Studies and Modern Analogues. American Association of Petrographic Geology Memoirs 50, 99-112. Thomas, B.M., Hanson, P., Stainforth, J.G., Stamford, P. and Taylor, L., 1990. Petroleum geology and exploration history of the Carpentaria Basin, Australia, and associated infrabasins. In: Leighton, M.W., Kolata, D.R., Oltz, D.F., Eidel, J.J. (Eds.), Interior Cratonic Basins, AAPG; 51, 709-724. Thomas, M.F., Nott, J. and Price, D.M., 2001. Late Quaternary stream sedimentation in the humid tropics: a review with new data from NE Queensland, Australia. Geomorphology 39, 53-68.

Thunell, R., Anderson, D., Gellar, D. and Miao, Q., 1994. Sea surface temperature estimates for the tropical western Pacific during the last glaciation and their implications for the Pacific warm pool. Quaternary Research 41, 255-265. Titterton, R. and Whatley, R.C., 1988. The provincial distribution of shallow water Indo-Pacific marine Ostracoda: Origins, antiquity, dispersal routes and mechanisms. In: Hanai, T., Ikeya, N. and Ishizaki, K. (Eds.) Evolutionary Biology of Ostracoda, its fundamentals and applications. Kodansha, Tokyo, pp. 759-787. Tomczak, M. and Godfrey, J.S., 1994. Regional Oceanography: an Introduction. Pergamon, Oxford. Torgersen, T., Hutchinson, M.F., Searle, D.E. and Nix, H.A., 1983. General bathymetry of the Gulf of Carpentaria and the Quaternary physiography of Lake Carpentaria. Palaeogeography, Palaeoclimatology, Palaeoecology 41, 207-225. Torgersen, T., Jones, M.R., Stephens, A.W., Searle, D.E. and Ullman, W.J., 1985. Late Quaternary hydrological changes in the Gulf of Carpentaria. Nature 313, 785-787. Torgersen, T., Luly, J., De Deckker, P., Jones, M.R., Searle, D.E., Chivas, A.R. and Ullman, W.J., 1988. Late Quaternary environments of the Carpentaria Basin, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 67, 245-261. Troup, A.J., 1965. The Southern Oscillation. Quarterly Journal of the Royal Meteorological Society 91, 490-506. Turpen, J.B. and Angell, R.W., 1971. Aspects of moulting and calcification in the ostracod Heterocypris. Biological Bulletin 140, 331-338. Usdowski, E., Michaelis, J., Bottcher, M.E. and Hoefs, J, 1991. Factors for the oxygen isotopic equilibrium fractionation between aqueous and gaseous o CO2, carbonic acid, bicarbonate, carbonate and water (19 C). Zeitschrift für Physikalische Chemie 170, 237-249. van Andel, T.H., Heath, G.R., Moore, T.C. and McGeary, F.R., 1967. Late Quaternary history, climate, and oceanography of the Timor Sea, northwestern Australia. American Journal of Science 265, 737-758. van der Kaars, S. and De Deckker, P., 2002. A Late Quaternary pollen record from deep-sea core Fr10/95, GC17 offshore Cape Range Peninsula, northwestern Western Australia. Review of Palaeobotany and Palynology 120, 17-39. van der Kaars, S., Penny, D., Tibby, J., Fluin, J., Dam, R.A.C. and Suparan, P., 2001. Late Quaternary Palaeoecology, palynology and palaeolimnology of a tropical lowland swamp: Rawa Danau, West-Java, Indonesia. Palaeogeography, Palaeoclimatology, Palaeoecology 171, 185-212.

van der Kaars, S., Wang, X., Kershaw, A.P., Guichard, F. and Setiabudi, D.A., 2000. A Late Quaternary palaeoecological record from the Banda Sea, Indonesia: patterns of vegetation, climate and biomass burning in Indonesia and northern Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 155, 135-153. van Harten, D., 1996. Cyprideis torosa (Ostracoda) revisited. Of salinity, nodes and shell size. Proceedings of the 2nd European Ostracodologists Meeting,

389

Glasgow, 1993. British Micropalaeontological Society, London, pp. 191-194. van Harten, D., 2000. Variable noding in Cyprideis torosa (Ostracoda): an overview, experimental results and a model from Catastrophe Theory. Hydrobiologia 419, 131-139. van Kreveld, S., Sarnthein, M., Erlenkeuser, H., Grootes, P., Jung, S., Nadeau, M.J., Pflaumann, U. and Voelker, A., 2000. Potential links between surging ice sheets, circulation changes, and the Dansgaard-Oeschger cycles in the Iminger Sea, 6018 kyr. Paleoceanography 15, 425-444. van Morkhoven, F.P.C.M., 1962. Post-Palaeozoic Ostracoda. Volume 1. Elsevier, Amsterdam, pp. 1-204. Vengosh, A., Kolodny, Y., Starinsky, A., Chivas, A.R. and McCulloch, M.T., 1991. Coprecipitation and isotopic fractionation of boron in modern biogenic carbonates. Geochimica et Cosmochima Acta 55, 2901-2910. Vesper, B., 1975. To the problem of noding on Cyprideis torosa (Jones, 1850). Bulletin of American Paleontology 65, 205-216. Victor, R., Dance, K.W. and Hynes, H.B.N., 1981. Drift of ostracod crustaceans in adjacent intermittent and permanent streams. Hydrobiologia 80, 219-223. Victor, R. and Fernando, C.H., 1982. Distribution of freshwater Ostracoda (Crustacea) in Southeast Asia. Journal of Biogeography 9, 281-288. Vidal, L., Layberis, L., Cortijo, E., Arnold, M., Duplessy, J.C., Michel, E., Becqué, S, and van Weering, T.C.E., 1997. Evidence for changes in the North Atlantic Deep Water linked to meltwater surges during the Heinrich events. Earth and Planetary Science Letters 146, 13-26. Visser, K., Thunnell, R. and Stott, L., 2003. Magnitude and timing of temperature change in the Indo-Pacific warm pool during deglaciation. Nature 421, 152-155. von Grafenstein, U., 2002. Oxygen-isotope studies of ostracods from deep lakes. In Holmes, J.A. and Chivas, A.R. (Eds.) The Ostracoda: Applications in Quaternary Research. American Geophysical Union Monograph 131. Washington D.C., pp. 249-266. von Grafenstein, U., Erlenkeuser, H., Kleinmann, A., Müller, J. and Trimborn, P., 1994. High-frequency climatic oscillations during the last deglaciation as revealed by oxygen-isotope records of benthic organisms (Ammersee, southern Germany). Journal of Paleolimnology 11, 349-357. von Grafenstein, U., Erlenkeuser, H., Müller, J., Jouzel, J. and Johnson, S., 1997. The cold event 8200 years ago documented in oxygen isotope records of precipitation in Europe and Greenland. Climate Dynamics 14, 73-81. von Grafenstein, U., Erlenkeuser, H., Müller, J. and Kleinmann-Eisenmann, A., 1992. Oxygen isotope records of benthic ostracods in Bavarian Lake sediments. Naturwissenschaften 79, 145-152. von Grafenstein, U., Erlenkeuser, H., Müller, J., Trimborn, P. and Alefs, J., 1996. A 200-year mid-European air temperature record preserved in lake 18 sediments: an extension of the air temperature-δ OP relation in the past. Geochimica et Cosmochimica Acta 60:21, 4025-4036. von Grafenstein, U., Erlenkeuser, H. and Trimborn, P., 1999. Oxygen and carbon isotopes in modern fresh-water ostracods valves: assessing vital offsets and autoecological effects of interest for palaeoclimatic studies. Palaeogeography, Palaeoclimatology, Palaeoecology 148, 133-152.

Waelbroeck, C., Labeyrie, L., Michel, E., Duplessy, J.C., McManus, J.F., Lambeck, K., Balbon, E. and Labracherie, M., 2002. Sea-level and deep water temperature changes derived from benthic foraminifera isotopic records. Quaternary Science Reviews 21, 295-305. Walker, D., 1972. Bridge and Barrier, The Natural and Cultural History of Torres Strait. Research School of Pacific Studies, Department of Biogeography and Geomorphology, The Australian National University, Canberra, ACT, Publ. BG/3. Wang, P., 1999. Response of Western Pacific marginal seas to glacial cycles: paleoceanography and sedimentological features. Marine Geology 156, 5-39. Wang, P. and Chappell, J., 2001. Foraminifera as Holocene environmental indicators in the South Alligator River, Northern Australia. Quaternary International 83-85, 47-62. Wasson, R.J., 1990. Desert dune building, dust raising and palaeoclimate in the Southern Hemisphere during the last 280,000 years. In: D.H. Donnelly and R.J. Wasson (Eds.), CLIMANZ 3: Symposium Proceedings, Melbourne, 1987. CSIRO Division of Water Resources, Canberra, pp. 123-137. Webster, P.J. and Streten, N.A., 1972. Aspects of Late Quaternary climate in tropical Australiasia. In: Walker, D. (Ed.), Bridge and Barrier, The Natural and Cultural History of Torres Strait. Research School of Pacific Studies, Department of Biogeography and Geomorphology, The Australian National University, Canberra, ACT, Publ. BG/3, pp. 39-59. Webster, P.J. and Streten, N.A., 1978. Late Quaternary Ice Age climates of tropical Australasia: interpretations and reconstructions. Quaternary Research 10, 279-309. Whatley, R., 1988. Ostracoda and palaeogeography. In: De Deckker, P., Colin, J.- P. and Peypouquet, J.-P. (Eds.), Ostracoda in the Earth Sciences. Elsevier, Amsterdam, pp. 103-123 Whatley, R.C. and Zhao, Q., 1987. Recent Ostracoda of the Malacca Straits. Part I – Rev. esp. Micropaleont. 19, 327-366. Whatley, R.C. and Zhao, Q., 1988. Recent Ostracoda of the Malacca Straits. Part II – Rev. esp. Micropaleont. 20(1), 5-37. Whitmore, T.J., Brenner, M., Curtis, J.H., Dahlin, B.H. and Leyden, B.W., 1996. Holocene climatic and human influences of the Yucatan Peninsula, Mexico: an interdisciplinary, palaeolimnological approach. The Holocene 6(3), 273-287. Williams, W.D. and Buckney, R.T., 1976. Chemical composition of some inland surface waters in South, Western and Northern Australia. Australian Journal of Marine and Freshwater Research 27, 379-397. Woodroffe, C.D. and Chappell, J., 1993. Holocene emergence and evolution of the McArthur River delta, southwestern Gulf of Carpentaria, Australia. Sedimentary Geology 83, 257-275. Woodroffe, C.D., Kennedy, D.M., Hopley, D., Rasmussen, C.E. and Smithers, S.G., 2000. Holocene reef growth in Torres Strait. Marine Geology 170, 331-346. Woodruff, F., Savin, S.M. and Douglas, R.G., 1980. Biological fractionation of oxygen and carbon isotopes by recent benthic foraminifera. Marine Microplaeontology 5, 3-11.

391

Woolfe, K.J., Larcombe, P., Orpin, A.R., Purdon, R.G., Michaelsen, P., McIntyre, C.M. and Amjad, N., 1998. Controls on inner-shelf sedimentation, Cape York Peninsula, in the region of 12oS. Australian Journal of Earth Sciences 45, 611-621. Wywroll, K.-H. and Miller, G.H., 2001. Initiation of the Australian summer monsoon 14,000 years ago. Quaternary International 83-85, 119-128. Xia, J., Ito, E. and Engstrom, D.R., 1997a. Geochemistry of ostracode calcite: part 1. An experimental determination of oxygen isotope fractionation. Geochimica et Cosmochimica Acta 61, 377-382. Xia, J., Engstrom, D.R. and Ito, E., 1997b. Geochemistry of ostracode calcite: part 2. The effects of water chemistry and seasonal temperature variation on Candona rawsoni. Geochimica et Cosmochimica Acta 61, 383-391. Xia, J., Haskell, B.J., Engstrom, D.R. and Ito, E., 1997c. Holocene climate reconstructions from tandem trace-element and stable-isotope composition of ostracodes from Coldwater Lake, North Dakota, USA. Journal of Paleolimnology 17, 85-100. Yassini, I. and Jones, B.G., 1995. Foraminiferida and Ostracoda from the estuarine and shelf environments on the southeastern coast of Australia. University of Wollongong Press, pp. 484. Yassini, I., Jones, B.G. and Jones, M.R., 1993. Ostracods from the Gulf of Carpentaria, northeastern Australia. Senckenbergiana Lethaea 73, 375-406. Yin, Y., Geiger, W. and Martens, K., 1999. Effects of genotype and environment on phenotypic variability in Limnocythere inopinata (Crustacea:Ostracoda). Hydrobiologia 400, 85-114. Yokoyama, Y., De Deckker, P., Lambeck, K., Johnston, P. and Fifield, L.K., 2001. Sea-level at the Last Glacial Maximum: evidence form northwestern Australia constrain ice volumes for oxygen isotope stage 2. Palaeogeography, Palaeoclimatology, Palaeoecology 165, 281-291. Yokoyama, Y., Lambeck, K., De Deckker, P., Johnston, P. and Fifield, L.K., 2000. Timing of the Last Glacial Maximum from observed sea-level minima. Nature 406, 713-716. Zazo, C., 1999. Interglacial sea levels. Quaternary International 55, 101-113. Zeebe, R.E., 1999. An explanation of the effect of seawater carbonate concentration on foraminiferal oxygen isotopes. Geochimica et Cosmochimica Acta 63, 2001-2007. Zhao, Q. and Wang, P., 1988. Distribution of modern Ostracoda in the shelf seas off China. In: Hanai, T., Ikeya, N. and Ishizaki, K. (Eds.) Evolutionary Biology of Ostracoda: its fundamentals and applications. Kodansha Ltd. Tokyo, pp. 805-821. Zhao, Q., Wang, P. and Zhang, Q., 1985. Ostracoda in bottom sediments of the South China Sea off Guangdong Province, China: their taxonomy and distribution. In Wang P. (Ed.) Marine Micropaleontology of China. China Ocean Press, Beijing, pp. 196-359. Zhao, Q. and Whatley, R.C., 1989a. A taxonomic revision of the new species of Ostracoda described by J.T. Kingma (1948) from the Late Cainozoic of Indonesia. Acta Micropalaeontologica Sinica 6 (3), 229-246. Zhao, Q. and Whatley, R.C., 1989b. Recent podocopid Ostracoda of the Sedili River and Jason Bay, southeastern Malay Peninsula. Micropaleontology 35 (2), 168-187

Zhu, Z.R., Wywroll, K.-H., Collins, L.B., Chen, J.H., Wasserburg, G.J. and Eisenhauer, A., 1993. High-precision U-series dating of Last Interglacial events by mass spectrometry: Houtman Abrolhos Islands, Western Australia. Earth and Planetary Science Letters 118, 281-293.

393

Appendices

395

Appendix 1a. Core log of core MD-28.

397

Appendix 1b. Core log of core MD-29.

399

Appendix 1c. Core log of core MD-30.

401

Appendix 1d. Core log of core MD-31.

403

Appendix 1e. Core log of core MD-33.

405

Appendix 2a. Sedimentary parameters of core MD-31.

407

Sedimentology Fauna r r d y d m m u Pyrite glass Quartz froste Gypsu Bolivina Rotaliids Miliolids subroun Ammonia Cyprideis Ilyocypris Pteropods Elphidium Iron-oxide Cytherella subangula >2%>63 Concretions Loxoconcha Charophytes Paranesidea Leptocythere Limnocythere Venericythere Fish fragments Organic matte Organic Shell fragments Echinoid spines Bivalve molluscs MD-31 forams Planktonic Facies om om om m m m m m mm mm mm mm mm br br br br fl fl fl Depth (m) 0 ++ + + + + + + + + + + + + 0.1 ++ + + + + + + + + + + + + 0.2 ++ + + + + + + + + + + + + 0.3 ++ + + + + + + + + + + + + 0.4 ++ + + + + + + + + + + + + 0.5 ++ + + + + + + + + + + 0.6 ++ + + + + + + + + + + + + + + 0.7 ++ + + + + + + + + + ++ 0.8 +++ + + + + + + + + + 0.9 +++ + + + + + + + + 1.0 +++ + + + + + + + 1.1 +++ + + + + + + + + + 1.2 +++ + + + + + + + + + 1.3 +++ + + + + + + + + + 1.4 +++ + + + + + + + + 1.5 +++ + + + + + + + + 1.6 +++ + + + + + ++ + + 1.7 +++ + + + + + + + + 1.8 +++ + + + ++ + + + 1.9 +++ + + + + + + + + + + 2.0 +++ + + + + + + + 2.1 +++ + + + + + + + + 2.2 +++ + + + + + + + + 2.3 +++ + + + + + + + + 2.4 +++ + + + + + + 2.5 +++ + + + + + + 2.6 ++ + + + + + +

2.7 ++ + + + + + + + 2.8 ++ + + + + + + 2.9 ++ + + + + + + 3.0 ++ + + + + + + + + + 3.1 +++ + + + + + + + + + + + + 3.2 +++ + + + + + + + + + + + 3.3 +++ + + + + + + + + + + + 3.4 +++ + + + + + + + + + + + 3.5 +++ + + + + + + + + + + + 3.6 +++ + + + + + + + + + + + 3.7 +++ + + + + + + + + + + + 3.8 +++ + + + + + + + + + + 3.9 +++ + + + + + + + + + + 4.0 +++ + + + + + + + + + + 4.1 +++ + + + + + ++ + ++ ++ + 4.2 +++ + + + + ++ ++ ++ 4.3 ++ + + + + + ++ + 4.4 + + + + + ++ + + ++ ++ 4.5 +++ + + + + + + 4.6 +++ + + + + + + 4.7 +++ + + + + + 4.8 +++ + + + + + + 4.9 +++ + + + + + 5.0 +++ + + + + 5.1 +++ + + + + 5.2 +++ + + + + + 5.3 +++ + + + + 5.4 +++ + + + + 5.5 +++ + + + + 5.6 +++ + + + + 5.7 +++ + + + + + + 5.8 +++ + + + + + 5.9 ++ + + + + + 6.0 ++ + + + + + + + 6.1 ++ + + + + + 6.2 ++ + + + + + + + 6.3 ++ + + + + + + + 6.4 ++ + + + + + + + + ++ ++ 6.5 ++ + + + + + + + + 6.6 ++ + + + + + + + + + + 409

6.7 ++ + + + + + + + + 6.8 ++ + + + + + + + + + 6.9 ++ + + + + + + + + 7.0 + + + + + +++ 7.1 +++ + + + + + + + + + + + 7.2 +++ + + + + + + + + + + + 7.3 +++ + + + + + + + + + + 7.4 +++ + + + + + + + + + + 7.5 +++ + + + + + + + + + + 7.6 +++ + + + + + + + + + + 7.7 + + + + + + + + 7.8 + + + + + ++ + + + + + + + + + + 7.9 + + + + + ++ + + + + + + + + + 8.0 + + + + + ++ + + + + + + + + + 8.1 + + + + + ++ + + + + + + + + + 8.2 + + + + + ++ + + + + ++ + + + + 8.3 + + + + + ++ + + + + + + + + 8.4 + + + + + + + + + + + + + 8.5 + + + + + + + + + + + + + 8.6 + + + + + + + + + + + + 8.7 + + + + + + + + + + + + 8.8 + + + + + + + + + + + + 8.9 + + + + + + + + + + + + 9.0 + + + + + + + 9.1 + + + + + + + + + + + + 9.2 + + + + + + + + + + + + 9.3 + + + + + + + + + + + + 9.4 + + + + + + + ++ + ++ + 9.5 + + + + + + + + + + + + + + 9.6 + + + + + + + + + + + + + + + + 9.7 + + + + + + + + + + + + 9.8 + + + + + + + + + + + + + 9.9 + + + + + + + + + + + + + 10.0 + + + + + + + + + + + + + 10.1 + + + + + + + + + + + + + 10.2 + + + + + + + + 10.3 ++ + + ++ + + + + 10.4 ++ + + + + + + + ++ + + + 10.5 ++ + + + + + + + + +++ + + ++ + + 10.6 ++ + + + + + + + + + + + ++ ++ ++

10.7 + + 10.8 + + 10.9 + + + + + + 11.0 + + + 11.1 + + + + + + + ++ + + + 11.2 + + + + + + + + + + + + + ++ + + + 11.3 + + + + + + + + + + + + + ++ + + + 11.4 + + + + + + + ++ + + + + + ++ ++ + + 11.5 + + + + + + + + + + + + + ++ + + + 11.6 + + + + + + + +++ 11.7 + + + + + + + + + + + + + + ++ + + + 11.8 + + + + + + + + + + + + + + + ++ + + + 11.9 + + + + + + + + ++ + + + + + + ++ + + + + 12.0 + + ++ + + 12.1 + + ++ + + 12.2 + + ++ + + 12.3 + + ++ + + 12.4 + + + ++ + + 12.5 + + ++ + + 12.6 + + ++ + + 12.7 + + + ++ + + + + 12.8 + ++ + + + 12.9 + ++ + + + 13.0 + + 13.1 ++ + + + + + + + 13.2 ++ + + + + + + + + 13.3 ++ + + + + + + + 13.4 ++ + + + + + 13.5 ++ + + + + +

411

Appendix 2b. Sedimentary parameters of core MD-32

413

Sedimentology Fauna r r d y d m m u Pyrite glass Quartz froste Gypsu Bolivina Rotaliids Miliolids subroun Ammonia Cyprideis Ilyocypris Pteropods Elphidium Iron-oxide Cytherella subangula >2%>63 Concretions Loxoconcha Charophytes Paranesidea Leptocythere Limnocythere Venericythere Fish fragments Organic matte Organic Shell fragments Echinoid spines Bivalve molluscs MD-32 forams Planktonic Facies om om om m m m m m mm mm mm mm mm br br br br fl fl fl Depth (m) 0 ++ + + + + ++ + ++ ++ + + + + + + + + + 0.1 + ++ + + + + ++ ++ ++ + + + + + + + + + 0.2 ++ + + + + ++ + ++ ++ + + + + + + + + 0.3 ++ + + + + + ++ + ++ ++ + + + + + + + + 0.4 + + + + ++ + + + + + + ++ + 0.5 + + + ++ + + + + + + + ++ + + 0.6 + + + +++ + + + + + + +++ +++ + +++ ++ + 0.7 +++ + + + + + + +++ +++ + +++ + + 0.8 +++ + + + + +++ + +++ +++ + + 0.9 + + + + ++ + + + + + + + + +++ ++ +++ +++ + 1.0 + + + + + + + + + +++ + +++ +++ + 1.1 + + + + +++ + + 1.2 + + + ++ + +++ 1.3 ++ +++ ++ +++ 1.4 ++ + +++ +++ +++ + + 1.5 + + + +++ + +++ ++ +++ +++ + 1.6 + + + + +++ + +++ +++ ++ +++ + 1.7 + ++ + +++ +++ +++ +++ + 1.8 + + +++ +++ +++ +++ 1.9 + +++ +++ +++ +++ 2.0 + + + + + +++ +++ ++ + 2.1 + + + +++ +++ ++ ++ 2.2 ++ +++ ++ +++ + 2.3 +++ +++ ++ 2.4 +++ +++ ++ + 2.5 + + + +++ +++ +++ + ++

2.6 + + +++ 2.7 + +++ ++ +++ +++ 2.8 +++ + +++ + 2.9 +++ +++ +++ + 3.0 +++ +++ +++ + 3.1 +++ + +++ +++ 3.2 ++ ++ + +++ + 3.3 +++ + ++ +++ + 3.4 + ++ +++ + +++ + 3.5 ++ +++ +++ ++ 3.6 +++ + +++ + +++ + + 3.7 + + + +++ + + +++ +++ +++ + 3.8 ++ +++ +++ +++ +++ 3.9 + ++ +++ +++ +++ +++ 4.0 + ++ +++ +++ +++ +++ 4.1 ++ + +++ +++ + +++ +++ 4.2 ++ + +++ +++ + +++ +++ 4.3 ++ + +++ +++ + +++ +++ 4.4 ++ + +++ +++ + +++ +++ + 4.5 ++ + +++ + +++ + +++ +++ 4.6 ++ ++ +++ + +++ + + +++ +++ 4.7 ++ + +++ + +++ + +++ +++ +++ 4.8 ++ + +++ +++ +++ +++ 4.9 ++ + +++ +++ +++ +++ 5.0 + ++ ++ +++ ++ + +++ 5.1 ++ ++ ++ +++ +++ ++ +++ 5.2 ++ ++ +++ +++ +++ +++ 5.3 ++ ++ +++ ++ + +++ 5.4 + + ++ ++ + +++ +++ +++ +++ 5.5 + + ++ + +++ +++ +++ +++ 5.6 + + + + + +++ +++ +++ +++ 5.7 +++ + + + + +++ + 5.8 +++ + + + + + +++ + 5.9 +++ + + + + + +++ + + + 6.0 +++ + + + + +++ + + 6.1 +++ + + + + +++ + + + + 6.2 +++ + + + + +++ + + + +

415

6.3 +++ + + + + +++ + + + + 6.4 +++ + + + + +++ + + + 6.5 +++ + + + + ++ + + + 6.6 +++ + + + + + ++ + + 6.7 +++ + + + ++ + + ++ + + + 6.8 +++ + + + + + + + + + + + + 6.9 + +++ + + + + + + + + + + + + + + 7.0 ++ + + + + + + + + 7.1 + + + + + + + + + + + 7.2 + + + + + + + + 7.3 + + + + + + + + + 7.4 + + + + + + + + 7.5 + + + + + + + + + + + + + + 7.6 + + + + + + + + + + + 7.7 + + + + + + + + + + 7.8 + + + + + + + + + + + + + 7.9 + + + + + + + + + + + 8.0 + + + + + + + + + + 8.1 + + + + + + + + ++ 8.2 + + + + + + + 8.3 + + + + + + + + + + + + + + + 8.4 + + + + + + + + + + + + + + + 8.5 + + + + + + + + + 8.6 + + + + + + + + + 8.7 + + + + + + + + + +++ 8.8 + + + + + + + + + 8.9 + + + + + + + + 9.0 + +++ +++ + +++ + +++ 9.1 + + + + + +++ ++ +++ + + +++ + +++ + 9.2 + + + + + +++ +++ + + +++ + +++ 9.3 + ++ + + + + +++ + + + + + ++ ++ + + 9.4 ++ + + + + ++ + + + + + + + + + + 9.5 ++ + + + + ++ + + + + + ++ + + + 9.6 + ++ + + + + ++ + + + + + + + + + + + 9.7 + ++ + + + + ++ + + + + + + + + ++ + 9.8 + ++ + + + + ++ + + + + + + + + + + + + 9.9 + ++ + + + + ++ + + + + + + + + + + + + +

10.0 + ++ + + + + ++ + + + ++ + + + + + + + + 10.1 + ++ + + + + ++ + + + + ++ + + + + + + + + + 10.2 + ++ + + + + ++ + + + + + ++ + + + + + + + + + 10.3 + ++ + + + + ++ + + + ++ ++ + + + + + + + + + + 10.4 + ++ + + + + ++ + + + ++ +++ + + + + + + + + + 10.5 + ++ + + + + ++ + + + + +++ + + + + + + + + + 10.6 + ++ + + + + ++ + + + + ++ + + + + + + + + 10.7 + ++ + + + + ++ + + + + ++ + + + + + + + + + 10.8 + + + + + ++ + + + + + + + + + + + 10.9 + + + + + ++ + + + + + + + + + 11.0 + + + + + + ++ + + + + ++ + + + + + + 11.1 ++ + + + + ++ + + + + + + + + + + + + + + 11.2 + + + + + ++ + + + + + + + + + + + + + 11.3 + + + + + + ++ + + + + ++ + + + +++ + + + + + + 11.4 ++ + + + + ++ + + + + + + +++ + + + + 11.5 + ++ + + + + ++ + + + + + +++ + + + + + 11.6 + ++ + + + + ++ + + + + + +++ + + + + 11.7 ++ + + + + ++ + + + + + +++ ++ + + + + 11.8 ++ + + + + ++ + + + + + + +++ + + + + + + 11.9 ++ + + + + ++ + + + + + + +++ + + + + + + 12.0 ++ + + + + ++ + + + + + +++ ++ + + + + 12.1 ++ + + + + ++ + + + +++ +++ + + + + + 12.2 ++ + + + + ++ + + + + +++ +++ + + + + + 12.3 ++ + + + + ++ + + + + +++ + + + + + 12.4 ++ + + + + ++ + + + +++ +++ + + + + + 12.5 ++ + + + + ++ + + + + + +++ +++ + + + + 12.6 ++ + + + + ++ + + + + + + +++ +++ + + + + 12.7 ++ + + + + ++ + + + + + + +++ +++ + + + + 12.8 ++ + + + + ++ + + + + + + + ++ + + + + 12.9 ++ + + + + ++ + + + + + + + + + + + + 13.0 + + + + + ++ + + + + + + + +++ + + + + + + 13.1 + + + + + ++ + + +++ + + + + + + + + + 13.2 + + + + + ++ + + ++ + + + + + + + + + 13.3 +++ + + + ++ + + + +++ + + + + + + + + 13.4 +++ + + + ++ +++ + + + + + + + 13.5 +++ + + + ++ +++ + + + + + + + + 13.6 +++ + + + ++ + + +++ + + + + + + + +

417

13.7 +++ + + + ++ + + +++ + + + + + + + 13.8 +++ + + + + ++ + + +++ + + + + + + + 13.9 +++ + + + + ++ + +++ + + + + + + + 14.0 + +++ + + + + +++ + 14.1 + +++ + + + + +++ 14.2 + +++ + + + + +++ 14.3 + +++ + + + + +++ 14.4 + +++ + + + + +++ 14.5 + +++ + + + + +++ 14.6 + +++ + + + + +++ 14.7 + +++ + + + + +++ 14.8 + +++ + + + + +++

419

Appendix 2c Sedimentary parameters of core MD-33.

421

Sedimentology Fauna

r r m u Pyrite glassy Quartz Quartz frosted frosted Bolivina Gypsum Gypsum Rotaliids Miliolids subround Ammonia Cyprideis Ilyocypris Ilyocypris Pteropods Pteropods Elphidium Elphidium Iron-oxide Iron-oxide Cytherella subangula >2%>63 Concretions Concretions Loxoconcha Loxoconcha Charophytes Paranesidea Paranesidea Leptocythere Leptocythere Limnocythere Limnocythere Venericythere Venericythere Fish fragments Fish fragments Organic matte Organic Shell fragments Echinoid spines Echinoid spines Bivalve molluscs MD-33 forams Planktonic Facies om om om m m m m m mm mm mm mm mm br br br br fl fl fl Depth (m) 0 + + + + + + + + + + + + + + + + 0.1 + + + + + + + + + + + + + + + + + 0.2 + + + + + + + + + + + + + + + + + + + 0.3 ++ + + + + + + + + + + + + + + + + + ++ 0.4 ++ + + + + + + + + + + + + + + + + ++ + 0.5 + + + + + + + + + + + + 0.6 ++ + + + + + + + +++ ++ ++ + 0.7 + ++ + + + + + + + ++ + 0.8 + + + + + + + + +++ ++ ++ + 0.9 + ++ + + + + + +++ ++ ++ + 1.0 + + + + + +++ ++ ++ + 1.1 + + +++ ++ ++ + 1.2 + + + + + + +++ ++ ++ + 1.3 ++ + + + + + + +++ ++ + 1.4 ++ + + + + + + +++ ++ ++ + 1.5 ++ + + + + + + +++ ++ + 1.6 + + + + + + + + ++ + 1.7 + + + + + + + + ++ + 1.8 + + + + + + + +++ +++ ++ + 1.9 + + + + +++ + ++ 2.0 + + + + + + +++ + ++ + 2.1 + + + + + + + +++ + ++ 2.2 + ++ + + + + + + + + + + + + +

2.3 + + + + + + + + + + + + + + + 2.4 + ++ + + + + + + + + + + + + + + 2.5 + ++ + + + + + + + + + + + + + 2.6 + ++ + + + + + + + + + + + + 2.7 ++ + + + + + + + + 2.8 + ++ + + + + + + + 2.9 ++ + + + + + + + + +++ 3.0 + ++ + + + + + + + 3.1 ++ + + + + + + 3.2 + ++ + + + + + + + 3.3 + +++ + + + + + + + 3.4 + +++ + + + + + + + + +++ 3.5 + +++ + + + + + + + ++ + + +++ 3.6 + +++ + + + + + + + + + ++ + + + + 3.7 + +++ + + + + + + + + ++ + + + 3.8 + +++ + + + + + + + + + ++ + + + 3.9 + +++ + + + + + + + + + ++ + + + 4.0 + +++ + + + + + + + + + ++ + + + 4.1 + +++ + + + + + + + ++ + 4.2 + +++ + + + + + + + + ++ + 4.3 + +++ + + + + + + + + ++ + 4.4 + +++ + + + + + + + + + ++ + 4.5 + +++ + + + + + + + + + + ++ + 4.6 + +++ + + + + + + + + + ++ + + + 4.7 + +++ + + + + + + + + + + ++ + + + 4.8 + +++ + + + + + + + + + + ++ + + + 4.9 + +++ + + + + + + + + + + ++ + + + 5.0 + +++ + + + + + + + + + + + ++ + + + 5.1 + +++ + + + + + + + + + + + ++ + + + 5.2 + +++ + + + + + + + + + + + ++ + + + 5.3 + +++ + + + + + + + + + + + ++ + + + 5.4 + +++ + + + + + + + + + + + ++ + + + 5.5 + +++ + + + + + + + + + + ++ + + + +

423

5.6 + +++ + + + + 5.7 + +++ + + + + 5.8 + +++ + + + + + + 5.9 + +++ + + + + + + 6.0 + +++ + + + + + + 6.1 + +++ + + + + + + 6.2 + +++ + + + + + + 6.3 + +++ + + + + + + 6.4 + +++ + + + + + + 6.5 + +++ + + + + + + +

425

Appendix 3a. Quartz grains of unit 7, MD-32.

Appendix 3b. Quartz grains of unit 4, MD-32.

427

Appendix 4a. Results of EDX analyses of material from core MD-32.

Sample Description Elements Identification 32/1400 Grey concretion Si, Al, Fe, Ti Pumice 32/1390 Grey concretion Si, Al, P, Ca, Fe Pumice 32/1140 Pink concretion Ca, P, Si, Al Calcite, Phosphate 32/1080 Orange concretion Ca, P, Si, Al, Fe Calcite, Phosphate, some iron- oxide 32/1060 Striated fragment Ca, Fe Shell fragment 32/1040 White concretions Ca, P Calcite, Phosphate 32/900 Blackened shell Ca, Fe, S Pyritised shell 32/860 Small crystals Ca, S Gypsum 32/660 Concretion Ca, Si Calcite, Quartz 32/630 Pink concretion Ca, P Calcite, Phosphate 32/40 "Tooth" Ca, P Phosphate 32/30 Black mineral S, Fe Pyrite 32/0 Metallic sphere Fe Iron ball

Appendix 4b. Results of XRD analyses of material of cores MD-28, 29, 30, 32.

Number Sample Compounds Mineral phases 32/630 Concretion CaCO3, FeCO3 Calcite with minor siderite 32/640 Concretion CaCO3, FeS2 Calcite with minor pyrite 32/900 Blackened gastropod CaCO3, FeS2 Aragonite with pyrite 32/930 Shiny black material Ca(PO4)3F, FeS2 Apatite (phosphate) 28/105 Concretion CaCO3 Calcite 28/150 Concretion CaCO3 Calcite 28/250 Frosted shell fragment CaCO3, CaSO4.2H2O Aragonite with gypsum 29/240 Concretion CaCO3 Calcite 29/300 Concretion CaCO3 Calcite 30/155 Pink concretion CaCO3 Calcite 30/240 Concretion CaCO3 Calcite 30/290 Sugary concretion BaSO4 Barite 30/500 Yellow concretion Ca(Fe,Mg)(CO3)2 Dolomite/ankerite 30/650 Orange concretion FeS2, FeO(OH) Pyrite/goethite

429

Appendix 5a 14C AMS dates from the Gulf of Carpentaria cores.

Core No. and ANSTO Number and identification of dated material. Conventional Calibrated depth (cm) No. Total sample weight (mg) 14C age (a BP)+1σ 14C age (a BP) MD28/0-1 OZE263 3: Cardiidae, Glycymeridae, Ostreidae (7.81 mg) 745+55 350* MD28/15-16 OZE260 60 ostracod (Paranesidea) valves (1.41 mg) 2930+50 2640* MD28/15-16 OZE261 3: Cardiidae, Ostreidae (2) (8.02 mg) 2590+40 2210* MD28/35-36 OZG379 1: Nuculanidae (24.46 mg) 9930+60 10660* MD28/50-51 OZE257 1: Lucinidae (8.4 mg) 9530+80 10130* MD28/60-61 OZG374 1: Corbulidae (2.48 mg) 10260+70 12040† MD28/66-67 OZE254 5: Corbulidae (7.21 mg) 14350+90 17200† MD28/70-71 OZG373 1: Corbulidae (3.14 mg) 14960+90 17900† MD28/75-76 OZG375 1: Corbulidae (2.12 mg) 14280+90 17120† MD29/5-6 OZF283 1: Lucidinae (1.05 mg) 820+50 410* MD29/20-21 OZE259 1: Veneridae (18.87 mg) 9810+90 10520* MD30/0-1 OZF284 4: Veneridae (1.13 mg) 8280+60 8730* MD30/5-6 OZG230 1: unidentified marine bivalve mollusc (0.87 mg) 1650+50 1160* MD30/30-31 OXG381 1: Ostreidae (1.69 mg) 4860+50 5100* MD30/60-61 OZF285 2: Corbulidae (1.40 mg) 9330+70 10530† MD30/65-66 OZE255 2: Nuculanidae (7.17 mg) 4310+60 4360* MD30/70-71 OZE250 4: Planorbidae (1.58 mg) 10410+80 12390† MD30/80-81 OZG231 1: Planorbidae (1.13 mg) 10450+80 12460† ΜD30/90-91 OZG382 2: Corbulidae (0.95 mg) 10680+70 12760† MD30/150-151 OZF286 1: Hydrobiidae (1.31 mg) 11880+170 13940† MD31/0-1 OZE262 2: Myochamidae, Nuculanidae (5.21 mg) 6840+50 7320* MD31/0-1 OZE262U2 1: Corbulidae (3.67 mg) 7790+60 8200* MD31/5-6 OZG220 1: unidentified marine bivalve (19.77 mg) 3760+40 3620* MD31/10-11 OZF287 1: Veneridae (1.81 mg) 10100+60 10850* MD31/30-31 OZG383 1: Veneridae (1.69 mg) 2290+50 1840* MD31/55-56 OZE256 1: Veneridae (14.2 mg) 6910+80 7370* MD31/55-56 OZG221 1: Arcidae (4.44 mg) 1920+40 1410*

431

MD31/60-61 OZG377 1: Nuculanidae (7.90 mg) 735+35 340* MD31/65-66 OZE251 1: Planorbidae (1.16 mg) 10350+100 12250† MD31/70-71 OZG222 1: Corbulidae (7.59 mg) 10340+60 12240† MD31/180-181 OZG232 3: mollusc, Eulimidae, echinoid spines (0.73 mg) 3810+100 4200† MD31/320-321 OZG233 1: Planorbidae (1.26 mg) 46000+200 MD31/330-331 OZG234 3: Planorbidae (2.61 mg) 45800+1700 MD32/0-1 OZF289 1: Veneridae (17.39 mg) 9720+50 10440* MD32/10-11 OZG235 1: unidentified marine bivalve (2.18 mg) 9700+80 10400* MD32/20-21 OZG384 2: Cardiidae, Nuculanidae (1.09 mg) 1820+50 1320* MD32/35-36 OZF290 1: Corbulidae (4.75 mg) 10390+70 12350† MD32/40-41 OZF291 1: Planorbidae (2.46 mg) 11440+80 13380† MD32/70-71 OZF292 1: Planorbidae (1.19 mg) 12330+80 14630† MD32/75-76 OZG378 1: Corbulidae (2.32 mg) 14390+80 17250† MD32/100-101 OZG385 1: Corbulidae (2.39 mg) 14190+130 17020† MD32/120-121 OZG386 1: Corbulidae (3.39 mg) 14330+90 17180† MD32/145-146 OZG388 1: Corbulidae (3.80 mg) 14330+100 17180† MD32/150-151 OZF293 1: Corbulidae (4.21 mg) 15390+110 18400† MD33/0-1 OZF294 1: Corbulidae (1.50 mg) 2010+80 1510* MD33/0-1 OZG236 1: unidentified marine mollusc (1.28 mg) 2570+50 2190* MD33/10-11 OZG237 2: Ostreidae (1.82 mg) 1300+40 790* MD33/20-21 OZE258 1: Nuclidae (17.95 mg) 465+50 480‡ MD33/20-21 OZF295 1: Arcidae (12.30 mg) 690+30 290* MD33/30-31 OZG389 1: Veneridae (21.30 mg) 820+50 410* MD33/77-78 OZE252 135 ostracod (Cyprideis) valves (5.19 mg) 15760+90 18820† MD33/77-78 OZE253 2: Corbulidae (8.10 mg) 14570+100 17450† MD33/210-211 OZG224 1: Corbulidae (18.2 mg) 40000+500 The calibrated 14C dates have been obtained utilising the CALIB v4.4 program, Stuiver and Reimer (1993). *MARINE98.14C calibration utilised for marine samples (460-20,760 14C yr BP)(Stuiver et al., 1998). In addition to the marine reservoir correction of 400 years, a further 50±31 year local correction has been applied based on the regional mean for NE Australia. † INTCAL98.14C calibration has been utilised for non-marine samples (0-20,265 14C yr BP)(Stuiver et al., 1998). ‡ SHCAL02.14C marine southern hemisphere calibration curve utilised for marine samples (0-460 14Cyr BP)(McCormac et al, 2002).

Appendix 5b. Marine reservoir correction for NE Australia.

Lat Long Site Calendar 14C age Reservoir ΔR‡ Refs. Age BP* age† 10oS 143oE Torres Strait 1875 553±68 401±69 77±68 1, 2 10oS 143oE Torres Strait 1875 536±85 384±86 60±85 1, 2 10oS 143oE Torres Strait 1909 443±84 304±85 -6±85 1, 2 12oS 141oE 80 km N Weipa 1903 576±60 437±61 122±60 3 12oS 141oE 80 km N Weipa 1903 436±68 297±61 -17±60 3 Regional mean 50±31 NE Australia

Date taken from Reimer, 2004. *14C age refers to conventional radiocarbon age with a Libby-half-life of 5568 a, corrected for isotopic fractionation (Stuiver and Polach, 1977). †Reservoir age refers to the measures marine 14C minus the atmospheric 14C at time t as defined by Stuiver et al., 1986. The atmospheric age is taken from a 5 point moving average of New Zealand cedar radiocarbon ages (McCormac et al., 1998). ‡ΔR refers to the difference between the regional and global marine 14C (Stuiver et al., 1998a). References: 1. Gillespie, 1977; 2. Gillespie and Pollach, 1979; 3. Rhodes et al., 1980.

433

Appendix 6. Taxonomic list of ostracods described from core MD-32

Phylum CRUSTACEA Pennant, 1777 Class OSTRACODA Latrielle, 1806 Subclass MYODOCOPA Sars, 1866 Order CLADOCOPIDA Sars, 1866 Suborder CLADOCOPINA Sars, 1866 Family POLYCOPIDAE Sars, 1866 Genus Eupolycope Chavtur, 1981 Eupolycope sp. Yassini et al., 1993 Eupolycope sp. Yassini et al., 1993: 382, pl. 2, fig. 23.

Genus Polycope Sars, 1866 Polycope favus Brady, 1880 Polycope ?favus Brady, 1880: 170, pl. 36, figs. 4a-4b; Polycope ?favus Puri and Hulings, 1976: 311, pl. 24, fig. 9; Polycope favus Yassini et al., 1993: 382, pl. 2, fig. 26. Material: MD972132-10. Plate 1, Fig. 1 (jmr48)

Polycope sp. I Yassini et al., 1993 Polycope sp. I Yassini et al., 1993: 382, pl. 2, fig. 24. Material: MD972132-10. Plate 1, Fig. 2 (jmr49)

Subclass PODOCOPA Order PLATYCOPIDA Sars, 1866 Suborder PLATYCOPINA Sars, 1866 Family CYTHERELLIDAE Sars, 1866 Genus Cytherella Jones, 1845 Cytherella semitalis Brady, 1868 Cytherella semitalis Brady, 1868: 72, figs. 23-24; Cytherella semitalis Kingma, 1948: 63, pl. 6, fig. 6; Cytherella semitalis Whatley and Zhao, 1987: 334, pl. 1, figs. 7-10; Cytherella semitalis Howe and McKenzie, 1989: 4, fig. 37; Cytherella semitalis Mostafawi, 1992: 133, pl. 1, fig. 4; Cytherella semitalis Yassini et al., 1993: 383, pl. 1, figs. 14-16, pl. 8, fig. 156. Cytherella semitalis Dewi, 1997: 55, figs. 11-13. Material: MD972132-1070. Plate 1, Fig 3. (jmr03)

Cytherella cf. C. hemipuncta Swanson, 1969 Cytherella cf. C. hemipuncta Whatley and Zhao, 1987: 333. pl. 1, figs. 2-4; Cytherella cf. C. hemipuncta Dewi, 1997: 54, fig. 6. Material: MD972132-1390. Plate 1, Fig. 4 (jmr72)

435

Genus Cytherelloidea Alexander, 1929 Cytherelloidea malaccaensis Whatley and Zhao, 1987 Cytherelloidea malaccaensis Whatley and Zhao, 1987: 335, pl. 1, figs. 19-21; Cytherelloidea malaccaensis Yassini et al., 1993: 383, pl. 1, figs. 12-13; Cytherelloidea malaccaensis Dewi, 1997: 56, figs. 20-21. Material: MD972132-70. Plate 1, Fig. 5 (jmr70)

Cytherelloidea cf. C. excavata Mostafawi, 1992 Cytherelloidea excavata Mostafawi, 1992: 135, pl. 1, figs. 9-11; Cytherelloidea excavata Dewi, 1997: 56, figs. 18-19. Similar to C.excavata, but ornamentation not as well defined. Material: MD972132-1160. Plate 1, Fig. 6 (jmr02)

Cytherelloidea sp. Description: Carapace medium-sized. Dorsum largely straight, convex at posterior end, ventrum medially concave. Anterior margin broadly and symmetrically rounded, wider than posterior. Posterior margin truncated. Ornamentation finely reticulate, anterior marginal ridge prominent and smooth. Dorsal ridge oblique toward muscle scar depression. Ventral marginal ridge less pronounced. Material: MD972132-10. Plate, Fig. 7 (jmr50)

Order PODOCOPIDA Sars, 1866 Superfamily BAIRDIACEA Sars, 1888 Family BAIRDIIDAE Sars, 1888 Subfamily BAIRDIINAE Sars, 1888 Genus Paranesidea Maddocks, 1969 Paranesidea onslowensis Hartmann, 1978 Paranesidea onslowensis Hartmann, 1978: 71, pl. 2, figs. 3-5; Paranesidea onslowensis Howe and McKenzie, 1989: 4, figs. 14, 43, 45; Paranesidea onslowensis Yassini et al., 1993: 383, pl. 1, figs. 1-4. Material: MD972132-990. Plate 1, Fig. 10 (jmr04)

Genus Neonesidea Maddocks, 1969 Neonesidea australis (Chapman, 1914) Neonesidea australis Yassini et al., 1993: 384, pl. 1, figs. 9, 10. Material: MD972132-1060. Plate 1, Fig. 8 (jmr06)

Superfamily CYPRIDACEA Baird, 1845 Family PARACYPRIDIDAE Sars, 1923 Genus Phlyctenophora Brady, 1880 Phlyctenophora zealandica Brady, 1880 Macrocypris orientalis Brady, 1868: 61-62, pl. 7, figs. 1-3; Phlyctenophora zealandica Brady, 1880: 33, pl. 3, figs. 1a-m; Paracypris zealandica Kingma, 1948: 67, pl. 6, fig. 18; Phlyctenophora cf. zealandica Hartmann, 1978: 147-8; pl. 14, figs. 21-23; Phlyctenophora orientalis Whatley and Zhao, 1987: 336-7, pl. 2, figs. 3,4; Phlyctenophora cf. zealandica Howe and McKenzie, 1989: 6, figs. 47-49; Phlyctenophora orientalis Mostafawi, 1992: 163, pl. 8, fig. 175; Phlyctenophora zealandica Yassini et al., 1993: 386, pl. 2, figs. 31, 32; Phlyctenophora orientalis Dewi, 1997: 57, figs. 31-32. Material: MD972132-980. Plate 1, Fig. 9 (jmr73)

Genus Phlyctocythere Keij, 1958 Phlyctocythere cf. P. pellucida (Müller, 1894) Phlyctocythere cf. P. pellucida Yassini et al., 1993: 386, pl. 3, figs. 43, 44. Material: MD972132-0. Plate 3, Fig. 4 (jmr97)

Family PONTOCYPRIDIDAE Müller, 1894 Genus Argilloecia Sars, 1866 Argilloecia cf. A. lunata Frydl, 1866 Argilloecia cf. A. lunata Mostafawi, 1992: 166, pl. 8, fig. 187; Argilloecia cf. A. lunata Dewi, 1997: 58, figs. 35-36. Material: MD972132-1310. Plate 1, Fig. 15 (jmr12)

Genus Pontocypris Sars, 1866 Pontocypris cf. P. attenuata (Brady, 1868) Pontocypris cf. P. attenuata Mostafawi, 1992: 163, pl. 8, fig. 177; Pontocypris cf. P. attenuata Dewi, 1997: 58, figs. 40-42. Material: MD972132-1130. Plate 1, Fig. 20 (jmr15)

Genus Pontocypria Müller, 1894 Pontocypria sp. Mostafawi, 1992 Pontocypria sp. Mostafawi, 1992: pl. 8, fig. 176; Pontocypria sp. Dewi, 1997: 59, fig. 43. Material: MD972132-1130. Plate 1, Fig. 17 (jmr13)

437

Superfamily CYTHERACEA Baird, 1850 Family CYTHERETTIDAE Triebel, 1952 Genus Alocopocythere Siddiqui, 1971 Alocopocythere goujoni (Brady, 1868) Cythere goujoni Brady, 1868: 78, pl. 10, figs. 9, 10; Neocytheretta pandirugosa Labutis, 1977 (unpublished): 186,pl. 38, figs. 1-4; Alocopocythere reticulata indoaustralica Hartmann, 1978: 90, pl. 7, figs. 1, 2; Alocopocythere goujoni Whatley and Zhao, 1988: 20-21, pl. 9, figs. 11, 12; Alocopocythere reticulata indoaustralica Howe and McKenzie, 1989: 46; Alocopocythere goujoni Mostafawi, 1992: 148, pl. 4, fig. 78; Alocopocythere goujoni Yassini et al., 1993: 400, pl. 8, figs. 149, 150; Alocopocythere goujoni Dewi, 1997: 72, figs. 194-195. Material: MD972132-1180. Plate 2, Fig. 5 (jmr25)

Genus Neocytheretta van Morkhoven, 1963 Neocytheretta adunca (Brady, 1880) Cythere adunca Brady, 1880: 97, pl. 25, figs 6a-d; Neocytheretta adunca Whatley and Zhao, 1988: 22, pl. 9, figs. 23-28; Neocytheretta adunca Yassini et al., 1993: 400, pl. 7, figs. 138, 139; Neocytheretta adunca Dewi, 1997: 74, figs. 205, 209-212. Material: MD972132-1130. Plate 2, Fig. 1 (jmr78)

Neocytheretta spongiosa (Brady, 1870) Cythere spongiosa Brady, 1870: 194, pl. 30, figs. 1-2; Neocytheretta spongiosa Whatley and Zhao, 1988: 22, pl. 9, figs. 20-22; Neocytheretta spongiosa Mostafawi, 1992: 150, pl. 5, fig. 97; Neocytheretta spongiosa Yassini et al., 1993: 400, pl. 7, figs. 140-142, pl. 8, fig. 153; Neocytheretta spongiosa Dewi, 1997: 73, figs. 201, 203. Material: MD972132-960. Plate 2, Fig. 4 (jmr77)

Neocytheretta spinobifurcata Yassini et al., 1993 Neocytheretta spinobifurcata Yassini et al., 1993: 398, pl. 7, figs. 126, 127. Material: MD972132-20. Plate 2, Fig. 3 (jmr53)

Neocytheretta vandijki (Kingma, 1948) Cythereis vandijki Kingma, 1948: 83-4, pl. 9, fig. 13; Neocytheretta cf. N. pandirugosa Labutis, 1977 (unpublished): 189, pl. 2, fig. 5, pl. 37, fig. 16-20; Neocytheretta vandijki Mostafawi, 1992: 150, pl. 5, fig. 99; Neocytheretta vandijki Dewi, 1997: 73, figs 198-199, 202. Material: MD972132-950. Plate 2, Fig. 2 (jmr81)

Neocytheretta cornea Labutis, 1977 (unpublished) Neocytheretta cornea Labutis, 1977(unpublished): 184, pl. 2, fig. 7, pl. 37, figs. 6-10. Material: MD972132-0. Plate 2, Fig. 6 (jmr79)

Subfamily ARCULACYTHERINAE Hartmann, 1981 Genus Hemikrithe van den Bold, 1950 Hemikrithe sp. Hemikrithe sp. Whatley and Zhao, 1988: 25, pl. 10, fig. 7. Materials: MD972132-1060. Plate 3, Fig. 15 (jmr21)

Hemikrithe sp. 1 Description: Carapace oblong, anterior broadly rounded, posterior rounded ventrally, approaching straight dorsally, peaked medially. Dorsum broadly convex, ventrum concave. Anterior marginal pore canals long, sinuous, clustered. Posterior pore canals short and well spaced. Irregular line of concresence. Hinge heterodont. Muscle scar pattern characteristic of the genus with a line of four adductor scars, a v-shaped frontal scar and two dorsal muscle scars. Carapace thin, surface pitted with deep, well-opened sieve pores, finely reticulate with longitudinal ridges. Materials: MD972132-1280 Plate 3, fig. 16 (jmr40)

Superfamily CYTHEROIDEA Baird, 1850 Family TRACHYLEBERIDIDAE Sylvester-Bradley, 1948 Subfamily TRACHYLEBERIDINAE Sylvester-Bradley, 1948 Genus Stigmatocythere Siddiqui, 1971 Stigmatocythere indica (Jain, 1977) Stigmatocythere indica Whatley and Zhao, 1988: 9, pl. 29, figs. 20, 21; Stigmatocythere indica Dewi, 1997: 68, figs. 151-2, 156. Materials: MD972132-1290. Plate 3, Fig 1 (jmr20)

Subfamily PTERYGOCYTHERINAE Puri, 1957 Genus Venericythere Mostafawi, 1992 Venericythere papuensis (Brady, 1880) Cythere papuensis Brady, 1880: 95, pl. 25, figs. 5a-d; Cythereis papuensis Kingma, 1948: 81-2, pl. 10, fig. 2; Cythere papuensis Puri and Hulings, 1976: 283, pl. 16, figs. 7, 11-18; Keijella papuensis Whatley and Zhao, 1988: 13, pl. 8, figs. 1-2; Venericythere papuensis Mostafawi, 1992: 146, pl. 4, figs. 71-72; Venericythere papuensis Yassini et al.,1993: 396, pl. 7, figs. 124-5; Venericythere papuensis Dewi, 1997: 70, figs. 170-171. Materials: MD972132-1220. Plate 2, Fig. 7 (jmr26)

439

Venericythere darwini (Brady, 1868) Cythere darwini Brady, 1868: 71, pl. 8, figs. 17, 18; Ruggiera darwini Whatley and Zhao, 1988: 16, pl. 8, fig. 10-13; Bicornucythere cf. darwini Howe and McKenzie, 1989: 42, figs. 34, 120, 138; Venericythere darwini Mostafawi, 1992: 146, pl. 4, figs. 69-70; Venericythere darwini Dewi, 1997: 71, figs. 174-5. Materials: MD972132-440. Plate 2, Fig. 8 (jmr67)

Genus Pistocythereis Gou 1983, in Gou et al. Pistocythereis bradyi (Ishizaki, 1968) Pistocythereis bradyi Whatley and Zhao, 1988: 19, pl. 9, figs 3-5; Pistocythereis bradyi Mostafawi, 1992: 146, pl. 4, fig. 75. Materials: MD972132-30. Plate 2, Fig. 9 (jmr54)

Pistocythereis euplectella (Brady, 1869) Cythere euplectella Brady, 1869: 157-8, pl. 16, figs. 5-7; Cythere euplectella Puri and Hulings, 1976, pl. 25, figs. 14-18; ?Lankacythere euplectella Whatley and Zhao, 1988: 18, pl. 9, figs. 1-2; Pistocythereis euplectella Mostafawi, 1992: 146, pl. 4, fig. 74. Pistocythereis euplectella Dewi, 1997: 72, figs. 190-3. Materials: MD972132-0. Plate 2, Fig. 10 (jmr85)

Pistocythereis cribriformis (Brady, 1866) Cythere cribriformis Brady, 1880: 96, pl. 19, figs. 3a-d; Cythereis cribriformis Kingma, 1948: 78-9: pl. 9, fig. 3; Pistocythereis cribriformis Whatley and Zhao, 1988: 19, pl. 9, figs. 6-7; Pistocythereis cribriformis Mostafawi, 1992: 146, pl. 4, fig. 73; Pistocythereis cribriformis Yassini et al., 1993: 398, pl. 7, fig. 128; Pistocythereis cribriformis Dewi, 1997: 72, fig. 189. Materials: MD972132-1040. Plate 2, Fig. 14 (jmr84)

Genus Pterygocythereis Blake, 1933 Pterygocythereis velivola (Brady, 1880) Cythere velivola Brady, 1880: 111, pl. 23, figs. 4a-c; Cythere velivola Puri and Hulings, 1976: 293, pl. 15, figs. 9-16; Pterygocythereis velivola Yassini et al., 1993: 398, pl. 8, figs. 146-148. Materials: MD972132-1130. Plate 3, Fig. 3 (jmr27)

Family BYTHOCYTHERIDAE Sars, 1926 Genus Bythocytheropteron Whatley and Zhao, 1987 Bythocytheropteron alatum Whatley and Zhao, 1987 Bythocytheropteron alatum Whatley and Zhao, 1987: 344, pl. 3, figs. 23-28; Bythocytheropteron alatum Mostafawi, 1992: 160, pl. 7, fig. 161; Bythocytheropteron alatum Dewi, 1997: 61, figs. 75-79. Material: MD972132-1050. Plate 3, Fig. 10 (jmr37)

Genus Rhombobythere Schornikov, 1982 Rhombobythere alata Schornikov, 1982 Rhombobythere alata Yassini et al., 1993: 393, pl. 6, figs. 119, 120. Materials: MD972132-250. Plate 3, Fig. 6 (jmr57)

Genus Bythoceratina Hornibrook, 1953 Bythoceratina hastata Mostafawi, 1992 Bythoceratina sp. 2 Whatley and Zhao, 1987: 343, pl. 3, fig. 17; Bythoceratina hastata Mostafawi, 1992: 159, pl. 7, figs. 148-150; Bythoceratina hastata Dewi, 1997: 60-1, figs. 65-67. Materials: MD972132-10. Plate 3, Fig. 9 (jmr95)

Family CYTHERIDEIDAE Sylvester-Bradley and Harding Genus Bishopina Bonaduce et al., 1976 Bishopina spinulosa (Brady, 1868) Cytheridea spinulosa Brady, 1868: 182 Cytheridea spinulosa Brady, 1880: 112 Bishopina spinulosa Howe and McKenzie, 1989: 16, figs. 62, 63. Materials: MD972132-1170. Plate 3, Fig. 5 (jmr90)

Family NEOCYTHERIDEIDAE Puri, 1957 Genus Copytus Skogsberg, 1939 Copytus posterosulcus Wang, 1985 Copytus posterosulcus Whatley and Zhao, 1988: 345, pl. 4, figs. 6-8; Copytus posterosulcus Mostafawi, 1992: 142, pl. 3, fig. 52; Copytus posterosulcus Yassini et al., 1993: 402, pl. 8, figs. 151, 152; Copytus posterosulcus Dewi, 1997: 62, figs. 92-3. Materials: MD972132-1130. Plate 1, Fig. 12 (jmr75)

Family KRITHIIDAE Mandelstam, 1960 Genus Parakrithella Hanai, 1961 Parakrithella pseudornata (Hanai, 1959) Parakrithella pseudornata Whatley and Zhao, 364, pl. 4, figs. 11-12; Parakrithella pseudornata Mostafawi, 1992: 142, pl. 2, fig. 48; Parakrithella pseudornata Dewi, 1997: 63, figs. 94, 95. Materials: MD972132-1200. Plate 2, Fig. 19 (jmr86)

Genus Pseudopsammocythere Carbonel, 1966 Pseudopsammocythere cf. P. reniformis (Brady, 1868) Paradoxostoma reniforme Brady, 1868: 224 Pseudopsammocythere cf. P. reniformis Howe and McKenzie, 1989: 26, fig. 27; Parakrithe placida Mostafawi, 1992: 142, pl. 3, figs. 49-51; Pseudopsammocythere cf. P. reniformis Yassini et al., 1993: 386, pl. 3, figs. 40-42; Pseudopsammocythere cf. P. reniformis Dewi, 1997: 63, figs. 96-7. Materials: MD972132-1170. Plate 1, Fig. 11 (jmr07)

441

Family CYTHERURIDAE Müller, 1894 Subfamily CYTHEROPTERONINAE Müller, 1894 Genus Cytheropteron Sars, 1866 Cytheropteron wrighti Yassini and Jones, 1987 Cytheropteron wrighti Yassini et al., 1993: 404, pl. 9, figs. 159, 160; Cytheropteron cf. C. wilmablomae Dewi, 1997: 64, figs. 107-9. Materials: MD972132-1020. Plate 3, Fig. 12 (jmr92)

Subfamily CYTHERURINAE Müller, 1894 Genus Semicytherura Wagner, 1957 Semicytherura sp. Yassini et al., 1993 Semicytherura sp. Yassini et al., 1993: 404, pl. 19, fig. 166. Materials: MD972132-970. Plate 3, Fig. 11 (jmr91)

Family HEMICYTHERIDAE Puri, 1953 Subfamily ORIONINAE Puri, 1953 Genus Caudites Coryell and Fields, 1937 Caudites cf. C. javana Kingma, 1948 Caudites medialis var. javana Kingma, 1948: 85, pl. 10, fig. 5; Caudites cf. C. javana Hartmann, 1978: 101, pl. 9, figs. 8-9; Caudites cf. C. javana Howe and McKenzie, 1989: 39, fig. 162. Materials: MD972132-1150. Plate 3, Fig. 7 (jmr87)

Caudites exmouthensis Hartmann, 1978 Caudites exmouthensis Hartmann, 1978: 102, pl. 9, figs. 10-11; Caudites exmouthensis Whatley and Zhao, 1988: 7, pl. 29, figs. 8, 9; Caudites sp. 2 Dewi, 1997: 67, fig. 143. Materials: MD972132-1060. Plate 3, Fig. 8 (jmr29)

Family CYTHERIDAE Baird, 1850 Subfamily CYTHERINAE Baird, 1850 Genus Microcytherura Müller, 1894 Microcytherura cf. M. punctatella Howe and McKenzie, 1989 Microcytherura punctatella Howe and McKenzie, 1989: 14, fig. 64. Materials: MD972132-90. Plate 3, Fig. 13 (jmr42)

Genus Mediocytherideis Mandelstam, 1956 Subgenus Mediocytherideis (Sylvestra) Dorak, 1973 Mediocytherideis (Sylvestra) jellineki Yassini et al., 1993 Tanella gracilis Kingma, 1948: 87, pl. 10, fig. 7; Tanella gracilis Hartmann, 1978: 80, pl. 4, figs. 4-13; Tanella cf. gracilis Whatley and Zhao, 1988: 6, pl. 6, figs. 5,6; Tanella gracilis Mostafawi, 1992: 139, pl. 2, fig. 40; Mediocytherideis (Sylvestra) jellineki Yassini et al., 1993: 388&405, pl. 3, figs. 48-52. Tanella gracilis Dewi, 1997: 67-8, fig. 147. Materials: MD972132-1300 Plate 4, Fig. 3 (jmr99)

Mediocytherideis (Sylvestra) jellineki carpentariensis Yassini et al., 1993 Mediocytherideis (Sylvestra) jellineki carpentariensis Yassini et al., 1993: 388&405, pl. 3, figs. 53-55. Materials MD972132-0. Plate 4, Fig. 7 (jmr100)

Subfamily LEPTOCYTHERIDAE Hanai, 1957 Genus Leptocythere Sars, 1925 Leptocythere hartmanni McKenzie, 1967 Leptocythere hartmanni Hartmann, 1978: 79, figs. 101-7; Leptocythere cf. hartmanni Howe and McKenzie, 1989: 30; Leptocythere hartmanni Yassini et al., 1993: 386, pl. 3, fig. 45. Materials: MD972132-560. Plate 4, Fig. 1 (jmr10)

Leptocythere hartmanni lacustris (De Deckker, 1981) Leptocythere lacustris De Deckker, 1981a: 29, figs. 24-25; Leptocythere cf. lacustris Howe and McKenzie, 1989: 30, fig. 92. Materials: MD972132-320. Plate 5, Fig. 1 (jmr60)

Genus Callistocythere Ruggieri, 1953 Callistocythere warnei Howe and McKenzie, 1989 Callistocythere warnei Howe and McKenzie, 1989: 28, fig. 89; Callistocythere warnei Yassini et al., 1993: 388, pl. 3, figs. 46, 47. Materials: MD972132-1130. Plate 4, Fig. 2 (jmr11)

Family LOXOCONCHIDAE Sars, 1925 Genus Loxoconcha Sars, 1866 Loxoconcha judithae Howe and McKenzie, 1989 Loxoconcha judithae Howe and McKenzie, 1989: 24, fig. 79; Loxoconcha judithae Yassini et al., 1993: 394, pl. 6, figs. 105-8. Materials: MD972132-1240. Plate 3, Fig. 14 (jmr18)

Family PARADOXOSTOMATIDAE Brady and Norman, 1889 Genus Cytherois Müller, 1884 Cytherois sp. Yassini et al., 1993 Cytherois sp. Yassini et al, 1993: 386, pl. 12, figs. 36, 37. Materials: MD972132-1390. Plate 1, Fig. 14 (jmr09)

Genus Paracytherois Müller, 1894 Paracytherois sp. Whatley and Zhao, 1988 Paracytherois sp. Whatley and Zhao, 1988: 27, pl. 10, fig. 26. Materials: MD972132-1310. Plate 1, Fig. 13 (jmr16)

Family CYTHEROMATIDAE Elofson, 1939 Genus Paracytheroma Juday, 1907 Paracytheroma mangrovicola (Hartmann, 1978) Cytheroma mangrovicola Hartmann, 1978: 81, pl. 4, figs. 15-16; Paracytheroma mangrovicola Howe and McKenzie, 1989: 12, figs. 2, 58; Paracytheroma mangrovicola Yassini et al., 1993: 384, pl. 2, figs. 38, 39. Materials: MD972132-1200. Plate 1, Fig. 16 (jmr14)

443

Genus Javanella Kingma, 1948 Javanella kendengensis Kingma, 1948 Javanella kendengensis Kingma, 1948: 89, pl. 10, fig. 6; Javanella kendengensis Mostafawi, 1992: 162, pl. 8, fig. 170; Javanella kendengensis Yassini et al., 1993: 384, pl. 2, figs. 33-35. Materials: MD972132-1130. Plate 1, Fig. 18 (jmr08)

Family PECTOCYTHERIDAE Hanai, 1957 Genus Labutisella McKenzie, 1989 in Howe and McKenzie Labutisella quadrata Howe and McKenzie, 1989 Labutisella quadrata Howe and McKenzie, 1989: 38, fig. 102; Labutisella quadrata Yassini et al., 1993: 388, pl. 4, figs. 59-61. Materials: MD972132-1100. Plate 4, Fig. 6 (jmr93)

Labutisella darwinensis Howe and McKenzie, 1989 Labutisella darwinensis Howe and McKenzie, 1989: 37, figs. 31, 99-101; Labutisella darwinensis Yassini et al., 1993: 388, pl. 4, figs. 62-4. Materials: MD972132-1050. Plate 4, Fig. 5 (jmr39)

Genus Praemunita Labutis, 1989 in Howe and McKenzie Praemunita broomensis (Hartmann), 1978 Pectocythere ?broomensis Hartmann, 1978: 145, pl. 14, figs. 15-20; Praemunita broomensis Howe and McKenzie, 1989: 36, fig. 96; Praemunita broomensis Yassini et al., 390, pl. 4, figs. 72-74. Materials: MD972132-1050. Plate 4, Fig. 4 (jmr94)

Praemunita capeyorkiana Yassini et al., 1993 Praemunita capeyorkiana Yassini et al., 1993: 392, pl. 3, figs. 56-58. Materials: MD972132-1050. Plate 4, Fig. 8 (jmr39)

Praemunita sp. Yassini et al., 1993 Praemunita sp. Yassini et al., 1993: 390, pl. 4, figs. 70, 71. Materials: MD972132-1130. Plate 4, Fig. 9 (jmr22)

Genus Keijia Teeter, 1975 Keijia nordaustraliae Howe and McKenzie, 1989 Keijia nordaustraliae Howe and McKenzie, 1989: 32, fig. 74; Keijia nordaustraliae Yassini et al., 1993: 392, pl. 4, figs. 78-80. Materials: MD972132-1100. Plate 4, Fig. 10 (jmr98)

Keijia australis Yassini et al., 1993 Keijia australis Yassini et al., 1993: 392, pl. 4, figs. 68, 69. Materials: MD972132-0. Plate 4, Fig. 11 (jmr14)

Family SCHIZOCYTHERIDAE Mandelstam, 1960 Subfamily SCHIZOCYTHERINAE Mandelstam, 1960 Genus Neomonoceratina Kingma, 1948 Neomonoceratina bataviana (Brady, 1868) Cytherura bataviana Brady 1868: 65, pl. 8, figs. 7-9; Neomonoceratina columbiformis Kingma, 1948: 95, pl. 10, figs. 8; Neomonoceratina bataviana Whatley and Zhao, 1989: 338-9, pl. 2, figs. 20; Neomonoceratina microreticulata Howe and McKenzie, 1989: 12, figs. 59; Neomonoceratina bataviana Mostafawi, 1992: 138, pl. 1, figs. 19, 20; Neomonoceratina bataviana Yassini et al., 1993: 393, pl. 5, figs. 84-90; Neomonoceratina bataviana Dewi, 1997: 59, figs. 44-6. Materials: MD972132-1150. Plate 4, Fig. 12 (jmr71)

Neomonoceratina porocostata Howe and McKenzie, 1989 Neomonoceratina porocostata Howe and McKenzie, 1989: 60, figs. 60, 61; Neomonoceratina porocostata Yassini et al., 1993: 393, pl. 5, figs. 93-7. Materials: MD972132-0. Plate 4, Fig. 13 (jmr89)

Subfamily PAIJENBORCHELLINAE Deroo, 1966 Genus Paijenborchella Kingma, 1948 Paijenborchella solitaria Ruggieri, 1962 Paijenborchella solitaria Yassini et al., 1993: 393, pl. 5, figs. 101-104. Materials: MD972132-10. Plate 3, Fig. 2 (jmr88)

Family XESTOLEBERIDIDAE Sars, 1928 Genus Xestoleberis Sars, 1928 Xestoleberis darwinensis Howe and McKenzie, 1989 Xestoleberis darwinensis Howe and McKenzie, 1989: 20, figs. 15, 51, 52. Materials: MD972132-1060. Plate 4, Fig. 16 (jmr33)

Genus Foveoleberis Malz, 1980 Foveoleberis cypraeoides (Brady, 1868) Cythere cypraeoides Brady, 1868: 72, pl. 8, figs. 21, 22; Xestoleberis foveolata Brady, 1880: 130, pl. 30, fig 1; Xestoleberis foveolata Kingma, 1948: 98, pl. 8, fig. 10; Foveoleberis cypraeoides Whatley and Zhao, 1988: 26, pl. 10, figs. 18, 19; Foveoleberis cypraeoides Mostafawi, 1992: 158, pl. 6, figs. 139; Foveoleberis cypraeoides Yassini et al., 1993: 402, pl. 9, figs. 171, 172; Foveoleberis cypraeoides Dewi, 1997: 75, figs. 224, 226-7, 229. Materials: MD972132-1060. Plate 4, Fig. 14 (jmr34)

445

Family LIMNOCYTHERIDAE Sars, 1925 Subfamily LIMNOCYTHERINAE Sars, 1925 Genus Limnocythere Brady, 1867 Limnocythere sp. Description: Carapace small to medium, rectangular, reticulate. Three main depressions on each valve, separating two smooth dorsal bosses. Vertical column of four muscle scars present. Sexual dimorphism pronounced ; female larger adult vale, male better defined reticulation. Materials: MD972132-60. Plate 5, Fig. 2 (jmr104)

Family CYTHERIDEIDAE Sars, 1925 Subfamily CYTHERIDEINAE Sars, 1925 Genus Cyprideis Jones, 1857 Cyprideis australiensis Hartmann, 1978 Cyprideis australiensis Hartmann, 1978: 85, pl. 5, figs. 1-5; Cyprideis australiensis Yassini et al., 1993: 392, pl. 4, figs. 81-3. Materials: MD972132-910. Plate 5, Fig. 4 (jmr66)

Superfamily CYPRIDOIDEA Baird, 1845 Family ILYOCYPRIDIDAE Kaufmann, 1900 Subfamily ILYOCYPRIDINAE Kaufmann, 1900 Genus Ilyocypris Brady and Norman, 1889 Ilyocypris australiensis Sars, 1889 Ilyocypris australiensis Yassini et al., 1993: 393, pl. 5, figs. 91, 92. Material: MD972132-100. Plate 5, Fig. 3 (jmr58)

Subfamily CYPRIDINAE Baird, 1845 Genus Cyprinotus Brady, 1886 Cyprinotus cf. C. cingalensis Brady, 1886 Cyprinotus cingalensis Neale, 1979: 78, Plates 1, 2. Description: Carapace large, subtriangular in lateral view, inequivalved. Right valve protrudes over left with hump, denticulate antero-ventrally and postero-ventrally. Hump smooth, translucent, thickened shell material, rest of carapace finely reticulate. Material: MD972132-30. Plate 5, Fig. 5 (jmr55)

Subfamily HERPETOCYPRIDINAE Kaufmann, 1900 Genus Candonocypris Sars, 1896 Candonocypris cf. C. novaezealandicae (Brady, 1843) Candonocypris novaezealandicae De Deckker, 1981: 54-4, fig. 6. Description: Carapace smooth and elongated ellipsoid with tapering ends. Dorsum arched, ventrum slightly concave in center. Narrow in dorsal view. Broad inner lamella anteriorly, prominent selvage. Material: MD972132-100. Plate 5, Fig. 6 (jmr109)

Subfamily CYPRETTINAE Hartmann, 1963 Genus Cypretta Vavra, 1895 Cypretta sp. Description: Carapace medium-sized, subovate in lateral view, bulbous in dorsal view. Dorsum arched, ventrum slightly inflexed, anterior and posterior rounded. Surface pitted throughout. Presence of radial striae in the inner lamella. Material: MD972132-80. Plate 5, Fig. 8 (jmr46)

Genus Zonocypretta De Deckker, 1981 Zonocypretta sp. Description: Carapace medium-sized, elongated in lateral view. Dorsum arched, ventrum slightly concave, anterior and posterior broadly rounded. Shell sculptured with longitudinal lines all over, divided by small transverse ridges. Material: MD972132-110. Plate 5, Fig. 9 (jmr56)

Superfamily DARWINULACEA Brady and Norman, 1889 Family DARWINULIDAE Brady and Norman, 1889 Genus Darwinula Brady and Robertson, 1885 Darwinula sp. Description: Carapace smooth, elongate, oblong in lateral view. Dorsum slightly convex, ventrum straight. Broadly rounded anterior, narrower, tapering posterior. Muscle scar in rosette arrangement, characteristic of the genus. Material: MD972132-80. Plate 5, Fig. 7 (jmr45)

447

449

Appendix 7. Ostracod plates.

PLATE 1

Fig. 1 Polycope favus Brady, 1880 Sample MD32-10; jmr48, RV ext. Fig. 2 Polycope sp. I Yassini et al., 1993 Sample MD32-10; jmr49, RV ext. Fig. 3 Cytherella semitalis Brady, 1868 Sample MD32-1070; jmr03, RV ext. Fig. 4 Cytherella cf. C. hemipuncta (Swanson, 1969) Sample MD32-1390; jmr72, RV (juv.) ext. Fig. 5 Cytherelloidea malaccaensis Whatley and Zhao, 1987 Sample MD32-70; jmr70, LV ext. Fig. 6 Cythereolloidea cf. C. excavata (Mostafawi, 1992) Sample MD32-1160; jmr02, LV ext. Fig. 7 Cytherelloidea sp. Sample MD32-10; jmr50, RV ext. Fig. 8 Neonesidea australis (Chapman, 1914) Sample MD32-1060; jmr06, RV ext. Fig. 9 Phlyctenophora zealandica Brady, 1880 Sample MD32-980; jmr73, LV ext. Fig. 10 Paranesidea onslowensis Hartmann, 1978 Sample MD32-990; jmr04, LV ext. Fig. 11 Pseudopsammocythere cf.P. reniformis (Brady, 1868) Sample MD32-1170; jmr07, RV ext. Fig. 12 Copytus posterosulcus Wang, 1985 Sample MD32-1130; jmr75, RV ext. Fig. 13 Paracytherois sp. Whatley and Zhao, 1988 Sample MD32-1310; jmr16, RV ext. Fig. 14 Cytherois sp. Yassini et al., 1993 Sample MD32-1390; jmr09, LV ext. Fig. 15 Argilloecia cf. A. lunata Frydl, 1866 Sample MD32-1310; jmr12, RV ext. Fig. 16 Paracytheroma mangrovicola (Hartmann, 1978) Sample MD32-1200; jmr14, RV ext. Fig. 17 Pontocypria sp. Mostafawi, 1992 Sample MD32-1130; jmr13, LV ext. Fig. 18 Javanella kendengensis Kingma, 1948 Sample MD32-1130; jmr08, RV ext. Fig. 19 Parakrithella pseudornata Hanai, 1959 Sample MD32-1200; jmr86, RV ext. Fig. 20 Pontocypris cf. P. attenuata (Brady, 1868) Sample MD32-1310; jmr12, LV ext.

451

453

PLATE 2

Fig. 1 Neocytheretta adunca (Brady, 1880) Sample MD32-1130; jmr78, LV ext. Fig. 2 Neocytheretta vandijki (Kingma, 1948) Sample MD32-950; jmr81, LV ext. Fig. 3 Neocytheretta spinobifurcata Yassini et al., 1993 Sample MD32-20; jmr53, LV ext. Fig. 4 Neocytheretta spongiosa (Brady, 1870) Sample MD32-960; jmr82, LV ext. Fig. 5 Alocopocythere goujoni (Brady, 1868) Sample MD32-1180; jmr25, LV ext. Fig. 6 Neocytheretta cornea Labutis, 1977 (unpublished) Sample MD32-0; jmr79, LV ext. Fig. 7 Venericythere papuensis (Brady, 1880) Sample MD32-1220; jmr26, LV ext. Fig. 8 Venericythere darwini (Brady, 1868) Sample MD32-440; jmr67, RV ext Fig. 9 Pistocythereis bradyi (Ishizaki, 1968) Sample MD32-30; jmr54, LV ext. Fig. 10 Pistocythereis euplectella (Brady, 1868) Sample MD32-0; jmr85, RV ext. Fig. 11 Pistocythereis cribriformis (Brady, 1866) Sample MD32-1040; jmr84, LV ext.

455

457

PLATE 3

Fig. 1 Stigmatocythere indica (Jain, 1977) Sample MD32-1290; jmr20, LV ext. Fig. 2 Paijenborcella solitaria Ruggieri, 1962 Sample MD32-10; jmr88, RV ext. Fig. 3 Pterygocythereis velivola (Brady, 1880) Sample MD32-1130; jmr27, RV ext. Fig. 4 Phlyctocythere cf. C. pellucida (Müller, 1894) Sample MD32-0; jmr97, LV ext. Fig. 5 Bishopina spinulosa (Brady, 1868) Sample MD32-1170; jmr90, RV ext. Fig. 6 Rhombobythere alata Schornikov, 1982 Sample MD32-250; jmr57, LV ext. Fig. 7 Caudites cf. C. javana (Kingma, 1947) Sample MD32-1150; jmr87, RV ext. Fig. 8 Caudites exmouthensis Hartmann, 1978 Sample MD32-1060; jmr29, RV ext. Fig. 9 Bythoceratina hastata Mostafawi, 1992 Sample MD32-10; jmr10, LV ext. Fig. 10 Bythocytheropteron alatum Whatley and Zhao, 1987 Sample MD32-1050; jmr37, LV ext. Fig. 11 Semicytherura sp. Yassini et al., 1993 Sample MD32-970; jmr91, LV ext. Fig. 12 Cytheropteron wrighti Yassini and Jones, 1987 Sample MD32-1020; jmr92, LV ext. Fig. 13 Microcytherura cf. M. punctaella Howe and McKenzie, 1989 Sample MD32-90; jmr42, RV ext. Fig. 14 Loxoconca judithae Howe and McKenzie, 1989 Sample MD32-1240; jmr18, LV ext. Fig. 15 Hemikrithe sp. Whatley and Zhao, 1988 Sample MD32-1060; jmr21, RV (juv) ext. Fig. 16 Hemikrithe sp.1 (sp. nov.) Samples MD32-1280; jmr40, RV ext.

459

461

PLATE 4

Fig. 1 Leptocyhere hartmanni McKenzie, 1967 Sample MD32-560; jmr10, RV ext. Fig. 2 Callistocythere warnei Howe and McKenzie, 1989 Sample MD32-1130; jmr11, RV ext. Fig. 3 Mediocytherideis (Sylvestra) jellineki Yassini et al., 1993 Sample MD32-1300; jmr99, LV ext. Fig. 4 Praemunita broomensis (Hartmann, 1978) Sample MD32-1050; jmr94, LV ext. Fig. 5 Labutisella darwinensis Howe and McKenzie, 1989 Sample MD32-1050; jmr39, RV ext. Fig. 6 Labutisella quadrata Howe and McKenzie, 1989 Sample MD32-1050; jmr39, RV ext. Fig. 7 Mediocytherideis (Sylvestra) jellineki carpentariensis Yassini et al., 1993 Sample MD32-0; jmr100, LV ext. Fig. 8 Praemunita capeyorkiana Yassini et al., 1993 Sample MD32-1050; jmr39, RV ext. Fig. 9 Praemunita sp. Yassini et al., 1993 Sample MD32-1130; jmr22, RV ext. Fig. 10 Keijia nordaustralie Howe and McKenzie, 1989 Sample MD32-1100; jmr98, RV ext. Fig. 11 Keijia australis Yassini et al., 1993 Sample MD32-0; jmr14, LV ext. Fig. 12 Neomonoceratina bataviana (Brady, 1868) Sample MD32-1150; jmr71, LV ext. Fig. 13 Neomonoceratina porocostata Howe and McKenzie, 1989 Sample MD32-0; jmr89, LV ext. Fig. 14 Xestoleberis darwinensis Howe and McKenzie, 1989 Sample MD32-1060; jmr33, RV ext. Fig. 15 Foveoleberis cypraeoides (Brady, 1868) Sample MD32-1060; jmr34, LV ext.

463

465

PLATE 5

Fig. 1 Leptocythere lacustris De Deckker, 1981 Sample MD32-320; jmr60, RV ext. Fig. 2 Limnocythere sp. Sample MD32-60; jmr104, LV female ext. Fig. 3 Ilyocypris australiensis Sars, 1889 Sample MD32-100; jmr58, LV ext. Fig. 4 Cyprideis australiensis Hartmann, 1978 Sample MD32-910; jmr66, LV ext. Fig. 5 Cyprinotus cf. C. cingalensis Brady, 1886 Sample MD32-30; jmr55, RV ext. Fig. 6 Candonocypris cf. C. novaezealandicae (Baird, 1843) Sample MD32-100; jmr109, LV ext. Fig. 7 Darwinula sp. Sample MD32-80; jmr45, RV ext. Fig. 8 Cypretta sp. Sample MD32-80; jmr46, LV ext. Fig. 9 Zonocypretta sp. Sample MD32-110; jmr56, LV ext.

467

Appendix 8. Ostracod genus distribution.

469

Sample Number 0 10 10 20 30 40 50 60 70 80 90 100 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410 420 430 440 450 460 470 480 490 500 510 520 530 540 550 560 670 680 690 Genera Polycope 0 7000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Cytherella 16 0020 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Cytherelloidea 5 2010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Paranesidea 31 20 3 9 0 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Neonesidea 13 20 1 3 0 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Phlyctenophora 3 0010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Phlyctocythere 8 5000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Argilloecia 3 2010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Pontocypris 8 2000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Neocytheretta 21 24 1 4 0 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Hemikrithe 5 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 1 0 1 V. papuensis 8 10 1 9 0 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 V. darwini 3 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 00000083# # 1152012# 921 8000000 0 0 0 0 0 0 Pistocythereis 5 2020 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Pterygocythereis 5 12 0 4 0 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Bythocytheriids 0 2000 0 0 0 0 0 0 0 00000000 000001 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Bishopina 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Copytus 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Parakrithella 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Pseudopsammocythere 5 7020 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Cytheropteron 5 2020 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Caudites 0 2000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Microcytherura 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 L. hartmanni 10 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 000000399842311628# 2 9744362648191 14 13 21 0 0 0 L. lacustris 0 0000 0 6 9 1 2 1 0 0261 924795 9162075 0200 0 0 000000000000000000316122 0 0 0 0 0 0 Callistocythere 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Mediocytherideis 3 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Loxoconcha 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Cytherois 0 0010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Paracytheroma 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Javanella 0 0010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Labutisella 8 5010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Praemunita 3 2010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Keijia 5 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Neomonoceratina 23 70100 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Paijenborchella 5 5010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Xestoleberis 3 2100 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Foveoleberis 5 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Limnocythere 0 0000 0 0 2 1 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Cyprideis 0 15 0 0 1 1 23 53 4 9 6 1 4 17 55 # 119 9 69 55 41 34 17 10 24 16 13726 7 4 332735933000000001590291485202 68 30 15 6 0 0 Ilyocypris 5 2062 34 5 13 4 4 2 0021311654 121212 0601 0 0 0023100000000000000000 0 0 0 0 0 0 Cyprinotus 0 0010 0 0 0 1 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Candonocypris 0 0000 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Cypretta 0 0010 0 0 0 0 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0 Zonocypretta 0 0000 0 0 0 1 0 0 0 00000000 000000 0000 0 0 0000000000000000000000 0 0 0 0 0 0

Genera 820 820 830 840 850 870 880 890 900 910 920 930 940 950 960 970 980 990 ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 00000000000000000000 0 00000000000000000Polycope 0 0 0 0 0 0 0 0 0 00000 0 1 0 0 1 2313331 203201221 0111 01 0 01100 1114 37613 586637Cytherella 0 0 0 0 0 0 0 0 0 00001 0 0 0 0 0 113110 0010 0100 1000 00 0 00000 0000 0010 0082Cytherelloidea 0 0 0 0 0 0 0 0 0 001012 1 4 8 4 6 1315220 0000 0640 0000 00 0 00000 0000 0000 0000Paranesidea 0 0 0 0 0 0 0 0 0 00216 2 6 5 2 3 4730652 104323150 0000 00 0 00000 0000 0000 0000Neonesidea 0 0 0 0 0 0 0 0 0 00001 0 1 0 0 0 100000 0000 0000 0000 00 0 00000 0000 0200 1152Phlyctenophora 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Phlyctocythere 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 102000 0000 0000 0000 00 0 00000 0100 0000 0000Argilloecia 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 001000 0000 0200 0000 00 0 00000 0000 0010 0020Pontocypris 0 0 0 0 0 0 0 0 0 00328 2 6 4 3 7 3427221 0011 0941 4125 01 0 01000 0001 1001 1134Neocytheretta 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 001000 0000 0000 0000 00 0 00000 1101 1221 11123Hemikrithe 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 1 101000 0000 0002 3212 23 5 09742 2021 1121 0000V. papuensis 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000V. darwini 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 001000 0000 0000 0000 00 0 00000 0000 0000 0000Pistocythereis 0 0 0 0 0 0 0 0 0 00010 2 1 1 1 0 538321 101001510 0001 00 0 00000 0000 0000 0000Pterygocythereis 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Bythocytheriids 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0200 1100 00 0 00000 0000 0000 0003Bishopina 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 1 101000 0000 0400 0000 01 1 01000 0000 0000 0000Copytus 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 1 112100 0001 0713 2343 31 1 12241 1221 2011 1000Parakrithella 0 0 0 0 0 0 0 0 0 00001 3 2 3 2 0 758421 3011 2511 0212 21 1 02111 1101 0000 0040Pseudopsammocythere 0 0 0 0 0 0 0 0 0 00000 3 0 0 2 1 1063620 303202110 1001 00 0 00000 0000 0000 0000Cytheropteron 0 0 0 0 0 0 0 0 0 00001 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Caudites 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Microcytherura 0 0 0 0 0 0 0 0 0 011100 0 0 0 0 0 010100 0000 0200 0000 00 0 00000 0000 0000 0000L. hartmanni 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000L. lacustris 0 0 0 0 0 0 0 0 0 02001 0 0 0 0 0 110100 0000 0101 0000 00 0 00000 0000 0000 0011Callistocythere 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Mediocytherideis 0 0 0 0 0 0 0 0 0 00001 0 0 0 0 0 113100 101001295141229383435 39 95719334201200000 0000Loxoconcha 0 0 0 0 0 0 0 0 0 01001 0 0 0 0 0 100000 0000 0100 0000 00 0 00000 0000 0000 0010Cytherois 0 0 0 0 0 0 0 0 0 00001 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Paracytheroma 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 001010 0000 0410 0000 00 0 00000 0000 0000 0000Javanella 0 0 0 0 0 0 0 0 0 00000 3 1 1 1 0 012110 0010 0000 0010 00 0 00000 0000 0000 0001Labutisella 0 0 0 0 0 0 0 0 0 01121 5 5 5 4 3 743332 202401253 5535 33 4 13232 2121 1211 2194Praemunita 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0010 0000 0000 00 0 00000 0000 0000 0000Keijia 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 220210 0001 0595 2102 01 1 01111 1110 1121 1174Neomonoceratina 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000100 0000 0511 0000 00 0 00000 0000 0000 0000Paijenborchella 0 0 0 0 0 0 0 0 0 00001 1 0 0 1 1 459972 303404652 2211 00 0 00000 0000 0000 0000Xestoleberis 0 0 0 0 0 0 0 0 0 00001 0 0 1 0 0 115110 0000 0210 1100 00 0 00000 0000 0000 0000Foveoleberis 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Limnocythere 0 0 0 0 0 0 0 # 42 516200 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Cyprideis 0 0 0 0 0 0 0 0 0 01000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Ilyocypris 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Cyprinotus 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Candonocypris 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Cypretta 0 0 0 0 0 0 0 0 0 00000 0 0 0 0 0 000000 0000 0000 0000 00 0 00000 0000 0000 0000Zonocypretta

471

Appendix 9. Stable isotope results from core MD-32.

Unit Depth cm Genus Valve Wt (μg) VPI* δ13C δ18O 1 0 Neocytheretta R 28 1.5 -1.9 0.0 1 0 Neocytheretta L 25 2.5 -3.5 5.7 1 0 Neocytheretta L 21 3 -2.1 -1.7 1 0 Neocytheretta L 27 3 -3.7 -2.9 1 0 Neocytheretta L 29 3 -3.1 -2.5 1 10 Neocytheretta R 41 1.5 -5.8 -2.0 1 10 Neocytheretta L 28 1.5 -4.7 -2.0 1 15 Neocytheretta L 34 1.5 -5.4 -2.2 1 15 Neocytheretta L 40 2 -4.7 -2.3 1 20 Neocytheretta L 30 2 -5.3 -1.5 1 25 Neocytheretta L 21 2 -2.5 -2.4 1 30 Cyprideis LF 59 2 -8.6 -1.2 1 30 Neocytheretta L 47 1.5 -4.0 -2.4 1 35 Neocytheretta L 29 1.5 -3.0 -2.2 2a 40 Ilyocypris L 17 1.5 -0.2 1.4 2a 40 Ilyocypris R 17 1.5 3.0 0.4 2a 40 Ilyocypris 3L 26 2.5 -4.0 1.9 2a 40 Ilyocypris 2L, 1R 36 2.5 -0.4 10.0 2a 40 Cyprideis RM 78 4 -4.7 -0.9 2a 40 Ilyocypris 2R (1A-1), 2L 28 3 -4.4 -0.2 2a 50 Ilyocypris 2L 24 1 -3.9 0.7 2a 50 Ilyocypris 4L 30 2 -3.3 7.5 2a 50 Ilyocypris 3L, 3R (A-1) 57 4 -1.8 1.6 2a 50 Ilyocypris 4L 33 2 -6.6 -1.6 2a 55 Ilyocypris 2L, 2R 25 1.5 -1.1 6.0 2a 55 Ilyocypris 4L (A-1) 42 3 -2.0 1.3 2a 55 Ilyocypris 4L, 1R (A-1/2) 57 3 -3.4 -0.2 2a 60 Ilyocypris L 19 1.5 -0.9 4.9 2a 60 Ilyocypris L 20 2 -1.8 -0.6 2a 60 Ilyocypris 2L, 1R 32 1.5 -2.0 9.4 2a 60 Ilyocypris 2L, 2R (A-1) 51 3 -3.7 -0.6 2a 65 Cyprideis RF 99 3 -3.4 1.5 2a 65 Cyprideis LF 98 3 -3.6 2.2 2a 65 Ilyocypris 2L 33 2 -1.5 3.7 2a 65 Ilyocypris 2R 33 2 -1.4 6.8 2a 70 Cyprideis RF 83 1.5 -3.8 -1.4 2a 70 Ilyocypris L, R 28 1.5 -2.8 0.9 2a 70 Neocytheretta L 32 1.5 -5.1 -2.0 2a 70 Cyprideis LF 117 2 -2.7 -1.7 2a 70 Ilyocypris L 12 1.5 -2.6 3.1 2a 70 Ilyocypris 3L 22 1.5 -2.9 10.5 2a 70 Ilyocypris 3R 19 1.5 -2.3 11.4 2a 70 Ilyocypris 3L, 3R (A-1) 56 3 -3.7 1.8 2a 75 Cyprideis LF 109 2 -3.5 0.5 2a 75 Ilyocypris 1L, 1R 11 2 -2.6 8.3 2a 80 Ilyocypris 2L 29 1.5 -3.1 2.8 2a 80 Ilyocypris 2L, 2R 26 1.5 -3.0 13.2 2a 85 Cyprideis LM 88 2 -4.2 2.2 2a 90 Ilyocypris R 14 1.5 -3.3 -2.7 2a 90 Ilyocypris 1L, 2R (A-1) 18 2 -3.3 13.4 2a 95 Cyprideis L (A-1) 92 1.5 -4.4 2.8

2a 95 Ilyocypris 2R, 1L 26 2.5 -1.9 8.5 2a 100 Cyprideis LF 88 2 -2.2 -0.6 2a 100 Cyprideis RF 71 1.5 -2.2 -0.8 2a 100 Ilyocypris L 22 1.5 -3.9 -1.8 2a 100 Ilyocypris 3R 22 2.5 -1.4 6.9 2a 100 Cyprideis RF 80 3.5 -3.5 2.5 2b 120 Cyprideis LF(A-1) 34 3 -2.9 -1.9 2b 130 Cyprideis LM 44 1 -4.3 -1.7 2b 130 Cyprideis RM 67 1.5 -2.7 -0.2 2b 130 Cyprideis L (A-1) 29 2 -2.8 4.0 2b 130 Cyprideis R (A-1) 28 2 -2.4 6.0 2b 130 Cyprideis 1(A-3), 2 (A-5) 51 2 -3.8 -1.6 2b 135 Cyprideis LM 80 2 -3.2 3.4 2b 135 Cyprideis RM 66 2 -3.9 4.8 2b 135 Cyprideis LM 76 2 -4.5 3.2 2b 135 Cyprideis RF 71 3 -2.8 -0.7 2b 135 Cyprideis LF 74 3 -2.6 -1.5 2b 140 Cyprideis LM 74 1.5 -2.9 -0.4 2b 140 Ilyocypris R 19 1.5 -1.4 -0.4 2b 145 Cyprideis RM 55 1.5 -2.8 2.5 2b 145 Cyprideis RF 79 1.5 -3.4 21.6 2b 145 Cyprideis RF 97 2.5 -2.6 1.6 2b 145 Cyprideis RF 64 2.5 -2.6 3.2 2b 145 Cyprideis L (A-4), R (A-5) 63 2 -4.1 -2.3 2b 145 Cyprideis LF 68 2 -3.2 -0.1 2b 150 Cyprideis R 57 1 -4.4 -0.7 2b 150 Cyprideis L 70 1 -3.4 -0.2 2b 150 Cyprideis RM 65 1 -3.1 -0.4 2b 150 Cyprideis LF 69 1 -4.8 -1.5 2b 150 Cyprideis RF 58 1 -4.5 -1.9 2b 150 Cyprideis RM 69 1 -3.2 0.1 2b 150 Cyprideis RF 83 1 -4.2 -0.9 2b 160 Cyprideis LM 63 1 -4.7 -0.6 2b 160 Cyprideis LM 74 1 -4.0 -1.8 2b 160 Cyprideis RM 52 1 -2.3 -1.0 2b 160 Cyprideis RM 48 1 -3.4 -1.0 2b 160 Cyprideis RF 62 1 -3.9 -1.9 2b 160 Cyprideis LM 65 1 -4.1 -2.2 2b 160 Cyprideis LF 68 1 -3.3 0.0 2b 160 Ilyocypris R 22 1.5 -2.1 -2.2 2b 160 Cyprideis LF 72 1 -3.4 -0.2 2b 160 Cyprideis RF 64 1 -3.4 -1.4 2b 165 Cyprideis LM 65 2 -3.8 5.0 2b 170 Ilyocypris R 27 1 -1.1 0.9 2b 170 Ilyocypris L 26 1.5 -1.1 0.7 2b 170 Ilyocypris 4L 74 1.5 -1.4 8.5 2b 180 Cyprideis LF 72 1 -4.5 -0.9 2b 180 Cyprideis LF 84 1 -3.9 -2.2 2b 180 Cyprideis LF 60 1.5 -3.3 1.5 2b 180 Cyprideis RF 68 1.5 -4.0 1.5 2b 190 Cyprideis LM 66 1 -4.3 -2.3 2b 190 Cyprideis RM 60 1 -4.3 -2.2

473

2b 190 Cyprideis LF 65 1.5 -3.4 4.0 2b 190 Cyprideis RF 76 1.5 -3.5 3.4 2b 190 Cyprideis RM 61 1.5 -3.9 2.5 2b 200 Cyprideis LF 86 1 -3.5 -0.1 2b 200 Cyprideis RF 75 1 -3.3 -0.2 2b 200 Cyprideis LF 71 1.5 -2.9 3.9 2b 200 Cyprideis RF 61 1.5 -2.9 1.8 2b 210 Cyprideis RM 63 1 -3.7 -0.4 2b 210 Cyprideis LM 73 1 -5.0 -0.9 2b 210 Cyprideis LF 77 1 -3.9 -0.9 2b 210 Cyprideis RF 65 1 -3.0 1.6 2b 210 Ilyocypris L 17 1 -1.4 1.7 2b 210 Cyprideis RM 58 1 -4.6 0.0 2b 210 Cyprideis LF 72 1 -3.7 0.1 2b 210 Cyprideis RM 53 1 -3.4 -1.1 2b 220 Cyprideis RM 61 3 -4.7 -1.6 2b 220 Cyprideis RF 60 3 -4.0 -1.5 2b 220 Cyprideis RF 53 3 -4.3 -1.7 2b 220 Ilyocypris 1L, 2R 54 1.5 -2.1 26.5 2b 230 Cyprideis LM 71 1 -4.0 -0.9 2b 230 Cyprideis RF 74 1 -4.1 -2.0 2b 230 Ilyocypris 1L, 2R 53 2 -2.3 12.1 2b 240 Cyprideis RM 63 3 -3.9 0.7 2b 240 Cyprideis RF 74 1.5 -4.1 -0.1 2b 240 Cyprideis RM 70 2.5 -3.8 -0.3 2b 240 Cyprideis LF 75 2.5 -3.6 -0.1 2b 250 Cyprideis LM 73 1.5 -4.7 -1.1 2b 250 Cyprideis RM 70 1.5 -4.8 -1.3 2b 250 Ilyocypris L 23 1.5 -1.4 -0.2 2b 250 Ilyocypris 2L, 1R 52 1.5 -1.7 -0.6 2b 265 Ilyocypris 2R 50 2 -1.7 -0.6 2b 270 Cyprideis LF 80 4 -4.8 -2.4 2b 270 Ilyocypris L 26 2 -1.0 1.3 2b 270 Ilyocypris 1L, 2R 70 2 -1.5 -0.4 2c 280 Ilyocypris L 15 1.5 -2.4 -0.5 2c 290 Ilyocypris L 16 1.5 -1.2 0.1 2c 290 Ilyocypris 2R 22 2 -2.6 -0.9 2c 300 Ilyocypris L 21 1.5 -1.5 -0.6 2c 350 Ilyocypris L 24 2 -3.2 -0.6 2c 350 Ilyocypris 1L, 1R 45 2.5 -1.7 -0.7 2c 350 Cyprideis RF 33 5 -5.0 0.0 3a 380 Venericythere RM 14 4 -5.5 -1.8 3a 380 Venericythere 2RM 42 5 -4.3 -0.3 3a 390 Venericythere LF 23 4 -6.0 -1.6 3a 400 Venericythere LF 7 3 -5.6 -2.4 3a 400 Venericythere LF, RF 36 2-3.5 -6.5 -2.8 3a 410 Venericythere RM 19 4 -7.0 -2.4 3a 410 Venericythere 1L, 1R 36 3 -5.9 -0.7 3a 420 Venericythere RM 19 4 -5.6 -2.6 3a 420 Venericythere LF, RM 31 2 -5.6 0.0 3a 430 Venericythere LF(A-1) 9 3 -6.5 -0.1 3a 440 Venericythere LF 29 2 -7.7 -4.0

3a 440 Venericythere 2LF 36 2 -6.4 -0.8 3a 440 Venericythere RF, LF 31 4.5 -6.7 1.0 3a 450 Venericythere LF, RM 28 5 -8.1 -4.3 3a 470 Venericythere LF, RM 45 2 -6.0 0.3 3b 480 Cyprideis RF 60 3 -5.0 -1.2 3b 480 Cyprideis LF 61 4 -4.9 -1.6 3b 490 Cyprideis LF 66 4 -4.5 -1.1 3b 490 Cyprideis LF 55 4 -6.0 -3.3 3b 500 Cyprideis LF 65 3.5 -4.1 -1.0 3b 500 Cyprideis LF 61 3.5 -4.8 -2.3 3b 500 Cyprideis LM 55 5 -5.3 -1.9 3b 500 Cyprideis RF 66 5 -4.8 4.2 3b 510 Cyprideis LM 64 5 -4.5 -1.9 3b 510 Cyprideis LF 64 3.5 -4.6 0.8 3b 510 Cyprideis LM 60 3.5 -5.0 -2.2 3b 520 Cyprideis LM 52 5 -5.2 0.2 3b 520 Cyprideis RM 46 4 -5.4 0.1 3b 520 Cyprideis LM 39 3.5 -5.0 -0.4 3b 520 Cyprideis LM 61 3.5 -5.3 -2.0 3b 540 Cyprideis LF 60 5 -3.9 1.0 3b 540 Cyprideis LM 51 5 -6.0 0.3 3b 540 Cyprideis RF 57 5 -4.3 0.4 3b 550 Cyprideis LF 83 5 -3.8 -1.3 3b 550 Cyprideis LF 40 5 -4.7 -0.6 3b 560 Cyprideis LM 22 5 -4.8 -1.2 5 897 Cyprideis LM 30 4 -6.6 -4.3 5 899 Cyprideis LM 40 2.5 -7.4 -4.0 5 899 Cyprideis LM 46 3 -7.2 -2.8 5 899 Cyprideis RF 41 3 -3.6 0.8 5 900 Cyprideis LM 34 4 -8.5 -5.0 5 900 Cyprideis RM 48 5 -10.8 -5.1 5 900 Cyprideis RM(A-1) 27 5 -6.1 -4.4 5 900 Cyprideis LF 47 3 -6.5 -2.2 5 900 Cyprideis RF 44 3 -3.7 -0.6 5 900 Cyprideis LM 24 3.5 -7.1 -4.5 5 900 Cyprideis RM 39 3.5 -9.1 3.4 5 900 Cyprideis LF 38 4.5 -7.1 -5.6 5 900 Cyprideis RM 25 4.5 -8.0 -7.8 5 900 Cyprideis RF 45 3 -7.6 5.0 5 900 Cyprideis RF 33 2.5 -4.9 -2.8 5 901 Cyprideis LF 52 3 -5.5 -0.9 5 901 Cyprideis LF 47 4.5 -10.0 -4.6 5 902 Cyprideis LF 42 3 -6.7 -0.2 5 902 Cyprideis LF 32 4 -9.9 1.8 5 903 Cyprideis LM 61 3 -5.8 5.2 5 903 Cyprideis LF 38 3 -6.7 -7.4 5 904 Cyprideis RM 40 3 -6.8 0.8 5 904 Cyprideis LF 41 4 -8.8 0.9 5 905 Cyprideis LF 33 2 -6.2 -3.1 5 905 Cyprideis LF 43 2.5 -6.2 -2.8 5 905 Cyprideis LF 51 3.5 -9.0 -5.2 5 905 Cyprideis LF 54 2.5 -8.6 0.9

475

5 906 Cyprideis LF 42 4 -4.3 0.8 5 906 Cyprideis LF 42 4 -8.4 1.6 5 906 Cyprideis RM 46 4 -5.1 -5.6 5 907 Cyprideis LM 51 3 -7.1 -2.0 5 907 Cyprideis RM 38 3 -8.2 2.1 5 907 Cyprideis LM 41 4 -8.6 4.4 5 907 Cyprideis LF 47 4 -7.9 1.4 5 908 Cyprideis LM 47 5 -5.8 -1.4 5 908 Cyprideis LM 22 3.5 -5.5 -1.3 5 908 Cyprideis LM 38 4.5 -3.9 1.2 5 909 Cyprideis LF 45 3 -4.0 0.5 5 909 Cyprideis RM 40 5 -3.8 1.3 5 910 Cyprideis LM 32 5 -3.5 2.3 5 910 Cyprideis RM 30 5 -3.4 2.1 5 910 Cyprideis RM 40 5 -6.1 -7.0 5 910 Cyprideis LF 22 3.5 -2.9 0.4 5 910 Cyprideis LF 36 4 -4.7 -2.9 5 910 Cyprideis RF 32 2.5 -3.7 -0.4 5 911 Cyprideis LM 50 3.5 -8.2 3.1 5 911 Cyprideis RF 44 3 -12.4 5.3 5 911 Cyprideis LF 49 3.5 -12.9 8.3 5 913 Cyprideis LF 38 4 -4.5 0.9 5 914 Cyprideis LF 55 3 -4.6 1.6 5 915 Cyprideis LF 46 4 -7.4 -5.4 5 915 Cyprideis LF 54 3 -6.7 -3.4 5 916 Cyprideis RM 40 5 -2.5 1.7 5 916 Cyprideis RM 43 5 -2.6 1.5 5 917 Cyprideis LF 55 3 -6.3 -3.0 5 917 Cyprideis LM 42 3 -7.1 -4.5 5 917 Cyprideis RM 43 3.5 -7.5 -0.3 5 917 Cyprideis RF 43 3.5 -6.6 -3.2 5 919 Cyprideis RF 42 4 -3.4 1.2 5 920 Cyprideis RF 36 5 -4.8 -0.7 5 920 Cyprideis RF 36 2.5 -4.7 -0.5 5 925 Cyprideis RM 32 4 -5.9 0.6 5 925 Cyprideis LF 47 4 -5.7 -0.6 5 930 Cyprideis RM(A-1) 9 4 -7.2 -4.6 5 930 Cyprideis LF 28 4 -9.1 -2.0 5 930 Cyprideis LF 40 4 -5.1 0.5 6a 940 Neocytheretta W 69 4 -5.7 -2.0 6a 950 Neocytheretta whole 72 3 -6.2 -2.2 6a 960 Neocytheretta L 34 3.5 -5.8 -1.7 6a 960 Neocytheretta R 30 3.5 -4.6 -2.3 6a 960 Neocytheretta L 33 3.5 -5.5 -2.0 6a 960 Neocytheretta L 24 2.5 -3.8 -2.8 6a 980 Neocytheretta L 33 3 -5.3 -2.2 6a 980 Neocytheretta R 36 3 -5.4 -2.5 6a 980 Neocytheretta L 29 3 -4.8 -1.7 6a 980 Neocytheretta R 27 3.5 -5.3 -1.8 6a 990 Neocytheretta R 31 4 -5.3 -2.2 6a 990 Neocytheretta R 32 3.5 -4.3 -2.1 6a 990 Neocytheretta L 30 4 -3.7 -2.5

6a 990 Neocytheretta R 36 4 -5.4 -2.3 6a 1000 Neocytheretta R 25 2 -4.7 -2.5 6a 1000 Neocytheretta R 27 3.5 -4.7 -2.3 6a 1010 Neocytheretta L 54 2 -6.8 -1.5 6a 1010 Neocytheretta R 25 2.5 -4.8 -2.6 6a 1010 Neocytheretta R 30 3 -4.1 -2.3 6b 1020 Neocytheretta L 19 4 -5.0 -2.7 6b 1030 Neocytheretta R 28 2 -4.9 -2.1 6b 1030 Neocytheretta L 21 2.5 -3.3 -2.8 6b 1040 Neocytheretta R 20 2 -4.0 -3.0 6b 1040 Neocytheretta R 35 4 -4.8 -1.8 6b 1040 Neocytheretta L 25 2 -3.1 -2.4 6b 1050 Neocytheretta L 22 2 -3.2 -2.6 6b 1055 Neocytheretta L 31 3 -5.3 1.2 6b 1055 Neocytheretta L 27 2 -2.6 -2.4 6b 1065 Neocytheretta L 30 3 -3.5 -1.8 6c 1070 Neocytheretta L 68 4 -4.5 -1.6 6c 1075 Neocytheretta L 25 4 -4.6 -2.2 6c 1075 Neocytheretta Whole 52 2 -4.5 -2.0 6c 1115 Neocytheretta L 24 2 -3.8 -1.6 6d 1130 Neocytheretta R 31 1.5 -4.4 -2.6 6d 1130 Neocytheretta L 26 1.5 -3.4 -2.3 6d 1130 Neocytheretta R 31 2.5 -5.4 -1.8 6d 1130 Neocytheretta L 26 2 -4.5 -2.4 6d 1140 Neocytheretta R 23 1.5 -4.4 -1.4 6d 1140 Neocytheretta R 16 1.5 -5.0 -1.9 6d 1150 Neocytheretta R 29 1.5 -5.9 -1.8 6d 1150 Neocytheretta R 31 1.5 -5.2 -1.4 6d 1160 Neocytheretta R 41 1 -6.3 -1.9 6d 1160 Neocytheretta R 38 1 -5.2 -2.1 6d 1160 Neocytheretta L 38 1 -5.9 -2.5 6e 1170 Neocytheretta R 29 1.5 -6.3 -1.6 6e 1170 Neocytheretta L 30 3 -6.2 -1.7 6e 1180 Neocytheretta L 31 2 -6.1 -2.0 6e 1190 Neocytheretta L 30 3 -6.2 -2.1 6e 1190 Neocytheretta L 38 2 -5.9 -2.1 6e 1210 Neocytheretta R 40 2 -4.6 -1.8 6e 1210 Neocytheretta L 41 2 -4.6 -1.8 6f 1310 Neocytheretta R 49 4 -5.2 -1.6 6f 1310 Neocytheretta L 51 4 -5.1 -1.6 6f 1320 Neocytheretta R 23 2 -4.9 0.0 6f 1330 Neocytheretta R 33 1.5 -4.6 -1.6 6f 1340 Neocytheretta R 32 3 -5.0 -2.0 6f 1350 Neocytheretta R 38 2 -5.1 -1.0 6f 1360 Neocytheretta L 36 2 -5.4 -0.5 6f 1380 Neocytheretta R 45 2 -5.5 -1.3 6f 1390 Neocytheretta R 35 2 -5.8 -1.0

VPI* refers to the preservation of the valve as per table 3.

477