The State of the Oligocene Icehouse World: Sedimentology, Provenance, and Stable Isotopes of Marine Sediments from the Antarctic Continental Margin

City University of New York (CUNY) CUNY Academic Works

All Dissertations, Theses, and Capstone Projects Dissertations, Theses, and Capstone Projects

5-2015

Daniel William Hauptvogel Graduate Center, City University of New York

How does access to this work benefit ou?y Let us know!

More information about this work at: https://academicworks.cuny.edu/gc_etds/966 Discover additional works at: https://academicworks.cuny.edu

This work is made publicly available by the City University of New York (CUNY). Contact: [email protected]

THE STATE OF THE OLIGOCENE ICEHOUSE WORLD: SEDIMENTOLOGY,

PROVENANCE, AND STABLE ISOTOPES OF MARINE SEDIMENTS FROM THE

ANTARCTIC CONTINENTAL MARGIN

DANIEL W. HAUPTVOGEL

A dissertation submitted to the Graduate Faculty in Earth and Environmental Sciences in partial

fulfillment of the requirements for the degree of Doctor of Philosophy, The City University of

New York

2015

This manuscript has been read and accepted for the Graduate Faculty in Earth and Environmental Sciences in satisfaction of the dissertation requirement for the degree of Doctor of Philosophy.

Dr. Stephen F. Pekar

Date Chair of Examining Committee

Dr. Cindy Katz

Date Executive Officer

Athanasios Koutavas

Cecilia M. Mchugh

Trevor Williams

Sidney R. Hemming

Supervisory Committee

THE CITY UNIVERSITY OF NEW YORK

iii

Abstract

The State of the Oligocene Icehouse World: Sedimentology, Provenance, and Stable Isotopes of

Marine Sediments from the Antarctic Continental Margin

Daniel W. Hauptvogel

Adviser: Dr. Stephen F. Pekar

Abstract

The Oligocene Epoch (34-23 Ma) was a dynamic time in Antarctica, with previous ice volume estimates suggesting fluctuations from below 50 % up to 140 % of modern all while atmospheric CO2 decreased from above 1,000 ppm in the Early Oligocene to near modern levels by the Late Oligocene. Most of what is known about the Oligocene Antarctic cryosphere however, is derived from distal sedimentary records that can only provide a generalized view of the cryospheric dynamics in Antarctica. To better understand regional differences in Antarctic glacial dynamics, proximal records are needed. This dissertation advances our understanding of these dynamics in Antarctica during the Oligocene by investigating three proximal, marine sediment cores from different regions of the continent.

Ice-rafted debris (IRD) concentrations, 40Ar/39Ar thermochronology, and stable isotope records combined from 3 proximal marine sediment cores reveal a large ice sheet existed throughout the Oligocene, with ice volume reaching up to 155 % of modern. Concentrations and

40Ar/39Ar thermochronology from IRD offshore of the Wilkes subglacial basin suggest the ice sheet was fairly stable on elevated portions such as the Adélie Craton, but the basin itself was more responsive to climate changes. These changes appear to be influenced by 405-kyr

iv eccentricity and 1.2 myr obliquity. In the Ross Sea, 40Ar/39Ar thermochronology from IRD show a large West Antarctic influence, indicative of a large ice sheet residing there during the Late

Oligocene. Stable isotopes from benthic foraminifera from the Maud Rise show ice volume fluctuations from below 50 % up to 155 % of modern, in agreement with modeling and far-field records. The isotope record is also influenced by 405-kyr and 100-kry eccentricity and does not show a warming trend during the Late Oligocene as seen in other isotope records. Together, these records are indicative of a near-modern size or larger ice sheet present in both East and

West Antarctica during the Oligocene, a time when the extent of Antarctica glaciation has been debated.

Acknowledgements

I would like to thank my Ph.D. advisor, Dr. Stephen F. Pekar, for his guidance, support, and allowing me the independence for completing this dissertation. He has been a mentor and has provided several opportunities for me over the past 4 years.

Dr. Sidney Hemming and Dr. Trevor Williams have been indispensible over the course of this research. Their knowledge and encouragement are one of the many reasons I have made it this far and I thank them for serving on my dissertation committee. I would also like to thank Dr.

Athanasios (Tom) Koutavas and Dr. Cecilia McHugh for serving on my dissertation committee.

A special thanks to Dr. Allan Ludman for always being someone to talk to and seek advice from, especially for my post-graduate future.

To my wife Christine, who has put up with me through my years of graduate school and has always supported me in my endeavors.

Sitting at a desk all day can become tedious, and I thank my office/lab-mate Matt for being there to keep me sane for the first 3 years of research. You are a good friend and I wish you all the luck in the future.

I would like to thank my mother, father, and the rest of my family for always believing in me and supporting me throughout my academic career.

Lastly I would not be where I am today if it weren’t for the encouragement of my high school chemistry and physics teacher, Joseph Fusco, whom truly opened my eyes to science.

Table of Contents Chapter 1. An Introduction to Antarctica and the Oligocene ...... 1

1.1 Geologic History of Antarctica ...... 3

1.2 The Early Icehouse World ...... 10

1.2.1 Trend in Cenozoic Climate ...... 11

1.2.3 Climate History of the Oligocene ...... 14

1.3 Ice-rafted debris history ...... 21

1.4 Use of 40Ar/39Ar in Ice-rafted Debris ...... 24

1.5 Previous Drilling Expeditions and Proximal Antarctic Records ...... 26

1.6 Questions addressed by this study ...... 30

1.7 Significance ...... 32

Chapter 2. Evidence for large East Antarctic Ice Sheet in the Wilkes Subglacial Basin during the Oligocene ...... 33

2.1 Introduction ...... 34

2.1.1 The Wilkes Sub-glacial Basin ...... 34

2.1.2 Lithostratigraphy of the Oligocene section at Site U1356 ...... 36

2.1.3 Regional geology of the WSB region ...... 40

2.1.4 Previous studies in the WSB region ...... 43

2.2 Methods ...... 45

2.2.1 Sampling ...... 45

2.2.2 Sediment Separation ...... 47

2.2.3 IRD concentrations ...... 48

2.2.4 40Ar/39Ar Thermochronology ...... 48 vii

2.3 Results ...... 50

2.4 Discussion ...... 54

2.4.1 Implications for ice volume during the Oligocene ...... 54

2.4.2 Gravity Flows: Evidence for a larger than modern ice sheet ...... 60

2.4.3 Ar Thermochronology ...... 61

2.4.4 Hornblende vs. Biotite as a provenance signal ...... 64

2.4.5 Small biotite vs. large biotite ...... 66

2.4.6 Age distribution in the high IRD bearing layers ...... 67

2.4.7 Age distribution in the low IRD abundance intervals ...... 69

2.4.8 Age Distribution of the Debris Flow Layers ...... 71

2.4.9 Sediment Transport during the Late Oligocene ...... 73

2.4.10 Implications for ice sheet changes the Oligocene ...... 74

2.5 Conclusion ...... 79

Chapter 3. Evidence from the Ross Sea DSDP Site 270 for a large West Antarctic Ice

Sheet during the Late Oligocene (25.8-25.3 Ma) ...... 81

3.1 Introduction ...... 81

3.1.1 Previous evidence for West Antarctic glaciation ...... 81

3.1.2 Importance and sensitivity of the WAIS ...... 82

3.1.3 DSDP Site 270 ...... 85

3.1.4 Ross Sea regional geology ...... 87

3.2 Methods ...... 90

3.2.1 Site 270 Age Model ...... 91

viii

3.2.2 Sediment Preparation ...... 92

3.3 Results ...... 93

3.3.1 Potential sources of IRD ...... 94

3.4 Discussion ...... 97

3.4.1 IRD from East vs. West Antarctica ...... 97

3.4.2 Biotite vs. Hornblende ...... 99

3.4.3 Downcore distribution of IRD ...... 100

3.4.4 West Antarctic Ice Sheet during the Oligocene ...... 107

3.5 Conclusion ...... 110

Chapter 4. Bottom water redistribution during the Late Oligocene: No evidence for a global warming trend ...... 112

4.1 Introduction ...... 112

4.1.1 Geology of the Maud Rise ...... 115

4.1.2 Previous studies from the MR drill sites ...... 118

4.1.3 Stable Isotopes ...... 120

4.2 Methods ...... 124

4.2.1 Site 690 sampling and age model ...... 124

4.2.2 Foraminifer methodology ...... 125

4.2.3 Calibration and Standardization of Foraminifera ...... 126

4.3 Results ...... 128

4.4 Discussion ...... 133

4.4.1 Stable oxygen and carbon isotopes ...... 133

4.4.2 Late Oligocene Apparent Sea Level and Ice Volume ...... 137

4.4.3 Late Oligocene Warming? ...... 143

4.4.4 Oligocene bottom water temperatures ...... 144

4.5 Conclusion ...... 146

Chapter 5. The state of the Oligocene icehouse world ...... 148

5.1 The evolution of the Antarctic ice sheet ...... 148

5.2 Sensitivity of the Wilkes Subglacial basin ...... 153

5.3 The West Antarctic Ice Sheet and the Ross Sea ...... 154

5.4 The Late Oligocene Warming ...... 155

5.5 Future Work ...... 156

Appendix ...... 158

References ...... 187

List of Figures

Figure 1-1 - Map of Antarctica with overlaid BEDMAP2 topography ...... 2

Figure 1-2 - Antarctic paleogeography in the early Mesoproterozoic ...... 4

Figure 1-3 - Assembly of Rodinia during the late Mesoproterozoic ...... 5

Figure 1-4 - Antarctic paleogeography from the middle Cambrian to the late Carboniferous ...... 6

Figure 1-5 - Rifting of Gondwana ...... 8

Figure 1-6 - Geography of the WARS ...... 9

Figure 1-7 - Reconstructed Oligocene alkenone-based CO2 records ...... 11

Figure 1-8 - The combined Oligocene benthic foraminiferal δ18O curve ...... 12

Figure 1-9 - Global Cenozoic sea-level record ...... 14

Figure 1-10 - Oligocene apparent sea-level and ice volum estimates ...... 16

Figure 1-11 - Maximum and minimum reconstructed Oligocene topography ...... 18

Figure 1-12 - Summary of Oligocene (34-25 Ma) recovery from proximal Antarctic drill sites. 27

Figure 1-13 - Biomagnetostratigraphic age-depth plot for Site U1356 ...... 29

Figure 2-1 - Drilling location map for IODP Expedition 318 ...... 35

Figure 2-2 - Recovery and lithology from IODP Site U1356 ...... 37

Figure 2-3 - Bedrock geology of the greater WSB region ...... 41

Figure 2-4 - Map of thermochronologic data from offshore sediment and on-land 40Ar/39Ar ages in the WSB region ...... 44

Figure 2-5 - Oligocene age model from IODP Site U1356 ...... 47

Figure 2-6 - Sedimentology and age model of IODP Site U1356 ...... 51

Figure 2-7 - Summary of all 40Ar/39Ar ages measured in the Late Oligocene sediment at IODP

Site U1356 ...... 54

Figure 2-8 - IRD concentration record from IODP Site U1356 ...... 56

Figure 2-9 - IRD intervals from well-recovered Late Oligocene strata ...... 57

Figure 2-10 - IRD intervals from well-recovered Early Oligocene strata ...... 58

Figure 2-11 - Distribution of all hornblende and biotite 40Ar/39Ar ages analyzed for this study. 66

Figure 2-12 - Distribution of all hornblende and biotite 40Ar/39Ar ages in higher IRD abundance layers (IRD concentrations >10 grains/gram)...... 68

Figure 2-13 - Downcore distribution of 40Ar/39Ar ages from higher IRD intervals ...... 69

Figure 2-14 - Distribution of hornblende and biotite 40Ar/39Ar ages in low IRD abundance layers

(IRD concentrations <10 grains/gram)...... 70

Figure 2-15 - Distribution of hornblende and biotite 40Ar/39Ar ages for the debris flow layers. . 72

Figure 2-16 - Downcore distribution of all hornblende and biotite 40Ar/39Ar ages...... 77

Figure 3-1 - Location map of DSDP Site 270 ...... 86

Figure 3-2 - Ross Sea Geology ...... 88

Figure 3-3 - Regional map of Ross Sea with offshore Holocene sediment and onland 40Ar/39Ar ages ...... 90

Figure 3-4 - DSDP Site 270 age model ...... 92

Figure 3-5 - Summary of all argon ages for DSDP Site 270...... 95

Figure 3-6 - Distribution of biotite and hornblende grains...... 100

Figure 3-7 - Downcore distribution of argon ages ...... 102

Figure 3-8 - Three scenarios for ice flow into the Ross Sea ...... 104

xii

Figure 4-1 - Location map of ODP Site 690 ...... 114

Figure 4-2 - Plate-tectonic reconstructions of the Maud Rise ...... 116

Figure 4-3 - Lagrangian particle tracks over the Maud Rise ...... 118

Figure 4-4 - Age model for ODP Site 690B for the Late Oligocene ...... 125

Figure 4-5 - Raw isotope for Site 690B relative to PDB ...... 130

Figure 4-6 - Cibicidoides spp. corrected values ...... 132

Figure 4-7 - Comparison of stable isotope data from Site 690B to other records ...... 134

Figure 4-8 - Oxygen isotope data against short and long eccentricity cycles ...... 137

Figure 4-9 - Apparent sea-level based on δ18O values from Figure 4-6 ...... 140

Figure 4-10 - Ice volume estimates based on the ASL values from Figure 4-9 ...... 141

Figure 5-1 - Summary and comparison of data collected during this study………………...…150

xiii

List of Tables

Table 2-1 – Paleomagnetic chrons from IODP Site U1356 for the Oligocene ...... 46

Table 4-1 – Stable isotope precision and offsets...... 131

xiv

Chapter 1. An Introduction to Antarctica and the Oligocene

The formation of continental scale ice sheets in the earliest Oligocene was concomitant with deep-sea cooling (e.g. Lear et al., 2004), biotic turnovers (e.g. Houben et al, 2013), and a gradual decline of atmospheric CO2 from ~1000 ppm to near preindustrial levels by the Late

Oligocene (Pagani et al., 2011). Our understanding of ice sheet dynamics during the Oligocene remains ambiguous, as most of what is known about the Antarctic cryosphere is derived from lower latitude, distal studies such as oxygen isotope records from foraminifera (e.g. Miller et al.,

1991; Zachos et al., 2001), shallow marine stratigraphy and sea-level records (e.g. Pekar et al.,

2002), and carbon isotope records (e.g. Pälike et al., 2006). Inherently, these data sets can only provide overall estimates of Antarctic ice volume rather than details of specific areas of the continent. While proximal data sets and drill sites from Antarctica that can detect the presence of ice in specific regions or determine the provenance of glacial sediment (e.g. Hauptvogel and

Passchier, 2012), the Oligocene has few proximal-Antarctic records. In addition, those that have been recovered are largely incomplete due to lack of recovery and/or contain large hiatuses (e.g.

Barrett, 1986a; Barrett, 1986b; Cape Roberts Science Team, 2000). This is due, in part, to overprinting and erosion from subsequent glaciations through the Neogene (e.g. Hambrey et al.,

1989). While the prominent ice streams that currently drain the Antarctic continent have been identified (e.g. Rignot et al., 2011) and those that existed in recent history (e.g. Livingstone et al.,

2011), we have little to no empirical evidence to identify paleo-ice streams or evaluate their dynamics during the inception of the ice house world. Additionally, the sensitivity of ice streams to climate forcing during the Oligocene, such as elevated atmospheric CO2 and orbital cyclicity,

is restricted to modeling results and limited proximal datasets. While distal records show

Oligocene ice volume was driven by 405 kyr eccentricity and 1.2 myr obliquity cycles (Wade and Pälike, 2004; Pälike et al., 2006), they cannot address how different areas of the Antarctic

Ice Sheet (AIS) responded.

− 30˚ 0˚ 30˚

60˚ DM − − 60˚ −60˚ 60˚

RnIS AP AIS PB East Antarctica

−90˚ 90˚ West Antarctica TAM

WARS

AB MBL RIS

RS AC WSB −120˚ 120˚ − 60˚ NVL

150˚ −60˚

150˚ 180˚ − Figure 1-1 – Map of Antarctica with overlaid BEDMAP2 topography (Fretwell et al., 2013). AB=Aurora Basin, AC=Adélie Craton, AIS=Amery Ice Shelf, AP=Antarctic Peninsula, DM=Dronning Maud Land, MBL=Marie Byrd Land, NVL=Northern Victoria Land, PB=Prydz Bay, RIS=Ross Ice Shelf, RnIS=Ronne Ice Shelf, RS=Ross Sea, TAM=Transantarctic Mountains, WARS=West Antarctic Rift System, WS=Weddell Sea.

1.1 Geologic History of Antarctica

Antarctica is approximately 14.0×106 km2 in area and is the fifth-largest continent on

Earth, almost the size of North America (Boger, 2010). It can be divided into two main regions,

East Antarctica or the East Antarctic Ice Sheet (EAIS) and West Antarctic or the West Antarctic

Ice Sheet (WAIS)(Figure 1-1). They are separated by the 3500 km long Transantarctic

Mountains (TAM). The geology of Antarctica is quite complex, and has evolved through accretionary and collisional additions and rift driven subtractions that began in the Archean

(Boger, 2010 and references therein). It is this evolution that has lead Antarctica to become part of at least two of the Earth's main supercontinents, Rodinia and Gondwana, and ultimately transforming the continent into its’ present day shape and size. The complete tectonic history of

Antarctica has already been summarized by Boger (2010), and a brief summary of major events described in that paper are presented below.

The nucleus of the Antarctic continent is believed to be the Late Archean/Early

Paleoproterozoic rocks of the George V and Terre Adélie Lands within the Gawler Craton

(Fanning et al., 1988; Flöttmann and Oliver, 1994; Oliver and Fanning, 2002; Peucat et al.,

1999), which is part of the larger Mawson Craton that extends into Antarctica (Figure 1-2).

While much of the inner portions of the continent are covered by kilometers thick ice, these

Archean rocks are thought to underlie much of the EAIS inland of these coastal areas (Fanning et al., 1995). Other Archean terrains exist in Antarctica, but they are composed of only fragments of much larger terrains incorporated into neighboring continents (i.e. Australia and Africa)

(Boger, 2010).

Figure 1-2 - Antarctic paleogeography in the early Mesoproterozoic (from Boger, 2010). Arc magmatism and accretion of terranes along the western margin of the North Australia, Curnamona–Beardmore, Laurentia and Mawson amalgam. Trans–Laurentian igneous suite intrudes Antarctica, Laurentia and possibly SW Australia

The oldest tectonic event on record to affect Antarctica was the Nimrod-Kimban orogenesis (Figure 1-2) and is the oldest continuous orogenic belt in Antarctica (Boger 2010).

This orogenesis has been dated between 1730 Ma and 1690 Ma and primarily resulted in generally high-grade metamorphism with the accretion of the Beardmore microcontinent, though regional differences exist (Brommer et al., 1999; Goodge et al., 2001; Will et al., 2009). Through the Proterozoic, Antarctica underwent several magmatic and accretionary events that ultimately

culminated in the Pinjarra Orogeny and assembly of Rodinia at 1,100 Ma (Grenville Orogeny in

Laurentia) (e.g. Goodge et al., 2008; McLelland et al., 2001)(Figure 1-3).

Figure 1-3 - Assembly of Rodinia during the late Mesoproterozoic (from Boger, 2010) by the Pinjarra Orogeny between the Crohn Block and the North Australia, Curnamona–Beardmore, Laurentia, Coompana-Nawa and Mawson amalgam. Domains signified with letters are: AF=Albany– Fraser, B=Beardmore, C=Coompana, Cur=Curnamona, M=Mazatzal and N=Nawa. Stars denote the known (black) and inferred (grey) plutons of the Trans–Laurentian igneous suite from Figure 2.

After the break-up of Rodinia by 680 Ma, East Antarctica began its final assembly in a series of Ediacaran and Cambrian collisions that lead to the Pan-African orogens, including the

East African-Antarctica and Kuunga orogens, and the assembly of Gondwana (580-490 Ma)

(McWilliams, 1981; Jacobs et al., 1998; Meert and Van der Voo, 1997; Cawood and Buchan,

2007). The Ross orogeny also began during this time, in which subduction and metamorphism

has been dated between 555 Ma and 510 Ma (e.g. Lira et al., 1997; Goodge et al., 1993; Rowell et al., 1993). Also during this time, the accretion of crust that would eventually make up West

Antarctica started (Cawood, 2005; Cawood et al., 2009). See Figure 1-4 for the paleogeographic interpretation of these events.

Figure 1-4 - Antarctic paleogeography from the middle Cambrian to the late Carboniferous (from Boger, 2010). Westward subduction under the Pacific margin of Gondwana marks onset of Terra Australis orogenesis (Ross Orogeny). Black stars mark exposures of the passive margin sequences: AF=Adelaide Fold Belt (Adelaidean and Kanmantoo Groups), CFB=Cape Fold Belt, EM=Ellsworth– Whitmore Mountains, NP=North Patagonia Terrane, PF=Puncoviscan Formation, PM=Pensacola Mountains (Hannah Range Formation), SFB=Sierra de la Ventana Fold Belt, TAM=Transantarctic Mountains (Beardmore Group and lower Byrd Group), VL=Victoria Land (Koettlitz Group). Terranes accreted from the Late Cambrian to the Late Carboniferous include: AP(e)=eastern domain Antarctic Peninsula, B=Bowers Terrane, Ch=Chilenia Terrane, ChP=Challenger Plateau (western domain New Zealand (NZ(w))), CP(w)=Campbell Plateau (western domain New Zealand), Cu=Cuyania Terrane, D=Deseado Terrane, LH=Lord Howe Rise, MB(r)=Ross Province Marie Byrd Land, NEFB=New England Fold Belt, P=Pampean Terrane, RBT=Robertson Bay Terrane and T=Tasmanides. Outboard terranes: AP(cw)=central and western domains of the Antarctic Peninsula, CP (e)=Campbell Plateau (eastern domain New Zealand) and MB(a)=Amundsen Province Marie Byrd Land.

By the Late Carboniferous, the Gondwanide Orogeny had begun, coinciding with the formation of Pangea (Cawood et al., 2009). This collision occurred along the northern and western margins of South America and Africa and was distal to Antarctica as it was already part of Gondwana Land (Boger, 2010). It is believed however, that this still had significant effects on plate motions (Cawood and Buchan, 2007) and can be seen in the folding of the Cape, Ellsworth-

Whitmore and the Sierra de la Ventana basins (Curtis, 2001). This folding coincides with the magmatic activity of the Antarctic Peninsula and a portion of Marie Byrd Land (Vaughan and

Storey, 2000; Pankhurst et al., 1998, 2006), which did not become a part of Antarctica until the

Cretaceous (Di Venere et al., 1995; Pankhurst et al., 1998; Vaughan et al., 2002).

Figure 1-5 - Rifting of Gondwana (from Boger, 2010). Boxed ages between continents give age of first appearance of seafloor between continents. DFZ=Davey Fracture Zone, KP=Keguelen Plume (black star), LG=Lambert Graben.

The last set of events that shaped Antarctica’s present form (especially West Antarctica) were extensional episodes during the Mesozoic and Cenozoic, in which Antarctica eventually became isolated from Earth’s landmasses (Figure 1-5). The first extensional event was the separation of Africa and South America from the Dronning Maud region of Antarctica during the

Jurassic (Boger, 2010). This was followed by the separation of Antarctica from India and

Australia in the mid-Jurassic and Cretaceous, between 160 and 125 million years ago (Stagg and

Wilcox, 1992) with new oceanic crust forming 95 million years ago (Boger, 2010; Veevers and

Eittreim, 1988). The final phase of rifting was when New Zealand separated from West

Antarctica during the Late Cretaceous (Reeves and de Witt, 2000) and the West Antarctic Rift

System (WARS) began to open the Ross Sea and separate the TAM from Marie Byrd Land

(Figure 1-6), triggering extensive volcanism of basalt to rhyolite (LeMasurier and Rex, 1991).

The rift valley was then filled with ~600 m of sediment eroded from the surrounding areas and marine sediment sedimentation (Rooney et al., 1991; Scherer et al., 1998).

Figure 1-6 - Geography of the WARS. East and West Antarctica are shown in present-day co- ordinates with Zealandia and Australia restored to their approximate pre-Gondwana breakup locations (Mortimer et al., 2011). Geographic features and sample sites: STR=South Tasman Rise, SGI=Surgeon Island, TAM=Transantarctic Mountains, NVL=northern Victoria Land, SVL=southern Victoria Land, IB=Iselin Bank, CAP=Campbell Plateau, CAI=Campbell Island, FDL=Fiordland, MBL=Marie Byrd Land, XC=Executive Committee Range, MM=Mount Murphy, EWM=Ellsworth Mountains. Geological units: do=Delamerian Orogen, lo=Lachlan Orogen, w=Wilson Terrane, b=Bowers Terrane, r5Robertson Bay Terrane, bu=Buller Terrane, tk=Takaka Terrane, mb=Median Batholith, ep=Eastern Province, wars=West Antarctic Rift

System, mg=Mulock Granite, lg=Liv Group, rp=Ross province (not to be confused with Ross Orogen), ap=Amundsen province. 1.2 The Early Icehouse World

There has been a recent push for the investigation of Earth’s climate during warmer or higher CO2 intervals. The last time atmospheric CO2 levels were as high as they are today was during the Pliocene (Seki et al., 2010), thus much attention has been given to this time period. It has been shown that the Pliocene was a dynamic time for Antarctica, and especially the East

Antarctic Ice Sheet. Climate-ice sheet modeling has shown that the Wilkes subglacial basin of

Antarctica may have witnessed a large retreat of the current grounding line anywhere from 100 to 1500 km inland (Dolan et al., 2011; Hill et al., 2007; Pollard and DeConto, 2009; Pollard et al., 2015). Sediment geochemistry from the marine continental shelf for this region indicates active erosion from further in the basin, suggesting a retreat of the ice sheet up to 500 km (Cook et al., 2013).

While atmospheric CO2 was similar to present in the Pliocene (~400 ppm), the last time levels were above 500 ppm was during the Oligocene (Hendriks and Pagani, 2008; Pagani et al.,

2009; 2011 [Figure 1-7]; Pearson et al., 2009). The Oligocene is the only time interval within the icehouse world that experienced atmospheric CO2 levels similar to what has been predicted for the end of this century by the IPCC 2007 report. It represents the start of the icehouse world, and was the result of climactic changes that began earlier in the Cenozoic (Zachos et al., 2001).

Figure 1-7 - Reconstructed alkenone-based CO2 records from Pagani et al. (2011). Gray line is the combined benthic foraminiferal δ18O record from Zachos et al. (2001).

1.2.1 Trend in Cenozoic Climate

The climate history of the Cenozoic can be broadly described as the transition from the greenhouse world of the Cretaceous and Early Paleogene to the icehouse world of today. This transition into the present-day climate can best be seen in the Eocene, which can be characterized as a period of long-term global cooling that lead to continental scale ice sheet growth through the

Eocene-Oligocene transition and can be seen in the combined oxygen isotope records of Zachos et al. (2008)(Figure 1-8).

Following the Eocene-Oligocene boundary, step-wise transitions during the Middle

Miocene (e.g. Holbourn et al., 2013) and an abrupt trend toward cooler temperatures following the mid-Pliocene (e.g. Seki et al., 2010) have brought us to the present-day. While an abrupt warming has been though to occur during the latest Oligocene (e.g. Zachos et al., 2001), it has

been shown that this warming is not as pronounced due to the geographic distribution of data between Miocene (lower latitude data sets) and Oligocene (higher latitude data sets) (Pekar and

Christie-Blick, 2008).

Figure 1-8 - The combined benthic foraminiferal δ18O curve based on records from Deep Sea Drilling Project and Ocean Drilling Program sites (from Zachos et al., 2008). The figure shows the 2-million-year-long Early Eocene Climatic Optimum, Mid-Eocene Climatic Optimum, and the very short-lived early Eocene hyperthermals such as the PETM (also known as Eocene Thermal Maximum 1, ETM1) and Eocene Thermal Maximum 2 (ETM2; also known as ELMO).

The first appearance of ice sheets on the Antarctic continent is a subject of debate. We know that continental scale glaciation occurred during the Eocene-Oligocene transition, but it is unclear how long a smaller or ephemeral ice sheet preceded this event. Sea-level estimates from the Western Desert of Egypt indicate the possibility of large-scale ice sheets briefly developing just prior to the Oligocene (Peters et al., 2010). Sequence stratigraphy from the South Tasman rise (Pekar et al., 2005) and sea-level estimates (Miller et al., 2005) suggest marked eustatic changes during the early to middle Eocene (54-49 Ma). Cyclostratigraphic data from the equatorial western Atlantic also suggests the presence of ice sheets during the Eocene (Miller et

al., 1998; Pekar et al., 2005, as early as 50 Ma (Westerhold and Rohl, 2009). These data compare well with stacked oxygen isotope records that show a steady increase that has been interpreted to suggest small, ephemeral ice sheets existed in Antarctica (Zachos et al., 2001; 2008). The combined global sea-level estimates from Miller et al. (2005) also suggest ephemeral ice sheets in Antarctica as far back as the Late Cretaceous, as indicated by eustatic changes of 10s of meters in less than 1 myr (Figure 1-9). However, pollen and spore climate proxies and temperature reconstructions from the Wilkes Land region of Antarctica indicate near-tropical forests existed there until at least 48 Ma, with mild, frost-free winters of about 10°C (Pross et al.,

2012).

Figure 1-9 - Global sea-level record from Miller et al., (2005).

1.2.3 Climate History of the Oligocene

The Eocene-Oligocene transition is the most researched interval of the Oligocene. It is well-documented with an abundant amount of evidence to support the timing of this transition, including stable isotope records (e.g. Zachos et al., 1996; 2001; 2008), Mg/Ca ratios (e.g. Lear et al., 2000), pCO2 records (e.g. Pagani et al., 2005), biotic turnovers (e.g. Houben et al, 2013), sea- surface temperature changes (Wade et al., 2012), apparent sea-level records from backstripped, continental margin stratigraphy (e.g. Miller et al., 2005; Pekar and Miller, 1996; Pekar et al.,

2001; Pekar et al., 2002) and others; all showing a marked change at ~34 Ma representing the transition into the icehouse world. The cause of this transition has been the subject of considerable debate. It has previously been suggested the opening of the Drake Passage and

Tasman Gateway resulted in the initiation of the Antarctic Circumpolar Current (ACC), causing the thermal isolation of Antarctica from the other continents and was responsible for the continental scale ice sheet growth (Kennett, 1977; Berggren and Hollister, 1977). Thermal isolation occurs when the development of strong zonal flow blocks pole-ward transport of heat by both oceans and atmosphere. The initiation of the ACC marks the development of such an ocean flow regime, but only occurs after Antarctica separated from Australia and South America

(Tasman and Drake passages). The initiation of the ACC has also been debated as well. It may have begun flow as early as the Late Eocene (Exon et al., 2004; Stickley et al., 2004), but did not reach full strength until the latest Oligocene (Pfuhl and McCave, 2005; Lyle et al., 2007).

Modeling studies have verified that substantial reorganization of ocean structure should be associated with the formation of a circumpolar current that spans the water column

(Toggweiler and Bjornsson, 2000; Nong et al., 2000; Sijp and England, 2004), and cooling was a likely result (Najjar et al., 2002). However, modeling of the Eocene-Oligocene transition indicated the growth of continental scale ice sheets was highly dependent on atmospheric CO2, and that concentrations had to drop below the threshold of ~750 ppm to initiate continental scale growth (DeConto and Pollard, 2003; DeConto et al., 2008). This model by DeConto and Pollard

(2003) was run with the effects of an open Drake Passage and a second time with a closed Drake

Passage. It was determined that the opening of the Drake Passage did not directly cause the continental scale glaciation, but only affected the timing whereas declining CO2 actually induced glacial growth (DeConto and Pollard, 2003).

Figure 1-10 - Apparent sea-level estimates from Pekar and Christie-Blick (2008) derived from δ18O records. The upper x-axis is the percent of the present-day EAIS (equivalent to ∼60 m ASL). The lower x-axis is ASL change, with zero representing sea level resulting from ice volume equivalent to the present-day EAIS, and with increasing values representing sea-level rise and negative numbers representing ice volume greater than the present-day EAIS volume.

Most of what is known about the Oligocene cryosphere after the E/O boundary is derived from low-to-mid latitude, deep-sea isotopic and backstripped stratigraphic records that indicate the East Antarctic Ice Sheet (EAIS) varied from 50% (e.g. Zachos et al., 2001) to 130% of modern (Kominz and Pekar, 2001; Lear et al., 2004; Pekar and Christie-Blick, 2008 [Figure 1-

10]). The ice volume maxima in oxygen isotope records (termed Oi-events [Miller et al., 1991;

Pekar and Miller, 1996]) have been interpreted as ice volume near or larger than modern (Pekar and Christie-Blick, 2008) with the “mid” Oligocene Oi2b event (~26.7 Ma) characterized as the largest ice volume (Pekar et al., 2006). It is important to recognize however, that due to age model uncertainties and resolution of δ18O records, these Oi events are not consistently seen in every oxygen isotope record, contain temporal offsets, and are more pronounced in some records and not others. Regardless, these larger ice volumes are consistent with reconstructed Oligocene topography of Antarctica that allows for a 30-40% larger ice sheet than present-day during at least the earliest Oligocene because more land surface area was above sea-level, primarily in the

West Antarctic sector (Wilson et al., 2012)(Figure 1-11).

Figure 1-11 - Maximum and minimum reconstructed topography (from Wilson et al., 2012), 1000-mcontour interval. Land area above present sea level is 13.0×106 km2 for the maximum and 12.4×106 km2 for the minimum, compared with 10.5×106 km2 for the ice-rebounded modern topography.

Sediment drilling on the Antarctic continental margin has provided some great insight to the state of the early icehouse world. Despite some of the low ice-volume estimates from oxygen isotope records, glacial-marine sediments and terrestrial palynology (e.g. Cape Roberts Project,

CIROS-1) indicate Antarctica was gradually cooling during the Oligocene and may have approached near tundra-like conditions by the Early Miocene (e.g., Raine, 1998; Raine and

Askin, 2001; Thorn, 2001; Roberts et al., 2003; Prebble et al., 2006). These data, as well as glacial sediments recovered in upper Oligocene strata from the western Ross Sea indicate

Antarctica was sufficiently cold to support the existence of ice sheets at sea level (e.g., Barrett,

1989; Cape Roberts Science Team, 1998, 1999, 2000; Naish et al., 2001). In Prydz Bay ice sheet grounding lines have been identified near the shelf edge as early as the Early Oligocene (Cooper et al., 1991; Bartek et al., 1997), which also suggests the presence of large continental ice sheets in Antarctica.

To generate a 30-40% larger than modern ice sheet in Antarctica, there would have to be significant West Antarctic glaciation (Wilson et al., 2012). The timing of West Antarctic glaciation has also been the subject of debate. The existing paradigm based on early oxygen isotope records previously suggested the build up of West Antarctic glaciation did not occur until

~6 Ma (Zachos et al., 2001). In contrast, seismic stratigraphic evidence for regional grounded ice in the eastern Ross Sea has been diversely interpreted as late Oligocene (Bartek et al., 1992), middle Miocene (Bart, 2003), and late Miocene (De Santis et al. 1999). Recent Seismic reflection data from the easternmost Ross Sea suggest erosion by glaciers from Marie Byrd Land during the Late Oligocene (Sorlien et al., 2007). Seismic facies analyses from the central Ross

Sea indicate the initial build-up of ice in West Antarctica first occurred during the Latest

Oligocene, with ice in the highland areas of Marie Byrd Land and ice caps on the elevated

(above sea-level) basement blocks in the central Ross Sea (Bart and De Santis, 2012). Glacial erosion between ca. 28 Ma and 15 Ma of dated volcanoes in the interior of northern Marie Byrd

Land as well as hydrovolcanic rocks suggest a small Oligocene ice cap there (Rocchi et al.,

2006). Most recently, the paleotopographic model from Wilson et al. (2012) suggests major West

Antarctic glaciation through the E-O boundary.

It has been determined the Oi events, which are associated with larger ice volume, have an excellent correlation with 1.2 myr obliquity nodes (Zachos et al., 2001; Pälike et al., 2006).

Sea-level estimates from backstripped sedimentary records also showed large changes at the 1.2 myr timescale (Pekar and Christie-Blick, 2008). Additionally, high-resolution stable oxygen and carbon isotopes from a low latitude marine sediment core show cyclic changes at the 1.2 myr and

400 kyr timescale (Pälike et al., 2006). These records show that ice volume was not always larger than modern, but fluctuated through the Oligocene and was primarily orbitally driven.

Despite the large ice-volume changes indicated by distal records and modeling, low- resolution CO2 estimates for the Oligocene show atmospheric levels below the modeled threshold of significant deglaciation (~840 ppm [Pollard and DeConto, 2005], revised to 750 ppm [DeConto et al., 2008]). Carbon dioxide estimates range from 300-1000 ppm (Pagani et al.,

2005; Hendericks and Pagani, 2008; Pierce et al., 2009; Pagani et al., 2011), but primarily remained below 800 ppm (Figure 1-7), similar to levels predicted for the end of this century

(IPCC, 2007). This creates a mismatch between coupled climate-ice sheet models and geochemical and stratigraphic data, resulting in uncertainty in terms of ice sheet growth and decay dynamics during the Oligocene. The lack of proximal records limits our ability to evaluate ice sheet changes at a sufficient spatial resolution, which is needed to address how the EAIS responded to climate forcing, especially under elevated CO2 conditions. This is especially true of the Oligocene, where stratigraphic intervals are still sparse (Pekar and Christie-Blick, 2008 and references therein) despite the successes of drilling projects on the Antarctic margin.

Understanding the regional extent of these Oligocene ice volume changes is critical not only for assessing how Earth’s climate system responded during the early “icehouse world”, but also for evaluating changes in future warming climates. Additionally, the timing and extent of glaciation

in West Antarctica is also uncertain. While the modeling of Wilson et al. (2012) can provide a fairly good estimate of West Antarctic glaciation, empirical data from sediment is not robust to define this during other parts of the Oligocene.

1.3 Ice-rafted debris history

One of the best ways to directly study the characteristics of the paleo-cryosphere is by analyzing ice-rafted debris (IRD) found in marine sediment cores (Connolly and Ewing; 1965).

IRD are terrigenous material deposited on the ocean floor by icebergs that calved from a glaciated coastline, and therefore can provide specific details about the coastline they are sourced from. IRD in glacial marine sediments can provide information about the distribution of icebergs, continental erosion by ice, sediment transport, and paleoclimatic conditions of polar regions (e.g. Grobe, 1987). Since IRD deposits can vary greatly in grain size, from silts and clays to gravel and cobbles, they are typically easy to identify in fine grained strata (i.e. silts and clays) that are within iceberg survival range of the Antarctic coastline. Perhaps one of the best examples of the usefulness of IRD from Antarctica was its identification at Ocean Drilling

Program (ODP) Site 748 on the Kerguelen Plateau, which lies almost 1200 km north of the

Antarctic coastline. Sediment grains larger than 250 µm were found in strata dated as the

Eocene-Oligocene boundary (Breza and Wise, 1992). The only way those coarse grained sediments could be deposited at Site 748 was by icebergs. In the northern hemisphere, the identification of episodic IRD deposits in the North Atlantic, Heinrich events (Heinrich, 1988), and their likely provenance (e.g. Hemming et al., 2002; 2004) have helped shape the way IRD can be used in paleoclimatic and cryospheric research.

A variety of techniques are used for analyzing IRD concentrations that include: estimating sand and gravel content (Connoly and Ewing, 1965; Smith et al, 1983), using the number of rock fragments in the 1-2 mm particle size for 100 g of dry sediment (Vorren et al.,

1983), gravel distribution in surface sediments in kg/m2 (Lisitzin, 1960), counting IRD grains in the 63-250 µm range (Watkins et al., 1974), counting ice-rafted quartz grains larger than 63 µm of a known sediment volume (Labeyrie, et al., 1986), using weight percentages of different size intervals (Cooke and Hays, 1982; Piper and Brisco, 1975; Kent et al., 1971; Bornhold, 1983), and using x-rays to identify IRD horizons and counting grains (Grobe, 1987).

These methods are primarily based on two different techniques: 1) using weight percent or mass accumulation rates and 2) counting of individual grains. Using weight percent of particle sizes can often bias data if there are several large particles present. Just a few grains in the 1 mm range are enough to offset the weight percent ratio of IRD to background sedimentation.

Counting of grains is time-consuming and involves a large amount of work, but ultimately proves more useful in obtaining reliable results (Grobe, 1987).

One of the biggest debates in studying IRD is what the appearance of the IRD in the marine sedimentary record means in terms of glacial development on the continent. Anderson

(1999) provides a lengthy summary of IRD studies in Antarctica, which show wildly contradicting interpretations in relationship to ice sheet growth or retreat. He concludes that studies of IRD concentrations or mass accumulation rates need to be very cautious, and would be more beneficial if they are directly compared with other proxies such as oxygen isotopes, lithology, or orbital parameters. Even recently the interpretation of IRD in terms of glacial growth or retreat has been an issue. IRD in Pliocene sediments from ODP Site 1165 in Prydz

Bay were interpreted to represent a deglacial signal, in which large iceberg armadas calved into

the Southern Ocean during warmer intervals (Williams et al., 2010). In direct contradiction,

Passchier (2011) compares the same Pliocene interval of IRD accumulation rates at ODP Site

1165 to oxygen isotope records and concludes the IRD have a correlation with increases in oxygen isotopes, interpreted as cooling intervals.

The deposition of IRD at a particular location is not just dependent on the iceberg calving during warming or cooling, but a multitude of other factors play major roles once the ice is freely floating in the ocean. Sea-surface temperatures can severely limit (warm) or enhance (cold) the survival rate of icebergs, which would control how far the IRD can be carried from the calving point (Cook et al., 2014). The surface currents also play a role in the iceberg trajectory and can help extend the range. Similarly, wind velocity and direction can control the range and location.

Generally speaking, most icebergs from East Antarctica travel west and north once in the ocean, primarily driven by the coastal counter current.

Another factor in determining to deposition of IRD is the geographic location of the study site. Sites farther from the continent will likely receive a lower abundance of IRD as compared to more proximal locations, however the closer a site is to the continent, the harder it would be to definitively resolve the sediment as IRD. The reasoning here is that basal debris entrained in moving ice streams is typically deposited near the grounding line (Powell, 1984; Alley et al.,

1986; Death et al., 2006). Sediment that is transported distances in excess of a few hundred kilometers from the Antarctic coastline must be incorporated higher in the ice column to prevent premature melting out. Icebergs sourced from outlet glaciers that extend through glacial valleys with steep bedrock walls can incorporate sediment into the ice column or on the surface of the glacier (Death et al., 2006). The production of icebergs is influenced by the thermal regime, subglacial hydrology and topography, geometry of the ice front, glacier flow speed, and calving

rate (Death et al., 2006). Additionally, icebergs can turn over if the mass above the water becomes greater than the mass below the water-level, increasing the distance the material is carried before melting out.

Though retreating ice shelves contain significantly more sand and gravel due to the lack of comminution (Passchier, 2011; Evans and Pudsey, 2002), the greatest depositional rates occur near the calving line (Domack and Harris, 1998) because the sediments will melt out due to the increased sea-surface temperature. It is likely that if icebergs were being produced during interglacial periods, the debris was being deposited closer to the continent due to increased sea- surface temperatures and reduced iceberg survival rate. The IRD signal produced at these locations would be a deglacial signal. At drill sites further away, icebergs would only reach if conditions were conducive for longer iceberg survival rates, such as during glacial periods, and the IRD layers would then represent a glacial signal.

1.4 Use of 40Ar/39Ar in Ice-rafted Debris

One of the most useful ways to analyze IRD in marine sediments is to identify their provenance by Roy et al. (2007) used 40Ar/39Ar from potassium bearing hornblende to yield the closure ages of the grains. It was also determined that biotite grains were a suitable alternative to hornblende (Pierce et al., 2011), suggesting other K-bearing minerals could have similar results.

Other techniques use the mineralogical assemblage of the IRD to represent source rock contribution (e.g. Hauptvogel and Passchier, 2012), but this is only useful for areas quite proximal to glaciated coasts that have an abundant amount of IRD and a variety of minerals distinct to source rocks.

The K/Ar dating method is one of the earliest isotope dating techniques, originally demonstrated by Aldrich and Nier (1948) and used for the first time by Smits and Genter (1950).

It is a more traditional method of analyzing K-bearing rocks (rocks with hornblende, biotite, k- feldspar, etc.), but it suffers a major disadvantage in that K and Ar isotopes need to be measured on separate aliquots and homogeneous samples are needed to generate reliable data, which means the assumption that the rock is homogeneous needs to be made. As described by Kelley

(2002), Thorbjorn Sigurgeirsson proposed the principles of Ar-Ar dating in an unpublished

Icelandic laboratory report in 1962, but never succeeded in publishing or testing the idea. The

Ar-Ar dating technique originated in the noble gas laboratory of John Reynolds in Berkeley, where Craig Merrihue and Grenville Turner recognized an 39Ar signal seen while working on neutron irradiated meteorite samples. They concluded the 39K was the result of neutron irradiation and published the idea in Merrihue and Turner (1966). The advantage of this style of dating is that K and Ar can be measured simultaneously on a heterogeneous sample, which provides greater resolution and the ability to analyze smaller samples or individual mineral grains (Kelley, 2000). This is particularly useful when dating material that is not in a large quantity (i.e. ice-rafted debris grains). The K/Ar dating method is still important however, as it serves as the dating method that standards in 40Ar/39Ar dating are compared to.

Single-step 40Ar/39Ar thermochronology involves two isotope conversions and measurements. The first is 40K, which is a radioactive isotope that naturally converts to 40Ar with a half-life of approximately 1.3 billion years. Here, 89% of the 40K is converted to 40Ca and 11% is converted to 40Ar. The second isotope measured is 39Ar, which is also radioactive and is produced by the transmutation of 39K during irradiation. The half-life of 39Ar is about 269 years, which leaves enough time to accurately measure the ratio.

For this isotope system to work, several assumptions do need to be made. First, potassium is a reactive metal commonly found in many minerals whereas argon is an inert gas that is not part of any minerals, it is therefore assumed that when a mineral forms its isotopic composition is that of the air. It is also assumed that the sample has remained a closed system since the event being dated and there has been no loss or gain of potassium or argon.

1.5 Previous Drilling Expeditions and Proximal Antarctic Records

Since 98% of Antarctica resides under ice, marine sediment cores provide the best insight into the geologic and climactic history of Antarctica. Approximately 77% of ocean sediment proximal to Antarctica is terrigenous in origin (Lisitzin, 1996) and glacial marine sediments can provide information about the distribution of icebergs, continental erosion by ice, sediment transport, and paleoclimatic conditions of polar regions (Grobe, 1987). Despite the successes of drilling projects on the Antarctic margin, stratigraphic intervals within the Oligocene are still inadequately sampled (Figure 1-12).

The earliest attempt at recovering sediment from Antarctica was the Dry Valleys Drilling

Project (DVDP) where multiple drill cores up to 325 m in depth were taken from the Dry Valleys region. This was followed a few years later by the McMurdo Sound Sediment and Tectonic

Studies (MSSTS), which drilled offshore cores in the same region as DVDP. While these projects had some Oligocene recovery, the age models are not well developed and have not been reanalyzed since the original work. Additional cores for this region include the CIROS-1 drill core which contained a ~4 million year-long hiatus starting from the Early Oligocene, and the

CRP core which has a 9 million year-long hiatus (Barrett, 1986a; Barrett, 1986b; Hambrey et al.,

1989; Barrett, 1991; Barrett et al., 2000; Cape Roberts Science Team, 2000). These cores are

located in the Ross Sea fairly close to the Transantarctic Mountain coastline where they are quite susceptible to erosional events by ice sheet expansion.

26 270

??? 28

34 ??? U1356 689B 690B 742 693B CIROS-1 Figure 1-12 - Summary of Oligocene (34-25 Ma) recovery from proximal Antarctic drill sites.

There have also been attempts at recovering Oligocene strata through ocean drill cores by the Deep Sea Drilling Project (DSDP), which became the Ocean Drilling Program (ODP), and the

Integrated Ocean Drilling Program (IODP). Unfortunately, extended hiatuses also occurred from the early Oligocene to early Miocene at ODP Sites 739 and 742 in Prydz Bay (Hambrey et al.,

1991; Cooper et al. 2004) and the drill sites on the Kerguelen Plateau are quite distal from the

Antarctica continent. DSDP Site 270 captures a portion of the Late Oligocene strata from the

Central Ross Sea, which was above sea-level until the Late Oligocene (Bart and DeSantis, 2012).

ODP Sites 689 and 690 from the Maud Rise have excellent Oligocene recovery, but they are

more distal (~400 km) and received very little terrigenous sedimentation (Barker and Kennett,

1988).

More recently, during the austral summer of 2010, IODP Expedition 318 drilled sites

U1355-1361 along the Wilkes Land Margin of Antarctica. Site U1356 is located approximately

300 km from the Antarctic coastline along the lower continental rise. Antarctic ice sheet modeling indicates that the Wilkes Land Margin was the last East Antarctic coastline to become glaciated (DeConto and Pollard, 2003) and would be the first affected by climate warming (Hill et al., 2007), but this has not yet been thoroughly investigated with proximal datasets. Provided this region is readily responsive to environmental changes, it would be an ideal place to study the effects of climate variations, especially during times with a dynamic ice sheet. Site U1356 has a nearly complete Oligocene section where recovered cores generally have moderate recovery, 60-

80% or better (Figure 1-13).

Figure 1-13 - Biomagnetostratigraphic age-depth plot for Site U1356 (from Escutia et al., 2010).

Sample sites from Antarctica were chosen for this study based upon recovery of

Oligocene strata. Site U1356, from the Wilkes Land continental rise, was selected as the starting point as it provides the most recent and complete recovery of Oligocene strata and it has become quite an important region of study due to its sensitivity to climate change. The Maud Rise Site

690B, and Ross Sea Site 270 are the other 2 sites with recovered Oligocene sediment selected for this study. Site U1356 and 270 were selected to investigate how sensitive regions of Antarctica responded during the Oligocene. Site 690 was selected for a semi-high resolution (9 kyr) benthic foraminiferal stable isotope record. Though high-resolution oxygen isotope records do exist for the Oligocene, a published record of this resolution does not exist this close to the Antarctic continent and may be able to provide additional insight into the early icehouse world. Together

these drill sites help to create a comprehensive representation of Late Oligocene glacial dynamics for a large portion of Antarctica.

1.6 Questions addressed by this study

To summarize, the outstanding questions of Antarctica during the Oligocene are presented below and are addressed in this study.

1. How did the AIS evolve through the Oligocene and was it similar to or larger than

today? δ18O records (e.g. Zachos et al., 2001) suggest a heavily deglaciated AIS during

portions of the Oligocene (<50%) while others suggest the AIS may have been

significantly reduced (~50%), but at times reached near present day size or larger during

glacial maxima, (e.g. Lear et al., 2004; Pekar et al., 2006). Most of the Late Oligocene

warming interpreted from the isotopic records reflects a divergence in isotopic records

between Southern Ocean sites and lower latitude sites (Pekar et al., 2006; Pekar and

Christie-Blick, 2008). While 1.2 myr obliquity and 400 kyr eccentricity have been found

to have the greatest influence on Oligocene ice volume (Wade and Pälike, 2004; Pälike et

al., 2006), the data are derived from a lower latitude drill site. Proximal Antarctic records

at orbital scale do not currently exist for the Oligocene. Using drill cores that have IRD

potentially sourced from sensitive regions of Antarctica (WSB and West Antarctica)

should be able to provide a representation of how these regions were changing (or not)

during the Oligocene. Additionally, a high-resolution stable isotope study from

Antarctica can place the dynamics of these regions into a broader view.

2. Is the Wilkes Land region a particularly responsive sector of East Antarctica?

Modeling studies indicate that the WSB coastline was the last region of the East Antarctic

coastline to become glaciated (DeConto and Pollard, 2003) and would be the first

coastline affected in a warming climate (Hill et al., 2007). The Wilkes Land basin

currently resides >500 m below sea-level (Wilson et al., 2012) and therefore should be

sensitive to changes. If this region is the most responsive area of East Antarctica, it is

important to understand how the ice sheet responded under climactic conditions similar to

what is predicted for the end of this century. The WSB is quite large, and drains a

sizeable portion of East Antarctica. Changes within this drainage region could have

profound effects or feedbacks for a large portion of the continent.

3. Was a West Antarctic Ice Sheet (WAIS) present during the Oligocene and what was

its influence on the flow of ice into the Ross Sea? While modeling (Wilson et al.,

2012), seismic interpretations (Sorlien et al., 2007), and on-land geology from Marie

Byrd Land (Rocchi et al., 2006) suggest the presence of a large WAIS during the

Oligocene, geochemical evidence from the glaciomarine sediment record does not exist.

Sediment from DSDP Site 270 will be used to investigate the potential for West Antarctic

glaciation for the Late Oligocene.

4. How prominent was the Late Oligocene warming in Antarctica? The stable isotope

stack (Zachos et al., 2001; Zachos et al., 2008) shows a large warming during the Late

Oligocene. However, Pekar and Christie-Blick (2008) show that most of this warming is

an artifact between the geographic distribution of the study sites and suggest the Late

Oligocene warming was not as prominent as the stacked data suggests. Continuous

isotopic records of the Late Oligocene show a general trend toward lighter oxygen

isotope values, suggestive of warming and decreased ice volume for the Late Oligocene.

1.7 Significance

In summary, the Oligocene represents a modern or near-future analog to the state of the

Antarctic cryosphere. Though the geographic positioning of the continents was slightly different than today, and the ocean current structure was in its early stages of modern development, the

Oligocene atmospheric CO2 levels were similar to what is expected for the end of this century

(500-800 ppm). Therefore we can look at the characteristics of ice sheet behavior during the

Oligocene to inform our understanding of the effects of elevated CO2 on the ice sheets of

Antarctica in the future.

The IRD and geochemical data gathered from these drill sites will be the first attempt at creating a spatial and temporal picture of ice sheet dynamics around the Antarctic continent during the Late Oligocene. This study uses the geochemical provenance of IRD to trace the erosional history of 2 important regions in Antarctica; the Wilkes Sub-glacial Basin and West

Antarctica. Additionally, the creation of a new semi-high-resolution stable isotope record proximal to Antarctica allows us to better assess ice volume and water mass changes because this record was taken directly from the Southern Ocean (~400 km from the continent). Together these Late Oligocene data will provide new insight into the state of the Late Oligocene icehouse world.

Chapter 2. Evidence for large East Antarctic Ice Sheet in the Wilkes

Subglacial Basin during the Oligocene

Abstract

Ice-rafted debris concentrations and 40Ar/39Ar thermochronology from Integrated Ocean

Drilling Program (IODP) Site U1356 reveal an ice sheet was consistently present in the Wilkes subglacial basin (WSB) and Adélie regions of Antarctica during the Late Oligocene, even though atmospheric CO2 levels were elevated. According to climate modeling, the WSB is one of the most sensitive regions of East Antarctica and is thought to be the first regions that experiences ice sheet retreat during warming. Concentrations of ice-rafted debris suggest iceberg delivery to

Site U1356 is paced by 405-kry eccentricity cycles, with IRD being delivered to the drill site during cooler times. A comparison with ice-volume estimates suggests IRD are delivered to Site

U1356 when ice volume is at least 80 % of modern. The lack of IRD during warmer periods suggests a retreat of the ice sheet from the coastline or a change in oceanographic conditions that prevent many icebergs from reaching Site U1356. Debris flow layers have a good correlation with oxygen isotope maxima and have been interpreted to represent an ice sheet larger than modern. The primary provenance of the IRD is from the Mertz Shear Zone region, which separates the Adélie craton from the WSB. Ages indicative of the Ross Orogeny comprise a secondary population that has been interpreted as being sourced from the Ninnis glacier or outlet glaciers further east. Additionally Grenville age grains comprise another population that can only be explained as being sourced from areas further inland that are currently obscured by the ice sheet. Together the concentrations and 40Ar/39Ar provenance of the IRD show a large ice sheet

consistently eroded the MSZ and Adélie craton and that the WSB was more dynamic, and may not have had a large ice sheet consistently present.

2.1 Introduction

Climate models suggest the Wilkes Land region, especially the Wilkes Sub-glacial Basin

(WSB), would have been the last region of East Antarctica to become glaciated (DeConto and

Pollard, 2003) and is predicted to be the first region to experience ice sheet retreat during warming (e.g. Hill et al., 2007; Pollard et al., 2015). Additionally, it has been shown that a retreat of the ice sheet in this region will be initiated once ice volume drops to ~80 % of modern

(DeConto and Pollard, 2003; Pollard and DeConto, 2005; DeConto et al., 2012; Hill et al., 2007;

Pollard et al., 2015). Sediment geochemical provenance data during the Pliocene warming indicates a significant retreat of the ice sheet margin inland of the current location (Cook et al.,

2013), which supports the sensitivity observed in the models (e.g. Pollard et al., 2015). During the Oligocene the dynamics of the ice sheet in this region are unclear because proximal records for the WSB do not exist. This study presents the first proximal evidence for the glacial dynamics of the Oligocene in the WSB region.

2.1.1 The Wilkes Sub-glacial Basin

The WSB is a large, low-lying sub-glacial basin that extends from the center of the East

Antarctic continent toward the George V/Adélie coastline. The sensitivity of this region is due to the fact that most of the WSB lies >500 m below sea-level (Mercer, 1978; Drewry, 1983) (Figure

2-1). Additionally, many sub-glacial lakes are present (Siegert et al., 2005a) and have a

hydrologic connection with the margin of the ice sheet (Siegert et al., 2005a, 2005b; Wingham et al., 2006; Jordan et al., 2010). These sub-glacial lakes play an important role in ice sheet dynamics (Bell, 2008) and modeling has shown them to be a key factor in ice stream surges because of the overlying pressure created as the ice sheet covers these lakes (Erlingsson, 1994;

Alley et al., 2006).

150˚

180˚ 120˚

TAM

AC m WSB 2000 1000 NVL AL GVL U1357 0 U1360 Site U1356 60˚ −1000 elevation ng mg U1358 − −70˚ U1359 U1355 −2000 U1361

0 500

150˚

180˚ Figure 2-1 - Drilling location map for IODP Expedition 318 with BEDMAP2 topography (Fretwell et al., 2013). Site U1356 is denoted by the black star, other sites are black circles. WSB=Wilkes Subglacial Basin, AC=Adélie Craton, AL=Adélie Land, GVL=George V Land, NVL=Northern Victoria Land, TAM=Transantarctic Mountains, mg=Mertz Glacier, ng=Ninnis Glacier, dashed line is the Mertz Shear Zone.

Two outlet glaciers currently drain most of the WSB, The Mertz Glacier and the Ninnis

Glacier (Figure 2-1), and terminate along the Adélie and George V Coasts of the eastern Wilkes

Land margin. These outlet glaciers extend seaward as ice tongues and play an important role in ice sheet drainage and sediment delivery to the ocean (Anderson et al., 1980; Drewry and

Cooper, 1981). Drainage velocities of outlet glaciers have been shown to range from >0.5 km/y

to ~3.7 km/yr (Lindstrom and Tyler, 1984; MacDonald et al., 1989), whereas drainage in the areas between outlet glaciers may range from a few meters to tens of meters every year

(Anderson, 1999). Recent satellite data shows that the Mertz and Ninnis glaciers drain at well over 1 km/yr (Rignot et al., 2011) and that the WSB is one of the areas in East Antarctica that is currently losing ice mass (Shepherd et al., 2012).

2.1.2 Lithostratigraphy of the Oligocene section at Site U1356

IODP Site U1356 was part of expedition 318 to the Wilkes and Adélie coastal regions of

Antarctica during the Austral Summer of 2010 (Figure 2-1). The goals of this cruise included investigating the timing and nature of the initial onset of glaciation at the Wilkes Land margin

(Eocene-Oligocene boundary), and to obtain records of large fluctuations of ice sheet size and regimes (i.e. wet vs. cold) and climate variability during the Late Cenozoic.

Site U1356 is located on the lower continental rise of the Wilkes Land margin (63.31°S,

135.99°E, between Antarctica and Australia) in 3992 m water depth and is about 300 km north of the coastline, the most distal site drilled during this expedition (Figure 2-1). The strata recovered at Site U1356 range from recent to Early Eocene in age. Two large hiatuses are present at Site U1356: an 8 myr hiatus exists from ~24-16 Ma (Late Oligocene to Early

Miocene) and a 14 myr hiatus exists from ~48-34 Ma (Early Eocene-Early Oligocene). The core recovery was generally poor throughout with the exception of Oligocene strata that have relatively good to excellent recovery, generally 60-80% or better (Figure 2-2). This study utilizes cores 68-95, which are the best-recovered Oligocene sections.

Unit V - interbedded bioturbated claystones and silt-laminated claystones (some limestone beds). Lacks gravel-sized clasts.

Unit VI - contorted beds of silty claystones with clasts.

Unit VII - interbedded bioturbated claystones and silt-laminated claystones (some limestone beds). Lacks gravel-sized clasts.

Unit VIII - interbedded bioturbated claystones and silt-laminated claystones (some limestone beds and convoluted intervals).

Unit IX - contorted and convoluted claystones.

Unit X - contorted claystones interbedded with massive and stratified sandstones and diamic- tites

Figure 2-2 - Recovery and lithology from IODP Site U1356. Unit descriptions are from the initial reports.

The dominant lithofacies at Site U1356 are bioturbated claystone and calcareous claystone (Expedition 318 Scientists, 2011). The Oligocene strata studied here include lithostratigraphic units V-IX, include cores 68-95 (640-895 mbsf) and have been described and interpreted by the shipboard scientific party (Expedition 318 Scientists, 2011). Below are summaries of each Oligocene lithostratigraphic unit studied (summarized from Expedition 318

Scientists, 2011):

Unit V is comprised of the interval 63R-1, 41 cm, through 73R-4, 72 cm (593.8–694.4 mbsf) and was estimated to be Late Oligocene in age. These strata consist of light greenish gray, strongly bioturbated claystones and micritic limestones interbedded with dark brown, laminated claystones. Minor cross-laminated sandstone interbeds are present as well. Unit V was

previously interpreted to represent pelagic and hemipelagic sedimentation. According to the initial reports, the absence of gravel and presence of rare dispersed sand grains indicate that ice rafting was not an important sedimentary process. While the amount of sand (wt%, described in the results section) in this interval is small, I suspect these grains were primarily derived by ice- rafting. The laminated claystones and ripple cross-laminated sandstones may indicate variations in bottom current strength and fine-grained terrigenous supply. The sharp-based facies successions indicate that sedimentation was strongly cyclic. The recovery of Unit V from core

63-68 was poor (<50 %), but cores 68-73 were much better, 57-84 % and averaging 76 % recovery. This was one of the best-recovered intervals at U1356.

Unit VI is the interval from 73R-4, 72 cm, through 76R-4, 134 cm (694.4–723.5 mbsf) and has been given an age of early Late Oligocene. This unit is primarily contorted beds of mudstones with dispersed (<1%) to common (1%–5%) clasts of granite, diorite and others. These strata are interbedded with thinner (~1–2 m thick) intervals of bioturbated claystones and carbonate-bearing claystones. Beds of mudstones with clasts may represent the distal ends of debris flows affecting an environment characterized by hemipelagic sedimentation, represented by the bioturbated claystones and carbonate-bearing claystones. Additionally the contorted beds could be indicative of slumping of the sediment. The abundant mudstone intraclasts are interpreted as rip-up clasts derived from the underlying bioturbated claystone lithologies. Core

74 only had 3% recovery, while the others ranged from 56-78%.

Unit VII includes the strata from 76R-4, 134 cm, through 82R-6, 37 cm (723.5–782.7 mbsf) and has been given an age of middle Oligocene. Unit VII consists of bioturbated claystones and calcareous claystones interbedded with laminated to massive claystones. The lithologies are similar to those observed in Unit V. Silt laminae are present as either pinstripe

(submillimeter scale) or parallel laminations and ripple cross-laminations up to 10 mm thick.

Gravel-sized clasts are absent, again suggesting ice rafting was not an important sedimentary process. The bioturbated and calcareous claystones and limestones represent pelagic and hemipelagic sedimentation. The pinstripe laminated and ripple cross-laminated claystones probably indicate variations in bottom current strength and fine-grained terrigenous supply.

Sediment recovery within this unit was fairly poor, with 3 cores only recovering 6% or less.

The interval from 82R-6, 37 cm, through 93R-3, 61 cm (782.7–879.7 mbsf) composes

Unit VIII. It has an age of Early Oligocene. Unit VIII consists of interbedded bioturbated claystones, calcareous claystones, laminated claystones, and massive structureless claystones.

The bioturbated claystones are characterized by Zoophycos ichnofacies. Ripple cross-lamination in laminated silty claystones is common. Locally, decimeter-thick sand and granule-rich interbeds are present. Contorted or inclined bedding, intraformational clasts (Fig. F12 ), and synsedimentary microfaults are common throughout this unit. The bioturbated claystones and laminated claystones indicate hemipelagic deposition. Ice rafting could be responsible for the gravel component in the sediments. Contorted bedding results from mass movement. The convolute bedding and possible repetition of stratigraphy with preservation of undeformed beds between deformed intervals suggests that submarine slides and slumps occurred periodically during deposition of the unit. The recovery of Unit VIII was mixed, with sections of good recovery of 79-97% and sections of poorer recovery (<30%).

Unit IX is comprised of strata from 93R-3, 61 cm, through 95RCC, 0 cm (879.7–895.5 mbsf) and has an age of earliest Early Oligocene. It is composed of light green laminated claystones, dark green massive (structureless) claystones, and contorted massive and stratified sandstones (closer to base of the Unit). It is distinguished from Unit VIII by a color change to

distinct green lithologies, the appearance of massive (structureless) claystones, and the presence of sandstone interbeds. Locally, beds are composed of convoluted mudstones and sandstones with vertical stratification and abundant intraformational mudstone clasts. Repetition of a conspicuous green mudstone lithology on a meter scale is also observed. Therefore, repetition of stratigraphy is suspected. The claystones probably indicate hemipelagic sedimentation, whereas the sandstones indicate stronger winnowing by currents or gravity flow. Repetition of stratigraphy and convolute bedding suggests that submarine slides and slumps were a dominant type of mass movement. Alternatively, proglacial or subglacial deformation of strata may have occurred. I suspect the latter interpretations better explain the sedimentary processes occurring here, especially with new paleotopographic models indicating a larger than modern ice sheet in the earliest Oligocene (Wilson et al., 2012). The recovery of this unit was only little better than

50%.

2.1.3 Regional geology of the WSB region

The WSB is bordered on the west by the Mertz Shear Zone (MSZ), which separates it from the Adélie craton, and on the east by the Transantarctic Mountains (TAM) (Figure 2-3).

The Satellite data of Rignot et al. (2012) shows that the Mertz Glacier primarily drains the

Adélie craton while the Ninnis glacier primarily drains the WSB region (Figure 2-1). The coastal geology of the WSB region is quite diverse, with three distinct sectors.

Figure 2-3 - Bedrock geology of the greater WSB region (from Cook, 2013, based on Bushnell and Craddock, 1970). Subglacial topography is also indicated (from Fretwell et al., 2013); gray areas are below sea-level. Light green shading offshore represents the approximate transport region of modern icebergs (Antarctic Iceberg Tracking Database (1978–2012); available at http://www.scp.byu.edu/data/iceberg/database1.html) with direction of movement shown with small grey arrows. Larger dashed arrows indicate the approximate location of major flows of Antarctic Bottom Water (from Orsi et al. 1999). AL=Adélie Land, NVL=Northern Victoria Land, SVL=Southern Victoria Land, TAM=Transantarctic Mountains, NG=Ninnis Glacier, MG=Mertz Glacier, MSZ=Mertz Shear Zone; CIS=Cook Ice Shelf; RI=Ross Island.

The first area is the Adélie craton, which is among the oldest rocks found in Antarctica and is thought to be the nucleus of the Antarctic continent (Boger, 2010). Its present day appearance and position formed during an extensional tectonic event between Antarctica and

Australia in the mid-Jurassic and Cretaceous, between 160 and 125 million years ago (Stagg and

Wilcox, 1992) that later resulted in the rifting of the two continents as Australia moved

northward at least 95 million years ago (Boger, 2010; Veevers and Eittreim, 1988). The bedrock here is primarily composed of metasedimentary rocks (~1700 Ma) and paragneiss and granitoids

(>2400 Ma) (Peucat et al., 2002; Goodge and Fanning, 2010). Adélie Land is located between

136°E and 142°E. To the West lies the boundary with the greater Wilkes Land and the east with

George V Land.

The second sector is the Mertz Shear Zone (MSZ), which separates the Precambrian

Adélie cratonic terrain from the Paleozoic rocks of the Ross Orogeny and is exposed in all of the easternmost outcrops of the East Antarctic shield (Talarico and Kleinschmidt, 2003). Argon dating of biotite grains from mylonitic rocks indicate a minimum age of ductile deformation of the MSZ at 1502 ± 9 Ma (Di Vincenzo et al., 2007). The terrain between Commonwealth Bay and Mertz Glacier comprises a composite domain belonging to the East Antarctic Craton which includes (Talarico and Kleinschmidt, 2003): the low-grade Cape Hunter Phyllite; an amphibolite- facies granitic orthogneiss at Cape Denison; amphibolites and enderbitic orthogneisses at

Madigan Nunatak; garnet bearing gneisses with minor garnet-bearing amphibolites and mafic granulites at Garnet Point and Cape Pigeon and; enderbitic orthogneiss, minor felsic granulites, subordinate amphibolite lenses and variably metamorphosed mafic dykes at Correll Nunatak,

Aurora Peak and Mt. Murchison. The eastern termination of this area, corresponding to the western side of the Mertz Glacier is marked by the Mertz Shear Zone (Talarico and

Kleinschmidt, 2003 ).

The third area is George V Land, which extends from 142°E to153°E and includes the

Mertz and Ninnis glaciers. Though the MSZ is within George V Land, its geology is distinct from the rest of the region. The coastal exposures between the Mertz and Ninnis glaciers consist mainly of various granitoids (sometimes slightly foliated) and subordinate low-grade

metamorphic rocks. Cape Webb represents the easternmost exposure of basement rocks in the region. Other outcrops along the eastern side of Ninnis Glacier (Organ Pipe Cliffs, Horn Bluff) consist of thick sills of Jurassic Ferrar Dolerite (Hanemann and Viereck-G¨otte, 2004 ); a thin layer of Permo–Triassic Beacon Sandstone is exposed at the bottom of the prominent cliff at

Horn Bluff (Klimov and Solove’v, 1958 ).

2.1.4 Previous studies in the WSB region

The most recent work done in the WSB was the geochemical provenance of fine-grained sediment from Cook et al. (2013). They used sediment from IODP Site U1361, which lies just northeast of Mertz Glacier terminus, to investigate any provenance changes during the Pliocene warm period. Using Sr and Nd isotopes, they determined the detrital material exhibited alternating sources during warmer and cooler periods. The isotopic signatures changed from a

Ross Orogeny signal to a FLIP signal during warming. Rocks of the FLIP have been interpreted by airborne geophysics to have larger exposures further inland, underneath the ice sheet

(Ferraccioli et al., 2009), which lead Cook et al. (2013) to interpret this signal as a reflection of the ice sheet retreating significantly inland.

150˚

80˚ 70˚ 120˚ − − 2000 Regional 40Ar/39Ar Thermochronology (or K/Ar) 1500 Holocene IRD ages

On-land hornblende ages 1000 On-land biotite ages −80˚ Inferred FLIP Group 500 Thermochronological age (Ma) age Thermochronological

MSZ 0 0

120˚

ELT37−16

ELT37−−1313 ˚ NBP 01−01 JPC11 60 − ng mg Site U1356 DF80−34 ELT37−09 2000 −70˚ DF80−35 ELT37−06 DF80−20 DF79−47 1000

DSDP 269 0

−1000 (m) Elevation NBP9802−3H −2000 0 500

150˚ −60˚

Figure 2-4 - Map of thermochronologic data from offshore sediment and on-land 40Ar/39Ar ages (Roy et al., 2007; Pierce et al., 2012; Di Vincenzo et al., 2007; Duclaux et al., 2007; Goodge, 2007; Phillips et al., 2007; Wilson et al., 2007). Gray area indicates inferred location of FLIP group (from Ferraccioli et al., 2009). mg=Mertz Glacier, ng=Ninnis Glacier, MSZ=Mertz Shear Zone.

Other published work from the region includes the construction of detailed sediment provenance signatures in proximal, piston-cored sediment where ages are not well constrained

(i.e. Roy et al., 2007; Pierce et al., 2011). These studies used 40Ar/39Ar thermochronology of detrital hornblende grains (Roy et al., 2007; Pierce et al., 2011) and biotite (Pierce et al., 2011) to interpret sediment provenance for coastal regions in Antarctica. (Figure 2-4) The study of Roy et al. (2007) examined the Nd isotope composition, coupled with the 40Ar/39Ar hornblende ages of 44

bulk sediments from around the Antarctic continent, laying the foundation for future marine sediment provenance work in Antarctica. The study of Pierce et al. (2011) focused on the portion of East Antarctica from Northern Victoria Land to Prydz Bay, which includes the WSB/Adélie

Craton regions in this study. Additionally, U-Pb work on rock clasts from dredge samples and detrital zircon from glaciomarine sediments (Goodge and Fanning, 2010) provided insight into the rocks of the WSB region, filling in gaps left by the limitations of Ar studies (i.e. thermal

“resetting” temperatures of minerals). These studies worked to characterize the geology of the near-by coastal regions and have now become the standard in geochronology for detrital mineral studies for this portion of East Antarctica.

Even though ice volume fluctuated during the Oligocene, uncertainty still exists in the size and extent of the ice sheet (see section 1.2.3). Seismic surveys from the area suggest the ice sheet did not extend far onto the continental shelf during the Oligocene, as indicated by fairly well stratified sediments (Donda et al., 2003; Escutia et al., 2005; Donda et al., 2008). Despite the stratified nature of the sediments, there are a few erosional surfaces, which may indicate glacial growth onto the continental shelf (Escutia et al., 2005). These surfaces may represent the

Oligocene glacial maximum at ~26.7 Ma (Oi2b) when δ18O reached 3.6‰ and the Oi1 glacial event, which are likely the only times ice advanced onto the continental shelf.

2.2 Methods

2.2.1 Sampling

For Site U1356, 151 core samples were originally taken between 25.3 and 33.8 Ma (cores

68-95, 750-895 mbsf), resulting in a moderate resolution of 25-50 kyr (~1 sample per meter). As the Late Oligocene sections had among the best recovery for the entire Oligocene (Cores 67-73,

70-80% recovery), a higher resolution sample set (additional 100 samples) was taken between

25.4 and 27.0 Ma at a 10-17 kyr resolution. These resolutions were selected to provide insight into Oligocene glacial dynamics on long-term orbital scales; however, this does alias the short- term processional and obliquity signals. Ages of the sediment sample were linearly interpolated between six paleomagnetic tie points from Tauxe et al. (2012) (Table 2-1, Figure 2-5), however, uncertainty exists in these ages owing to incomplete recovery, partial paleomagnetic overprints, and moderate confidence in the correlation of these reversals to the Geomagnetic Polarity Time

Scale. Therefore, caution must be taken when making quantitative correlations to other data sets or timescales.

Table 2-1 – Paleomagnetic chrons from IODP Site U1356 for the Oligocene (Tauxe et al., 2012). Top Average Bottom Chron Age Depth Depth Depth C6Cn. 3n (y) 23.249 444.72 444.72 444.72 C7An (y) 24.915 596.94 597.115 597.29 C7An (o) 25.091 604.3 604.325 604.35 C8n. 1n (o) 25.444 643.1 643.375 643.65 C8n. 2n (y) 25.492 652.55 652.575 652.6 C8n. 2n (o) 26.154 678.06 678.98 679.9 C9n (y) 26.714 694.57 701.66 708.75 C9n (o) 27.826 725.06 725.085 725.11 C11n. 2n (o) 30.217 782.65 782.675 782.7 C13n (y) 33.266 877.77 878 878.23

The higher resolution sampling set also includes the Oligocene oxygen isotope maximum and has been interpreted as the period of greatest ice volume during the Oligocene. This isotope maximum is further investigated in this study.

640 Paleomagnetic Tie Points FO Foraminifera 660 FO Dinocysts 680 LO Nannofossils

700 Rectuvigerina stonei

720

740

760

780 Depth (mbsf) 800

820

840

860 Reticulofenestra umbilicus | Isthmolithus recurvus 880 Malvinia escutiana | Stoveracysta kakanuiensis 900 26 28 30 32 34 Age (Ma) Figure 2-5 - Oligocene age model from IODP Site U1356. Data is from Tauxe et al. (2012). FO=first occurrence, LO=last occurrence.

2.2.2 Sediment Separation

Approximately 10 grams of dry sediment were taken from each sample and soaked in a sodium metaphosphate (NaPO3)6 solution (5.5 g of sodium metaphosphate to 1 L water) for 24 hours to promote disaggregation. If sediment proved difficult to disaggregate it was first gently crushed with a mortar and pestle, avoiding any grinding motion so that individual grains would

not break. The sample was then allowed to soak for another 24 hours in a sodium metaphosphate solution. Samples that remained difficult were placed in a solution of sodium pyrophosphate

(Na4P2O7) and slowly brought to a boil. Then they were placed in an ultrasonic bath while in a sodium metaphosphate solution. Following disaggregation the sediment was washed with water and sieved to remove the <63 µm fraction. The remaining sediment was then dried in an oven overnight at 60°C.

2.2.3 IRD concentrations

Core samples were then dry sieved to obtain the >150 µm fraction. This study used a minimum size constraint of 150 µm for a grain to be classified as IRD. This size constraint was chosen because it is the minimum grain size required to perform 40Ar/39Ar analyses on the hornblende grains and has been used in several recent IRD studies in Antarctica (e.g. Williams et al., 2010; Pierce et al., 2011). It also ensures that these grains could only be deposited at the drill site by ice rafting because the U1356 is far enough offshore (~300 km) that grains of this size would not reach the drill site under alternative, terrigenous sedimentation (with the exception of gravity flows). Grains were counted and identified using a Zeiss Stemi 2000-C binocular microscope.

2.2.4 40Ar/39Ar Thermochronology

This study employed the use of single-step fusion 40Ar/39Ar thermochronology to trace the sediment provenance of the IRD. This method has several advantages, including the release of 40Ar, or “resetting”, as the temperature rises above the closure temperature of the minerals during a metamorphic event. This temperature varies for each mineral. For example, hornblende has a closure temperature of ~550°C and biotite ~300°C. The resetting of the 40Ar means the

ages obtained during 40Ar/39Ar thermochronology are that of cooling from an igneous or metamorphic event through the closure temperature of the mineral analyzed which can vary between different minerals depending on the cooling rate. This allows a better characterization of source rocks as compared to other techniques such as Sm/Nd, which is resistant to resetting and provides an age of crustal emplacement rather than metamorphic events. Additionally, this method has a significant advantage over K/Ar dating because both Ar isotopes can be measured simultaneously rather than on separate aliquots.

Hornblende and biotite grains >150 µm and in some cases biotite grains between 63 and

150 µm were hand-picked using a binocular microscope for 40Ar/39Ar thermochronology to trace the terrigenous provenance of the IRD. While magnetic separation is a very reliable way of isolating these minerals from other grains in the sediment, the number of grains above 150 µm was fairly low (generally less than 1000), making hand-picking a quicker alternative.

Additionally, most of the grains were white or clear quartz, which made for easier task of finding the hornblende and biotite grains. Both hornblende and biotite were selected in an attempt to minimize any lithological bias generated by using only one mineral.

Grains were loaded into metal disks and sent to the U.S. Geological Survey (USGS)

TRIGA reactor in Denver, CO where they were co-irradiated so that 39Ar was produced from 39K during irradiation. The grains were co-irradiated for 8 hours and were given a minimum of 2 weeks to “cool down” before being sent back for analysis. Once received, grains were individually loaded into single spots on a disk for 40Ar/39Ar analysis. Ages for the grains were obtained using CO2 laser fusion attached to a VG 5400 Mass Spectrometer at the Lamont-

Doherty Earth Observatory (L-DEO) argon geochronology lab (AGES: Argon Geochronology for the Earth Sciences). The CO2 laser heats the grains, releasing the gasses to be analyzed.

Gasses released from the heating of samples are scrubbed of reactive gases such as H2 CO2, CO and N2 by exposure to Zr-Al sintered metal alloy getters. The remaining inert gasses, principally

Ar, are then admitted to the mass spectrometer. The Ar-isotopic ratios are determined using automated data collection software (MassSpec). J values used to correct for neutron flux during irradiation were calculated using the co-irradiated Mmhb-1 hornblende standard (525 Ma

[Samson and Alexander, 1987]). A series of blank and air samples were also measured to ensure precision and accuracy.

2.3 Results

The sedimentary record at Site U1356 shows a highly variable flux of sand-sized grains being delivered to the lower continental rise, with weight percentages ranging mainly between 0 and 18%, with one data point from 895.24 mbsf at ~75% (Figure 2-6). The IRD are mainly comprised of 90% or more of quartz, with minor amounts of hornblende and biotite, and even less of feldspars. The grains range from sub-rounded to primarily angular. The quartz grains were clear, white, and in some instances pink. Iron staining of the quartz grains did not appear to be present. Hornblende grains were easily recognizable based on cleavage and were black in color. Biotite grains that were identified consisted of black and brown colors. A range of biology was also encountered, including foraminifera, diatoms, and fish teeth. The foraminifers were not abundant, and those that were present were heavily altered and likely not suitable for stable isotopic studies. The few diatoms that were present did appear to be well preserved, but they were rare. No discernible pattern could be found in the abundances and diversity of the biology.

ab c d e C7An 25 C8n.1n 650 C8n.2n 26

27 700 C9n

28 C10n.1n C10n.2n 750

) 29

s b

IRD Gravity Flow A

ge (Ma) ge (

C11n.1n

t p

e C11n.2n 30 D 800 C12n 31

32 850

85 C13n 900 0510 15 34 RSand (wt%) P Chrons Clay/Silt Sandy mud Limestone Nannofossil ooze Sand

Figure 2-6 - Sedimentology and age model of IODP Site U1356. Depth is in meters below sea floor (mbsf). A: Core recovery for Site U1356 (R). Black intervals indicate recovered strata, white indicates intervals of no recovery. B: Lithology (Expedition 318 Scientists). C: Weight percent of sediment >63 µm. Gaps in the record reflect gaps in core recovery. Dashed line separates IRD from gravity flows. D: Polarity (P) (Tauxe et al., 2012). Black intervals indicate normal polarity, white reverse, grey reflects no data, blue within 10° of the horizontal. E: Polarity interpretations and ages (Tauxe et al., 2012).

Most of the sand-bearing intervals were interpreted as IRD, however grains from several intervals were interpreted to be the result of downslope processes (Expedition 318 Scientists,

2011) (Figure 2-6, sandy mud intervals) and are represented in our record by sand abundances rising above ~4% and a large increase in the number of grains >150 µm to over 1,000 grains per gram. Seismic data over the drill site (Expedition 318 Scientists, 2011) are not of high enough

resolution to identify sediment waves or other contourite characteristics at the scale of these flow layers (less than a few meters thick). Seismic profiles south and west of Site U1356 however, show that sediment re-working did not occur until the Miocene and that Oligocene sediments are fairly well stratified on the shelf (Donda et al., 2003; Donda et al., 2008). These gravity flows are interpreted as distal turbidites, which have been identified on the continental rise east of Site

U1356 (Escutia et al., 2005).

The samples interpreted to be the result of iceberg detritus are based on IRD bearing layers with less than 300 grains per gram (less than 4 wt% sand) with a poorly sorted sand fraction. The flow layers were greater than 500 grains per gram (5-15 wt% sand) and had a somewhat better sorted sand fraction compared to IRD layers. For debris flow layers, the >150

µm fraction contained many thousands of grains, making them fairly distinct, but difficult to count. The interval from ~26.5 to 28.0 Ma had very poor recovery and is where most of the uncertainty in the age model is located. This is also where most of the debris flows in the IRD record are occurring, therefore all of the samples from this interval have been interpreted to be related to these debris flows. For these layers grains were spread out on a tray that was separated into 1 cm2 boxes. A single box was counted fully and multiplied by the total number of boxes to achieve the estimated total number of grains.

There was no indication of a lithological bias in the presence of IRD grains. These layers were present in both light and dark green-grey silty clays as well as more carbonate bearing layers. This would suggest that the deposition of IRD at Site U1356 was independent to changes in the fine-grained sediments. Geochemical studies of the fine-grained material in IRD-bearing and non-bearing layers is needed to interpret this further.

A total of 165 grains were measured for 40Ar/39Ar analysis from the Late Oligocene sediment at Site U1356 from ~27.7-25.4 Ma. Of the grains measured 72 were hornblendes >150

µm, 44 were biotites >150 µm, and 49 were biotites in the 63-150 µm range. Grains were selected from this interval for several reasons: 1) The Late Oligocene contains the Oi2a and Oi2b isotopic events, with the latter representing the Oligocene oxygen isotope maximum and interpreted large ice volume, 2) This Late Oligocene interval is one of the best recovered sections at Site U1356, 3) Though there were IRD layers in the Early Oligocene section, they did not contain grains suitable for 40Ar/39Ar analysis.

The layers selected for Ar analysis included intervals with high IRD abundance (6 samples, 106 grains), low IRD abundance (4 samples, 24 grains), and debris flow sections (7 samples, 36 grains). Biotite grains from both size fractions (63-150 µm and >150 µm) were picked whenever possible to investigate any potential grain size bias in the provenance data.

The summary of all hornblende and biotite grains measured from Site U1356 is presented in Figure 2-7. The grain ages ranged from ~340 Ma to 2000 Ma. The most dominant age population of the grains was clustered around 1400-1600 Ma, with a smaller cluster around 400-

600 Ma. The errors on the grain ages were generally less than ±100 Ma and most were less than

±50 Ma. Grains with an error of greater than ±100 Ma were not included in the interpretations.

Due to the large Proterozoic population of grains, this error would not have an effect on the interpretations. The Ca/K ratio was used to confirm the identity of the hornblende and biotite grains; hornblendes typically have a Ca/K between 1 and 60 while biotites usually have a ratio significantly less than 1.

25 All Ar ages measured n=155 20

10 number

0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Age (Ma) Figure 2-7 - Summary of all 40Ar/39Ar ages measured in the Late Oligocene sediment at IODP Site U1356. Samples used are denoted by the black triangles in Figure 2-9.

2.4 Discussion

The occurrence of IRD at Site U1356 is a direct indicator of the existence of ice along the coastline in the vicinity of the drill site and the 40Ar/39Ar ages of the grains can be used to constrain the likely source areas.

2.4.1 Implications for ice volume during the Oligocene

The IRD record from Site U1356 indicates the EAIS was at least periodically reaching the WSB coast, with IRD being deposited in clusters, especially in the upper section (Figure 2-

8). This suggests there is some controlling mechanism on the delivery of IRD to Site U1356.

While limitations in the age model prevent a quantitative correlation of the IRD to parameters such as orbital forcing or oxygen isotope records, the IRD concentrations can still inform us about the dynamics of the WSB during the Oligocene.

Qualitatively, the deposition of IRD in the upper interval (26.6-25.4 Ma) occurs during low variation in insolation and low 405-kyr eccentricity, which can be interpreted as less seasonality and a relatively colder interval (Figure. 2-9). Higher IRD concentrations seem to correlate with low-resolution sea-level estimates of Pekar and Christie-Blick (2008), representing an EAIS of at least 80 % of present-day size. In contrast, lower concentrations (<1 grain/gram) seem to occur during insolation maxima and relate to ice volume below 80%. The lower interval

(32.4-30.6 Ma) is not as clear (Figure 2-10), which may be due to the lower sampling resolution of this section and limitations of the age model.

Antarctic ice sheet modeling based on CO2 changes (DeConto and Pollard, 2003; Pollard and DeConto, 2005), modeling from the Pleistocene (DeConto et al., 2012), and modeling and sediment geochemistry from the Pliocene (Hill et al., 2007; Cook et al., 2013; Pollard et al.,

2015) show ice volume shrank to between 80 and 90% of modern before a retreat from the

Wilkes Land coastline was initiated. Though these models have boundary conditions that would be different from the Oligocene, especially in paleogeography and West Antarctic physiognomy, they provide clues to the sensitivity of the Wilkes Land coastline.

Larger Than IRD Debris Flow Cooler Modern 25

Oi2b 27

Oi2a 28

Age (Ma) 30

Oi1 34 0 500 1000 1.0 1.5 2.0 0 2 0 4 0 6 0 80100120140 IRD (grains/gram) Site 1218 δ18O (‰) % of Present-day Ice Volume acb

Figure 2-8 - IRD concentration record from IODP Site U1356). A: IRD concentrations in grains per gram of dry sediment >150 µm. Dashed line separates IRD from debris flow. A concentration peak at 31.1 Ma (white circle) is composed of carbonate particles that we interpret to have precipitated during diagenesis. B: Benthic δ18O record from ODP Site 1218, eastern equatorial Pacific (Pälike et al., 2006). Oxygen isotope (Oi) events (Miller et al., 1991; Pekar et al., 1996) are labeled Oixx. C: Estimated global ice volume compared to modern based on apparent sea-level records (Sites 522, 689, 690, and 1218) for the Oligocene (Pekar and Christie- Blick, 2008) (modern is 66 m [Wilson et al., 2013]).

bd Daily Insolation % Present-day Ice (w/m2) Volume 450 475 500 525 550 40 60 80 100 120 140 R

25.4

25.6

25.8

26.0 Age (Ma)

26.2

26.4

26.6 0 1 1 0 100 1000 1 -1 405 kyr IRD (grains/gram) less seasonality ac Figure 2-9 - IRD intervals from well-recovered Late Oligocene strata. Recovery (R). A: IRD concentrations for 26.6-25.2 Ma. Triangles denote samples used for 40Ar/39Ar analysis B: Daily insolation record for 65°S, December 21 (Laskar et al., 2004). C: Filtered 405,000-year cycle from eccentricity record using Analyseries (Laskar et al., 2004). Values close to -1 mark near- circular orbits (minimum 405,000-year eccentricity). D: Estimated global ice volume (see Figure 2-8c for details).

bd Daily Insolation % Present-day Ice (w/m2) Volume

R 450 475 500 525 550 0 20 40 60 80 100 120

30.6

30.8

31.0

31.2

31.4

Age (Ma) 31.6

31.8

32.0

32.2

32.4

0 1 1 0 100 1000 1 -1 405 kyr IRD (grains/gram) less seasonality a c Figure 2-10 - IRD intervals from well-recovered Early Oligocene strata. Recovery (R). A: IRD concentrations for 30.5-32.5 Ma. A concentration peak at 31.1 Ma (white circle) is composed of carbonate particles that we interpret to have precipitated during diagenesis. B: Daily insolation record for 65°S, December 21 (Laskar et al., 2004). C: Filtered 405,000-year cycle from eccentricity record using Analyseries (Laskar et al., 2004). Values close to -1 mark near-circular orbits (minimum 405,000-year eccentricity). D: Estimated global ice volume (see Figure 2-8c for details).

Based on the modeling results and comparison of the IRD abundance to moderate resolution sea-level records, the presence of IRD at Site U1356 is interpreted to indicate an EAIS size of at least 80%. This occurs more frequently than what has been suggested by ice volume estimates (Pekar and Christie-Blick, 2008) and oxygen isotope maxima (Oi- events) (Miller et al., 1991; Pekar and Miller, 1996; Lear et al., 2004). Below this threshold, it is likely the coastline in the WSB retreated inland and that oceanographic conditions were not as favorable for iceberg delivery to the Site U1356.

The higher IRD intervals at Site U1356 are interpreted to represent glacial growth in the

WSB region, which is supported by the qualitative correlation between the IRD abundances and the insolation minima and sea-level records. IRD as a glacial signal has previously been interpreted in offshore sediment cores from Antarctica (e.g. Breza and Wise, 1992; Knies et al.,

2001; Kanfoush et al., 2002; Passchier, 2011) This may not be a reflection of more icebergs being discharged along the ice sheet margin, which would be expected during warming (e.g.

Williams et al., 2010), but an indication that more icebergs are reaching Site U1356 during these cooler intervals. It has been suggested that subglacial lakes can cause outburst flooding when subjected to great pressure by an overlying ice sheet (Alley et al., 2006), which could produce large iceberg armadas during cooler intervals. Subglacial lakes do exist within the WSB today, which may suggest they existed during the Oligocene.

The interpretation of IRD as a glacial signal may not hold for drill sites closer to the continent because although retreating ice shelves contain significantly more sand and gravel due to the lack of comminution (Evans and Pudsey, 2002; Passchier 2011), the greatest depositional rates occur near the calving line (Domack and Harris, 1998). Site U1356 was likely not near the calving line during the Oligocene, which is supported by percentages of sand and gravel grains

less than 15% and seismic surveys (Donda et al., 2003; Escutia et al., 2005; Donda et al., 2008).

If icebergs were being produced during interglacial periods, most of the debris at the bottom of the ice column would be deposited before the iceberg calves. After calving, the debris would likely be deposited closer to the continent due to increased sea-surface temperatures and reduced iceberg survival rate. The IRD concentrations however, do not allow for an interpretation of the location of the paleocoastline or the exact amount of ice mass loss or gain, but they do provide us with clues that the coastline may have retreated inland when IRD is not present at Site U1356.

During the Early Oligocene interval (32.4-30.6 Ma), it is more difficult to interpret IRD layers as a glacial signal. Based on the current age model, it appears most of the higher IRD layers are corresponding with orbital parameters that suggest warming and more seasonality, opposite of the Late Oligocene interval. This offset suggests there was a change in the delivery of icebergs to Site U1356 between 30.6 and 28.0 Ma, where recovery of sediment is the lowest.

There are a few things factors that are limiting this interpretation. While recovery of this section was generally good, it was sampled at a lower resolution because the main focus was the Late

Oligocene. Also, the age model is not well constrained in this section (see Figure 2-6d,e).

2.4.2 Gravity Flows: Evidence for a larger than modern ice sheet

Intervals of some high IRD concentrations and gravity flow layers were correlated to

δ18O maximum events (Figure 2-8), which have been related to 1.2 myr low obliquity nodes

(Zachos et al., 2001; Pälike et al., 2006). Though the depositional ages of these flows have larger uncertainty than other intervals, they occur fairly close to these Oi events, where larger ice volumes may have been responsible for the generation of these flows.

One IRD event at 31.8 Ma (Oi1b) and three sediment slumps/gravity flows at 33.6 Ma

(Oi1), 27.9 Ma (Oi2a), and 26.6 Ma (partial Oi2b) correlate with Oi events. The interval that contained Oi1a (32.9 Ma) was sampled at a much lower resolution and Oi2 (30.0 Ma) was not recovered at Site U1356. A sediment flow also occurs at an unidentified glacial interval at 27.2

Ma, which is positioned between the two largest isotopic events, Oi2a and Oi2b, when stable isotopes suggest the greatest ice volumes of the Oligocene occur (Pekar and Christie-Blick,

2008).

Each of these events is interpreted to be the result of glacial growth larger than modern in the WSB region. While gravity flows could occur any time the upper slope becomes unstable, their timing coincides with Oi events, suggesting a glacially influenced mechanism. However, it is unlikely the ice sheet advanced all the way to the shelf edge after the earliest Oligocene, as seismic surveys along the Wilkes Land continental shelf (Donda et al., 2003; Donda et al., 2008;

Escutia et al., 2005) and the sandy mud and other fine-grained lithologies of these flow intervals

(Expedition 318 Scientists, 2011) do not support this. It is possible however, that an ice sheet could have expanded partially onto the shelf, where increased meltwater flow and sediment load could have resulted in over steepening of the slope, triggering gravity flows. Additionally, the propagation of the peripheral bulge due to increased isostatic loading (Stocchi et al., 2013) may have triggered flows as well. If the continental slope gradient was similar to today (~8-10°), then only minimal perturbation would be needed to trigger a gravity flow (Escutia et al., 2005).

2.4.3 Ar Thermochronology

The summary of all grain ages analyzed from Site U1356 is shown in Figure 2-7 and ranged from ~340 to 2000 Ma. The most dominant population of grains is in the 1400 to 750 Ma

range, which indicates a primary source region of the Adélie Craton. The ages of this sector are well constrained by onshore 40Ar/39Ar thermochronology (Di Vincenzo et al., 2007) and offshore

Holocene sediment 40Ar/39Ar thermochronology (Roy et al., 2007; Pierce et al., 2011; 2014)

(Figure 2-4). Grain ages around ~1500 Ma are most similar to those found in the Mertz Shear

Zone, which borders the eastern flank of the Adélie Craton and is the only known locale in this region for that age. Here, ductile deformation has been dated to be no younger than ~1500 Ma and regional-scale retrograde metamorphism under amphibolite-facies has been found to occur at

~1.7 Ga (Di Vincenzo et al., 2007). Much of the Mertz Glacier overrides the MSZ and drains a portion of the Adélie Craton, which makes it the most likely source of this sediment.

The second cluster of ages, which is significantly smaller than the dominant group, is from ~400 to 600 Ma. There are several potential locations for these grains to be derived from.

The closest is from the Early Paleozoic granites (Granite Harbour Intrusive Suite) and metasandstones near the Ninnis Glacier, where 40Ar/39Ar dates on white mica have ranged between 530 and 640 (Di Vincenzo et al., 2007). These rocks are located on the coast to the east of the MSZ. The Mertz Glacier essentially separates the exposures of Archean and Proterozoic bedrock of the Adélie Craton from the Paleozoic bedrock in the WSB. Additionally, Holocene marine sediments located directly downstream of the Ninnis Glacier contain hornblendes with

40Ar/39Ar ages between 480 and 560 Ma (Roy et al. 2007; Pierce et al. 2011). Another likely sources for grains within this age range are the high-grade rocks related to the Ross Orogeny, most likely coming from Northern Victoria Land. Peak metamorphism during the Ross Orogeny varied slightly between Northern Victoria Land (460-500 Ma), Southern Victoria Land (480-550

Ma) and the Transantarctic Mountains that border West Antarctica/Ross Sea (480-545 Ma)

(Goodge, 2007). Rocks eroded by the Ninnis Glacier are the best explanation for the older grains

in this cluster (>500 Ma), while rocks from Northern Victoria Land best explain the younger grains (460-500). Areas further south in the Transantarctic Mountains and their exposures on the

Ross Sea coast could also be potential sources, as it is possible for icebergs to travel as far as

1500 km for their source (Williams et al., 2010), but it is difficult to resolve them during the

Oligocene because the characteristic Cenozoic volcanism of the McMurdo Volcanic Group had not yet initiated. The 8 grains between 340 and 460 Ma cannot be easily resolved with either of these potential sources. They are best explained by the Bowers Terrane in Northern Victoria

Land, where whole-rock K-Ar ages range between 320 and 450 Ma (Adams, 2006).

The seventeen grains between 1000 and 1300 Ma (interpreted here as Grenville in age) in the Oligocene sediments at IODP Site U1356 are difficult to resolve. There are no exposures of rocks between ~800 and ~1400 Ma on the Wilkes, Adélie, or George V coasts or in Northern

Victoria Land, and there is no evidence that the MSZ region was reactivated during or before the

Ross Orogeny (Di Vincenzo et al., 2007). Di Vincenzo et al. (2007) did find that amphiboles within the MSZ contained excess Ar (parentless 40Ar), which would cause grains to exhibit older ages, however no evidence for this was found in the Ross Orogeny age rocks. These ages however, do closely resemble thermochronological ages for exposures west of the Adélie Craton

(Sheraton et al. 1992; Post et al. 1996, 2000; Möller et al. 2002; Fitzsimons, 2003) and 40Ar/39Ar ages of hornblende in Holocene marine sediments offshore of the Wilkes Land margin (Roy et al. 2007; Pierce et al. 2011; 2014) west of the Adélie Craton (and Site U1356). Iceberg delivery from this area though would not be in agreement with initiation of the Polar current and modern iceberg tracks, as they would have had to travel east upon calving to be delivered to the drill site.

The most reasonable explanation is that rocks retaining a Grenville-age signature likely exist underneath the ice sheet in the interior of the continent where erosion includes a path through the

WSB or Adélie Craton. Grenville-aged zircons dated using U-Pb have been found in glacial sediments off the Oates and George V coasts, where the ages suggest unexposed Grenville rocks exist inland of this region (Goodge et al. 2010). Therefore the Grenville aged grains at Site

U1356 are interpreted as being sourced from a continental interior source currently covered by the ice sheet. This hypothesis of an unknown bedrock source obscured by the ice sheet may also explain the 7 grains with ages between 800 and 1000 Ma.

Interestingly there is a lack of FLIP group ages (~178 Ma) at Site U1356, which have a small exposure just to the east of the Ross Orogeny aged rocks. While their known exposure on the coastline is small, large exposures underneath the ice sheet within the WSB have been interpreted from fine-grained Nd isotopes (Cook et al., 2013) and airborne geophysics

(Ferraccioli et al., 2009). This would suggest a large FLIP signature should be seen in the

40Ar/39Ar ages at Site U1356. The reason for not seeing the FLIP signature is likely due to the fact that mafic igneous rocks (i.e. FLIP) do not contain biotite and only have rare hornblende.

The hornblende and biotite ages are biased against the input of rocks from this group, but if grains with a Ross signature were coming from the Ninnis Glacier area, it is likely that FLIP sediments were also included. Additional work with the fine-grained geochemical signature of this sediment is needed, similar to that of Cook et al. (2013).

2.4.4 Hornblende vs. Biotite as a provenance signal

The distribution of hornblende and biotite grains shows there is a difference in the provenance signal based on the mineral being analyzed (Figure 2-9). The hornblende grains mainly exhibit ages in the 1400-1700 Ma range, suggesting a primary input of the MSZ and

Adélie Craton. Only 2 hornblende grains give an age between 300 and 500 Ma, indicating Ross

Orogeny age rocks and Northern Victoria Land (NVL) rocks were not contributing much coarse- grained sediment to Site U1356 during the Oligocene. The biotite grains on the other hand have a better distribution between 1400-1750 Ma and 400-600 Ma, with the Proterozoic grains being slightly more dominant. The older grains indicate an input from the MSZ and Adélie Craton, similar to the hornblende grains, but also show a higher population of grains representing the

Ross Orogeny or NVL. In selecting both hornblende and biotite grains for 40Ar/39Ar analysis, I had hoped to minimize biases that may exist by only using one type of mineral. It is clear that work done in this region needs to include a multi-mineral analysis to minimize the biasing of the hornblende grains.

Previous work on Holocene marine sediments from the Wilkes Land region also investigated the use of coupled hornblende and biotite 40Ar/39Ar analyses (i.e. Pierce et al., 2011;

2014). Here they found hornblende and biotite grains from sediment very proximal to the Mertz

Glacier had a dominant 1400-1750 signal, with no input of Ross Age material. A core of similar longitude, but further offshore of the Mertz Glacier contains a hornblende distribution very similar to that at Site U1356, but no biotite. Another core a bit further west showed biotite with a dominant Ross aged signal and minor Proterozoic while the hornblende was predominantly

Proterozoic.

10 Biotite 63−150 µm, n=48 5

number 0 10 Biotite >150 µm n=41 5

number 0 10 Hornblende >150 µm n=65 5

number 0 20 All Hornblende and Biotite 15 n=155 10 5 number 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Age (Ma) Figure 2-11 - Distribution of all hornblende and biotite 40Ar/39Ar ages analyzed for this study.

2.4.5 Small biotite vs. large biotite

Biotite grains of >150 µm and 63-150 µm were selected to see if there would be any grain size bias in the 40Ar/39Ar derived ages and to increase the number of grains analyzed because the >150 µm fraction did not contain many hornblende or biotite grains. A total of 41 large and 48 small biotite grains were analyzed (Figure 2-11). Their age distributions are quite similar to each other, suggesting that a grain size bias does not exist within the sand sized biotite grains at this site. The lack of grain size bias also indicates the biotites were being sourced from the same areas and have similar erosional and depositional histories. Therefore they can be treated as one population of grains. Hornblende grains between 63-150 µm were not chosen for

analysis because the smaller grains do not contain enough K, which means for there would not be enough Ar for the mass spectrometer to measure accurately. Other grains suitable for

40Ar/39Ar dating, such as feldspars or white mica, were not found or were only in very trace amounts.

2.4.6 Age distribution in the high IRD bearing layers

The age population amongst the higher IRD bearing layers (Figure 2-12) is quite similar to the overall distribution of all grains measured at Site U1356 (Figure 2-7). The hornblende population indicates a predominant source of the MSZ and Adélie Craton with 6 grains suggesting input of Grenville and other obscured rocks and 1 grain of Ross Orogeny age. The biotites also have a dominant Mertz and Adélie signal, but have more Ross aged grains and also

6 grains from obscured rocks within the continent. It is clear that during the deposition of these higher IRD layers, the icebergs were primarily being sourced from the Mertz Glacier with a minor input from the Ninnis glacier and others from further east. As suggested by the comparison of different sized biotites, they can be treated as one population here and can be combined with the hornblende grains to construct the downcore distribution among the 6 intervals sampled for 40Ar/39Ar analysis.

10 Biotite 63−150 µm n=27 5

number 0 10 Biotite >150 µm n=25 5

number 0 10 Hornblende >150 µm n=46 5

number 0 20 All Hornblende and Biotite 15 n=97 10 5 number 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Age (Ma) Figure 2-12 - Distribution of all hornblende and biotite 40Ar/39Ar ages in higher IRD abundance layers (IRD concentrations >10 grains/gram).

The downcore distribution within the higher IRD abundance layers (Figure 2-13) shows what may be a marked change in the delivery of icebergs to Site U1356 at ~26 Ma. Prior to the sample 672.12 mbsf, Ross/NVL aged grains are absent from the high IRD layers, aside from a single hornblende grain at 686.61 mbsf, which immediately follows a debris flow layer. The

Ross/NVL signal is primarily exhibited by the biotite grains. Within the lower three IRD layers analyzed, 17 biotite grains were analyzed, but none exhibited an age indicative of Ross/NVL. In contrast, 9 of the 36 biotites and measured in the upper three layers show a Ross/NVL age. No hornblendes from the upper 3 layers exhibited Paleozoic ages. These higher abundance IRD age are a great example of why it is important to utilize a multi-mineral approach.

10 646.97 mbsf, 25.4 Ma n=15 5

number 0 10 656.38 mbsf, 25.6 Ma n=26 5

number 0 10 672.12 mbsf, 26.0 Ma n=16 5

number 0 10 672.73 mbsf, 26.0 Ma n=20 5

number 0 10 686.53 mbsf, 26.3 Ma n=7 5

number 0 10 686.61 mbsf, 26.3 Ma n=22 5

number 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Age (Ma)

Figure 2-13 - Downcore distribution of 40Ar/39Ar ages from higher IRD intervals. Hornblende and biotite grains are combined for each sample.

2.4.7 Age distribution in the low IRD abundance intervals

Only 4 intervals of low or no IRD were sampled, with 24 grains measured (Figure 2-14).

Of these 24 grains, only 2 were hornblende from a single interval. The interpretations from these data have to be taken very carefully because the number of measurements is quite low for each

interval. This data set was expected to be smaller because the intervals analyzed contained few or no grains >150 µm, but it is nonetheless important to see what the ages of the grains were doing when the IRD counts were lower.

5 Biotite 63−150 µm n=11

number 0 5 Biotite >150 µm n=9

number 0 5 Hornblende >150 µm n=2

number 0 5 All Hornblende and Biotite n=22

number 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Age (Ma) Figure 2-14 - Distribution of hornblende and biotite 40Ar/39Ar ages in low IRD abundance layers (IRD concentrations <10 grains/gram).

The distribution among the biotite grains appears to remain quite similar to that of the biotites in the higher IRD intervals, including fairly equal amounts of Proterozoic and Ross

Orogeny ages. The 2 hornblende grains measured are Proterozoic in age, which is the dominant population in hornblendes from the higher IRD-bearing layers. Though this data set is quite small, it can be interpreted to mean that these areas still had at least some ice at the coastline, but

climactic and/or oceanic conditions may have prevented icebergs from being delivered to Site

U1356 or iceberg discharge rate was lowered.

2.4.8 Age Distribution of the Debris Flow Layers

The debris flow layers exhibit a distinct difference compared to the high and low abundance IRD layers. One of the main distinctions is the debris flow layers have a strong signal of ages in the 800-1400 range (Figure 2-15), which have been interpreted to represent rocks of

Grenville age and younger that are currently obscured by the ice sheet. In these layers, the

Grenville ages are best seen in the hornblende and 63-150 µm biotites. While there does appear to be a grain size bias between the biotites (not seen in IRD intervals), the overall age distribution of the smaller biotite grains resembles what can be seen in the hornblende grains, which is counter intuitive. The larger biotite grains seem to be restricted to the MSZ/Adélie signal. With only 9 and 10 grains analyzed for large and small biotites respectively, the most likely reason for this apparent grain size bias is that not enough grains have been analyzed to create an age distribution that is robust enough to make an interpretation.

Even though only 36 grains were analyzed from these debris flow layers, the 800-1400

Ma signal is comparatively stronger here than in the high and low abundance IRD intervals.

These debris flow layers were already interpreted to reflect an ice sheet that is near or larger than modern (see section #). This brings up an interesting question as to where these grains were coming from because there is not a large signal in the IRD layers. These are debris flow layers, so the direct source of these grains is likely from higher up on the continental rise on the continental shelf where the sediment was redeposited downslope due to a glacial mechanism.

5 Biotite 63−150 µm n=10

number 0 5 Biotite >150 µm n=9

number 0 5 Hornblende >150 µm n=17

number 0 5 All Hornblende and Biotite n=36

number 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Age (Ma) Figure 2-15 - Distribution of hornblende and biotite 40Ar/39Ar ages for the debris flow layers.

The geographic location of Site U1356 is on the border between Adélie Land and Wilkes

Land proper, which means it is still east of Grenville age rocks in Wilkes Land. It may be possible that icebergs initially carried this material to the outer continental shelf directly from

Wilkes Land. While the Polar current does move westward, icebergs can be deflected northward due to bathymetric highs and can become entrained in the eastward Antarctic Circumpolar

Current (ACC), which has been observed in modern iceberg tracks near the Wilkes Land margin

(Gladstone et al., 2001; Stuart and Long, 2011). This is difficult to resolve however, because it is highly dependent on the strength a latitudinal position of the ACC.

The best explanation is this material was eroded from somewhere in the interior of the

WSB/Adélie region and that the expansion of the ice sheet caused the source rocks to become more susceptible to erosion. Perhaps a change in the thermal regime of the ice sheet caused an increase in erosion. The original source of this sediment and its erosional and transport history remain a very important question. Additional work is needed in both Ar thermochronology and geochemistry of fine-grained sediments to further develop the source of these grains in the debris flow sections.

2.4.9 Sediment Transport during the Late Oligocene

In addition to the provenance of the IRD and ice sheet erosional history, the Ar thermochronology can also provide some insight into sediment and iceberg transport during the

Late Oligocene. Seismic sections from Prydz Bay (Kuvaas and Leitchenkov, 1992) and modeling of ice sheets and wind fields (DeConto et al., 2007) indicate the westward flowing Polar current was already in effect during the Oligocene, indicating iceberg transport was westwards. This is a wind driven current, primarily by the katabatic winds coming from the higher elevations of ice sheet. As these pressure gradient winds move northward along the ice sheet, the Coriolis Force causes the winds to turn left (southern hemisphere), thereby generating the Polar current as the air moves over water. Since ice streams and outlet glaciers are the primary sources of IRD (O

Cofaigh et al., 2001; Death et al., 2006), the most likely source region for the IRD at Site U1356 were the outlet glaciers draining the WSB (i.e. Mertz and Ninnis Glaciers) or further east.

The overall age ranges from the 40Ar/39Ar thermochronology were expected based on the anticipated sediment source regions, as defined by the local geology of the WSB coastline and surrounding regions. The actual ages of the IRD and the expected source regions of WSB and

Adélie Craton are in agreement. This indicates icebergs travelled westward upon calving from the ice sheet margin, and that these icebergs likely originated east of Site U1356. This is in agreement with modern iceberg drift models (Gladstone et al., 2001; Stuart and Long 2011) and supports the work of DeConto et al. (2007) and Kuvaas and Leitchenkov (1992) that indicate ocean currents during the Oligocene were similar to today, at least for this region of Antarctica.

In addition to supporting the existence of the Polar current during the Late Oligocene, this data set can also place constraints on its northerly extent and iceberg transport. Since the abundance of IRD at Site U1356 was low, it is likely the Polar current did not fully extend as far north as the drill site. Icebergs that were delivered to the drill site were most likely deflected northward out of the Polar current as described by Williams et al. (2010) and Stuart and Long

(2011).

2.4.10 Implications for ice sheet changes the Oligocene

The combined data sets of the IRD concentrations with their qualitative correlation to insolation and eccentricity changes and the 40Ar/39Ar ages of grains from several different sediment layers provides an excellent proxy for ice sheet changes occurring in the WSB region during the Oligocene. Beginning at the base of the IRD concentration record in the Early

Oligocene, evidence of the Oi1 glaciation is present, indicated by the sandy lithology and high abundance of grain counts (Figures 2-6, 2-8). This suggests the ice sheet did reach the WSB coastline during the initial glaciation. Following Oi1, IRD concentrations drop, indicative of a warming immediately after the initial large growth. Moving into the well-recovered Early

Oligocene interval (Figure 2-10), IRD is present which indicates the ice sheet was at the coastline in the WSB, even under what may be warming conditions. While there are no

provenance data for this interval in this study, the source distribution is very similar to what has been found during the Late Oligocene (Tina van de Flierdt, personal communication).

For the Late Oligocene, the downcore distribution of all the sediment layers analyzed for

Ar thermochronology is presented in Figure 2-16, allowing a more detailed interpretation of ice sheet changes. Starting from the debris flows, which are the oldest strata where Ar data was collected, it is clear there is a very mixed signal in the population of grain ages during the first flow at ~27.6 Ma, which correlates to the Oi2a isotope maximum. Following this flow, are two additional ones at ~27.2 and 26.6 Ma. These two flows had a lower number of grains measured, but do show a primary age representing sediment from the MSZ/Adélie region. Since these layers are not IRD, it is difficult to tell exactly when the sediment was first deposited on the shelf

(prior to the flow occurring). The grains from these debris flow layers were likely in the most recent sediment deposits up on the shelf and likely ice-rafted in origin as the ice sheet expanded through the oxygen isotope maximum. Either way, these flow layers were most likely triggered by a glacial mechanism based on their correlation to the large increases in oxygen isotope records.

After the debris flows, the IRD concentrations reflect 405-kyr cooling cycles with high

IRD-bearing layers (>10 grains/gram) occurring during times where estimated ice volume is at least 80 % of modern and layers with low or no IRD below 80 % of modern (Figure 2-9). The ages of grains in the higher IRD abundance layers from ~26.5 Ma to 26.0 Ma primarily reflect iceberg input from the Mertz Glacier, which is the closest major outlet glacier to Site U1356. In contrast, the low abundance IRD layers show an increased input of Ross/NVL sources, indicating outlet glaciers such as the Ninnis. It is difficult to interpret the low IRD abundance layers because of the low number of grains analyzed. The low IRD layers however, do contain

more grains of the Ross/NVL origin than the higher abundance intervals that have more grains analyzed. Therefore the age distribution of the lower abundance layers can still help to provide a robust interpretation of glacial dynamics.

The appearance of the Ross/NVL aged grains during the low IRD intervals (concurrent with higher insolation variance and eccentricity) may be interpreted as the deglacial signal of the

WSB, of which there is evidence of at least 2 episodes of retreat at ~26.5 Ma and ~26.2 Ma. The former immediately follows the Oligocene isotope maximum and associated ice volume maximum (e.g. Palike et al., 2006) and both occur during large variance in insolation (higher seasonality). Since this area would be the first to retreat under warming (e.g. Pollard et al., 2015), the icebergs produced as the ice sheet collapses are mainly going to be sourced from the Ninnis

Glacier and possibly the Cook Glacier. This scenario may fit with the current models and understanding of glacial retreat in the WSB region (i.e. Pollard et al., 2015). Even if a large retreat occurs in the WSB however, the Adélie and Northern Victoria Land regions would not experience a significant retreat (Pollard et al., 2015). This could mean that the Mertz Glacier would still be active, which is the dominant source of IRD at Site U1356, and is still present during most low IRD intervals. During these retreats the Ninnis Glacier may release iceberg armadas (Williams et al., 2010), some of which may travel to Site U1356. If the Mertz Glacier is more stable because it primarily drains a region above sea-level, iceberg production would likely be less than that of the collapsing Ninnis Glacier and can explain the increase of the Ross/NVL signature.

5 High IRD: 646.97 mbsf 25.4 Ma n=14

number 0 5 High IRD: 656.38 mbsf 25.6 Ma n=26

number 0 5 Low IRD: 660.88 mbsf 25.7 Ma n=2

number 0 5 High IRD: 672.12 mbsf 26.0 Ma n=16

number 0 5 High IRD: 672.73 mbsf 26.0 Ma n=20

number 0 5 Low IRD: 682.58 mbsf 26.2 Ma n=5

number 0 5 Low IRD: 682.90 mbsf 26.3 Ma n=9

number 0 5 High IRD: 686.53 mbsf 26.3 Ma n=7

number 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Age (Ma)

40 39 Figure 2-16 - Downcore distribution of all hornblende and biotite Ar/ Ar ages.

5 High IRD: 686.61 mbsf 26.3 Ma n=15

number 0 5 Low IRD: 691.08 mbsf 26.5 Ma n=6

number 0 5 Debris Flow: 694.95 mbsf 26.6 Ma n=11

number 0 5 Debris Flow: 710.73−712.16 mbsf ~27.2 Ma n=6

number 0 5 Debris Flow: 719.79−723.25 mbsf ~27.6 Ma n=19

number 0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Age (Ma)

Figure 2-16 (continued)

At ~26.0 Ma, the Ross/NVL signature appears in the higher abundance IRD layers, suggesting input from the Ninnis Glacier or others further east. Also at about 26.0 Ma is the Late

Oligocene warming, which can be seen in the combined oxygen isotope record of Zachos et al.

(2001), but it was found to be overstated due to the geographic distribution of the data sets

(Pekar and Christie-Blick, 2008). A recent, complete high-resolution stable isotope record of the

Oligocene from IODP Site 1218 (Palike et al., 2006) shows a decrease in δ18O values to below

2‰ after ~26.0 Ma and values above 2.5‰ from ~29-26.5 Ma. Our dataset suggests the Late

Oligocene warming may have affected the WSB, as indicated by the increase in Ross/NVL aged grains after 26.0 Ma.

2.5 Conclusion

Our interpretations place new constraints on EAIS extent during the Late Oligocene and show that the WSB was dynamic and sensitive to climate change during the early icehouse world. Our coupled IRD concentrations and 40Ar/39Ar thermochronology show the ice sheet frequently reached the coastline, at least partially, within the WSB during the Oligocene, suggesting a recurrent, near present-day size (at least 80%) ice sheet under elevated CO2. Ice volume greater than modern is interpreted from the gravity flows seen in our record, which have a similar timing with δ18O maxima that in turn correlate well with the 1.2 myr obliquity cycle.

Insolation and 405-kyr eccentricity may have influenced the deposition of IRD, but uncertainties in portions of the age model for Site U1356 do not permit a quantitative correlation.

The Ar thermochronology results suggest the Ross/NVL signal is best seen in the biotite grains, which have a wider distribution between that and the MSZ/Adélie signal. The hornblende grains however, are almost exclusively from the MSZ/Adélie region. Additionally, there was no distinct difference between the age distributions of biotite of different grain size. This data suggests that the biotites could be treated as one population and, when combined with hornblende, provide an excellent record for provenance interpretations.

The ages of the IRD grains shows the primary source of icebergs to Site U1356 was the

Mertz Glacier, indicated by the dominant ages populations reflecting the MSZ and Adélie

Craton, with a secondary location of the Ninnis Glacier or outlet glaciers as far east as NVL. The

downcore distribution of the IRD ages is indicative of a steady input from the Mertz Glacier, and an increased supply of grains from the Ninnis during warmer periods. Additionally, the use of biotite as a provenance indicator in the WSB region was evaluated. Our IRD study also provides evidence to support some periods of warming during Late Oligocene, but this record stops at

~25.4 Ma, making it difficult to see an overall trend towards warming. Our IRD dataset however, still has the potential to act as a constraint for numerical models of ice sheet dynamics in the

WSB region by forcing models to include some form of glacial ice reaching the coastline and calving under elevated CO2.

Chapter 3. Evidence from the Ross Sea DSDP Site 270 for a large West

Antarctic Ice Sheet during the Late Oligocene (25.8-25.3 Ma)

Abstract

Ice-rafted debris (IRD) 40Ar/39Ar ages from Deep Sea Drilling Project (DSDP) Site 270 from the central Ross Sea indicate a substantial input of grains sourced from rocks that can only be explained by the geology of West Antarctica, suggesting the presence of a large ice sheet in this sector of Antarctica. Two populations of grains exist, ~460-560 Ma and ~100-450 Ma, which have been interpreted as Ross Orogeny/Transantarctic Mountains and West

Antarctica/MBL respectively. The distribution of these ages in the 500-kyr interval of this study shows distinct changes in ice flow into the Ross Sea, suggesting ice coming out of West

Antarctica was primarily dominant. While the presence of an ice on West Antarctica during the

Oligocene has previously been suggested, this study presents the first geochemical evidence from the Ross Sea.

3.1 Introduction

3.1.1 Previous evidence for West Antarctic glaciation

Previous evidence for the timing of large scale West Antarctica glaciation has varied significantly from as early as the Eocene (45 Ma) where evidence suggests that mountain glaciers may have formed in the Peninsula region (Birkenmajer et al., 2005), to as late as the

Pliocene (~6 Ma), as interpreted from the oxygen isotope stack of Zachos et al. (2001; 2008).

Evidence has been mounting for a largely glaciated West Antarctica during the Oligocene.

Paleotopographic models indicate much of West Antarctica would have been above sea-level during the Early Oligocene, allowing ice volume as high as 130-140% of modern, (Wilson et al.,

2012; 2013), which support ice-volume estimates of Pekar et al. (2006). Dinoflagellate biostratigraphy and strontium isotopes from the Antarctic Peninsula suggest major glaciation at the Eocene-Oligocene boundary (Ivany et al., 2006), which is in agreement with the modeling of

Wilson et al. (2012). Patterns in the seismic reflection data of the easternmost Ross Sea are indicative of glacial erosion and deposition before 25 Ma (Sorlien et al., 2007). Seismic facies analyses from the central Ross Sea indicate the initial build-up of ice in West Antarctica first occurred during the Latest Oligocene, with ice in the highland areas of Marie Byrd Land and ice caps on the elevated (above sea-level) basement blocks in the central Ross Sea (Bart and De

Santis, 2012). Glacial erosion between ca. 28 Ma and 15 Ma of dated volcanoes in the interior of northern Marie Byrd Land as well as hydrovolcanic rocks suggest at least small Oligocene ice cap there (Rocchi et al., 2006). In the eastern Ross Sea, seismic stratigraphic evidence for regional grounded ice has been diversely interpreted as Late Oligocene (Bartek et al., 1992), middle Miocene (Bart, 2003), and late Miocene (De Santis et al. 1999). While the consensus of these studies suggest the presence of ice in West Antarctica during the Oligocene, it remains unclear whether a large WAIS existed or if glaciation was more localized in elevated regions with perhaps some small outlet glaciers.

3.1.2 Importance and sensitivity of the WAIS

Most of the West Antarctic Ice Sheet (WAIS) is grounded below sea-level on a bed that deepens inland, which makes it quite sensitive to climate forcings and oceanographic changes

(e.g. Pollard et al., 2015). It has been suspected that the WAIS would be subjected to

catastrophic collapse and rapid disintegration with only a small change in conditions, and that such a collapse would be responsible for 3.3 m sea-level rise out of the total volume contained in the WAIS (Bamber et al., 2009; Joughin and Alley, 2011). Evidence exists to support the WAIS has undergone partial collapse in the past, possibly as recently as 400 Ka as a consequence of moderate warming (Fox, 2008; Scherer et al., 1998).

The instability and rapid collapse of the WAIS is a consequence of the marine ice sheet instability (MISI) hypothesis, which suggests the removal of fringing ice shelves will result in the rapid and irreversible inland migration of the grounding line where the bedrock is below sea level and slopes downward from the margins toward the interior (Thomas, 1979; Weertman,

1974). This results in the glacier becoming grounded in deeper water with a greater ice thickness because the grounding line has a reverse-bed gradient that becomes deeper inland. In fact stable grounding lines cannot be located on upward-sloping portions of seafloor (Schoof, 2007). This hypothesis was first suggested in the 1970s, and has recently been supported by theoretical analysis of grounding line stability (Schoof, 2007). It is highly likely the WAIS was not grounded below sea-level during the Late Oligocene. Paleotopographic models of the earliest

Oligocene indicate West Antarctica was mainly above sea-level (Wilson et al., 2013) and up to 3 km of sediment has been eroded from West Antarctica (Wilson and Luyendyk, 2009) (see Figure

1-11). This would allow for an ice sheet significantly larger than modern to develop, up 140 %

(Wilson et al., 2012).

Determining the WAIS’s capacity to respond to climate change is an important part of

Antarctic and climate research because of its potential effects on the globe, due to its sensitivity.

Similar to sensitive areas of East Antarctica (e.g. Wilkes Subglacial Basin), the best way to determine the potential response of the WAIS is to document changes that have occurred in the

past. Extensive research has been conducted on the WAIS, especially with the collapse of large ice shelf masses, such as the Larson B ice shelf in 2002 (e.g. Rack and Rott, 2004) and there have been numerous expeditions and sediment cores recovered that document the history of the

WAIS. Previous evidence for WAIS dynamics from marine sediments along the continental margin include cores drilled by the Nathaniel B. Palmer cruise ship, the SHALDRIL program,

ANDRILL program, and Ocean Drilling Program. While these expeditions have been successful at recovering sediment derived from the WAIS, especially in the peninsula and Weddell Sea regions, they mainly contain strata of Miocene age or younger. Interest has also been generated in the Pine Island Glacier region, as this region is one of the most susceptible areas of West

Antarctica (Hughes, 1981; Joughin et al., 2010, Rignot et al., 2011).

The most recent work done in the Ross Sea is from two sediment cores in the McMurdo

Sound region (western Ross Sea) by the ANDRILL program. The record preserved in the AND-

1B core is quite good and produced an excellent record of the Pliocene. While interpretations were made in regards to a WAIS influence on the sediment and the collapse of ice in the Ross

Sea (e.g. Naish et al., 2009), the drill site is quite far from West Antarctica. Similarly in the

ANDRILL AND-2A core, certain intervals during the Middle Miocene have been interpreted to contain sediments from further south in the Ross Sea and that the deposition of the sediment was influenced by the movement of ice out of West Antarctica (Hauptvogel and Passchier, 2012;

Talarico et al. 2011). Under this scenario, ice flow in the Ross Sea is primarily controlled by

West Antarctic ice instead of East Antarctic dominance or a more balanced flow between West

Antarctica and the outlet glaciers through the Transantarctic Mountains (TAM).

3.1.3 DSDP Site 270

Deep Sea Drilling Project (DSDP) Site 270 was part of Leg 28 in the Ross Sea. Currently

Site 270 is centrally located between East and West Antarctica in the middle of the Ross Sea

(77.44°S, 178.50°W) in 634 m water depth (Figure 3-1). The sedimentary record of this site includes the Late Oligocene, Early Miocene and Pliocene to recent. A large hiatus exists between the Early Miocene and Pliocene, which may be related to the expansion of the West Antarctica

Ice Sheet during the early Middle Miocene (Bart, 2003). While the overall recovery of this core is only 64%, the average recovery for Oligocene strata is ~85%, making these sediments useful for investigating the Oligocene cryosphere. The Oligocene strata recovered at Site 270 span a 3- myr period from 26 to 23 Ma.

The drillsite is located on the eastern flank of the Central High, an uplifted crustal block that is part of the fault-block basin structure of the Ross Sea, a process that began forming during rifting in the Mesozoic. The base of the core is composed of metamorphic basement rock, which is calcsilicate gneiss and marble, in which apatite grains show a U/Pb age of ~436 Ma (Mortimer et al., 2011). Above the basement rock lies a glauconite layer that shows a minimum K/Ar age of

~26 Ma (McDougall, 1977; Kulhanek et al., 2014), and is indicative of submergence of the drill site and initiation of marine conditions. This is followed by a thick sequence of glaciomarine sediments.

It is important to recognize K/Ar ages of glauconite typically exhibit values that are 10-

20 % younger than the age of sedimentation, due to the ease of in which radiogenic argon is lost from the mineral and replaced by potassium (Hurley et al., 1960; Thompson and Hower, 1973).

Although a method has been proposed to correct for this Ar loss (Thompson and Hower, 1973), this was not applied by McDugall (1977). Additionally, it has been shown that 40Ar/39Ar

− 30˚

70˚ 80˚

− − − 0˚ 60˚ 30˚ 60˚ −90˚ 90˚

120˚

m 2000 1000 0

−1000 elevation −120˚ Site 270 −2000

−70˚ 150˚ 0 500

− 60˚60 ˚ 180˚

150˚ 180˚ − − 80˚ 3 − − − 00

−7 00

−600 Site 270

500

−700 −600

500 0 − − 0 500 6 −400 −

− 300 − 400 −

− 5 0 0

Figure 3-1 - Location map of DSDP Site 270. Upper map is overlaid with BEDMAP2 topography (Fretwell et al., 2013).

chronology is not a reliable measure for glauconite age due to large 39Ar loss (recoil) because of the fine blades that actually make up the mineral (Foland et al., 1984). Therefore, it can be assumed that 26 Ma age is the youngest possible age for this interval.

The glauconite layer and bedrock suggest the drillsite was above sea-level prior to the deposition of marine sediments. Paleotopographic reconstructions from the Early Oligocene do show that the Central High was likely above sea-level (Wilson et al., 2012). This suggests the

Central High had begun subsiding by at least 26 Ma, the time the glauconite layer was deposited.

3.1.4 Ross Sea regional geology

The Ross Sea region is bound to the west and south by the Transantarctic Mountains

(TAM) of East Antarctica, and east by West Antarctica and Marie Byrd Land (MBL) (Figure 3-

2). Since the Proterozoic, the TAM have undergone repeated tectonic and magmatic events

(Fitzgerald, 2002; Goodge et al., 2004; Goodge, 2007). The present-day Transantarctic

Mountains were shaped as a result of three episodes of uplift occurring during the Early

Cretaceous, Late Cretaceous, and early Cenozoic (Fitzgerald, 2002). Each of those events is related to the breakup of Gondwana and the separation of Antarctica from other continents as it moved to its present polar position. The basement rocks of the Transantarctic Mountains are primarily composed of late Proterozoic-Cambrian amphibolite facies, metamorphic rocks of the

Nimrod, Byrd, and Beardmore Groups, which are intruded by the Cambrian-Ordovician granitoids of the Granite Harbour Intrusive Suite (Fitzgerald, 2002; Sandroni and Talarico,

2006; Goodge, 2007).

Figure 3-2 - Ross Sea Geology (from Cook, 2013). TAM geology is the same from Figure 2-3.

The basement rocks are unconformably overlain by the Beacon Supergroup, which is composed of Devonian-Triassic glacial, alluvial, and shallow marine sandstones, siltstones, conglomerates and minor coal measures (Barrett, 1991). The Ferrar Large Igneous Province

(FLIP) are Jurassic age intrusive (Ferrar Dolerite) and extrusive (Kirkpatrick Basalt) igneous rocks related to initial rifting between East and West Antarctica, that intruded the basement complex and Beacon Supergroup simultaneously (Elliot, 1975). These rocks have been dated using 40Ar/39Ar of feldspar grains to be 176.6 ± 1.8 Ma at several locations throughout the TAM

(Fleming et al., 1997; Heimann et al., 1994). It was determined that the FLIP group was a short- lived, wide-spread lateral intrusion into the TAM.

This is followed by a 160 Ma gap in the geologic record until rift-related igneous rocks of the Erebus Volcanic Province (sometimes referred to as the McMurdo Volcanic Group) started forming over the past 20 myr (e.g. LeMasurier and Thomson, 1990; Sandroni and Talarico,

2006) with the eruption of trachytic rocks 19-12 Ma and basanitic to phonolitic rocks during the past 12 Myr (Di Vincenzo et al., 2010; Martin et al., 2010). Prior to 12 Ma, the only known volcanic centers were located south of the McMurdo Sound in the Mt. Morning volcanic area and under the ice sheet in the southern Transantarctic Mountains (Stump et al., 1980).

Today, the TAM form a range of peaks with elevations that exceed 4,000 m and impede outflow of the East Antarctic ice sheet into the Ross Sea embayment (Kerr and Huybrechts,

1999). Cenozoic uplift has been attributed to either normal faulting or strike-slip faulting or a combination with an isostatic response to glacial erosion (Stern et al., 2005; Miller et al., 2010).

Rapid uplift and denudation occurred in the Transantarctic Mountains between 40 and 26 Ma and fluvial and/or glacial erosion of the mountain front in late Oligocene-early Miocene time (<26

Ma - 14 Ma) produced several large-scale erosion surfaces (Sugden and Denton, 2004; Miller et al., 2010). These erosion surfaces were later dissected by rivers and outlet glaciers (Passchier,

2001; Hicock et al., 2003; Lewis et al., 2006; Jamieson and Sugden, 2008) , but rock and surface uplift is estimated to be less than 800 m since ~14 Ma (Wilch et al., 1993; Miller et al., 2010).

Bart and DeSantis (2011) have summarized the geologic history of the Ross Sea from the

Oligocene to present. During the Late Oligocene, they suspect the Central High (and other uplifted blocks in the Ross Sea) was ice capped and created a well-developed glacial drainage system, waxing and waning into the surrounding waters. This would indicate that the Central

High should be the source of most of the sediment deposited at DSDP Site 270. They also suggest that elevated areas of West Antarctica may have also been ice capped.

In addition to the regional geology, Holocene sediment from core tops around the Ross

Sea region have previously been investigated for 40Ar/39Ar provenance of K-bearing minerals

(Figure 3-3). As a compliment to the limited on-land data, these sediment ages provide valuable information about the chronology of the nearby coastlines, and can be used to help trace the likely sources of IRD in other marine sediment cores.

− 30˚

70˚ 80˚80˚ 0˚ − −− − 60˚ 30˚ 60˚ −90˚ 90˚

ELT11−19 m 2000

ELT11−18 NBP99−02 53PC 1000

DF85 96−1 0 DF85 PC109 elevation 120˚ −1000 ELT11−17 1 −2000

NBP00−01 KC2

ELT33−12 C

ELT33−11 ArAr NBP99−02 21P 2000 Ma

1500 − −120˚ DSDP 270 age − 1000 Regional 40Ar/39Ar Thermochronology (or K/Ar) NBP9802−2H 70˚ Holocene IRD ages − 500 150˚ thermochron ELT27−20 On-land hornblende ages

DF80−34 0 On-land biotite ages DF80−35 DF80−20 0 500 − 60˚

150˚ 180˚ −

Figure 3-3 - Regional map of Ross Sea. Colored circles are on-land 40Ar/39Ar ages of hornblendes and micas. Pie charts are the distribution of 40Ar/39Ar ages in core-top sediment.

3.2 Methods

A total of 14 sediment samples were taken from Late Oligocene sections of DSDP Site

270 between cores 36 and 43 (~320-385 mbsf) and 3 samples from the recovered bedrock from cores 47, 48, and 49. The recovery at Site 270 was generally poor and this low-resolution sample

set (5-8 m/sample) was chosen as an initial look into the provenance of Ross Sea IRD during the

Late Oligocene. Specifically the goal was to identify any West Antarctic IRD input to the central

Ross Sea to see if there was a large ice sheet in West Antarctica.

3.2.1 Site 270 Age Model

During the original development of this study, the depositional ages for the sediment samples were linearly calculated from the glauconite at the base of the glacial sequence (located in cores 43 and 44), which was previously dated using K/Ar to obtain an age of ~26 Ma

(McDougall, 1977), and paleomagnetic interpretations (Allis et al., 1975) and forminiferal evidence (Shipboard Scientific Party, 1977) which place the Oligocene-Miocene boundary (23

Ma) between 260 and 300 mbsf, well above the samples chosen for study. The original age model for DSDP Site 270 also included a linear sedimentation rate which suggested the base of the sedimentary sequence was ~26 Ma (Shipboard Scientific Party, 1977), in agreement with the

K/Ar age minimum for the glauconite sequence.

Recently, new age constraints were developed for Site 270 (Kulhanek et al., 2014) because of renewed interest in drilling future sediment records in this region (i.e. ANDRILL

Coulman High). Kulhanek et al. (2014) re-analyzed the glauconite at the base of the core, which did agree with the original analysis from McDougall (1977) of no younger than 26 Ma.

Additionally, Kulhanek et al. (2014) re-evaluated the biostratigraphy and magnetostratigraphy that now places the Oligocene-Miocene boundary at ~75 mbsf, which is about 200 m higher than the original interpretation from the Shipboard Scientific Party (1977). They do point out however, that this boundary could be deeper, so more work is being done to improve the age interpretations. This new age interpretation for Site 270 is the current model used for this study

(Figure 3-4). This means the 14 sample, low-resolution study now becomes somewhat higher resolution, spanning only ~500 kyr instead of 3 myr.

100 LO Kisseleviella Kulhanek et al., 2014 tricoronata preferred age model 150

200

250 Depth (mbsf) LO Dictyococcites bisectus 300

350 LO Chiasmolithus altus Glauconite Age 400 20 21 22 23 24 25 26 27 Age (Ma) Figure 3-4 - DSDP Site 270 age model. Data points are from Kulhanek et al. (2014). Solid line represents the preferred age model. LO=last occurrence.

3.2.2 Sediment Preparation

The 14 sediment samples were sieved and separated in an identical way to Site U1356

(See section 2.2.2). IRD grains were not counted for this site, but particle size wt% data was collected on the sand sized sediment. Potassium-bearing minerals were hand-picked under a binocular microscope for 40Ar/39Ar thermochronology. These grains were handled and analyzed the same way as those from Site U1356.

In addition to the 14 sediment samples, three bedrock samples were also chosen for

40Ar/39Ar analysis of any hornblende or biotite grains. We had hoped to obtain an age of the bedrock in addition to the U/Pb of Mortimer et al. (2011) to directly compare sediment

provenance data to. The bedrock samples were crushed and magnetically separated in an attempt to isolate any hornblende and biotite grains for 40Ar/39Ar chronology.

3.3 Results

A total of 151 grains were run for 40Ar/39Ar ages from the Oligocene sediment at Site

270, and only 11 of the 14 intervals selected contained K-bearing minerals in the > 150 µm fraction. Biotite comprised most of the grains analyzed (128 grains) with very little hornblende found (23 grains). The 63-150 µm fraction was not picked for biotite grains for Site 270. The ages exhibit a range from 100 to 800 Ma, with the largest cluster of ages around 500 Ma. There are also smaller age peaks around 100, 200, and 300 Ma. The distribution in particle size at Site

270 does not show any correlation with distribution in 40Ar/39Ar ages. The range of the sand sized sediment varies from 0 to ~36%, however sandier intervals did not always yield more K- bearing minerals.

The bedrock samples from the base of the core did not produce any K-bearing minerals.

Hornblende was not present (as expected for calcsilicate) and the biotite grains were altered to chlorite, as previously suggested by petrographic descriptions (Mortimer et al., 2011). Therefore, there are no 40Ar/39Ar data from the bedrock of the Central High to directly compare our IRD ages to see if they were being sourced from the Central High as previously suggested (Bart and

DeSantis, 2011). There is reason to believe the Central High was not a large contributor for the sand sized sediment because there was no biotite in the bedrock samples since it was all altered the chlorite, and biotite was the main K-bearing mineral in the Oligocene sediment. This indicates the biotite grains found in the Oligocene sediment are not sourced from the Central

High. At this point other basement highs in the Ross Sea cannot be eliminated as sources of the

Oligocene sediment because the ages of the bedrock south of Site 270 that lie under the Ross Ice

Shelf are unknown.

3.3.1 Potential sources of IRD

The age populations of the IRD grains indicate that both East and West Antarctica may have had a significant contribution of IRD into the Ross Sea (Figure 3-5). Rocks relating to the

Ross Orogeny best explain the largest population for the IRD grain ages (~460-500). As discussed in Chapter 1, the Ross Orogeny has varying peak metamorphic ages depending on the location. They vary slightly between Northern Victoria Land (460-500 Ma), Southern Victoria

Land (480-550 Ma) and the Transantarctic Mountains that border West Antarctica/Ross Sea

(480-545 Ma) (Goodge, 2007). Based on the current transport of icebergs and ocean currents in the northern Ross Sea, it would be unlikely that icebergs calving from Northern Victoria Land or

Southern Victoria Land would be brought to Site 270. It is possible however, that these grains could be sourced from further south in the TAM, near the southwestern portion of the Ross Sea or as far south as where the TAM separate East Antarctica from West Antarctica.

The three smaller peaks between 100 and 400 Ma are best explained as being sourced from West Antarctica. Rocks of these ages are not known in East Antarctica and there is no indication of tectonic activity after the Ross Orogeny aside from the emplacement of the FLIP group during the opening of the WARS (Boger, 2010). Despite an active rift system bordering the TAM, there is no indication of widespread metamorphism resulting from this, only the tectonic uplift of the mountains at ~200 Ma, as indicated by apatite fissure track dating (e.g.

Balestrieri et al., 1994).

40 All 40Ar/39Ar Ages 35 n=151 30 25 20

number 15 10 5 0 0 200 400 600 800 Age (Ma)

Figure 3-5 - Summary of all argon ages for DSDP Site 270.

Within West Antarctica, there are two probable source areas for the 100-400 Ma grains.

The first is the MBL block, where outcrops have been extensively studied and dated. The rocks here are younger in age, ranging between ~100 and 450 Ma (Adams, 1986, 1987; Luyendyk et al., 1996; Mukasa and Dalziel, 2000). Rocks exposed in northwestern Marie Byrd Land are

Paleozoic and Mesozoic in age with some similarities in petrography and age to those in

Northern Victoria Land of the TAM and rocks of New Zealand and Australia (references). The

Mesozoic rocks present are related to the extension, intrusion, and uplift of Marie Byrd Land as the WARS continued to spread (e.g. Luyendyk et al., 1996). Though this region is slightly north and further east of Site 270, it has been shown that terrigenous material from this region can be transported across the Ross Sea and around the tip of Northern Victoria Land (Cook et al., in prep).

Within MBL, the Ford granodiorite is the most exposed rock and has been dated at 353-

379 Ma using Rb-Sr whole rock chemistry, 321-376 Ma based on K/Ar ages of hornblende and biotite (Adams, 1987), and 40Ar/39Ar ages of 356, 356, and 346 Ma for hornblende, biotite, and

muscovite respectively (Richard et al., 1994). Metamorphic rocks within the Swanson formation of MBL have been dated using Rb-Sr whole rock analysis with an age of 421-432 Ma (Adams et al., 1995) and K/Ar dating has indicated this formation has undergone thermal resetting with the emplacement of mid-Cretaceous granites, and show cooling ages of 113-440 Ma (Adams et al.,

2005). This suggests only parts of the Swanson formation were thermally reset by the granitic intrusion. The Swanson formation has also been correlated with the Greenland Group of New

Zealand, interpreted from the lithologic similarities and metamorphic ages. Permian-Triassic magmatism has been identified in the Kohler Range (276 ± 2 Ma, Pankhurst et al., 1998).

The second possible region for the Mesozoic sediment from West Antarctica is from within the West Antarctica Rift System (WARS), which separates the MBL block from the

TAM. Little is known about the geology within the WARS because there are no outcrops above the ice surface. Magnetic and seismic studies show that subglacial rocks within the WARS are primarily interbedded volcanic and sedimentary strata, (Jankowski and Drewry, 1981; Bentley,

1991). Cenozoic basalts associated with rifting are present (Behrendt et al., 1994) and overlying the basalt are Oligocene or younger terrigenous and glacial–marine sediment, which may be as thick as 600 m (Rooney et al., 1991; Tulaczyk et al., 1998). It is suspected that the sources of the sediment within the WARS are from the uplifted blocks of West Antarctica (e.g. Marie Byrd

Land) and the TAM (Licht et al., 2005). This can be supported by the work of Farmer et al.

(2006) who have sampled onshore till within the WARS. Till samples from this region exhibit

εNd values of -7 to -9.9, which is consistent with values from tills found in the southern TAM

(Farmer et al., 2006). Samples from West Antarctica are within the known range of Nd and Sr isotopic compositions of Granite Harbor Intrusive rocks from the southern portions of the TAM

(Farmer et al., 2006). A source from these rocks, or sediments derived from these rocks that

infilled the West Antarctic rift zone, is also consistent with the petrography of the till coarse fractions which are dominated by quartz, feldspar, and sedimentary and felsic intrusive rock fragments (Licth et al., 2005; Farmer et al., 2006).

The 40Ar/39Ar ages for Site 270 can be divided into two major groups or source regions:

1) West Antarctica (90-360 Ma) and 2) Ross Orogeny/Transantarctic Mountains (420-520 Ma).

While the TAMs do contain Mesozoic rocks of the FLIP group, ages characteristic of the FLIP group (~176 Ma) are not found among the ages of the IRD at Site 270 because these mafic igneous rocks do not contain hornblende or biotite.

3.4 Discussion

3.4.1 IRD from East vs. West Antarctica

Despite the apparent contribution from both major regions of Antarctica, interpreting the age populations remains complex because of the surrounding geology. Ages younger than ~400

Ma are almost certainly sourced from rocks within West Antarctica. The primary peak however, is difficult to interpret because the potential sources of these rocks have a large spatial variability that extends the entire length of the TAM. While NVL and SVL can be eliminated as potential sources, it is difficult to determine if the Ross Orogeny aged IRD (~500 Ma) were sourced from the central or southern TAM. Rocks within the TAM that contain these aged grains include the

GHIS and Beacon Supergroup. The GHIS is a suite of granitic, intrusive igneous rocks that should contain an abundant amount of hornblende and biotite. This suite is more abundant in the southern TAM (Farmer et al., 2006), where they border West Antarctica and the WARS. The

Beacon Supergroup is a large group of sedimentary rocks spread throughout the TAM and were

deposited from the Devonian to Earliest Jurassic (e.g. Barrett, 1981). They unconformably lay

(~50 Myr gap) on top of the crystalline bedrock. Though these rocks were deposited after the

Ross Orogeny, they may contain recycled hornblende or biotite. Despite MBL containing rocks of Early Paleozoic age and being similar to the geology in NVL, there is no evidence for any rocks older than ~450 Ma.

For IRD to be delivered to Site 270 from the Central TAM, sea-level would need to be higher and/or subsidence within the Ross Sea would have had to occur fairly rapidly with respect to the modeled topography of the earliest Oligocene (Wilson et al., 2012). According to the topographic models, the Central High was above sea-level during the Early Oligocene, and extended to almost 83°S. If the Central High was still above sea-level, it would have deflected any icebergs from the Central TAM. If the basement highs were below sea-level during the Late

Oligocene, icebergs derived from the Central TAM may have travelled along the TAM coastline rather than moving into the open Ross Sea. This would require ice flow to be fairly balanced between East and West Antarctica, similar to what was determined for the last glacial maximum

(Licht et al., 2005). While Site 270 is in a position where it could record the convergence of ice coming from East and West Antarctica, the basement highs of the Ross Sea may not have subsided enough to allow this.

The sediment supplied to the Central Ross Sea was most likely heavily dominated by erosion of material from West Antarctica. The geomorphology and geography of the uplifted basement blocks within the Ross Sea help to support this. The ages of the IRD sediment can also be well resolved from a strictly West Antarctic source, including the southern TAM sediment transported through the WARS, similar to how the tills from Farmer et al (2006) were deposited.

West Antarctica contains or can transport sediment from rocks that can explain all of the

40Ar/39Ar ages seen in the IRD from Site 270. Additional techniques are needed to further support a dominant West Antarctic source, or to distinguish sources from central vs. southern

TAM, such as fine-grained sediment geochemistry.

3.4.2 Biotite vs. Hornblende

The distribution between the hornblende and biotite grains is remarkably similar (Figure

3-6). In contrast to what was seen at Site U1356, there does not appear to be a mineral bias in the sources of these grains. Both exhibit a strong peak indicative or Ross Orogeny related rocks.

While the biotites have more developed peaks in the 100-400 Ma range, the hornblendes only have a few clustered grains. Two grains at ~100 Ma, and 4 at ~300 Ma do correlate to the smaller peaks in the biotites, but there is an absence of grains in the ~200 Ma range. This is likely due to the low amount of hornblende grains analyzed compared to the biotites. Curiously one hornblende grain exhibited an age of ~700 Ma, which likely indicates a source from the interior of East Antarctica. Similar ages were found in the Central TAM by Grindley and McDougall

(1969) who came to a similar conclusion. The similarities of the biotite and hornblende age populations suggest they can be treated as one population.

10 Hornblende >150 um n=23

5 number

0 30 Biotite >150 um n=128 20

10 number

0 40 All Hornblende and Biotite 30 n=151 20

number 10 0 0 200 400 600 800 Age (Ma)

Figure 3-6 - Distribution of biotite and hornblende grains.

3.4.3 Downcore distribution of IRD

It is clear from the downcore distribution of IRD ages (Figure 3-7) that significant changes in the source rock contribution occur, which can be broadly described as a shift from a dominant West Antarctica/MBL source with a minor contribution from the central TAM to largely central TAM dominated. The number of grains analyzed for these samples are generally low (10-20 grains), which is not a robust signature, but these grain populations do provide us

100

with a general idea to the distribution of the IRD sources. It is unlikely the age populations would change significantly if more grains were analyzed, but the possibility does exist.

The following discussion places each interval into 1 of 3 primary ice flow regimes into the central Ross Sea (Figure 3-8). The first is a scenario in which ice flow and sediment delivery is dominated by outlet glaciers from MBL. The second scenario is a combination of ice flowing from MBL and from within the WARS. Under this scenario there is a somewhat balanced distribution between West Antarctic and TAM sourced grains. The third scenario includes ice flow and sediment from the WARS and central TAM to the central Ross Sea.

Beginning from the base of the recovered marine sediment, the first two intervals at 25.86 and 25.84 Ma (370.66 and 368.91 mbsf) show a dominant contribution of IRD sourced from

West Antarctica/MBL. While the former may have had a small contribution from the central

TAM, the latter is most likely entirely derived from MBL, with only 1 of 16 grains representing a contribution from the TAM. The first of these samples (25.86 Ma) is interpreted as a balanced input between West Antarctica and TAM sources. Because of this balance, most of the TAM grains are likely coming from the southern TAM via outlet glaciers that travel through the

WARS to the Ross Sea. If there was a balanced input of central TAM and West Antarctic derived ice, the TAM signal of the sediment should be somewhat greater than that of West

Antarctic ages. This initial sample fits into the second scenario.

101

West Antarctica/MBL TAM/Ross Orogeny

15 320.36 mbsf, 25.37 Ma 10 n=26 S3 5 number 0 10 324.81 mbsf, 25.42 Ma n=19 5 S3 number 0 5 331.01 mbsf, 25.48 Ma n=2 S2 number 0 5 339.66 mbsf, 25.56 Ma n=14 S2 number 0 5 343.46 mbsf, 25.60 Ma n=10 S3 number 0 10 348.26 mbsf, 25.64 Ma n=10 5 S3 number 0 5 352.96 mbsf, 25.69 Ma n=16 S1 number 0 5 358.21 mbsf, 25.74 Ma n=14 S2 number 0 0 200 400 600 800 Age (Ma)

Figure 3-7 - Downcore distribution of argon ages. S refers to corresponding scenario (see text).

102

West Antarctica/MBL TAM/Ross Orogeny 5 360.80 mbsf, 25.76 Ma n=17 S1 number 0

5 368.91 mbsf, 25.84 Ma n=11 S1 number 0

5 370.66 mbsf, 25.86 Ma n=12 S2 number 0 0 200 400 600 800 Age (Ma)

Figure 3-7 (continued)

The next two samples at 25.84 and 25.76 Ma are clearly dominated by a West

Antarctic/MBL source, suggesting the flow of ice out of West Antarctica was providing most of the sediment deposited into the central Ross Sea and fit into the first flow regime. This scenario for ice flow would be a regime that is primarily dominated by ice flow out of West Antarctica, similar to what has been interpreted for the Middle Miocene (Hauptvogel and Passchier, 2012) and could be described as a heavily dominant flow of outlet glaciers from MBL. Under this scenario, ice flow coming from the outlet glaciers from the TAM would be influenced by the ice coming from the WARS, pushing the TAM derived ice westward along the coastline. Here there would be no sedimentation from ice sourced in the central TAM at Site 270. Instead, any Ross aged grains would be sourced from the southern TAM through the WARS. The MBL grains could still be derived from outlet glaciers on the coast where icebergs come across the Ross Sea.

103

The possibility that the MBL aged grains could be coming from within the WARS still exists as well, but the lack of TAM aged grains would be difficult to explain.

Scenario 1 - MBL dominated Scenario 2 - WARS/MBL

southern TAM southern TAM

central TAM WARS WARS central TAM

MBL MBL

0 500 0 500 Scenario 3 - WARS/central TAM southern TAM 40Ar/39Ar 2000

WARS central TAM m 1500 2000

MBL 1000 1000 0 Age (Ma)

−1000 elevation 500 −2000

0 0 500

Figure 3-8 - Three scenarios for ice flow into the Ross Sea. Blue dashed line is inferred Oligocene ice sheet/shelf edge.

The next interval at 25.74 Ma (358.21 mbsf) exhibits a more even distribution amongst the grain populations, indicating slightly less input from MBL (8 grains) and an increased contribution from the central or southern TAM (6 grains). Since there are grains with a MBL signature, the Ross aged grains in this interval are mainly dominated by the southern TAM. This interval is interpreted as a return to flow regime 2, indicating icebergs out of the WARS and

104

MBL were likely being delivered into the central Ross Sea. A major shift does occur following this, to a dominantly West Antarctic aged source in the interval 352.96 mbsf (25.69 Ma), which can be best explained by our first scenario. This interval shows a very dominant West

Antarctic/MBL signal with only 1 of the 14 grains analyzed interpreted to have a likely source of

TAM. Here, outlet glaciers coming from West Antarctica (likely the southern MBL region) are the primary source for this sediment in the central Ross Sea with a small input from ice derived in the WARS (which explains the 1 Ross aged grain).With 14 grains analyzed for this interval, lack of TAM signal is a not likely a reflection of the number of grains analyzed

There is another dramatic shift seen in the following two intervals at 348.26 and 343.46 mbsf (25.64 and 50.60 Ma respectively) where the IRD age distribution is a dominantly Ross aged source, with only 2 of the 10 grains indicating a West Antarctica/MBL contribution. Here the distribution of grains suggests the TAM would be the source of most of the sediment delivered to the central Ross Sea. Based on the argon ages alone, it is difficult to interpret the type of flow regime of ice delivered to the Ross Sea when the IRD ages have a primary input from the TAM and a secondary West Antarctica/MBL source. There is clearly a West Antarctic input into the central Ross Sea, but the specific TAM input is not so apparent because these sediments could have come directly from the central TAM or from the southern TAM through the WARS. For the latter to be true, more of a contribution of West Antarctic aged sediment would be expected, which these two intervals do exhibit. Despite the difficulties in determining the provenance of the TAM sourced grains, it seems likely that a balanced ice sheet regime in which ice from the central TAM and from West Antarctica are both providing sediment into the central Ross Sea (scenario 3). This balanced ice flow regime in the central Ross Sea would be

105

similar to what was found for the last glacial maximum (Farmer et al., 2006; Anderson et al.,

2014).

The next interval at 339.66 mbsf (25.56 Ma) is a transition into sediment ages evenly distributed between West Antarctic/MBL and TAM sources (scenario 2). Similar to some previous intervals, this distribution likely represents ice flow and sediment to the central Ross

Sea out of the WARS and MBL. The next interval at 331.01 mbsf, 25.48 Ma) contains only 2 grains analyzed and as mentioned earlier, this likely represents a flow regime similar to the preceding sample. The interval at 324.81 mbsf (25.42 Ma) contains 11 grains representing a

TAM source and 8 grains that are best explained by West Antarctica/MBL sources. The interval also contains the hornblende grain that had an age ~700 Ma, which likely came from the Central

TAM. Argon ages older than ~450 Ma are not currently found in MBL, which is similar in geology to parts of NVL on the other side of the Ross Sea. The most likely scenario for this interval would be number 3, where East and West Antarctic ice converges towards the central

Ross Sea.

The last interval studied at 320.36 mbsf (25.37 Ma), also has a dominant TAM source.

Here only 6 of the 26 grains analyzed exhibit an age that can only be explained by a West

Antarctic/MBL source. This is interpreted to represent a central Ross Sea flow regime that is balanced between WARS and central TAM sourced ice streams (scenario 3).

An additional source for the MBL aged grains are from icebergs calving from the northern MBL coast, where they could have travelled across the Ross Sea to Site 270, similar to modern iceberg drift patterns (Gladstone et al., 2001; Stuart and Long, 2011). It is also important to remember that the age of the bedrock within the WARS is unclear, but it is clear that this region plays a definite role in the distribution of sediment ages and delivery at Site 270.

106

One important trend in the downcore distribution of these IRD grains is that the grains interpreted as being sourced from MBL exhibit a very similar distribution in every interval they are present. This would suggest most of the grains interpreted as MBL are likely being eroded from the same location, despite the large differences in age. The alternative to this would be that even though the sources of the MBL grains were some distance apart, ice was actively eroding them simultaneously, suggesting that each of these areas responded to cryospheric changes in a similar manner. Unfortunately the coastal geology of MBL is primarily defined by offshore

Holocene sediment geochronology and some onland geology ages, especially the northwestern region, but little is known about the rocks in the interior or the distribution of this geology underneath the ice sheet.

3.4.4 West Antarctic Ice Sheet during the Oligocene

The range of 40Ar/39Ar ages of the IRD are indicating that a large WAIS existed during the Late Oligocene. The paleotopographic models already suggest the presence of a larger than modern WAIS during the earliest Oligocene, mainly due to larger surface land area above sea- level (Wilson et al., 2012). Unfortunately, topographic models of Antarctica do not exist after the earliest Oligocene. Despite the interpretation of localized ice caps during the Late Oligocene

(Bart and DeSantis, 2012), our data show that the sand sized sediment has variable sources and that the basement highs of the Ross Sea may not be the likely sources. Our data can be interpreted to show that a large WAIS still existed in at least the Late Oligocene, as indicated by the significant contribution of IRD grain ages between 100 and 450 Ma. As discussed earlier, there are few lines of evidence to support a large WAIS during the Oligocene and there is no marine sedimentary evidence from the Ross Sea to suggest it. This chapter presents the first

107

geochemical evidence from the Ross Sea that clearly indicates a WAIS had to be reaching the coastline. The implications for this are quite large because it changes our understanding of West

Antarctic glaciation, Oligocene climate studies, and reconciles the larger than modern ice volume estimates from distal records (Pekar et al., 2006).

The elevations of the Ross Sea region and West Antarctica likely played a large role in supporting a large WAIS during the Oligocene, and perhaps larger than modern. Modeling results support a larger WAIS during the Early Oligocene (Wilson et al., 2012) and it is unlikely that West Antarctica assumed its present day appearance during the Oligocene because up to 3 km of material would need to be eroded by the Late Oligocene (Wilson and Luyendyk, 2009).

Even though subsidence and erosion may have been occurring since the development of the

WAIS, the elevations were still likely much higher than modern with more surface area above sea-level, allowing the larger growth and sustainment of a large WAIS. This then raises questions in regards to the history of subsidence and erosion in West Antarctica through the Late

Cenozoic. It is difficult to determine when the large volume of sediment was eroded and what the rate was, and the response of the lithosphere from the isostatic loading for this period is unknown (i.e. rate of subsidence). Miocene topography in Antarctica is currently being investigated (Doug Wilson, personal communication) and when available may be able to help constrain the rate of subsidence and erosion in West Antarctica during the Oligocene.

Our current data indicate IRD at Site 270 were derived from both East and West

Antarctic sources and the age range coincides with the observed Late Oligocene warming that begins at ~26 Ma, which is interpreted from Oligocene oxygen isotope records (e.g. Zachos et al., 2001; 2008). The downcore distribution of the 40Ar/39Ar ages indicates large provenance changes between MBL/West Antarctic and TAM, suggesting a dynamic glacial flow regime.

108

This could be a result of this observed warming, but these changes are occurring within a ~500 kyr period, making the Late Oligocene warming difficult to interpret from the current data.

Though the number of grains analyzed for each sample was generally between 10 and 20, it should be suitable for determining what the dominant signature in the ice-rafted sediment is, and therefore interpreting the dynamics of ice flow into the Ross Sea during the Oligocene.

While it is possible that an ice shelf developed in the Oligocene, it is difficult to speculate on the location of the grounding line. The grounding line was likely further south of Site 270 because there are very limited grains larger than 1 mm. If the grounding line was closer to the drill site, it should be reflected in the grains size because most glacial debris is deposited at the grounding line (Death et al., 2006). Additionally, ice-rafted debris accounts for a significantly less percentage of the sediment entrained in the ice (e.g. Powell 1984; Alley et al., 1989; Death,

2006). It is unlikely the ice sheet advanced to or covered Site 270 because the lithology of the core does not support this. Above the glauconite layer is a thick sequence of silty clay with scattered granules and pebbles. Typically if an ice sheet or ice shelf were overriding a drill site, more coarse sediment would be present in a fine-grained matrix like glacial diamict, similar to the ANDRILL SMS core in the western Ross Sea. Glacial scouring occurred during the Late

Oligocene, in the easternmost Ross Sea along the MBL coast (Sorlien et al., 2007), but there is no evidence to suggest such glacial erosion in the Ross Sea (Bart and DeSantis, 2011). With this in mind, the location of the ice shelf edge in the scenarios is a best estimate without any additional evidence.

It is interesting that these changes in IRD sources are occurring over such a short timescale. This may be an indication of a controlling mechanism on the production of icebergs in the Ross Sea region. One of the possibilities includes short-term orbital cyclicity like obliquity.

109

This is mere speculation however, because it is not possible to quantitatively correlate these.

Other factors that could explain the changes in our proposed ice flow regimes include changes in precipitation rates and locations, which is a factor ascribed to the currently increasing amount of ice along the Dronning Maud coastline (Shepherd et al., 2012). Additionally, this sediment was deposited during the Late Oligocene warming, which could also explain the drastic changes in the flow regimes and confluence and East and West Antarctic derived ice in the Ross Sea.

3.5 Conclusion

Site 270 provides an excellent record of ice sheet dynamics in the Ross Sea during the

Late Oligocene. The 40Ar/39Ar ages indicate the existence of a large West Antarctic ice sheet during a time when West Antarctic glaciation is not well constrained. The IRD age populations clearly indicate 2 primary sources for the grains; Ross aged grains (500 Ma) from the TAM and grains that clearly represent West Antarctica/MBL (100-450 Ma). Though it is difficult to interpret where in the TAM the grains are sourced from (i.e. central vs. southern), it does provide an important record of glacial erosion in the region. Three primary scenarios of glacial erosion in the Ross Sea and iceberg delivery into the central Ross Sea were proposed, including 1) dominantly MBL sourced IRD, 2) MBL/WARS sourced IRD and 3) WARS/central TAM sourced IRD. The findings of this study indicate the Ross Sea was quite dynamic in terms of glacial changes during the Late Oligocene. The new age model suggests the sediment intervals studied here only span ~500 kyr, which means that noticeable changes in the IRD deposition in the central Ross Sea are occurring at short time scales. This may represent evidence for orbitally paced changes in iceberg delivery (i.e. obliquity controlled), however this is only speculation.

The recovery of Site 270 is fairly poor, and no correlation can be made to orbital forcing, but

110

these sediments do indicate a large and highly dynamic WAIS from 26.0 to 25.5 Ma. This is in disagreement with the observed Late Oligocene warming from oxygen isotope records, which is explored thoroughly in the next chapter. This dataset and interpretations are the first geochemical evidence from the Ross Sea of a large WAIS during the Late Oligocene.

111

Chapter 4. Bottom water redistribution during the Late Oligocene: No

evidence for a global warming trend

Abstract

A 9-kyr resolution stable isotope record was developed from benthic foraminifera from

Ocean Drilling Program (ODP) Site 690B from the Maud Rise, Antarctica for the Late

Oligocene. This is the southernmost-deepest, Antarctic proximal, high-resolution record for this time interval. The δ18O record shows values consistent with a near or larger than modern sized

Antarctic ice sheet up to 155 %, in agreement with paleotopographic models and previous ice volume estimates from far-field studies. A strong 405-kyr and 100-kyr eccentricity signal is present, indicating orbital forcing was largely responsible for cryospheric changes. Additionally the δ18O record shows no indication of the Late Oligocene warming, suggesting the warming trend seen in other records is a reflection of changes in water mass distribution. The δ13C record indicates climate changes did not have a direct effect on ocean circulation, as evidenced by the dissimilarities between the δ18O and δ13C record trends.

4.1 Introduction

Most of what is known about the Late Oligocene cryosphere is derived from low-to-mid latitude, deep-sea isotopic and backstripped stratigraphic records with estimates of Antarctica ice volume between 50% (e.g. Zachos et al., 2001) to 130% of modern (Kominz and Pekar, 2001;

Lear et al., 2004; Pekar and Christie-Blick, 2008). Eustatic estimates from sequence stratigraphic records (Kominz and Pekar, 2001; Pekar and Christie-Blick, 2008) indicate repeated sea-level lowerings during the Late Oligocene, consistent with a heavily glaciated East Antarctic 112

continent. Oxygen isotope records of the Late Oligocene indicate a large warming after the Oi2b isotopic event (~26.7 Ma) that lasts until at least 24.5 Ma (e.g. Zachos et al., 2001), however

Pekar and Christie-Blick (2008) have shown this warming is not as prominent due to the geographic distribution of data between high and low latitude records, yet a warming trend exists in all records that span the Late Oligocene (e.g. Zachos et al., 2001; 2008; Palike et al., 2006).

Proximal isotopic records for the Late Oligocene are needed to properly evaluate the Antarctic cryosphere in terms of climactic and cryospheric changes.

Ocean Drilling Program (ODP) Site 690B presents an excellent opportunity to study the

Antarctic cryosphere during the Oligocene because it is one of two exceptionally well-recovered

Oligocene records (ODP Site 689 is the other). Site 690B was drilled during ODP Leg 113 and is located on the Maud Rise (65.16°S, 1.20°E), northeast of the Weddell Sea, about 400 km from the Dronning Maud coastline (Figure 4-1). The water depth at which this site was drilled was

2,914 m, the deepest Antarctic proximal drill site with well-preserved foraminifers suitable for isotopic analysis. Site 690B is ~834 m deeper than the popular Site 689 of the same leg and is located on the southwestern flank of the Maud Rise, making Site 690 a better register of deep water. Three holes were drilled at Site 690, with Site 690B successful at recovering Oligocene strata. Sediment recovered at Site 690B ranges in age from Recent to Late Paleocene. Overall recovery in this hole was 100 % and includes a Late Oligocene section. Hiatuses do exist in the

Early (~31-29 Ma) and latest Oligocene (~24.5-23 Ma), however the target age of sediment for this study was between 29 and 25 Ma to compliment the IRD work done at Site U1356 and Site

270.

113

−

10˚

10˚ 20˚

−60˚

Site 689

m 2000 −65˚ Site 690 1000 0

0 500 −1000 elevation −2000 −70˚

− −

20˚ 10˚ 0˚ 0˚

10˚ 20˚ 30˚

−3500 −64˚ − −4500 2500 4500 − Site 689 −3000 − 64˚

4000 − −

3500 − −2500 2000 2000 − −4000

3500 − Site 690 − −66˚ 3000 −66˚

0 100

−

2˚ 0˚ 2˚ 4˚ 6˚ 8˚ Figure 4-1 - Location map of ODP Site 690. Upper has BEDMAP2 topography overlaid (Fretwell et al., 2013).

The lithology of the Site 690B sediment from this study is mainly pelagic in origin with a mixture of some terrigenous material and has been described by the Shipboard Scientific Party

(1988) as follows. The lithostratigraphic units identified as Oligocene include Units IIA and IIB, which can be broadly described as diatom rich and foraminifer rich respectively. Unit IIA is

114

described as a diatom-bearing nannofossil ooze that accounted for 50% of the sediment. It extends from 24.4 to 53.4 mbsf. Other sediment types included olive brown mud-bearing diatom ooze; muddy diatom ooze; mud and radiolarian-bearing diatom ooze; light brownish gray and grayish brown nannofossil-bearing diatom ooze; gradational from white to brown nannofossil diatom ooze; and white and light gray diatom nannofossil ooze. The primary sediment included in lithostratigraphic Unit IIB is a nannofossil ooze that contains some diatom rich layers, but is distinguished from Unit IIA by an increase in the number of foraminifers. It is present from 53.4 to 92.9 mbsf. Secondary lithologies include light brownish gray diatom-bearing nannofossil ooze, white foraminifer and radiolarian-bearing nannofossil ooze, and additional similar lithologies. Units IIA and IIB contained sufficient species, numbers, and preservation of foraminifers for stable isotope analysis.

4.1.1 Geology of the Maud Rise

The Maud Rise (MR) is a seamount that is about 100 km in diameter and rises ~3,400 m above the sea floor, 1,600 m below the sea-surface (Figure 4-1 lower). The base of hole 690C drilled into the bedrock of the MR, which was found to be alkali basalt. This basalt is thought to be the result of hot spot activity prior to 84 Ma, which is the age of the oldest sediment found at

Site 690C (Schandl et al., 1990). Basalts are not uncommon around the perimeter of Antarctica.

Alkali basalts have been found previously, including Kerguelen Island (Watkins et al., 1974),

Kerguelen Plateau (Storey et al., 1988), the Antarctic Peninsula (Hole, 1988; Smellie, 1987);

Ross Island, Antarctica (i.e. the Erebus Volcanic Province) (e.g. Goldich et al., 1975), Bouvetoya

Island (Imsland et al., 1977) and P-type basalts from several localities at and near the Bouvet triple junction (LeRoex et al., 1983).

115

The MR is geologically interesting and significant for several reasons. Firstly, it is a

Cretaceous Quiet Zone anomaly (CQZ, Fullerton et al., 1994; Kim et al., 2005). The CQZ anomalies were created during a long span of normal geomagnetic polarity that lasted 35 Ma

(Fullerton et al., 1994). Such anomalies are usually visible by satellites as opposed to the narrow, alternating seafloor-spreading anomalies. These anomalies have been observed previously over ocean plateaus, rises, and subduction zones (Frey, 1985; Bradley and Frey, 1988; Toft and

Arkani-Hamed, 1992) and anomaly maxima usually correlate with plateaus and rises while anomaly minima correlate with basins (Frey, 1982; Hinze et al., 1991). The MR is thought to have been part of a large KQZ that extended across the southwest Indian Ocean from southern

Africa to the northeastern Weddell Sea (Purucker et al., 1998; 1999; Marks and Tikku, 2001) and separated from the Agulhas Plateau by 93 Ma (Martin and Hartnady, 1986; Fullerton et al., 1994;

Parsiegla et al., 2008; Figure 4-2). The analysis of these CQZs and specifically the MR, are important for reconstructions of plate tectonics (Kim et al., 2005).

Figure 4-2 - Plate-tectonic reconstructions of the Maud Rise from Parsiegla et al. (2008). (a) 100 Ma: The reconstructions of Agulhas Plateau (AP), Northeast Georgia Rise (NEGR), and Maud Rise (MR) (but with the recent boundaries). (b) 94 Ma: The formation of the whole large igneous province (LIP; AP, NEGR, MR) is complete. Subsequent spreading causes separation of the three fragments.

116

Another reason the MR is an important region is that much of the initial mixing of bottom waters that flow into the Weddell Sea from the Antarctic Circumpolar Current occurs near the

MR (i.e. mixing of Weddell Sea Warm Deep Water (WDW) and Circumpolar Deep Water

(CDW)) (Muench et al., 2001). The encroachment of these waters upon the MR causes a number of dynamic events relating to ocean transport. Firstly, the MR creates a quasi-stationary eddy that resides over it, partially trapping the water (Ou, 1991; Alverson and Owens, 1996). The water in this column is colder, less saline, and slightly denser than the surrounding waters causing it to block horizontal flow over the rise (Muench et al., 2001) (Figure 4-3). Southwestward flow becomes accelerated around the perimeter of the column, causing a closed circulation around the rise with the relatively warm and saline mixture of WDW and CDW. This closed circulation has been referred to a “halo” which has a complex structure with sharp fronts and multiple warm cores (Gordon and Huber, 1990; Bersch et al., 1992). The surrounding regional flow continues southwestward past the MR and contributes to a downstream region of water with elevated temperature that is situated southwest of the MR (Muench et al., 2001). These features impact local ocean processes and strongly affect the upper ocean heat fluxes and associated overlying air-sea exchanges (McPhee et al., 1996), sea-ice movement and production (Lindsay et al.,

2008), and recurring polynyas (open water surrounded by ice) (Comiso and Gordon, 1987). The effect of the MR on bottom water flow however, is yet to be determined. Regionally speaking, the MR can influence upper ocean heat and ice budget (Meunch et al., 2001).

117

Figure 4-3 - Lagrangian particle tracks from the Parallel Ocean Climate Model based on annual averaged mean flow for 1994 for the layer centered at 310m (from Muench et al., 2001). Open circles along 10°E show the initial positions of each particle.

4.1.2 Previous studies from the MR drill sites

A number of important, seminal studies have been published from the sedimentary records of Sites 689 and 690 that have furthered our understanding of the MR and the climactic and paleoceanographic history of the region. The recovery of Cretaceous sediment on the rise has led to an interpretation of a cooling trend during the Early Maestrichtian (74-70 Ma) (Barrera and Huber, 1990). Recovery of Paleogene sediments provides evidence for the Paleocene and

Eocene climactic events (e.g. Thomas, 1990; Pospichal and Wise, 1990; Kennett and Stott, 1990;

Bohaty and Zachos, 2003). Neodymium isotopes from the MR have been interpreted as the opening of the Drake Passage to intermediate depths at ~37 Ma (Scher and Martin, 2004).

118

Results from clay mineral studies from Site 689 indicate the Early Oligocene on the MR was a period of very high productivity whereas the Late Oligocene is a transition to lower productivity and an increased intensity of physical weathering on Antarctica, which is thought to be the result of ice sheet expansion (Diester-Haass et al., 1993). Clay mineral indices also have cyclic variations, suggesting 400-450 kyr variation (Diester-Haass et al., 1996). Additionally they suggest at least six periods of strong carbonate dissolution associated with low productivity levels during the Oligocene, and have been interpreted to be the result of incursions of cold, corrosive proto-Antarctic Bottom Water (Diester-Haas et al., 1996). Also at Site 689, a shift to temperate-water taxa in the calcareous nannofossil record during the Late Oligocene suggests an influx of warmer waters to the MR (Persico and Villa, 2004).

Delivery of terrigenous sediment to the MR has been found to be the result of bottom currents (Diester-Haass, 1991) rather than ice-rafting. This was interpreted from mica contents up to 10 times higher on the flank of the MR compared to the crest. Quantitative calcareous nannofossil assemblages indicate stable, cool water conditions persisted until the Late Oligocene where there was a decrease in the number of cool-water indicators (Persico and Villa, 2004).

Strontium isotopes from Site 689 provide evidence for the onset of Antarctic glaciation, increased mountain building in the Himalayan-Tibetan Plateau, and decreased hydrothermal activity (Mead and Hodell, 2010). High resolution magnetobiochronology has also been produced for portions of Sites 689 and 690, highlighting the fact that multiple offset holes need to be drilled to gain continuous records (Florindo and Roberts, 2004).

119

4.1.3 Stable Isotopes

As summarized by Katz et al. (2003), one of the most commonly used proxies for reconstructing past climates is using stable isotope measurements on benthic foraminifera (i.e.

δ18O and δ13C). Foraminifera build their tests (shell) by precipitating calcium carbonate, which incorporates the oxygen isotopic composition of the seawater, which is dependent on temperature and ice volume. Three isotopes of oxygen exist, of which 16O is by far the most common (99.759%), followed by 18O (0.204%), with 17O being the most rare (0.037%) (Rosman and Taylor, 1998). Of these isotopes, 16O and 18O are used for paleoclimate estimates based on their abundance.

Paleotemperature and ice volume estimates from δ18O are the result of isotopic fractionation (Urey, 1947 Epstein et al., 1953). Fractionation is due to the slightly different physiochemical properties each isotope has owing to different atomic masses, which results in differences in the velocities of isotopic molecules of the same compound (i.e., vibrational energy). The difference in vibrational energy is the dominant source of isotopic fraction. At high temperatures, the equilibrium constant for isotopic exchange tends to be similar, because small differences in mass are less important when all molecules have very high kinetic and vibrational energies. The ʻnormalʼ ratio of 18O/16O is about 400; therefore, variations of this ratio are usually multiplied by a large number (1000), so the values are small whole numbers. Oxygen isotope ratios (δ18O) are expressed as the difference from a standard (‰) (Equation 4-1).

Equation 4-1

18 As evaporation of surface water occurs, the H2O vapor becomes depleted in O compared to seawater. This isotopic fractionation during evaporation is temperature dependent.

120

This ʻlight waterʼ (water depleted in 18O) that is evaporated at low latitudes is transported into the higher latitudes. With each precipitation event, the remaining water vapor becomes more depleted. Isotopic fractionation that occurs during condensation of water vapor can be described by Rayleigh Distillation, where at a given percent of water vapor remaining in the atmosphere, the condensed liquid will be heavier than the remaining vapor (Pekar and Lear, 2009). This effect is especially significant over continents because the water vapor cannot get recharged with the lighter isotope from evaporation. At increasing latitude and altitude, the remaining vapors become increasingly enriched in 16O as more 18O precipitate out.

The isotopic composition of the foraminifera tests are routinely used to reconstruct past temperatures and circulation patterns in the deep oceans, as well as to infer glacial histories, but they are reliant on assumptions that these isotopes reflect environmental changes (Pekar and

Lear, 2009). The primary assumptions are the δ18O value recorded in the benthic foraminifera

(δ18O calcite) is a reflection of the ambient temperature and the δ18O value of seawater (δ18O seawater), and the δ13C value is primarily a function of the dissolved inorganic carbon (DIC)

δ13C value (e.g. Epstein et al., 1953; Emiliani, 1955; Graham et al., 1981; McCorkle et al., 1990).

Problems arise however, when trying to isolate temperature from ice volume in the δ18O signal, especially during older periods of time. To calculate temperature from δ18O values, the

δ18O of the seawater at the time the foraminifera lived must be known. One technique that has been used to calculate temperature from foraminifera is using the Mg/Ca ratio of their tests, which is temperature dependent (e.g. Martin et al., 2002; Lear et al., 2002). As temperature decreases below 5°C however, the Mg/Ca ratio changes very little, making temperature interpretations for cold water unclear. Additionally there are concerns about the carbonate ion saturation of the seawater, which acts as a secondary control on Mg/Ca ratios (Martin et al.,

121

2002; Elderfield et al., 2006). An attempt has been made to calculate the δ18O of seawater through the Oligocene by using Mg/Ca temperatures and δ18O from foraminifers to isolate the

δ18O value for seawater (Lear et al., 2004). Concerns exist however, owing to the difficulties and reliability of Mg/Ca in cold waters and older time periods (e.g. Elderfield et al., 2006).

One of the concerns with stable isotope studies is that δ18O and δ13C measurements of modern benthic foraminiferal species are often consistently offset from each other and from calcite precipitated in equilibrium with the surrounding water (e.g., Duplessy et al. , 1970;

Shackleton , 1974; Belanger et al. , 1981; Graham et al. , 1981; McCorkle et al. , 1990). These interspecies offsets have been attributed to microhabitat preferences (test growth in isotopically distinct environments) (e.g., Belanger et al., 1981; Corliss, 1985; McCorkle et al., 1990) and species-specific metabolic variation in isotopic fractionation (e.g., Duplessy et al., 1970;

McCorkle et al., 1990)

Many stable isotope records already exist for portions of the Oligocene, especially for the

Eocene-Oligocene transition, but fewer records have been developed that encompass the entire

Oligocene. Those that exist are mainly from lower latitude sites and/or are of low resolution (e.g.

Keigwin and Keller, 1984; Shackleton et al., 1984; Miller and Fairbanks, 1985; Miller et al.,

1987; Kennett and Stott, 1990; Mackenson and Ehrmann, 1992). These records and others have been commonly combined in an isotope stack to provide a complete Oligocene record and to interpret the global trend of climate (Miller et al., 1987; Zachos et al., 2001; 2008). While this isotope stack is valuable for the general trend, it has been found that large interbasin and latitudinal differences in Cenozoic stable isotope records exist, presumably reflecting multiple sources of deep water formation (Pekar et al., 2006). In their work they concluded the Late

122

Oligocene warming was much more pronounced in the combined records because of the geographic distribution of the stable isotope records.

Only one complete record exists for the Oligocene, from the equatorial Pacific (ODP Site

1218), and is high resolution throughout (4 kyr intervals, Lear et al., 2004; Palike et al., 2006).

Despite the distance of this record from the Antarctic continent it provides a more detailed record of Oligocene climate. The major findings from spectral analysis of the stable isotopes from this site indicate Oligocene climate was paced by 1.2 myr obliquity and 400 kyr eccentricity (Palike et al., 2006), which was previously interpreted from isotope maxima from earlier, lower- resolution records (Miller et al., 1991; Pekar and Miller, 1996). Additionally, Palike et al. (2006) also found that the onset of glaciation in the Oligocene was independent of the timing of CO2 reduction and was triggered by astronomical forcing once atmospheric CO2 levels decreased to a threshold value. This record also contained a prominent warming during the Late Oligocene.

Despite the high-resolution record of the Oligocene, Site 1218 is distal from the Antarctic continent and isotopic values could be influenced by more localized factors (i.e. mixing of different water masses). To provide a robust and reliable interpretation of cryospheric and climactic changes, higher resolution proximal records need to be developed around Antarctica.

Proximal Oligocene isotopic records that have been developed along the Antarctic continental margin are of low resolution (e.g. Kennett and Stott, 1990; Mackenson and Ehrmann, 1992).

Reasons for this are due to the low recovery of Oligocene sediments from Antarctica and the low number of sediment cores that contain the Oligocene. Even though Sites 689 and 690 had great

Oligocene recovery and improved age models since the original drilling (Florindo and Roberts,

2005), they have yet to published re-investigations with improved isotopic records.

123

4.2 Methods

4.2.1 Site 690 sampling and age model

For ODP Site 690B, 226 sediment samples were taken between 25.5 and 28.0 Ma at an approximate 9 kyr resolution (cores 7 and 8, 50-70 mbsf). The ages for the strata were derived from the paleomagnetic interpretations of Spiess (1990) and converted to the GTS 2012 timescale (Figure 4-4). While high-resolution magnetics were completed at Site 690C for some of the Oligocene section recovered (Florindo, 2005), it was difficult to transfer this data to Hole

B because the offset between the two holes seems to change downcore. The initial reports show the offset between the two holes at the start of drilling is 6 m, but the offset of the lithostratigraphic unit change in the Late Oligocene is only 3.7 m, suggesting that even though

Hole C was only 16 m away from Hole B, there is a some variation in the sediment supply.

Additionally while correlating the results to other stable isotope records or orbital forcing, it became increasingly difficult to explain data based on the Florindo (2005) age model. While that model is excellent for Site 690C, it is not ideal for Site 690B until the offset between the 2 holes is better understood. Therefore, the age model used for this study is based on the magnetostratigraphy of Speiss (1990), even though it is not as high-resolution as the Florindo

(2005) magnetics. Biostratigraphic points were rare for the Oligocene and there are two hiatuses in the sedimentary record in the Early Oligocene and latest Oligocene that are not found at Site

689.

The sediment recovery from this section of Site 690 was 100%. The 9 kyr resolution provides a low-resolution obliquity signal, ~4 data points for every cycle, and higher resolution data for longer orbital timescales. This resolution is not adequate for identifying any precessional

124

influence. A small section from ~26.3 to 26.0 Ma sees a drop in sedimentation rate from 7 m/myr to 2 m/myr, so this interval is of lower resolution.

50 Chron Depth Age C7ar 53.25 25.099 C8n1 54.75 25.264 C8r1 55 25.304 55 C8n2 60.11 25.987 C8r2 60.99 26.42 C9n1/2 68.28 27.439

70 24.5 25.0 25.5 26.0 26.5 27.0 27.5 28.0 Figure 4-4 - Age model for ODP Site 690B for the Late Oligocene. Paleomagnetic ties are from Speiss (1990).

4.2.2 Foraminifer methodology

Site 690B was the only drill site in this study with an abundant amount of well-preserved foraminifers. Similar sediment disaggregation methods as the IRD studies were used, but did not include further steps after the initial soaking in a sodium metaphosphate solution. This was due to the ease of disaggregation of the sediment so that no further means were needed. Additionally, further sediment handling runs the risk of breaking foraminifers’ tests. These samples were also washed and sieved to obtain sediments sizes >150 µm. Benthic foraminifera were identified under a binocular microscope and prepared for stable isotope analysis (δ18O, δ13C).

125

The most abundant species was Nuttalides umbonifera, followed by Cibicidoides spp.

(mainly C. mundulus) and Oridorsalis umbonifera. Isotopic analyses of modern benthic foraminifera show that epifaunal forms such as Nuttalides spp. and Cibicidoides spp. more closely reflect seawater δ13C values, whereas infaunal taxa such as Bulimina spp. and Uvigerina spp. reflect the lower δ13C values of the pore water (McCorkle et al., 1990). While Cibicidoides spp. are common for stable isotope analysis, their lower abundance at Site 690B did not allow for a mono-specific study of the genus. Instead, this study mainly used N. umbonifera for stable isotope analyses with minor amounts of Cibicidoides. spp. and O. umbonifera when needed.

Once identified, samples were hand-picked and weighed on a Thermo Orion Cahn C-35 microbalance, which was calibrated with a 200 mg weight. The range of foraminifers’ weights used for stable isotope analysis were primarily 100-150 µg and in some instances 70-100 µg. The foraminifers were then sonicated in distilled water for 5 seconds to remove any remaining fine- grained particles. The weight of any silt particles removed at this stage would be negligible compared to the overall weights of the samples and associated standards. Visual inspections ensured the sonicating process did not break the foraminifera. The foraminifers were loaded into glass tubes and analyzed at Columbia University’s Lamont-Doherty Earth Observatory. I was not involved in the process of measuring the samples.

4.2.3 Calibration and Standardization of Foraminifera

The N. umbonifera have been used previously and calibrated to the Pee Dee Belemnite standard (PDB) with an offset of +0.4‰ (Shackleton et al., 1984; Zachos et al., 1992).

Similarly, calibrations for Cibicidoides spp. and O. umbonifera are also well known. Despite the research on these species calibrations, interspecies offsets have likely changed throughout the

126

Cenozoic as a result of evolution or as an artifact of statistical methods and data set sizes used to determine the offsets (Katz et al., 2003), making it difficult to compare isotopic values for different species in older time periods even when calibrating to PDB.

Katz et al. (2003) provided Early Cenozoic correction factors for some common benthic foraminifera, including N. truempyi, Cibicidoides spp, and Oridorsalis spp. The Katz et al.

(2003) study was from the Paleocene and Eocene Epochs and they cautioned the use of their correction factors for other time periods. I have found however, that the oxygen isotope correction factors work quite well for the Oligocene and that their correction factor for N. truempyi works well for N. umbonifera. Similarly, the PDB correction factor is the same among different Nuttalides spp, suggesting minimal offset between the two. Additionally, values of N. umbonifera and Cibicidoides spp. measured in the same interval were more similar using the

Katz et al. (2003) correction in place of converting directly to PDB. Since the calibration to PDB is difficult to resolve during the Oligocene, oxygen isotopic values from N. umbonifera and O. umbonifera were converted to Cibcidoides spp. values using the following equations from Katz et al. (2003)

(Nutt + 0.10)/0.89 = Cib Equation 4-2

Orid - 0.28 = Cib Equation 4-3

where “Nutt” is the raw oxygen isotopic value for Nuttalides spp. and “Orid” is the value for

Oridorsalis spp. All raw values and corrected values can be found in Appendix 4-1.

The calibrations from Katz et al. (2003) for δ13C however, were not suitable for the

Oligocene because values were consistently offset when converted to Cibicidoides spp. values.

127

Instead, calibrations were developed based on the values from this study. Since Cibicidoides spp. fractionate carbon isotopes in their tests in equilibrium with calcite (e.g. Katz et al., 2003), a correction factor for N. umbonifera and O. umbonifera can be calculated from samples where either of those species were analyzed with Cibicidoides spp. Of all the samples analyzed, 11 had both N. umbonifera and Cibicidoides spp. and the average offset was about 0.104 ‰ for δ13C.

For O. umbonifera only 2 samples contained Cibicidoides spp. and had an offset of about 1.1 ‰.

In an attempt to make this value more robust, samples containing O. umbonifera and N. umbonifera were used. For these 4 additional samples, the δ13C value of N. umbonifera was converted to Cibicidoides spp. (by adding our new offset of 0.104 ‰) and the offset for O. umbonifera was determined in reference to that value. The result of the 6 samples of O. umbonifera with Cibicidoides spp. or N. umbonifera is an offset of +1.08 ‰, which is very similar to original value of the first 2 O. umbonifera samples. The equations used to convert δ13C values to Cibicidoides spp. are

Nutt + 0.104 = Cib Equation 4-4

Orid + 1.08 = Cib Equation 4-5

where “Nutt” and “Orid” refer to the respected raw carbon isotopic values. Converting all of the foraminifera analyzed to the Cibicidoides spp. values creates a much more accurate record.

4.3 Results

A total of 332 foraminifera samples from 282 intervals were analyzed for stable isotopes, including replicates and multiple species samples. The raw data can be found in Appendix 4-1

128

and is displayed in Figure 4-5. Overall, the isotope values obtained in this study look to be reliable and compare well with what was originally obtained in the low-resolution study of 690B

(Kennett and Stott, 1990; Mackensen and Ehrmann, 1992), which are included in all figures and interpretations. About 8.5 % (24 samples) of the samples analyzed had replicates run to check for precision with 18 samples having duplicates and 4 having triplicates. Most of the replicates were

N. umbonifera with 1 set of replicates from C. spp. The offsets of the δ18O replicates varied from as low as 0.01 up to 0.63 ‰, with the average offset of the replicates at 0.15 ‰. The precision of the samples was good, with twelve of the replicates having an offset of < 0.1 ‰, 7 replicates between 0.1 and 0.25 ‰, and 5 were > 0.25 ‰ (Table 4-1). This suggests the values measured are reliable to interpret paleo-conditions. The offsets in the δ13C values were similarly good, with an average of 0.072 ‰. Large excursions in excess of ± 0.5 ‰ δ18O between samples were determined to be outliers and were removed. While they are not presented in the figures, the values are included in Appendix 4-1.

Using the equations from Katz et al. (2003) proved to be the best way to standardize the values between the three genera analyzed. As a check, raw isotope values were also calibrated straight to PDB using their known correction values to see if there would be any advantage over the Katz et al. (2003) conversion method (Appendix 4-1). Samples in which N. umbonifera and

Cibicidoides spp. were analyzed had values most similar to each other using the Katz et al.

(2003) conversion method (converting isotope values to Cibicidoides spp. first, rather than straight to PDB; e.g. samples 52.23, 55.53, and 68.49 mbsf). A similar conclusion can be made when using O. umbonifera and Cibicidoides spp. or N. umbonifera in the same samples. This indicates the methods used by Katz et al. (2003) during the Paleocene and Eocene can also work for Oligocene benthic foraminifera and that the conversion factor for N. truempyi works well for

129

N. umbonifera. Once all values were converted to Cibicidoides spp., they were then corrected to

PDB using the standard of +0.64 ‰ (Figure 4-6). Intervals that contained replicates or had multiple genera analyzed had their values averaged together.

60 Depth (mbsf)

70 0123−10 1 18 13 δ O (‰) δ C (‰)

Figure 4-5 - Raw isotope for Site 690B relative to PDB.

130

Table 4-1 – Stable isotope precision and offsets (see text for details).

Oxygen Isotopes Carbon isotopes Depth Species Sample R1 R 2 Offset Sample R1 R 2 Offset 52.96 N. umbonifera 2.31 2.29 0.02 0.558 0.507 0.051 54.11 N. umbonifera 2.38 2.35 0.04 0.563 0.572 0.009 54.99 N. umbonifera 2.05 1.88 2.12 0.24 0.161 0.161 0.273 0.112 55.53 N. umbonifera 2.15 1.83 0.33 0.476 0.532 0.056 56.43 N. umbonifera 2.33 2.39 2.36 0.06 0.776 0.817 0.751 0.066 57.16 N. umbonifera 2.14 2.18 0.04 0.666 0.718 0.052 57.58 N. umbonifera 2.33 1.71 0.62 0.784 0.746 0.038 57.76 N. umbonifera 2.24 2.18 0.06 0.7 0.763 0.063 58.73 N. umbonifera 2.02 2.16 0.14 0.768 0.806 0.038 59.71 N. umbonifera 2.14 2.13 2.26 0.13 0.602 0.601 0.572 0.030 60.94 C. mundulus 2.61 2.59 0.01 0.719 0.735 0.016 61.89 N. umbonifera 2.39 2.23 0.16 0.618 0.647 0.029 62.09 N. umbonifera 2.38 2.24 0.14 0.672 0.63 0.042 62.55 N. umbonifera 2.39 2.48 0.09 0.622 0.657 0.035 62.85 N. umbonifera 2.06 2.69 0.63 0.589 0.667 0.078 64.05 N. umbonifera 2.39 2.17 0.22 0.429 0.281 0.148 64.23 N. umbonifera 2.41 2.50 0.09 0.393 0.49 0.097 64.83 N. umbonifera 2.34 2.36 2.38 0.02 0.67 0.559 0.563 0.111 65.6 N. umbonifera 2.63 2.66 0.03 0.418 0.614 0.196 65.97 N. umbonifera 1.98 2.09 0.11 0.066 0.141 0.075 66.89 N. umbonifera 2.57 2.29 0.28 0.615 0.452 0.163 67.04 N. umbonifera 2.16 2.20 0.04 0.315 0.434 0.119 67.24 N. umbonifera 2.39 2.42 0.03 0.103 0.17 0.067 68.49 N. umbonifera 2.20 2.09 0.12 0.503 0.457 0.046

Average 0.15 0.072

131

Phase III

26 Age (Ma) Phase II

Phase I

2.5 3.0 3.5 0 0.5 1.0 1.5 18 13 δ O (‰) δ C (‰)

Figure 4-6 - Cibicidoides spp. corrected values (see text for calibrations and calculations).

It has previously been shown that δ18O values can be used to back out apparent sea-level

(ASL) changes (Pekar et al. 2002). Calibrations to the ASL changes have also been made to interpret approximate total ice volume relative to modern (Figure 4-7). The Pekar et al. (2002) study created calibrations for ice volume for each drill site analyzed, which included Site 690B.

Different calibrations for each location were created using oxygen isotope amplitudes for Oi isotopic events. These amplitudes are the difference between maximum value of the δ18O event

132

and preceding minimum (or average minimum) δ18O. The approximate sea-level equivalent in total Antarctic ice volume used for this Site 690B study is ~66m (Wilson et al., 2013).

4.4 Discussion

The δ18O record from Site 690B is an excellent register of the state of the Antarctic cryosphere during the Late Oligocene. For this discussion all values will be in reference to the

Cibicidoides spp. converted values (i.e. Figure 4-6). It is clear from this new record that δ18O values describe a largely glaciated Antarctic continent and that a prominent warming seen in other records during the Late Oligocene is absent.

4.4.1 Stable oxygen and carbon isotopes

The stable isotope records can be separated into 3 distinct phases for the Late Oligocene

(Figure 4-6). Phase I can be defined by the large amplitude changes in the δ18O record and slightly decreasing trend in the δ13C from 27.6-27.0 Ma. The large variation in the δ18O values, up to 0.5 ‰, may be real a real signal because this portion of the record occurs between the Oi2a

(27.9 Ma) and Oi2b (26.7 Ma) isotopic events, which was a dynamic time between two large glacial intervals. Based on the age of this section, it is likely the Oi2a (~27.9 Ma) isotopic event is not recorded in this record. If there are large melt pulses during this time as the ice sheet begins the shrink, the additional freshwater could have influenced the production of Antarctic bottom water which would lower δ18O values. The rate at which these values change is also suggestive of short-term orbital forcing (i.e. 40 kyr obliquity) affecting the flux of meltwater.

These large amplitude changes are also seen in the record from Site 1218 (Figure 4-7) from the

133

equatorial Pacific and on the continental slope off the coast of North Carolina for the same interval.

25 25

26 26 Age (Ma)

690 689 27 744 27 1218 ASP-5

0.5 1.0 1.5 2.0 2.5 3.0 3.5 −0.5 0.0 0.5 1.0 1.5 18 13 δ O (‰) δ C (‰)

Figure 4-7 - Comparison of stable isotope data from Site 690B to other records. ODP Site 689 from the MR (Zachos et al., 2001 from Kennett and Stott, 1990), ODP Site 744 from the Kerguelen Plateau (Zachos et al., 2001 from Barrera and Huber, 1991), ODP Site 1218 (Palike et al., 2006), ASP-5 from the North Carolina continental slope (Katz et al., 2011).

There is no correlation between these amplitudes and the species of foraminifera used to generate the δ18O value and most of these large changes occur between samples of N. umbonifera. Additionally, it has already been shown that converting N. umbonifera values to

134

Cibicidoides spp. values is a reliable way to interpret this record. As indicated in the methods, the foraminifera selected for analysis were as pristine as possible, but it is possible some alteration or recrystallization could have taken place, which can also lead to variance in the data.

Initially, the δ13C signal has a slight trend toward lighter values to ~27.3 Ma, but then is similar to the δ18O. The trend of these values is opposite to what has been observed by Palike et al. (2006) in the equatorial Pacific, where δ13C trends towards heavier values during the same interval. It is clear that δ13C is not necessarily a global signal and is influenced by water mass distribution. The slight trend in the δ13C toward lighter values may be explained by the warming taking place after Oi2a, and suggests a weakening in AABW and a lowering of productivity.

This negative trend is not seen in the δ18O data, which could suggest a divergence in ice volume and ocean circulation. It is difficult to further interpret the relationship between δ18O and δ13C for this phase without extending the record to an older age.

Phase II of the stable isotope record has been assigned to the interval 27.0-26.0 Ma, which contains the Oligocene δ18O maximum at 3.59 ‰ (value from Kennett and Stott, 1990).

During this interval δ18O values primarily remain well over 3.0 ‰, indicative of a near or greater than modern ice sheet present in Antarctica. The current age model places the Oi2b event at

26.76 Ma, in agreement with what has been found elsewhere (e.g. Pekar and Miller, 1996; Palike et al., 2006). This large δ18O excursion (Oi2b) is not reflected in the δ13C record though. Instead, there is a gradual increase in values from 0.5 to 0.9 ‰. This suggests bottom waters on the southwestern flank of the MR were not immediately affected by the glacial maximum.

The interval from 26.0-24.7 Ma comprises Phase III of the Site 690B stable isotope record and should include the Late Oligocene warming. This phase shows δ18O and δ13C becoming more in sync, however it does appear the δ13C signal is slightly lagging with respect to

135

the δ18O values at 25.3 Ma. More importantly, the δ18O values vary fairly consistently between

3.2 and 2.6 ‰, suggesting the ice sheet is relatively stable through this interval. Other δ18O records show trends toward lighter values (Figure 4-7), which has been interpreted as the Late

Oligocene warming, but this trend is not present at Site 690B. Rather the δ18O record suggests the ice sheet in Antarctica consistently grew to modern size or larger, as indicated by values above 3.0 ‰. There are decreases in the δ18O record to as low as 2.5 ‰, but these are in line with orbital forcing and do not represent a larger trend. The δ13C record more closely matches

δ18O during this phase, but there are some distinct differences. The overall trend is towards lighter values, indicating that even though the climate was fairly stable, changes in ocean circulation or productivity were still occurring. The δ13C record also seems to have a distinct

405-kyr cyclicity in the data.

The isotopic data at Site 690B appears to be paced by orbital cyclicity, especially the

405-kyr eccentricity (Figure 4-8), which was also found at Site 1218 from the equatorial Pacific

(Palike et al., 2006). The comparison also suggests 100-kyr eccentricity cycles are playing a large role. The resolution of this record is approximately 9-kyr, which is robust enough for interpretations of eccentricity. It is more than likely that an obliquity signal is also present in this record. With this high resolution, it should be possible to orbitally tune the record of Site 690B to short-term eccentricity (not done during this study).

136

100-kyr 0.07 0.00

Phase III 25

Age (Ma) Phase II

Phase I

2.5 3.0 3.5 0.8 0.0 -0.8 18 δ O (‰) 405-kyr

Figure 4-8 - Oxygen isotope data against short and long eccentricity cycles (Analyseries from Laskar et al., 2004). Towards the right indicates less seasonality and cooler conditions.

4.4.2 Late Oligocene Apparent Sea Level and Ice Volume

Ice volume estimates of the Late Oligocene can be estimated from the δ18O record from

Site 690B using the calibrations from Pekar et al. (2006). The modern Cibicidoides spp. isotopic value in cold bottom waters (2°C) is 2.7‰ (Shackleton and Kennett, 1975), which is 3.34 ‰

137

(+0.64 ‰) when adjusted to equilibrium with calcite. Of that value, the present-day contribution of global ice volume is ~ 1.0 ‰ (Williams and Ferrigno, 1999), where 0.91‰ is attributed to all

Antarctic ice, 0.07‰ attributed to Greenland, and the remaining 0.02‰ to other sources. While it may have been possible to have northern hemisphere glaciation during the Oligocene, there are few pre-Middle Miocene records from the high northern latitudes because of poor age control, low sedimentation rates, and widespread hiatuses (Eldrett et al., 2007). A few recent studies suggest the presence of ice rafted debris during the Eocene and Early Oligocene (e.g. Tripati et al., 2005; Eldrett et al., 2007), but these findings are not supported by marine isotope records

(Edgar et al., 2007). Even as the existence of an ice sheet on Greenland during the Oligocene is debated, the isotopic signal from its ice is less than 0.1‰. For this study, it is assumed there is no ice sheet on Greenland, which would reduce the overall isotopic value of the Oligocene from

3.34‰ to 3.27‰.

Additionally if a polythermal ice sheet existed in Antarctica during the Oligocene, it could further reduce the isotopic value by another 0.25 ‰, bringing the isotopic contribution of global ice to 3.02 ‰ (Pekar et al., 2006). The isotope record from Site 690B however, does not indicate a polythermal ice sheet, but more likely a cold based one. If a dry-based ice sheet did exist during the Oligocene, then the isotopic contribution of the Antarctica ice sheets would be

3.27 ‰. It is likely a combination of these two thermal regimes existed, so they provide the maximum and minimum constraints for the isotopic signal of the ice sheet. This means that a fully glaciated Antarctica during the Oligocene would be reflected in the calcite corrected isotopic record of Cibicidoides spp. by a value of 3.02 ‰ for a wet-based ice sheet or 3.27 ‰ for a dry-based ice sheet. Bottom water temperatures are thought to vary very little in Antarctica

138

during the Oligocene (Pekar et al., 2002; Pekar et al., 2006), so it is assumed that any variation in the isotopic record at Site 690 is mainly attributed to ice volume changes in Antarctica.

Using the interpreted Oligocene equivalent of modern ice volume δ18O value 3.27 ‰ or

3.02 ‰ (dry vs. wet), apparent sea-level (ASL) changes can be calculated (Figure 4-9). Pekar et al. (2006) have provided ASL calibrations for several marine drill sites based on linear relationships in δ18O and ASL amplitude changes, which include a temperature component. For

Site 690B they calculated a 0.12 ‰ change in δ18O was equivalent to 10 m of ASL change.

Assuming that Antarctic bottom waters change very little, the ASL calibration for Site 690B is

18 18 (δ OMIS-δ OCib)/0.012=ASL Equation 4-6

18 where δ OMIS is the interpreted isotopic signal of a modern-sized ice sheet for the Oligocene

18 (3.02 ‰ or 3.27 ‰, depending on the thermal regime of the ice sheet), δ OCib is the value of

Cibicidoides spp. analyzed after adding 0.64 ‰ to correct for disequilibrium with calcite, and

0.012 is the slope of the line for Site 690 that relates the δ18O amplitude change between Oi events and their preceding minima with known ASL changes on from the New Jersey continental margin (Pekar et al. 2002; Pekar et al. 2006). Simply this calibration for Site 690 indicates a change in δ18O of 0.12 ‰ is equivalent to an ASL change of 10 m. The difference between wet- based and dry-based is 20.83 m, meaning the best estimate has an error of ± 10.4 m. The ASL estimates are presented in Figure 4-10. The ASL record shows Antarctic ice volume primarily remained at or below modern during the Late Oligocene with periodic excursions to lower than modern and a steady, large ice sheet from 27.0 to 26.0 Ma.

139

26 Age (Ma)

−50 −40 −30 −20 −10 0 10 20 30 40 50 60 70 Lower than modernASL (m) Higher than modern Modern = 66 m

Figure 4-9 - Apparent sea-level based on δ18O values from Figure 4-6 using the calculations of Pekar et al. (2006), see text for details. 0 ASL is a modern sized ice sheet (equivalent to 66 m [Wilson et al., 2013]). Blue indicates ice volume larger than modern, red indicates ice volume less than modern.

Ice volume estimates were calculated from the ASL based on the modern Antarctic ice volume equivalent of 66 m (Wilson et al., 2013).

(66-ASL)/66*100 = % of modern ice volume Equation 4-7

140

In this equation, ASL is the calculated value from Equation 4-6. The ice volume estimates using the calibration from Pekar et al. (2006) are presented in Figure 4-10 and include the maximum and minimum estimates. Since they are directly derived from the δ18O, interpretations about climate and changes seen in the ice volume record are the same as the previous section for

Cibicidoides spp. corrected isotopic values (Figure 4-6).

26 Age (Ma)

0 10 20 30 40 50 60 70 80 90 100110120130140150160170 % Present Day AIS

Figure 4-10 - Ice volume estimates based on the ASL values from Figure 4-9 using the calculation from Pekar et al. (2006).

141

The ice-volume estimates however, can add to the interpretations of the isotopic values by placing constraints on Late Oligocene ice volume. The ice-volume estimates range from a maximum value of ~155 % and minimum of 20 % of modern. The uncertainty in ice volume is due to the uncertainty in wet vs. dry based ice sheet so the best estimate has an acceptable range of ±15.7 % of modern, or ±10.4 m ASL. It is also possible there is an additional temperature signal not accounted for in the Pekar et al. (2006) calibration, but that effect would be small. The

155 % of modern value is very large and coincides with the Oi2b glacial event (~26.7 Ma).

Though this estimate is quite large, Early Oligocene ice volume has been interpreted to be greater than 140 % of modern (Wilson et al, 2013), adding support to the large ice volume estimates for portions of the Late Oligocene.

The lower portion of this record from ~27.6 to 27.0 Ma (Phase I) has large variations in estimated ice volume, which are derived from large variations in the δ18O values for this interval.

Best estimate ice-volume here ranges from as high as 120 % during glacial maxima to as low as

20 % during minima, which is quite large. As mentioned in the previous section, this interval is a dynamics time between two large isotopic events (Oi2a and Oi2b), and the ice volume estimates suggest the ice sheets in Antarctica were highly sensitive to climate changes.

During Phase II ice volume estimates consistently remain high, near or larger than modern. This interval includes the Oi2b isotopic event, which has been suggested as the period of largest ice volume during the Oligocene, larger than initial glaciation at the Eocene-Oligocene boundary. The best estimate suggests ice volume was up to 155 % of modern. For Antarctica to contain that much ice, West Antarctica would have to be heavily glaciated and above sea-level

(Wilson et al., 2013). Following the Oi2b event, ice volume estimates are somewhat lower, but fluctuate above and below modern size. Phase III is from 26.0 to 24.7 Ma where ice volume is

142

mainly below modern, but approaches 110 % during cooling periods. As discussed earlier, this portion of the record correlates very well with the 405 kyr eccentricity (Figure 4-8). The low ice volumes pairs well with intervals of more elliptical orbit, which promotes a higher seasonality.

Due to the seasonal variation in temperature, ice sheets are unable to expand because the warmer summer temperature is melting more ice than is being created during the winter.

4.4.3 Late Oligocene Warming?

The most important part of the stable isotope and ice volume records from Site 690B is the lack of Late Oligocene warming. The Late Oligocene warming is seen in many oxygen isotope records of the Late Oligocene (e.g. Site 1218 [Palike et al., 2006], Site 689 [Kennett and

Stott, 1990], Site 744 [Barrera and Huber, 1991]), however this event is unclear in the δ18O record at Site 690B. In the other records there is a clear trend in the δ18O values toward lighter values after the Oi2b glacial maximum event (~26.6 Ma in most isotope records) until the increase just before the Oligocene-Miocene boundary (Figure 4-7). As described earlier, the oxygen isotope stack (e.g. Zachos et al., 2001) shows a large decrease in δ18O values for the Late

Oligocene, which was explained by Pekar et al. (2006) as an artifact of the geographic distribution of the study sites spliced together. Minimum ice volume estimates for this warming are placed at ~40-50% (Pekar et al., 2006) suggesting the Late Oligocene warming was not as pronounced as originally thought.

In the δ18O record at Site 690B, values after the Oi2b event remain fairly consistent right around 3.0 ‰, which can be interpreted as the isotopic composition reflecting near modern ice volume size for the Late Oligocene (70-90 %). The record does show several drops in ice volume to below 50 %, but this is not part of an overall warming trend. These drops seem to be directly

143

related to orbital parameters. The original, minimum estimates of ice volume for this interval by

Pekar et al. (2006) are reasonably accurate, but this study shows the Late Oligocene warming was not an interval of gradual global warming at all. The only warming that can be identified is the transition from Phase II to Phase III at 26.0 Ma, which was an abrupt transition and not a long-term trend. It is more likely the warming trend seen by the other isotopic records is a result of ocean circulation changes, shifting the distribution of water masses.

Site 689 seems to exhibit this warming even though it was also drilled on the Maud Rise.

This could be explained by the water depth at Site 690B being deeper by almost 1,000 m. Site

689 is on the northeastern flank of the MR and may have been affected by the halo and trapping of waters caused by the Maud Rise (See section 4.1.#). It is also possible that this warming may have occurred in Antarctica after 24.7 Ma, which is not recorded in the strata at Site 690B, but the other marine isotope records indicate this warming should have already been happening prior to 24.7 Ma. If this event was not recorded in the current strata, it would have to have happened in a ~200 kyr period because isotope records shift toward more positive values following 24.5

Ma, at least in lower latitude records. This is highly unlikely as this additional 200 kyr period is likely another warming interval related to orbital parameters, rather than a large deglaciation.

4.4.4 Oligocene bottom water temperatures

Since the isotopic changes in the Site 690B record can be attributed to ice volume and there is no overall warming trend during the Late Oligocene, the observed warming in other isotopic records must be derived from changes in bottom water temperature. The ice volume contribution should be seen globally, but temperature changes can vary significantly depending on location, water depth, and which bottom water masses are bathing the drill site. This would

144

suggest a warming of bottom water masses was occurring outside of the deep waters proximal to

Antarctica. Based on the Site 690B oxygen isotopes, Antarctic bottom water remained cold and therefore are not responsible for the observed warming in other records. Rather, a redistribution of bottom water to the lower latitudes can best explain the observed Late Oligocene warming.

What caused the redistribution of bottom water is more difficult to explain. One explanation could be the initiation of deep-water flow in the Antarctic Circumpolar Current

(ACC), which began deep-water flow during the Late Oligocene (Lyle et al., 2007). This could have restricted the flow of AABW to the lower latitudes, allowing a higher flux of warmer proto

North Atlantic Deep Water (NADW) to the lower latitudes and reducing the δ18O values. Such a scenario was interpreted from isotopic records in the Atlantic (Katz et al., 2011) for the Late

Oligocene. The increase in lower latitude isotopic values through the Oligocene-Miocene boundary is likely a reflection of cooling in the proto NADW related to the glacial event over this boundary. Also occurring during the Late Oligocene was a reorganization of the

Mediterranean and Paratethys seaway (Rogl, 1999). A release of warm saline waters from this region could help explain the patterns seen in the Atlantic lower latitude records.

While it can be determined that most of the observed decrease in δ18O records for the

Late Oligocene is primarily caused by bottom water temperature changes, it remains difficult to calculate water temperatures for the Oligocene because the SMOW value cannot be measured.

The Oligocene SMOW values have been estimated based on Mg/Ca records and δ18O (Lear et al., 2004), however concerns arise over the validity of using Mg/Ca in sediment older than ~15

Ma (reference) due to carbonate ion saturation described earlier.

145

4.5 Conclusion

The new δ18O and δ13C benthic foraminiferal dataset from ODP Site 690B has changed the understanding of the Late Oligocene in Antarctica. Perhaps the most important finding is that this new record shows that while the Late Oligocene was a dynamic time for Antarctica where ice volume varied from less than 50 % up to as high as 155 % of modern, the δ18O values after the Oi2b event do not trend toward lighter values as seen in other records. This indicates the observed Late Oligocene warming in other records is most likely due to bottom water temperature changes. A trend toward lighter δ18O values seen in other records even though ice volume is steadily fluctuating can only be explained by ocean current changes and water mass distribution. During the time of this observed warming, the ACC has been suspected of deepening, which would restrict AABW to higher latitudes and allow the incursion of proto

NADW further south, thereby decreasing the δ18O value of lower latitude sites. Additionally Site

689 shows this observed warming, but it is located higher up on the Maud Rise and is likely being subjected to the halo of warmer water that surrounds the rise. There is also some evidence to suggest δ18O changes are leading δ13C changes, suggesting changes in ocean circulation may be forced by climate or ice volume.

Additional findings indicate 405-kyr eccentricity played a major role in the overall fluctuations in the records, in agreement with other Oligocene records (e.g. Palike et al., 2006).

A 100-kyr eccentricity signal is also present and could prove useful for orbitally tuning this record to produce a more accurate age model. Overall, the ice volume estimates from Site 690B suggest a large ice sheet was present in Antarctic during the Late Oligocene, in agreement with estimates from Pekar et al. (2006). The Oi2b isotopic event at Site 690B indicates Antarctic ice

146

volume was up to 155 % of modern during the glacial maximum, consistent with values from the earliest Oligocene (Wilson et al., 2013).

147

Chapter 5. The state of the Oligocene icehouse world

The Oligocene (34-23 Ma) can be characterized as the inception of the icehouse world of the Cenozoic. Since initial continental-scale growth of the ice sheets at the Eocene-Oligocene boundary (34 Ma), there has been considerable debate about the state of the Oligocene Antarctic cryosphere. Chapter 1 summarized some important questions that remained unanswered about the Oligocene Antarctica, which this study set out to investigate:

1. How did the ice sheet in Antarctica evolve through the Oligocene and was it similar

to or larger than today?

2. Is the Wilkes Land region a particularly responsive sector of East Antarctica?

3. Was a West Antarctic Ice Sheet (WAIS) present during the Oligocene and what was

its influence on the flow of ice into the Ross Sea?

4. How prominent was the Late Oligocene warming in Antarctica?

The research completed in Chapters 2, 3, and 4 provide adequate data to answer these questions and to pose new ones for future research. The following sections specifically address these questions.

5.1 The evolution of the Antarctic ice sheet

The current understanding of Oligocene cryospheric evolution in Antarctica is limited to far-field records that show considerable variation in interpretations. Oxygen isotope records (e.g.

Zachos et al., 2001) suggest a heavily deglaciated ice sheet in Antarctica during portions of the

Oligocene, mainly ice volumes less than 50 % of modern. Others suggest the ice sheet may have been significantly reduced, below 50 %, but at times reached near present day size or larger

148

during glacial maxima, up to 130 % (e.g. Lear et al., 2004; Pekar et al., 2006). While 1.2 myr obliquity and 400 kyr eccentricity have been found to have the greatest influence on Oligocene ice volume (Wade and Pälike, 2004; Pälike et al., 2006), the data are derived from a lower latitude drill site. Proximal Antarctic records at orbital scale do not currently exist for the

Oligocene, but are needed to understand the evolution of the ice sheet. These distal records could be subjected to local factors that are not an Antarctic signal (i.e. water mass changes, fresh water runoff, subsidence, nutrient availability, and others). Additionally, any interpretations about the

Antarctic cryosphere are generalized, and cannot speak to specific regions of the continent.

Proximal sedimentary records from the continental margin are the only way to investigate how specific regions of Antarctica were responding to climate changes. To address the evolution of the ice sheet, ice-rafted debris (IRD) records from the Wilkes subglacial basin (WSB) and the

Ross Sea were used to interpret cryospheric changes in in these sensitive regions of Antarctica, and a stable isotope record from Site 690B places the dynamics of these regions into a broader view.

Conclusions from Sites U1356, 270, and 690B all indicate a large Antarctic ice sheet existed during the Late Oligocene despite CO2 levels that were higher than present (Figure 5-1).

The ice sheet was present on two margins that are thought to be particularly responsive to climate changes, the WSB and the Ross Sea. The IRD provenance ages from Site U1356 and Site

149

270 indicate a dynamic and primarily large ice sheet existed during the Late Oligocene,

Site 690 δ18O (‰) 2.5 3.0 3.5

25.0

25.5 WAIS 270 Site

26.0 Age (Ma) 26.5

27.0

27.5

0.1 1 10 100 1000 Site U1356 IRD (grains/gram)

Figure 5-1 – Summary and comparison of data collected during this study. The blue rectangle indicates the existence of the WAIS only during the intervals studied. It is likely the WAIS was consistently present during the Oligocene. supported by ice volume estimates from Site 690B. Site U1356 is off the coast of the WSB, currently one of the most sensitive regions of East Antarctica (DeConto and Pollard, 2003; Hill et al., 2007; Pollard et al., 2015). The provenance of Late Oligocene sediment at this site is

150

mainly dominated by ages representing the Mertz Shear Zone, which is the boundary between the WSB and Adélie craton. This signal is present in high and low abundance IRD layers, suggesting a stable ice sheet resided at the coastline in the MSZ and Adélie regions. The other signal in this provenance record is indicative of rocks within the WSB or further east, but they are not present in all of the IRD layers, suggesting this region is more sensitive to climate forcings (i.e. orbital cyclicity), even though this region was likely above sea-level during the

Late Oligocene (Wilson et al., 2012).

Even though Site 270 only has a record that only spans ~500-kyrs, it shows that both East and West Antarctica were contributing IRD to the central Ross Sea. The distribution of ages indicates ice flow from East and West Antarctica into the Ross Sea was somewhat dynamic, but primarily dominated by West Antarctica. The scale of this sensitivity however, is at a higher resolution than the samples analyzed. It is likely that short-term obliquity and precession play an important role. The resolution is low because this study’s primary goal was to investigate if a

WAIS existed during the Late Oligocene. For this same interval at Site U1356, a dominant Mertz

Shear Zone signal is present, with a smaller input from rocks further east. Ice volume estimates from Site 690B during this interval suggest an ice sheet that was mainly between 70 and 100 % of modern, and may have dropped below 50 % during warmer periods, which could be related to short term forcing that is obscured by these records. This is in agreement with far-field estimates

(Kominz and Pekar, 2001; Lear et al., 2004; Pekar et al., 2006; Pekar and Christie-Blick, 2008).

The stable isotope record from ODP Site 690B (Chapter 4) has been interpreted to reflect an ice volume signal with little reflection of temperature change.. This is due to the water depth

(~3,000 m) and its proximity to the Antarctic coastline (~400 km). This site was most likely bathed by cold, Antarctic Bottom Water in which the temperature did not vary much from 2°C

151

(modern cold value). Ice volume was calculated from the δ18O values using the calibrations of

Pekar et al. (2006). Results indicated ice volume fluctuated up to 155 % of modern during cold intervals and as low as 20 % during warm intervals. The values were heavily influenced by orbital forcing, especially 405-kyr eccentricity. The ~9-kry resolution of this study makes it difficult to interpret short-term obliquity, though it is suspected as a major control.

In addition to the provenance, IRD concentrations from Site U1356 were also presented, which show high IRD concentrations correlating with 405-kyr eccentricity indicative of cooling periods and less seasonality. These IRD pulses appear to have been influenced by insolation and

405-kyr eccentricity, but limitations of the age model created difficulties for quantitative correlation. These high IRD periods also have a good correlation with the ice volume estimates from Site 690B showing a near or larger than modern ice sheet (Figure 5-1). Modeling studies suggest a retreat in the WSB would not occur until ice volume dropped below at least 80 to 90 % of modern (Deconto and Pollard, 2003; Pollard and DeConto, 2005; DeConto et al., 2012, Hill et al., 2007; Pollard et al., 2015). (Need additional data to interpret further).

Additionally, debris flows that were identified correlate with Oi events, suggesting an increase in the size of the ice sheet was responsible for the generation of these flows. At Site

690B, only the Oi2b event was recorded (26.7 Ma), the largest isotopic value of the Oligocene at

3.59 %. The appearance of the debris flow during this time at Site U1356 and the large ice volume indicated by the 690B record reflect a large ice sheet, one that was reaching the coast of the Wilkes subglacial basin, the most sensitive region in East Antarctica. Provenance of the grains in the debris flow layers shows a significant contribution from the Mertz Shear Zone as well as ages indicative of rocks currently obscured by ice. These sediments were deposited

152

further up on the continental shelf and provide insight into how sediment deposition evolved on the shelf prior to these debris flows.

IRD concentrations from the Early Oligocene at Site U1356 appear to correlate with warming signatures in orbital forcing, suggesting a change production and transport of icebergs occurred during the larger recovery gap of this core (30-28 Ma). Grains suitable for 40Ar/39Ar dating were extremely rare in these samples so no provenance data was collected. The Early

Oligocene interval is of lower resolution than the Late Oligocene, which is limiting the interpretations and age model uncertainty and recovery gaps may also be a factor. The presence of IRD during the Early Oligocene however, does indicate the ice sheet was at the WSB coastline.

Together, these records show that ice volume during the Oligocene fluctuated in response to long-term orbital forcing and that overall ice volume was large, which suggests that disagreement from the distal records may be influenced by localized factors. The data suggest a large ice sheet was frequently present and shows evidence of changes in sediment sources, indicating a dynamic ice sheet was present in both the WSB and in the Ross Sea, two sensitive regions of Antarctica.

5.2 Sensitivity of the Wilkes Subglacial basin

Modeling studies indicate that the WSB coastline was the last region of the East Antarctic coastline to become glaciated (DeConto and Pollard, 2003) and would be the first coastline affected in a warming climate (Hill et al., 2007). The Wilkes Land basin currently resides >500 m below sea-level (Fretwell et al., 2013) and therefore should be sensitive to changes. If this region is the most responsive area of East Antarctica, it is important to understand how the ice

153

sheet responded under climactic conditions similar to what is predicted for the end of this century. The WSB is quite large, and drains a sizeable portion of East Antarctica (Rignot et al.,

2011). Changes within this drainage region could have profound effects or feedbacks for a large portion of the continent. The WSB was primarily above sea-level during the Oligocene (Wilson et al., 2012), but still lower elevation than the surrounding Adélie craton or Transantarctic

Mountains.

The IRD concentrations from Site U1356 indicate long-term orbital forcing influenced iceberg discharge or transport. It is highly likely that short-term forcing (obliquity) is also a controlling mechanism, but the resolution of this study and the age model are not suitable for this. These changes at orbital scale indicate the WSB region was sensitive to long-term climate variations. The provenance data of the IRD shows input from the WSB and areas further east was variable, suggesting changes in ice sheet erosion and extent in the area. While 40Ar/39Ar ages of grains indicative of the MSZ were the dominant population in most layers analyzed, ages of grains indicative of the WSB were comparatively more abundant in layers containing low concentrations of IRD. This could be interpreted to signify a retreat of the ice sheet in the WSB.

The 40Ar/39Ar ages do indicate the WSB is a dynamic region in terms of iceberg production.

5.3 The West Antarctic Ice Sheet and the Ross Sea

The presence of a large West Antarctic Ice Sheet (WAIS) during the Oligocene has been the subject of debate. While modeling (Wilson et al., 2012), seismic interpretations (Sorlien et al., 2007), and on-land geology from Marie Byrd Land (Rocchi et al., 2006) suggest the presence of a large WAIS during the Oligocene, geochemical evidence from the glaciomarine sedimentary

154

record does not exist. The Oligocene sediment at DSDP Site 270 provides an excellent opportunity to investigate the presence of a WAIS.

The 40Ar/39Ar data from IRD at Site 270 shows a large input of source ages that can only be explained by the geology of West Antarctica (100-440 Ma). In the 500 kyr interval studied, large changes seem to occur in the flow of ice into the Ross Sea. During the Oligocene it is mainly dominated by ice flowing out of West Antarctica. Any ice streams from the central

Transantarctic Mountains (TAM) is primarily forced west and north along the coastline the western boundary of the Ross Sea. The primarily dominant West Antarctic flow also includes a portion of the southern TAM that first drain into the West Antarctic Rifts System before entering the Ross Sea. A West Antarctic dominated flow into the Ross Sea is in agreement with what was interpreted for the Middle Miocene (Hauptvogel and Passchier, 2012).

The stable isotope record from Site 690B indicates that a WAIS is still present in the

Ross Sea region when Antarctic ice volume is as low as 70 to 80 % of modern, suggesting much of West Antarctica was still above sea-level during the Late Oligocene.

5.4 The Late Oligocene Warming

All stable isotope records of the Late Oligocene (e.g. Zachos et al., 2001; 2008; Palike et al., 2006) show a large warming after the Oi2b (26.7 Ma) isotopic event that lasts until just before the Miocene boundary, however Pekar and Christie-Blick (2008) show a significant portion of this warming is an artifact between the geographic distribution of the study sites and suggest the Late Oligocene warming was not as prominent as the stacked data suggest. The stable isotope record from Site 690B shows no evidence of a warming trend during the Late

155

Oligocene, suggesting the warming interpreted in other records is primarily due to changes in bottom water temperature and not ice volume.

Instead of an overall warming, the record at Site 690B shows fluctuations that are driven by orbital forcing. At the ~9 kry sampling resolution, changes can be seen at the 405-kyr eccentricity scale, in agreement with previous interpretations from lower latitudes (Palike et al.,

2006). There are also smaller variations, which can be linked to short-term eccentricity and most likely obliquity.

5.5 Future Work

The research presented in this dissertation presents new questions to be investigated about the dynamics of the Late Oligocene cryosphere. At Site U1356, the next steps will be to investigate the difference in fine-grained sediment (<63 µm) geochemistry between high and low

IRD-bearing intervals. I propose to further probe the glacialogical history preserved in these sediments by applying multi-proxy provenance methods that have been previously used successfully in other time intervals (Sr and Nd isotopes on fine sediment fractions, 40Ar/39Ar chronology of individual sand grains of hornblende and mica, major and trace element analyses) as well as to develop and apply less known strategies for defining sediment provenance (K/Ar ages of fine sediment fractions, 40Ar/39Ar chronology of bulk silt). This suite of geochemical fingerprinting techniques will allow us to obtain a comprehensive understanding of the glacial dynamic history of the East Antarctic Ice Sheet in the Late Oligocene. From these data, interpretations can be made about the location(s) of the ice sheet margin, paleo-ice streams during glacial and interglacial intervals, as well as erosional dynamics on the continent.

156

Similar work at DSDP Site 270 is also planned. With the updated age model from

Kulhanek et al. (2014), this site has a much more expanded Oligocene section than previously thought. Using fine-grained geochemistry and IRD provenance can provide a more robust interpretation for the WAIS and dynamics of ice flow into the Ross Sea.

157

Appendix

Appendix 2-1 – Site U1356 IRD weights and counts

318 1356A Depth Age Dry Weight Sand Rock Grains Sample (mbsf) (Ma) Weight >63 wt % Qtz Hbl Mv Bt K Frag per gram 68R 1W 15/17 641.55 25.368 9.78 0.18 1.85 1 1 0.21

68R 1W 70/71 642.1 25.374 9.56 0.00 0.04 1 0.11

68R 1W 126/127 642.66 25.381 9.00 0.00 0.01 0 0.00

68R 2W 115/116 644.04 25.396 9.15 0.00 0.01 0 0.00

68R 3W 77/80 645.04 25.407 10.21 0.00 0.04 0 0.00

68R 4W 28/31 645.84 25.416 9.68 0.00 0.02 0 1 0.10

68R 4W 37/38 645.93 25.417 9.62 0.01 0.07 1 1 0.21

68R 4W 65/66 646.21 25.420 8.82 0.01 0.10 0 0.00

68R 4W 77/80 646.33 25.422 9.95 0.03 0.26 0 0.00

68R 4W 141/143 646.97 25.429 9.19 0.22 2.44 112 10 2 2 13.71

68R 5W 5/7 647.08 25.430 9.06 0.04 0.49 8 1 4 1.44

68R 5W 20/21 647.23 25.432 8.46 0.00 0.05 0 0.00

68R 5W 67/68 647.7 25.437 9.46 0.01 0.11 0 0.00

68R 5W 68/71 647.71 25.437 9.92 0.00 0.04 1 0.10

68R 6W 25/26 648.01 25.441 9.28 0.01 0.13 1 0.11

68R 6W 28/31 648.04 25.441 10.05 0.00 0.03 0 0.00

69R 1W 26/27 651.26 25.477 9.34 0.01 0.05 1 0.11

69R 1W 77/80 651.77 25.483 9.76 0.01 0.09 0 0.00

69R 1W 93/94 651.93 25.485 9.24 0.00 0.02 3 0.33

69R 1W 106/107 652.06 25.486 9.52 0.01 0.05 1 0.11

69R 1W 124/125 652.24 25.488 9.41 0.01 0.05 0 1 0.11

69R 1W 128/131 652.28 25.489 9.69 0.01 0.08 1 0.10

69R 2W 23/25 652.58 25.492 9.96 0.03 0.29 8 0.80

69R 2W 26/27 652.61 25.493 9.51 0.04 0.44 18 1 4 2.42

69R 2W 28/31 652.63 25.493 9.84 0.02 0.16 1 0.10

69R 2W 77/80 653.12 25.506 9.89 0.01 0.08 2 0.20

69R2W 102/104 653.37 25.512 9.04 0.02 0.18 176 4 19.91

69R 3W 17/18 653.84 25.524 9.41 0.00 0.03 0 0.00

69R 3W 28/31 653.95 25.526 10.29 0.01 0.12 1 0.10

69R 3W 47/48 654.14 25.531 9.19 0.02 0.26 57 1 6.32

69R 3W 72/75 654.39 25.537 9.87 0.01 0.07 2 0.20

69R 3W 128/131 654.95 25.551 9.97 0.01 0.14 1 0.10

69R 4W 56/57 655.66 25.569 8.63 0.01 0.10 0 0.00

69R 4W 77/80 655.87 25.574 10.17 0.27 2.67 389 7 3 39.23

69R 4W 128/131 656.38 25.587 10.10 0.03 0.26 283 3 28.31

69R 5W 17/18 656.77 25.597 9.35 0.00 0.04 1 0.11

69R 5W 28/31 656.88 25.600 9.65 0.00 0.00 0 0.00

69R 5W 66/68 657.26 25.609 8.99 0.00 0.02 1 0.11

69R 5W 128/131 657.88 25.625 9.74 0.08 0.80 46 1 4.83

158

70R 1W 28/31 660.88 25.700 9.74 0.02 0.25 20 1 4 2.57

70R 1W 81/84 661.41 25.713 9.77 0.18 1.87 84 8.60

70R 1W 128/131 661.88 25.725 9.19 0.00 0.02 0 0.00

70R 2W 30/33 662.35 25.737 9.90 0.03 0.26 2 0.20

70R 2W 73/76 662.78 25.748 9.72 0.00 0.01 0 0.00

70R 3W 28/31 663.73 25.772 9.95 0.01 0.08 0 0.00

70R 3W 77/80 664.22 25.784 9.94 0.05 0.54 3 0.30

71R 1W 16/18 670.18 25.933 9.54 0.01 0.08 0 0.00

71R 1W 55/58 670.57 25.943 9.39 0.00 0.04 0 0.00

71R 1W 75/78 670.77 25.948 9.41 0.03 0.33 2 0.21

71R 1W 106/108 671.08 25.956 9.44 0.01 0.13 3 2 0.53

71R 2W 25/28 671.60 25.969 9.21 0.01 0.14 1 0.11

71R 2W 65/67 672.00 25.979 8.99 0.01 0.08 1 0.11

71R 2W 77/79 672.12 25.982 9.32 0.34 3.64 1600 13 5 173.59

71R 2W 86/88 672.31 25.987 8.85 0.01 0.08 10 1.13

71R 2W 116/118 672.51 25.992 9.26 0.01 0.10 2 0.22

71R 2W 138/141 672.73 25.997 9.72 0.47 4.83 2992 10 308.82

71R 3W 26/28 673.07 26.006 9.49 0.02 0.20 60 1 6.43

71R 3W 66/67 673.47 26.016 8.96 0.01 0.11 0.00

71R 4W 78/81 674.92 26.052 9.94 0.05 0.54 3 0.30

71R 4W 127/130 675.41 26.064 9.79 0.01 0.06 1 1 0.21

71R 5W 28/31 675.82 26.075 9.80 0.08 0.86 0 0.00

71R 5W 78/81 676.32 26.087 9.83 0.04 0.36 0 1 1 0.20

71R 6W 28/31 677.22 26.110 9.91 0.15 1.46 0 0.00

71R 6W 77/80 677.71 26.122 10.08 0.02 0.24 0 0.00

72R 1W 26/28 680.06 26.181 9.12 0.00 0.04 0 0.00

72R 1W 55/57 680.35 26.188 9.02 0.01 0.06 0 0.00

72R 1W 117/118 680.97 26.203 9.12 0.01 0.07 0 0.00

72R 2W 16/18 681.46 26.215 9.20 0.01 0.07 0 0.00

72R 2W 28/31 681.58 26.218 9.98 0.08 0.84 0 0.00

72R 2W 36/38 681.66 26.220 9.54 0.02 0.19 0 0.00

72R 2W 78/81 682.04 26.230 10.06 0.03 0.25 0 0.00

72R 2W 85/87 682.11 26.231 9.62 0.00 0.03 0 0.00

72R 2W 122/123 682.52 26.241 9.26 0.00 0.03 0 0.00

72R 2W 128/131 682.58 26.243 9.75 0.01 0.07 4 4 0.82

72R 3W 28/31 682.90 26.251 10.38 0.10 0.99 30 1 4 3.37

72R 3W 56/58 683.18 26.258 9.26 0.02 0.21 3 1 0.43

72R 3W 119/121 683.81 26.273 9.12 0.00 0.02 0 0.00

72R 4W 16/18 683.99 26.278 8.92 0.00 0.02 0 0.00

72R 4W 35/37 684.18 26.282 8.82 0.01 0.09 0 0.00

72R 4W 55/57 684.38 26.287 9.35 0.03 0.34 5 3 0.86

72R 4W 78/81 684.61 26.293 9.79 0.04 0.37 3 3 0.61

72R 4W 95/98 684.78 26.297 8.97 0.04 0.41 5 1 0.67

72R 4W 115/117 684.98 26.302 9.22 0.01 0.13 0 1 0.11

72R 4W 136/138 685.19 26.307 8.90 0.01 0.12 1 0.11

72R 5W 9/11 685.34 26.311 9.23 0.02 0.25 0 0.00

159

72R 5W 18/21 685.43 26.313 9.11 0.00 0.03 0 0.00

72R 5W 28/31 685.53 26.316 9.43 0.01 0.13 13 1.38

72R 5W 55/58 685.80 26.322 9.48 0.01 0.08 0 0.00

72R 5W 78/81 686.03 26.328 9.86 0.01 0.09 3 0.31

72R 5W 95/97 686.20 26.332 8.89 0.00 0.03 0 0.00

72R 5W 127/128 686.52 26.340 9.38 0.31 3.34 480 7 8 52.78

72R 5W 128/131 686.53 26.340 10.56 0.46 4.39 710 3 1 8 68.38

72R 5W 136/138 686.61 26.342 9.23 0.03 0.36 230 14 1 26.54

72R 6W 26/28 687.00 26.352 9.09 0.02 0.24 0 0.00

72R 6W 28/31 687.02 26.353 10.02 0.04 0.37 2 0.20

72R 6W 78/81 687.52 26.365 9.96 0.03 0.32 4 0.40

73R 1W 28/31 689.68 26.418 9.81 0.01 0.11 31 1 1 3.37

73R 1W 67/68 690.07 26.428 9.44 0.01 0.15 2 1 0.32

73R 2W 25/27 691.08 26.453 9.57 0.17 1.81 29 4 3.45

73R 2W 78/81 691.61 26.466 10.16 0.00 0.04 0 0.00

73R 2W 94/96 691.77 26.470 8.94 0.00 0.00 0 0.00

73R 2W 115/117 691.98 26.475 9.14 0.00 0.03 0 0.00

73R 2W 136/138 692.19 26.480 9.14 0.02 0.19 0 2 0.22

73R 3W 56/58 692.40 26.485 9.05 0.01 0.09 10 1.11

73R 4W 28/31 693.94 26.523 9.92 0.03 0.32 13 1 1.41

73R 4W 55/57 694.21 26.530 9.20 0.01 0.15 32 3.48

73R 4W 78/81 694.44 26.536 11.54 0.05 0.45 422 1 36.65

73R 4W 86/88 694.52 26.538 9.18 0.69 7.54 10127 15 4 3 1105.32

73R 5W 5/8 694.68 26.542 9.28 0.39 4.15 1120 14 1 15 123.90

73R 5W 32/34 694.95 26.548 8.63 0.99 11.47 11438 15 9 8 1328.62

73R 5W 43/45 695.06 26.551 8.94 0.01 0.15 6 0.67

73R 5W 65/68 695.28 26.556 9.56 0.05 0.52 5 3 0.84

73R 5W 68/71 695.31 26.557 10.04 0.06 0.61 7 0.70

75R 1W 85/87 709.45 27.084 9.46 0.02 0.22 13 1.38

75R 1W 128/131 709.88 27.104 10.02 0.01 0.05 5 2 0.70

75R 2W 28/31 710.23 27.121 9.70 0.07 0.76 123 2 5 13.40

75R 2W 78/81 710.73 27.144 9.63 0.14 1.46 314 12 1 33.94

75R 2W 129/131 711.24 27.169 9.68 0.20 2.04 1353 3 10 3 141.47

75R 3W 28/31 711.66 27.189 10.05 0.04 0.42 282 1 28.16

75R 3W 56/58 711.88 27.199 8.61 0.12 1.37 64 7.43

75R 3W 78/81 712.16 27.212 9.78 0.35 3.62 4959 15 1 5 509.10

75R 4W 28/30 712.85 27.245 9.42 0.10 1.09 1304 4 1 1 139.13

75R 4W 78/81 713.35 27.269 10.02 0.16 1.63 131 1 13.17

76R 1W 45/48 718.65 27.520 9.41 0.45 4.81 5339 19 3 5 570.31

76R 1W 116/118 719.36 27.554 9.72 0.52 5.35 8493 18 7 876.25

76R 2W 35/37 719.79 27.574 9.89 0.30 2.99 3036 20 3 309.46

76R 2W 78/81 720.22 27.595 12.95 0.43 3.28 5126 18 9 398.07

76R 2W 105/107 720.49 27.608 10.08 0.05 0.45 3 1 1 0.50

76R 3W 62/63 721.36 27.649 9.52 0.01 0.15 50 1 5.36

76R 3W 78/81 721.52 27.657 8.90 0.29 3.25 224 5 1 3 1 26.31

76R 3W 101/103 721.75 27.667 9.44 0.02 0.17 2 0.21

160

76R 4W 28/31 722.37 27.697 13.96 0.06 0.46 5 0.36

76R 4W 55/57 722.64 27.710 9.25 0.03 0.27 15 3 1.95

76R 4W 78/81 722.97 27.725 10.86 0.01 0.13 18 1.66

76R 4W 106/108 723.25 27.739 9.33 0.43 4.61 7201 12 6 13 775.47

76R 5W 25/28 723.82 27.766 9.36 0.00 0.03 0 0.00

76R 5W 28/31 723.85 27.767 9.00 0.02 0.18 2 0.22

76R 5W 78/81 724.35 27.791 15.99 0.02 0.11 0 0.00

76R 6W 15/17 724.86 27.815 9.48 0.03 0.26 4 1 0.53

76R 6W 28/31 724.99 27.821 12.72 0.01 0.11 0 0.00

77R 1W 20/23 728.00 27.947 11.28 0.04 0.38 27 1 6 3.02

78R 1W 28/31 737.68 28.349 9.49 0.01 0.07 2 0.21

78R 1W 77/80 738.17 28.369 11.33 0.02 0.13 1 1 0.18

78R 2W 28/31 739.12 28.408 10.69 0.03 0.32 2 1 0.28

78R 2W 77/80 739.61 28.429 9.86 0.08 0.81 0.00

79R 1W 28/31 747.28 28.747 9.59 0.01 0.15 0.00

79R 1W 77/80 747.77 28.768 9.36 0.04 0.45 2 1 0.32

79R 2W 28/31 748.74 28.808 9.36 0.04 0.41 0.00

79R 2W 77/80 749.23 28.828 10.01 0.03 0.25 0.00

79R 3W 77/80 750.58 28.884 9.89 0.03 0.30 2 1 0.30

79R 4W 28/31 751.55 28.925 9.58 0.02 0.16 0.00

79R 4W 76/79 752.03 28.944 9.60 0.01 0.10 0.00

79R 4W 128/131 752.55 28.966 9.41 0.00 0.03 1 1 0.21

79R 5W 28/31 753.04 28.986 9.44 0.18 1.88 0.00

79R 5W 77/80 753.53 29.007 9.78 0.02 0.19 0.00

79R 5W 128/131 754.04 29.028 9.73 0.11 1.10 0.00

82R 2W 28/31 776.08 29.943 9.75 0.00 0.03 0.00

82R 2W 77/80 776.57 29.963 10.01 0.01 0.07 0.00

82R 4W 28/31 780.16 30.112 9.81 0.00 0.01 0.00

82R 4W 77/80 780.65 30.133 9.70 0.00 0.02 0.00

82R 5W 28/31 781.41 30.164 9.91 0.06 0.58 0.00

82R 5W 77/80 781.90 30.185 9.47 0.01 0.08 0.00

82R 6W 77/80 783.08 30.230 9.81 0.01 0.07 0.00

84R 1W 28/31 795.38 30.623 9.12 0.00 0.02 11 1 1.32

84R 2W 28/31 796.77 30.668 9.33 0.01 0.11 2 1 0.32

84R 2W 77/80 797.26 30.683 9.69 0.04 0.41 34 3.51

84R 3W 77/80 798.77 30.732 9.35 0.06 0.64 62 1 6.74

84R 3W 128/131 799.28 30.748 9.62 0.02 0.21 23 2.39

84R 4W 28/31 799.84 30.766 8.94 0.01 0.08 4 0.45

84R 4W 77/80 800.33 30.782 9.16 0.00 0.04 1 0.11

84R 4W 128/131 800.84 30.798 9.70 0.01 0.11 0.00

84R 5W 28/31 801.34 30.814 9.58 0.03 0.27 0.00

84R 5W 77/80 801.83 30.830 9.21 0.03 0.36 2 1 0.33

84R 6W 28/31 802.82 30.861 9.78 0.07 0.73 0.00

84R 7W 28/31 803.93 30.897 9.48 0.04 0.38 4 0.42

84R 7W 66/69 804.31 30.909 9.48 0.00 0.02 0.00

85R 1W 28/31 804.98 30.930 9.79 0.06 0.61 69 7.05

161

85R 2W 28/31 806.26 30.971 9.46 0.02 0.21 2 2 0.42

85R 2W 77/80 806.75 30.987 9.30 0.13 1.40 3 0.32

85R 3W 28/31 807.27 31.004 9.48 0.04 0.42 2 0.21

85R 3W 77/80 807.76 31.019 9.77 0.04 0.41 4 0.41

85R 3W 128/131 808.13 31.031 9.79 0.07 0.72 5 0.51

85R 4W 28/31 808.77 31.052 9.68 0.05 0.52 86 8.89

85R 4W 77/80 809.26 31.067 9.75 0.05 0.51 54 5.54

85R 4W 128/131 809.77 31.084 9.12 0.01 0.11 7 0.77

85R 5W 28/31 810.27 31.100 9.73 0.16 1.64 670 68.86

85R 5W 77/80 810.76 31.115 9.47 1.26 13.31 10000 1055.97

86R 1W 28/31 814.58 31.237 9.07 0.05 0.55 62 6.84

86R 1W 77/80 815.07 31.253 9.22 0.02 0.22 15 1 1.74

86R 1W 128/131 815.58 31.269 9.17 0.02 0.22 17 1.85

86R 2W 77/80 816.55 31.300 8.74 0.08 0.92 0 0.00

86R 3W 28/31 817.27 31.324 9.08 0.06 0.66 4 0.44

86R 3W 77/80 817.76 31.339 9.01 0.09 1.00 57 6.33

86R 4W 27/30 818.25 31.355 8.94 0.16 1.79 236 26.40

86R 4W 77/80 818.75 31.371 8.99 0.10 1.11 5 0.56

86R 4W 128/131 819.26 31.387 9.11 0.12 1.32 8 0.88

86R 5W 28/31 819.63 31.399 8.79 0.01 0.11 1 1 0.23

86R 5W 77/80 820.12 31.415 8.90 0.05 0.56 0 0.00

86R 5W 126/129 820.61 31.430 9.19 0.28 3.05 0 0.00

86R 6W 28/31 820.92 31.440 11.91 0.02 0.17 0 0.00

86R 6W 77/80 821.41 31.456 9.01 0.01 0.11 0 0.00

86R 7W 27/80 822.05 31.476 8.81 0.01 0.11 0 0.00

86R 7W 57/80 822.35 31.486 8.94 0.01 0.11 0 0.00

87R 1W 28/31 824.18 31.545 9.02 0.02 0.22 9 1 1.11

87R 1W 77/80 824.67 31.560 9.08 0.02 0.22 0 0.00

87R 1W 128/131 825.18 31.577 8.96 0.05 0.56 0 0.00

87R 2W 28/31 825.69 31.593 10.21 0.01 0.10 1 0.10

87R 2W 77/80 826.18 31.609 9.70 0.19 1.96 3 0.31

87R 2W 128/131 826.69 31.625 9.08 0.01 0.11 0 0.00

87R 3W 28/31 827.20 31.641 8.95 0.07 0.78 5 2 0.78

87R 3W 77/80 827.69 31.657 9.30 0.07 0.75 4 0.43

87R 4W 28/31 828.36 31.678 9.35 0.17 1.82 245 2 26.42

87R 4W 77/80 828.85 31.694 9.13 0.09 0.99 142 2 15.77

87R 4W 128/131 829.36 31.710 9.37 0.04 0.43 3 1 0.43

87R 5W 28/31 829.86 31.726 8.91 0.03 0.34 0 0.00

87R 5W 77/80 830.35 31.742 9.24 0.05 0.54 4 0.43

87R 5W 128/131 830.86 31.758 9.37 0.03 0.32 0 0.00

87R 6W 77/80 831.84 31.790 7.87 0.01 0.13 4 0.51

87R 7W 28/31 832.54 31.812 8.81 0.01 0.11 0 0.00

87R 7W 57/60 832.83 31.821 8.81 0.02 0.23 4 0.46

88R 1W 28/31 833.78 31.852 9.69 0.04 0.41 7 2 0.93

88R 1W 128/131 834.78 31.884 8.91 0.02 0.22 2 1 0.34

88R 2W 28/31 835.14 31.895 10.96 0.03 0.27 3 1 0.37

162

88R 2W 77/80 835.63 31.911 9.15 0.01 0.11 3 0.33

88R 2W 116/119 836.02 31.923 7.71 0.01 0.13 11 1 2 1.82

89R 1W 28/31 843.38 32.159 14.85 0.10 0.67 4 1 0.34

89R 1W 77/80 843.87 32.174 15.83 0.41 2.59 182 1 11.56

89R 2W 28/31 844.69 32.201 9.66 0.53 5.49 2200 227.74

89R 2W 77/80 845.18 32.216 15.66 0.03 0.19 4 0.26

89R 3W 126/129 846.97 32.274 8.15 0.00 0.05 260 31.90

89R 4W 28/31 847.28 32.284 12.19 0.18 1.48 21 1.72

89R 4W 77/80 847.77 32.299 12.38 0.48 3.88 0 0.00

89R 5W 28/31 848.53 32.324 10.13 0.02 0.20 6 0.59

89R 5W 112/115 849.37 32.350 7.94 0.01 0.13 1 2 0.38

91R 2W 28/31 863.44 32.800 10.63 0.15 1.39 0 0.00

91R 3W 28/31 864.68 32.840 22.48 0.02 0.09 4 0.18

91R 5W 28/31 867.07 32.917 7.79 0.51 6.55 5 0.64

91R 6W 28/31 868.43 32.960 10.82 0.81 7.49 6 0.56

92R 2W 28/31 873.34 33.117 13.54 0.02 0.15 17 1.26

92R 3W 28/31 874.30 33.148 19.73 0.04 0.20 8 0.41

93R 1W 28/31 876.88 33.230 9.19 0.01 0.11 1 1 0.22

93R 3W 28/31 879.33 33.314 10.40 0.01 0.10 31 7 3.65

94R 1W 28/31 881.68 33.397 8.33 0.02 0.24 12 1.44

94R 2W 28/31 883.16 33.450 6.74 0.02 0.30 37 1 5.64

94R 4W 28/31 885.91 33.549 8.79 0.12 1.37 147 12 18.09

95R 1W 28/31 891.28 33.740 9.89 0.16 1.64 2730 276.18

95R 2W 28/31 892.76 33.793 9.68 0.12 1.27 880 2 4 3 91.89

95R 3W 28/31 894.15 33.843 9.59 1.72 17.88 10000 1042.43

95R 4W 28/31 895.24 33.882 10.18 7.60 74.67 10000 982.71

163

Appendix 2-2 – Site U1356 argon thermochronology data

Depos itional Grain Age Depth age Age error Size Run ID (mbsf) (Ma) (Ma) (±) Ca/K Mol 39Ar % Ar40* Mineral (µm) 15030-01A 646.97 25.429 1416 8 0.835675 0.076511 99.63 Bt >150 15030-02A 646.97 25.429 1486 8 7.046828 0.017295 93.94 Hbl >150 15030-03A 646.97 25.429 1341 9 0.610140 0.004483 88.50 Bt >150 15030-04A 646.97 25.429 1486 8 10.521610 0.009026 98.93 Hbl >150 15030-05A 646.97 25.429 900 6 2.577133 0.016609 93.39 Hbl >150 15030-06A 646.97 25.429 911 31 0.295793 0.000268 45.52 Bt >150 15030-07A 646.97 25.429 1601 167 0.557523 0.000044 63.48 Bt >150 15030-08A 646.97 25.429 1550 10 23.384700 0.004177 95.80 Hbl >150 15677-01A 646.97 25.429 171 4 111.429900 0.000954 41.45 Hbl >150 15677-02A 646.97 25.429 1439 6 22.360620 0.002186 98.16 Hbl >150 15677-03A 646.97 25.429 1448 10 11.077540 0.000773 96.54 Hbl >150 15677-04A 646.97 25.429 1034 3 4.867694 0.005569 95.61 Hbl >150 17306-01A 646.97 25.429 1277 5 4.831190 0.058000 99.10 Hbl >150 17306-02A 646.97 25.429 1297 24 8.777300 0.005000 97.50 Hbl >150 16636-01A 646.97 25.429 1597 78 0.270465 0.000105 99.66 Bt >150 16636-02A 646.97 25.429 1550 31 0.035469 0.000283 97.81 Bt >150 16636-03A 646.97 25.429 1591 29 0.049869 0.000322 97.69 Bt >150 16636-04A 646.97 25.429 512 22 0.019225 0.000160 98.38 Bt >150

15680 -01A 656.38 25.587 1650 23 59.913530 0.000272 92.99 Hbl >150 15680-02A 656.38 25.587 1493 18 14.004950 0.000347 100.20 Hbl >150 15680-03A 656.38 25.587 1743 40 18.313740 0.000144 98.92 Hbl >150 15680-04A 656.38 25.587 1453 11 10.919880 0.000649 98.25 Hbl >150 15680-05A 656.38 25.587 1647 10 6.952369 0.000752 100.08 Hbl >150 15680-06A 656.38 25.587 1560 26 33.618030 0.000259 98.88 Hbl >150 15680-07A 656.38 25.587 1182 24 0.254947 0.000197 89.95 Bt >150 15681-01A 656.38 25.587 1529 13 0.151605 0.000483 99.43 Bt 63-150 15681-02A 656.38 25.587 487 17 0.071206 0.000183 83.10 Bt 63-150 15681-03A 656.38 25.587 1378 25 -0.075734 0.000202 99.94 Bt 63-150 15681-04A 656.38 25.587 1668 60 -0.477893 0.000100 96.77 Bt 63-150 15681-05A 656.38 25.587 1705 61 -0.226919 0.000092 92.13 Bt 63-150 15681-06A 656.38 25.587 456 12 0.074246 0.000262 96.78 Bt 63-150 15681-07A 656.38 25.587 512 14 0.084658 0.000229 91.27 Bt 63-150 15681-08A 656.38 25.587 1504 41 0.093955 0.000126 98.88 Bt 63-150 15682-01A 656.38 25.587 1481 4 0.009258 0.004555 99.94 Bt >150 15682-02A 656.38 25.587 1491 6 0.059560 0.001881 99.91 Bt >150 16637-01A 656.38 25.587 1514 63 -0.323835 0.000118 98.44 Bt >150 16637-02A 656.38 25.587 594 37 -0.114361 0.000110 97.19 Bt >150

164

16637-03A 656.38 25.587 1431 13 0.017863 0.000778 99.68 Bt >150 16637-04A 656.38 25.587 1461 19 -0.033007 0.000486 99.36 Bt >150 16637-05A 656.38 25.587 1720 30 0.082490 0.000307 99.90 Bt >150 16637-06A 656.38 25.587 1607 40 -0.201313 0.000227 100.37 Bt >150 16637-07A 656.38 25.587 1650 25 -0.027093 0.000382 99.75 Bt >150 16637-08A 656.38 25.587 523 22 0.036123 0.000155 97.17 Bt >150 16637-09A 656.38 25.587 1670 65 -0.500870 0.000105 99.14 Bt >150

15678 -01A 660.88 25.700 1481 10 8.033887 0.000849 96.92 Hbl >150 15678-02A 660.88 25.700 1633 6 9.023823 0.002061 98.54 Hbl >150

15684 -01A 672.12 25.982 1508 6 24.693240 0.003067 98.83 Hbl >150 15684-02A 672.12 25.982 1459 10 18.490530 0.001350 98.86 Hbl >150 15684-03A 672.12 25.982 1441 35 27.603730 0.000328 91.99 Hbl >150 15684-04A 672.12 25.982 188 306 -5.914611 0.000013 4.82 Hbl >150 15684-05A 672.12 25.982 1479 23 19.813960 0.000520 98.09 Hbl >150 15684-06A 672.12 25.982 1540 18 12.642380 0.000696 99.36 Hbl >150 15684-07A 672.12 25.982 1435 56 16.051930 0.000200 95.27 Hbl >150 15684-08A 672.12 25.982 1547 71 18.578770 0.000165 97.37 Hbl >150 16638-01A 672.12 25.982 1666 32 -0.289624 0.000232 99.50 Bt 63-150 16638-02A 672.12 25.982 1316 13 0.024110 0.000559 99.62 Bt 63-150 16638-03A 672.12 25.982 514 25 -0.176493 0.000154 100.22 Bt 63-150 16638-04A 672.12 25.982 560 38 -0.628122 0.000108 100.00 Bt 63-150 16638-05A 672.12 25.982 1617 43 0.175103 0.000201 100.23 Bt 63-150 16638-06A 672.12 25.982 1486 38 -0.073220 0.000212 99.76 Bt 63-150 16638-07A 672.12 25.982 1579 39 0.111876 0.000209 99.32 Bt 63-150 16639-01A 672.12 25.982 1602 4 0.007350 0.033194 99.90 Bt >150 16639-02A 672.12 25.982 401 12 0.105449 0.000434 86.95 Bt >150

15679 -01A 672.73 25.997 1487 11 30.507090 0.000602 97.32 Hbl >150 15679-02A 672.73 25.997 1233 6 17.615180 0.001442 98.91 Hbl >150 15679-03A 672.73 25.997 -432 344 3.612124 0.000015 -3.13 Hbl >150 15679-04A 672.73 25.997 1505 20 31.609100 0.000371 98.07 Hbl >150 15679-05A 672.73 25.997 1515 11 14.961510 0.000709 95.96 Hbl >150 15679-06A 672.73 25.997 1635 15 10.442090 0.000440 99.40 Hbl >150 15679-07A 672.73 25.997 1508 38 44.020220 0.000138 93.26 Hbl >150 15679-08A 672.73 25.997 1482 27 25.626410 0.000199 97.98 Hbl >150 15679-09A 672.73 25.997 1394 35 47.913430 0.000137 80.40 Hbl >150 15679-10A 672.73 25.997 1491 7 8.361547 0.001040 97.21 Hbl >150 15679-11A 672.73 25.997 1501 31 26.710020 0.000173 97.20 Hbl >150 15679-12A 672.73 25.997 1528 17 20.719740 0.000394 99.20 Hbl >150 15679-13A 672.73 25.997 1501 13 11.108280 0.000532 99.39 Hbl >150 16640-01A 672.73 25.997 1262 47 26.591040 0.000190 96.00 Hbl >150

165

16640-02A 672.73 25.997 1626 39 0.102015 0.000271 99.49 Bt >150 16640-03A 672.73 25.997 1383 33 0.010278 0.000272 90.99 Bt >150 16640-04A 672.73 25.997 1384 25 16.268340 0.000355 98.35 Hbl >150 16640-05A 672.73 25.997 1468 44 12.257770 0.000209 99.63 Hbl >150 16640-06A 672.73 25.997 1428 54 10.634610 0.000160 96.97 Hbl >150 16640-07A 672.73 25.997 1477 15 0.096036 0.000713 99.80 Bt >150

15031 -01A 682.58 26.243 446 15 0.046020 0.000408 90.80 Bt 63-150 15031-02A 682.58 26.243 1694 29 0.002926 0.000322 98.83 Bt 63-150 15031-03A 682.58 26.243 1445 32 -0.049243 0.000260 97.00 Bt 63-150 15031-04A 682.58 26.243 1468 41 -0.061725 0.000192 96.40 Bt 63-150 15031-05A 682.58 26.243 1639 30 -0.037987 0.000311 97.25 Bt 63-150

15032 -01A 682.9 26.251 1493 18 0.042230 0.000688 99.56 Bt >150 15032-02A 682.9 26.251 478 16 0.004436 0.000384 98.25 Bt >150 15032-03A 682.9 26.251 1660 20 -0.031864 0.000615 100.58 Bt >150 15032-04A 682.9 26.251 1678 24 0.003540 0.000446 100.85 Bt >150 15032-05A 682.9 26.251 497 24 0.011244 0.000253 107.14 Bt >150 15032-06A 682.9 26.251 1278 48 -0.035489 0.000144 99.91 Bt >150 15032-07A 682.9 26.251 1579 30 -0.030684 0.000306 101.36 Bt >150 15032-08A 682.9 26.251 533 30 -0.011275 0.000195 105.52 Bt >150 15032-09A 682.9 26.251 468 19 0.029657 0.000336 91.86 Bt >150

15038 -01A 686.53 26.340 1423 8 0.007932 0.007937 99.23 Bt >150 15038-02A 686.53 26.340 -625 581 0.342551 0.000018 -50.17 Bt >150 15038-03A 686.53 26.340 1619 19 0.004196 0.000573 99.83 Bt >150 15038-04A 686.53 26.340 1416 36 0.025575 0.000223 100.38 Bt >150 15038-05A 686.53 26.340 1441 34 0.031336 0.000224 101.74 Bt >150 15038-06A 686.53 26.340 1544 66 0.107657 0.000106 102.32 Bt >150 15038-07A 686.53 26.340 1435 11 0.040194 0.001741 100.29 Bt >150 15038-08A 686.53 26.340 1610 24 0.029504 0.000404 102.23 Bt >150

15042 -01A 686.61 26.342 2972 165 2.249462 0.000052 90.04 Hbl >150 15042-02A 686.61 26.342 477 16 1.348190 0.000394 96.42 Hbl >150 15042-03A 686.61 26.342 1505 9 12.933640 0.004974 99.17 Hbl >150 15042-04A 686.61 26.342 2821 2706 53.112330 0.000003 107.51 Hbl >150 15042-05A 686.61 26.342 1531 16 15.767710 0.000825 99.79 Hbl >150 15042-06A 686.61 26.342 1955 238 1.353167 0.000032 123.62 Hbl >150 15042-07A 686.61 26.342 1535 886 140.362800 0.000008 145.74 Hbl >150 15042-08A 686.61 26.342 3210 832 2.576187 0.000011 116.98 Hbl >150 15042-09A 686.61 26.342 1540 50 0.049922 0.000161 100.07 Bt >150 15042-10A 686.61 26.342 1503 11 6.537432 0.002251 101.17 Hbl >150 15683-01B 686.61 26.342 1543 11 38.426030 0.001413 97.03 Hbl >150

166

15683-02A 686.61 26.342 1426 44 15.384040 0.000266 96.01 Hbl >150 15683-03A 686.61 26.342 1449 39 26.014610 0.000304 97.31 Hbl >150 15683-04A 686.61 26.342 1495 16 6.383564 0.000789 99.06 Hbl >150 15683-05A 686.61 26.342 1411 106 21.497950 0.000108 88.30 Hbl >150 15683-06A 686.61 26.342 4519 1319 17.826580 0.000014 90.28 Hbl >150 15683-07A 686.61 26.342 1434 49 21.177670 0.000235 95.18 Hbl >150 15683-08A 686.61 26.342 1453 43 12.619410 0.000273 97.55 Hbl >150 16641-01A 686.61 26.342 1468 43 0.048229 0.000216 98.55 Bt >150 16641-02A 686.61 26.342 1646 57 -0.120949 0.000169 100.37 Bt >150 16641-03A 686.61 26.342 1607 62 -0.137749 0.000142 100.16 Bt >150 16641-04A 686.61 26.342 1441 32 -0.004948 0.000265 99.46 Bt >150 16641-05A 686.61 26.342 1735 32 0.129142 0.000314 99.84 Bt >150

15037 -01A 691.08 26.453 394 28 -0.098993 0.000209 120.72 Bt 63-150 15037-02A 691.08 26.453 669 62 -0.142306 0.000094 177.12 Bt 63-150 15037-03A 691.08 26.453 526 23 -0.007043 0.000267 109.89 Bt 63-150 15037-04A 691.08 26.453 1609 21 0.092174 0.000458 101.71 Bt 63-150 15037-05A 691.08 26.453 485 29 0.040244 0.000221 110.01 Bt 63-150 15037-06A 691.08 26.453 453 114 -0.089403 0.000051 146.63 Bt 63-150 15037-07A 691.08 26.453 372 16 0.026706 0.000396 86.86 Bt 63-150

17307 -01A 694.95 26.548 895 57 4.582070 0.001000 103.80 Hbl >150 17307-02A 694.95 26.548 1587 5 0.019380 0.098000 99.90 Bt >150 17307-03A 694.95 26.548 1523 5 0.023620 0.198000 99.90 Bt >150 17307-04A 694.95 26.548 1355 5 0.038650 0.477000 99.40 Bt >150 17307-05A 694.95 26.548 1554 10 12.309880 0.015000 99.20 Hbl >150 17307-06A 694.95 26.548 1595 13 11.094100 0.013000 99.20 Hbl >150 17307-07A 694.95 26.548 1529 9 7.512210 0.022000 99.50 Hbl >150 17307-08A 694.95 26.548 1527 6 0.023710 0.048000 99.90 Bt >150 17308-01A 694.95 26.548 1375 23 0.005120 0.005000 99.30 Bt 63-150 17308-02A 694.95 26.548 1483 22 -0.052090 0.005000 100.20 Bt 63-150 17308-03A 694.95 26.548 1704 22 0.034640 0.004000 100.30 Bt 63-150

17309 -01A 710.73 27.144 1380 7 9.593810 0.034000 98.20 Hbl >150 17309-02A 710.73 27.144 1407 10 8.722750 0.031000 98.70 Hbl >150 17309-03A 710.73 27.144 1507 9 9.011690 0.023000 98.70 Hbl >150

17310 -01A 712.16 27.212 1499 8 13.270540 0.037000 99.20 Hbl >150 17310-02A 712.16 27.212 1225 16 9.914190 0.010000 97.70 Hbl >150 17310-03A 712.16 27.212 1407 5 2.976620 0.227000 97.50 Hbl >150

17311 -01A 719.79 27.574 1417 12 11.192950 0.009000 98.40 Hbl >150 17311-02A 719.79 27.574 354 20 3.599610 0.002000 61.10 Hbl >150

167

17311-03A 719.79 27.574 1483 10 3.245360 0.013000 98.80 Hbl >150 17312-01A 719.79 27.574 1435 14 0.199190 0.009000 98.50 Bt 63-150 17312-02A 719.79 27.574 1071 20 0.251810 0.004000 93.10 Bt 63-150 17312-03A 719.79 27.574 1266 25 0.156970 0.003000 98.60 Bt 63-150

17313 -01A 720.22 27.595 1654 7 6.922340 0.019000 99.80 Hbl >150 17313-02A 720.22 27.595 1443 5 7.752150 0.063000 99.70 Hbl >150 17313-03A 720.22 27.595 1134 13 10.801570 0.007000 96.60 Hbl >150

17314 -01A 721.52 27.657 1480 8 0.031600 0.016000 99.80 Bt 63-150 17314-02A 721.52 27.657 960 21 0.196500 0.003000 94.00 Bt 63-150 17314-03A 721.52 27.657 1486 6 0.013730 0.067000 99.80 Bt 63-150 17314-04A 721.52 27.657 876 7 0.536630 0.016000 94.80 Bt 63-150 17314-05A 721.52 27.657 834 10 0.225760 0.009000 90.60 Bt 63-150 17314-06A 721.52 27.657 867 16 0.125460 0.004000 95.60 Bt 63-150 17314-07A 721.52 27.657 445 30 -0.310700 0.001000 126.80 Bt 63-150

17315 -01A 723.25 27.739 1508 5 0.000410 0.113000 100.00 Bt >150 17315-02A 723.25 27.739 1479 7 0.053460 0.032000 99.30 Bt >150 17315-03A 723.25 27.739 1451 9 8.286500 0.020000 99.10 Hbl >150

168

Appendix 3-1 – Site 270 argon thermochronology data

Grain Age Depth Depositional Age error Mol Run ID (mbsf) age (Ma) (Ma) (±) Ca/K 39Ar % Ar40* Mineral 16629-01A 320.36 25.37 479 6 10.8593 0.0007 98.40 Hbl 16629-02A 320.36 25.37 482 71 5.4573 0.0001 90.12 Hbl 16629-03A 320.36 25.37 473 7 12.1936 0.0009 94.48 Hbl 16629-04A 320.36 25.37 491 4 16.3006 0.0030 98.55 Hbl 16629-05A 320.36 25.37 491 3 8.6845 0.0017 99.66 Hbl 16629-06A 320.36 25.37 474 2 -0.0010 0.0036 99.69 Bt 16629-07A 320.36 25.37 488 7 -0.0031 0.0006 94.56 Bt 16629-08A 320.36 25.37 476 2 0.4156 0.0063 99.03 Bt 16629-09A 320.36 25.37 489 2 5.7346 0.0105 99.61 Hbl 16629-10A 320.36 25.37 487 3 0.0871 0.0019 98.87 Bt 16629-11A 320.36 25.37 485 3 21.3030 0.0019 97.26 Hbl 17297-01A 320.36 25.37 338 3 18.8001 0.0700 91.10 Hbl 17297-02A 320.36 25.37 485 3 8.9147 0.1260 99.20 Hbl 17297-03A 320.36 25.37 300 6 44.5401 0.0220 67.50 Hbl 17297-04A 320.36 25.37 259 18 0.3546 0.0020 44.30 Bt 17297-05A 320.36 25.37 485 6 0.0270 0.0060 98.80 Bt 17297-06A 320.36 25.37 478 4 0.1084 0.0130 98.90 Bt 17297-07A 320.36 25.37 290 32 1.6766 0.0010 47.80 Hbl 17297-08A 320.36 25.37 312 6 12.5775 0.0050 91.50 Hbl 17297-09A 320.36 25.37 523 9 0.1714 0.0050 79.50 Bt 17297-10A 320.36 25.37 477 4 0.0174 0.0120 99.30 Bt 17297-11A 320.36 25.37 133 32 13.1838 0.0020 11.30 Hbl 17297-12A 320.36 25.37 496 3 0.1171 0.0230 98.60 Bt 17297-13A 320.36 25.37 490 4 0.0082 0.0190 99.40 Bt 17297-14A 320.36 25.37 489 3 7.1467 0.0760 99.50 Hbl 17297-15A 320.36 25.37 481 3 1.4117 0.0240 98.10 Hbl 16631-01A 324.81 25.42 438 29 0.6866 0.0002 62.16 Bt 16631-02A 324.81 25.42 470 3 0.0167 0.0022 97.26 Bt 16631-03A 324.81 25.42 478 6 -0.0085 0.0008 98.95 Bt 16631-04A 324.81 25.42 356 3 0.0328 0.0022 78.17 Bt 16631-05A 324.81 25.42 102 2 0.0678 0.0064 53.37 Bt 16631-06A 324.81 25.42 313 16 0.0267 0.0133 99.54 Bt 16631-07A 324.81 25.42 441 4 0.0085 0.0022 94.23 Bt 16631-08A 324.81 25.42 114 6 0.1272 0.0010 31.71 Bt 16631-09A 324.81 25.42 479 2 0.0105 0.0093 99.47 Bt 16631-10A 324.81 25.42 703 56 1.1100 0.0001 88.86 Hbl 16631-11A 324.81 25.42 483 2 1.1624 0.0072 96.39 Hbl 17298-01A 324.81 25.42 464 3 0.0048 0.0610 99.60 Bt

169

17298-02A 324.81 25.42 465 3 0.0039 0.0570 99.30 Bt 17298-03A 324.81 25.42 236 27 0.3766 0.0010 28.30 Bt 17298-04A 324.81 25.42 221 30 0.5774 0.0010 19.50 Bt 17298-05A 324.81 25.42 473 8 0.0133 0.0040 96.50 Bt 17298-06A 324.81 25.42 468 3 0.0236 0.0210 99.40 Bt 17298-07A 324.81 25.42 471 4 0.0082 0.0100 98.20 Bt 17298-08A 324.81 25.42 473 4 0.0115 0.0160 99.10 Bt 17299-01A 331.01 25.48 489 5 0.2812 0.0080 96.90 Bt 17299-02A 331.01 25.48 291 7 0.7083 0.0040 83.20 Bt 16632-01A 339.66 25.56 463 2 0.6396 0.0038 99.38 Bt 16632-02A 339.66 25.56 300 12 0.9425 0.0003 75.76 Bt 16632-03A 339.66 25.56 208 4 0.1981 0.0010 92.49 Bt 16632-04A 339.66 25.56 225 6 0.0403 0.0006 93.85 Bt 16632-05A 339.66 25.56 488 3 0.1850 0.0025 99.58 Bt 16632-06A 339.66 25.56 509 3 0.0215 0.0019 99.99 Bt 16632-07A 339.66 25.56 479 2 0.0711 0.0053 98.98 Bt 16632-08A 339.66 25.56 506 2 0.7674 0.0052 99.39 Bt 16632-09A 339.66 25.56 491 4 0.4611 0.0015 97.39 Bt 16632-10A 339.66 25.56 349 5 0.3026 0.0009 94.32 Bt 17300-01A 339.66 25.56 476 2 0.9196 0.1280 97.90 Bt 17300-02A 339.66 25.56 450 15 15.6993 0.0030 45.30 Hbl 17300-03A 339.66 25.56 117 16 1.7334 0.0020 19.90 Hbl 17300-04A 339.66 25.56 543 4 0.1477 0.0210 90.60 Bt 16642-01A 343.46 25.60 386 15 26.3976 0.0002 86.09 Hbl 16642-02A 343.46 25.60 482 3 0.0220 0.0028 98.92 Bt 16642-03A 343.46 25.60 477 2 0.0424 0.0033 98.69 Bt 16642-04A 343.46 25.60 478 5 16.2729 0.0010 97.93 Hbl 16642-05A 343.46 25.60 476 5 5.0309 0.0011 100.07 Hbl 16642-06A 343.46 25.60 479 2 0.0411 0.0026 98.96 Bt 16642-07A 343.46 25.60 449 2 3.4409 0.0066 99.50 Hbl 16642-08A 343.46 25.60 485 3 0.0982 0.0021 99.26 Bt 16642-09A 343.46 25.60 481 4 0.1487 0.0012 99.18 Bt 16642-10A 343.46 25.60 483 2 0.0642 0.0056 99.54 Bt 17301-01A 343.46 25.60 314 3 0.1380 0.0120 88.80 Bt 17301-02A 343.46 25.60 313 2 0.0188 0.0400 97.00 Bt 17301-03A 343.46 25.60 93 8 0.3893 0.0100 19.40 Bt 17301-04A 343.46 25.60 189 4 -0.0047 0.0140 70.70 Bt 17301-05A 343.46 25.60 580 16 1.1153 0.0060 56.80 Bt 16634-01A 348.26 25.64 438 4 -0.0218 0.0010 99.00 Bt 16634-02A 348.26 25.64 497 7 0.0671 0.0006 99.27 Bt 16634-03A 348.26 25.64 485 6 -0.0044 0.0008 91.69 Bt 16634-04A 348.26 25.64 204 22 6.2305 0.0001 38.23 Hbl 16634-05A 348.26 25.64 501 20 0.8031 0.0002 88.36 Bt

170

16634-06A 348.26 25.64 493 3 0.0218 0.0023 99.46 Bt 16634-07A 348.26 25.64 482 3 0.0182 0.0030 99.01 Bt 16634-08A 348.26 25.64 473 3 0.1107 0.0017 99.13 Bt 16634-09A 348.26 25.64 487 3 0.2421 0.0025 99.52 Bt 16634-10A 348.26 25.64 498 3 17.3643 0.0027 99.14 Hbl 16627-01A 352.96 25.69 263 2 0.0098 0.0055 98.24 Bt 16627-02A 352.96 25.69 205 5 0.0134 0.0008 78.21 Bt 16627-03A 352.96 25.69 202 25 0.0051 0.0002 88.01 Bt 16627-04A 352.96 25.69 196 8 0.3314 0.0005 83.78 Bt 16627-05A 352.96 25.69 130 3 -0.0095 0.0015 85.13 Bt 16627-06A 352.96 25.69 98 1 0.0577 0.0033 89.97 Bt 16627-07A 352.96 25.69 216 3 0.0562 0.0017 84.24 Bt 16627-08A 352.96 25.69 342 9 0.1359 0.0005 63.49 Bt 16627-09A 352.96 25.69 102 3 0.0018 0.0013 92.99 Bt 16627-10A 352.96 25.69 354 14 0.3966 0.0003 78.33 Bt 16627-11A 352.96 25.69 103 2 0.0900 0.0022 96.93 Bt 16626-01A 358.21 25.74 378 4 -0.0082 0.0014 97.50 Bt 16626-02A 358.21 25.74 469 5 0.0221 0.0012 97.41 Bt 16626-03A 358.21 25.74 491 3 0.0222 0.0028 98.20 Bt 16626-04A 358.21 25.74 460 3 0.0231 0.0045 98.32 Bt 16626-05A 358.21 25.74 484 4 0.0096 0.0019 98.37 Bt 16626-06A 358.21 25.74 424 9 0.1275 0.0006 72.77 Bt 16626-07A 358.21 25.74 511 5 0.0023 0.0012 99.43 Bt 16626-08A 358.21 25.74 485 3 0.0083 0.0036 98.87 Bt 16626-09A 358.21 25.74 469 2 0.0029 0.0047 99.82 Bt 17302-01A 358.21 25.74 446 5 0.0447 0.0090 92.50 Bt 17302-02A 358.21 25.74 427 6 0.0631 0.0080 85.90 Bt 17302-03A 358.21 25.74 254 2 0.2701 0.0370 89.50 Bt 17302-04A 358.21 25.74 111 5 0.0526 0.0070 64.40 Bt 17302-05A 358.21 25.74 132 9 0.4260 0.0050 47.60 Bt 17303-01A 358.21 25.74 219 1 0.4648 0.0600 93.60 Bt 17303-02A 358.21 25.74 220 2 0.0341 0.0290 95.40 Bt 17303-03A 358.21 25.74 204 4 0.1211 0.0090 72.60 Bt 17303-04A 358.21 25.74 388 6 0.1473 0.0070 89.70 Bt 16633-01A 360.80 25.76 313 2 0.0194 0.0031 97.72 Bt 16633-02A 360.80 25.76 114 1 0.0145 0.0032 95.34 Bt 16633-03A 360.80 25.76 359 4 0.0258 0.0010 100.52 Bt 16633-04A 360.80 25.76 111 2 0.0279 0.0016 97.30 Bt 16633-05A 360.80 25.76 469 3 0.0110 0.0023 98.54 Bt 16633-06A 360.80 25.76 478 2 0.0139 0.0074 99.23 Bt 16633-07A 360.80 25.76 445 5 0.0617 0.0024 94.40 Bt 16633-08A 360.80 25.76 200 1 0.0614 0.0108 98.39 Bt 16633-09A 360.80 25.76 110 5 0.0013 0.0008 97.40 Bt

171

16633-10A 360.80 25.76 247 9 0.0351 0.0005 60.88 Bt 16633-11A 360.80 25.76 457 13 -0.0871 0.0003 70.65 Bt 16633-12A 360.80 25.76 122 7 0.0119 0.0005 52.45 Bt 16633-13A 360.80 25.76 291 1 0.0294 0.0297 99.09 Bt 16628-01A 368.91 25.84 100 1 0.0081 0.0045 92.73 Bt 16628-02A 368.91 25.84 115 1 0.0558 0.0093 72.78 Bt 16628-03A 368.91 25.84 319 5 0.0319 0.0011 94.48 Bt 16628-04A 368.91 25.84 460 2 0.0220 0.0062 99.23 Bt 16628-05A 368.91 25.84 205 9 0.0238 0.0005 84.91 Bt 16628-06A 368.91 25.84 321 2 0.0316 0.0035 98.10 Bt 16628-07A 368.91 25.84 412 9 0.0348 0.0006 73.70 Bt 16628-08A 368.91 25.84 317 3 0.0151 0.0022 94.63 Bt 17304-01A 368.91 25.84 415 3 0.0383 0.0320 93.40 Bt 17304-02A 368.91 25.84 450 4 0.0757 0.0170 93.60 Bt 17304-03A 368.91 25.84 455 3 0.1869 0.0390 95.90 Bt 17305-01A 368.91 25.84 418 36 0.9727 0.0010 80.40 Bt 16623-01A 370.66 25.86 447 5 0.0558 0.0014 98.64 Bt 16623-02A 370.66 25.86 347 8 0.0505 0.0008 94.32 Bt 16623-03A 370.66 25.86 229 4 0.0250 0.0013 93.28 Bt 16623-04A 370.66 25.86 460 6 -0.0121 0.0011 94.01 Bt 16623-05A 370.66 25.86 100 3 0.0372 0.0015 85.30 Bt 16625-01A 374.91 25.90 111 2 0.0309 0.0028 95.30 Bt 16625-02A 374.91 25.90 496 8 0.0133 0.0008 97.04 Bt 16625-03A 374.91 25.90 491 4 0.1985 0.0019 96.78 Bt 16625-04A 374.91 25.90 481 15 0.0014 0.0004 93.31 Bt 16625-05A 374.91 25.90 498 5 0.0140 0.0012 95.93 Bt 16625-06A 374.91 25.90 397 7 0.0077 0.0008 91.42 Bt

172

Appendix 4-1 – oxygen isotope conversions

Cib converted values +0.64 Depth N. N. r1 N. r2 O. C. C. r1 Avg Calcite Age 50.46 2.38 2.41 2.40 3.04 -24.792 50.52 2.43 2.43 3.07 -24.799 50.58 2.25 2.25 2.89 -24.805 50.62 2.40 2.40 3.04 -24.810 50.7 2.37 2.37 3.01 -24.819 50.74 2.53 2.46 2.50 3.14 -24.823 50.75 3.13 -24.824 51.02 2.19 2.19 2.83 -24.854 51.10 1.99 1.99 2.63 -24.863 51.14 2.04 2.04 2.68 -24.867 51.2 2.21 2.21 2.85 -24.874 51.38 2.15 2.15 2.79 -24.893 51.44 2.37 2.37 3.01 -24.900 51.51 2.12 2.12 2.76 -24.908 51.65 2.29 2.29 2.93 -24.923 51.71 2.45 2.45 3.09 -24.930 51.77 2.40 2.40 3.04 -24.936 51.84 2.35 2.35 2.99 -24.944 51.99 2.27 2.27 2.91 -24.960 52.05 2.29 2.29 2.93 -24.967 52.11 2.23 2.23 2.87 -24.974 52.15 2.38 2.38 3.02 -24.978 52.23 2.47 2.23 2.45 2.38 3.02 -24.987 52.25 3.23 -24.989 52.3 2.54 2.54 3.18 -24.995 52.42 2.27 2.27 2.91 -25.008 52.48 2.35 2.35 2.99 -25.014 52.60 2.15 2.15 2.79 -25.028 52.64 2.34 2.34 2.98 -25.032 52.79 2.43 2.43 3.07 -25.048 52.85 2.52 2.52 3.16 -25.055 52.91 2.40 2.40 3.04 -25.062 52.96 2.42 2.41 2.41 3.05 -25.067 53.00 3.04 -25.072 53.02 2.43 2.43 3.07 -25.074 53.09 2.39 2.39 3.03 -25.081 53.14 2.45 2.45 3.09 -25.087 53.43 2.32 2.32 2.96 -25.119 53.49 2.30 2.30 2.94 -25.125 53.55 2.35 2.35 2.99 -25.132 53.62 2.34 2.34 2.98 -25.140

173

53.71 2.38 2.38 3.02 -25.150 53.73 2.33 2.33 2.97 -25.152 53.75 3.26 -25.154 53.78 2.50 2.50 3.14 -25.157 53.84 2.36 2.36 3.00 -25.164 53.9 2.22 2.22 2.86 -25.171 53.96 2.42 2.42 3.06 -25.177 54.11 2.50 2.46 2.48 3.12 -25.194 54.18 2.36 2.36 3.00 -25.201 54.23 2.28 2.28 2.92 -25.207 54.3 2.28 2.28 2.92 -25.215 54.34 2.33 2.33 2.97 -25.219 54.41 2.28 2.28 2.92 -25.227 54.46 2.41 2.41 3.05 -25.232 54.52 2.34 2.34 2.98 -25.239 54.7 2.54 2.54 3.18 -25.259 54.76 2.27 2.27 2.91 -25.266 54.81 2.28 2.28 2.92 -25.274 54.84 2.17 2.17 2.81 -25.278 54.93 2.27 2.27 2.91 -25.293 54.99 2.16 1.99 2.23 2.13 2.77 -25.302 55.06 2.16 2.16 2.80 -25.312 55.11 2.20 2.20 2.84 -25.319 55.18 2.05 2.05 2.69 -25.328 55.22 2.03 2.03 2.67 -25.333 55.25 2.97 -25.337 55.29 2.09 2.09 2.73 -25.343 55.37 2.15 2.15 2.79 -25.353 55.41 1.85 1.85 2.49 -25.359 55.48 2.01 2.01 2.65 -25.368 55.53 2.27 1.94 2.13 2.29 2.16 2.80 -25.375 55.61 2.06 2.18 2.12 2.76 -25.386 55.67 2.24 2.24 2.88 -25.394 55.78 2.28 2.28 2.92 -25.408 55.79 2.08 2.08 2.72 -25.410 55.85 2.22 2.22 2.86 -25.418 55.91 2.34 2.34 2.98 -25.426 55.96 2.33 2.33 2.97 -25.432 56.00 3.02 -25.438 56.02 2.29 2.29 2.93 -25.440 56.08 2.26 2.26 2.90 -25.448 56.2 2.63 2.63 3.27 -25.464 56.26 2.47 2.47 3.11 -25.472 56.32 2.50 2.50 3.14 -25.480 56.38 2.42 2.42 3.06 -25.488 56.43 2.44 2.51 2.47 2.55 2.49 3.13 -25.495

174

56.49 2.49 2.49 3.13 -25.503 56.55 2.22 2.22 2.86 -25.511 56.67 2.46 2.46 3.10 -25.527 56.75 2.55 2.54 2.54 3.18 -25.538 56.81 2.46 2.46 3.10 -25.546 56.88 2.47 2.47 3.11 -25.555 56.96 2.35 2.35 2.99 -25.566 57.01 2.32 2.32 2.96 -25.573 57.06 2.28 2.28 2.92 -25.579 57.10 2.19 2.19 2.83 -25.585 57.11 2.36 2.36 3.00 -25.586 57.16 2.25 2.29 2.27 2.91 -25.593 57.29 1.93 1.93 2.57 -25.610 57.35 2.37 2.37 3.01 -25.618 57.41 2.31 2.31 2.95 -25.626 57.46 2.45 2.45 3.09 -25.633 57.50 2.89 -25.638 57.52 2.27 2.27 2.91 -25.641 57.58 2.44 1.82 2.13 2.77 -25.649 57.62 2.50 2.50 3.14 -25.654 57.7 2.39 2.39 3.03 -25.665 57.76 2.35 2.29 2.32 2.96 -25.673 57.81 2.32 2.32 2.96 -25.680 57.87 2.44 2.44 3.08 -25.688 57.99 2.45 2.45 3.09 -25.704 58.05 2.37 2.37 3.01 -25.712 58.11 2.31 2.31 2.95 -25.720 58.17 1.88 1.88 2.52 -25.728 58.21 2.40 2.09 2.24 2.88 -25.733 58.25 2.91 -25.738 58.29 2.21 2.21 2.85 -25.744 58.33 2.48 2.48 3.12 -25.749 58.4 2.49 2.49 3.13 -25.758 58.58 2.38 2.38 3.02 -25.783 58.60 2.29 2.29 2.93 -25.785 58.61 2.29 2.29 2.93 -25.787 58.67 2.26 2.26 2.90 -25.795 58.73 2.13 2.27 2.51 2.30 2.94 -25.803 58.79 2.32 2.32 2.96 -25.811 58.84 2.42 2.42 3.06 -25.817 58.91 2.31 2.31 2.95 -25.827 58.96 2.26 2.26 2.90 -25.833 59 2.44 2.19 2.31 2.95 -25.839 59.08 2.46 2.46 3.10 -25.849 59.13 2.54 2.54 3.18 -25.856 59.19 2.24 2.24 2.88 -25.864

175

59.23 2.46 2.51 2.48 3.12 -25.869 59.33 2.07 2.07 2.71 -25.883 59.46 2.32 2.32 2.96 -25.900 59.52 2.27 2.27 2.91 -25.908 59.58 2.14 2.14 2.78 -25.916 59.64 2.31 2.31 2.95 -25.924 59.71 2.25 2.24 2.37 2.29 2.93 -25.934 59.75 2.88 -25.939 59.76 2.27 2.27 2.91 -25.940 59.82 2.13 2.13 2.77 -25.948 59.88 2.18 2.18 2.82 -25.956 59.94 2.23 2.23 2.87 -25.964 60.2 2.54 2.54 3.18 -26.031 60.32 2.74 2.74 3.38 -26.090 60.36 2.66 2.66 3.30 -26.110 60.43 2.31 2.31 2.95 -26.144 60.46 2.96 -26.159 60.51 2.50 2.50 3.14 -26.184 60.57 2.57 2.57 3.21 -26.213 60.63 2.30 2.30 2.94 -26.243 60.69 2.14 2.14 2.78 -26.272 60.75 2.31 2.31 2.95 -26.302 60.80 2.51 2.51 3.15 -26.327 60.81 2.47 2.47 3.11 -26.331 60.91 2.57 2.57 3.21 -26.381 60.94 2.61 2.59 2.60 3.24 -26.395 61.05 2.41 2.41 3.05 -26.428 61.1 2.68 2.68 3.32 -26.435 61.17 2.64 2.64 3.28 -26.445 61.20 3.11 -26.449 61.23 2.41 2.41 3.05 -26.454 61.31 2.47 2.47 3.11 -26.465 61.37 2.28 2.28 2.92 -26.473 61.39 2.47 2.47 3.11 -26.476 61.49 2.39 2.39 3.03 -26.490 61.67 2.73 2.73 3.37 -26.515 61.71 2.77 2.77 3.41 -26.521 61.77 2.50 2.50 3.14 -26.529 61.83 2.57 2.57 3.21 -26.537 61.89 2.51 2.35 2.43 3.07 -26.546 61.94 2.53 2.53 3.17 -26.553 62.01 2.29 2.29 2.93 -26.563 62.09 2.38 2.24 2.31 2.95 -26.574 62.14 2.47 2.47 3.11 -26.581 62.19 2.26 2.26 2.90 -26.588 62.24 2.30 2.30 2.94 -26.595

176

62.30 2.41 2.41 3.05 -26.603 62.38 2.42 2.42 3.06 -26.614 62.45 2.53 2.53 3.17 -26.624 62.49 2.71 2.71 3.35 -26.630 62.55 2.50 2.59 2.55 3.19 -26.638 62.61 2.62 2.62 3.26 -26.646 62.68 2.77 2.77 3.41 -26.656 62.7 2.68 2.36 2.52 3.16 -26.659 62.73 2.62 2.62 3.26 -26.663 62.85 2.06 2.69 2.38 3.02 -26.680 62.91 2.60 2.60 3.24 -26.688 62.97 2.61 2.61 3.25 -26.697 63.03 2.51 2.51 3.15 -26.705 63.15 2.44 2.44 3.08 -26.722 63.21 2.38 2.38 3.02 -26.730 63.25 2.21 2.21 2.85 -26.736 63.33 2.48 2.48 3.12 -26.747 63.39 2.56 2.56 3.20 -26.755 63.43 2.68 2.68 3.32 -26.761 63.46 3.59 -26.765 63.51 2.59 2.59 3.23 -26.772 63.57 2.81 2.81 3.45 -26.781 63.61 2.87 2.87 3.51 -26.786 63.69 2.76 2.76 3.40 -26.797 63.75 2.52 2.52 3.16 -26.806 63.81 2.56 2.56 3.20 -26.814 63.84 2.70 2.70 3.34 -26.818 63.93 2.36 2.36 3.00 -26.831 63.99 2.15 2.15 2.79 -26.839 64.05 2.39 2.17 2.28 2.92 -26.848 64.11 2.27 2.27 2.91 -26.856 64.19 2.56 2.56 3.20 -26.867 64.20 3.07 -26.869 64.23 2.41 2.50 2.45 3.09 -26.873 64.29 2.41 2.41 3.05 -26.881 64.33 2.66 2.66 3.30 -26.887 64.41 2.72 2.72 3.36 -26.898 64.65 2.60 2.60 3.24 -26.932 64.71 2.43 2.43 3.07 -26.940 64.77 2.65 2.65 3.29 -26.948 64.83 2.45 2.47 2.49 2.47 3.11 -26.957 64.89 2.44 2.44 3.08 -26.965 64.95 2.31 2.31 2.95 -26.974 65.01 2.50 2.50 3.14 -26.982 65.09 2.32 2.32 2.96 -26.993 65.13 2.47 2.47 3.11 -26.999

177

65.19 2.32 2.32 2.96 -27.007 65.22 2.24 2.24 2.88 -27.011 65.30 2.27 2.27 2.91 -27.022 65.31 2.32 2.32 2.96 -27.024 65.34 2.28 2.28 2.92 -27.028 65.43 2.37 2.37 3.01 -27.041 65.49 2.40 2.40 3.04 -27.049 65.54 1.90 1.90 2.54 -27.056 65.6 2.63 2.66 2.65 3.29 -27.064 65.68 2.03 2.03 2.67 -27.076 65.73 2.53 2.53 3.17 -27.083 65.79 2.37 2.37 3.01 -27.091 65.85 2.35 2.35 2.99 -27.099 65.91 2.19 2.19 2.83 -27.108 65.97 2.09 2.20 2.15 2.79 -27.116 66.03 2.49 2.49 3.13 -27.124 66.14 2.40 2.40 3.04 -27.140 66.22 2.03 2.03 2.67 -27.151 66.29 2.01 2.01 2.65 -27.161 66.33 2.28 2.28 2.92 -27.166 66.39 2.31 2.31 2.95 -27.175 66.44 2.25 2.25 2.89 -27.182 66.51 2.35 2.35 2.99 -27.192 66.57 2.49 2.49 3.13 -27.200 66.63 2.37 2.37 3.01 -27.208 66.64 2.39 2.39 3.03 -27.210 66.69 2.26 2.26 2.90 -27.217 66.75 2.57 2.59 2.58 3.22 -27.225 66.81 2.67 2.67 3.31 -27.234 66.89 2.57 2.29 2.47 2.44 3.08 -27.245 66.93 2.32 2.32 2.96 -27.250 66.99 2.28 2.28 2.92 -27.259 67.04 2.27 2.32 2.26 2.28 2.92 -27.266 67.17 2.20 2.22 2.21 2.85 -27.284 67.20 2.85 -27.288 67.24 2.51 2.53 2.26 2.43 3.07 -27.294 67.31 2.32 2.32 2.96 -27.303 67.38 2.08 2.09 2.09 2.73 -27.313 67.4 2.26 2.26 2.90 -27.316 67.47 2.36 2.36 3.00 -27.326 67.53 2.48 2.48 3.12 -27.334 67.83 2.39 2.39 3.03 -27.376 67.94 2.65 2.65 3.29 -27.391 68.01 2.58 2.58 3.22 -27.401 68.07 2.44 2.44 3.08 -27.410 68.13 2.42 2.42 3.06 -27.418

178

68.19 2.27 2.27 2.91 -27.426 68.29 2.15 2.15 2.79 -27.440 68.30 1.87 1.87 2.51 -27.442 68.32 1.97 1.97 2.61 -27.445 68.38 2.09 2.09 2.73 -27.453 68.43 2.21 2.21 2.85 -27.460 68.49 2.31 2.20 2.27 2.26 2.90 -27.468 68.55 2.11 2.11 2.75 -27.477 68.61 2.21 2.31 2.26 2.90 -27.485 68.67 2.09 2.09 2.73 -27.494 68.73 2.10 2.10 2.74 -27.502 68.79 2.13 2.08 2.11 2.75 -27.510 68.83 2.11 2.11 2.75 -27.516 68.89 2.17 2.17 2.81 -27.524 68.97 2.23 2.23 2.87 -27.535 69.03 2.01 2.01 2.65 -27.544 69.15 2.57 2.57 3.21 -27.561 69.25 2.48 2.48 3.12 -27.575 69.33 2.46 2.46 3.10 -27.586 69.39 2.40 2.40 3.04 -27.594 69.44 2.46 2.46 3.10 -27.601 69.46 2.91 -27.604 69.51 2.49 2.49 3.13 -27.611 69.51 2.51 2.51 3.15 -27.611 69.57 2.34 2.34 2.98 -27.619 69.63 2.40 2.40 3.04 -27.628

179

Appendix 4-2 – Apparent sea-level and ice volume estimates

Pekar calibration Avg Avg Depth Age d18O ASL wet %AIS ASL dry %AIS ASL %AIS 50.46 -24.792 3.04 -1.38 102.08 19.46 70.52 9.04 86.30 50.52 -24.799 3.07 -4.11 106.23 16.72 74.67 6.30 90.45 50.58 -24.805 2.89 11.22 83.00 32.05 51.43 21.64 67.22 50.62 -24.810 3.04 -1.70 102.57 19.14 71.00 8.72 86.79 50.7 -24.819 3.01 0.89 98.66 21.72 67.09 11.30 82.87 50.74 -24.823 3.14 -9.61 114.57 11.22 83.00 0.80 98.78 50.75 -24.824 3.13 -9.17 113.89 11.67 82.32 1.25 98.11 51.02 -24.854 2.83 15.58 76.39 36.42 44.82 26.00 60.61 51.10 -24.863 2.63 32.50 50.76 53.33 19.19 42.92 34.97 51.14 -24.867 2.68 28.00 57.58 48.83 26.01 38.42 41.79 51.2 -24.874 2.85 14.17 78.54 35.00 46.97 24.58 62.75 51.38 -24.893 2.79 19.47 70.50 40.30 38.93 29.89 54.72 51.44 -24.900 3.01 1.14 98.28 21.97 66.71 11.55 82.49 51.51 -24.908 2.76 21.92 66.79 42.75 35.23 32.33 51.01 51.65 -24.923 2.93 7.80 88.18 28.64 56.61 18.22 72.39 51.71 -24.930 3.09 -6.11 109.26 14.72 77.70 4.30 93.48 51.77 -24.936 3.04 -1.45 102.19 19.39 70.63 8.97 86.41 51.84 -24.944 2.99 2.58 96.09 23.42 64.52 13.00 80.30 51.99 -24.960 2.91 9.47 85.65 30.30 54.09 19.89 69.87 52.05 -24.967 2.93 7.30 88.93 28.14 57.37 17.72 73.15 52.11 -24.974 2.87 12.75 80.68 33.58 49.12 23.17 64.90 52.15 -24.978 3.02 0.39 99.41 21.22 67.85 10.80 83.63 52.23 -24.987 3.02 -0.21 100.32 20.62 68.76 10.20 84.54 52.25 -24.989 3.23 -17.50 126.52 3.33 94.95 -7.08 110.73 52.3 -24.995 3.18 -13.11 119.87 7.72 88.30 -2.70 104.09 52.42 -25.008 2.91 8.80 86.66 29.64 55.10 19.22 70.88 52.48 -25.014 2.99 2.39 96.38 23.22 64.82 12.80 80.60 52.60 -25.028 2.79 19.17 70.96 40.00 39.39 29.58 55.18 52.64 -25.032 2.98 3.55 94.62 24.39 63.05 13.97 78.83 52.79 -25.048 3.07 -4.03 106.11 16.80 74.54 6.39 90.32 52.85 -25.055 3.16 -11.36 117.22 9.47 85.65 -0.95 101.43 52.91 -25.062 3.04 -1.70 102.57 19.14 71.00 8.72 86.79 52.96 -25.067 3.05 -2.82 104.28 18.01 72.71 7.60 88.49 53.00 -25.072 3.04 -1.67 102.53 19.17 70.96 8.75 86.74 53.02 -25.074 3.07 -4.11 106.23 16.72 74.67 6.30 90.45 53.09 -25.081 3.03 -0.95 101.43 19.89 69.87 9.47 85.65 53.14 -25.087 3.09 -5.70 108.63 15.14 77.07 4.72 92.85 53.43 -25.119 2.96 5.22 92.09 26.05 60.53 15.64 76.31 53.49 -25.125 2.94 7.00 89.39 27.83 57.83 17.42 73.61 53.55 -25.132 2.99 2.64 96.00 23.47 64.44 13.05 80.22

180

53.62 -25.140 2.98 3.30 94.99 24.14 63.43 13.72 79.21 53.71 -25.150 3.02 -0.36 100.55 20.47 68.98 10.05 84.77 53.73 -25.152 2.97 4.05 93.86 24.89 62.29 14.47 78.08 53.75 -25.154 3.26 -20.00 130.30 0.83 98.74 -9.58 114.52 53.78 -25.157 3.14 -9.61 114.57 11.22 83.00 0.80 98.78 53.84 -25.164 3.00 1.80 97.27 22.64 65.70 12.22 81.48 53.9 -25.171 2.86 13.42 79.67 34.25 48.11 23.83 63.89 53.96 -25.177 3.06 -3.28 104.97 17.55 73.40 7.14 89.19 54.11 -25.194 3.12 -8.07 112.23 12.76 80.66 2.35 96.45 54.18 -25.201 3.00 1.97 97.02 22.80 65.45 12.39 81.23 54.23 -25.207 2.92 8.30 87.42 29.14 55.85 18.72 71.64 54.3 -25.215 2.92 8.05 87.80 28.89 56.23 18.47 72.02 54.34 -25.219 2.97 4.17 93.69 25.00 62.12 14.58 77.90 54.41 -25.227 2.92 8.33 87.37 29.17 55.81 18.75 71.59 54.46 -25.232 3.05 -2.25 103.41 18.58 71.84 8.17 87.63 54.52 -25.239 2.98 2.97 95.50 23.80 63.93 13.39 79.72 54.7 -25.259 3.18 -13.53 120.50 7.30 88.93 -3.11 104.72 54.76 -25.266 2.91 9.14 86.16 29.97 54.59 19.55 70.37 54.81 -25.274 2.92 8.05 87.80 28.89 56.23 18.47 72.02 54.84 -25.278 2.81 17.72 73.15 38.55 41.59 28.14 57.37 54.93 -25.293 2.91 8.80 86.66 29.64 55.10 19.22 70.88 54.99 -25.302 2.77 21.19 67.89 42.03 36.32 31.61 52.11 55.06 -25.312 2.80 18.05 72.65 38.89 41.08 28.47 56.86 55.11 -25.319 2.84 14.80 77.57 35.64 46.00 25.22 61.79 55.18 -25.328 2.69 27.39 58.50 48.22 26.94 37.80 42.72 55.22 -25.333 2.67 29.39 55.47 50.22 23.91 39.80 39.69 55.25 -25.337 2.97 4.17 93.69 25.00 62.12 14.58 77.90 55.29 -25.343 2.73 23.80 63.93 44.64 32.37 34.22 48.15 55.37 -25.353 2.79 19.22 70.88 40.05 39.31 29.64 55.10 55.41 -25.359 2.49 44.05 33.25 64.89 1.69 54.47 17.47 55.48 -25.368 2.65 31.22 52.70 52.05 21.13 41.64 36.91 55.53 -25.375 2.80 18.67 71.71 39.51 40.14 29.09 55.93 55.61 -25.386 2.76 21.94 66.75 42.78 35.19 32.36 50.97 55.67 -25.394 2.88 12.05 81.74 32.89 50.17 22.47 65.95 55.78 -25.408 2.92 8.00 87.88 28.83 56.31 18.42 72.10 55.79 -25.410 2.72 24.97 62.17 45.80 30.60 35.39 46.38 55.85 -25.418 2.86 13.72 79.21 34.55 47.65 24.14 63.43 55.91 -25.426 2.98 3.30 94.99 24.14 63.43 13.72 79.21 55.96 -25.432 2.97 4.22 93.61 25.05 62.04 14.64 77.82 56.00 -25.438 3.02 0.00 100.00 20.83 68.43 10.42 84.22 56.02 -25.440 2.93 7.30 88.93 28.14 57.37 17.72 73.15 56.08 -25.448 2.90 10.39 84.26 31.22 52.70 20.80 68.48 56.2 -25.464 3.27 -20.78 131.48 0.05 99.92 -10.36 115.70 56.26 -25.472 3.11 -7.78 111.79 13.05 80.22 2.64 96.00 56.32 -25.480 3.14 -10.28 115.58 10.55 84.01 0.14 99.79 56.38 -25.488 3.06 -3.53 105.35 17.30 73.78 6.89 89.57

181

56.43 -25.495 3.13 -9.29 114.08 11.54 82.52 1.12 98.30 56.49 -25.503 3.13 -9.28 114.06 11.55 82.49 1.14 98.28 56.55 -25.511 2.86 13.05 80.22 33.89 48.66 23.47 64.44 56.67 -25.527 3.10 -6.53 109.89 14.30 78.33 3.89 94.11 56.75 -25.538 3.18 -13.68 120.73 7.15 89.16 -3.26 104.95 56.81 -25.546 3.10 -6.53 109.89 14.30 78.33 3.89 94.11 56.88 -25.555 3.11 -7.78 111.79 13.05 80.22 2.64 96.00 56.96 -25.566 2.99 2.89 95.63 23.72 64.06 13.30 79.84 57.01 -25.573 2.96 5.05 92.34 25.89 60.78 15.47 76.56 57.06 -25.579 2.92 8.55 87.04 29.39 55.47 18.97 71.26 57.10 -25.585 2.83 15.83 76.01 36.67 44.44 26.25 60.23 57.11 -25.586 3.00 1.64 97.52 22.47 65.95 12.05 81.74 57.16 -25.593 2.91 9.26 85.97 30.10 54.40 19.68 70.18 57.29 -25.610 2.57 37.39 43.35 58.22 11.79 47.80 27.57 57.35 -25.618 3.01 0.47 99.29 21.30 67.72 10.89 83.50 57.41 -25.626 2.95 5.72 91.33 26.55 59.77 16.14 75.55 57.46 -25.633 3.09 -6.11 109.26 14.72 77.70 4.30 93.48 57.50 -25.638 2.89 10.83 83.59 31.67 52.02 21.25 67.80 57.52 -25.641 2.91 9.47 85.65 30.30 54.09 19.89 69.87 57.58 -25.649 2.77 20.76 68.54 41.60 36.98 31.18 52.76 57.62 -25.654 3.14 -9.86 114.94 10.97 83.38 0.55 99.16 57.7 -25.665 3.03 -0.78 101.18 20.05 69.62 9.64 85.40 57.76 -25.673 2.96 4.93 92.53 25.76 60.97 15.35 76.75 57.81 -25.680 2.96 4.97 92.47 25.80 60.90 15.39 76.69 57.87 -25.688 3.08 -4.95 107.49 15.89 75.93 5.47 91.71 57.99 -25.704 3.09 -5.61 108.50 15.22 76.94 4.80 92.72 58.05 -25.712 3.01 0.83 98.74 21.67 67.17 11.25 82.95 58.11 -25.720 2.95 5.72 91.33 26.55 59.77 16.14 75.55 58.17 -25.728 2.52 41.55 37.04 62.39 5.47 51.97 21.26 58.21 -25.733 2.88 11.32 82.85 32.15 51.29 21.74 67.07 58.25 -25.738 2.91 9.17 86.11 30.00 54.55 19.58 70.33 58.29 -25.744 2.85 14.47 78.08 35.30 46.51 24.89 62.29 58.33 -25.749 3.12 -8.33 112.63 12.50 81.06 2.08 96.84 58.4 -25.758 3.13 -9.08 113.76 11.75 82.20 1.33 97.98 58.58 -25.783 3.02 -0.11 100.17 20.72 68.61 10.30 84.39 58.60 -25.785 2.93 7.50 88.64 28.33 57.07 17.92 72.85 58.61 -25.787 2.93 7.55 88.56 28.39 56.99 17.97 72.77 58.67 -25.795 2.90 10.14 84.64 30.97 53.08 20.55 68.86 58.73 -25.803 2.94 6.26 90.52 27.09 58.95 16.67 74.74 58.79 -25.811 2.96 4.89 92.60 25.72 61.03 15.30 76.81 58.84 -25.817 3.06 -3.53 105.35 17.30 73.78 6.89 89.57 58.91 -25.827 2.95 6.14 90.70 26.97 59.14 16.55 74.92 58.96 -25.833 2.90 10.22 84.52 31.05 52.95 20.64 68.73 59 -25.839 2.95 5.57 91.56 26.40 60.00 15.99 75.78 59.08 -25.849 3.10 -6.70 110.15 14.14 78.58 3.72 94.36 59.13 -25.856 3.18 -13.03 119.74 7.80 88.18 -2.61 103.96

182

59.19 -25.864 2.88 11.97 81.86 32.80 50.30 22.39 66.08 59.23 -25.869 3.12 -8.64 113.09 12.19 81.53 1.78 97.31 59.33 -25.883 2.71 25.72 61.03 46.55 29.46 36.14 45.25 59.46 -25.900 2.96 5.08 92.30 25.92 60.73 15.50 76.52 59.52 -25.908 2.91 9.55 85.53 30.39 53.96 19.97 69.74 59.58 -25.916 2.78 19.89 69.87 40.72 38.30 30.30 54.09 59.64 -25.924 2.95 5.55 91.59 26.39 60.02 15.97 75.80 59.71 -25.934 2.93 7.61 88.47 28.44 56.91 18.03 72.69 59.75 -25.939 2.88 11.67 82.32 32.50 50.76 22.08 66.54 59.76 -25.940 2.91 8.97 86.41 29.80 54.84 19.39 70.63 59.82 -25.948 2.77 21.05 68.10 41.89 36.54 31.47 52.32 59.88 -25.956 2.82 16.72 74.67 37.55 43.10 27.14 58.88 59.94 -25.964 2.87 12.47 81.11 33.30 49.54 22.89 65.32 60.2 -26.031 3.18 -13.03 119.74 7.80 88.18 -2.61 103.96 60.32 -26.090 3.38 -30.25 145.83 -9.42 114.27 -19.83 130.05 60.36 -26.110 3.30 -23.17 135.10 -2.33 103.54 -12.75 119.32 60.43 -26.144 2.95 5.67 91.41 26.50 59.85 16.08 75.63 60.46 -26.159 2.96 5.00 92.42 25.83 60.86 15.42 76.64 60.51 -26.184 3.14 -9.67 114.65 11.17 83.08 0.75 98.86 60.57 -26.213 3.21 -15.67 123.74 5.17 92.17 -5.25 107.95 60.63 -26.243 2.94 7.05 89.31 27.89 57.75 17.47 73.53 60.69 -26.272 2.78 19.67 70.20 40.50 38.64 30.08 54.42 60.75 -26.302 2.95 6.08 90.78 26.92 59.22 16.50 75.00 60.80 -26.327 3.15 -10.83 116.41 10.00 84.85 -0.42 100.63 60.81 -26.331 3.11 -7.58 111.49 13.25 79.92 2.83 95.71 60.91 -26.381 3.21 -15.67 123.74 5.17 92.17 -5.25 107.95 60.94 -26.395 3.24 -18.21 127.59 2.62 96.02 -7.79 111.81 61.05 -26.428 3.05 -2.33 103.54 18.50 71.97 8.08 87.75 61.1 -26.435 3.32 -25.33 138.38 -4.50 106.82 -14.92 122.60 61.17 -26.445 3.28 -21.92 133.21 -1.08 101.64 -11.50 117.42 61.20 -26.449 3.11 -7.50 111.36 13.33 79.80 2.92 95.58 61.23 -26.454 3.05 -2.17 103.28 18.67 71.72 8.25 87.50 61.31 -26.465 3.11 -7.45 111.28 13.39 79.72 2.97 95.50 61.37 -26.473 2.92 8.30 87.42 29.14 55.85 18.72 71.64 61.39 -26.476 3.11 -7.45 111.28 13.39 79.72 2.97 95.50 61.49 -26.490 3.03 -0.70 101.06 20.14 69.49 9.72 85.27 61.67 -26.515 3.37 -28.86 143.73 -8.03 112.17 -18.45 127.95 61.71 -26.521 3.41 -32.61 149.41 -11.78 117.85 -22.20 133.63 61.77 -26.529 3.14 -9.86 114.94 10.97 83.38 0.55 99.16 61.83 -26.537 3.21 -16.11 124.41 4.72 92.85 -5.70 108.63 61.89 -26.546 3.07 -3.82 105.79 17.01 74.22 6.60 90.01 61.94 -26.553 3.17 -12.45 118.86 8.39 87.29 -2.03 103.08 62.01 -26.563 2.93 7.80 88.18 28.64 56.61 18.22 72.39 62.09 -26.574 2.95 5.76 91.27 26.60 59.70 16.18 75.49 62.14 -26.581 3.11 -7.61 111.54 13.22 79.97 2.80 95.75 62.19 -26.588 2.90 10.39 84.26 31.22 52.70 20.80 68.48

183

62.24 -26.595 2.94 6.30 90.45 27.14 58.88 16.72 74.67 62.30 -26.603 3.05 -2.50 103.79 18.33 72.22 7.92 88.01 62.38 -26.614 3.06 -2.95 104.46 17.89 72.90 7.47 88.68 62.45 -26.624 3.17 -12.86 119.49 7.97 87.92 -2.45 103.71 62.49 -26.630 3.35 -27.61 141.84 -6.78 110.27 -17.20 126.06 62.55 -26.638 3.19 -14.03 121.26 6.80 89.69 -3.61 105.47 62.61 -26.646 3.26 -19.70 129.84 1.14 98.28 -9.28 114.06 62.68 -26.656 3.41 -32.53 149.29 -11.70 117.72 -22.11 133.50 62.7 -26.659 3.16 -11.85 117.95 8.99 86.39 -1.43 102.17 62.73 -26.663 3.26 -20.03 130.35 0.80 98.78 -9.61 114.57 62.85 -26.680 3.02 0.39 99.41 21.22 67.85 10.80 83.63 62.91 -26.688 3.24 -18.20 127.57 2.64 96.00 -7.78 111.79 62.97 -26.697 3.25 -19.36 129.34 1.47 97.77 -8.95 113.56 63.03 -26.705 3.15 -10.53 115.95 10.30 84.39 -0.11 100.17 63.15 -26.722 3.08 -4.70 107.12 16.14 75.55 5.72 91.33 63.21 -26.730 3.02 0.14 99.79 20.97 68.23 10.55 84.01 63.25 -26.736 2.85 13.89 78.96 34.72 47.39 24.30 63.18 63.33 -26.747 3.12 -8.20 112.42 12.64 80.85 2.22 96.64 63.39 -26.755 3.20 -15.11 122.90 5.72 91.33 -4.70 107.12 63.43 -26.761 3.32 -24.86 137.67 -4.03 106.11 -14.45 121.89 63.46 -26.765 3.59 -47.50 171.97 -26.67 140.40 -37.08 156.19 63.51 -26.772 3.23 -17.53 126.56 3.30 94.99 -7.11 110.78 63.57 -26.781 3.45 -36.03 154.59 -15.20 123.03 -25.61 138.81 63.61 -26.786 3.51 -40.86 161.91 -20.03 130.35 -30.45 146.13 63.69 -26.797 3.40 -31.53 147.77 -10.70 116.21 -21.11 131.99 63.75 -26.806 3.16 -11.28 117.09 9.55 85.53 -0.86 101.31 63.81 -26.814 3.20 -14.95 122.65 5.89 91.08 -4.53 106.86 63.84 -26.818 3.34 -26.45 140.07 -5.61 108.50 -16.03 124.29 63.93 -26.831 3.00 1.64 97.52 22.47 65.95 12.05 81.74 63.99 -26.839 2.79 19.39 70.63 40.22 39.06 29.80 54.84 64.05 -26.848 2.92 8.30 87.42 29.14 55.85 18.72 71.64 64.11 -26.856 2.91 9.39 85.78 30.22 54.21 19.80 69.99 64.19 -26.867 3.20 -15.20 123.03 5.64 91.46 -4.78 107.24 64.20 -26.869 3.07 -4.17 106.31 16.67 74.75 6.25 90.53 64.23 -26.873 3.09 -6.24 109.45 14.60 77.89 4.18 93.67 64.29 -26.881 3.05 -2.78 104.21 18.05 72.65 7.64 88.43 64.33 -26.887 3.30 -23.53 135.65 -2.70 104.09 -13.11 119.87 64.41 -26.898 3.36 -28.53 143.23 -7.70 111.66 -18.11 127.44 64.65 -26.932 3.24 -18.36 127.82 2.47 96.26 -7.95 112.04 64.71 -26.940 3.07 -4.45 106.74 16.39 75.17 5.97 90.95 64.77 -26.948 3.29 -22.61 134.26 -1.78 102.70 -12.20 118.48 64.83 -26.957 3.11 -7.81 111.83 13.03 80.26 2.61 96.05 64.89 -26.965 3.08 -5.36 108.13 15.47 76.56 5.05 92.34 64.95 -26.974 2.95 5.89 91.08 26.72 59.52 16.30 75.30 65.01 -26.982 3.14 -10.20 115.45 10.64 83.88 0.22 99.67 65.09 -26.993 2.96 5.22 92.09 26.05 60.53 15.64 76.31

184

65.13 -26.999 3.11 -7.53 111.41 13.30 79.84 2.89 95.63 65.19 -27.007 2.96 4.72 92.85 25.55 61.28 15.14 77.07 65.22 -27.011 2.88 11.47 82.62 32.30 51.06 21.89 66.84 65.30 -27.022 2.91 9.17 86.11 30.00 54.55 19.58 70.33 65.31 -27.024 2.96 4.72 92.85 25.55 61.28 15.14 77.07 65.34 -27.028 2.92 8.14 87.67 28.97 56.11 18.55 71.89 65.43 -27.041 3.01 1.22 98.15 22.05 66.59 11.64 82.37 65.49 -27.049 3.04 -2.03 103.08 18.80 71.51 8.39 87.29 65.54 -27.056 2.54 40.14 39.19 60.97 7.62 50.55 23.40 65.6 -27.064 3.29 -22.36 133.88 -1.53 102.32 -11.95 118.10 65.68 -27.076 2.67 29.55 55.22 50.39 23.66 39.97 39.44 65.73 -27.083 3.17 -12.86 119.49 7.97 87.92 -2.45 103.71 65.79 -27.091 3.01 0.97 98.53 21.80 66.96 11.39 82.75 65.85 -27.099 2.99 2.14 96.76 22.97 65.20 12.55 80.98 65.91 -27.108 2.83 16.05 75.68 36.89 44.11 26.47 59.89 65.97 -27.116 2.79 19.47 70.50 40.30 38.93 29.89 54.72 66.03 -27.124 3.13 -9.45 114.31 11.39 82.75 0.97 98.53 66.14 -27.140 3.04 -2.03 103.08 18.80 71.51 8.39 87.29 66.22 -27.151 2.67 28.92 56.19 49.75 24.62 39.33 40.40 66.29 -27.161 2.65 30.80 53.33 51.64 21.76 41.22 37.55 66.33 -27.166 2.92 8.64 86.91 29.47 55.35 19.05 71.13 66.39 -27.175 2.95 5.97 90.95 26.80 59.39 16.39 75.17 66.44 -27.182 2.89 10.64 83.88 31.47 52.32 21.05 68.10 66.51 -27.192 2.99 2.22 96.64 23.05 65.07 12.64 80.85 66.57 -27.200 3.13 -9.28 114.06 11.55 82.49 1.14 98.28 66.63 -27.208 3.01 0.75 98.86 21.58 67.30 11.17 83.08 66.64 -27.210 3.03 -0.45 100.68 20.39 69.11 9.97 84.89 66.69 -27.217 2.90 9.64 85.40 30.47 53.83 20.05 69.62 66.75 -27.225 3.22 -16.68 125.28 4.15 93.71 -6.26 109.49 66.81 -27.234 3.31 -24.45 137.04 -3.61 105.47 -14.03 121.26 66.89 -27.245 3.08 -5.16 107.82 15.67 76.25 5.26 92.03 66.93 -27.250 2.96 4.89 92.60 25.72 61.03 15.30 76.81 66.99 -27.259 2.92 7.97 87.92 28.80 56.36 18.39 72.14 67.04 -27.266 2.92 8.04 87.82 28.87 56.26 18.45 72.04 67.17 -27.284 2.85 13.86 79.00 34.69 47.43 24.28 63.22 67.20 -27.288 2.85 14.17 78.54 35.00 46.97 24.58 62.75 67.24 -27.294 3.07 -4.46 106.76 16.37 75.20 5.95 90.98 67.31 -27.303 2.96 4.72 92.85 25.55 61.28 15.14 77.07 67.38 -27.313 2.73 24.53 62.84 45.36 31.27 34.94 47.06 67.4 -27.316 2.90 9.89 85.02 30.72 53.45 20.30 69.24 67.47 -27.326 3.00 1.30 98.03 22.14 66.46 11.72 82.24 67.53 -27.334 3.12 -8.61 113.05 12.22 81.48 1.80 97.27 67.83 -27.376 3.03 -1.11 101.69 19.72 70.12 9.30 85.90 67.94 -27.391 3.29 -22.53 134.14 -1.70 102.57 -12.11 118.35 68.01 -27.401 3.22 -16.78 125.42 4.05 93.86 -6.36 109.64 68.07 -27.410 3.08 -5.20 107.87 15.64 76.31 5.22 92.09

185

68.13 -27.418 3.06 -3.20 104.84 17.64 73.28 7.22 89.06 68.19 -27.426 2.91 8.97 86.41 29.80 54.84 19.39 70.63 68.29 -27.440 2.79 19.55 70.37 40.39 38.81 29.97 54.59 68.30 -27.442 2.51 42.50 35.61 63.33 4.04 52.92 19.82 68.32 -27.445 2.61 34.33 47.98 55.17 16.41 44.75 32.20 68.38 -27.453 2.73 24.55 62.80 45.39 31.23 34.97 47.02 68.43 -27.460 2.85 14.14 78.58 34.97 47.02 24.55 62.80 68.49 -27.468 2.90 9.92 84.96 30.76 53.40 20.34 69.18 68.55 -27.477 2.75 22.89 65.32 43.72 33.76 33.30 49.54 68.61 -27.485 2.90 10.28 84.43 31.11 52.86 20.69 68.65 68.67 -27.494 2.73 24.14 63.43 44.97 31.86 34.55 47.65 68.73 -27.502 2.74 23.64 64.19 44.47 32.62 34.05 48.40 68.79 -27.510 2.75 22.69 65.62 43.53 34.05 33.11 49.83 68.83 -27.516 2.75 22.47 65.95 43.30 34.39 32.89 50.17 68.89 -27.524 2.81 17.64 73.28 38.47 41.71 28.05 57.49 68.97 -27.535 2.87 12.80 80.60 33.64 49.04 23.22 64.82 69.03 -27.544 2.65 30.72 53.45 51.55 21.89 41.14 37.67 69.15 -27.561 3.21 -15.86 124.04 4.97 92.47 -5.45 108.25 69.25 -27.575 3.12 -8.58 113.01 12.25 81.44 1.83 97.22 69.33 -27.586 3.10 -6.36 109.64 14.47 78.08 4.05 93.86 69.39 -27.594 3.04 -1.67 102.53 19.17 70.96 8.75 86.74 69.44 -27.601 3.10 -6.45 109.77 14.39 78.20 3.97 93.98 69.46 -27.604 2.91 9.17 86.11 30.00 54.55 19.58 70.33 69.51 -27.611 3.13 -8.86 113.43 11.97 81.86 1.55 97.65 69.51 -27.611 3.15 -10.50 115.91 10.33 84.34 -0.08 100.13 69.57 -27.619 2.98 3.05 95.37 23.89 63.81 13.47 79.59 69.63 -27.628 3.04 -1.36 102.07 19.47 70.50 9.05 86.28

186

References

Adams, C.J., 1987, Geochronology of granite terranes in the Ford Ranges, Marie Byrd Land,

West Antarctica, New Zealand Journal of Geology and Geophysics, v. 30, p. 51-72.

Adams, C.J., Seward, D., and Weaver, S.D., 1995, Geochronology of Cretaceous granites and

metasedimentary basement on Edward VII Peninsula, Marie Byrd Land, West Antarctica,

Antarctic Science, v. 7, p. 265-276.

Adams, C. J., Style of uplift of Paleozoic terranes in Northern Victoria Land, Antarctica:

evidence from K-Ar Age patterns, In: Antarctica: contributions to global earth sciences,

(Springer: Berlin, Heidelberg, New York, 2006).

Aldrich, L.T., Nier, A.O., 1948, Argon 40 in potassium minerals, Physical Review, v. 74, p. 876-

877.

Alley R.B., Blankenship, D.D., Rooney, S.T., and Bentley, C.R., 1989, Sedimentation beneath

ice shelves – the view from ice stream B, Marine Geology, v. 85, p. 101-120.

Alley, R. B., Dupont, T.K., Parizek, B.R., Anandakrishnan, S., Lawson, D.E., Larson, G.J., and

Evenson, E.B., 2006, Outburst flooding and the initiation of ice-stream surges in response

to climatic cooling: A hypothesis, Geomorphology, v. 75, p. 76–89.

Alverson, K., and Owens, W.B., 1996, Topographic preconditioning of open-ocean deep

convection, Journal of Physical Oceanography, v. 26, p. 2,196-2,213.

Anderson, J.B., 1999, Antarctic Marine Geology, Cambridge (Cambridge Univ. Press)

Anderson, J.B., Kurtz, D.D., Domack, E.W., and Balshaw, K.M., 1980, Glacial and glacial

marine sediments of the Antarctic continental shelf. Journal of Geology, v. 88, p. 399-

414.

187

Anderson, J.B., Conway, H., Bart, P.J., Witus, A.E., Greenwood, S.L., McKay, R.M., Hall, B.L.,

Ackert, R.P., Licht, K., Jakobsson, M., and Stone, J.O., 2014, Ross Sea paleo-ice sheet

drainage and deglacial history during and since the LGM, Quaternary Science Reviews,

v. 100, p. 31-54.

Balestrieri, M.L., Bigazzi, G., Ghezzo, C., and Lombardo, B., 1994, Fission track dating of

apatites from the Granite Harbour Intrusive Suite and uplift-denundation history of the

Transantarctic Mountains in the area between the Mariner and David Glaciers (Northern

Victoria Land, Antarctica), Terra Antarctica, v. 1, p. 82-87.

Bamber, J.L., Riva, R.E.M., Vermeersen, B.L.A., and LeBroq, A.M., 2009, Reassessment of the

potential sea-level rise from a collapse of the West Antarctic Ice Sheet, Science, v. 324,

p. 901-903.

Barker, P. F., Kennett, J. P., et al., 1988, Proc. ODP, Init. Repts., 113: College Station, TX

(Ocean Drilling Program).

Barrera, E., and Huber, B.T., 1990, Evolution o fAntarctic waters during the Maestrichtian:

foraminifer oxygen and carbon isotope ratios, Leg 113, in Barker, P.F., Kennett, J.P., et

al, Proceedings of the Ocean Drilling Program, Scientific Results, v. 113.

Barrett, P.J., (ed) 1986. Antarctic Cenozoic history from the MSSTS-1 drillhole, McMurdo

Sound, Barrett, P.J., ed., Wellington: DSIR Bulletin, 237, Science Information Publishing

Centre, Wellington, 174 p.

Barrett, P.J., 1981, History of the Ross Sea region during the deposition of the Beacon

Supergroup 400 - 180 million years ago, Journal of the Royal Society of New Zealand,

v. 11, p. 447-458

188

Barrett, P.J., (ed) 1989. Antarctic Cenozoic history from the CIROS-1 drillhole, McMurdo

Sound: DSIR Bulletin, 245, Science Information Publishing Centre, Wellington, 254 p.

Barrett P.J., 1991. Antarctica and global climate change - a geological perspective. In: Harris

C.M. & Stonehouse B. (eds.), Antarctica and Global Climate Change, Belhaven Press and

Scott Polar Research Institute, Cambridge, p. 35-50.

Barrett, P., Sarti, M., and Sherwood, W, 2000. Studies from the Cape Roberts Project, Ross Sea,

Antarctica, Initial Report on CRP-3. Terra Antartica , 7(1/2):209 pp.

Bart, P.J., 2003, Were West Antarctic Ice Sheet grounding events in Ross Sea a consequence of

East Antarctic Ice Sheet expansion during the middle Miocene?, Earth and Planetary

Science Letters, v. 216, p. 93–107.

Bart, P.J., and De Santis, L., 2012, Glacial intensification during the Neogene: A review of

seismic stratigraphic evidence from the Ross Sea, Antarctica, continental shelf,

Oceanography, v. 25, p. 166-183.

Bartek, L.R., Sloan, L.C., Anderson, J.B., and Ross, M.I., 1992, Evidence from the Antarctic

continental margin of late Paleogene ice sheets: A manifestation of plate reorganization

and synchronous changes in atmospheric circulation over the emerging Southern Ocean?,

in Prothero, D.R., and Berggren, W.A., eds., Eocene-Oligocene Climatic and Biotic

Evolution: Princeton, New Jersey, Princeton University Press, p. 131–159.

Behrendt, J.C., Blankenship, D.D., Finn, C.C., Bell, R.E., Sweeney, R.E., Hodge, S.N., and

Brozena, J.M., 1994, CASERTZ aeromagnetic data reveal late Cenozoic flood basalts (?)

in the West Antarctic rift system, Geology, v. 22, 527-530.

189

Belanger, P.E., Curry, W.B., and Matthews, R.K., 1981, Core-top evaluation of benthic

foraminiferal isotopic ratios for paleo-oceanographic interpretations, Palaeogeography,

Palaeoclimatology, Palaeoecology, v. 33, p. 205-220.

Bell, R.E., 2008, The role of subglacial water in ice-sheet mass balance, Nature, v. 1, p. 297–

304.

Bentley, C.R., 1991, Configuration and structure of the subglacial crust. In: Tingey, R.J. (Ed),

The Geology of Antarctica, Oxford University Press, New York, p. 1077-1078.

Berggren, W.A., and Hollister, C.D., 1977, Plate tectonics and paleocirculation—Commotion in

the ocean, Tectonophysics, v. 38, p. 11–48.

Bersch, M., Becker, G.A., Frey, H., and Koltermann K.P., 1992, Topographic effects of the

Maud Rise on the stratification and circulation of the Weddell Gyre, Deep Sea

Research, v. 39, p. 303-331.

Billups, K., and Schrag, D.P., 2002, Paleotemperatures and ice volume of the past 27 Myr

revisited with paired Mg/Ca and 18O/16O measurements on benthic foraminifera,

Paleoceanography, v. 17, 1003.

Birkenmajer, K., Gazdzicki, A., Krajewski, K.P., Przybycin, A., Solecki, A., Tatur, A., Yoon,

H.I., 2005, First Cenozoic glaciers in West Antarctica, Polish Polar Research, v. 26, p. 3-

12.

Boger, S.D., 2010. Antarctica-Before and after Gondwana. Gondwana Research, v. 19, p. 335-

371.

Bohaty, S.M., and Zachos, J.C., 2003, Significant Southern Ocean warming event in the late

middle Eocene, Geology, v. 31, p. 1017-1020.

190

Bornhold, B.D., 1983. Ice-rafted debris in sediments from leg 71, southwest Atlantic Ocean, in

Ludwig, W.J., Krasheninikov, V.A. et al. (eds) Initial Reports of theDeep Sea Drilling

Project, v. 71, p. 307-316, Washington.

Bradley, L.M., and Frey H., 1988, Constraints on the crustal nature and tectonic history of the

Kerguelen Plateau from comparative magnetic modeling using Magsat data,

Tectonophysics, v. 145, p. 243-251.

Breza, J.R., Wise, S.W., 1992. Lower Oligocene ice-rafted debris on the Kerguelen Plateau:

Evidence for East Antarctic continental glaciation, in: Wise, S.W., Schlich, J.R. et al.,

Proceedings of the Ocean Drilling Program, Scientific Results, v. 120, p. 161-178.

Brommer, A., Millar, I.L., Zeh, A., 1999. Geochronology, structural geology and petrography of

the northwestern La Grange Nunataks, Shackleton Range, Antarctica. Terra Antarctica, v.

6, p. 269-278.

Cape Roberts Science Team, 1998. Initial Report on CRP-1, Cape Roberts Project, Antarctica,

Terra Antarctica, v. 5 (187 pp.).

Cape Roberts Science Team, 1999. Studies from the Cape Roberts Project, Ross Sea, Antarctica,

Initial Report on CRP-2/2A, Terra Antarctica, v. 6 (187 pp.).

Cape Roberts Science Team, 2000. Studies from the Cape Roberts Project, Ross Sea, Antarctica,

Initial Report on CRP-3, Terra Antarctica, v. 7 (209 pp.).

Cawood, P.A., 2005. Terra Australis Orogen: Rodinia breakup and development of the Pacific

and Iapetus margins of Gondwana during the Neoproterozoic and Paleozoic. Earth

Science Reviews, v. 69, p. 249-279.

Cawood, P.A., Buchan, C., 2007. Linking accretionary orogenesis with supercontinent assembly.

Earth Science Reviews, v. 82, p. 217-256.

191

Cawood, P.A., Kröner, A., Collins, W.J., Kusky, T.M., Mooney, W.D., Windley, B.F., 2009.

Accretionary orogens through earth history. In: Cawood, P.A., Kröner, A. (Eds.), Earth

Accretionary Systems in Space and Time: Geological Society, London, Special

Publication, 318, p. 1-36.

Comiso, J.C., and Gordon, A.L., 1987, Recurring polynyas over the Cosmonaut Sea and the

Maud Rise, Journal of Geophysical Research, v. 92, p. 2,819-2,833.

Connolly, J. R. & Ewing, M., 1965. Ice-rafted detritus as a climatic indicator in Antarctic deep

sea cores. Science, v. 150, p. 1822-1824.

Cooke, D.W., Hays, J.D. 1982. Estimates of Antarctic ocean seasonal sea-ice cover during

glacial intervals, in Craddock, C. (ed) Antarctic Geoscience, p. 1017-1025, Madison.

Cooper, A.K., O’Brien, P.E., 2004. Leg 188 Synthesis: Transitions in the glacial history of the

Prydz Bay region, East Antarctica, from ODP Drilling: Cooper, A.K., O’Brien, P.E., and

Richter, C. (eds), Proceedings of the Ocean Drilling Program, Scientific Results Volume

188.

Curtis, M.L., 2001. Tectonic history of the Ellsworth Mountains, West Antarctica: reconciling a

Gondwana enigma. Geological Society of America Bulletin, v. 113, p. 939-958.

Death, R., Siegert, M. J., Bigg, G. R., and Wadley, M. R., 2006, Modelling iceberg trajectories,

sedimentation rates and meltwater input to the ocean from the Eurasian Ice Sheet at the

Last Glacial Maximum, Paleogeography, Paleoclimatology, Paleoecology, v. 236, p. 135-

150.

DeConto, R.M., and Pollard, D., 2003. Rapid Cenozoic glaciation of Antarctica induced by

declining atmospheric CO2. Nature, 421, p. 245-249.

192

DeConto, R.M., Pollard, D., Wilson, P.A., Palike, H., Lear, C.H., Pagani, M., 2008. Thresholds

for Cenozoic bipolar glaciation. Nature, v. 455, p. 652-656.

De Santis, L., Prato, S., Brancolini, G., Lovo, M., and Torelli, L., 1999, The Eastern Ross Sea

continental shelf during the Cenozoic: Implications for the West Antarctic Ice Sheet

development: Global and Planetary Change, v. 23, p. 173–196.

Di Venere, V., Kent, D.V., Dalziel, I.W.D., 1995. Early Cretaceous paleomagnetic results from

Marie Byrd Land, West Antarctica: implications for the Weddellia collage of crustal

blocks. Journal of Geophysical Research, v. 100, p. 8133-8151.

Diester-Haass, L., Robert, C., and Chamley, H., 1993, Paleoceanographic and paleoclimatic

evolution in the Weddell Sea (Antarctica) during the middle Eocene-late Oligocene, from

a coarse sediment fraction and clay mineral data (ODP Site 689), Marine Geology, v.

114, p. 233-250.

Dolan, A.M., Haywood, A.M., Hill, D.J., Dowsett, H.J., Hunter, S.J., Lunt, D.J., and Pickering,

S.J., 2011, Sensitivity of Pliocene ice sheets to orbital forcing, Palaeogeography,

Palaeoclimatology, Palaeoecology, v. 309, p. 98-110.

Domack, E. W., and Harris, P. A., 1998, A new depositional model for ice shelves based upon

sediment cores from the Ross Sea and the MacRobertson shelf, Antarctica, Annals of

Glaciology, v. 27, p. 281-284.

Donda F., Brancolini, G., De Santis, L., Trincardi. F., 2003. Seismic facies and sedimentary

processes on the continental rise off Wilkes Land (East Antarctica): evidence of bottom

current activity. Deep-Sea Research II, v. 50, p. 1509-1527.

193

Donda, F., O’Brien, P.E., De Santis, L., Rebesco, M., Brancolini., G., 2008. Mass wasting

processes in the Western Wilkes Land margin: Possible implications for East Antarctic

glacial history. Palaeogeography, Palaeoclimatology, Palaeoecology, v. 260, p. 77-91.

Drewry, D.J. (Ed.), 1983, Antarctica: Glaciological and Geophysical Folio, Scott Polar Res. Inst.,

Cambridge, U. K.

Drewry, D.J., and Cooper, A.P.R., 1981, Processes and models of Antarctic glaciomarine

sedimentation, Annals of Glaciology, v. 2, p. 117-122.

Duplessy, J.C., Lalou, C., and Vinot, A.C., 1970, Differential isotopic fractionation in benthic

foraminifera and paleotemperatures reassessed, Science, v. 168, p. 250-251.

Edgar, K.M., Wilson, P.A., Sexton, P.F., and Suganuma, Y., 2007, No extreme bipolar glaciation

during the main Eocene calcite compensation shift, Nature, v. 448, p. 908-911.

Elderfield, H. and Ganssen, G., 2000, Past temperature and δ18O of surface ocean waters

inferred from foraminiferal Mg/Ca ratios, Nature, v. 405, p. 442-445.

Eldrett, J.S., Harding, I.C., Wilson, P.A., Butler, E., and Roberts, A.P., 2007, Continental ice in

Greenland during the Eocene and Oligocene, Nautre, v. 8, p. 176-179.

Emiliani, C., 1955, Pleistocene temperatures, Journal of Geology, v. 63, p. 538-578.

Epstein, S., Buchsbaum, R., Lowenstam, H.A., and Urey, H.C., 1953, Revised carbonate-water

isotopic temperature scale, Geological Society of America Bulletin, v. 64, p. 1,315-1,325.

Erlingsson, U., 1994, The “Captured Ice Shelf” hypothesis and its applicability to the

Weichselian glaciation, Geografiska Annaler Series A-physical Geography, v. 76, p. 1-

12,

194

Escutia, C., De Santis, L., Donda, F., Dunbar, R.B., Cooper, A.K., Brancolini, G., Eittreim, S.L.,

2005. Cenozoic ice sheet history from East Antarctic Wilkes Land continental margin

sediments. Global and Planetary Change, v. 45, p. 51-81.

Evans, J., and Pudsey, C. J., 2002, Sedimentation associated with Antarctic Peninsula ice

shelves: Implications for palaeoenvironmental reconstructions of glacimarine sediments,

Journal of the Geological Society, v. 159, p. 233-237.

Exon, N.F., Kennett, J.P., and Malone, M.J., 2004, Leg 189 synthesis: Cretaceous–Holocene

history of the Tasmanian Gateway, in Exon, N.F., et al., Proceedings of the Ocean

Drilling Program, Scientific results, Volume 189: College Station, Texas, Ocean Drilling

Program, p. 1-37.

Expedition 318 Scientists, 2011. Site U1356. In Escutia, C., Brinkhuis, H., Klaus, A., and the

Expedition 318 Scientists, Proc. IODP, 318: Tokyo (Integrated Ocean Drilling Program

Management International, Inc.).

Fanning, C.M., Flint, R.B., Parker, A.J., Ludwig, K.R., Blissett, A.H., 1988. Refined Proterozoic

evolution of the Gawler Craton, South Australia, through U–Pb zircon geochronology.

Precambrian Research, v. 40–41, p. 363–386.

Fanning, C.M., Dally, S.J., Bennett, V.C., Ménot, R.P., Peucat, J.J., Oliver, G.J.H., Monnier, O.,

1995. The “Mawson Block”: once contiguous Archaean to Proterozoic crust in the East

Antarctic Shield and the Gawler Craton. In: Ricci, C.A. (Ed.), Abstracts, 7th International

Symposium on Antarctic Earth Sciences, Sienna.

Farmer, G.L., Licht, K., Swope, R.J., Andrews, J., 2006, Isotopic constraints on the provenance

of fine-grained sediment in LGM tills from the Ross Embayment, Antarctica, Earth and

Planetary Science Letters, v. 249, p. 90-107.

195

Fitzsimons, I.C.W., 2000. Grenville-age basement provinces in East Antarctica: Evidence for

three separate collisional orogens. Geology, v. 28, p. 879-882.

Fitzsimons, I.C.W., 2003, Proterozoic basement provinces of southern and southwestern

Australia, and their correlation with Antarctica. Special Publication – Geological Society

of London, Proterozoic East Gondwana: Supercontinent Assembly and Breakup, v. 206,

p. 93-130.

Florindo, F, and Roberts, A.P., 2005, Eocene-Oligocene magnetobiochronology of ODP Sites

689 and 690, Maud Rise, Weddell Sea, Antarctica, Geological Society of America

Bulletin, v. 117, p 46-66.

Flöttmann, T., James, P., Rogers, J., Johnson, T., 1994. Early Palaeozoic foreland thrusting and

basin reactivation at the palaeo-Pacific margin of the southeastern Australian

Precambrian Craton: a reappraisal of the structural evolution of the southern Adelaide

Fold-Thrust Belt. Tectonophysics, v. 234, p. 95–116.

Foland, K.A., Linder, J.S., Laskowski, T.E., and Grant, N.K., 1984, 40Ar/39Ar dating of

glauconites: Measured 39Ar recoil loss from well-crystallized specimens, Isotope

Geoscience, v. 2, p. 241-264.

Fretwell, P., and 59 others, 2013, Bedmap2: improved ice bed, surface, and thickness datasets for

Antarctica, The Cryosphere, v. 7, p. 375-393.

Frey, H., 1982, Magsat scalar anomaly distribution: the global perspective, Geophysical

Research Letters, v. 9, p. 277–280.

Frey, H., 1985, Magsat and POGO anomalies over the Lord Howe Rise: Evidence against a

simple continental crustal structure, Journal of Geophysical Research, v. 90, p. 2,631-

2,639.

196

Fullerton, L.G., Frey, H.V., Roark, J.H., and Thomas, H.H., 1994, Contributions of Cretaceous

Quiet Zone natural remanent magnetization to Magsat anomalies in the Southwest Indian

Ocean, Journal of Geophysical Research, v. 99, p. 11,923-11,936.

Goldich, S.S., Treves, S.B., Suhr, N.H., and Stuckless, J.S., 1975, Geochemistry of the Cenozoic

volcanic rocks of Ross Island and vicinity, Antarctica, Journal of Geology, v. 83, p. 415-

435.

Goodge, J.W., Walker, N.W., Hansen, V.L., 1993. Neoproterozoic–Cambrian basement involved

orogenesis within the Antarctic margin of Gondwana. Geology, v. 21, p. 37-40

Goodge, J.W., Fanning, C.M., Bennett, V.C., 2001. U–Pb evidence of 1.7 Ga crustal tectonism

during the Nimrod Orogeny in the Transantarctic Mountains, Antarctica: implications for

Proterozoic plate reconstructions. Precambrian Research, v. 112, p. 261-288.

Goodge, J. W., 2007, Metamorphism in the Ross Orogen and its bearing on Gondwana margin

tectonics, Geological Society of America Special Paper: Convergent Margin Tectonics

and Associated Regions: A tribute to W. G. Ernst.

Goodge, J.W., Vervoort, J.D., Fanning, C.M., Brecke, D.M., Farmer, G.L., Williams, I.S.,

Myrow, P.M., DePaolo, D.J., 2008. A positive test of East Antarctica–Laurentia

juxtaposition within the Rodinia supercontinent. Science, v. 321, p. 235-239.

Goodge, J. W., and Fanning, C. M., 2010, Composition and age of the East Antarctic Shield in

eastern Wilkes Land determined by proxy from Oligocene-Pleistocene glaciomarine

sediment and Beacon Supergroup sandstones, Antarctica. Geological Society of America

Bulletin, v. 122, no. 7-8, p. 1135-1159.

Gordon, A.L., and Huber, B.A., 1990, Southern Ocean winter mixed layer, Journal of

Geophysical Research, v. 95, p. 11,655-11,672.

197

Graham, D.W., Corliss, B.H., Bender, M.L., and Keigwin Jr., L.D., 1981, Carbon and oxygen

isotopic disequilibria of recent deep-sea benthic foraminifera, Marine Micropalentology,

v. 6, p. 483-497.

Grobe H., 1987. A simple method for determination of ice-rafted debris in sediment cores.

Polarforschung, v. 57 (3), p. 123-126.

Hambrey, M.J., Barrett, P.J., and Robinson, P.H., 1989, Stratigraphy: In Barrett, P.J., (ed),

Antarctic Cenozoic history from the CIROS-1 drillhole, western McMurdo Sound: DSIR

Miscellaneous Bulletin, , Science Information and Publishing Centre, 247:23-47.

Hambrey, M.J., Ehrmann, W.U., and Larsen, B., 1991, Cenozoic glacial record of the Prydz Bay

continental shelf, east Anatarctica: Barron, J., Larsen, B., et al. (eds), Proceedings of the

Ocean Drilling Program, Scientific results, 119:77-132.

Hauptvogel, D.W., Passchier, S., 2012. Early–Middle Miocene (17–14 Ma) Antarctic ice

dynamics reconstructed from the heavy mineral provenance in the AND-2A drill core,

Ross Sea, Antarctica. Global and Planetary Change, v. 82-83, p. 38-50.

Heinrich, H., 1988. Origin and consequences of cyclic ice rafting in the Northeast Atlantic Ocean

during the past 130,000 years. Quaternary Research, v. 29(2), p. 142-152.

Hemming. S.R., Hall, C.M., Biscaye, P.E., Higgins, S.M., Bond, G.C., McManus, J.F., Barber,

D.C., Andrews, J.T., Broecker, W.S., 2002. 40Ar/39Ar and 40Ar concentrations of fine-

grained sediment fractions from North Atlantic Heinrich layers. Chemical Geology, v.

182, p. 583-603.

Hill, D. J., Haywood, A. M., Hindmarsh, R. C. A. and Valdes, P. J. 2007. Characterizing ice

sheets during the Pliocene: evidence from data and models. In Deep time perspectives on

climate change: marrying the signals from computer models and biological proxies (eds

198

M. Williams, A. M. Haywood, J. Gregory & D. Schmidt). Micropalaeontological Society,

Special Publication, London, UK: Geological Society of London, p. 600.

Hinze, W.J., von Frese, R.R.B., and Ravat, D.N. 1991, Mean magnetic contrasts between oceans

and continents, Tectonophysics, v. 192, p. 117–127.

Holbourn, A., Kuhnt, W., Clemens, S., Prell, W., and Andersen, N., 2013, Middle to Late

Miocene stepwise climate cooling: Evidence from a high-resolution deep water isotope

curve spanning 8 million years, Paleoceanography, v. 28, p. 688-699.

Hole, M.J., 1988, Post-subduction alkaline volcanism along the Antarctic Peninsula, Jounral of

the Geological Society of London, v. 145, p. 985-998.

Hughes, T.J., 1981, The weak underbelley of the West Antarctic Ice-Sheet, Journal of

Glaciology, v. 27, p, 518–525.

Imsland, P., Larsen, J.G., Prestwick, T, and Sigmond, E., 1977, The geology and petrology of

Buovetoya, South Atlantic Ocean, Lithos, v. 10, p. 213-34.

Ivany, L.C., Van Simaeys, S., Domack, E.W., and Samson, S.D., 2006, Evidence for and earliest

Oligocene ice sheet on the Antarctic Peninsula, Geology, v. 34, p. 377-380.

Jacobs, J., Fanning, C.M., Henjes-Kunst, F., Olesch, M., Paech, H., 1998. Continuation of the

Mozambique Belt into East Antarctica: Grenville-age metamorphism and polyphase Pan-

African high-grade events in central Dronning Maud Land. Journal of Geology, v. 106, p.

385-406.

Jacobs, J., Fanning, C.M., and W. Bauer, 2003. Timing of Grenville-age vs. Pan-African

medium- to high grade metamorphism in western Dronning Maud Land (East Antarctic)

and significance for correlations in Rodinia and Gondwana. Precambrian Research, v.

125, p. 1-20. DOI:10.1016/S0301-9268(03)00048-2

199

Jankowski, E.J., Drewry, D.J., 1981, The structure of West Antarctica from geophysical studies,

Nature, v. 291, p. 17-21.

Jordan, T. A., Ferraccioli, F., Corr, H., Graham, A., Armadillo, E., and Bozzo, E., 2010,

Hypothesis for mega-outburst flooding from a palaeo-subglacial lake beneath the East

Antarctic Ice Sheet, Terra Nova, v. 22, p. 283–289.

Joughin, I. and Alley, R.B., 2011, Stability of the West Antarctica ice sheet in a warming world,

Nature Geoscience, v. 4, p. 506-513.

Joughin, I., Smith, B.E., and Holland, D.M., 2010, Sensitivity of 21st century sea level to ocean-

induced thinning of Pine Island Glacier, Antarctica, Geophysical Research Letters, v.

37, L20502.

Katz, M.E., Katz, D.R., Wright, J.D., Miller, K.G., Pak, D.K., Shackleton, N.J., and Thomas, E.,

2003, Early Cenozoic benthic foraminiferal isotopes: Species reliability and interspecies

correction factors, Paleoceanography, v. 18, PA000798.

Katz, M.E., Miller, K.G., Wright, J.D., Wade, B.S., Browning, J.V., Cramer, B.S., Rosenthal, Y.

2008. Stepwise transition from the Eocene greenhouse to the Oligocene icehouse, Nature

Geoscience, v. 1, p. 329-334.

Keigwin, L.D., and Keller, G., 1984, Middle Oligocene climatic changes from equatorial Pacific

DSDP Site 77, Geology, v. 12, p. 16-19.

Kelley, S., 2002, K-Ar and Ar-Ar dating, Reviews in Mineralogy and Geochemistry, v. 47, p.

785-818.

Kennett, J.P., 1977, Cenozoic evolution of Antarctic glaciation, the circum-Antarctic Ocean, and

their impact on global paleoclimate: Journal of Geophysical Research, v. 82, p. 3843–

3860.

200

Kennett, J.P., and Stott, L.D., 1990, Proteus and proto-oceanus: ancestral Paleogene oceans as

revealed from Antarctic stable isotopic results; ODP Leg 113, in Barker, P.F., Kennett,

J.P., et al, Proceedings of the Ocean Drilling Program, Scientific Results, v. 113.

Kent, D., Opdyke, N.D., Ewing, M., 1971. Climatic change in the North Pacific using ice-rafted

detritus as a climatic indicator. Geological Society of America Bulletin, v. 82, p. 2741-

2754.

Kim, H.R., von Frese, R.R.B., Golynsky, A.V., Taylor, P.T., and Kim, J.W., 2005, Crustal

analysis of maud rise from combined satellite and near-surface magnetic survey data,

Earth Planets Space, v. 57, p. 717-726.

Kominz, M.A., Pekar, S.F., 2001. Oligocene eustasy from two-dimensional sequence

stratigraphic backstripping, Geological Society of America Bulletin, 113:291-304.

Kulhanek, D., Levy, R., Morgans, H., Zwingmann, H., Wilson, D., and Luyendyk, B., 2014,

New age constraints for Paleogene sediments in the Ross Sea and implications for

Coulman High, Scientific Committee on Antarctic Research Open Science Conference

Abstract, p. 128.

Labeyrie, L.D., Pichon, J.J., Labracherie, M., Ippolito, P., Duprat, J., Duplessy, J.C., 1986.

Melting history of Antarctica during the past 60,000 years. Nature, v. 322, p. 701-706.

Lear, C. H., H. Elderfield, Wilson, P.A., 2000. Cenozoic deep-sea temperatures and global ice

volumes from Mg/Ca in benthic foraminiferal calcite, Science, v. 287, p. 269-272.

Lear, C. H., Rosenthal, Y., Slowey, N., 2002, Benthic foraminiferal Mg/Ca-paleothermometry: A

revised core-top calibration, Geochim. Cosmochim. Acta, v. 66, p. 3375-3387.

201

Lear, C.H., Rosenthal, Y., Coxall, H.K., and Wilson, P.A., 2004. Late Eocene to early Miocene

ice sheet dynamics and the global carbon cycle, Paleoceanography 19, PA4015,

doi:10.1029/2004PA001039.

LeMasurier, W.E., and Rex, D.C., 1991, The Marie Byrd Land Volcanic Province and its

relation to the Cenozoic West Antarctic rift system, in: R.J. Tingey (Ed.), The Geology of

Antarctica, Oxford University Press, New York, 1991, pp. 249–284.

LeRoex, A.P., Dick, H.J.B., Erlank, A.X, Reid, A.M., Frey, FA., and Hart, S.R., 1983,

Geochemistry, Mineralogy and petrogenesis of lavas erupted along the southwest Indian

Ridge between the Bouvet triple junction and 11 degrees east, Journal of Petrology, v. 24,

p. 267-318.

Licth, K.J., Lederer, J.R., and Swope, R.J., 2005, Provenance of LGM glacial till (sand fraction)

across the Ross embayment, Antarctica, Quaternary Science Reviews, v. 24, p. 1499-

1520.

Lindsay, R.W., Kwok, R., de Steur, L., and Meier, W., 2008, Halo of ice deformation observed

over the Maud Rise seamount, Geophysical Research Letters, v. 35, L15501.

Lindstrom, D., and Tyler, D., 1984, Preliminary results of Pine Island and Thwaites Glaciers

study, Antarctic Journal of the United States, v. 19, p. 53-55.

Lira, R., Millone, H.A., Kirschbaum, A.M., Moreno, R.S., 1997. Calc-alkaline arc granitoid

activity in the Sierra Norte–Ambargasta Ranges, central Argentina. Journal of South

American Earth Sciences, v. 10, p. 157-177.

Lisitzin, A.P., 1960. Bottom sediments of the eastern Antarctic and the Southern Indian Ocean.

Deep-Sea Research, v. 7, p. 89-99.

202

Lyle, M., Gibbs, S., Moore, T. C., and Rea, D. K., 2007, Late Oligocene initiation of the

Antarctic Circumpolar Current: Evidence from the South Pacific, Geology, v. 35, p. 691-

694.

Lythe, M.B., Vaughan, D.G, and the BEDMAP consortium, 2001, BEDMAP: A new ice

thickness and subglacial topographic model of Antarctica. Journal of Geophysical

Research, v. 106, p. 11335-11351.

MacDonald, T.R., Ferrigno, J.G., Williams, R.S., Jr., and Luchitta, B.K., 1989, Velocities of

Antarctic outlet glaciers determined from sequential Landsat images, Antarctic Journal of

the United States , v. 24, p. 105-106.

Mackensen, A., and Ehrmann, W.U., 1992, Middle Eocene through Early Oligocene climate

history and paleoceanography in the Southern Ocean: Stable oxygen and carbon isotopes

from ODP Sites on Maud Rise and Kerguelen Plateau, Marine Geology, v. 108, p. 1-27.

Marks, K.M., and Tikku, A.A., 2001, Cretaceous reconstructions of East Antarctica, Africa and

Madagascar, Earth and Planetary Science Letters, v. 186, p. 479-495.

Martin, A.K., and Hartnady, C.J.H., 1986, Plate tectonic development of the South West Indian

Ocean: A revised reconstruction of East Antarctica and Africa, Journal of Geophysical

Research, v. 91, p. 4,767-4,786.

Martin, P.A., Lea, D.W., Rosenthal, Y., Shackleton, N.J., Sarnthein, M., and Papenfuss, T.,

2002, Quaternary deep sea temperature histories derived from benthic foraminiferal

Mg/Ca, Earth Planetary Science Letters, v. 198, p.193-209.

McCorkle, D.C., Keigwin, L.D., Corliss, B.H., and Emerson, S.R., 1990, The influence of

microhabitats on the carbon isotopic composition of deep sea benthic foraminifera,

Paleoceanography, v. 5, p. 161-185.

203

McLelland, J., Hamilton, M., Selleck, B., McLelland, J., Walker, D., Orrell, S., 2001. Zircon U–

Pb geochronology of the Ottawan Orogeny, Adirondack Highlands, New York: regional

and tectonic implications. Precambrian Research, v. 109, p. 39-72.

McPhee, M.G., Ackley, S.F., Guest, P., Huber, B.A., Martinson, D.G., Morison, J.H., Muench,

R.D., Padman, L, and Stanton, T.P., 1996, The Antarctic Flux Zone Experiment, Bulletin

of the American Meteorological Society, v. 77, p. 1,221-1,232.

McWilliams, M.O., 1981. Palaeomagnetism and Precambrian tectonic evolution of Gondwana.

In: Kröner, A. (Ed.), Precambrian Plate Tectonics. Elsevier, Amsterdam, p. 649-687.

Mead, G.A., and Hodell, D.A., 1995, Conrols on the 87Sr/86Sr composition of seawater from the

Middle Eocene to Oligocene: Hole 689B, Maud Rise Antarctica, Paleoceanography, v.

10, p. 327-346.

Meert, J.G., Van der Voo, R., 1997. The assembly of Gondwana 800-500 Ma. Journal of

Geodynamics, v. 23, p. 223-235.

Mercer, J. H., 1978, West Antarctic ice sheet and CO2 greenhouse effect: A threat of disaster,

Nature, v. 271, p. 321–325.

Merrihue, C.M., Turner, G., 1966, Potassium-argon dating by activation with fast neutrons,

Journal of Geophysical Research, v. 71, p. 2852-2857.

Miller, K.G., and Fairbanks, R.G., 1985, Oligocene-Miocene global carbon and abyssal

circulation changes, in The carbon cycle and atmospheric CO2: Natural variations

Archean to Present (eds Sundquist E.T., and Broecker W.S.), American Geophysical

Union, Washington D.C.

Miller, K.G., Fairbanks, R.G., and Mountain, G.S., 1987, Tertiary oxygen isotope synthesis, sea

level history, and continental margin erosion, Paleoceanography, v. 2, p. 1-19.

204

Miller, K.G., Kominz, M.A., Browning, J.V., Wrigth, J.D., Mountain, G.S., Katz, M.E.,

Sugarman, P.J., Cramer, B.S., Christie-Blick, N., and Pekar S.F., 2005, The Phanerozoic

record of global sea-level change, Science, v. 310, p. 1293-1298.

Miller, K.G., Wright, J.D., and Fairbanks, R.G., 1991, Unlocking the ice house: Oligocene-

Miocene oxygen isotopes, eustasy, and margin erosion, Journal of Geophysical Research,

v. 96, p. 6829–6848.

Möller, A., Post, N. J., and Hensen, B. J., 2002, Crustal residence history and garnet Sm-Nd ages

of high-grade metamorphic rocks from the Windmill Islands area, East Antarctica.

International Journal of Earth Sciences, v. 91, p. 993-1004.

Muench, R.D., Morison, J.H., Padman, L., Martinson, D., Schlosser, P., Huber, B., and

Hohmann, R., 2001, Maud Rise revisited, Journal of Geophysical Research, v. 106, p.

2,423-2440.

Najjar, R., Nong, G.T., Seidov, D., and Peterson, W.T., 2002, Modeling geographic impacts on

early Eocene ocean temperature, Geophysical Research Letters, v. 29.

Naish, T.R., Wolfe, K.J., Barrett, P.J., Wilson, G.S., Cliff, A., Bohaty, S.M., Bucker, C.J., Claps,

M., Davey, F.J., Dunbar, G.B., Dunn, A.G., Fielding, C.R., Florindo, F., Hannah, M.J.,

Harwood, D.M., Henrys, S.A., Krissek, L.A., Lavelle, M., van der Meer, J., McIntosh,

W.C., Niessen, F., Passchier, S., Powell, R.D., Roberts, A.P., Sagnotti, L., Scherer, R.P.,

Strong, P.C., Talarico, F., Verosub, K.L., Villa, G., Watkins, D.K., Webb, P.N., Wonik,

T., 2001, Orbitally induced oscillations in the East Antarctic ice sheet at the

Oligocene/Miocene boundary, Nature, v. 413, p. 719-723.

Naish, T. and 55 others, 2009, Obliquity-paced Pliocene West Antarctic Ice Sheet oscillations,

Nature, v. 458, p. 322-329.

205

Nong, G.T., Najjar, R.G., Seidov, D., and Peterson, W.H., 2000, Simulation of ocean

temperature change due to the opening of Drake Passage, Geophysical Research Letters,

v. 27, p. 2689-2692.

Oliver, R.L., Fanning, C.M., 2002. Proterozoic geology east and southeast of Commonwealth

Bay, George V Land, Antarctica, and its relationship to that of adjacent Gondwana

terranes. In: Gamble, J.A., Skinner, D.N.B., Henrys, S. (Eds.), Antarctica at the Close of

the Millennium: Royal Society of New Zealand, Bulletin, v. 35, p. 51-58. Wellington.

Ou, H.W., 1991, Some effects of a seamount on oceanic flows, Journal of Physical

Oceanography, v. 21, p. 1,835-1,845.

Pagani, M., Huber, M., Liu, Z., Bohaty, S., Hendriks., Sijp, W., Krishnan, S., DeConto, R.M.,

2011. The role of carbon dioxide during the onset of Antarctic glaciation. Science, v. 334,

p. 1261-1264.

Pagani, M., Zachos, J.C., Freeman, K.H., Tipple, B., Bohaty, S., 2005. Marked decline in

atmospheric carbon dioxide concentrations during the Paleogene, Science, 309, 600-603.

Pälike, H., Norris, R.D., Herrle, J.O., Wilson, P.A., Coxall, H.K., Lear, C.H., Shackleton, N.J.,

Tripati, A.K., Wade, B.S., 2006. The Heartbeat of the Oligocene Climate System.

Science 314, 1894–1898.

Pankhurst, R.J., Weaver, S.D., Bradshaw, J.D., Storey, B.C., Ireland, T.R., 1998. Geochronology

and geochemistry of pre-Jurassic superterranes in Marie Byrd Land, Antarctica. Journal

of Geophysical Research, v. 103, p. 2529-2547.

Pankhurst, R.J., Rapela, C.W., Fanning, C.M., Márquez, M., 2006. Gondwanide continental

collision and the origin of Patagonia. Earth Science Reviews, v. 76, p. 235-257.

206

Parsiegla, N., Gohl, K., and Uenzelmann-Neben, G., 2008, The Agulhas Plateau: structure and

evolution of a Large Igneous Province, Geophysical Journal International, v. 174, p. 336-

350.

Passchier, S., 2011, Linkages between East Antarctic Ice Sheet extent and Southern Ocean

temperatures based on a Pliocene high-resolution record of ice-rafted debris off Prydz

Bay, East Antarctica, Paleoceanography, v. 26, PA4204.

Pearson, N.P., Foster, G.L., Wade, B.S., 2009. Atmospheric carbon dioxide through the Eocene-

Oligocene climate transition. Nature, v. 461, p. 1110-1113.

Pekar, S.F., Christie-Blick, N., 2008. Linking pCO2 Estimates and Oceanographic and Antarctic

Climate Records for the Early Icehouse World (34-16 Ma), Palaeogeography,

Palaeoclimatology, Palaeoecology, v. 260, p. 41-49.

Pekar S.F., Christie-Blick, N., Kominz, M.A., Miller, K.G., 2001, Evaluating the stratigraphic

response to eustasy from Oligocene strata in New Jersey, Geology, v. 29, p. 55-58.

Pekar S.F., Christie-Blick, N., Kominz, M.A., Miller, K.G., 2002. Calibration between eustatic

estimates from backstripping and oxygen isotopic records for the Oligocene. Geology, v.

30 (10), p. 903-906.

Pekar, S.F., DeConto, R.M., Harwood, D.M., 2006. Resolving a late Oligocene conundrum:

deep-sea warming versus Antarctic glaciation: Palaeogeography, Palaeoclimatology,

Palaeoecology, 231:29-40.

Pekar, S.F., Hucks, A., Fuller, M., and Li, S., 2005, Glacioeustatic changes in the early and

middle Eocene (51-42 Ma): Shallow-water stratigraphy from ODP Leg 189 Site 1171

(South Tasman Rise) and deep-sea δ18O records, Geological Society of America Bulletin,

v. 117, p. 1081-1093.

207

Pekar, S.F., Lear, C.H., 2009, Estimating Cenozoic ice volume from deep-sea records, in

Florindo. F., and Siegert, M. (eds), Antarctic Climate Evolution, Elsevier

Pekar, S. F. Miller, K. G., 1996. New Jersey Oligocene "Icehouse" sequences (ODP 150X)

correlated with global d18O and Exxon eustatic records. Geology, v. 24, p. 567-570.

Persico, D., and Villa, G., 2004, Eocene-Oligocene calcareous nannofossils from Maud Rise and

Kerguelen Plateau (Antarctica): paleoecological and paleoceanographic implications,

Marine Micropaleontology, v. 52, p. 153-179.

Peters, S.E., Carlson, A.E., Kelly, D.C., Gingerich, P.D., 2010. Large-scale glaciation and

deglaciation of Antarctica during the Late Eocene. Geology, v. 38, p. 723-726.

Peucat, J.J., Ménot, R.P., Monnier, O., Fanning, C.M., 1999. The Terra Adélie basement in the

East-Antarctica Shield: geological and isotopic evidence for a major 1.7 Ga thermal

event; comparison with the Gawler Craton in South Australia. Precambrian Research, v.

94, p. 205-224.

Pfuhl, H.A., and McCave, I.N., 2005, Evidence for late Oligocene establishment of the Antarctic

Circumpolar Current, Earth and Planetary Science Letters, v. 235, p. 715–728.

Pierce, E. L., T. Williams, T. van de Flierdt, S. R. Hemming, S. L. Steven, and Brachfeld, S. A.,

2012. Characterizing the sediment provenance of East Antarctica's weak underbelly: the

Wilkes and Aurora sub-glacial basins. Paleoceanography, v. 26(4), PA4217.

Piper D.J.W., Brisco, C.D. 1975. Deep-water continental-margin sedimentation, DSDP leg 28,

Antarctica, in Hayes, D.E. Frakes, L.A., et al. (eds) Initial Reports of the Deep Sea

Drilling Project, v. 28, p. 727-755, Washington.

Pollard D., DeConto, R., 2005. Hysteresis in Cenozoic Antarctic ice-sheet variations, Global and

Planetary Change, 45, 9-21.

208

Pollard, D., and DeConto, R.M., 2009, Modelling West Antarctic ice sheet growth and collapse

through the past five million years, Nature, v. 458, p. 329-332.

Pollard, D., DeConto, R.M., and Alley, R.B., 2015, Potential Antarctic Ice Sheet retreat driven

by hydrofracturing and ice cliff failure, Earth and Planetary Science Letters, v. 412, p.

112-121.

Pospichal, J.J., and Wise, S.W., 1990, Paleocene to Middle Eocene calcareous nannofossil of

ODP Sites 689 and 690, Maud Rise, Weddell Sea, in Barker, P.F., Kennett, J.P., et al,

Proceedings of the Ocean Drilling Program, Scientific Results, v. 113.

Post, N. J., Hensen, B. J., Kinny, P. D., 1996, Two metamorphic episodes during a 1340-1180

Ma convergent tectonic event in the Windmill Islands, East Antarctica. In: The Antarctic

region; geological evolution and processes, proceedings of the VII ISAES, p. 157-161.

Post, N. J., 2000, Unravelling Gondwana fragments; an integrated structural, isotopic and

petrographic investigation of the Windmill Islands, Antarctica (PhD Thesis: University of

New South Wales).

Powell, R.D., 1984, Glacimarine processes and inductive lithofacies modelling of ice shelf and

tidewater glacier sediments based on Quaternary examples, Marine Geology, v. 57, p. 1-

52.

Prebble, J.G., Raine, J.I., Barrett, P.J., Hannah, M.J., 2006. Vegetation and climate from two

Oligocene glacioeustatic sedimentary cycles (31 and 24 Ma) cores by the Cape Roberts

Project, Vitoria Land Basin Antarctica, Palaeogeography, Palaeoclimatology,

Palaeoecology, v. 231, p. 41-57.

Pross., J., and 17 others, 2012, Persistent near-tropical warmth on the Antarctic continent during

the early Eocene epoch, Nature, v. 488, p. 73-77.

209

Purucker, M.E., Langel, R.A., Rajaram, M., and Raymond, C., 1998, Global magnetization

models with a priori information, Journal of Geophysical Research, v. 103, p. 2,563-

2,584.

Purucker, M.E., von Frese, R.R.B., and Taylor, P.T., 1999, Mapping and interpretation of

satellite magnetic anomalies from POGO data over the Antarctic region, Annali di

geofisicia, v. 42, p. 215–228.

Rack, W., and Rott, H., 2001, Pattern of retreat and disintegration of the Larsen B ice shelf,

Antarctic Peninsula, Annals of Glaciology, v. 39, p. 505-510.

Raine, J.I., 1998, Terrestrial palynomorphs from Cape Roberts Project drillhole CRP-1, Ross

Sea, Antarctica. Terra Antarctica, v. 5, p. 539-548.

Raine, J.I., and Askin, R.A., 2001, Oligocene and Early Miocene terrestrial palynology of the

Cape Roberts Drillhole CRP-2/2A, Victoria Land Basin, Antarctica, Terra Antarctica, v.

7, p. 389-400.

Reeves, C., de Wit, M., 2000. Making ends meet in Gondwana: retracing the transforms of the

Indian Ocean and reconnecting continental shear zones. Terra Nova, v. 12, p. 272-280.

Richard, S.M., Smith, C.H., Kimbrough, D.L., Fitzgerald, P.G., Luyendyk, B.P., and

McWilliams, M.O., 1994, Cooling history of the northern Ford Ranges, Marie Byrd

Land, West Antarctica, Tectonics, v. 13, p. 837-857.

Rignot, E., Mouginot, J., and Scheuchl, B., 2011. Ice flow of the Antarctic Ice Sheet. Science,

333, 1427-1430.

Roberts, A.P., Wilson, G.S., Harwood, D.M., Verosub, K.L., 2003, Glaciation across the

Oligocene–Miocene boundary in Southern McMurdo Sound, Antarctica: new chronology

210

from the CIROS-1 drill hole, Palaeogeography, Paleoclimatology, Palaeoecology, v. 198,

p. 113– 130.

Rocchi, S., LeMasurier, W.E., and Di Vincenzo, G., 2006, Oligocene to Holocene erosion and

glacial history in Marie Byrd Land, West Antarctica, inferred from exhumation of the

Dorrel Rock intrusive complex and from volcano morphologies, Geological Society of

America Bulletin, v. 118, p. 991-1005.

Rooney, S.T., Blankenship, D.D., Alley, R.B., and Bentley, C.R., 1991, Seismic reflection

profiling of a sediment-filled graben beneath ice stream B, West Antarctica, In:

Thomson, M.R.A., Crame, J.A., and Thomson, J.W. (Eds.), Geological evolution of

Antarctica, Cambridge University Press, New York, p. 261-265.

Rowell, A.J., Rees, M.N., Duebendorfer, E.M., Wallin, E.T., Van Schmus, W.R., Smith, E.I.,

1993. An active Neoproterozoic margin: evidence from the Skelton Glacier area,

Transantarctic Mountains. Journal of the Geological Society of London v. 150, p. 677-

682.

Roy, M., van de Flierdt, T., Hemming, S.R., and Goldstein, S.L., 2007. 40Ar/39Ar ages of

hornblende grains and bulk Sm/Nd isotopes of Circum-Antarctic glacio-marine

sediments: Implications for sediment provenance in the Southern Ocean, Chem. Geol., v.

244, p. 507-519.

Schandl, E.S., Gorton, M.P., and Wicks, F.J., 1990, Mineralogy and geochemistry of alkali

basalts from Maud Rise, Weddell Sea, Antarctica, in Barker, P.F., Kennett, J.P., et al.,

Proceedings of the Ocean Drilling Program, Scientific Results, v. 113.

211

Scher, H.D., and Martin, E.E., 2004, Circulation in the Southern Ocean during the Paleogene

inferred from neodymium isotopes, Earth and Planetary Science Letters, v. 228, p. 391-

405.

Scherer, R.P., Aldahan, A., Tulaczyk, S., Possnert, G., Engelhardt, H., and Kamb, B., 1998,

Pleistocene collapse of the West Antarctic ice sheet, Science, v. 281, p. 82–85.

Schoof, C., 2007, Ice sheet grounding line dynamics: Steady states, stability, and hysteresis,

Journal of Geophysical Research, v. 112, p. F03S28.

Seki, O., Foster, G.L., Schmidt, D.N., Mackensen, A., Kawamura, K., and Pancost, R.D., 2010,

Alkenone and boron-based Pliocene pCO2 records, Earth and Planetary Science Letters,

v. 292, p. 201-211.

Shackleton, N.J., 1974, Attainment of isotopic equilibrium between ocean water and the

benthonic foraminifera genus Uvigerina: Isotopic changes in the ocean during the last

glacial, Colloques internationaux Centre national de la recherche scientifique, v. 219, p.

203-225.

Shackleton, N.J., Hall, M.A., and Boersma, A., 1984, Oxygen and carbon isotope data from Leg

74 foraminifers, Initial Reports of the Deep Sea Drilling Project, v. 74, p. 599-612.

Shackleton, N.J., and Kennett, J.P., 1975, Late Cenozoic oxygen and carbon isotopic changes at

DSDP Site 284: implications for glacial history of the Northern hemisphere and

Antarctic. In: Kennett, J.P., Houtz, R.E., et al., (Eds.), Initial Reports of the Deep Sea

Drilling Project, vol. 29, pp. 801– 808.Sheraton, J. W., Black, L. P.,Tindle, A. G., 1992,

Petrogenesis of plutonic rocks in a Proterozoic granulite-facies terrane - the Bunger Hills,

East Antarctica. Chemical Geology, v. 97, p. 163-198.

212

Siegert, M.J., Carter, S., Tabacco, I., Popov, S., and Blankenship, D.D., 2005a, A revised

inventory of Antarctic subglacial lakes, Antarctic Science, v. 17, p. 453–480.

Siegert, M.J., Taylor, J., and Payne, A.J., 2005b, Spectral roughness of subglacial topography

and implications for former ice-sheet dynamics in East Antarctica, Global Planetary

Change, v. 45, p. 249–263.

Sijp, W.P., and England, M.H., 2004, Effect of the Drake Passage throughflow on global climate,

Journal of Physical Oceanography, v. 34, p. 1254–1266.

Smellie, J. L., 1987, Geochemistry and tectonic setting of Cenozoic alkali basalts from

Alexander Island, Antarctic Peninsula, Journal of Volcanology and Geothermal Research,

v. 32, p. 269-285.

Smith, D.G., Ledbetter, G.T., Ciesielski, P.F., 1983. Ice rafted volcanic ash in the Southern

Ocean during the last 100,100 years. Marine Geology, v. 53, 291-312.

Smits, F., Gentner, W., 1950, Argonbestimmummungen an Kalium-Mineralien I. Bestimmungen

an tertiären Kalisalzen. Geochim Cosmochim Acta, v. 1, p. 22-27.

Sorlien, C.C., Luyendyk, B.P., Wilson, D.S., Decesari, R.C., Bartek, L.R., and Diebold, J.B.,

2007, Oligocene development of the West Antarctic Ice Sheet recorded in eastern Ross

Sea strata, Geology, v. 35, 467-470.

Spies, V., 1990, Cenozoic Magnetostratigraphy of Leg 113 drill sites, Maud Rise, Weddell Sea,

Antarctica, in Barker, P.F., Kennett, J.P. et al., Proceedings of the Ocean Drilling

Program, Scientific Results, v. 113.

Stagg, H.M.J., and J.B. Willcox, 1992. A case for Australia-Antarctica separation in the

Neocomian (125 Ma). Tectonophysics, v. 210, p. 21-32. DOI: 10.1016/0040-

1951(92)90125-P

213

Stickley, C.E., Brinkhuis, H., Schellenberg, S.A., Sluijs, A., Rohl, U., Fuller, M., Grauert, M.,

Huber, M., Warnaar, J., and Williams, G.L., 2004, Timing and nature of the deepening of

the Tasmanian Gateway, Paleoceanography, v. 19, p. 1–18.

Storey, M., Saunders, A.D., Tarney, J., Leat, P., Thirlwall, M.F., Thompson, R.N., Menzies,

M.A., and Marriner, G.F., 1988, Geochemical evidence for plume-mantle interactions

beneath Kerguelen and Heard Islands, Indian Ocean, Nature, v. 336, p. 371-374.

Swain, G., Woodhouse, A., Hand, M., Barovich, K., Schwarz, M., and Fanning, C.M., 2005,

Provenance and tectonic development of the late Archaean Gawler Craton, Australia; U–

Pb zircon, geochemical and Sm–Nd isotopic implications, Precambrian Research, v. 141,

p. 106–136.

Talarico, F.M., Sandroni, S., 2011, Early Miocene basement clasts in ANDRILL AND-2A core

and their implications for paleoenvironmental changes in the McMurdo Sound region

(western Ross Sea, Antarctica), Global and Planetary Change, v. 78, p. 23-35.

Thomas, E., 1990, Late Cretaceous through Neogene deep-sea benthic foramnifers (Maud Rise,

Weddell Sea, Antarctica, in Barker, P.F., Kennett, J.P., et al, Proceedings of the Ocean

Drilling Program, Scientific Results, v. 113.

Thompson, G.R., and Hower, J., 1973, An explaniation for low radiometric ages from

glauconite, Geochimica et Cosmochimica Acta, v. 37, p. 1273-1491.

Thorn, V.C., 2001, Oligocene and Early Miocene phytoliths from CRP-2/2A and CRP-3,

Victoria Land Basin, Antarctica, Terra Antarctica, v. 8, p. 407-422.

Toft, P.B., and Arkani-Hamed, J., 1992, Magnetization of the Pacific Ocean lithosphere deduced

from Magsat data, Journal of Geophysical Research, v. 97, p. 4,387-4,406.

214

Tulaczyk, S., Kamb, B., Scherer, R.P, and Engelhardt, H.F., 1998, Sedimentary processes at the

base of a West Antarctic ice stream: constraints from textural and composition properties

of subglacial debris, Journal of Sedimentary Research, v. 68, p. 487-496.

Toggweiler, J.R., and Bjornsson, H., 2000, Drake Passage and palaeoclimate: Journal of

Quaternary Science, v. 15, p. 319–328.

Tripati, A., Backman, J., Elderfield, H., and Ferretti, P., 2005, Eocene bipolar glaciation

associated with global carbon cycle changes, Nature, v. 436, p. 341-346.

Vaughan, A.P.M., Storey, B.C., 2000. The eastern Palmer Land shear zone: a new terrane

accretion model for the Mesozoic development of the Antarctic Peninsula. Journal of the

Geological Society of London, v. 157, p. 1243-1256.

Vaughan, A.P.M., Pankhurst, R.J., Fanning, C.M., 2002. Amid-Cretaceous age for the Palmer

Land event, Antarctic Peninsula: implications for terrane accretion timing and Gondwana

palaeolatitudes. Journal of the Geological Society of London, v. 159, p. 113-116.

Veevers, J.J., and S.L. Eittreim, 1988. Reconstruction of Antarctica and Australia at breakup (95

± 5 Ma) and before rifting (160 Ma). Australian Journal of Earth Sciences, v. 35, p. 955-

362. DOI: 10.1080/08120098808729453

Vorren, T.O., Hald, M., Edvardsen, M., Lind-Hansen, O.W., 1983. Glacigenic sediments and

sedimentary environments on continental shelves: general principles with a case study

from the Norwegian shelf, in: J. Ehlers (ed) Glacial deposits in northern Europe, p. 61-73,

Rotterdam.

Wade, B.S., Pälike, H., 2004. Oligocene climate dynamics. Paleoceanography, v. 19, PA4019.

Wade, B.S., Houben, A.J.P., Quaijtaal, W., Schouten, S., Rosenthal, Y., Miller, K.G., Katz,

M.E., Wright, J.D., and Brinkhuis, H., 2012, Multiproxy record of abrupt sea-surface

215

cooling across the Eocene-Oligocene transition in the Gulf of Mexico, Geology, v. 40, p.

159-162.

Watkins, N.D., Gunn, B.M., Nougier, J., and Baksi, A.K., 1974, Kerguelen: Continental

fragment or oceanic island?, Geological Society of America Bulletin, v. 85, p. 201-212.

Watkins, N.D., Keany, J., Ledbetter, M.T., Huang, T.C., 1974. Antarctic glacial history from

analyses of ice rafted deposits in marine sediments: new model and initial tests. Science,

v. 186, p. 533-536.

Westerhold, T., Rohl, U., 2009. High resolution cyclostratigraphy of the early Eocene – new

insights into the origin of the Cenozoic cooling trend. Climate of the Past, v. 5, p. 309-

327.

Will, T.M., Zeh, A., Gerdes, A., Frimmel, H.E., Millar, I.L., Schmädicke, E., 2009.

Palaeoproterozoic to Palaeozoic magmatic and metamorphic events in the Shackleton

Range, East Antarctica: constraints from zircon and monazite dating, and implications for

the amalgamation of Gondwana. Precambrian Research, v. 172, p. 25-45.

Williams, R.S., and Ferrigno, J.G., 1999, Satellite Image Atlas of Glaciers of the World, Chapter

A: Introduction. U.S. Geological Survey Professional Paper 1386-A.

Williams T., van de Flierdt, T., Hemming, S.R., Chung, E., Roy, M., Goldstein, S.L., 2010.

Evidence for ice berg armadas from East Antarctica in the Southern Ocean during the late

Miocene and early Pliocene. Earth and Planetary Science Letters, v. 290, p. 251-361.

Wilson, D.S., and Luyendyk, B.P., 2009, West Antarctic paleotopography estimated at the

Eocene-Oligocene climate transition, Geophysical Research Letters, v. 36, L16302.

216

Wilson, D.S., Jamieson, S.S.R., Barrett, P.J., Leitchenkov, G., Gohl, K., Larter, R.D., 2012.

Antarctic topography at the Eocene-Oligocene boundary. Palaeogeography,

Palaeoclimatology, Palaeoecology, doi:10.1016/j.palaeo.2011.05.028

Wingham, D.J., Siegert, M.J., Shepherd, A., and Muir, A.S., 2006, Rapid discharge connects

Antarctic subglacial lakes, Nature, v. 440, p. 1033–1036.

Zachas J.C., Breza, J., and Wise, S.W., 1992, Early Oligocene ice-sheet expansion on Antarctica,

Sedimentological and isotopic evidence from Kerguelen Plateau, Geology, v. 20, p. 569-

573.

Zachos, J. C., T. M. Quinn, and K. A. Salamy, 1996. High-resolution (104 years) deep-sea

foraminiferal stable isotope records of the Eocene- Oligocene climate transition,

Paleoceanography, v. 11, p. 251–266.

Zachos, J.C., Pagani, M., Sloan, L., Thomas, E., and K. Billups, 2001. Trends, rhythms, and

aberrations in global climate 65 Ma to present. Science, v. 292, p. 686-693.

Zachos, J.C., Dickens, G.R., and Zeebe, R.E., 2008, An early Cenozoic perspective on

greenhouse warming and carbon-cycle dynamics, Nature, v. 451, p. 279-283.

217