University of Southern California Dissertations and Theses

Antarctic climate variability from greenhouse to icehouse world

(USC Thesis Other)

Antarctic climate variability from greenhouse to icehouse world

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Copyright 2022 Emily Jane Tibbett ANTARCTIC CLIMATE VARIABILITY FROM GREENHOUSE TO ICEHOUSE WORLD by Emily Jane Tibbett A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (GEOLOGICAL SCIENCES) May 2022 i Acknowledgements I would not have reached this accomplishment without the numerous friends, family, colleagues, and mentors in my life who have continued to support and encourage me. First, I must thank my advisor Prof. Sarah Feakins for her guidance and mentorship throughout my time at USC. Her dedication and prompt responsiveness to emails shaped my doctoral research and I am grateful for her support during my graduate degree. I also want to thank my committees and co-authors who have helped shaped my research and provided encouraging feedback on all questions and drafts. I also need to thank my mentors throughout my life who were willing to share their own experiences with me and took the time to guide me through high school, undergraduate and now through a doctorate degree. Without the mentorship I have received, I would not have sought a graduate degree and I am thankful for their insight and guidance. I want to thank my labmates both past and present such as Christine Wu, Hyejung Lee, Mark Peaple, Rachel So, Patrick Cho, and Efrain Vidal for all the laughs and comradery. I want to thank my friends inside the department and inside MEB for all the fun adventures from the dinner outings to hiking. I want to thank Jayme Feyhl-Buska, Tarryn Cawood and her husband Gareth Cawood, and Didi Bojanova for being my climbing partners and for the numerous outdoor adventures. I also want to thank Didi for being a close friend since our first year in the department and for her unwavering encouragement to try new things outside of my comfort zone that usually led to injuries on her part. I also need to thank my movie watching crew Thomas, Sarah and baby Arthur Luckie, Naomi Rodgers, and Jess Stellmann whose commentary always made me laugh. This thesis is also for my parents who always had confidence in me even when they did not understand the research I did at USC. I am thankful to my sister Sarah who has been there ii through all the ups and downs academically and personally, whose always on board for traveling adventures, and who I will always stand in long lines with for novelty food. I am thankful to my friends who predate my graduate career, Tara Mokhtarzadeh from undergrad, Nisa Dang, Samantha Huynh, and Kimberly Striegel from our high school swim team days. Their continuing lifelong friendship, snow adventures, and random zoom shenanigans have helped me through the roughest times in my life and through a pandemic and I am very grateful to have them in my life. Also, of course, I need to thank my dog, Nita, who continues to remind me there is more to life than work even when writing a thesis. This was already said but needs repeating: I am so grateful for the people from all areas of my life whom, without them, this journey to a doctorate degree would not have been the same. This thesis is dedicated to all my friends, family, collaborators, colleagues, and mentors whose kind sentiments and words of encouragement made this possible. iii Table of Contents ACKNOWLEDGEMENTS ............................................................................................................. i TABLE OF CONTENTS ............................................................................................................... iii LIST OF TABLES ....................................................................................................................... viii LIST OF FIGURES ....................................................................................................................... ix ABSTRACT ................................................................................................................................. xiii Chapter 1: INTRODUCTION..........................................................................................................1 1.1 Cenozoic History of the Antarctic .............................................................................................1 1.2 Use of Biomarkers in Antarctic Paleoclimate research .............................................................2 1.2.1 Plant Wax ..........................................................................................................................3 1.2.2 GDGTs ..............................................................................................................................5 1.2.2.1 brGDGTs ..............................................................................................................5 1.2.2.2 isoGDGTs .............................................................................................................6 1.2.3 Alkenones .........................................................................................................................7 1.3 Reconstruction of ocean temperature in the Southern Ocean ....................................................7 1.4 Biomarker sourcing and depositional settings ...........................................................................8 1.4.1 Plant wax ...........................................................................................................................8 1.4.2 GDGTs ...........................................................................................................................10 References ......................................................................................................................................13 CHAPTER 2: SNAPSHOTS OF PRE-GLACIAL PALEOENVIRONMENTAL CONDITIONS ALONG THE SABRINA COAST, EAST ANTARCTICA: NEW PALYNOLOGICAL AND BIOMARKER EVIDENCE ..............................................................20 Abstract ..........................................................................................................................................20 2.1 Introduction ..............................................................................................................................21 2.2 Material and methods ...............................................................................................................24 2.2.1 Material ...........................................................................................................................24 2.2.2 Age control: a review of uncertainties and hard facts .....................................................27 2.2.2.1 NBP14-02 JPC-54 and -55 .................................................................................27 iv 2.2.2.2 NBP14-02 JPC-30 and -31 .................................................................................29 2.2.3 Palynology .....................................................................................................................30 2.2.4 Biomarkers .....................................................................................................................31 2.2.5 Statistical Analysis .........................................................................................................33 2.3 Results ......................................................................................................................................33 2.3.1 Palynology .....................................................................................................................33 2.3.2 Biomarkers .....................................................................................................................46 2.4 Discussion ................................................................................................................................49 2.4.1 Palynomorph and biomarker provenance and regional hydroclimate ...........................49 2.4.2 Sabrina Coast paleoenvironments ..................................................................................53 2.5 Conclusions ..............................................................................................................................56 Acknowledgements ........................................................................................................................57 References ......................................................................................................................................58 CHAPTER 3: LATE EOCENE RECORD OF HYDROLOGY AND TEMPERATURE FROM PRYDZ BAY, EAST ANTARCTICA ..............................................................................64 Abstract ..........................................................................................................................................64 3.1 Introduction ..............................................................................................................................65 3.2 Material and Methods ..............................................................................................................68 3.2.1 Site Selection ..................................................................................................................68 3.2.2 Age Model ......................................................................................................................70 3.2.3 Palynology ......................................................................................................................71 3.2.4 Extraction of Lipids and Quantification of n-Alkanes and n-Alkanoic Acids ................72 3.2.5 Compound Specific Isotopic Analyses: δ 13 C and δD .....................................................73 3.2.6 GDGT Analyses ..............................................................................................................74 3.3 Results ......................................................................................................................................77 3.3.1 Pollen ..............................................................................................................................77 3.3.2 Plant Wax ........................................................................................................................79 3.3.2.1 Abundance ..........................................................................................................79 v 3.3.2.2 Compound Specific Isotopic Analyses ...............................................................81 3.3.3 GDGTs ............................................................................................................................82 3.4 Discussion ................................................................................................................................85 3.4.1 Excluding the Permian Influence in Late Eocene Sediments .........................................85 3.4.2 Terrestrial Inputs and Marine Productivity at Continental Margins ...............................89 3.4.3 Late Eocene Warmth between PrOM and the EOT ........................................................90 3.4.4 Hydroclimate of the Late Eocene ...................................................................................92 3.4.5 Glacial Expansion at the EOT .........................................................................................96 3.5 Conclusions ............................................................................................................................101 Acknowledgements ......................................................................................................................103 References ....................................................................................................................................104 CHAPTER 4: CENOZOIC ANTARCTIC PENINSULA TEMPERATURES AND GLACIAL EROSION SIGNALS FROM A MULTI-PROXY BIOMARKER STUDY OF SHALDRIL SEDIMENTS ..........................................................................................................113 Abstract ........................................................................................................................................113 4.1 Introduction ............................................................................................................................114 4.2 Study Location .......................................................................................................................116 4.3 Methods..................................................................................................................................117 4.3.1 Biomarker Extraction and Purification .........................................................................117 4.3.2 Liquid Chromatography Analyses ................................................................................118 4.3.3 Gas Chromatography Analyses .....................................................................................121 4.4 Results ....................................................................................................................................122 4.4.1 GDGTs ..........................................................................................................................122 4.4.1.1 BIT ....................................................................................................................122 4.4.1.2 brGDGTs ..........................................................................................................123 4.4.1.3 isoGDGTs .........................................................................................................123 4.4.2 Alkyl lipids....................................................................................................................124 vi 4.4.2.1 Alkanoic acids ...................................................................................................124 4.4.2.2 Alkanes .............................................................................................................125 4.4.2.3 Hopanes.............................................................................................................126 4.5 Discussion ..............................................................................................................................129 4.5.1 Eocene ..........................................................................................................................129 4.5.2 Oligocene .....................................................................................................................131 4.5.3 Miocene........................................................................................................................132 4.5.4 Pliocene ........................................................................................................................134 4.5.5 Iceberg transport in the Weddell Sea ...........................................................................135 4.5.6 Cenozoic Cooling.........................................................................................................136 4.6 Conclusions ............................................................................................................................141 Acknowledgements ......................................................................................................................142 References ....................................................................................................................................143 CHAPTER 5: PROXY-MODEL COMPARISON FOR THE EOCENE-OLIGOCENE TRANSITION IN SOUTHERN HIGH LATITUDES ................................................................152 Abstract ........................................................................................................................................152 5.1 Introduction ............................................................................................................................153 5.2 Methods..................................................................................................................................156 5.2.1 Proxy data .....................................................................................................................156 5.2.2 Models...........................................................................................................................160 5.3 Results ....................................................................................................................................165 5.3.1 Proxy-Model comparison..............................................................................................165 5.3.1.1 pCO2 ..................................................................................................................165 5.3.1.2 Ice sheet ............................................................................................................166 5.3.1.3 Paleogeography .................................................................................................167 5.3.2 CO2 scaling ....................................................................................................................168 5.4 Discussion ..............................................................................................................................171 vii 5.4.1 Ice sheet extent ..............................................................................................................171 5.4.2 Southern Ocean .............................................................................................................172 5.4.3 Declining pCO2 .............................................................................................................174 5.5 Conclusions ............................................................................................................................175 Acknowledgements ......................................................................................................................176 References ....................................................................................................................................177 CHAPTER 6: DISSERTATION CONCLUSIONS.....................................................................184 REFERENCES ............................................................................................................................186 APPENDIX A: SUPPLEMENTARY INFORMATION.............................................................211 A.1 Supplementary information for Chapter 3 ............................................................................211 A.2 Supplementary information for Chapter 4 ............................................................................220 A.3 Supplementary information for Chapter 5 ............................................................................223 viii List of Tables Table 2.1 Four-group SIMPER analysis with Bray-Curtis similarity matrix conducted for groups G1, G2, G3, and G4 .......................................................................................................................36 Table 5.1 Surface air temperature proxy compilation for the Eocene and Oligocene .................159 Table 5.2 Sea surface temperature proxy compilation for the Eocene and Oligocene ................160 Table 5.3 Model run parameters ..................................................................................................162 Table 5.4 Climate model RMSE for the surface air and sea surface temperature proxy-model comparison for the best fit pCO2 forcing .....................................................................................169 Table 5.5 Experiment to compare the effects of proxy uncertainty on pCO2 forcing estimates .......................................................................................................................................174 ix List of Figures Figure 1.1 Cenozoic Climate History ..............................................................................................2 Figure 1.2 Map of Antarctica with sites discussed in this thesis .....................................................3 Figure 2.1 Paleogeographic reconstruction for the Australian-Antarctic margins at 50 Ma .........23 Figure 2.2 Bathymetry of Sabrina Coast continental shelf ............................................................25 Figure 2.3 Core photos, core x-rays, lithologies and sample distributions for JPC-54, 55, 30, and 31 .............................................................................................................................................26 Figure 2.4 UPGMA cluster analysis of a Bray-Curtis similarity matrix with no stratigraphic constraint ........................................................................................................................................35 Figure 2.5 Palynomorph absolute and relative abundances, Shannon Diversity Index and assemblage data from JPC-55 (MS-III older core) ........................................................................38 Figure 2.6 UPGMA cluster analysis of a Bray-Curtis similarity matrix with stratigraphic constraint for all samples in JPC-55 ..............................................................................................39 Figure 2.7 Palynomorph absolute and relative abundances, Shannon Diversity Index and assemblage data from JPC-54 (MS-III, younger core) ..................................................................40 Figure 2.8 UPGMA cluster analysis of a Bray-Curtis similarity matrix with stratigraphic constraint for all samples in JPC-54 ..............................................................................................41 Figure 2.9 Palynomorph abundance, Shannon Diversity Index and relative abundance of dominant species for JPC-30 .........................................................................................................42 x Figure 2.10 Palynomorph abundance, Shannon Diversity Index and relative abundance of dominant species for JPC-31 .........................................................................................................42 Figure 2.11 Correspondence analysis of JPC-30, JPC-31, JPC-54, and JPC-55 assemblages ......43 Figure 2.12 Factorial space related to the correspondence analysis shown in Fig. 11 ..................44 Figure 2.13 Compilation of biomarker data for Core JPC 54 and 55 ............................................48 Figure 3.1 Map of Prydz Bay Study location ................................................................................68 Figure 3.2 Age model for the Prydz Bay sedimentary sequence compile from sections of 1166A, 742A, and 739C ................................................................................................................70 Figure 3.3 Multiproxy latest Eocene reconstruction from Prydz Bay ...........................................84 Figure 3.4 A detailed view of microfossil and biomarker evidence for the change in East Antarctica across the EOT .............................................................................................................90 Figure 4.1 Map of the present day Antarctic Peninsula bed elevation ........................................117 Figure 4.2 Dendrogram of n-alkane chain lengths with molecular abundance distribution ........126 Figure 4.3 Multi-proxy reconstruction of terrestrial environmental changes from the SHALDRIL II cores .....................................................................................................................128 Figure 4.4 Compiled Cenozoic proxy records from the continent of Antarctica and the surrounding Southern Ocean........................................................................................................138 Figure 5.1 Southern hemisphere surface air temperature model ensemble means and model RMSE for pCO2 surface air temperature runs ..............................................................................163 xi Figure 5.2 Southern hemisphere sea surface temperature model ensemble means and model RMSE for pCO2 sea surface temperature runs ..................................................................164 Figure 5.3 Summary of average RMSE across the model experiments ......................................165 Figure 5.4 Best proxy-model fit for both surface air and sea surface temperatures after scaling pCO2 ................................................................................................................................170 Figure A.1 Oblique view of the 3D landscape showing the catchment from the Gamburtsev Mountains delivering sediments to Prydz Bay ........................................................211 Figure A.2 Lithostratigraphic summary for Ocean Drilling Program (ODP) Holes 739C, 742A, and 1166A .........................................................................................................................212 Figure A.3 Lithostratigraphic summary for Ocean Drilling Program (ODP) Holes 739C, 742A, and 1166A with new multiproxy biomarker and microfossil data ....................................214 Figure A.4 Example chromatograms from Prydz Bay sediments ................................................216 Figure A.5 n-Alkanoic acid chain length distribution averaged across all samples with δ 13 C by chain length .............................................................................................................................218 Figure A.6 n-Alkanoic acid δ 13 C and δD values downcore .........................................................219 Figure A.7 Present day Antarctic bed elevation ..........................................................................220 Figure A.8 Data from SHALDRIL II sedimentary summary .....................................................221 Figure A.9 Dendrogram of n-alkanoic acid chain lengths with molecular abundance distribution ...................................................................................................................................222 xii Figure A.10 Southern hemisphere high latitude surface air temperature proxy comparison to pCO2 model runs for the Eocene, Oligocene, and the difference across the transition ...............223 Figure A.11 Southern hemisphere high latitude sea surface temperature proxy comparison to pCO2 model runs for the Eocene, Oligocene, and the difference across the transition ...............224 Figure A.12 Southern hemisphere high latitude surface air temperature proxy comparison to ice model runs for the Eocene, Oligocene, and the difference across the transition ...................225 Figure A.13 Southern hemisphere high latitude sea surface temperature proxy comparison To ice model runs for the Eocene, Oligocene, and the difference across the transition ..............226 Figure A.14 Southern hemisphere high latitude surface air temperature proxy comparison to paleogeography model runs for the Eocene, Oligocene, and the difference across the transition ......................................................................................................................................227 Figure A.15 Southern hemisphere high latitude sea surface temperature proxy comparison to paleogeography model runs for the Eocene, Oligocene, and the difference across the transition ......................................................................................................................................228 Figure A.16 Proxy-model comparison for models scaled to a 25% reduction in pCO2 across the Eocene-Oligocene Transition .................................................................................................229 xiii Abstract The Cenozoic can be divided into a Greenhouse (Paleocene and Eocene) and an Icehouse world (Oligocene to present). This division is based on the presence of permanent ice sheets in Antarctica; however, the glacial history of Antarctica is more complex with periods of ice sheet growth and retreat of the East, West, and Antarctic Peninsula ice sheets. On the fully glaciated continent today, there are few records of past climate accessible, however the marine margins provide evidence for conditions on land. With the development of biomarker methodologies, these marginal sediments from around the Antarctic continent can now yield new proxy evidence for the fluctuating climate history of Antarctica. This thesis revisits legacy cores drilled over recent decades and finds new evidence for Antarctic climate across the Cenozoic. The first record, Chapter 2, captures a snapshot of the Paleocene/Eocene offshore of the Sabrina Coast, East Antarctica, where few records exist. Application of plant wax dual isotopes (δD and δ 13 C) in combination with pollen analysis suggest that the Sabrina Coast region consisted of an open canopy woodland or shrubby tundra with δD of precipitation similar to today. To evaluate how the terrestrial climate shifted from the Greenhouse to the Icehouse world, Chapter 3 generates a record spanning the Eocene-Oligocene Transition (EOT) from Prydz Bay, at the outflow of a major drainage basin for the East Antarctic ice sheet. Soil biomarkers revealed a 5°C cooling on land prior to the EOT and ocean paleothermometers revealed a decrease of 4°C at the EOT with additional changes in marine productivity. In addition, we found plant wax n-alkanoic acid whose δD and δ 13 C values identify increasing aridity across this transition. Whereas plant wax n- alkanoic acids are penecontemporary, other biomarkers show inputs of older material reworked by glacial erosion (n-alkanes and hopanes). To compare timeslices of the Cenozoic, Chapter 4, generates a biomarker record offshore of the Antarctic Peninsula to capture evidence for xiv terrestrial and marine conditions for the late Eocene, late Oligocene, mid-Miocene, and Pliocene. This record captures declining vegetation with increasing glacial erosional inputs of older strata identified by hopanes. Chapter 5 compiles proxy temperature records from around the Antarctic for the late Eocene and Early Oligocene including those generated in this thesis. These records are compared to surface air and sea surface temperatures generated from previous model simulations across the EOT. As EOT model experiments were forced with larger than expected pCO2 decreases, the model experiments were scaled to a more realistic pCO2 drop, assessed by fitting to the temperature decreases found in the proxies. The temperature-scaled pCO2 drop used for the revised EOT model scenario, now better fits pCO2 proxy reconstructions. In summary, this thesis adds new proxy data for the Antarctic, reinforces the value of multi-biomarker and multi-proxy comparisons, and demonstrates how screening for glacial reworking allows us to reconstruct past climate in polar regions across glacial transitions. The added records help to provide a proxy comparison to climate model simulations and thus to test the mechanisms needed to drive the glaciation of Antarctica. 1 Chapter 1 Introduction 1.1 Cenozoic history of Antarctica The Cenozoic (the last 63 million years) consists of the greenhouse climate state of the Palaeocene and Eocene and icehouse conditions (with polar glaciation) from the Oligocene to present (Fig 1.1). Overall, the Cenozoic was a time of long term cooling. During the warmth of the early Cenozoic, the Palaeocene-Eocene Thermal Maximum was a rapid global warming event around 56 Ma and lasted 200 kyr (Kennett & Stott, 1991). This hyperthermal was associated with a carbon isotope excursion that occurred due to the rapid release of 13 C depleted carbon (Meissner et al., 2014), followed by a gradual increase in temperature into the Early Eocene Climate Optimum (Zachos et al., 1996) which was a long term warming trend (Huber & Caballero, 2011). This was then followed by a long term cooling across the late Eocene into the Oligocene. The late Eocene was punctuated by the Priabonian Oxygen Minimum (37.3 Ma) (Carter et al., 2017) suggesting a temporary expansion of ice in Antarctica prior to the Eocene- Oligocene Transition. The Eocene-Oligocene Transition (EOT) is a period of 500 kyr (34.1 to 33.6 Ma) that marks the shift from the greenhouse conditions of the Eocene to the icehouse conditions of the Oligocene and onward. A two-step increase in benthic foraminiferal δ 18 O of ~1.5‰ (Zachos et al., 2001; Coxall et al., 2005) reflects a 2.5°C cooling of bottom waters and growth of a continent-wide ice sheet on Antarctica (Bohaty et al., 2012; Lear et al., 2008). The Oligocene climate is unique compared to the late Neogene icehouse as there was a return to warm surface temperatures similar to the Eocene greenhouse (O’Brien et al., 2020) as noted by a warm nannofossil assemblage (Villa et al., 2014) and late Oligocene deep-water temperatures similar to the late Eocene supported by Mg/Ca estimates (Lear et al., 2004). On Antarctica late 2 Oligocene to early Miocene pollen assemblages from the Ross Sea suggest a tundra landscape with a relatively stable terrestrial environment (Kulhanek et al., 2019). The Oligocene-Miocene Boundary is a period of glacial expansion due to low seasonality orbit (Zachos et al., 2001). The Miocene is punctuated by the Mid-Miocene Climate Optimum (17-14.7 Ma) with elevated temperatures (Pound et al., 2012) and reduced Antarctic ice sheet volumes based on benthic δ 18 O (Holbourn et al., 2013) and palynology suggesting a warmer Antarctic at this time (S. Warny et al., 2009). As we enter the Pliocene, we have the Northern Hemisphere ice sheet formation and the growth of the West Antarctic Ice sheet. Across the Pleistocene there are fluctuations in the extent of the Antarctic Ice Sheet during interglacial intervals with a retreat and thinning of the East Antarctic Ice Sheet (Wilson et al., 2018). Fig 1.1: Overview of the Cenozoic climate history and the Antarctic ice sheets. δ 18 Obenthic record spline (Westerhold et al., 2020) and pCO2 record compiled from δ 11 B (blue) and alkenones (red) proxies (Rae et al., 2021). MMCO (Mid-Miocene climate optimum), EOT (Eocene-Oligocene Transition), PrOM (Priabonian Oxygen Maximum), PETM (Paleocene-Eocene Thermal Maximum), and AA (Antarctica) 1.2 Use of biomarkers in Antarctic paleoclimate research On Antarctica, terrestrial archives of the early Cenozoic are limited by glacial ice cover due to glacial erosion leading to missing geological units or are covered by the ice sheet. Accessible 3 Eocene outcrops are only found on the very northern limits at 64°S on Cockburn Island (Askin et al., 1991); on Seymour Island (Ivany et al., 2011); King George Island (Sophie Warny et al., 2019) or on exposed land in the Transantarctic Mountains marking the boundary between East and West Antarctica (Ashworth et al., 2007). Other records of conditions prior to glaciation come from drilling expeditions around the continental margin such as the SHALDRIL expedition off the tip of the Antarctic Peninsula (63°S) (Fig 1.2) that captured a short window of time prior to the EOT (Anderson et al., 2011; S. S. Warny & Askin, 2011). Most applications of biomarkers to reconstruct Antarctic paleoclimate come from drill cores from the Southern Ocean. Fig 1.2: Map of Antarctica from NOAA National Centers for Environmental Information. Yellow stars are locations studied in this dissertation 1.2.1 Plant wax A handful of plant wax records around Antarctica have been generated across the Cenozoic. The Sabrina Coast, Wilkes Land record with possible ages from the late Cretaceous to the Paleocene- Eocene recovered pollen and plant wax with the plant wax isotope record indicating sourcing 4 from the coast due to more enriched δDwax values (Chapter 2) (Fig 1.2). The combination of plant wax isotopes and pollen suggest a drier, more open type of coastal vegetation (Duffy et al., 2021) versus a tropical rainforest as previously thought. Sediment from the Antarctic Peninsula yielded a snapshot of climate prior to the EOT from pollen and plant wax, with carbon isotopic analyses on the pollen yielding evidence for cooling and drying (Feakins et al., 2014; Griener et al., 2015). Hydrogen isotopic analyses on the plant waxes yielded precipitation isotopic estimates similar to modern, i.e. low sensitivity to the huge changes between ephemeral glaciation and the glaciated continent today (Feakins et al., 2014). Modeling comparison suggested 12°C prior to the EOT similar to other estimates based on pollen assemblages found in drill cores from Prydz Bay. Oligocene and Miocene sediment from the McMurdo glacial erractics, the Transantarctic Mountain range, the Cape Roberts Project from Victoria land, DSDP 270 and DSDP 274 were compared to assess reworking inputs in various deposition settings and catchments and determined that there was a shift to climate cooling from the Eocene to Oligocene (Duncan et al., 2019). Miocene record from ANDRILL AND-2A core from the Ross Sea identified periods of vegetation expansion coinciding with peak warmth (Feakins et al., 2012). Combined with model simulations summer temperatures were estimated to be 11°C, warmer than today with increase moisture delivery based on δD values higher than modern (Feakins et al., 2012). The aliphatic lipid record from the Transantarctic Mountains indicates a mixed vegetation during the Neogene (Rees-Owen et al., 2018). Holocene plant wax records from the Faroe Islands suggest a drying trend across the Holocene with the last interglacial period as warmer and wetter than present (Curtin et al., 2019). An additional Holocene record from South Georgia island suggest cooling across the Holocene with less vegetation cover (Berg et al., 2019). 5 1.2.2 GDGTs GDGTs (glycerol dialkyl glycerol tetraethers) are membrane lipids produced by bacteria and archaea. There are two types of GDGTs the isoprenoidal (isoGDGTs) and the branched (brGDGTs) each with a large number of structural forms or moieties. Applying the branched isoprenoidal tetraether index (BIT) allows distinguishing between terrestrial versus marine dominated sourcing. The isoGDGTs are commonly sourced from Thaumarchaeota (the dominant producer of Crenarchaeol) in the water column and used for reconstruction of ocean and sea surface temperatures using TEX86 (Tetraether index 86 carbon atoms) (Schouten et al., 2002). While the brGDGTs are primarily soil derived and reflect mean air temperature calculated using the methylated index of branched tetraethers (MBT ' 5Me) (De Jonge et al., 2014). The organisms that produce branched GDGTs in soils are unknown but likely to include acidobacteria (Damsté et al., 2011). The composition of the membrane lipids for both brGDGT and isoGDGTs changes in response to temperature predominantly, related to the need to adjust membrane fluidity. For brGDGT there is a secondary correlation with soil pH, both parameters can be reconstructed revealing both temperature and aridity. 1.2.2.1 brGDGTs In Antarctica the brGDGT temperature reconstruction using MBT ' 5Me has been applied across the Cenozoic. Paleogene temperatures after the K-Pg boundary estimate soil mean air temperature from 15-8°C suggesting a cool temperate climate on the Antarctic Peninsula indicating temporal heterogeneities in the southern high latitudes (Kemp et al., 2014). Application of brGDGTs for mean annual air temperature on Antarctica suggests summer temperatures of 3-7°C during the Miocene (Rees-Owen et al., 2018). Temperature reconstruction was generated for the first time on East Antarctic from Prydz Bay (Tibbett et al., 2021) with temperature reconstructed using 6 MBT ' 5Me and applying BayMBT (Dearing Crampton-Flood et al., 2020) estimating the late Eocene at 13°C decreasing to 8°C in the early Oligocene (Chapter 3). 1.2.2.2 isoGDGTs For the PETM TEX86 estimates placed SST near the East Tasman Plateau around 33°C (Sluijs et al., 2011). For the late Eocene into the early Oligocene TEX86 estimates range from 22-16°C (Houben et al., 2019). Middle to late Eocene ocean temperature from Seymour Island range from 10-17°C (Douglas et al., 2014). Throughout the Oligocene SST was 10-17°C in the Ross Sea (Hoem et al., 2020). Sea surface temperature was variable from the Oligocene to Miocene (23 to 8°C ) with variations on glacial-interglacial timescales suggested that the Antarctic ice volume may be less dynamic than originally thought (Hartman et al., 2018). Another record from the late Oligocene from Wilkes Land suggest SSTs of 14-9°C with SST variations corelating with glacial-interglacial cyclicity (Evangelinos et al., 2020). Miocene sea surface temperatures from TEX86 suggest temperatures ranging from 18 to 4°C from Wilkes Land (Sangiorgi et al., 2018) and 8 to 2°C from the Ross Sea (Levy et al., 2016). With other estimates placed temperature as high as 22°C to 12°C from the South Tasman Rise (Leutert et al., 2020). The temperature change occurred in conjunction with ice sheet expansion suggested that thermal isolation did not precede ice sheet expansion (Leutert et al., 2020). More recent applications of isoGDGTs are utilized to understand current changes in ice shelf around Antarctica. A 0.3 to 1.5°C increase in subsurface ocean temperature over the last 9000 years have driven major collapse and recession of the regional ice shelf (Etourneau et al., 2019). Another Holocene record indicates a cooling across the past 10kyr with a peak temperature of 3°C and the lowest temperature of 0°C (Etourneau et al., 2019). 7 1.2.3 Alkenones Alkenones are commonly utilized for the U k' 37 index which provide an estimate of sea surface temperatures and has been calibrated for the Southern Ocean with the lower limit at 5°C (Sikes et al., 1997; Sikes & Volkman, 1993). For the PETM U k' 37 temperature estimates ranging from 19 to 9°C across the same TEX86 record previously discussed (Houben et al., 2019). From the late Eocene to the early Oligocene U k' 37 indicates temperature upward of 22°C in the late Eocene to 8°C in the early Oligocene (Plancq et al., 2014). Pleistocene records from the Pacific sector of the Southern Ocean suggest glacial cooling of 8°C in the subantarctic which is comparable to other sectors of the Southern Ocean suggesting a uniform circumpolar cooling at that time (Ho et al., 2012). Alkenones are present in sediment throughout the Eocene to present and Cretaceous alkenones have been reported; however, their alkenones differ from modern (Yamamoto et al., 1996) so caution should be used when applying SST farther back in time Alkenones have potential to reconstruct SSTs throughout the Cenozoic; however, there is a notable change in the ketone distribution in older sediment (de Bar et al., 2019). 1.3 Reconstruction of ocean temperatures in the Southern Ocean Southern Ocean temperature reconstructions have predominantly been determined through TEX86 and Mg/Ca with some U k' 37 records. For the PETM TEX86 estimates placed SST around 33C (Sluijs et al., 2011). For the Southern Ocean records from SST range from 22°C–12°C from the late Eocene to post-EOT from Maud Rise (Petersen & Schrag, 2015) and on mixed layer planktic foraminifera (Chiloquembelina cubensis) from Kerguelen Plateau sites 738 and 744, which are more proximal to the continent, with δ 18 O values that indicated temperatures between 5 and 10°C (Zachos et al., 1994). Similar temperatures are also suggested in the region based on modelling of ocean temperatures aligned with temperature reconstructions from further afield 8 (La Meseta formation, Seymour Island, Antarctic Peninsula) using clumped isotopes on bivalve shells and TEX86 (Douglas et al., 2014). For the late Eocene into the early Oligocene TEX86 estimates range from 22 to 16°C with U k’ 37 reporting temperatures ranging from 19 to 9°C across the same record (Houben et al., 2019). Miocene Southern Ocean sea surface temperatures range from 18 to 14°C fusing Mg/Ca from the South Tasman Rise (Shevenell et al., 2004). Seawater temperature from TEX86 suggest temperatures ranging from 4 to 18°C from Wilkes Land (Sangiorgi et al., 2018) and 8 to 2°C Ross Sea (Levy et al., 2016). Across the Cenozoic there is a cooling of Southern Ocean temperatures with discrepancies accounting for differences in locality and proximity to the continent. Decreasing pCO2 associated with global cooling and the thermal isolation of Antarctica due to an unhindered Antarctic Circumpolar Current explains the general decrease in ocean temperatures from the Paleocene to present. 1.4 Biomarker sourcing and depositional settings 1.4.1 Plant wax Plant waxes are protective coatings on plant leaves that are used to limit evaporation from leaves. The plant wax is deposited in soil and then is eroded from the soil and from leaves through precipitation and transported into rivers where they are transported downstream and offshore. Modern catchment studies inform on catchment sourcing indicating differences in riverine organic carbon cycling depending on the geomorphological controls (Galy et al., 2011) and that the plant wax from river-suspended sediment indicates uniform spatial integration (Feakins et al., 2018). However, it is noted that the lowland signal can dominate in more distal archives (Feakins et al., 2018). 9 Leaf waxes in large lakes and marine basins should consider the transport time of leaf wax compounds into the sedimentary record, ideally through radiocarbon dating (Li et al., 2011). There can be a lag in formation of organic biomarkers and delivery to sediments in marine settings. This lag can be attributed to extended residence times in systems prior to deposition this can occur in the open ocean due to high residence time in shelf sediments. Lag in terrestrial inputs and marine sediment δ 13 C were identified for the carbon isotope excursion for the Paleocene-Eocene Thermal Maximum indicating a delay of 4-5 kyr between n-alkane formation and deposition in marine sediment (Tipple et al., 2011). Also noted in modern catchment studies where the average age of the plant derived fatty acids from 50 to 1300 years (Galy & Eglinton, 2011). Terrestrial organic matter is delivered either via aeolian, fluvial, or glacial processes. When applying and interpreting plant wax and other biomarkers in glacially-influenced settings comparison to additional sedimentological and paleontological indicators are needed (Duncan et al., 2019). Sediment records from around Antarctica can be reworked due to glacial erosion of sediment and bedrock leading to older material than the sediment age. Evaluating the organic proxies utilizing known parameters will ensure the climate interpretation reflects contemporary environments. Marine records integrate catchment-scale conditions smoothing over local signals to reflect overall climate conditions in the catchment. Marginal setting terrestrial organic carbon deposition is controlled by fluvial runoff, catchment dynamics, and vegetation (Pancost & Boot, 2004). These can impact the residence time of biomarkers prior to deposition in marginal marine sediments. Fluvial delivery is more important in marginal settings versus open ocean where aeolian transport is more likely to dominated. Aeolian deposits in open ocean settings are 10 controlled by wind strength and direction and needs to be considered to determine source of plant wax in open ocean drill cores (Martínez-Garcia et al., 2011) Marine marginal settings are valuable archives of terrestrial conditions due to proximity, however they are complicated by changing delivery mechanisms (fluvial vs. glacial, proximity to shore, sea-level rise and fall). Detailed comparisons of the facies grain size, lithology paired with biomarker studies can aid interpretations of sourcing for terrestrial organic matter allowing accurate interpretations of the reconstruction temperatures and additional proxies. This was noted in Prydz Bay (Chapter 3) (Fig 1.2) where an increase in marine archeal productivity corresponds to a major depositional system change from prograding diamictite to diatomaceous mudstones. 1.4.2 GDGTs To recreate terrestrial climates, it is ideal to have coring locations at the outflow of drainage basins. In these environments it is expected that terrestrial material flowing from the continent into the ocean will dominate the record; however it is important to consider sourcing of brGDGTs when applied to lacustrine and marine sedimentary archives (Freeman & Pancost, 2014). This can be done by applying the branch and isoprenoidal tetraether (BIT) index (Hopmans et al., 2004) which assesses the input of terrestrial predominantly brGDGTs and marine predominantly isoGDGTs. Temperature and soil pH reconstructions utilize brGDGTs. It is It is necessary to understand possible environmental impacts on interpreting brGDGTs to ensure that the temperatures are not biased including in situ aquatic brGDGT production which can skew temperatures (Inglis et al., 2019). However, in environments where terrestrially material dominates the record, brGDGT production in the water column is negligible. Therefore, in Prydz Bay a record of mean annual air temperature and soil pH was generated across the Eocene-Oligocene Transition (Tibbett et al., 2021; Chapter 3). 11 In addition, isoGDGTs production can occur at depth. Whether they are produced in the subsurface or surface is dependent on where Thaumarchaeota resides. Application of TEX86 in regions with steep thermoclines and nutriclines better reflect subsurface temperatures (Kim et al., 2015; Lopes dos Santos et al., 2010). When applying TEX86 it is important to consider whether production occurs at the surface or subsurface to better interpret temperatures generated utilizing the proxy. Coastal margins have a limited temperature gradient in Antarctica and therefore, sea surface temperatures can be reconstructed which were generated for both Prydz Bay and the Antarctic Peninsula (Fig 1.2) (Tibbett et al., 2021) (Chapter 3 and Chapter 4). Sea surface temperatures from GDGTs are limited in settings where most of the sediment input is from land. This can be assessed through the BIT index and the Ring index (ΔRI). The ring index assesses additional non-thermal influences (land derived isoGDGTs, growth rates, etc) on the isoGDGT distribution which can lead to inaccurate TEX86 SST reconstructions. Due to this, limited SST reconstructions were viable in Prydz Bay along the coastal margin (Chapter 3). To create SSTs record for the Eocene-Oligocene Transition we can assess marine cores further offshore in the open ocean. This dissertation contains the following published and in-preparation manuscripts: Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., & Feakins, S. J. (2021). Late Eocene Record of Hydrology and Temperature From Prydz Bay, East Antarctica. Paleoceanography and Paleoclimatology, 36(4), e2020PA004204. https://doi.org/10.1029/2020PA004204 (Chapter 2) The co-authors contributed to project conceptualization, methodology, funding acquisition, and reviewing of the manuscript. 12 Duffy, M., Tibbett, E. J., Smith, C., Warny, S., Feakins, S. J., Escarguel, G., Askin, R., Leventer, A., & Shevenell, A. E. (2021). Snapshots of pre-glacial paleoenvironmental conditions along the Sabrina Coast, East Antarctica: New palynological and biomarker evidence. Geobios. https://doi.org/10.1016/j.geobios.2021.09.001 (Chapter 3) Duffy (pollen) and Tibbett (biomarkers) co-led their respective sections of the manuscript including performing analyses and writing of the methodological and results components of the manuscript. The co-authors contributed to project conceptualization, methodology, funding acquisition, and reviewing of the manuscript. Tibbett, E.J., Warny, S., Tierney, J. E., & Feakins, S. J. Cenozoic Antarctic Peninsula temperatures and glacial erosion signals from a multi-proxy biomarker study of SHALDRIL sediments, in review Paleoceanography and Paleoclimatology (Chapter 4) The co-authors contributed to project conceptualization, methodology, funding acquisition, and reviewing of the manuscript. Tibbett, E.J., Burls, N.J., Hutchinson, K.D., & Feakins, S. J. Proxy-Model Comparison for the Eocene-Oligocene Transition in Southern High Latitudes (Chapter 5) The co-authors contributed to project conceptualization, methodology, funding acquisition, and reviewing of the manuscript. 13 References Anderson, J. B., Warny, S., Askin, R. A., Wellner, J. S., Bohaty, S. M., Kirshner, A. E., Livsey, D. N., Simms, A. R., Smith, T. R., Ehrmann, W., Lawver, L. A., Barbeau, D., Wise, S. W., Kulhanek, D. K., Weaver, F. M., & Majewski, W. (2011). Progressive Cenozoic cooling and the demise of Antarctica’s last refugium. Proceedings of the National Academy of Sciences, 108(28), 11356. https://doi.org/10.1073/pnas.1014885108 Ashworth, A., Lewis, A., Marchant, D., Askin, R., Cantrill, D., Francis, J., Leng, M., Newton, A., Raine, J., & Williams, M. (2007). The Neogene biota of the Transantarctic Mountains. Askin, R. A., Elliot, D. H., Stilwell, J. D., & Zinsmeister, W. J. (1991). Stratigraphy and paleontology of Campanian and Eocene sediments, Cockburn Island, Antarctic Peninsula. Journal of South American Earth Sciences, 4(1), 99–117. https://doi.org/10.1016/0895- 9811(91)90021-C Belt, S., Smik, L., Brown, T., Kim, J.-H., Rowland, S., Allen, C., Gal, J.-K., Shin, K.-H., Lee, J., & Taylor, K. (2016). Source identification and distribution reveals the potential of the geochemical Antarctic sea ice proxy IPSO 25. Nature Communications, 7(1), 1–10. Belt, S. T. (2018). Source-specific biomarkers as proxies for Arctic and Antarctic sea ice. Organic Geochemistry, 125, 277–298. https://doi.org/10.1016/j.orggeochem.2018.10.002 Belt, S. T., Massé, G., Rowland, S. J., Poulin, M., Michel, C., & LeBlanc, B. (2007). A novel chemical fossil of palaeo sea ice: IP25. Organic Geochemistry, 38(1), 16–27. Berg, S., White, D. A., Jivcov, S., Melles, M., Leng, M. J., Rethemeyer, J., Allen, C., Perren, B., Bennike, O., & Viehberg, F. (2019). Holocene glacier fluctuations and environmental changes in subantarctic South Georgia inferred from a sediment record from a coastal inlet. Quaternary Research, 91(1), 132–148. Cambridge Core. https://doi.org/10.1017/qua.2018.85 Bohaty, S. M., Zachos, J. C., & Delaney, M. L. (2012). Foraminiferal Mg/Ca evidence for Southern Ocean cooling across the Eocene–Oligocene transition. Earth and Planetary Science Letters, 317–318, 251–261. https://doi.org/10.1016/j.epsl.2011.11.037 Carter, A., Riley, T. R., Hillenbrand, C.-D., & Rittner, M. (2017). Widespread Antarctic glaciation during the Late Eocene. Earth and Planetary Science Letters, 458, 49–57. https://doi.org/10.1016/j.epsl.2016.10.045 Collins, L. G., Allen, C. S., Pike, J., Hodgson, D. A., Weckström, K., & Massé, G. (2013). Evaluating highly branched isoprenoid (HBI) biomarkers as a novel Antarctic sea-ice proxy in deep ocean glacial age sediments. Quaternary Science Reviews, 79, 87–98. Curtin, L., D’Andrea, W. J., Balascio, N., Pugsley, G., de Wet, G., & Bradley, R. (2019). Holocene and Last Interglacial climate of the Faroe Islands from sedimentary plant wax 14 hydrogen and carbon isotopes. Quaternary Science Reviews, 223, 105930. https://doi.org/10.1016/j.quascirev.2019.105930 Damsté, J. S. S., Rijpstra, W. I. C., Hopmans, E. C., Weijers, J. W., Foesel, B. U., Overmann, J., & Dedysh, S. N. (2011). 13, 16-Dimethyl octacosanedioic acid (iso-diabolic acid), a common membrane-spanning lipid of Acidobacteria subdivisions 1 and 3. Applied and Environmental Microbiology, 77(12), 4147–4154. de Bar, M. W., Rampen, S. W., Hopmans, E. C., Sinninghe Damsté, J. S., & Schouten, S. (2019). Constraining the applicability of organic paleotemperature proxies for the last 90 Myrs. Organic Geochemistry, 128, 122–136. https://doi.org/10.1016/j.orggeochem.2018.12.005 De Jonge, C., Hopmans, E. C., Zell, C. I., Kim, J.-H., Schouten, S., & Sinninghe Damsté, J. S. (2014). Occurrence and abundance of 6-methyl branched glycerol dialkyl glycerol tetraethers in soils: Implications for palaeoclimate reconstruction. Geochimica et Cosmochimica Acta, 141, 97–112. https://doi.org/10.1016/j.gca.2014.06.013 Dearing Crampton-Flood, E., Tierney, J. E., Peterse, F., Kirkels, F. M. S. A., & Sinninghe Damsté, J. S. (2020). BayMBT: A Bayesian calibration model for branched glycerol dialkyl glycerol tetraethers in soils and peats. Geochimica et Cosmochimica Acta, 268, 142–159. https://doi.org/10.1016/j.gca.2019.09.043 Douglas, P. M. J., Affek, H. P., Ivany, L. C., Houben, A. J. P., Sijp, W. P., Sluijs, A., Schouten, S., & Pagani, M. (2014). Pronounced zonal heterogeneity in Eocene southern high- latitude sea surface temperatures. Proceedings of the National Academy of Sciences, 111(18), 6582. https://doi.org/10.1073/pnas.1321441111 Duncan, B., McKay, R., Bendle, J., Naish, T., Inglis, G. N., Moossen, H., Levy, R., Ventura, G. T., Lewis, A., Chamberlain, B., & Walker, C. (2019). Lipid biomarker distributions in Oligocene and Miocene sediments from the Ross Sea region, Antarctica: Implications for use of biomarker proxies in glacially-influenced settings. Palaeogeography, Palaeoclimatology, Palaeoecology, 516, 71–89. https://doi.org/10.1016/j.palaeo.2018.11.028 Etourneau, J., Sgubin, G., Crosta, X., Swingedouw, D., Willmott, V., Barbara, L., Houssais, M.- N., Schouten, S., Damsté, J. S. S., Goosse, H., Escutia, C., Crespin, J., Massé, G., & Kim, J.-H. (2019). Ocean temperature impact on ice shelf extent in the eastern Antarctic Peninsula. Nature Communications, 10(1), 304. https://doi.org/10.1038/s41467-018- 08195-6 Evangelinos, D., Escutia, C., Etourneau, J., Hoem, F., Bijl, P., Boterblom, W., van de Flierdt, T., Valero, L., Flores, J.-A., Rodriguez-Tovar, F. J., Jimenez-Espejo, F. J., Salabarnada, A., & López-Quirós, A. (2020). Late Oligocene-Miocene proto-Antarctic Circumpolar Current dynamics off the Wilkes Land margin, East Antarctica. Global and Planetary Change, 191, 103221. https://doi.org/10.1016/j.gloplacha.2020.103221 15 Feakins, S. J., Warny, S., & DeConto, R. M. (2014). Snapshot of cooling and drying before onset of Antarctic Glaciation. Earth and Planetary Science Letters, 404, 154–166. https://doi.org/10.1016/j.epsl.2014.07.032 Feakins, S. J., Warny, S., & Lee, J.-E. (2012). Hydrologic cycling over Antarctica during the middle Miocene warming. Nature Geoscience, 5(8), 557–560. https://doi.org/10.1038/NGEO1498 Feakins, S. J., Wu, M. S., Ponton, C., Galy, V., & West, A. J. (2018). Dual isotope evidence for sedimentary integration of plant wax biomarkers across an Andes-Amazon elevation transect. Geochimica et Cosmochimica Acta, 242, 64–81. https://doi.org/10.1016/j.gca.2018.09.007 Freeman, K. H., & Pancost, R. D. (2014). 12.15—Biomarkers for Terrestrial Plants and Climate. In H. D. Holland & K. K. Turekian (Eds.), Treatise on Geochemistry (Second Edition) (pp. 395–416). Elsevier. https://doi.org/10.1016/B978-0-08-095975-7.01028-7 Galy, V., & Eglinton, T. (2011). Protracted storage of biospheric carbon in the Ganges– Brahmaputra basin. Nature Geoscience, 4(12), 843–847. https://doi.org/10.1038/ngeo1293 Galy, V., Eglinton, T., France-Lanord, C., & Sylva, S. (2011). The provenance of vegetation and environmental signatures encoded in vascular plant biomarkers carried by the Ganges– Brahmaputra rivers. Earth and Planetary Science Letters, 304(1–2), 1–12. Griener, K. W., Warny, S., Askin, R., & Acton, G. (2015). Early to middle Miocene vegetation history of Antarctica supports eccentricity-paced warming intervals during the Antarctic icehouse phase. Global and Planetary Change, 127, 67–78. https://doi.org/10.1016/j.gloplacha.2015.01.006 Hartman, J. D., Sangiorgi, F., Salabarnada, A., Peterse, F., Houben, A. J. P., Schouten, S., Brinkhuis, H., Escutia, C., & Bijl, P. K. (2018). Paleoceanography and ice sheet variability offshore Wilkes Land, Antarctica – Part 3: Insights from Oligocene–Miocene TEX86-based sea surface temperature reconstructions. Climate of the Past, 14(9), 1275– 1297. https://doi.org/10.5194/cp-14-1275-2018 Ho, S. L., Mollenhauer, G., Lamy, F., Martínez-Garcia, A., Mohtadi, M., Gersonde, R., Hebbeln, D., Nunez-Ricardo, S., Rosell-Melé, A., & Tiedemann, R. (2012). Sea surface temperature variability in the Pacific sector of the Southern Ocean over the past 700 kyr. Paleoceanography, 27(4). https://doi.org/10.1029/2012PA002317 Hoem, F. S., Valero, L., Evangelinos, D., Escutia, C., Duncan, B., McKay, R. M., Brinkhuis, H., Sangiorgi, F., & Bijl, P. K. (2020). Temperate Oligocene surface ocean conditions offshore Cape Adare, Ross Sea, Antarctica. Climate of the Past Discussions, 2020, 1–32. https://doi.org/10.5194/cp-2020-139 Holbourn, A., Kuhnt, W., Clemens, S., Prell, W., & Andersen, N. (2013). Middle to late Miocene stepwise climate cooling: Evidence from a high-resolution deep water isotope curve 16 spanning 8 million years. Paleoceanography, 28(4), 688–699. https://doi.org/10.1002/2013PA002538 Hopmans, E. C., Weijers, J. W. H., Schefuß, E., Herfort, L., Sinninghe Damsté, J. S., & Schouten, S. (2004). A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraether lipids. Earth and Planetary Science Letters, 224(1), 107–116. https://doi.org/10.1016/j.epsl.2004.05.012 Houben, A. J. P., Bijl, P. K., Sluijs, A., Schouten, S., & Brinkhuis, H. (2019, May 1). Late Eocene Southern Ocean Cooling and Invigoration of Circulation Preconditioned Antarctica for Full‐Scale Glaciation. Geochemistry, Geophysics, Geosystems. https://doi.org/10.1029/2019GC008182 Huber, M., & Caballero, R. (2011). The early Eocene equable climate problem revisited. Climate of the Past. Inglis, G. N., Farnsworth, A., Collinson, M. E., Carmichael, M. J., Naafs, B. D. A., Lunt, D. J., Valdes, P. J., & Pancost, R. D. (2019). Terrestrial environmental change across the onset of the PETM and the associated impact on biomarker proxies: A cautionary tale. Global and Planetary Change, 181, 102991. https://doi.org/10.1016/j.gloplacha.2019.102991 Ivany, L. C., Brey, T., Huber, M., Buick, D. P., & Schöne, B. R. (2011). El Niño in the Eocene greenhouse recorded by fossil bivalves and wood from Antarctica. Geophysical Research Letters, 38(16). https://doi.org/10.1029/2011GL048635 Kemp, D. B., Robinson, S. A., Crame, J. A., Francis, J. E., Ineson, J., Whittle, R. J., Bowman, V., & O’Brien, C. (2014). A cool temperate climate on the Antarctic Peninsula through the latest Cretaceous to early Paleogene. Geology, 42(7), 583–586. https://doi.org/10.1130/G35512.1 Kennett, J. P., & Stott, L. (1991). Abrupt deep-sea warming, palaeoceanographic changes and benthic extinctions at the end of the Palaeocene. Nature, 353(6341), 225–229. Kim, J.-H., Schouten, S., Rodrigo-Gámiz, M., Rampen, S., Marino, G., Huguet, C., Helmke, P., Buscail, R., Hopmans, E. C., Pross, J., Sangiorgi, F., Middelburg, J. B. M., & Sinninghe Damsté, J. S. (2015). Influence of deep-water derived isoprenoid tetraether lipids on the TEX86H paleothermometer in the Mediterranean Sea. Geochimica et Cosmochimica Acta, 150, 125–141. https://doi.org/10.1016/j.gca.2014.11.017 Kulhanek, D. K., Levy, R. H., Clowes, C. D., Prebble, J. G., Rodelli, D., Jovane, L., Morgans, H. E. G., Kraus, C., Zwingmann, H., Griffith, E. M., Scher, H. D., McKay, R. M., & Naish, T. R. (2019). Revised chronostratigraphy of DSDP Site 270 and late Oligocene to early Miocene paleoecology of the Ross Sea sector of Antarctica. Global and Planetary Change, 178, 46–64. https://doi.org/10.1016/j.gloplacha.2019.04.002 Lear, C. H., Bailey, T. R., Pearson, P. N., Coxall, H. K., & Rosenthal, Y. (2008). Cooling and ice growth across the Eocene-Oligocene transition. Geology, 36(3), 251–254. https://doi.org/10.1130/G24584A.1 17 Lear, C. H., Rosenthal, Y., Coxall, H. K., & Wilson, P. A. (2004). Late Eocene to early Miocene ice sheet dynamics and the global carbon cycle. Paleoceanography, 19(4). https://doi.org/10.1029/2004PA001039 Leutert, T. J., Auderset, A., Martínez-García, A., Modestou, S., & Meckler, A. N. (2020). Coupled Southern Ocean cooling and Antarctic ice sheet expansion during the middle Miocene. Nature Geoscience, 13(9), 634–639. https://doi.org/10.1038/s41561-020-0623- 0 Levy, R., Harwood, D., Florindo, F., Sangiorgi, F., Tripati, R., Von Eynatten, H., Gasson, E., Kuhn, G., Tripati, A., & DeConto, R. (2016). Antarctic ice sheet sensitivity to atmospheric CO2 variations in the early to mid-Miocene. Proceedings of the National Academy of Sciences, 113(13), 3453–3458. Li, C., Sessions, A. L., Valentine, D. L., & Thiagarajan, N. (2011). D/H variation in terrestrial lipids from Santa Barbara Basin over the past 1400years: A preliminary assessment of paleoclimatic relevance. Organic Geochemistry, 42(1), 15–24. https://doi.org/10.1016/j.orggeochem.2010.09.011 Lopes dos Santos, R. A., Prange, M., Castañeda, I. S., Schefuß, E., Mulitza, S., Schulz, M., Niedermeyer, E. M., Sinninghe Damsté, J. S., & Schouten, S. (2010). Glacial–interglacial variability in Atlantic meridional overturning circulation and thermocline adjustments in the tropical North Atlantic. Earth and Planetary Science Letters, 300(3), 407–414. https://doi.org/10.1016/j.epsl.2010.10.030 Martínez-Garcia, A., Rosell-Melé, A., Jaccard, S. L., Geibert, W., Sigman, D. M., & Haug, G. H. (2011). Southern Ocean dust–climate coupling over the past four million years. Nature, 476(7360), 312–315. https://doi.org/10.1038/nature10310 Meissner, K. J., Bralower, T. J., Alexander, K., Jones, T. D., Sijp, W., & Ward, M. (2014). The Paleocene-Eocene Thermal Maximum: How much carbon is enough? Paleoceanography, 29(10), 946–963. https://doi.org/10.1002/2014PA002650 O’Brien, C. L., Huber, M., Thomas, E., Pagani, M., Super, J. R., Elder, L. E., & Hull, P. M. (2020). The enigma of Oligocene climate and global surface temperature evolution. Proceedings of the National Academy of Sciences, 117(41), 25302. https://doi.org/10.1073/pnas.2003914117 Pancost, R. D., & Boot, C. S. (2004). The palaeoclimatic utility of terrestrial biomarkers in marine sediments. New Approaches in Marine Organic Biogeochemistry: A Tribute to the Life and Science of John I. Hedges, 92(1), 239–261. https://doi.org/10.1016/j.marchem.2004.06.029 Petersen, S. V., & Schrag, D. P. (2015). Antarctic ice growth before and after the Eocene- Oligocene transition: New estimates from clumped isotope paleothermometry: Antarctic ice growth at the E/O transition. Paleoceanography, 30(10), 1305–1317. https://doi.org/10.1002/2014PA002769 18 Plancq, J., Mattioli, E., Pittet, B., Simon, L., & Grossi, V. (2014). Productivity and sea-surface temperature changes recorded during the late Eocene–early Oligocene at DSDP Site 511 (South Atlantic). Palaeogeography, Palaeoclimatology, Palaeoecology, 407, 34–44. https://doi.org/10.1016/j.palaeo.2014.04.016 Pound, M. J., Haywood, A. M., Salzmann, U., & Riding, J. B. (2012). Global vegetation dynamics and latitudinal temperature gradients during the Mid to Late Miocene (15.97– 5.33Ma). Earth-Science Reviews, 112(1), 1–22. https://doi.org/10.1016/j.earscirev.2012.02.005 Rees-Owen, R. L., Gill, F. L., Newton, R. J., Ivanović, R. F., Francis, J. E., Riding, J. B., Vane, C. H., & Lopes dos Santos, R. A. (2018). The last forests on Antarctica: Reconstructing flora and temperature from the Neogene Sirius Group, Transantarctic Mountains. Organic Geochemistry, 118, 4–14. https://doi.org/10.1016/j.orggeochem.2018.01.001 Sangiorgi, F., Bijl, P. K., Passchier, S., Salzmann, U., Schouten, S., McKay, R., Cody, R. D., Pross, J., van de Flierdt, T., Bohaty, S. M., Levy, R., Williams, T., Escutia, C., & Brinkhuis, H. (2018). Southern Ocean warming and Wilkes Land ice sheet retreat during the mid-Miocene. Nature Communications, 9(1), 317. https://doi.org/10.1038/s41467- 017-02609-7 Schouten, S., Hopmans, E. C., Schefuß, E., & Damste, J. S. S. (2002). Distributional variations in marine crenarchaeotal membrane lipids: A new tool for reconstructing ancient sea water temperatures? Earth and Planetary Science Letters, 204(1–2), 265–274. Shevenell, A. E., Kennett, J. P., & Lea, D. W. (2004). Middle Miocene southern ocean cooling and Antarctic cryosphere expansion. Science, 305(5691), 1766–1770. Sikes, E. L., & Volkman, J. K. (1993). Calibration of alkenone unsaturation ratios (Uk’37) for paleotemperature estimation in cold polar waters. Geochimica et Cosmochimica Acta, 57(8), 1883–1889. https://doi.org/10.1016/0016-7037(93)90120-L Sikes, E. L., Volkman, J. K., Robertson, L. G., & Pichon, J. J. (1997). Alkenones and alkenes in surface waters and sediments of the Southern Ocean: Implications for paleotemperature estimation in polar regions. Geochimica Et Cosmochimica Acta, 61(7), 1495–1505. https://doi.org/10.1016/S0016-7037(97)00017-3 Sluijs, A., Bijl, P. K., Schouten, S., Röhl, U., Reichart, G.-J., & Brinkhuis, H. (2011). Southern ocean warming, sea level and hydrological change during the Paleocene-Eocene thermal maximum. Climate of the Past, 7(1), 47–61. https://doi.org/10.5194/cp-7-47-2011 Sophie Warny, C. Madison Kymes, Rosemary Askin, Krzysztof P. Krajewski, & Andrzej Tatur. (2019). Terrestrial and Marine Floral Response to Latest Eocene and Oligocene Events on the Antarctic Peninsula. Palynology, 43(1), 4–21. https://doi.org/10.1080/01916122.2017.1418444 Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., & Feakins, S. J. (2021). Late Eocene Record of Hydrology and Temperature From Prydz Bay, East Antarctica. 19 Paleoceanography and Paleoclimatology, 36(4), e2020PA004204. https://doi.org/10.1029/2020PA004204 Tipple, B. J., Pagani, M., Krishnan, S., Dirghangi, S. S., Galeotti, S., Agnini, C., Giusberti, L., & Rio, D. (2011). Coupled high-resolution marine and terrestrial records of carbon and hydrologic cycles variations during the Paleocene–Eocene Thermal Maximum (PETM). Earth and Planetary Science Letters, 311(1), 82–92. https://doi.org/10.1016/j.epsl.2011.08.045 Villa, G., Fioroni, C., Persico, D., Roberts, A. P., & Florindo, F. (2014). Middle Eocene to Late Oligocene Antarctic glaciation/deglaciation and Southern Ocean productivity. Paleoceanography, 29(3), 223–237. https://doi.org/10.1002/2013PA002518 Warny, S., Askin, R. A., Hannah, M. J., Mohr, B. A. R., Raine, J. I., Harwood, D. M., Florindo, F., & the SMS Science Team. (2009). Palynomorphs from a sediment core reveal a sudden remarkably warm Antarctica during the middle Miocene. Geology, 37(10), 955– 958. https://doi.org/10.1130/G30139A.1 Warny, S. S., & Askin, R. R. (2011). Vegetation and Organic-Walled Phytoplankton at the End of the Antarctic Greenhouse World: Latest Eocene Cooling Events. In Tectonic, Climatic, and Cryospheric Evolution of the Antarctic Peninsula (pp. 193–210). American Geophysical Union (AGU). https://doi.org/10.1029/2010SP000965 Wilson, D. J., Bertram, R. A., Needham, E. F., van de Flierdt, T., Welsh, K. J., McKay, R. M., Mazumder, A., Riesselman, C. R., Jimenez-Espejo, F. J., & Escutia, C. (2018). Ice loss from the East Antarctic Ice Sheet during late Pleistocene interglacials. Nature, 561(7723), 383–386. https://doi.org/10.1038/s41586-018-0501-8 Yamamoto, M., Ficken, K., Baas, M., Bosch, H.-J., & de Leeuw, J. W. (1996). Molecular palaeontology of the earliest Danian at Geulhemmerberg (the Netherlands). Geologie En Mijnbouw, 75(2), 255–267. Zachos, J. C., Quinn, T. M., & Salamy, K. A. (1996). High‐resolution (104 years) deep‐sea foraminiferal stable isotope records of the Eocene‐Oligocene climate transition. Paleoceanography, 11(3), 251–266. Zachos, J. C., Shackleton, N. J., Revenaugh, J. S., Pälike, H., & Flower, B. P. (2001). Climate Response to Orbital Forcing Across the Oligocene-Miocene Boundary. Science, 292(5515), 274. https://doi.org/10.1126/science.1058288 Zachos, J. C., Stott, L. D., & Lohmann, K. C. (1994). Evolution of Early Cenozoic marine temperatures. Paleoceanography, 9(2), 353–387. https://doi.org/10.1029/93PA03266 20 Chapter 2 Snapshots of pre-glacial paleoenvironmental conditions along the Sabrina Coast, East Antarctica: New palynological and biomarker evidence This chapter was published in 2021 as: Duffy, M., Tibbett, E. J., Smith, C., Warny, S., Feakins, S. J., Escarguel, G., Askin, R., Leventer, A., & Shevenell, A. E. (2021). Snapshots of pre-glacial paleoenvironmental conditions along the Sabrina Coast, East Antarctica: New palynological and biomarker evidence. Geobios. https://doi.org/10.1016/j.geobios.2021.09.001 Abstract The Aurora Subglacial Basin (ASB) catchment contains 3–5 m of sea-level equivalent ice volume that drains to the Sabrina Coast, East Antarctica via the Totten Glacier system. Observed thinning and retreat of Totten Glacier indicate regional sensitivity to oceanographic and atmospheric warming. Paleoclimate studies of climatically sensitive catchments are required to understand the evolution of the East Antarctic Ice Sheet (EAIS) and its outlet glacier systems. Recent seismic and sediment studies from the Sabrina Coast document the evolution of the EAIS in the ASB catchment, suggesting that the region has long been sensitive to climatic changes. This study presents new palynological and biomarker data from Sabrina Coast continental shelf sediments. Detailed palynological records were obtained from four short jumbo piston cores (JPC; NBP14-02 JPC-30, -31, -54 and -55), enabling reconstructions of regional vegetation and environments prior to and during Cenozoic EAIS development. The Sabrina Flora is dominated 21 by angiosperms, with Gambierina spp. often exceeding 40% of the assemblage, and diverse Proteaceae, Battenipollis spp., Forcipites spp., Nothofagidites spp., fern, and conifer palynomorphs indicative of an open shrubby ecosystem. Excellent preservation and frequent occurrence of Gambierina spp. clusters suggest that a majority of the Sabrina Flora assemblage is penecontemporaneous with sedimentation; however, some uncertainties remain whether this sedimentation occurred in the Late Cretaceous or the Paleogene. Despite that uncertainty, high abundances of Gambierina spp. and Battenipollis spp., in combination with relatively low (<10%) Nothofagidites spp. abundances indicate that the Sabrina Flora is unique in Antarctica. Evaluation of biomarkers finds evidence for penecontemporaneous and reworked components. The penecontemporaneous C30 n-alkanoic acids have ẟ 13 C values of −30.2 ± 0.5‰, consistent with ẟ 13 C values in an open canopy woodland or shrubby open vegetation. Their hydrogen isotope (ẟD) values of −215 ± 4.5‰, indicate precipitation isotopic composition (ẟDprecip) of −130‰, similar to coastal snow in the same region today. Together, Sabrina Flora palynomorph and plant wax data suggest a drier, more open coastal vegetation in the Aurora Basin of East Antarctica rather than the closed rainforest vegetation often described from other parts of Antarctica for the Cretaceous to Paleogene. To directly compare records from the circum- Antarctic, additional long sedimentary records with improved biostratigraphic constraints are required. Such records will enable identification of regional climate gradients or micro-climates, and allow assessment of the environmental conditions and mechanisms driving observed differences. 2.1 Introduction The East Antarctic Ice Sheet (EAIS), which contains 53 m of sea level equivalent ice, is one of the largest potential contributors to modern sea level rise (Fretwell et al., 2013, DeConto and 22 Pollard, 2016, Morlighem et al., 2020). The Aurora Subglacial Basin (ASB) (Fig. 2.1) contains 3.9 m of sea level equivalent ice that is drained by the Totten Glacier and its tributaries that terminate at the Sabrina Coast, East Antarctica (Young et al., 2011, Wright et al., 2012, Greenbaum et al., 2015, Morlighem et al., 2020). The Sabrina Coast (115° to 121°E, 67°S) is located on the Wilkes Land continental margin (Young et al., 2011), which formed as Antarctica rifted from Australia in the mid-Cretaceous (Cande and Mutter, 1982, Escutia et al., 2011). As a major drainage outlet of the ASB, sediments deposited on the Sabrina Coast shelf likely contain historical records of glacial evolution in the ASB (Gulick et al., 2017, Montelli et al., 2020). Geologic and oxygen isotope data from deep-sea benthic foraminifera indicate continental-scale ice sheets were present in East Antarctica by at least the early Oligocene (Kennett, 1977, Coxall et al., 2005, Francis et al., 2009); however, the history of the EAIS prior to this time remains poorly understood. Recent airborne and marine geophysical and geological studies of the ASB and the sediments from the Sabrina Coast continental shelf document the Cenozoic evolution of the EAIS in the ASB, revealing that regional outlet glaciers expanded progressively seaward and became marine terminating as early as the middle Eocene (Aitken et al., 2016, Gulick et al., 2017, Montelli et al., 2020, Wright et al., 2012, Young et al., 2011). Large volumes of sediment eroded from the ASB were transported to the Sabrina Coast by glacio-fluvial systems, resulting in continental shelf progradation in the early Paleogene (Gulick et al., 2017). Paired marine seismic and sedimentologic studies of short jumbo piston cores from the Sabrina Coast shelf indicate that regional glaciers advanced and retreated across the shelf at least 11 times in the Oligocene and Miocene, suggesting the EAIS was not as stable as previously thought (Gulick et al., 2017, Montelli et al., 2020). 23 Here we present new pollen assemblage and biomarker records from short sediment cores recovered from the Sabrina Coast shelf that provide insights into the Late Cretaceous to Paleogene paleoclimate and vegetation history of the ASB catchment. Our data increase the resolution of previously published palynological records (Gulick et al., 2017, Smith et al., 2019). We also add new biomarker data that suggest an environmental significance for several palynomorphs with unknown botanical affinities and provide insight to past East Antarctic climate. Additionally, late Miocene to early Pliocene sediments provide insight to sedimentary reworking processes along the Sabrina Coast. Figure 2.1. Paleogeographic reconstruction for the Australian-Antarctic margins at 50 Ma. Aurora Subglacial Basin and Eucla Basin are outlined in blue. Study location indicated by star. 24 2.2 Material and Methods 2.2.1 Material This study is based upon four short (<2 m) jumbo piston cores (JPC-30, -31, -54, and -55 from cruise NBP14-02; Gulick et al., 2017; Fig. 2.2). Seismic imaging of the Sabrina Coast continental shelf identified three distinct sedimentary packages, termed Megasequences I-III (MS-I, MS-II, and MS-III; Gulick et al., 2017). MS-I, which overlies basement, contains 620 m of seaward-dipping, low-amplitude discontinuous reflectors with at least two clinoforms indicating periods of high sediment flux. An undulating eroded surface separates MS-I from MS- II and is interpreted to reflect evidence for the first grounded ice on the continental shelf (Gulick et al., 2017, Montelli et al., 2020). MS-II is 675 m thick and contains at least 11 erosive surfaces within a sequence of laminated to acoustically transparent sediments indicative of ice proximal to open marine sediments (Gulick et al., 2017, Montelli et al., 2020). These erosive surfaces indicate a minimum of 11 glacial advances and retreats across the middle shelf during the deposition of MS-II (Gulick et al., 2017, Montelli et al., 2020). A regional unconformity separates MS-II from MS-III, with evidence of significant glacial erosion into MS-II. Glacial erosion of the seafloor allowed Gulick et al. (2017) to sample exposed sediment from MS-I (NBP14-02 JPC-54 and -55) and through the regional unconformity (NBP14-02 JPC-30 and - 31). 25 Figure 2.2. Bathymetry of Sabrina Coast continental shelf, mapped during NBP 14-02 using a Kongsberg EM 120 multibeam system. Locations of JPC-30, 31, 54 and 55 are denoted by blue circles. Black line represents seismic line 17. Modified from Smith et al. (2018). As described by Gulick et al. (2017), JPC-54 (66.28°S, 120.67°E; water depth: 442 m) and JPC- 55 (66.35°S, 120.51°E; water depth: 520 m) are 121 cm and 169 cm in length, respectively (Fig. 2.2, Fig. 2.3). Core sites were chosen based on high-resolution seismic data from NBP14-02 seismic line 17, which revealed pre-glacial strata outcropping at the seafloor. JPC-54 and JPC-55 were collected from MS-I, above and below a clinoform, respectively. Each core consists of two units separated by a sharp contact. In both cores, Unit I (20 cm in JPC-54 and 40 cm in JPC-55) consists of late Quaternary diatom-rich mud, which overlies a partially consolidated sandy interval (Unit II). In JPC-54, Unit II (101 cm) consists of structureless diamict to silty coarse sands with centimeter-scale angular lonestones, while in JPC-55 Unit II (129 cm) consists of 26 more consolidated black micaceous silty sands with organic detritus, macro- and microfossils, and rare pyrite nodules (Gulick et al., 2017). Core lithologies are shown in Fig. 2.3. Figure 2.3. Core photos, core x-rays, lithologies and sample distributions for JPC-54, 55, 30 and 31. 27 Thirteen samples collected from JPC-54 and JPC-55 for preliminary palynological analysis (Gulick et al., 2017, Smith et al., 2019) contributed biostratigraphic information used in conjunction with foraminiferal samples to provide initial age-control for the Unit II sediments in each jumbo piston core. Smith et al. (2019) documented an unexpectedly abundant, diverse, and well-preserved terrestrial palynomorph assemblage, named the Sabrina Flora. This flora, dominated by two new pollen species, has the potential to contribute to limited existing knowledge of the paleobotanical history of East Antarctica. Preliminary analysis showed that the Sabrina Flora is dominated by angiosperms, with Gambierina (G.) rudata and G. edwardsii complexes often exceeding 40% of the assemblage (Smith et al., 2019). Additionally, diverse Proteaceae, Battenipollis sectilis, Forcipites spp., Nothofagidites (N.) spp., fern and conifer palynomorphs contribute to the assemblage. Smith et al. (2019) also described two new species: Battenipollis sabrinae and Gambierina askiniae. Pristine pollen preservation, the frequent occurrence of Gambierina spp. clusters, and the geomorphology of the ASB and Sabrina shelf, led Smith et al. (2019) to interpret a majority of the Sabrina Flora assemblage as penecontemporaneous with sedimentation. 2.2.2 Age Controls: a review of uncertainties and hard facts 2.2.2.1. NBP14-02 JPC -54 and -55 To constrain the age of Unit II in each short sediment core collected from the Sabrina Shelf, Gulick et al., 2017, Smith et al., 2019 used published age ranges of pollen when available and data on foraminifer species observed in NBP14-02 JPC-54 and -55. Due to the paucity of palynostratigraphic data from the East Antarctic margin, they incorporated age constraints from a limited number of key species from southern Australian and New Zealand (e.g., Stover and Partridge, 1973, Jarzen and Dettmann, 1992, Partridge, 2006), the Antarctic Peninsula (e.g., 28 Francis et al., 2009, Warny and Askin, 2011, Warny et al., 2019), and McMurdo Sound erratics (e.g., Askin, 2000, Francis, 2000, Levy and Harwood, 2000). A detailed discussion of the methods used to establish preliminary age control in JPC-54 Unit II and JPC-55 Unit II can be found in Gulick et al., 2017, Smith et al., 2019. In these publications, JPC-54 was determined to have an age of early to middle Eocene, based on pollen biostratigraphy, while JPC-55 was determined to be of latest Paleocene age, based on a combination of pollen and foraminiferal biostratigraphy. The presence of benthic foraminifera in JPC-55 indicate that sediment was deposited in a marine shelf environment and pollen clusters suggest that the terrestrial-derived sedimentary component was not likely transported over significant distances or substantially reworked (Gulick et al., 2017, Smith et al., 2019). The overlap in foraminifer species ranges indicates a Paleocene age for JPC-55 Unit II (Gulick et al., 2017). The Paleocene age of JPC-55 was further refined to late Paleocene using the presence of Microalatidites paleogenicus, which has a first Antarctic occurrence in the early Paleocene (Truswell and Macphail, 2009, Raine et al., 2011) and the first occurrences of Nothofagidites lachlaniae, Proteacidites tenuiexinus, and N. flemingii-rocaensis complex in the late Paleocene (Stover and Partridge, 1973; Stover and Evans, 1973). Recently an alternative interpretation of the Sabrina Flora (Macphail, 2021) suggested a Late Cretaceous age, based on comparison to pollen from the Great Australian Bight sub-basins (Dettmann and Jarzen, 1990). Macphail (2021) also cited unpublished evidence for abundant pollen identical to Battenipollis sabrinae and Gambierina askiniae in Campanian to Maastrichtian sediments recovered from the Gippsland Basin, Australia, and reinterpreted the Sabrina Flora of Smith et al. (2019) to be also Late Cretaceous in age. While we acknowledge the possibility that the Sabrina Flora is of Late Cretaceous age and reworked (Macphail, 2021), 29 we argue that it is more plausible that Unit II in JPC-55 is of Paleocene age, when considering the totality of the published pollen and foraminifer data, the seismic facies, and the stratigraphic context of the Sabrina shelf sedimentary sequence (Gulick et al., 2017, Smith et al., 2019 Montelli et al., 2020). Further, because JPC-54 was recovered ∼30 m above JPC-55, stratigraphically, we can state with high confidence that JPC-54 sediments are younger than those in JPC-55, but older than the unconformity interpreted to reflect the first expansion of glacial ice across the Sabrina Coast shelf (Gulick et al., 2017, Smith et al., 2019 Montelli et al., 2020). This said, because the two studied sequences are <2 m in length and the extent of reworking cannot definitively be assessed, we will not exclude the Late Cretaceous as a possible source of the floral organic matter recovered. But until longer sedimentary sequences are drilled on the Sabrina Coast shelf, we accept the published Paleocene age for JPC-55 and the larger early to middle Eocene age range for JPC-54 (Gulick et al., 2017, Smith et al., 2019). 2.2.2.2 NBP14-02 JPC-30 and -31 To assess sedimentary reworking in Sabrina Coast sediments, we analyzed cores JPC-30 (66.45°S, 120.33°E; water depth: 548 m) and JPC-31 (66.45°S, 120.34°E, water depth: 534 m), which recovered the MS-II/MS-III contact and are late Miocene to early Pliocene in age, as indicated by well-defined diatom biostratigraphy (Gulick et al., 2017; Fig. 2.2, Fig. 2.3). As such, MS-II erosive surfaces, interpreted to reflect glacial advances and retreats, were likely formed during the Oligocene and Miocene (Gulick et al., 2017). Current knowledge of Antarctic plant evolution indicates that vegetation essentially disappeared from the Antarctic Peninsula at 12.8 Ma (Anderson et al., 2011) and from the Dry Valleys at ca. 13.85 Ma (Lewis et al., 2008, Rau, 2017). In the ASB, it is likely that palynomorph production ceased following the large-scale 30 glaciation evidenced by the regional unconformity observed in the Sabrina Coast seismic profiles (Gulick et al., 2017). We suggest that all palynomorphs observed in JPC-30 and -31 are reworked and do not provide biostratigraphic information. However, analysis of reworked palynomorphs is useful in understanding sediment transport to the Sabrina Coast and provides insight to the type of sediments eroded by ice advance (e.g., Baudoin, 2018, Coenen et al., 2019). 2.2.3 Palynology To quantify absolute abundance of terrestrial palynomorphs and discern palynomorph assemblages from JPC-30, -31, -54 and -55, six samples from JPC-30, five samples from JPC- 31, 23 samples from JPC-54 and 34 samples from JPC-55 (Fig. 3) were collected and processed for terrestrial palynomorphs at Global Geolab Limited (Alberta, Canada). For each sample preparation, ∼5 g of dried sediment was spiked with a known quantity of Lycopodium spores to allow for the quantitative assessment of terrestrial palynomorph concentrations. Acid soluble minerals (carbonates and silicates) were digested in HCl and HF, followed by a controlled oxidation of the residue. Rinsing to neutrality with DI water was performed between each steps. Residues were concentrated by filtration on a 10 μm mesh sieve. The 68 samples from JPC-30, -31, -54, and -55 were examined using an Olympus BX41 microscope with ×60 and ×100 oil immersion lenses. A minimum of 300 terrestrial palynomorphs were counted per sample, using a snaking transect method. In low abundance samples, the entire slide was counted; Lycopodium spores were also counted. A database of all palynomorphs recovered was prepared and key species documented photographically using a Q- Color 5 Olympus camera system with Q-Capture (v. 3.1.1) software. Taxonomic evaluation of palynomorphs utilized established literature (e.g., Cookson, 1950, Cookson and Pike, 1954, Couper, 1960, Stover and Partridge, 1973, Truswell, 1983, Jarzen and Dettmann, 1992, Macphail 31 and Truswell, 2004, Hou et al., 2006, Truswell and Macphail, 2009, Raine et al., 2011, Pross et al., 2012, Contreras et al., 2013, Smith et al., 2019) and collections curated at the Louisiana State University Center for Excellence in Palynology (CENEX). Shannon Diversity Index (H) was calculated as = − ∑ 𝑝𝑝𝑝𝑝 • ln (𝑝𝑝𝑝𝑝 ) where pi is the proportion of individuals belonging to the ith species in the dataset. 2.2.4 Biomarkers Eleven samples from JPC-54 and -55 were freeze-dried and 4–21 g of sediment was homogenized for extraction via an Accelerated Solvent Extraction (ASE 350, Dionex) using a 9:1 Dichloromethane (DCM) to methanol (MeOH) mixture (v/v). Resulting total lipid extracts were separated into neutral and acid fractions via column chromatography through LC-NH2 gel (columns were 5 cm × 40 mm Pasteur pipette with NH2 sepra bulk packing) using 2:1 DCM:isopropanol and 4% formic acid in diethyl ether to separate out the neutral and acid fractions, respectively. The neutral fraction was separated via column chromatography using 5% deactivated silica gel, eluting the n-alkanes in the hexane fraction, and the polar biomarkers with DCM and MeOH. The acid fraction, containing the n-alkanoic acids, were methylated using 95% MeOH of a known isotopic composition with 5% hydrochloric acid for 12 hours at 70 °C. The fatty acid methyl esters (FAMEs) were separated using 2:1 hexane:MilliQ water (v/v) and dried by passing through anhydrous sodium sulfate, and further purified over silica gel column chromatography eluting first with hexane to exclude any further impurities and then with DCM to recover the FAMEs. Both n-alkanes and FAMEs fractions were analyzed by gas chromatography mass spectrometry (GC–MS) for compound identification and the n-alkanes and n-alkanoic acids as methyl esters were quantified via Flame Ionization Detection, in comparison to in-house alkanes and FAME standards, respectively. We quantified the abundance of 32 individual n-alkanes and n-alkanoic acids and calculated the average chain length (ACL) and carbon preference index (CPI) as follows: ACL: ∑(𝑛𝑛 ∗ [ 𝐶𝐶 𝑛𝑛 ]) ∑ ⌊ 𝐶𝐶 𝑛𝑛 ⌋ � CPI: ∑ 𝐶𝐶 𝑛𝑛 (∑ 𝐶𝐶 𝑛𝑛 − 1 ) � where n = 24–30 for n-alkanoic acids and 23–31 for the n-alkanes. The CPI is calculated as the even over odd preference for n-alkanoic acids, the odd over even preference for alkanes, and the even over odd preference for n-alkanes. Within the alkanes fraction, we measured the relative abundances of hopanes, biomarkers that derive from membrane lipids in bacteria and that undergo isomerization with increasing thermal maturity (Inglis et al., 2020). We identified these compounds using their diagnostic mass fragments and comparison to published spectra (Inglis et al., 2018, Sessions et al., 2013, Uemura and Ishiwatari, 1995). Compound specific isotope analysis was performed on the FAMEs fraction using a gas chromatography isotopic ratio mass spectrometer (GC-IRMS) using a Thermo Scientific Trace gas chromatograph connected to a Delta V Plus mass spectrometer via an Isolink combustion furnace at 1000°C for ẟ 13 C and a pyrolysis furnace at 1400 °C for ẟD. The peak amplitude was 1–7 V. The ẟ 13 C linearity was recorded each day and had an average standard deviation of 0.042‰. H3 factor was recorded every day with an average value of 9.518 ± 0.377 ppm mV −1 . Samples were normalized to Vienna Pee Dee Belemnite (VPDB) and Vienna Standard Mean Ocean Water/Standard Light Antarctic Precipitation (VSMOW/SLAP) by an external standard mixture of 16 n-alkanes with ẟ 13 C values that range from −25.9 to −33.7‰ and ẟD values from −17 to −256‰ (A6 mix obtained from A. Schimmelmann, Indiana University). Corrections were 33 made for the methyl group added during methylation for the n-alkanoic acids (ẟ 13 C of −24.7 ± 0.2‰ and ẟD of −186.9 ± 3.7‰) by mass balance to calculate the isotopic composition of the corresponding acid. 2.2.5 Statistical analysis Statistical analyses on palynomorph abundance and assemblage data were conducted using PAST v. 2.17c freeware (Hammer et al., 2001). Stratigraphically-unconstrained and stratigraphically-constrained analyses using the Bray-Curtis similarity index and Correspondence Analysis (Legendre and Legendre, 2012) were used to examine changes in palynomorph assemblages between and among the four cores. Analysis of Similarity (ANOSIM; Clarke, 1993) also based on the Bray–Curtis similarity matrix was done to identify sample clusters with different taxonomical compositions, followed by Similarity Percentage analysis (SIMPER; Clarke, 1993) in order to determine which taxa control the significant differences among clusters. 2.3 Results 2.3.1 Palynology The samples from NBP14-02 JPC-54 and -55 (Unit II) reveal a Sabrina Flora with abundant, well-preserved palynomorphs in high abundances that range from ∼5,000 to 19,000 palynomorphs per gram of dried sediment, with a minimum of 62 species present. The assemblages are dominated by angiosperms, with Gambierina (G.) spp. often exceeding 40% of the assemblage. Diverse Proteaceae, Battenipollis (B.) spp., Forcipites spp., Nothofagidites spp., fern and conifer palynomorphs are also notable in the JPC-54 and -55 assemblages. Clusters of Gambierina spp., Battenipollis spp., Proteacidites spp., Forcipites sp., and Nothofagidites sp. are 34 also observed throughout Unit II in both JPC-54 and -55. Low abundances of darker, reworked Cretaceous and Permian palynomorphs are observed, including Cicatricosisporites ludbrooki, Granulatisporites sp., and Leiotriletes directus. A UPGMA cluster analysis on a Bray-Curtis similarity matrix with no stratigraphic constraint identified four statistically significant clusters (G1, G2, G3, G4) in the complete dataset from all four cores (Fig. 2.4): • cluster G1 includes all samples from JPC-30 and -31; • cluster G2 includes all JPC-54 Unit II samples; • cluster G3 includes all Unit I samples from JPC-55 and one sample from JPC-54; • cluster G4 includes all JPC-55 Unit II samples; • the two uppermost (Unit I) samples from JPC-54 do not fall within a statistically significant group. 35 Figure 2.4. UPGMA cluster analysis on a Bray-Curtis similarity matrix with no stratigraphic constraint. Y-axis shows similarity between samples. Samples are indicated by JPC#-Unit#-Depth (cm) with JPC-54 shown in blue, JPC-55 in red, JPC-30 in green and JPC-31 in brown. 36 Table 2.1 Four-group SIMPER analysis with Bray-Curtis similarity matrix conducted for groups G1, G2, G3, and G4. “Taxa A” taxa accounts for 50.5% of compositional variability between the four groups. “Taxa B” and “Taxa C” taxa together explains 30% of compositional variability. These three groups together account for 80.5% of compositional differences between the four groups. 37 Eighteen of 62 identified taxa explain >80% of the overall compositional differences between the four clusters; all other taxa individually explain <1.2% and together explain <20% of the overall compositional differences between the clusters (Table 2.1). In core NBP14-02 JPC-55, palynomorph abundance averages 362 (Unit I) and 10,212 (Unit II) palynomorphs per gram of dried sediment. Diversity in Unit I is low (H = 1.7) and significantly higher in Unit II (H = 2.8; Fig. 2.5). For all JPC-55 samples, a stratigraphically-constrained UPGMA cluster analysis on a Bray-Curtis similarity matrix reveals three distinct groups (55-1, 55-2, and 55-3) which are statistically different from one another (Fig. 2.6). Group 55-1 contains all samples from Unit I, where the assemblage is dominated by G. rudata, with B. sectilis subdominant. The sample at 49 cm depth, close to the contact between Units I and II, did not fall in any of the three groups. Groups 55-2 and 55-3 contain all other samples from lithologic Unit II. In this unit, G. askinae and B. sabrinae are co-dominant, with notable contributions from Proteacidites spp. (especially P. parvus), Forcipites spp., B. sectilis, Nothofagidites spp., and gymnosperm pollen, including Podocarpidites spp. 38 Figure 2.5. Palynomorph absolute and relative abundances, Shannon Diversity Index and assemblage data from JPC-55 (MS-III, older core). Relative abundances of G. rudata, G.askinae, B. sectilis, B. sabrinae, Forcipites spp., Proteacidites spp., Nothofagidites spp., L. ovatus and Gymnosperms are shown. Unit I (orange shading) consists of late Pleistoceneto Holocene glaciomarine sandy silt with low palynomorph abundances. Unit II (yellow shading) consists of organic-rich silty sand containing the diverse Sabrina Floraassemblage. Note that Unit I relative abundance curves are not a reflection of vegetation change, but are proportions of a very sparse, reworked assemblage. 39 Figure 2.6. UPGMA cluster analysis of a Bray-Curtis similarity matrix with stratigraphic constraint for all samples in JPC-55. Y-axis shows similarity level between samples. Sample depth is shown horizontally. Group 55-1 contains all samples from JPC-55, Unit I. Groups 55-2 and 55-3 contain samples from JPC-55, Unit II. Groups 55-1, 55-2 and 55-3 are all statistically different (ANOSIM test), indicating a clear distinction in palynomorph assemblage between Unit I and II, as well as a change in assemblage within Unit II. 40 Figure 2.7. Palynomorph abundance, Shannon Diversity Index and assemblage data from JPC-54 (MS-III, younger core). Relative abundances of G. rudata, G. askinae, B. sectilis, B sabrinae, Forcipites spp., Proteacidites spp. Nothofagidites spp., L. ovatus and gymnosperms are shown. Unit I (orange shading) consists of late Pleistocene to Holocene glaciomarine sandy silt with low palynomorph abundances. Unit II (yellow shading) consists of organic-rich silty sand containing the diverse Sabrina Flora assemblage. Note that Unit I relative abundance curves are not a reflection of vegetation change, but are proportions of a very sparse, reworked assemblage. 41 Figure 2.8. UPGMA cluster analysis on a Bray-Curtis similarity matrix with stratigraphic constraint for all samples in JPC-54. Y-axis shows similarity between samples. Sample depth is shown on X-axis. Group 54-1 contains only samples from JPC-54, Unit I. Groups 54-2 and 54-3 contain all samples from JPC-54, Unit II. Group 54-1 is statistically different from groups 54-2 and 54-3, but groups 54-2 and 54-3 are not statistically different from each other (ANOSIM test). This indicates a clear distinction between the palynomorph assemblages of Unit I and Unit II. 42 Figure 2.9. Palynomorph abundance, Shannon Diversity Index and relative abundance of dominant species for JPC-30. Figure 2.10. Palynomorph abundance, Shannon Diversity Index and relative abundance of dominant species for JPC-31. 43 Figure 2.11. Correspondence analysis of JPC-30, JPC-31, JPC-54, and JPC-55 assemblages, using only the 18 major taxa driving the between-group variability (Table 1). Factorial Axis 1 (red) and Factorial Axis 2 (green) account for 50.7% and 21.7% of the overall between-sample variability, respectively. The sparse, reworked assemblages of JPC-30, JPC31, JPC-54 Unit I and JPC-55 Unit I group together while the in situ assemblages of JPC-54 Unit II and JPC-55 Unit II both plot separately. 44 Figure 2.12. Factorial space resulting from correspondence analysis shown in Figure 3.11. Factorial Axis 1 (red) shows 50.7% of the overall between-sample variation; it is related to changes in abundance of B. sabrinae and G. askinae vs B. sectilis and G. rudata. Factorial Axis 2 (green) shows 21.7% of the overall between-sample variation; it is related to changes in abundance of B. sectilis, Forcipites stipulatus and Proteacidites parvus vs L. ovatus, N. lachlaniae, Liliacidites spp., Proteacidites spp., Arecipites spp., and trilete cryptogram spores. 45 In core NBP14-02 JPC-54, palynomorph abundance averages 441 (Unit I) and 8,287 (Unit II) palynomorphs per gram of dried sediment. As in JPC-55, JPC-54 Unit I diversity is low (H = 1.8) and is significantly greater in Unit II (H = 2.8) (Fig. 2.7). For all JPC-54 samples, a stratigraphically-constrained UPGMA cluster analysis on a Bray-Curtis similarity matrix reveals three groups (54-1, 54-2, and 54-3). Group 54-1 is statistically distinct from groups 54-2 and 54- 3 but groups 54-2 and 54-3 are not statistically different from each other (Fig. 2.8). Samples from 17 cm (close to the Unit I/II boundary) and 68 cm depth did not fit into any of the three groups. Group 54-1 contains samples from lithologic Unit I, whose assemblage is dominated by G. rudata, with B. sectilis subdominant. Groups 54-2 and 54-3 contain samples from lithologic Unit II, where G. rudata is dominant, with notable contributions from Nothofagidites spp. (especially N. lachlaniae), Proteacidites spp., B. sectilis, Forcipites spp., and gymnosperm pollen including Podocarpidites spp. The cores NBP14-02 JPC-30 and -31 both contain low abundances of palynomorphs, with concentrations averaging 49 and 44 palynomorphs per gram of dried sediment, respectively (Fig. 2.9, Fig. 2.10). The palynomorph assemblages are somewhat low in diversity (H = 2.2 and 2.3, respectively) and are statistically indistinguishable from one another (Fig. 2.4), with G. rudata dominant, and Cyathidites minor and B. sectilis sub-dominant. Because the significant compositional differences between groups in all four sediment cores were driven by 18 main taxa (Table 2.1), a Correspondence Analysis was performed on only those taxa. This analysis shows groups G1, 55-1 and 54-1 clearly separated from 55-2, 55-3, 54- 2, and 54-3 (Fig. 2.11). The resulting factorial space (two first factorial axes) shows the driving taxa of this sample distribution along each factorial axis (Fig. 2.12). Factorial Axis 1 (50.7% of the overall between-sample variation) is driven by changes in abundance of B. sabrinae and G. 46 askinae vs. B. sectilis and G. rudata. Groups G1, 54-1 and 55-1 are characterized by high abundances of B. sectilis and G. rudata relative to B. sabrinae and G. askinae. Groups 54-2 and 54-3 are characterized by moderate to low relative abundances of B. sectilis and G. rudata. Groups 55-2 and 55-3 are characterized by very low relative abundances of B. sectilis and G. rudata relative to B. sabrinae and G. askinae. Factorial Axis 2 (21.7% of the overall between- sample variation) is driven by changes in abundance of B. sectilis, Forcipites stipulatus and Proteacidites parvus vs. Laevigatosporites ovatus, N. lachlaniae, Liliacidites spp., Proteacidites spp., Arecipites spp., and trilete cryptogam spores. 2.3.2 Biomarkers Both n-alkane and n-alkanoic acid fractions were evaluated in Unit II of cores JPC-54 and JPC- 55. All samples contained n-alkanes and hopanes, except for in sample JPC-55-147 cm, where concentrations were below detection limits. We identified C17 to C31 n-alkanes above an uncharacterized complex mixture (UCM) in all samples. The concentration of detected (C17 to C31) n-alkanes ranged from 128 to 440 ng/gdw with long chain n-alkanes, likely derived from plants (C23-C31), ranging between 71 to 265 ng/gdw (Fig. 2.13(B)). ACL was in the range of 26 to 27 (Fig. 2.13(C)). The CPI was low, averaging 2.2 and 1.8 for JPC-54 and -55, respectively (Fig. 2.13(D)), consistent with evidence for maturity from the UCM. Within the alkanes fraction, we also identified C27 to C31 hopanes and their isomers. We report the hopane index ββ/(αβ + βα + ββ) for the available isomers of C29 hopanes only, C31 hopanes only, and due to variations in the ability to detect C29 and C31 stereoisomers across the two cores, we computed the index across all available C27 to C31 hopane isomers (Fig. 2.13(A)), similar to Duncan et al. (2019). Applying the hopane index across all long chain hopanoids yields average values of 0.62 and 0.41 for JPC-54 and -55, respectively, suggesting generally low to moderate thermal maturity; a 47 single low value (0.06) at 52 cm in JPC-55 suggests the presence of high thermal maturity carbon. However, variability between samples suggests variable inputs of reworked components. The n-alkanoic acids were more abundant than the n-alkanes. We found C16 to C30 n-alkanoic acid homologues with an even over odd chain length distribution, indicating penecontemporary inputs, with the shorter chains likely from aquatic production and the longer chain homologues most likely to be exclusively derived from plant wax. The long chain n-alkanoic acids (C24-C30) had concentrations ranging from 313 to 3,330 ng/gdw (Fig. 2.13(B)) with an average concentration of 1,929 ng/gdw and 542 ng/gdw for JPC-54 and -55, respectively. The C30 n- alkanoic acids are present with concentrations ranging from 78 to 1,172 ng/gdw. The ACL averaged 26.4 and 26.2 (Fig. 2.13(C)), with a CPI average of 8.2 and 4.9 for JPC-54 and -55, respectively (Fig. 2.13(D)); these distributions are indicative of penecontemporary plant wax inputs. Given their abundance and penecontemporary interpretation, the n-alkanoic acids were analyzed for compound specific carbon and hydrogen isotopic compositions. 48 Figure 2.13. Compilation of biomarker data for Core JPC 54 and 55. A) Hopane indices computed for the C 29 only, C 31 only, and all hopanoids (C 27 to C 31 ), showing the direction of greater maturity. B) Total plant wax concentrations for both n-alkanes (orange) and n-alkanoic acids (green), C) average chain length (ACL) of n-alkane and n-alkanoic acids, D) carbon preference index (CPI) for n-alkane and n- alkanoic acids, where lower values indicate maturity and higher values indicate fresher or penecontemporary inputs E) δD values and F) δ 13 C values for the long chain n-alkanoic acids, with putative marine and terrestrial plant contributions denoted. In cores JPC-54 and -55, the ẟ 13 C values for the C24, C26 and C28 n-alkanoic acids range from -29 to -25‰, which suggests a shifting proportion of marine or microbial sources across the homologous series. C30 and C32 have similar ẟ 13 C values, which are offset from C24, C26 and C28 n-alkanoic acid ẟ 13 C values, supporting the use of C30 and C32 as terrestrial plant biomarkers (Fig. 2.13(F)). The longer chain C30 and C32 n-alkanoic acid compounds are thus suitable for reconstructions of the hydrogen isotopic composition of precipitation as recorded by terrestrial plants. Downcore ẟ 13 C30 values in JPC-54 range from −31.2 to −29.9‰ with an average value of −30.2 ± 0.5‰ (Fig. 2.13). The ẟ 13 C30 was not measured in JPC-55 due to low C30-acid abundance and prioritization of hydrogen isotopic analyses. The C30 n-alkanoic acid has a mean ẟD value of −216 ± 5‰, and ranges between −222 and −207‰ across both JPC-54 and -55 (Fig. 2.13(E)). A net fractionation of −100‰ was the 49 approximation used for open vegetation environments in other Antarctic margin settings (Feakins et al., 2012, Feakins et al., 2014). More recently a suite of studies of n-alkanoic acids in high latitude boreal ecosystems suggested fractionations for the C28 n-alkanoic acid in Siberia of −107 ± 12‰ (Wilkie et al., 2013) and −95 ± 11‰ in the boreal forest ecosystems of the Yukon, Alaska and Northwest Territories (Bakkelund et al., 2018). Greenland surveys suggest consistency of high latitude plants net fractionations with a global mean estimate of −99‰ (McFarlin et al., 2019), hence all estimates are equivalent within uncertainties. Given the lack of calibration of austral high latitude ecosystems, uncertainty in the appropriate value is unavoidable and unknown. Therefore, we use the approximate net fractionation of −100‰, for comparability with prior Antarctic paleoclimate reconstructions and consistent with the boreal calibration ranges. Uncertainty in the net fractionation is estimated to be on the order of 10‰ based on boreal calibrations. This indicates ẟDprecip values range from −136 to −118‰, with a mean ẟDprecip of −129 ± 6‰ across both JPC-54 and -55. 2.4 Discussion 2.4.1 Palynomorph and biomarker provenance and regional hydroclimate Seismic data and benthic foraminifers indicate that siliciclastic sediments with terrestrial organic debris recovered in JPC-55 were deposited in a proximal marine shelf setting seaward of a low- lying fluvial and/or glaciofluvial environment (Young et al., 2011, Wright et al., 2012, Aitken et al., 2016, Gulick et al., 2017; Montelli et al., 2020). Although longer records are required to definitively assess reworking, we suggest that a majority of the Sabrina Flora assemblage was likely locally sourced from the low-lying coastal plain and delivered to the shelf via low-energy fluvial outflow during time intervals of relatively high sea level (e.g., the late Paleocene). Support for this interpretation comes from the following lines of evidence: 50 • pristine preservation of palynomorphs in Unit II sediments in both JPC-54 and −55, which indicates an assemblage that has not been vigorously reworked; • values obtained for the hopane index across all long chain hopanoids and the light color of the grains, both indicating low thermal maturity in contrast to the species that we know are definitely reworked from Permian and Lower Cretaceous sediments; • the high concentration in pollen recovered in Unit II as opposed to concentrations recovered from Unit I and cores JPC-30 and -31; • the frequent occurrence of angiosperm palynomorph clusters that suggests deposition close to the parent plant, with little opportunity for dispersion of individual grains (Smith et al., 2019). However, fluvial/glaciofluvial processes leave open the possibility of selective erosion and deposition of older reworked terrestrial sediments sourced from the central Aurora and Sabrina Basins (Young et al., 2011; Aitken et al., 2016). Plant derived biomarkers from Sabrina Coast shelf sediments provide further information on both transport history and paleoenvironment. Plant wax δD values are relatively stable across both JPC-54 and -55, a stability which is remarkable given their difference in age. Paleoprecipitation estimates indicate a δDprecip ranging from −136 to −118‰, with a mean of −129 ± 6‰, slightly more positive than coastal snow in the same region today (−133‰) and equal to modern snow values from the Antarctic Peninsula (Masson-Delmotte et al., 2008). This result is initially surprising considering the shift in climate regime from temperate in the Paleogene to polar in the present day. However, precipitation changes at the coast are often minimal, as demonstrated by comparison of modern and modelled precipitation isotope gradients used to interpret late Eocene samples from the margins of the Antarctic Peninsula (Feakins et al., 51 2014). Thus, we infer from δD values that biomarkers in JPC-54 and -55 were most likely sourced locally from plants in the proximal coastal lowland regions of the Sabrina Coast, as more D-depleted values would be expected for biomarkers sourced from farther inland or higher elevations (Masson-Delmotte et al., 2008). Further support for a local plant source comes from the presence of plant wax n-alkanoic acids, which have high abundances of C16 and C18 that indicate fresh biomass inputs (likely aquatic production) and high CPI long chain plant derived n-alkanoic acids, which indicate fresh inputs of those biomarkers without thermal alteration (Fig. 2.13). Plant wax n-alkanoic acids are likely sourced from wind or fluvial transport from the adjacent continent. Modern studies of catchment sourcing are not possible in Antarctica today, given ice cover. However, studies elsewhere show that, while rivers are catchment integrators, plant waxes are typically derived from lowland sources, due to the generally greater areal extent of lowlands and their proximity to the continental shelf, where these compounds are generally deposited; proximal sourcing is especially prominent for the n-alkanoic acids (Galy et al., 2011, Hemingway et al., 2016, Feakins et al., 2018). Studies of modern temperate forest tree leaves in Chile noted that the n-alkanoic acids were more abundant than n-alkanes in Nothofagus dombeyi (Cerda-Peña et al., 2020), supporting plant inputs of these long chain n-alkanoic acids from ancient Nothofagus spp. Plant wax ẟ 13 C provides further insight into the hydroclimate of the Sabrina Coast. Water supply is the dominant factor affecting carbon isotope fractionation in plants (Diefendorf et al., 2010). For the plant waxes analyzed in Sabrina shelf sediments, ẟ 13 C30 values from JPC-54 (−31 to −29‰) are compared to the global distribution of modern ẟ 13 C from leaf waxes in a range of climatic conditions (Diefendorf and Freimuth, 2017). The more positive ẟ 13 C values obtained here are best explained as being derived from open habitats (not closed canopy) and/or tundra 52 conditions, with reduced discrimination against 13 C likely due to low moisture availability (Diefendorf et al., 2010). The ẟ 13 C values are inconsistent with a closed canopy woodland, where respired CO2 leads to more depleted 13 C values. These ẟ 13 C values and climatic interpretation are consistent with the few Paleogene records from Antarctica, including the ẟ 13 C30 from ODP Site 1166 in Prydz Bay (Units III and IV: −28 to −25‰; Tibbett et al., 2021) and pollen ẟ 13 C from the Antarctic Peninsula (Griener et al., 2013). While the long chain n-alkanoic acids in proximal marine sediments are dominantly plant derived, with similar distributions noted in extant Nothofagus elsewhere (Cerda-Peña et al., 2020), shorter chain lengths may have microbial sources (Chen et al., 2019), as corroborated by their higher ẟ 13 C values. However, we cannot rule out some contribution from microbial sources to the C30 n-alkanoic acids (Chen et al., 2019), but they are most likely plant derived based on the abundant pollen described here. The presence of abundant, ‘fresh’ or penecontemporaneous n-alkanoic acids and lesser amounts of low CPI n-alkanes and partially matured hopanes detected in Sabrina Coast shelf sediments indicate both penecontemporaneous inputs of plant materials as well as an admixed component of reworked materials (Fig. 2.13). We can discount the possibility of thermal alteration in situ, as n-alkanoic acids would not be preserved. We therefore suggest that JPC-54 and -55 contain alkanes and hopanes derived from older, mature sediments that were likely eroded from the Sabrina and Aurora Basins and deposited on the Sabrina Coast shelf, in conjunction with long chain n-alkanoic acids that reflect penecontemporaneous vegetation inputs, similar to the biomarker findings and interpretations in other marginal settings around Antarctica (Duncan et al., 2019, Tibbett et al., 2021). It is the combination of evidence for both fresh and reworked components in both the biomarkers and pollen studies, also described in other Antarctic marginal sediments (Tibbett et al., 2021), that are particularly compelling. Here, the biomarker mixtures 53 echo that of the palynology with evidence for both penecontemporaneous and reworked pollen (Smith et al., 2019) as well as lignite in JPC-55 Unit II sediments (Gulick et al., 2017). We do not find any evidence that the sediments in these short cores are homogenous, as might occur through mixing, as many of the biomarker proxies show considerable variability in abundance and in some of the measured indices, although isotopes are invariant (Fig. 2.13). To assess reworking in the short Sabrina Coast sedimentary sequences, we studied the palynomorphs in the late Quaternary-age Unit I of JPC-54 and -55 as well as the late Miocene to Pliocene sediments recovered in JPC-30 and -31 (Gulick et al., 2017). In Unit I and in JPC-30 and -31, palynomorphs are in low abundance and correspondence analysis indicate that these statistical groups (G1, 54-1, and 55-1) are similar and may be at least partially derived from the same sedimentary source (Fig. 2.11). 2.4.2 Sabrina Coast paleoenvironments The Sabrina Flora is a unique assemblage, with high abundances of Gambierina spp., Battenipollis spp., and Proteaceae (Fig. 2.5, Fig. 2.7). Low abundances of Nothofagidites spp. (<10%) make the Sabrina Flora distinct from other Antarctic margin records, where Nothofagidites spp. dominate assemblages (e.g., Askin, 1988, Francis et al., 2009, Truswell and Macphail, 2009, Anderson et al., 2011, Warny and Askin, 2011, Contreras et al., 2013). The high abundances of palynomorphs with unknown botanical affinities make environmental reconstruction based on the Sabrina Flora alone difficult. A multi-proxy approach can provide insights into the Late Cretaceous to Cenozoic evolution of paleoenvironmental conditions on the Sabrina Coast. These insights will likely evolve as longer records are recovered by drilling, dating is refined, and reworking is assessed. However, short cores collected from the Sabrina 54 Coast provide a unique first glimpse at the region’s pre-glacial to glacial paleoenvironment (Gulick et al., 2017; Smith et al., 2019; Montelli et al., 2020). Sediment core JPC-55 was collected from a seismically stratified facies with subparallel reflectors, 15–20 m below the youngest of a series of progradational wedge-shaped clinoforms interpreted as deltas of fluvial and/or glaciofluvial origin, suggesting regression-transgression sequences that reflect changes in relative sea level and/or sediment flux to the shelf (Gulick et al., 2017; Montelli et al., 2020). The seismic signature, coupled with comparisons to previously drilled Antarctic margin sequences (e.g., McKay et al., 2019) indicate that JPC-55 was recovered from an ice-distal hemipelagic sedimentary sequence seaward of a low lying fluvial coastal plain (Young et al., 2011, Aitken et al., 2016, Gulick et al., 2017 Montelli et al., 2020). Preliminary age assessments indicate that JPC-55 contains sediments of Paleocene age (Gulick et al., 2017, Smith et al., 2019); this estimate will be further refined when longer sedimentary sequences are recovered from the Sabrina Coast margin (Macphail, 2021; (McKay, 2022)). The Sabrina Flora assemblage in JPC-55 Unit II (groups 55-2 and 55-3; Fig. 2.6) is dominated by G. askinae and B. sabrinae with unusually rugulate sculptured exines, relatively high abundances of P. parvus (4.7%), low abundances of Nothofagidites spp., and relatively low cryptogam spore abundances (3.2–9.5%). Taken together, the assemblage is consistent with limited plant moisture availability and a moderately dry environment (Griener et al., 2013). A complicating factor is that the dominant species in the Sabrina Flora (e.g., Gambierina, Battenipollis, Forcipites) have no known botanical affinities or relationships with modern Austral flora. Dettmann and Jarzen (1988) suggested that Gambierina, Battenipollis, and Forcipites may have inhabited a forest environment adjacent to an estuary. Correspondence analysis shows B. sectilis and F. stipulatus likely inhabited similar environments to P. parvus, which typically indicates open, shrubby 55 environments (Bowman et al., 2014). In total, the palynomorph, seismic data (Gulick et al., 2017; Montelli et al., 2020), subglacial topography and geomorphology (Young et al., 2011, Aitken et al., 2016), and plant wax δ 13 C data suggest that a moderately dry open coastal vegetation existed along the Sabrina Coast in the Late Cretaceous (Macphail, 2021) to early Paleogene (Gulick et al., 2017, Smith et al., 2019), and that terrestrial sediment was likely transported to the shelf by fluvial processes. Core JPC-54, dated to the early to middle Eocene, was recovered 13 m above the youngest clinoform, and 25–30 m above JPC-55, from a seismic facies with moderate laterally variable reflectors indicative of a more ice-proximal setting, but one where ice had not yet advanced across the continental shelf. Lonestones recovered in JPC-54 are interpreted as ice rafted debris, indicative of marine terminating ice along the Sabrina coast (Gulick et al., 2017, Smith et al., 2019 Montelli et al., 2020). The palynomorph assemblage in JPC-54 differs significantly from that of JPC-55, as it is dominated by G. rudata (Fig. 2.7). This species exhibits a smooth exine structure as compared to the rugulate species dominant in the JPC-55 assemblage. The JPC-54 assemblage has higher abundances of N. lachlaniae (4.6%) compared to JPC-55 (1.7%). Correspondence analysis shows N. lachlaniae plotting together with L. ovatus, Liliacidites spp., Proteaceae, and Arecipites spp. (Fig. 2.12). This correlation could be indicative of a complex open forest environment containing relatively high abundances of both overstory (i.e., Nothofagus and Arecipites [palms]) and understory (i.e., ferns, including L. ovatus) vegetation. Nothofagidites spp. abundances are slightly higher than in JPC-55, although still relatively low when compared with Eocene sequences in the Antarctic Peninsula, Transantarctic Mountains, Prydz Bay, and Wilkes Land (e.g., Askin, 1988, Francis et al., 2009, Truswell and Macphail, 2009, Anderson et al., 2011, Warny and Askin, 2011, Contreras et al., 2013). Cryptogam spore 56 abundances remain relatively low (4.6–15.7%), but higher than in JPC-55. Taken together, the palynological data from JPC-54 suggest a more humid Sabrina Coast environment with an open canopy forest in the early to middle Eocene. However, the average plant wax alkanoic acid δ 13 C value (30.2 ± 0.5‰) and low abundances of spores and Nothofagidites spp. suggest that regional conditions were still relatively dry. The co-existence of marine-terminating glaciers (Gulick et al., 2017) and an ecosystem with highly diverse vegetation could be indicative of an environment similar to modern-day Patagonia, southern New Zealand, or southeast Alaska. 2.5 Conclusions The Sabrina Flora is composed of a unique assemblage never previously recovered from Antarctica. Detailed species distribution results and new statistical analyses were performed on the assemblage and discussed to extrapolate potential paleoenvironmental affinities of Battenipollis sabrinae and Gambierina askiniae, the two recently described species that dominate the assemblage (Smith et al., 2019). We add analysis of biomarkers to find evidence for penecontemporaneous and reworked components of the biomarkers as a mixture in these sediments, from which we can interpret the in situ component (n-alkanoic acids) for signals of paleoenvironment. The palynomorph assemblage and plant wax n-alkanoic acid δ 13 C values suggest that the Sabrina Flora represents an open canopy woodland or shrubby environment. Pollen assemblage differences between the older JPC-55 and younger JPC-54 suggest a shift in vegetation and climate from somewhat drier shrubby, open vegetation (JPC-55) to a more open canopy forest (JPC-54). Global climate trends may play a role in this change in vegetation as well as potential influence from the opening of the Tasman gateway. However, improved age controls will be necessary for further interpretations. This study highlights the need for longer sedimentary sequences from East Antarctica, both from the continental shelf and from proximal 57 subglacial basins. Such records will enable researchers to assess the impact of reworking on proximal sequences, increase chronological confidence, and improve understanding of Late Cretaceous to Recent paleoenvironments in the Aurora and Wilkes Subglacial Basins. Acknowledgements Core acquisition during NBP14-02 was supported by the National Science Foundation Antarctic Integrated Systems Science Project (grants NSF PLR1143836, PLR-1143837, PLR-1143843, PLR-1430550, and PLR-1048343). We thank the NBP14-02 crew and scientific party for their roles in collecting the samples. Processing of all palynologic samples was funded by US NSF CAREER grant ANT-1048343 to S. Warny. Plant wax work was funded by NSF-OPP-1908548 to S. Feakins. 58 References Aitken, A. R. A., J. L. Roberts, T. D. Van Ommen, D. A. Young, N. R. Golledge, J. S. Greenbaum, D. D. Blankenship, and M. J. Siegert. (2016). Repeated large-scale retreat and advance of Totten Glacier indicated by inland bed erosion. Nature, 533(7603), 385- 389. Anderson, J. B., Warny, S., Askin, R. A., Wellner, J. S., Bohaty, S. M., Kirshner, A. E., Livsey, D. N., Simms, A. R., Smith, T. R., Ehrmann, W. and Lawver, L. A. (2011). Progressive Cenozoic cooling and the demise of Antarctica’s last refugium. Proceedings of the National Academy of Sciences, 108(28): 11356-11360. Askin, R. A. (1988). Campanian to Paleocene palynological succession of Seymour and adjacent islands, northeastern Antarctic Peninsula. In R. M. Feldman and M. O. Woodburne, eds., Geology and Paleontology of Seymour Island, Antarctic Peninsula. Geological Society of America Memoir 169, 131-153. Askin, R. A. (2000). Spores and pollen from the McMurdo Sound erratics, Antarctica. AGU Antarctic Research Series, 76:161-181. Bakkelund, A., Porter, T.J., Froese, D.G., Feakins, S.J. (2018). Net fractionation of hydrogen isotopes in n-alkanoic acids from soils in the northern boreal forest. Organic Geochemistry 125, 1-13. Baudoin, P. (2018). Reconstruction of Antarctic Cenozoic Paleoenvironments Through Palynological Analysis of Subglacial Lake and Ice Stream Sediments. LSU Master’s thesis. https://digitalcommons.lsu.edu/gradschool_theses/4721/. Bowman, V. C., Francis, J. E., Askin, R. A., Riding, J. B., Swindles, G. T. (2014). Latest Cretaceous-earliest Paleogene vegetation and climate change at the high southern latitudes: palynological evidence from Seymour Island, Antarctic Peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology 408, 26-47. Cande, S. C., Mutter, J. C. (1982), A Revised Identification of the Oldest Sea-Floor Spreading Anomalies between Australia and Antarctica. Earth and Planetary Science Letters, 58:151-160. Cerda-Peña, C., Contreras, S., & Rau, J. R. (2020). Molecular n-alkyl leaf waxes of three dominant plants from the temperate forest in South America. Organic Geochemistry, 149, 104105. Chen, X., Liu, X., Wei, Y., Huang, Y. (2019). Production of long-chain n-alkyl lipids by heterotrophic microbes: new evidence from Antarctic lakes. Organic Geochemistry, 103909. Clarke, K.R. (1993). Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18, 117–143. Coenen, J. J., Scherer, R., Baudoin, P., Warny, S., Castañeda, I. S., & Askin, R. (2019). Paleogene marine and terrestrial development of the West Antarctic Rift System Geophysical Research Letters, 47, 3. 59 Contreras, L., Pross, J., Bijl, P. K., Koutsodendris, A., Raine, J. I., van de Schootbrugge, B., Brinkhuis, H. (2013). Early to middle Eocene vegetation dynamics at the Wilkes Land Margin (Antarctica). Review of Palaeobotany and Palynology, 197: 119-142. Contreras, L., Pross, J., Bijl, P. K., O’Hara, R. B., Raine, J. I., Sluijs, A., Brinkhuis, H. (2014). Southern high-latitude terrestrial climate change during the Palaeocene-Eocene derived from a marine pollen record (ODP 1172, East Tasman Plateau). Climate of the Past, 10(4):1401-1420. Cookson, I.C. (1950). Fossil pollen grains of Proteaceous type from Tertiary deposits in Australia. Australian Journal of Science, series B, 3: 166-177. Cookson, I.C.; Pike, K.M. (1954). Some dicotyledonous pollen types from Cainozoic deposits in the Australian region. Australian Journal of Botany 2: 197-219. Couper, R.A. (1960). New Zealand Mesozoic and Cainozoic plant microfossils. New Zealand Geological Survey Palaeontological Bulletin 32: 87. Coxall, H. K., Wilson, P. A., Pälike, H., Lear, C. H., & Backman, J. (2005). Rapid stepwise onset of Antarctic glaciation and deeper calcite compensation in the Pacific Ocean. Nature, 433(7021), 53-57. DeConto, R. M., Pollard, D. (2016). Contribution of Antarctica to past and future sea-level rise. Nature, 531(591), doi: 10.1038/nature17145. Dettmann, M. E., Jarzen, D. M. (1988). Angiosperm pollen from uppermost Cretaceous strata of southeastern Australia and the Antarctic Peninsula. Memoirs of the Association of Australasian Palaentologists 5, 217-237. Diefendorf, A. F., Mueller, K. E., Wing, S. L., Koch, P. L., Freeman, K. H. (2010). Global patterns in leaf 13 C discrimination and implications for studies of past and future climate. Proceedings of the National Academy of Science, 107, 13. Duncan, B., McKay, R., Bendle, J., Naish, T., Inglis, G. N., Moossen, H., Levy, R., Ventura, G. T., Lewis, A., Chamberlain, B., & Walker, C. (2019). Lipid biomarker distributions in Oligocene and Miocene sediments from the Ross Sea region, Antarctica: Implications for use of biomarker proxies in glacially-influenced settings. Palaeogeography, Palaeoclimatology, Palaeoecology, 516, 71–89. Escutia, C., Brinkhuis, H., & Klaus, A. (2011). IODP expedition 318: From greenhouse to icehouse at the Wilkes Land Antarctic Margin. Scientific Drilling, 12, 15-23. Feakins, S. J., Warny, S., Lee, J. (2012). Hydrologic cycling over Antarctica during the middle Miocene warming. Nature Geoscience, 5, 557-560. Feakins, S. J., Warny, S., DeConto, R. M. (2014). Snapshot of cooling and drying before onset of Antarctic Glaciation. Earth and Planetary Science Letters, 404, 154-166. Feakins, S.J., Wu, M.S., Ponton, C., Galy, V. and West, A.J. (2018) Dual isotope evidence for sedimentary integration of plant wax biomarkers across an Andes-Amazon elevation transect. Geochimica et Cosmochimica Acta, 242, 64-81. 60 Francis, J. E. (2000). Fossil wood from Eocene high latitude forests McMurdo Sound, Antarctica. AGU Antarctic Research Series. 76:253–260. Francis, J. E., Marenssi, S., Levy, R., Hambrey, M., Thorn, V. C., Mohr, B., Brinkhuis, H., Warnaar, J., Zachos, J., Bohaty, S., DeConto, R. (2009). From greenhouse to icehouse– the Eocene/Oligocene in Antarctica. Developments in Earth and Environmental Science, 8:309-368. Fretwell, P., Pritchard, H. D., Vaughan, D.G., Bamber, J. L., Barrand, N. E., Bell, R., Bianchi, C., Bingham, R. G., Blankenship, D. D., Casassa, G., Catania, G., Callens, D., Conway, H., Cook, A. J., Corr, H. F. J., Damaske, D., Damm, V., Ferraccioli, F., Forsberg, R., Fujita, S., Gogineni, P., Griggs, J. A., Hindmarsh, R. C. A., Holmlund, P., Holt, J. W., Jacobel, R. W., Jenkins, A., Jokat, W., Jordan, T., King, E. C., Kohler, J., Krabill, W., Riger-Kusk, M., Langley, K. A., Leitchenkov, G., Leuschen, C., Luyendyk, B. P., Matsuoka, Mouginot, K. J., Nitsche, F. O., Nogi, Y., Nost, O. A., Popov, S. V., Rignot, E., Rippin, D. M., Rivera, A., Roberts, J., Ross, N., Siegert, M. J., Smith, A. M., Steinhage, D., Studinger, M., Sun, B., Tinto, B. K., Welch, B. C., Wilson, D., Young, D. A., Xiangbin, C., Zirizzotti, A. (2013). Bedmap2: improved ice bed, surface and thickness datasets for Antarctica. The Cryosphere, 7, 375–393, doi:10.5194/tc-7-375- 2013. Galy, V., Eglinton, T., France-Lanord, C. and Sylva, S. (2011) The provenance of vegetation and environmental signatures encoded in vascular plant biomarkers carried by the Ganges- Brahmaputra rivers. Earth and Planetary Science Letters 304, 1-12. Greenbaum, J. S., D. Blankenship, A. Young, T. G. Richter, J. L. Roberts, A. R. A. Aitken, B. Legresy, D. M. Schroeder, R. C. Warner, T. D. van Ommen, M. J. Siegert (2015). Ocean access to a cavity beneath Totten Glacier in East Antarctica. Nature, 8, 294-298, doi:10.1038/NGEO2388. Griener, K. W., Nelson, D. M., Warny, S. (2013). Declining moisture availability on the Antarctic Peninsula during the late Eocene. Palaeogeography, Palaeoclimatology, Palaeoecology, 383:72-78. Gulick, S. P. S., Shevenell, A. E., Montelli, A., Fernandez, R., Smith, C., Warny, S., Bohaty, A. M., Sjunneskog, C., Leventer, A., Frederick, B., Blankenship, D. D. (2017). Initiation and long-term instability of the East Antarctic Ice Sheet. Nature, (552)225, doi: 10.1038/nature25026. Hammer, Ø., Harper, D., Ryan, P. (2001). PAST: paleontological statistics software package for education and data analysis. Palaeontologia Electronica 4. Hemingway, J.D., Schefuß, E., Dinga, B.J., Pryer, H. and Galy, V.V. (2016) Multiple plant-wax compounds record differential sources and ecosystem structure in large river catchments. Geochimica et Cosmochimica Acta, 184, 20-40. Hou, B., Alley, N. F., Frakes, L. A., Stoian, L, Cowley, W. M. (2006). Eocene stratigraphic succession in the Eucla Basin of South Australia and correlation to major regional sea- level events. Sedimentary Geology, 183:297-583. 61 Inglis, G. N., Naafs, B. D. A., Zheng, Y., McClymont, E. L., Evershed, R. P., & Pancost, R. D. (2018). Distributions of geohopanoids in peat: Implications for the use of hopanoid-based proxies in natural archives. Geochimica et Cosmochimica Acta, 224, 249–261. Inglis, G. N., Carmichael, M. J., Farnsworth, A., Lunt, D. J., & Pancost, R. D. (2020). A long- term, high-latitude record of Eocene hydrological change in the Greenland region. Palaeogeography, Palaeoclimatology, Palaeoecology, 537, 109378. https://doi.org/10.1016/j.palaeo.2019.109378 Jarzen, D. M., Dettmannn, M. E. (1992). Structure and form of austral Cretaceous Normapolles- like pollen. Geobios, 25:569-583. Kennett, J. P. (1977). Cenozoic evolution of Antarctic glaciation, the Circum Antarctic Ocean, and their impact on global paleoceanography. Journal of Geophysical Research, 82, 3843–3860. Legendre, P., Legendre, L. (2012). Numerical ecology, 3rd Edition. Developments in Environmental Modelling 24, 1006. Levy, R. H., Harwood, D. M. (2000). Tertiary marine palynomorphs from the McMurdo Sound erratics, Antarctica. AGU Antarctic Research Series. 76:183-242. Lewis, A. R., Marchant, D. R., Ashworth, A. C., Hedenäs, L., Hemming, S. R., Johnson, J. V., Leng, M. J., Machlus, M. L., Newton, A. E., Raine, I., Willenbring, J. K., Williams, M., Wolfe, A. P. (2008). Mid-Miocene cooling and the extinction of tundra in continental Antarctica. Proceedings of the National Academy of Science, 105, 31. MacPhail, M. K. (2021). The Sabrina Floras of East Antarctica: Late Cretaceous, Paleogene or reworked? Palynology, 45, doi: 10.1080/01916122.2021.1921070. MacPhail, M. K., Truswell, E. M. (2004). Palynology of Neogene slope and rise deposits from ODP Sites 1165 and 1167, East Antarctica. Pp.1-20 in: Cooper, A.K.; O'Brien, P.E. Richter, C. (eds). Proceedings of the Ocean Drilling Program, scientific results, volume 188. Masson-Delmotte, V., Hou, S., Ekaykin, A., Jouzel, J., Aristarain, A., Bernardo, R. T., Bromwich, D., Cattani, O., Delmotte, M., Falourd, S., Frezzotti, M., Gallée, H., Genoni, L., Isaksson, E., Landais, A., Helsen, M. M., Hoffmann, G., Lopez, J., Morgan, V., Motoyama, N., Noone, D., Oerter, H., Petit, J. R., Royer, A., Uemura, R., Schmidt, G. A., Schlosser, E., Simöes, J. C., Steig, E. J., Stenni, B., Stievenard, M., van den Broeke, M. R., van de Wal, R. S. W., van de Berg, W. J., Vimeux, F., White, J. W. C. (2008). A review of Antarctic surface snow isotopic composition: Observations, atmospheric circulation, and isotopic modeling. Journal of Climate 21, 3359–3387. McFarlin, J.M., Axford, Y., Masterson, A.L., Osburn, M.R. (2019). Calibration of modern sedimentary δ2H plant wax-water relationships in Greenland lakes. Quaternary Science Reviews 225, 105978. M. McKay, R., Escutia, C., De Santis, L., Donda, F., Duncan, B., Gohl, K., Gulick, S., Hernández-Molina, J., Hillenbrand, C.-D., Hochmuth, K., Kim, S., Kuhn, G., Larter, R., Leitchenkov, G., H. Levy, R., R. Naish, T., O’Brien, P., F. Pérez, L., E. Shevenell, A., & 62 Williams, T. (2022). Chapter 3—Cenozoic history of Antarctic glaciation and climate from onshore and offshore studies. In F. Florindo, M. Siegert, L. D. Santis, & T. Naish (Eds.), Antarctic Climate Evolution (Second Edition) (pp. 41–164). Elsevier. https://doi.org/10.1016/B978-0-12-819109-5.00008-6 McKay, R. M., De Santis, L., & Kulhanek, D. K. (2019). Ross Sea West Antarctic Ice Sheet History. Proceedings of the International Ocean Discovery Program, 374. Montelli, A., Gulick, S., Fernandez-Vasquez, R., Frederick, B., Shevenell, A., Leventer, A., Blankenship, D. (2019). Seismic stratigraphy of the Sabrina Coast shelf, East Antarctica: history of early dynamic glaciations. GSA Bulletin, https://doi.org/10.1130/B35100.1. Morlighem, M., Rignot, E., Binder, T., Blankenship, D., Drews, R., Eagles, G. (2020). Deep glacial troughs and stabilizing ridges unveiled beneath the margins of the Antarctic ice sheet. Nature Geoscience, 13(2), 132-137. Partridge, A. D. (2006). Late Cretaceous–Cenozoic palynology zonations, Gippsland Basin. In: Monteil, E. (coord.), Australian Mesozoic and Cenozoic Palynology Zonations. Pross, J., Contreras, L., Bijl, P. K., Greenwood, D. R., Bohaty, S. M., Schouten, S., Bendle, J. A., Rohl, U., Trauze, L. (2012). Persistent near-tropical warmth on the Antarctic continent during the Early Eocene epoch. Nature, 488 (7409):73-77. Raine, J. I., Mildenhall, D. C., Kennedy, E. M., (2011). New Zealand fossil spores and pollen: an illustrated catalogue. 4 th edition. GNS Science miscellaneous series no. 4. http://data.gns.cri.nz/sporespollen/index.htm. Raine, J. I. (1984). Outline of a palynological zonation of Cretaceous to Paleogene terrestrial sediments in west coast region, South Island, New Zealand. Republic of New Zealand Geological Survey. 109:1-82. Rau, D. (2017). In search of Antarctica’s last vegetation refugium within the McMurdo Dry Valleys. LSU Master’s Thesis. https://digitalcommons.lsu.edu/gradschool_theses/4468/. Sessions, A. L., Zhang, L., Welander, P. V., Doughty, D., Summons, R. E., & Newman, D. K. (2013). Identification and quantification of polyfunctionalized hopanoids by high temperature gas chromatography-mass spectrometry. Organic Geochemistry, 56, 120– 130. PubMed. https://doi.org/10.1016/j.orggeochem.2012.12.009. Smith, C. (2016). An Early Paleogene Palynological Assemblage from the Sabrina Coast, East Antarctica: New Species and Implications for Depositional History, University of South Florida Master’s thesis. Smith, C., Warny, S., Shevenell, A. E., Gulick, S. P. S., Leventer, A. (2019). New species from the Sabrina Flora: an early Paleogene pollen and spore assemblage from the Sabrina Coast, East Antarctica. Palynology, 43, 650-659, doi: 10.1080/01916122.2018.1471422. Stover, L.E., Partridge, A.D. (1973). Tertiary and Late Cretaceous spores and pollen from the Gippsland Basin, southeastern Australia. Proceedings of the Royal Society of Victoria 85: 237-286. 63 Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., & Feakins, S. J. (2021). Late Eocene Record of Hydrology and Temperature From Prydz Bay, East Antarctica. Paleoceanography and Paleoclimatology, 36(4), e2020PA004204, doi: 10.1029/2020PA004204. Truswell, E. M. (1983). Recycled Cretaceous and Tertiary pollen and spores in Antarctic marine sediments: a catalogue. Palaeontographica Abteilung B. 186:121-174. Truswell, E. M., MacPhail, M. K. (2009). Polar forests on the edge of extinction: what does the fossil spore and pollen evidence from East Antarctica say? Australian Systematic Botany 22: 57-106. Uemura, H., & Ishiwatari, R. (1995). Identification of unusual 17β(H)-moret-22(29)-ene in lake sediments. Organic Geochemistry, 23(7), 675–680, doi: 10.1016/0146-6380(95)00036-E. Warny S., Askin, R. (2011). Vegetation and organic-walled phytoplankton at the end of the Antarctic greenhouse world: latest cooling events. AGU Special Publication 63:193-210. Warny, S., Kymes, C. M., Askin, R., Krajewski, K. P., & Tatur, A. (2019). Terrestrial and marine floral response to latest Eocene and Oligocene events on the Antarctic Peninsula. Palynology, 43(1), 4-21. Wilkie, K.M.K., Chapligin, B., Meyer, H., Burns, S., Petsch, S., Brigham-Grette, J. (2013). Modern isotope hydrology and controls on δD of plant leaf waxes at Lake El'gygytgyn, NE Russia. Climate of the Past 9, 335-352. Wright, A. P., Young, D. A., Roberts, J. L., Schroeder, D. M., Bamber, J. L., Dowdeswell, J. A., Young, N. W., Le Brocq, A. M., Warner, R. C., Payne, A. J., Blankenship, D. D., van Ommen, T. D., Siegert, M. J. (2012). Evidence of a hydrological connection between the ice divide and ice sheet margin in the Aurora Subglacial Basin, East Antarctica. Journal of Geophysical Research, 117, F01033, doi:10.1029/2011JF002066. Young, D.A., Wright, A. P., Roberts, J. L., Warner, R. C., Young, N. W., Greenbaum, J. S., Schroeder, D. M., Holt, J. W., Sugden, D. E., Blankenship, D.D., van Omen, T. D., Siegert, M. J. (2011). A dynamic early East Antarctic Ice Sheet suggested by ice-covered fjord landscapes. Nature, 474, 72-75. 64 Chapter 3 Late Eocene record of hydrology and temperature from Prydz Bay, East Antarctica This chapter was published in 2021 as: Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., & Feakins, S. J. (2021). Late Eocene Record of Hydrology and Temperature From Prydz Bay, East Antarctica. Paleoceanography and Paleoclimatology, 36(4), e2020PA004204. https://doi.org/10.1029/2020PA004204. Abstract The Eocene-Oligocene Transition (EOT) marks the onset of Antarctic glaciation at 33.7 Ma. Although the benthic oxygen isotope record defines the major continental ice sheet expansion, recent sedimentary and geochemical evidence suggests the presence of earlier ephemeral ice sheets. Sediment cores from Ocean Drilling Program (ODP) Legs 119 and 188 in Prydz Bay provide an archive of conditions in a major drainage system of East Antarctica. We study biomarker and microfossil evidence to discern how the vegetation and climate shifted between 36–33 Ma. Pollen were dominated by reworked Permian Glossopterid gymnosperms; however, penecontemporaneous Eocene pollen assemblages indicate that some vegetation survived the glacial advances. At the EOT, brGDGT soil biomarkers indicate abrupt cooling from 13–8°C and soil pH increases from 6.0 to 6.7, suggesting drying which is further supported by plant wax hydrogen and carbon isotopic with shifts of 20‰ and 1.1‰ respectively, and evidence for drying from weathering proxies. Although the terrestrial soil biomarker influx mostly precludes use of TEX86, we find sea surface temperatures of 12°C in the late Eocene cooling to 8°C at the EOT. 65 Marine productivity undergoes a sustained increase after the glacial advance, likely promoted by enhanced ocean circulation. Between the two glacial surge events of the Priabonian Oxygen Maximum (PrOM) at 37.3 Ma and the EOT at 33.7 Ma, we observe warming of 2–5°C at 35.7 and 34.7 Ma, with increases in penecontemporaneous pollen and enhanced marine productivity, capturing the last flickers of Antarctic warmth. 3.1 Introduction The Eocene-Oligocene Transition (EOT) is a period of ~790 kyr (34.2 to 33.5 Ma) that encompasses the chronostratigraphic Eocene-Oligocene Boundary at 33.9 Ma (Coxall & Pearson, 2007; Gradstein et al., 2012; Hutchinson et al., 2020). The EOT marks the shift from the greenhouse conditions of the Eocene to the icehouse conditions of the Oligocene. A two-step increase in benthic foraminiferal oxygen isotopes (δ 18 Obenthic) of ~1.5‰ (Zachos et al., 2001; Coxall et al., 2005), the latter and larger of which is now referred to as the Earliest Oligocene Isotope Step (EOIS; Hutchinson et al., 2021) reflects a 2.5°C cooling of bottom waters and growth of a continent-wide ice sheet on Antarctica (Bohaty et al., 2012; Lear et al., 2008). The cooling of bottom waters (Lear et al., 2008) leaves 0.6‰ of the δ 18 Obenthic shift as an increase in seawater (δ 18 Oseawater), suggesting an ice sheet 60–130% of modern East Antarctic Ice Sheet at 33.7 Ma (Bohaty et al., 2012). Clumped isotopes applied to microfossils from ODP Site 689 determined a similar increase in δ 18 Oseawater of 0.75‰, suggesting an ice sheet 110–120% of modern (Petersen & Schrag, 2015). Evidence for the growth of ice sheets comes from ice-rafted debris (Zachos et al., 1992), development of ice-proximal fjord sediments in the Prince Charles Mountains (McKelvey et al., 2001), and a negative excursion in seawater εNd ( 143 Nd/ 144 Nd sample ratio relative to the chondritic uniform reservoir in parts per 10,000) in the Southern Ocean corresponding to a surge of glacial erosion (Scher et al., 2011, 2014). 66 One of the leading theories for the glaciation of Antarctica at the EOT is that a decline in atmospheric pCO2 led to ice initiation and expansion on the southern continent (DeConto et al., 2008; Hutchinson et al., 2020). Alkenone δ 13 C evidence suggests a decline in atmospheric pCO2 from 1200–1000 ppmv pre-EOT to 700–600 ppmv post-EOT (Pagani et al., 2011; Zhang et al., 2013). Independent estimates of post-EOT atmospheric pCO2 range from 930–550 ppmv, inferred from the boron isotopic composition (δ 11 B) of planktic foraminifera (Anagnostou et al., 2016; Pearson et al., 2009). Climate modeling supports the hypothesis that pCO2 drove cooling of the southern continent and ocean, which would have intensified Antarctic bottom water formation explaining the paleoceanographic changes seen further afield. The earlier hypothesis that suggested the opening of the Drake Passage and Tasman Gateway led to the thermal isolation of Antarctica (Kennett, 1977), is no longer considered viable, since the timing of gateway opening does not support such a linkage (Hill et al., 2013; Scher et al., 2015). Instead, εNd reconstructions from the Kerguelen Plateau offshore from Prydz Bay, reveal no change in the Southern Ocean circulation across the EOT (Wright et al., 2018), and instead find intensification of circulation over the Kerguelen Plateau as well as between Tasmania and Antarctica over two million years before the EOT from 35.7 Ma (Wright et al., 2018; Houben et al., 2019). On Antarctica, archives of the latest Eocene are limited by modern glacial ice cover, with accessible outcrops only on the very northern limits at 64°S on Cockburn Island (Askin et al., 1991); on Seymour Island (Ivany et al., 2011); King George Island (Sophie Warny et al., 2019) or on exposed land in the Transantarctic Mountains marking the boundary between East and West Antarctica (Ashworth et al., 2007). Other records of conditions prior to glaciation come from drilling expeditions around the continental margin such as the SHALDRIL expedition off 67 the tip of the Antarctic Peninsula (63°S) that captured a short window of time prior to the EOT (Anderson et al., 2011; Warny & Askin, 2011). Those sediments yielded pollen and plant wax, with carbon isotopic analyses (δ 13 C) on the pollen yielding evidence for cooling and drying (Feakins et al., 2014; Griener et al., 2015). Hydrogen isotopic (δD) analyses on the plant waxes yielded precipitation isotopic estimates similar to modern, i.e. low sensitivity to the huge changes between ephemeral glaciation and the glaciated continent today (Feakins et al., 2014). Larger environmental changes are expected in the continental interior, and the largest drainage basin of East Antarctica is a strong candidate for delivering interior sediments to the margins. Between the Kerguelen Plateau and the East Antarctic margin lies Prydz Bay (Fig. 3.1), drilled by ODP Legs 119 and 188 (Barron et al., 1991; O’Brien et al., 2001), ideally placed to assess the exported signals of conditions on land in the glacio-fluvial catchment from the Gamburtsev mountains, today the largest outflow from the EAIS (Shepherd et al., 2018). Ice sheets expanded at the EOT, and had likely already initiated at the Gamburtsev Mountain peaks (in the southern corner of Fig. 3.1B) in the latest Eocene (Rose et al., 2013), expanding toward the sea, via the Lambert Graben (Fig. 3.1B; Fig. A.1), during the EOT ice expansion. During the late Eocene, weathering proxies from the Prydz Bay sediments suggest mean annual air temperatures (MAAT) were >10°C at 35 Ma and suggest a gradual cooling to 8°C across two million years, without any abrupt cooling inflections detected although with two warm spikes during the late Eocene including at 35.7 Ma (Passchier et al., 2013, 2017). Pollen analysis from 35.7 Ma indicate a late Eocene Nothofagus-gymnosperm community with species tolerances suggesting MAAT below 12°C and precipitation around 1200–2500 mm/year (Macphail & Truswell, 2004; Truswell & Macphail, 2009). We return to these legacy cores with new and classic biomarker 68 and microfossil techniques to reconstruct paleoenvironments in the three million years of the latest Eocene and across the EOT. Fig 3.1. Map of Prydz Bay study location. A) Modern conditions on the East Antarctic Ice Sheet (NOAA National Centers for Environmental Information); inset map shows whole Antarctic continent, red box for region shown in A. Plotting δD values of modern snow samples from the Lambert Glacier (Delmotte, 1997) and Dome A (~80°S,~77°E; Masson-Delmotte et al., 2008) with D-depletion inland (circle symbols, color legend). Black box denotes region expanded in panel B. B) Reconstructed topography and maximum bathymetry at 34 Ma (Hochmuth, Paxman, et al., 2020) showing the Gamburtsev Mountains (GM) Lambert Graben (LG), Prydz Bay (PB) and Kerguelen Plateau (KP), black contour denotes the EOT coastline. White stars indicate the Prydz Bay continental shelf locations of ODP Sites 739C (67°16.57'S, 75°04.91'E, 412 m), 742A (67°32.98'S, 75°24.27'E,416 m), 1166A (67°41.77´S, 74°47.22´E, 475 m) and Kerguelen Plateau Site 744A (61°34.66´S, 80°35.46´E, 2307 m), studied here. Arrow indicates direction of the Indian Ocean. 3.2 Materials and Methods 3.2.1 Site Selection For this study, we selected Ocean Drilling Program (ODP) Leg 119 Sites 739, 742 and Leg 188 Site 1166 from Prydz Bay located in the Indian Ocean sector of the Southern Ocean offshore of the modern coastline of East Antarctica (Fig. 3.1A). Sites 739, 742, and 1166 are geographically adjacent, on the continental shelf at water depths of 412, 416, and 475 m respectively and together form a chronostratigraphy that has been previously used to describe sedimentology and track weathering proxies across 36 to 33 Ma, including 34.2–33.5 Ma encompassing the EOT 69 (Passchier et al., 2013, 2017). The continental margin is at the outflow of the Lambert Glacier Amery Ice Shelf System that originates from the Gamburtsev Mountains and flows through the Lambert Graben (O’Brien et al., 2001) draining much of what is today covered by the EAIS (Fig. 3.1A). The Lambert Graben is part of the Indian Mahanadi Rift System that formed during the separation of India and Antarctica at roughly 130 Ma (Harrowfield et al., 2005). Paleotopographic reconstructions indicate coastline transgressions and regressions around the Lambert Graben varies with isostatic and eustatic changes (Hochmuth, Gohl, et al., 2020; Paxman et al., 2019) (Fig. 3.1B; see also 3D map Fig. A.1). The latter aligns with fluvial/deltaic sandstones in the oldest sediments within Hole 1166A (O’Brien et al., 2001). Hole 742A, slightly inland and younger, consists primarily of mudstone and sandstone in the late Eocene suggesting a proximal glaciolacustrine or protected marine environment (Erohina et al., 2004). Approaching and spanning the EOT, in Hole 739C the lithology shifts to diamictite with mud intraclasts (Erohina et al., 2004; Passchier et al., 2017) suggesting input from glaciers and proximity to grounded ice. Samples for the biomarker and pollen analyses here were selected to complement prior work on major element geochemistry. The age model is based on previous work on biostratigraphy, magnetostratigraphy, and seismostratigraphy with ages updated to GTS2012 geologic timescale (Passchier et al., 2017 and references within). The 61 samples span from 35.8–32.9 Ma at ~50 ka resolution. In addition, the Kerguelen Plateau, 700 km offshore from the Prydz Bay sites, was assessed for biomarker presence. ODP Site 744 (2307 m water depth) is located on the Southern Kerguelen Plateau and is dominated by nannofossil ooze. We selected 10 samples from Hole 744A across the same 32.9 to 35.8 Ma time span to sample a more distal location that had yielded a strong signal of glacial advance based on the εNd transported by ocean currents (Scher et al., 2014). 70 The goal was to assess the presence of wind-blown terrigenous plant waxes and to seek ocean temperature reconstructions using the isoGDGT-based TEX86 biomarker approach in the open Southern Ocean. However, no detectable biomarkers were recovered from 20 g samples, thus no record of conditions at this site was possible. Fig 3.2. Age model for the Prydz Bay sedimentary sequence compiled from sections of 1166A, 742A and 739C, with age datums based on biostratigraphy including first and last occurrence (directional triangles) of named nannofossils (blue) and dinoflagellates (green) and seismic tie points (orange squares) relative to GTS 2012 (Gradstein et al., 2012). Chron 13n occurs within the early Oligocene, the remainder within the late Eocene, and the approximate timing of the Eocene Oligocene Transition (EOT) is denoted alongside the paleomagnetic boundaries for reference. Species labelled in gray are not used in the construction of the age model but are provided for context, tie points used to connect depth to age are illustrated with dashed lines. Diagram modified from Passchier et al., (2017) to show the linear fit age model (thick black line) and implied sedimentation rates. 3.2.2 Age Model We used a previously reported paleomagnetic and biostratigraphic age model from Passchier et al., (2017) on GTS2012 (Gradstein et al., 2012). Age control tie points occur every 100–200 m in the ~400 m sedimentary sequence compiled from three sites. The age control includes paleomagnetic tie points constrained by biostratigraphy in the mudstones at the top of Hole 1166A. 1166A includes very high sedimentation rates (50.9 cm/kyr) with 100 m deposited 71 between two age control points separated by 0.2 Ma at 35.9 and 35.7 Ma; between which, the ages for individual samples are constrained to within ~200 ka. Hole 742A bottom age of 35.7 Ma is constrained by a stratigraphic tie at 304.29 mbsf in 742A to 136.32 mbsf in Hole 1166A where the 15r chron reversal is found. Although a fossil biostratigraphic datum (LO Vozzhenikovia sp. and Deflandrea sp. A) was described in 742A, the higher stratigraphic last appearance in 739C (shown) would be closer to the extinction in the water column, but as the duration of chron 13r to which the FO is tied is > 1 Myr, it does not further constrain the age model. Across 35.7 to 33.6 Ma sedimentation rates average 14.8 cm/kyr; lithology indicates likely changes in sedimentation rate, but they cannot be constrained due to the lack of additional age constraints. Biostratigraphic constraints are sparse and the first or last appearance of species have long spans (>1 Ma), as a result, uncertainties may be as large as 1 Ma. The long duration of chron 13r and the lack of additional biostratigraphic age control between 35.7 and 33.7 Ma, prohibit high-resolution temporal correspondence with regional and global records across the entire 2-Myr interval. However, the 35.7 Ma warming near the beginning of this record and the EOT cooling at 33.7 Ma are constrained to within ~200 ka. 3.2.3 Palynology For palynology, samples were processed for terrestrial palynomorphs at Global Geolab Limited (Alberta, Canada). Dry sediment was weighed and spiked with a known quantity of Lycopodium spores to allow for evaluation of palynomorph concentrations. Dry sediment was successively treated with hydrochloric acid, hydrofluoric acid, and heavy liquid separation (Brown et al., 2008). Samples were sieved between a 10 and 250 µm fraction, and the remaining residue was mounted on microscope slides using glycerin jelly. Palynological analysis was conducted on a subset of 39 samples in the Louisiana State University’s Center for Excellence in 72 Palynology (CENEX) lab. Recovery was excellent and 300 palynomorphs were tabulated for each sample using an Olympus BX41 microscope. Concentrations were calculated as: C=(Pc×Lt×T)/(Lc×W) (1) where C = concentration (counts per gram of dried sediment, gdw –1 ), Pc = the number of palynomorphs counted, Lt = the number of Lycopodium spores per tablet, T = the total number of Lycopodium tablets added per sample, Lc = the number of Lycopodium spores counted, W = the weight of dried sediment. Palynomorphs were identified to the lowest taxonomic level possible. Data reported provides counts of spores, southern beech tree (i.e., Nothofagus), other angiosperms, gymnosperms and reworked pollen grains (likely of Permian age but potentially as young as Jurassic age), each reported as a % of all terrestrial spores and pollen. Where present in unusual quantity, charcoal or coal fragments were noted. 3.2.4 Extraction of lipids and quantification of n-alkanes and n-alkanoic acids Prydz Bay sediment samples ranging from 10–50 g were ground with a mortar and pestle then weighed. Leaf waxes were extracted using a DIONEX Accelerated Solvent Extractor using a 9:1 (v/v) dichloromethane (DCM) to methanol solution at 100°C and 1500 psi for two 15 minute cycles. The total lipid extract was separated into a neutral and acid fraction using a NH2 sepra column eluted with a 2:1 (v/v) ratio of DCM:isopropanol (neutral fraction) and 4% formic acid in diethyl ether (acid fraction). The neutral fraction was separated over silica gel eluting alkanes with hexanes and then the polar fraction eluting first with DCM and then with methanol. The acid fraction was methylated overnight with methanol of a known isotopic concentration and hydrochloric acid (95:5) and dried using an anhydrous sodium sulfate column yielding the fatty acid methyl esters (FAMEs). The FAMEs were purified using a silica gel column eluting first with hexane, and then collecting the FAMEs with DCM. An additional silver nitrate column was 73 used to remove unsaturated compounds from the FAMEs fraction that would coelute and interfere with isotopic determinations. Alkane and FAMEs were identified by Gas Chromatography mass spectrometry (GC-MS). We also screened the alkanes fraction for the presence of hopanes, biomarkers that derived from membrane lipids in bacteria and undergo isomerization with increasing thermal maturity, using their diagnostic mass fragments and published spectra (Inglis et al., 2018; Sessions et al., 2013; Uemura & Ishiwatari, 1995), and noted where those samples included an uncharacterized complex mixture. The samples were quantified by GC-FID (Gas chromatography flame ionization detector) relative to in house standards of alkanes and FAMEs of known concentration. We report the concentration of C16– C30 n-alkanoic acids and C17–C29 n-alkanes and calculate the summed C17 to C29 alkanes and summed C16 to C30 FAMEs in ng/gdw. We also calculate the average chain length (ACL) and carbon preference index (CPI) using the following formulae: ACL: ∑(𝑛𝑛 ∗ [ 𝐶𝐶 𝑛𝑛 ]) ∑ ⌊ 𝐶𝐶 𝑛𝑛 ⌋ � (2) CPI: 2 ∑ 𝐶𝐶 𝑛𝑛 (∑ 𝐶𝐶 𝑛𝑛 + 1 + ∑ 𝐶𝐶 𝑛𝑛 − 1 ) � (3) where n is 24–30 for n-alkanoic acids and 23–29 for n-alkanes. 3.2.5 Compound Specific Isotopic Analyses: δ 13 C and δD Compound specific isotope analysis was performed using a gas chromatography isotopic ratio mass spectrometer (GC-IRMS) using a Thermo Scientific Trace gas chromatograph connected to a Delta V Plus mass spectrometer via an Isolink combustion furnace at 1000°C for δ 13 C and a pyrolysis furnace at 1400°C for δD. The peak amplitude was 1–7 V. ẟ 13 C linearity was recorded each day and had an average standard deviation of 0.042‰. H3 factor was recorded every day 74 with an average value of 9.518±0.377 ppm mV -1 . Samples were normalized to Vienna Pee Dee Belemite (VPDB) and Vienna Standard Mean Ocean Water/Standard Light Antarctic Precipitation (VSMOW/SLAP) by an external standard mixture of 16 n-alkanes with δ 13 C values that range from –25.9 to –33.7‰ and δD values from –17 to –256‰ (A6 mix obtained from A. Schimmelmann, Indiana University). The RMS error determined from replicate measurements of the standards run during the course of these analyses average 0.8 and 5.1‰ for δ 13 C and δD respectively, and constitute uncertainty in terms of accuracy. Replicate measurements of individual analytes average 0.2 and 3‰ for δ 13 C and δD respectively and constitute measurement precision. Corrections were made for the methyl group added during methylation for the n- alkanoic acids (δ 13 C of –24.7±0.2‰ and ẟD of –186.9±3.7‰) by mass balance. 3.2.6 GDGT Analyses An internal C46 GDGT standard (Huguet et al., 2006) was added to the polar fractions containing glycerol dialkyl glycerol tetraethers (GDGTs) for quantification. These fractions were then dissolved in hexane: isopropanol (99:1) and filtered (0.45 m PTFE) prior to injection on an Agilent 1260 High-Performance Liquid Chromatography (HPLC) coupled to an Agilent 6120 mass spectrometer at the University of Arizona. GDGTs were analyzed using two BEH HILIC silica columns (2.1×150 mm, 1.7 μm; 90 Waters) and the methodology of Hopmans et al., (2016). Single Ion Monitoring (SIM) of the protonated molecules (M + H + ions) was used to detect and quantify GDGTs with abundances determined by comparison to an internal standard at m/z 744. We report total (Σ) concentrations of branched (brGDGT) and isoprenoidal (isoGDGTs) GDGTs in the sediments as measures of terrestrial and marine inputs respectively, and calculate the Branched and Isoprenoidal Tetraether (BIT) index: 75 𝐵𝐵𝐵𝐵 𝐵𝐵 = 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐶𝐶𝐶𝐶𝐶𝐶 𝑛𝑛 (4) where Ia, IIa and IIIa represent the abundances of both the 5´ and 6´ methyl isomers of the non- cyclic terrestrial brGDGTs (from soil acidobacteria) and Cren represents the abundance of crenarchaeol (mostly produced by marine archaea) (Hopmans et al., 2004). In all samples, the 5-methyl brGDGT index, MBT´5Me (de Jonge et al., 2014, Hopmans et al., 2016) was calculated as 𝑀𝑀 𝐵𝐵𝐵𝐵 5 𝑀𝑀𝐶𝐶 ′ = 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 (5) and converted to mean annual air temperature using the BayMBT calibration (Dearing Crampton-Flood et al., 2020). For the prior distribution, we used a mean of 10 ± 15°C based on the mean summer temperature of 10°C estimated from the S-index from the same sites (Passchier et al., 2017) which is supported by pollen assemblage MAAT estimates below 12°C (Macphail & Truswell, 2004; Truswell & Macphail, 2009). In addition, the cyclization of branched teraether (CBT') index (De Jonge et al., 2014) was calculated and utilized to estimate soil pH. 𝐶𝐶 𝐵𝐵𝐵𝐵 ′ = 𝑙𝑙𝑙𝑙 𝑙𝑙 1 0 � 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 � (6) 𝑝𝑝𝑝𝑝 = 7.15 + 1.59𝐶𝐶 𝐵𝐵 𝐵𝐵 ′ (7) IsoGDGTs with 0–3 cyclopentane moieties (GDGT-0 to GDGT-3) and crenarchaeol (Cren) with an additional cyclohexane moiety and its regioisomer crenarchaeol´ (Cren´) are dominantly from marine archaeal production. Following (Schouten et al., 2007), TEX86 was calculated using the equation: 76 𝐵𝐵𝑇𝑇𝑇𝑇 8 6 = [ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐺𝐺 − 2] +[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐺𝐺 − 3] +[ 𝐶𝐶𝐶𝐶𝐶𝐶 𝑛𝑛 ′ ] [ 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 1] +[ 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 2] +[ 𝐺𝐺𝐺𝐺 𝐺𝐺 𝐺𝐺 − 3] +[ 𝐶𝐶𝐶𝐶𝐶𝐶 𝑛𝑛 ′ ] (8) We converted the TEX86 record to mean annual sea surface temperatures (SSTs) using the BAYSPAR calibration (Tierney & Tingley, 2014). For the prior distribution, we used 15°C±10°C based on SST estimates of 22°C–12°C from the late Eocene to post-EOT from Maud Rise (Petersen & Schrag, 2015) and on mixed layer planktic foraminifera (Chiloguembelina cubensis) from Kerguelen Plateau sites 738 and 744 with δ 18 O values that indicated temperatures between 5 and 10°C (Zachos et al., 1994). Similar temperatures are also suggested in the region based on modelling of ocean temperatures aligned with temperature reconstructions from further afield (La Meseta formation, Seymour Island, Antarctic Peninsula) using clumped isotopes on bivalve shells and TEX86 (Douglas et al., 2014). Samples with a BIT>0.4 were excluded from ocean temperature estimates. In addition, samples with a delta ring index (ΔRI) >0.3 were excluded as a high ΔRI implies non-analogue distributions, following (Zhang et al., 2016): 𝑅𝑅 𝐵𝐵 𝑠𝑠 𝐼𝐼 𝑠𝑠𝑠𝑠 𝑠𝑠 𝐶𝐶 = (0[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 0] + 1[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 1] + 2[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 2] + 3[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 3] + 4[ 𝑐𝑐𝑐𝑐𝑐𝑐 𝑛𝑛 ] + 4[ 𝑐𝑐𝑐𝑐𝑐𝑐 𝑛𝑛 ′ ]) (9) 𝑅𝑅 𝐵𝐵 𝐺𝐺 𝑇𝑇 𝑇𝑇 = −0.77 𝐵𝐵𝑇𝑇𝑇𝑇 8 6 + 3.32( 𝐵𝐵𝑇𝑇𝑇𝑇 8 6 ) 2 + 1.59 (10) ∆𝑅𝑅 𝐵𝐵 = 𝑅𝑅 𝐵𝐵 𝐺𝐺 𝑇𝑇 𝑇𝑇 − 𝑅𝑅 𝐵𝐵 𝑠𝑠 𝐼𝐼 𝑠𝑠𝑠𝑠 𝑠𝑠 𝐶𝐶 (11) 77 3.3 Results In generating a new microfossil and biomarker multi-proxy study of conditions in East Antarctica before the onset of the EOT, we revisit cores that have been studied for several mineralogical and inorganic geochemical weathering proxies. The Prydz Bay late Eocene to Oligocene sedimentary sequence has previously been described by Passchier et al., (2017). The results of our organic multi-proxy study are reported on that lithostratigraphic framework in the supplementary information (Fig. A.2 and A.3). The results of this organic study are presented on the age scale here (Fig. 3.3), alongside summary contextual information about the regional erosion context and marine depositional type (Fig. 3.3A and 3.3B). 3.3.1 Pollen Palynological assemblage recovered include some penecontemporaneous angiosperm (mostly Nothofagus), gymnosperm, and spores, along with a dominating reworked assemblage composed mostly of taeniate bisaccate pollen and spores (e.g., Protohaploxypinus spp., Striatopodocarpidites spp, Lunatisporites pellucidus, Playfordiaspora crenulatus and Triplexisporites playfordii) of Permian and Permo-Triassic transition age. Most of the bisaccate pollen recovered have a diagnostic “taeniate” (striped) appearance due to ridges on the wall of the central structure, many of these forms are related to Glossopteridales (extinct seed plants) common during the Permian on Gondwana. Glossopteridales pollen was also found in a terrestrial coal sample from Lamping Peak, Central Transantarctic Mountains, Antarctica, which we tested as a potential source rock for the reworked material delivered to Prydz Bay. Examining the penecontemporary pollen assemblage in the Prydz Bay sediments (Fig. 3.3C), we find Nothofagus to be dominant (mean 56%, 1σ = 18%, n = 39), with a secondary presence of 78 gymnosperms, and lesser amounts of spores and angiosperms (other than Nothofagus). Nothofagus proportions peak at 84% at 35.7 Ma and decrease to an average of 54% (1σ = 17%, n = 33) across 35.7 to 32.9 Ma. At 34.1 Ma gymnosperm pollen reach 55%, their highest proportions. While there are temporary declines in the dominance of Nothofagus after 34.7 and 34.3 and 33.6 Ma, their proportions recover afterwards, suggesting that the surviving vegetation remains a Nothofagus-gymnosperm mixture. Penecontemporary pollen grains are well preserved, with light colored grains indicating the absence of thermal alteration. Penecontemporaneous pollen concentrations average 2x10 3 count/gdw and peak at 12.2 x10 3 counts/gdw at 35.7 Ma, when they exceed reworked pollen abundances, suggesting a flourishing ecosystem (Fig. 3.3D). Penecontemporary pollen counts then decline by two orders of magnitude to an average of 745 counts/gdw indicating deteriorating conditions for plants. Reworked pollen concentrations average 5x10 3 count/gdw and exceed penecontemporary pollen (by 1-60 fold) in almost all samples after 34.7 Ma (Fig. 3.3D). In one sample at 34.8 Ma with anomalously low concentrations of reworked pollen grains, microscopy reveals charcoal or coal fragments, likely also signals of reworking (marked with an asterisk on Fig. 3.3, Fig. A.3). Although there are variable concentrations of reworked pollen throughout the record, there is an abrupt increase in reworked pollen concentration at 33.6 Ma to peak concentrations in the record that coincides with major depositional system change to diatomaceous mudstones and massive diamictite at the Eocene Oligocene Glacial Maximum (EOGM; Fig. 3.3B). Reworked concentrations increase after the EOT and peak at 12.6x10 3 counts/gdw at 33.5 Ma (Fig. 3.3D). 79 3.3.2 Plant Wax 3.3.2.1 Abundance We quantified individual n-alkanoic acids with C16 to C30 chain length (dataset: Tibbett et al., 2021). The n-alkanoic acids are co-dominated by C16 and C18 compounds, common in many marine and terrestrial organisms and long chain n-alkanoic acids with an even over odd predominance characteristic of terrestrial higher plants. Summarizing the molecular abundance distributions of the long chain C24-C30 n-alkanoic acids presumed to derive from plant wax sources, we calculate their summed concentration (Σacid) averaging 8.5±2.7 μg/gdw (Fig. A.3), with a mean ACL of 26.4 and CPI of 5.89 (range 3.8-6.9). The lowest CPI of 3.8 is found at 34.8 Ma corresponding to low concentrations of Σacid (1.39 μg/gdw). The n-alkanoic acid fraction has a clean baseline, high CPI and long ACL and is interpreted to represent penecontemporaneous inputs, from marine and terrestrial production, and we therefore focus on the n-alkanoic acids for compound specific isotopic analyses, similar to prior work in Antarctic marginal sediments of Miocene and late Eocene age in the Ross Sea and Antarctic Peninsula respectively (Feakins et al., 2012, 2014). In particular, we select the C30 n-alkanoic acid as most likely indicative of terrestrial plant sources based on molecular abundance and carbon isotopic differentiation of sourcing between mid and long chain n-alkanoic acids (see Supplementary Information Fig. A.5 and A.6). We quantified individual n-alkanes with C17 to C29 chain length. Many samples contained short to mid-chain n-alkanes in equal or greater abundance to the longer chains, a common symptom of thermal maturity. Specific signs of maturation include a modal n-alkane shorter than C23, with short-mid chain n-alkanes having a CPI~1, emerging above an uncharacterized complex mixture 80 (UCM; see Supplementary Information Fig. A.4). All Prydz Bay alkanes fractions contained substantial UCMs and short chain alkanes, except for one sample with a typical plant wax distribution in Site 1166, during the peak warmth at 35.7 Ma. For the long chain C23-C29 n- alkanes, presumed to derive from plant waxes, we calculated a mean ACL of 25.5 and CPI of 3.7 (range 1.7–4.6), and Σalk was 2.1±0.6 μg/gdw, a quarter of the Σacid concentrations. Given the low abundance of long chain n-alkanes and signs of maturity noted here and previously in these Prydz Bay sediments (Kvenvolden et al., 1991), we suspect that at least some of the n-alkanes are thermally mature. We expect these are reworked mixtures from earlier sedimentary deposits, rather than in situ alteration given the co-occurrence with n-alkanoic acids that are incompatible with thermal alteration. Within the alkane fraction, we detected hopanes, bacterial membrane lipids that are used as thermal maturity markers (Inglis et al., 2020). We found hopanes in the alkane fractions of all Prydz Bay sediments, including the most mature samples (high abundance of short chain n- alkanes with low CPI, UCM and no alkanes >C25) from the coal-rich layer. C27, C29, C30 and C31 hopanes were identified (Fig. A.4). We focused on the C31 hopane which is the most commonly reported and determined ratios of ββ/(αβ+βα+ ββ) averaging 0.45 ± 0.07, indicating moderate thermal maturity in glacio-fluvially-eroded sediments. Downcore variations indicated elevated maturity in the coal-bearing sample (0.34) and lower maturity (0.57 to 0.76) in the warm intervals (Fig. A.3). We obtained a potential source rock, a Permian coal sample, for comparison to the reworked alkanes. Corroborating the interpretation of reworked n-alkanes in the Prydz Bay sediments, the terrestrial, pollen-bearing coal sample yielded low abundances of mature, short to mid chain (CPI~1) n-alkanes. Homologues detected above the uncharacterized complex mixture range from 81 C18 to C29, short chains dominate (with a CPI~1) and the modal chain length n-alkane was C20 (52 ng/gdw). Corroborating the interpretation of penecontemporary n-alkanoic acids, these were below detection levels in the Permian coal. This is as expected as coals typically contain mature n-alkanes that are the product of thermal alteration, whereas fatty acids are considered more labile and would be altered and lost during coalification. We did not find hopanes in the terrestrial Permian coal sample, but its alkanes are at very low concentrations and thus we infer that hopanes are below detection limit or poorly preserved. 3.3.2.2 Compound Specific Isotopic Analyses The δD of the C24, C26, C28, and C30 n-alkanoic acids had an average value of –239‰ (range – 280‰ to –164‰), –257‰ (range –287‰ to –216‰), –232‰ (range –269‰ to –196‰), and – 222‰ (range –245‰ to –206‰). Due to microbial production of n-alkanoic acids which has been evaluated in Antarctic glacial lakes previously (Chen et al., 2019), and consistent with ẟ 13 C values (reported below and Fig. A.5), we assume the C30 best represents a plant wax source, hereafter δDwax. The downcore record of δDwax (Fig. 3.3H) shows that there is a trend toward D- enrichment up section. The most depleted value is in the oldest sample at 35.8 Ma and the most enriched value occurs at 33.3 Ma. δDwax record indicates large variability prior to and during the EOT with fluctuations as high as 30‰ from 35.2 to 35.0 Ma. At the EOT, there is a high of – 208‰ for δD corresponding to a 20‰ increase from 33.8 to 33.5 Ma. We found the C24, C26, C28, and C30 n-alkanoic acids had a mean δ 13 C value of –25.5 (range– 27.2‰ to –24.4‰), –25.6 (range –27.1‰ to –24.9‰), –27.3 (range –28.2‰ to –25.6‰), – 29.7‰ (–31.7‰ to –28.5‰). The tendency is to 13 C-depletion in longer chain lengths (see supplementary information, Fig. A.5 and A.6). Microbial inputs of n-alkanoic acid may be present in the shorter chain lengths as indicated by their more enriched values as previously 82 noted in Antarctic lakes (Chen et al., 2019); therefore, we report and discuss δ 13 C of the C30 n- alkanoic acid, hereafter δ 13 Cwax. Across the Prydz Bay record, there is no temporal trend in the δ 13 Cwax record; however, from 35.3 to 34.9 Ma there is a 1.8‰ positive shift followed by a 2.4‰ decrease in δ 13 Cwax at 34.7 Ma (Fig. 3.3G). After this event δ 13 Cwax increases into and across the EOT reaching a maximum of –28.6‰ after the EOT at 33.6 Ma. There is a decrease in δ 13 Cwax around 33.3 Ma from –28.9‰ to –31.3‰ before returning to the baseline of ~–29.0‰. The ẟ 13 C of the C30 n-alkanoic acid ranged from –31.9 to –28.6‰, with a 3.3‰ positive trend across the Prydz Bay record. 3.3.3 GDGTs ΣbrGDGTs had an average concentration of 42 ng/gdw with a range of 9–154 ng/gdw. The maximum concentrations correspond to Site 1166 with values of 115–154 ng/gdw at 35.7 Ma (Fig. A.3). The rest of the record from Sites 742 and 739, the brGDGT concentrations remain lower than 100 ng/gdw. For all samples we calculated MBT´5Me and used BayMBT calibration to estimate mean annual air temperatures over the adjacent continent between 8–20°C. We found soil pH, based on CBT', spanned pH 5.7 at 37.5 Ma to 6.7 at 33.5 Ma with a step increase at 33.7 Ma from pH 6.1–6.5. ΣisoGDGTs had an average concentration of 150 ng/gdw with a range of 12–730 ng/gdw (Fig. 3.3E). Site 1166 had the highest concentrations with 730 ng/gdw at 35.7 Ma. Site 742 had an initial increase in ΣisoGDGT concentration from 25 to 109 ng/gdw that overlaps with an increase from Site 1166 (35.8–35.7 Ma). In the rest of the record there were notable increases in the ΣisoGDGTs from 39 to 266 ng/gdw at site 742 at 35.0 Ma, from 21 to 344 ng/gdw at site 739C at 34.7 Ma, and values remained above 100 ng/gdw starting at 33.7 Ma and reaching a maximum 83 of 515ng/gdw at 33.6 Ma and of 526 ng/gdw at 33.4 Ma. The lowest concentrations reached after 33.7 Ma are 178 ng/gdw at 33.1 Ma and 126 ng/gdw 32.9 Ma. BIT values ranged between 0.2 (marine-dominated) and 1.0 (terrestrial-dominated), however most samples were terrestrially-dominated precluding ocean temperature reconstructions. A few samples have low BIT, and that change is driven by increased marine productivity as denoted by ΣisoGDGT abundance spikes. In the Southern Ocean, GDGT-1, 2 and 3 are minimal, and distributions are dominated by GDGT-0 and crenarchaeol (Zhang et al., 2016) this can lead to a loss of sensitivity in the coldest waters (Tierney & Tingley, 2015), however with TEX86 of 0.4 to 0.6 in these late Eocene samples, temperature estimates from these samples are likely to be robust. The ΔRI ratio allows us to check for normalcy of the distributions obtained for the isoGDGTs. The ΔRI ranged from 0.14 to 1.44 and denote samples where the TEX86 estimates would be biased by distributions outside those known for the modern ocean. For samples with BIT<0.4 and ΔRI<0.3, TEX86 values were used to estimate ocean temperatures using the BAYSPAR calibration finding temperatures from 8 to 14°C (n=6), most of the samples were dominated by terrestrial signals such that ocean temperatures could not be obtained. 84 Fig 3.3. Multi-proxy latest Eocene reconstruction from Prydz Bay showing A) conditions in the Gamburtsev Mountains from ephemeral glaciation to accelerated erosion during glacial expansion to cold-based ice sheets at the EOT and B) depositional systems inferred from Prydz Bay stratigraphy (both from Passchier et al., 2017 and references therein). Palynology results showing C) penecontemporaneous pollen assemblage proportions and D) total penecontemporaneous and reworked pollen proportions. Biomarker results including E) ΣisoGDGT concentrations as evidence for marine production (light blue circles), F) and temperature estimates based on two soil-proxies for land conditions yielding MAAT estimates from the biomarker based MBT´5me with BayMBT calibration (orange downward triangles, this study) compared to the S-index weathering proxy (purple diamonds, Passchier et al., 2017), and limited ocean temperature estimates from the BAYSPAR calibration of TEX86 (dark blue circles). Plant wax records include G) δ 13 Cwax from the C30-acid (upward brown triangles), H) δDwax from the C30-acid (upward green triangles). I) soil pH was reconstructed from CBT´. J) MAP using the calibration of Sheldon et al., (2002) to the CIA-K weathering proxy from Passchier et al., (2017). For comparison, we show regional records of glacial ice volume and ice expansion based on K) εNd in fish teeth (Scher et al., 2011, 2014) from ODP Site 738 (purple circles)as well as L) δ 18 Obenthic in Southern Ocean Sites 689, 738, 743 and 744 (Bohaty et al., 2009; Bohaty & Zachos, 2003; Cramer et al., 2009; Scher et al., 2014). Purple line highlights the diamict to diatomite depositional transition in Prydz Bay. 85 3.4 Discussion 3.4.1 Excluding the Permian influence in late Eocene sediments When reconstructing vegetation from these Prydz Bay sediments we find a large and variable proportion of reworked pollen and spores of primarily Permian age, although a few Cretaceous angiosperms were recovered as well. The majority of the palynological yield is composed of extinct seed plant’s bisaccate taeniate grains related to Glossopteridales, spores, and other Permian to Permo-Triassic transition age species. These reworked palynomorphs are likely sourced from the Prince Charles Mountains along the Lambert Glacier-Amery ice shelf system. Within the Lambert Graben there is a wide distribution of Permian sediment consisting of Permian coal and sandstone (Veevers and Saeed, 2008) with upper Permian coals within the Bainmedart Coal Measures that contain plant material including pollen, wood, leaves, and charcoal material from Glossopterid gymnosperms (McLoughlin et al., 1997). Fluvial or glacial erosion of these Permian sediments likely explain the reworked Permian pollen detected in Prydz Bay cores (Fig. 3.3D), with glacial erosion expected to explain the increased erosion of reworked pollen at and after the EOT (see Section 4.5). Prior to the erosive flux from marine-terminating glaciers detected in marine geochemistry at the EOT (Scher et al., 2011), there was an earlier glacial advance at the PrOM (Scher et al., 2014) with glacio-fluvial erosion (non-marine terminating glaciers) with geochemical signals detected at the Kerguelen Plateau, before the start of this Prydz Bay record. After the PrOM warm events would have melted glacial ice and led to fluvial erosion, and then the cirque mountain glaciers in the Gamburtsev Mountains (Rose et al., 2013) would have re-initiated and expanded as temperatures dropped. Thermochronometry provides constraints on this phase of erosion and suggests that while glaciers were still wet based, before the EOT, they had their greatest erosive power (Tochilin et 86 al., 2012). We expect that the climatic fluctuations and the erosive power of advancing ice, increased the erosion of ancient sediments (bearing pollen, alkanes and more thermally mature hopanes). At 34.8 Ma (483.93 mbsf, Fig. A.3) there is a large input of coal and reworked pollen, in the nearest biomarker sample (at 484.05 mbsf) we find no isoprenoidal or branched GDGTs detected, very low amounts of long chain n-alkanoic acids. This sample had the highest concentration of short- to mid-chain (C17-C25) n-alkanes (an order of magnitude above other samples), with a modal concentration in C18, a low CPI, and no n-alkanes >C25 detected, consistent with alkanes from coal. We compared a sample of pollen-bearing (Warny, pers. comm.), Permian-age strata, from Lamping Peak in the Beardmore Glacier region of the Central Transantarctic Mountains (Flaig, pers. comm.) to assess possible reworking of organic proxies along with reworked pollen. The coal sample did not contain GDGTs or n-alkanoic acids, suggesting no ancient contributions to the brGDGT or isoGDGT proxies or the plant wax n- alkanoic acids found in Prydz Bay. The coal sample included mostly short-chain n-alkanes (<C25 n-alkanes) with no odd over even preference above an uncharacterized complex mixture, typical of thermally mature samples. The fraction included low abundances of long chain n-alkanes from C25-C29 with the expected odd over even dominance suggesting plant-derived distribution (including detectable amounts of the C29 n-alkane). Although Permian plant waxes are not well known, during the Permian, gymnosperms would have been the plants present and gymnosperms produce some n-alkanes (Diefendorf et al., 2015). Although peaks are barely above detection limits it remains possible that erosion of Permian coals can contribute to the long chain n- alkanes, therefore we avoid this compound class for isotopic reconstructions in Prydz Bay. Furthermore mature n-alkane distributions were found in Prydz Bay samples in this study as well as in the original reports (Kvenvolden et al., 1991). The coal sample did not yield long chain n- 87 alkanoic acids or brGDGTs, thus we infer that these biomarkers and their isotopes reflect penecontemporaneous inputs into Prydz Bay. Penecontemporary pollen in the Prydz Bay sediments can be readily differentiated from reworked pollen, based on their taxon identification, surface features (specifically the Permo- Triassic “taeniate” features described previously for the reworked grains), and preservation (with good preservation for the penecontemporary pollen) and low thermal maturity (light colored grains). Species present include Nothofagus and other angiosperms in small amounts, gymnosperms (likely Podocarpus) and some spores. Spore concentrations are low, and we note that mosses are extant on Antarctica in the more humid regions such as the Antarctic Peninsula, whereas other plants are locally extinct. The plant community of the late Eocene is low diversity and its composition does not appreciably change across the 3 Myr period including the EOT, suggesting that southern beech-podocarp tundra assemblages, was the last surviving vestiges of forests of the same composition. Like the pollen, the biomarkers in the same sediments also carry components of penecontemporary and reworked contributions that can be differentiated. We do not suspect any thermal alteration of the n-alkanoic acids due to their “fresh” molecular abundance distributions. Variability in δD values between chain lengths and downcore (Fig. A.6), would also tend to be lost during any thermal alteration that drives H-exchange (Schimmelmann et al., 2006), thus variable δD values further support our interpretation of penecontemporary and not reworked n- alkanoic acids. We avoid low abundance, mature alkanes as they are unlikely to be penecontemporary, and as they may have undergone H-exchange during maturation. Given their low abundance they are unlikely to be useful indicators of reworking (unlike the reworked pollen), with the exception of 88 one sample at 34.8 Ma, which has the highest abundance of short and mid chain n-alkanes, no chain lengths >C25, an uncharacterized complex mixture, and this is the same sample where microscopy identified substantial coal or charcoal - likely a deposit of a remobilized coal eroded from land. Reworked n-alkanes have been also reported in Neogene age sediments on the Antarctic margin in the Ross Sea (Duncan et al., 2019). The presence of thermally mature hopanes confirms the likely thermal alteration of the alkanes on land and remobilization by erosion, with mineral-associated alkanes and hopanes in sediments that had already lost alkanoic acids during maturation. We do not suspect in situ thermal alteration given that the n-alkanoic acids have a fresh molecular abundance distribution as may be expected in unaltered marine sediments in a terrestrial margin setting. Can the GDGTs be disturbed by reworking noted in the microfossil and alkanes fractions? By definition, the samples that have marine-dominated GDGT production are free from concerns of glacial erosion of terrestrial sediments. GDGT fractions with high terrestrial soil inputs (high BIT), likely derived from penecontemporary soils in the catchment. We exclude the possibility of reworked GDGT signals, on the basis that 1) the Permian terrestrial pollen-bearing coal sample contained no measurable br- or iso-GDGTs, concentrations (below detection), and 2) we are aware of no prior reports of Permian GDGT preservation. Thus, these brGDGT biomarkers, and proxy information they contain, are less likely to be influenced by reworked sedimentary erosion than mineralogical proxies, as for example kaolinite in the clay size fraction has been shown to be a reworked mineral from earlier greenhouse climate states (Passchier et al., 2017). 89 3.4.2 Terrestrial inputs and marine productivity at continental margins The membrane lipid biomarkers of bacteria and archaea provide an indicator of terrestrial influx from soils and of marine productivity. Terrestrial vs. marine influxes are commonly tracked via the BIT index and variations in the BIT index between the sites may be partly due to their proximity to shore, but any such differences are clearly overwhelmed by changes between lithological facies within each segment (Fig. A.2). The BIT variations instead reflect facies, responding to the changing depositional environment, the changing terrestrial export and marine productivity (Fig. A.3). Site 1166 is the shallowest site and closest to shore but sees a large shift in BIT associated with lithological changes. Site 742 has high and stable BIT values. Site 739 has similar BIT at the outset, despite being farthest offshore there is a shift from terrestrial to marine dominance at the EOT. As glaciers advance, shorelines could shift associated with glacial isostatic and eustatic changes (Hochmuth, Gohl, et al., 2020; Paxman et al., 2019), we would expect an initial regression followed by a transgression with more marine influence at this site (Stocchi et al., 2013). However, the BIT ratio here responds more to variations in the isoGDGT flux, and thus it may be more informative to look at the ΣbrGDGT and ΣisoGDGT concentrations independently (Fig. A.3). While ΣbrGDGTs are fairly invariant there are interesting features in the ΣisoGDGT concentrations (Fig. 3.3E) that suggest variations in marine productivity (Fietz et al., 2011). 90 Fig 3.4. A detailed view of microfossil and biomarker evidence for change in East Antarctica across the EOT-1 and -2. A) Glacial conditions in the Gamburtsev Mountains and B) Prydz Bay stratigraphy (legend as in Fig. 4), C) pollen concentration (counts/gram of dry sediment), D) ΣisoGDGT concentration, E) BayMBT based MAAT, F) MAP from CIA-K index, G) soil pH, and comparison to H) Kerguelen Plateau εNd records from Site 738 (Scher et al., 2011) and I) δ 18 Obenthic (sources as in Fig. 3). Labels indicate EOT-1 and -2 (glacial expansion) and EOGM (formerly Oi-1; glaciation) features in I. Purple line highlights the diamict to diatomite depositional transition in Prydz Bay. 3.4.3 Late Eocene warmth between PrOM and the EOT Comparison data from the Kerguelen Plateau from the εNd in fossil fish teeth (Fig. 3.3L) indicates two major glacial expansions, and corresponding erosion pulses, at 37.5 Ma for the PrOM (Scher et al., 2014) and at 33.7 Ma for the EOT (Scher et al., 2011). Between the PrOM and the EOT glacial expansions, oxygen isotopes in benthic foraminifera are generally <2.0‰ and fairly invariant indicating warmth and reduced ice volume lasting several million years (Fig. 3.3K). In the available records from Prydz Bay we find evidence for warm conditions on land and marine proxies with some abrupt (<50 ka) warmings at 35.7 and 34.7 Ma within the generally warm two million years from 35.7 to 33.7 Ma (Fig. 3.3C-J). 91 Using the BayMBT temperature proxy for soil biomarkers, we find late Eocene MAAT averages 14°C with a peak temperature of 20°C (a warming of 4°C above background) at 35.7 Ma (Fig. 3.3F). Late Eocene warmth was previously identified by the weathering proxy (S-index), showing average temperatures of 10°C, and abrupt warming to a peak temperatures of 12°C at 35.7 Ma (Passchier et al., 2017). During the warm spell at 35.7 Ma, we find the proportion of penecontemporaneous pollen increases to a maximum of 12x10 3 counts/gdw or 90% of all pollen (Fig. 3.3D), much higher than elsewhere in the record (typically <1x10 3 counts/gdw and <20%) suggesting a flourishing of plants in that warm interlude in particular among the Nothofagus species (Fig. 3.3C). Both branched and isoprenoidal GDGT concentrations increase at that time suggesting a time period of high productivity. This warming event is also supported by the mineralogy of the Prydz Bay cores with kaolinite dominating (>50%) from 36–34.4 Ma (Fig 3.3b) (Forsberg et al., 2008). The additional independent temperature estimates from GDGTs and the pollen record adds additional evidence of a warming interval in the late Eocene around 35.7 Ma prior to the glacial transition. Around 35.0 Ma there is a gradual warming detected in the BayMBT which peaks at 17°C whereas the (cool-offset) S-index spikes to 12°C in what looks to be an abrupt warming (Passchier et al., 2017) and penecontemporary pollen inputs increase at 34.7 Ma to 3.5x10 3 counts/gdw. There is also an increase in δDwax of 10‰ consistent with warming and/or drying. The timing of the warming is estimated by the linear age model to be ~34.7 Ma, but is not well constrained (Fig. 3.2). At the warm events at 35.7 and 34.7 Ma, we observe spikes in the ΣisoGDGT concentration suggesting marine production increases (though not as persistent as observed at the EOT). In these events we find land plants flourish, chemical weathering increases, and soil bacterial lipid 92 biomarkers indicate a 3–6°C rise in ambient air temperatures over ~100 ka for 35.7 Ma and a 3°C warming over ~500 ka for 34.7 Ma. The rise in marine productivity in response to temperature increases, may suggest an increase in fluvial runoff and nutrient supply to the coast promoting marine production during warmer and wetter times for eastern Antarctica. In this the marginal setting, soil biomarkers inputs detected by the BIT and ΔRI preclude ocean temperature reconstructions in most samples. Contrasting available BAYSPAR estimates before and after the EOT at 34.2 Ma and 33.5 Ma, we find SSTs cool by ~6°C from 14 to 8°C (Fig. 3.3F). The absolute values of these Prydz Bay SSTs are comparable to model estimates of <10°C for the Southern Ocean during the Priabonian (37.8–33.9 Ma, pre-EOT) conducted using the HadCM3L model at 4 x preindustrial atmospheric pCO2 (Inglis et al., 2015) as well as ocean temperature reconstructions estimating a 4°C cooling from 13–9°C on the Falkland Plateau abased on UK ' 37 and TEX86 (Houben et al., 2019; Liu et al., 2009). After the EOT, the Prydz Bay SSTs rebound slightly (to 10–12°C) between 33.3–33.2 Ma, and there is a small increase in penecontemporaneous pollen concentration. A post-EOT warming of 1–2°C was previously suggested in the Southern Ocean from the Kerguelen Plateau based on planktic foraminifera Mg/Ca (Bohaty et al., 2012). 3.4.4 Hydroclimate of the late Eocene We assessed plant wax abundance, ẟ 13 C and ẟD (Fig. 3.3G and H). The ẟ 13 C of the C30 n- alkanoic acid ranged from –31.9 to –28.6‰, with a 3.3‰ positive trend which is much greater than the 0.4–0.6‰ reconstructed change in atmospheric ẟ 13 C during the same time period (Tipple et al., 2010), suggesting an overall drying. The ẟD of the C30 n-alkanoic acid varied from –245 to –206‰, a c. 40‰ range across the EOT the δDwax shifts from –233‰ to –209‰ (+24‰) 93 from 33.7 to 33.5 Ma (Fig. 3.3H). The hydrogen isotopic composition of plant wax records the isotopic composition of rainfall, subject to a large biosynthetic fractionation. The plant wax n- alkanes and n-alkanoic acids have been calibrated in a variety of grass, shrub and forested ecosystems at high latitudes in the Arctic (McFarlin et al., 2019). The most relevant of these the calibration are those of plant wax n-alkanoic acids in a moss-conifer ecosystem archived in soils (Bakkelund et al., 2018) and archived in Greenland lakes (McFarlin et al., 2019) as they are the same compound class used here. However, plant wax hydrogen isotopic fractionations were not calibrated in austral plant communities that have been locally extinct from the continent of Antarctica for at least fifteen million years (Warny et al., 2009). Although some of these late Eocene Antarctic species are today found in Tasmania and Patagonia, they have not yet been calibrated in terms of isotopic composition. However, a recent study of several temperate forest taxa in Chile (Cerda-Peña et al., 2020), reports n-alkyl lipid concentrations in Nothofagus dombeyi, and notes that n-alkanoic acids dominate over n-alkanes, unlike in most North American angiosperms, finding C22–C32 n-alkanoic acids with the expected even over odd dominance and with the maximum abundance in C28. This helps to establish Nothofagus spp., represented in the pollen in Prydz Bay, as likely producers of the plant wax n-alkanoic acids in the same samples, given their similar molecular abundance distributions, with additional contributions expected from the gymnosperms. Estimating δDprecip using the Arctic and globally-assessed fractionation of –99‰ (McFarlin et al., 2019) leads to estimates of δDprecip ranging from –162 to –104‰ across the Prydz Bay record. As expected, these values from the relatively warm Eocene are much more enriched compared to modern snow samples from glaciated Antarctica adjacent to Prydz Bay with snow samples from the East Antarctic Ice Sheet that range from –219 to –448‰ across an elevation transect of 94 1040–4093m and from 46 to 1183 km from the coast (Masson-Delmotte et al., 2008) (Fig. 3.1A). The variability within the late Eocene record is not clearly mirroring other proxies in the reconstruction, and may represent variability in atmospheric conditions and thus precipitation isotopes, changes in plant communities and their fractionation or catchment sourcing in ways that are not presently understood, we thus do not interpret the variability in much of the plant wax record. Although it is not an abnormal shift among the variability in the record, there is a shift associated with the EOT: δDprecip shifted from –149 to –122‰. This shift does appear to correspond to expected environmental changes. A similar 20‰ shift was estimated from the hydration waters of volcanic glass from the Sarmiento Formation, Patagonia (Colwyn & Hren, 2019). This shift occurs at the time of the ~+1.5‰ δ 18 Obenthic shift identified from Kerguelen Plateau sites (738, 744, and 748) at 33.7 Ma indicating enrichment in 18 O-seawater associated with glacial expansion. Globally a 1.5‰ increase in δ 18 Obenthic is registered, 0.9‰ of this is thought to reflect a decrease in ocean bottom water temperatures (Lear et al., 2008) and 0.6‰ is a change in seawater due to the buildup of continental ice sheets. Based on the seawater change, and mass dependent fractionation, we expect a D-enrichment in seawater of 5‰, which should lead to a commensurate 5‰ D-enrichment in meteoric water and plant wax. This would only explain part of the +20‰ shift, leaving the remainder as a sign of the enrichment of δDprecip. As we can rule out warming, based on numerous temperature reconstructions across the glacial expansion discussed previously, we infer the positive shift in δDprecip must represents increasing aridity, similar to interpretations of δDwax in the latest Eocene records from the Antarctic Peninsula (Feakins et al., 2014). In the same location, corroborating information from increasing δ 13 C of Nothofagus pollen grains were interpreted as signs of moisture stress in vegetation communities (Griener et al., 2013). Here too, aridity is also consistent with an increase in δ 13 Cwax 95 of +1.1‰ in conjunction with the increase in δDwax. While there are additional possible influences on δ 13 Cwax including atmospheric δ 13 C, estimates based on δ 13 C of planktonic foraminifera suggest variations of no more than 0.4–0.6‰ in the latest Eocene (Tipple et al., 2010). We also find no evidence for a major change in vegetation assemblage, such as could drive a change in fractionation of either carbon or hydrogen isotopes. The vegetation is dominantly Southern beech (Fig. 3.3C), with declining penecontemporaneous pollen abundances between a peak of 12.2x10 3 counts/gdw at 35.7 Ma with a steep decline and then low levels with the exception of 34.7 Ma (Fig. 3.3D). A Student’s t test indicates a significant decline in penecontemporaneous pollen when comparing the record before and after the EOT, however there is no strong change associated with the main ice expansion, instead the decline occurs with the earlier cooling and drying, detected by the mineralogical proxies after 35.7 and 34.7 Ma. The decline in pollen amounts indicate deteriorating conditions for tree growth, consistent with a cooling and drying climate. Penecontemporary pollen concentration decreased by an order of magnitude from highs of 12x10 3 counts/gdw at 35.7 Ma to mostly <1x10 3 counts/gdw thereafter with a low of 0.2 x10 3 counts/gdw at 33.5 Ma (Fig. 3.3D). Perhaps the strongest correlating change is to the mineralogical proxies showing a decline in MAP (Fig. 3.3J) from a high of 1200 mm to below 900 mm, as the penecontemporaneous pollen counts decline well before the EOT. MAP and penecontemporaneous pollen were observed to have a positive correlation (r=0.47, p<0.05) based on non-parametric methods that account for serial correlation (Ebisuzaki, 1997). MAP continues to drop to as low as 700 mm after the EOT as expected in a glacial climate state. These MAP estimates derive from the correlation to precipitation based on North American soils (Sheldon et al., 2002), as applied to the CIA-K index, a moisture responsive subset of the bulk 96 detrital geochemical oxides that together are responsive to weathering. Previously the S-index, based on the molar ratio of Na2O and K2O to Al2O3, was used to infer gradual cooling from ~10°C before 35 Ma to a low of 7°C at 33.2 Ma (Passchier et al., 2017). As warm and wet conditions drive stronger weathering and cooler and drier conditions drive lesser weathering these two proxies are related and moisture has a stronger correlation than temperature in the modern dataset. We report both moisture and temperature interpretations here. The drying pattern can be compared to the soil pH estimates from the CBT′ index, (Fig 3.3I) which shows a highly variable record in the late Eocene, with acidic conditions resulting from lowland soils during the warm and wet events described previously, as well as less acidic signals from drier conditions during colder times or upland locations as alpine glaciers initiate and expand in the catchment. In the late Eocene the temperatures from the weathering proxy are cool-biased relative to the soil biomarker proxy and experience a more gradual cooling trend. This may also reflect catchment sourcing with rock erosion in the Gamburtsev Mountains as glaciers grow. After the EOT, when ice sheets cover the upper catchments, the absolute temperatures estimates are the same within uncertainties for the S-index and BayMBT (Fig. 3.3F), suggesting a more consistent lowland source, with similar temperatures recorded. 3.4.5 Glacial expansion at the EOT Next, we examine the changes across the EOT recorded by several of the microfossil and biomarker proxies in Prydz Bay sediments (Fig. 3.4), to assess their timing relative to glacial expansions of the EOIS detected in the δ 18 Obenthic (Hutchinson et al., 2021). For context from prior work in the region, Apatite Fission Track dating indicates maximum erosion from 34.2 to 33.8 Ma (Tochilin et al., 2012) across the EOIS as glaciers incised and eroded the formerly fluvial drainage basin of Gamburtsev Mountains and the Lambert Graben into the sediments of 97 Prydz Bay, before cold-based glacial conditions were established at 33.8 Ma (Rose et al., 2013; Passchier et al., 2017, Fig. 3.4A). Depositional systems in Prydz Bay transition from prograding diamictites to diatomaceous sediments a little later at 33.7 Ma (Fig. 3.4B). Further offshore, the Kerguelen Plateau receives the εNd signal of glacial surge associated with EOT-1 and -2 (Fig. 3.4H; Scher et al., 2011) and the glacial ice volume and cooling of bottom waters are reflected in the benthic oxygen isotope record (Fig. 3.4I). Here, we find the persistence, but low abundance, of penecontemporary pollen across the full EOT (Fig. 3.4C), suggesting southern beech/podocarp vegetation were already restricted to refugia and were not eliminated as glaciers advanced to maximal extent at the EOGM (formerly Oi-1). The glacial surge of EOIS led to an increase in erosion of reworked pollen at 33.7 Ma and then a peak (12x10 3 counts/gdw) from 33.6–33.5 Ma perhaps indicating maximum glacial erosion, with sustained erosion of reworked pollen through 33.1 Ma. Soil biomarkers indicate that MAAT cooled by 6°C abruptly at 33.8 Ma from an average of ~14°C before to 8–10°C thereafter (Fig. 3.4D). This cooling occurred ~100 ka prior to the main inflection in the marine oxygen isotope curve of the EOIS, indicating that cooling of the catchment in East Antarctica preceded the continental-scale glacial ice volume and bottom water cooling. This makes sense as the Gamburtsev Mountains were the locus of glacial initiation (Rose et al., 2013) with valley glaciers descending via the Lambert Graben (Tochilin et al., 2012), which would have cooled the local catchment, before the continent-wide signals detected in the deep sea. The timing of the cooling appears earliest in the mineralogical proxies, which likely carry higher elevation erosion signals, and then abruptly in the soil biomarkers at the same time as glacial reconstructions suggest a shift to cold-based glaciers, suggesting ice-albedo feedbacks led to further cooling, including the lowland reaches of the catchment. The magnitude 98 of this cooling is consistent with climate model experiments, that decrease carbon dioxide concentrations from 750 ppmv pre-EOT and 560 ppmV post-EOT to achieve EOT cooling and suggest a drop of 5°C across the EOT using the 2011 GENESIS 3.0 GCM combined with a BIOME4 equilibrium vegetation model (Feakins et al., 2014). This cooling incorporates responses to the decrease in atmospheric CO2 as well as the ice-albedo feedback. The soil biomarkers further indicate that cooling was accompanied by drying. Soil pH shifted from 6.0 to 6.5, with less acidic soils denoting reduced plant and microbial activity to acidify the soil and known to be consistent with drying. The shift is abrupt and may be linked to abrupt changes in climate and/or erosion associated with the abrupt expansion of the cryosphere (Fig. 3.4H, I). Quantitative estimates of a reduction in mean annual precipitation can be calculated from the detrital mud geochemistry (CIA-K index, Passchier et al., 2013) using the calibration of Sheldon et al, (2002) to suggest gradually declining MAP from 900 to 700 mm across the EOT from 34.2 to 33.5 Ma. These reconstructions constitute some of the few estimates for terrestrial conditions in the latest Eocene and add to the available evidence for proxy-model comparisons (Deep-MIP-EOT; Hutchinson et al., 2021). While the four climate models used different forcing and geography and diverge greatly in their absolute temperature in the interior of the Antarctic continent, their marginal temperature estimates for the pre- and post-EOT simulations are fairly consistent with each other. In their proxy-model comparison they showed no proxy data from Antarctica, suggesting that their Antarctic predictions are untested and that this constitutes some of the first proxy constraints on absolute and relative temperature change to validate climate model experiments for the greenhouse-icehouse transition. Their climate model ensemble estimated a 3°C surface air temperature cooling at the edge of the Antarctic continent, which is the same 99 (within uncertainties) as the magnitude of cooling observed, from both MAAT over land (from brGDGTs) and SSTs (from TEX86) from our biomarker reconstructions using their broad integration windows pre- and post-EOT. If we compare our two temperature proxies in Prydz Bay sediments in detail, the soil biomarker- based cooling and drying is larger and more abrupt than reconstructed from weathering proxies and this may reflect the differences in the catchment sourcing of organic and inorganic proxies. The biomarker approach finds a step cooling of 6°C within 50 ka at 33.8 Ma, whereas the weathering proxy records a cooling of half that magnitude gradually across one million years. The organic proxy records a step-shift in soil pH whereas the weathering proxy records gradual change. These offsets may reflect the different response times of soil organic microbial activity and parent rock weathering rates, or differential organic-inorganic sourcing from the drainage basin to the offshore archive. Catchment studies elsewhere show that erosion rates tend to have upland dominance (Scher et al., 2014; van de Flierdt et al., 2008), leading to expectations of a cold-bias for the site of deposition. Early erosional and mechanical weathering signals are recorded for the inorganic proxies due to upland alpine glaciation (Rose et al., 2013). In comparison, lowland dominance is broadly reported in catchment studies of biomarkers (Freymond et al., 2018; Galy et al., 2011) based on areal extent and degradation in transit; therefore, a more proximal and rapidly responding biomarker signal is expected as conditions change. Whereas inorganic weathering proxies track aggregate conditions in the entire catchment, including uplands, biomarkers may not record the early phases of alpine glaciation. However, as ice sheets expand, the ice-albedo and ice-elevation feedbacks would have altered the local climate, including the lowlands, more strongly. The abrupt biomarker and microfossil changes recorded in Prydz Bay (Fig. 3.4) appear to coincide with geochemical evidence for a 100 glacial surge at EOT-2 recorded in εNd at the Kerguelen Plateau that marks the influence of the expanding EAIS on ocean chemistry (Scher et al., 2011), however additional corroboration of the relative timing of landscape and erosional shifts would benefit from those analyses in the same sediments. Between 33.8 and 33.7 Ma, εNd decreases by around 2.5 (Fig. 3.4H) indicating glacial erosion of radiogenic rocks influencing marine chemistry with both EOT-1 and 2 detected with timing matching that of the δ 18 Obenthic record of glacial ice volume expansion and deep ocean cooling (Fig. 3.4I). A change in marine productivity also occurs in the ocean waters of Prydz Bay. We find sedimentary concentrations of ΣisoGDGTs are an order of magnitude more abundant after 33.7 Ma compared to consistently low abundances before (Fig. 3.4D). This increase in marine archaeal productivity corresponds to a major depositional system change from prograding diamictite to diatomaceous mudstones (Fig. 3.4A) at EOT-2 (Fig. 3.4I). Increased diatom productivity along with other changes in marine ecosystems are reported around the Southern Ocean at the EOT (Houben et al., 2013). The detected marine ecological shift and productivity increase at the EOT in Prydz Bay may be a response to regional ocean circulation changes with enhanced upwelling and sea ice formation (Houben et al., 2019), resulting in enhanced overturning, resulting in changing bottom water (Fig. 3.4I). Climate model experiments show that an increase in sea ice, following cooling linked to atmospheric carbon dioxide changes, could stimulate deep water production and Southern Ocean circulation (Goldner et al., 2014). Local effects from the glacial surge detected in the Kerguelen Plateau (Fig. 3.4H), may have also locally stimulated increased marine productivity in Prydz Bay, as marine-terminating glaciers in Greenland today are noted as sources of nutrients that increase marine productivity (Meire et al., 2017). However, the glacial surges of EOT-1 and 2 (Fig. 3.4G) predates the main marine 101 productivity increase (Fig. 3.4C) which appears to coincide with EOGM and glacial volume, suggesting the connection to a larger regional phenomenon associated sea ice and ocean overturning. Changes in ocean circulation associated with the transition to icehouse conditions may explain fertilization of marine productivity that continues well beyond the glacial erosion surge. It then follows that a slight decrease in ΣisoGDGT concentrations occurs at 33.2 Ma, associated with a reduction in EAIS size (Galeotti et al., 2016) visible in the oxygen isotope record (Fig. 3.4I), presumably linked to reduced sea ice formation and ocean overturning. This is consistent with evidence for 1–2°C of warming based on planktic foraminiferal Mg/Ca from the Kerguelen Plateau (Bohaty et al., 2012). Locally we find Prydz Bay SSTs warm slightly to 10– 12°C (Fig. 3.4E) between 33.3–33.2 Ma, and there is a small increase in penecontemporaneous pollen concentration including spores suggesting slightly wetter conditions (Fig. 3.4C) associated with regional oceanic changes and the reduction in EAIS extent. 3.5 Conclusions Latest Eocene sediments from ODP Sites 739, 742 and 1166, from Prydz Bay, Antarctica encompassing the late Eocene through the EOT provide records of both terrestrial and marine climate change. Prior mineralogical evidence from Prydz Bay indicates a gradual cooling and drying in the catchment since the last warm pulse at 34.7 Ma (Passchier et al., 2017). The East Antarctic Ice Sheet began with cirque glaciers in the Gamburtsev Mountains (Rose et al., 2013), with maximum erosion via the Lambert Graben as ice sheets surged (Tochilin et al., 2012), before the transition to cold-based glaciers as glaciation proceeded. Prior geochemical evidence from further offshore in the Kerguelen Plateau indicates marine-terminating glaciers reach the coastline by 33.7 Ma. 102 In the lead up to the icehouse transition, the biomarker and pollen results from this study support the presence of a 3–6°C warming in the late Eocene at 35.7 Ma, followed by another warm spike at 34.7 Ma before the beginning of the EOT. The presence of penecontemporary pollen during and after the EOT suggest southern beech/podocarp vegetation refugia, with a vegetation community struggling to survive, as noted by concentration in pollen decreasing from 3.5x10 3 counts/gdw during the warm pulse at 34.7 Ma declining to ~0.4 x10 3 counts/gdw before and during the EOT. The declines in penecontemporary pollen correlate with mineralogical evidence for cooling and drying in the catchment. The plant wax shifts in δD and δ 13 C (+20‰ and +1.1‰) also suggest an increase in aridity, showing signs that drying was felt by plants. The biomarker proxies record an abrupt cooling 100 ka before the EOT, with terrestrial brGDGTs showing a MAAT decrease of 6°C, with a commensurate magnitude of cooling of SSTs based on available TEX86. The same soil biomarkers also show an abrupt drying accompanied that cooling (based on soil pH). The soil biomarker transition is notably more abrupt and later than the catchment mineralogical evidence likely reflecting differences in catchment averaging of the two signals with earlier upland cooling detected in the rock derived signal and lowland change occurring later as detected in the soil biomarkers. Marine biomarkers detect cooling of SSTs from 14 to 8°C associated with the EOT although the timing of the SST drop is poorly resolved as high terrestrial inputs obscure the proxy record in many samples with high BIT in this marginal setting. The ΣisoGDGT concentrations are a compelling aspect of the biomarker contribution to the Prydz Bay reconstructions, they indicate increased marine productivity after the EOT suggesting increased nutrient supply into the Southern Ocean with intensification of ocean circulation at the EOT. 103 Acknowledgements We declare no financial conflicts of interests for any author or their affiliations. This research was funded by the U.S. National Science Foundation NSF-OPP-1908548 to SJF, NSF-OPP- 1908399 to HDS, and NSF-OPP-1743643 to SP. This research used samples and/or data collected by the International Ocean Discovery Program, supported by funding from the U.S. National Science Foundation and other member nations. Thanks to Peter Flaig for providing a sample of Permian-age coal. This manuscript was improved with the constructive comments of Carlos Jaramillo, Aaron Diefendorf and Gordon Inglis. Data files are archived at the NOAA paleoclimatology database at https://www.ncdc.noaa.gov/paleo/study/32052 (Tibbett et al., 2021). 104 References Anagnostou, E., John, E. H., Edgar, K. M., Foster, G. L., Ridgwell, A., Inglis, G. N., et al. (2016). Changing atmospheric CO2 concentration was the primary driver of early Cenozoic climate. Nature, 533, 380. https://doi.org/10.1038/nature17423 Anderson, J. B., Warny, S., Askin, R. A., Wellner, J. S., Bohaty, S. M., Kirshner, A. E., et al. (2011). Progressive Cenozoic cooling and the demise of Antarctica’s last refugium. Proceedings of the National Academy of Sciences, 108(28), 11356. https://doi.org/10.1073/pnas.1014885108 Ashworth, A., Lewis, A., Marchant, D., Askin, R., Cantrill, D., Francis, J., Leng, M., Newton, A., Raine, J., & Williams, M. (2007). The Neogene biota of the Transantarctic Mountains. Online Proceedings of the ISAES X, edited by A.K. Cooper and C.R. Raymond et al., USGS Open-File Report 2007–1047, Extended Abstract 071, 4 p. Askin, R. A., Elliot, D. H., Stilwell, J. D., & Zinsmeister, W. J. (1991). Stratigraphy and paleontology of Campanian and Eocene sediments, Cockburn Island, Antarctic Peninsula. Journal of South American Earth Sciences, 4(1), 99–117. https://doi.org/10.1016/0895- 9811(91)90021-C Bakkelund, A., Porter, T. J., Froese, D. G., & Feakins, S. J. (2018). Net fractionation of hydrogen isotopes in n-alkanoic acids from soils in the northern boreal forest. Organic Geochemistry, 125, 1–13. https://doi.org/10.1016/j.orggeochem.2018.08.005 Barron, J., Larsen, B., & et al. (Eds.). (1991). Proceedings of the Ocean Drilling Program, 119 Scientific Results (Vol. 119). Ocean Drilling Program. https://doi.org/10.2973/odp.proc.sr.119.1991 Bohaty, S. M., & Zachos, J. C. (2003). Significant Southern Ocean warming event in the late middle Eocene. Geology, 31(11), 1017–1020. https://doi.org/10.1130/G19800.1 Bohaty, S. M., Zachos, J. C., & Delaney, M. L. (2012). Foraminiferal Mg/Ca evidence for Southern Ocean cooling across the Eocene–Oligocene transition. Earth and Planetary Science Letters, 317–318, 251–261. https://doi.org/10.1016/j.epsl.2011.11.037 Bohaty, S. M., Zachos, J. C., Florindo, F., & Delaney, M. L. (2009). Coupled greenhouse warming and deep-sea acidification in the middle Eocene. Paleoceanography, 24(2). https://doi.org/10.1029/2008PA001676 Brown, C. A., Riding, J. B., & Warny, S. (2008). Palynological techniques. American Association of Stratigraphic Palynologists Foundation; 2nd ed., 137 pp., AASP Special Publications. Cerda-Peña, C., Contreras, S. & Rau, J. R. (2020). Molecular n-alkyl leaf waxes of three dominant plants from the temperate forest in South America. Organic Geochemistry, 149, 104105. 105 Chen, X., Liu, X., Wei, Y., & Huang, Y. (2019). Production of long-chain n-alkyl lipids by heterotrophic microbes: New evidence from Antarctic lakes. Organic Geochemistry, 103909. https://doi.org/10.1016/j.orggeochem.2019.103909 Colwyn, D. A., & Hren, M. T. (2019). An abrupt decrease in Southern Hemisphere terrestrial temperature during the Eocene-Oligocene transition. Earth and Planetary Science Letters, 512, 227–235. https://doi.org/10.1016/j.epsl.2019.01.052 Coxall, H. K., & Pearson, P. N. (2007). The Eocene-Oligocene transition. In Deep time perspectives on climate change: marrying the signal from computer models and biological proxies, edited by M. Williams, A. M. Haywood, F. J. Gregory & D. N. Schmidt, Geological Society of London 351–387. https://doi.org/10.1144/TMS002.16 Cramer, B. S., Toggweiler, J. R., Wright, J. D., Katz, M. E., & Miller, K. G. (2009). Ocean overturning since the Late Cretaceous: Inferences from a new benthic foraminiferal isotope compilation. Paleoceanography, 24(4). https://doi.org/10.1029/2008PA001683 De Jonge, C., Hopmans, E. C., Zell, C. I., Kim, J.-H., Schouten, S., & Sinninghe Damsté, J. S. (2014). Occurrence and abundance of 6-methyl branched glycerol dialkyl glycerol tetraethers in soils: Implications for palaeoclimate reconstruction. Geochimica et Cosmochimica Acta, 141, 97–112. https://doi.org/10.1016/j.gca.2014.06.013 Dearing Crampton-Flood, E., Tierney, J. E., Peterse, F., Kirkels, F. M. S. A., & Sinninghe Damsté, J. S. (2020). BayMBT: A Bayesian calibration model for branched glycerol dialkyl glycerol tetraethers in soils and peats. Geochimica et Cosmochimica Acta, 268, 142–159. https://doi.org/10.1016/j.gca.2019.09.043 DeConto, R. M., Pollard, D., Wilson, P. A., Palike, H., Lear, C. H., & Pagani, M. (2008). Thresholds for Cenozoic bipolar glaciation. Nature, 455(7213), 652–656. https://doi.org/10.1038/nature07337 Delmotte, M. (1997). Enregistrements climatiques à Law Dome: Variabilité pour les périodes récentes et pour la déglaciation. PhD Thesis. Glaciologie. Université Joseph-Fourier - Grenoble I. Français. https://tel.archives-ouvertes.fr/tel-00759854/ Diefendorf, A. F., Leslie, A. B., & Wing, S. L. (2015). Leaf wax composition and carbon isotopes vary among major conifer groups. Geochimica et Cosmochimica Acta, 170, 145– 156. https://doi.org/10.1016/j.gca.2015.08.018 Douglas, P. M. J., Affek, H. P., Ivany, L. C., Houben, A. J. P., Sijp, W. P., Sluijs, A., et al. (2014). Pronounced zonal heterogeneity in Eocene southern high-latitude sea surface temperatures. Proceedings of the National Academy of Sciences, 111(18), 6582. https://doi.org/10.1073/pnas.1321441111 Duncan, B., McKay, R., Bendle, J., Naish, T., Inglis, G. N., Moossen, H., et al. (2019). Lipid biomarker distributions in Oligocene and Miocene sediments from the Ross Sea region, Antarctica: Implications for use of biomarker proxies in glacially-influenced settings. 106 Palaeogeography, Palaeoclimatology, Palaeoecology, 516, 71–89. https://doi.org/10.1016/j.palaeo.2018.11.028 Ebisuzaki, W. (1997). A Method to Estimate the Statistical Significance of a Correlation When the Data Are Serially Correlated. Journal of Climate, 10(9), 2147–2153. https://doi.org/10.1175/1520-0442(1997)010<2147:AMTETS>2.0.CO;2 Erohina, T., Cooper, A., Handwerger, D., & Dunbar, R. (2004). Seismic stratigraphic correlations between ODP Sites 742 and 1166: Implications for depositional paleoenvironments in Prydz Bay, Antarctica. Proceedings of the Ocean Drilling Program: Scientific Results, 188, 1–21. https://doi.org/10.2973/odp.proc.sr.188.2004 Feakins, S. J., Warny, S., & DeConto, R. M. (2014). Snapshot of cooling and drying before onset of Antarctic Glaciation. Earth and Planetary Science Letters, 404, 154–166. https://doi.org/10.1016/j.epsl.2014.07.032 Feakins, S. J., Warny, S., & Lee, J.-E. (2012). Hydrologic cycling over Antarctica during the middle Miocene warming. Nature Geoscience, 5(8), 557–560. https://doi.org/10.1038/NGEO1498 Fietz, S., Martínez-Garcia, A., Rueda, G., Peck, V. L., Huguet, C., Escala, M., & Rosell-Melé, A. (2011). Crenarchaea and phytoplankton coupling in sedimentary archives: Common trigger or metabolic dependence? Limnology and Oceanography, 56(5), 1907–1916. https://doi.org/10.4319/lo.2011.56.5.1907 Forsberg, C., Florindo, F., Gruetzner, J., Venuti, A., & Solheim, A. (2008). Sedimentation and aspects of glacial dynamics from physical properties, mineralogy and magnetic properties at ODP Sites 1166 and 1167, Prydz Bay, Antarctica. Palaeogeography, Palaeoclimatology, Palaeoecology, 260(1–2), 184–201. https://doi.org/10.1016/j.palaeo.2007.08.022 Freymond, C. V., Kündig, N., Stark, C., Peterse, F., Buggle, B., Lupker, M., et al. (2018). Evolution of biomolecular loadings along a major river system. Geochimica et Cosmochimica Acta, 223, 389–404. https://doi.org/10.1016/j.gca.2017.12.010 Galy, V., Eglinton, T., France-Lanord, C., & Sylva, S. (2011). The provenance of vegetation and environmental signatures encoded in vascular plant biomarkers carried by the Ganges– Brahmaputra rivers. Earth and Planetary Science Letters, 304(1), 1–12. https://doi.org/10.1016/j.epsl.2011.02.003 Gradstein, F. M., Ogg, J. G., Schmitz, M., & Ogg, G. (2012). The Geologic Time Scale 2012. The Geologic Time Scale 2012, 1144 pp., Elsevier, Boston. Griener, K. W., Nelson, D. M., & Warny, S. (2013). Declining moisture availability on the Antarctic Peninsula during the Late Eocene. Palaeogeography, Palaeoclimatology, Palaeoecology, 383, 72–78. http://dx.doi.org/10.1016/j.palaeo.2013.05.004 Griener, K. W., Warny, S., Askin, R., & Acton, G. (2015). Early to middle Miocene vegetation history of Antarctica supports eccentricity-paced warming intervals during the Antarctic 107 icehouse phase. Global and Planetary Change, 127, 67–78. https://doi.org/10.1016/j.gloplacha.2015.01.006 Harrowfield, M., Holdgate, G. R., Wilson, C. J., & McLoughlin, S. (2005). Tectonic significance of the Lambert Graben, East Antarctica: Reconstructing the Gondwanan rift. Geology, 33(3), 197–200. https://doi.org/10.1130/G21081.1 Hill, D. J., Haywood, A. M., Valdes, P. J., Francis, J. E., Lunt, D. J., Wade, B. S., & Bowman, V. C. (2013). Paleogeographic controls on the onset of the Antarctic circumpolar current. Geophysical Research Letters, 40(19), 5199–5204. https://doi.org/10.1002/grl.50941 Hochmuth, K., Gohl, K., Leitchenkov, G., Sauermilch, I., Whittaker, J., Uenzelmann-Neben, G., et al. (2020). The evolving paleobathymetry of the circum‐Antarctic Southern Ocean since 34 Ma – a key to understanding past cryosphere‐ocean developments. Geochemistry, Geophysics, Geosystems. https://doi.org/10.1029/2020GC009122 Hochmuth, K., Paxman, G. J. G., Gohl, K., Jamieson, S. S. R., Leitchenkov, G. L., Bentley, M. J., et al. (2020). Combined palaeotopography and palaeobathymetry of the Antarctic continent and the Southern Ocean since 34 Ma. PANGAEA. https://doi.org/10.1594/PANGAEA.923109 Hopmans, E. C., Weijers, J. W. H., Schefuß, E., Herfort, L., Sinninghe Damsté, J. S., & Schouten, S. (2004). A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraether lipids. Earth and Planetary Science Letters, 224(1), 107– 116. https://doi.org/10.1016/j.epsl.2004.05.012 Houben, A. J. P., Bijl, P. K., Pross, J., Bohaty, S. M., Passchier, S., Stickley, C. E., et al. (2013). Reorganization of Southern Ocean Plankton Ecosystem at the Onset of Antarctic Glaciation. Science, 340(6130), 341. https://doi.org/10.1126/science.1223646 Houben, A. J. P., Bijl, P. K., Sluijs, A., Schouten, S., & Brinkhuis, H. (2019, May 1). Late Eocene Southern Ocean cooling and invigoration of circulation preconditioned Antarctica for full‐scale Glaciation. Geochemistry, Geophysics, Geosystems. https://doi.org/10.1029/2019GC008182 Huguet, C., Hopmans, E. C., Febo-Ayala, W., Thompson, D. H., Sinninghe Damsté, J. S., & Schouten, S. (2006). An improved method to determine the absolute abundance of glycerol dibiphytanyl glycerol tetraether lipids. Organic Geochemistry, 37(9), 1036–1041. https://doi.org/10.1016/j.orggeochem.2006.05.008 Hutchinson, D. K., Coxall, H. K., Lunt, D. J., Steinthorsdottir, M., de Boer, A. M., Baatsen, M., et al. (2020). The Eocene-Oligocene transition: A review of marine and terrestrial proxy data, models and model-data comparisons. Climate of the Past, 2020, 17(1), 269-315. https://doi.org/10.5194/cp-17-269-2021. Inglis, G. N., Carmichael, M. J., Farnsworth, A., Lunt, D. J., & Pancost, R. D. (2020). A long- term, high-latitude record of Eocene hydrological change in the Greenland region. 108 Palaeogeography, Palaeoclimatology, Palaeoecology, 537, 109378. https://doi.org/10.1016/j.palaeo.2019.109378 Inglis, G. N., Farnsworth, A., Lunt, D., Foster, G. L., Hollis, C. J., Pagani, M., Jardine, P. E., Pearson, P. N., Markwick, P., Galsworthy, A. M. J., Raynham, L., Taylor, Kyle. W. R., & Pancost, R. D. (2015). Descent toward the Icehouse: Eocene sea surface cooling inferred from GDGT distributions. Paleoceanography, 30(7), 1000–1020. https://doi.org/10.1002/2014PA002723 Inglis, G. N., Naafs, B. D. A., Zheng, Y., McClymont, E. L., Evershed, R. P., & Pancost, R. D. (2018). Distributions of geohopanoids in peat: Implications for the use of hopanoid-based proxies in natural archives. Geochimica et Cosmochimica Acta, 224, 249–261. https://doi.org/10.1016/j.gca.2017.12.029 Ivany, L. C., Brey, T., Huber, M., Buick, D. P., & Schöne, B. R. (2011). El Niño in the Eocene greenhouse recorded by fossil bivalves and wood from Antarctica. Geophysical Research Letters, 38(16). https://doi.org/10.1029/2011GL048635 Kennett, J. P. (1977). Cenozoic evolution of Antarctic glaciation, the circum-Antarctic Ocean, and their impact on global paleoceanography. Journal of Geophysical Research (1896-1977), 82(27), 3843–3860. https://doi.org/10.1029/JC082i027p03843 Kvenvolden, K., D. Hostettler, F., B. Rapp, J., & Frank, T. (1991). Aliphatic hydrocarbons in sediments from Prydz Bay, Antarctica. Proceedings of Ocean Drilling Program: Part B (119), 417–423. Lear, C. H., Bailey, T. R., Pearson, P. N., Coxall, H. K., & Rosenthal, Y. (2008). Cooling and ice growth across the Eocene-Oligocene transition. Geology, 36(3), 251–254. https://doi.org/10.1130/G24584A.1 Liu, Z., Pagani, M., Zinniker, D., DeConto, R., Huber, M., Brinkhuis, H., Shah, S. R., Leckie, R. M., & Pearson, A. (2009). Global Cooling During the Eocene-Oligocene Climate Transition. Science, 323(5918), 1187. https://doi.org/10.1126/science.1166368 Macphail, M. K., & Truswell, E. M. (2004). Palynology of Neogene slope and rise deposits from ODP Sites 1165 and 1167, East Antarctica. Proceedings of the Ocean Drilling Program: Scientific Results, 188, 1–20. Masson-Delmotte, V., Hou, S., Ekaykin, A., Jouzel, J., Aristarain, A., Bernardo, R. T., Bromwich, D., Cattani, O., Delmotte, M., Falourd, S., Frezzotti, M., Gallee, H., Genoni, L., Isaksson, E., Landais, A., Helsen, M. M., Hoffmann, G., Lopez, J., Morgan, V., … White, J. W. C. (2008). A review of Antarctic surface snow isotopic composition: Observations, atmospheric circulation, and isotopic modeling. Journal of Climate, 21(13), 3359–3387. https://doi.org/10.1175/2007JCLI2139.1 McFarlin, J. M., Axford, Y., Masterson, A. L., & Osburn, M. R. (2019). Calibration of modern sedimentary δ2H plant wax-water relationships in Greenland lakes. Quaternary Science Reviews, 225, 105978. https://doi.org/10.1016/j.quascirev.2019.105978 109 McKelvey, B., Hambrey, M., Harwood, D., Mabin, M., Webb, P.-N., & Whitehead, J. (2001). The Pagodroma Group—A Cenozoic record of the East Antarctic ice sheet in the northern Prince Charles Mountains. Antarctic Science, 13(4), 455–468. https://doi.org/10.1017/S095410200100061X McLoughlin, S., Lindström, S., & Drinnan, A. N. (1997). Gondwanan floristic and sedimentological trends during the Permian–Triassic transition: New evidence from the Amery Group, northern Prince Charles Mountains, East Antarctica. Antarctic Science, 9(3), 281–298. Meire, L., Mortensen, J., Meire, P., Juul-Pedersen, T., Sejr, M. K., Rysgaard, S., et al. (2017). Marine-terminating glaciers sustain high productivity in Greenland fjords. Global Change Biology, 23(12), 5344–5357. https://doi.org/10.1111/gcb.13801 O’Brien, P. E., Cooper, A. K., & Richter, C. (2001). Leg 188 summary: Prydz Bay-Cooperation Sea, Antarctica. Proceedings of the Ocean Drilling Program. Initial Reports, 188, 1–65. Pagani, M., Huber, M., Liu, Z., Bohaty, S. M., Henderiks, J., Sijp, W., Krishnan, S., & DeConto, R. M. (2011). The role of carbon dioxide during the onset of Antarctic glaciation. Science, 334(6060), 1261. https://doi.org/10.1126/science.1203909 Passchier, S., Bohaty, S. M., Jimenez-Espejo, F. J., Pross, J., Roehl, U., van de Flierdt, T., Escutia, C., & Brinkhuis, H. (2013). Early Eocene to middle Miocene cooling and aridification of East Antarctica. Geochemistry Geophysics Geosystems, 14(5), 1399–1410. https://doi.org/10.1002/ggge.20106 Passchier, S., Ciarletta, D. J., Miriagos, T. E., Bijl, P. K., & Bohaty, S. M. (2017). An Antarctic stratigraphic record of stepwise ice growth through the Eocene-Oligocene transition. Geological Society of America Bulletin, 129(3–4), 318–330. https://doi.org/10.1130/B31482.1 Paxman, G. J. G., Jamieson, S. S. R., Hochmuth, K., Gohl, K., Bentley, M. J., Leitchenkov, G., & Ferraccioli, F. (2019). Reconstructions of Antarctic topography since the Eocene– Oligocene boundary. Palaeogeography, Palaeoclimatology, Palaeoecology, 535, 109346. https://doi.org/10.1016/j.palaeo.2019.109346 Pearson, P. N., Foster, G. L., & Wade, B. S. (2009). Atmospheric carbon dioxide through the Eocene–Oligocene climate transition. Nature, 461(7267), 1110–1113. https://doi.org/10.1038/nature08447 Petersen, S. V., & Schrag, D. P. (2015). Antarctic ice growth before and after the Eocene- Oligocene transition: New estimates from clumped isotope paleothermometry: Antarctic ice growth at the E/O transition. Paleoceanography, 30(10), 1305–1317. https://doi.org/10.1002/2014PA002769 Rose, K. C., Ferraccioli, F., Jamieson, S. S. R., Bell, R. E., Corr, H., Creyts, T. T., Braaten, D., Jordan, T. A., Fretwell, P. T., & Damaske, D. (2013). Early East Antarctic Ice Sheet growth 110 recorded in the landscape of the Gamburtsev Subglacial Mountains. Earth and Planetary Science Letters, 375, 1–12. https://doi.org/10.1016/j.epsl.2013.03.053 Scher, H. D., Bohaty, S. M., Smith, B. W., & Munn, G. H. (2014). Isotopic interrogation of a suspected late Eocene glaciation. Paleoceanography, 29(6), 628–644. https://doi.org/10.1002/2014PA002648 Scher, H. D., Bohaty, S. M., Zachos, J. C., & Delaney, M. L. (2011). Two-stepping into the icehouse: East Antarctic weathering during progressive ice-sheet expansion at the Eocene– Oligocene transition. Geology, 39(4), 383–386. https://doi.org/10.1130/G31726.1 Scher, H. D., Whittaker, J. M., Williams, S. E., Latimer, J. C., Kordesch, W. E. C., & Delaney, M. L. (2015). Onset of Antarctic Circumpolar Current 30 million years ago as Tasmanian Gateway aligned with westerlies. Nature, 523(7562), 580–583. https://doi.org/10.1038/nature14598 Schimmelmann, A., Sessions, A. L., & Mastalerz, M. (2006). Hydrogen isotopic (D/H) composition of organic matter during diagenesis and thermal maturation. Annual Review of Earth and Planetary Sciences, 34(1), 501–533. https://doi.org/10.1146/annurev.earth.34.031405.125011 Schouten, S., Huguet, C., Hopmans, E. C., Kienhuis, M. V. M., & Sinninghe Damsté, J. S. (2007). Analytical Methodology for TEX86 Paleothermometry by High-Performance Liquid Chromatography/Atmospheric Pressure Chemical Ionization-Mass Spectrometry. Analytical Chemistry, 79(7), 2940–2944. https://doi.org/10.1021/ac062339v Sessions, A. L., Zhang, L., Welander, P. V., Doughty, D., Summons, R. E., & Newman, D. K. (2013). Identification and quantification of polyfunctionalized hopanoids by high temperature gas chromatography-mass spectrometry. Organic Geochemistry, 56, 120–130. https://doi.org/10.1016/j.orggeochem.2012.12.009 Sheldon, N. D., Retallack, G. J., & Tanaka, S. (2002). Geochemical Climofunctions from North American Soils and Application to Paleosols across the Eocene‐Oligocene Boundary in Oregon. The Journal of Geology, 110(6), 687–696. https://doi.org/10.1086/342865 Shepherd, A., Ivins, E., Rignot, E., Smith, B., van den Broeke, M., Velicogna, I., et al. (2018). Mass balance of the Antarctic Ice Sheet from 1992 to 2017. Nature, 558(7709), 219–222. https://doi.org/10.1038/s41586-018-0179-y Stocchi, P., Escutia, C., Houben, A. J. P., Vermeersen, B. L. A., Bijl, P. K., Brinkhuis, H., et al. (2013). Relative sea-level rise around East Antarctica during Oligocene glaciation. Nature Geoscience, 6(5), 380–384. https://doi.org/10.1038/ngeo1783 Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., Feakins, S. J. (2020). Prydz Bay East Antarctica, biomarker and pollen, 36-33 million years. NOAA Paleoclimate Database. https://www.ncdc.noaa.gov/paleo/study/32052. Deposited December 17th 2020. 111 Tierney, J. E., & Tingley, M. P. (2014). A Bayesian, spatially-varying calibration model for the TEX86 proxy. Geochimica et Cosmochimica Acta, 127, 83–106. https://doi.org/10.1016/j.gca.2013.11.026 Tierney, J. E., & Tingley, M. P. (2015). A TEX86 surface sediment database and extended Bayesian calibration. Scientific Data, 2(1), 150029. https://doi.org/10.1038/sdata.2015.29 Tipple, B. J., Meyers, S. R., & Pagani, M. (2010). Carbon isotope ratio of Cenozoic CO2: A comparative evaluation of available geochemical proxies. Paleoceanography, 25(3). https://doi.org/10.1029/2009PA001851 Tochilin, C. J., Reiners, P. W., Thomson, S. N., Gehrels, G. E., Hemming, S. R., & Pierce, E. L. (2012). Erosional history of the Prydz Bay sector of East Antarctica from detrital apatite and zircon geo- and thermochronology multidating. Geochemistry, Geophysics, Geosystems, 13(11). https://doi.org/10.1029/2012GC004364 Truswell, E. M., & Macphail, M. K. (2009). Polar forests on the edge of extinction: What does the fossil spore and pollen evidence from East Antarctica say? Australian Systematic Botany, 22(2), 57–106. https://doi.org/10.1071/SB08046 Uemura, H., & Ishiwatari, R. (1995). Identification of unusual 17β(H)-moret-22(29)-ene in lake sediments. Organic Geochemistry, 23(7), 675–680. https://doi.org/10.1016/0146- 6380(95)00036-E van de Flierdt, T., Hemming, S. R., Goldstein, S. L., Gehrels, G. E., & Cox, S. E. (2008). Evidence against a young volcanic origin of the Gamburtsev Subglacial Mountains, Antarctica. Geophysical Research Letters, 35(21). https://doi.org/10.1029/2008GL035564 Warny, S. S., & Askin, R. R. (2011). Vegetation and Organic-Walled Phytoplankton at the End of the Antarctic Greenhouse World: Latest Eocene Cooling Events. In Tectonic, Climatic, and Cryospheric Evolution of the Antarctic Peninsula (pp. 193–210). American Geophysical Union (AGU). https://doi.org/10.1029/2010SP000965 Warny, S. Kymes, C. M., Askin, R. Krajewski, K. P. & Tatur, A. (2019). Terrestrial and Marine Floral Response to Latest Eocene and Oligocene Events on the Antarctic Peninsula. Palynology, 43(1), 4–21. https://doi.org/10.1080/01916122.2017.1418444 Wright, N. M., Scher, H. D., Seton, M., Huck, C. E., & Duggan, B. D. (2018). No change in Southern Ocean circulation in the Indian Ocean from the Eocene through Late Oligocene. Paleoceanography and Paleoclimatology, 33(2), 152–167. https://doi.org/10.1002/2017PA003238 Zachos, J. C., Stott, L. D., & Lohmann, K. C. (1994). Evolution of Early Cenozoic marine temperatures. Paleoceanography, 9(2), 353–387. https://doi.org/10.1029/93PA03266 Zhang, Y. G., Pagani, M., Liu, Z., Bohaty, S. M., & DeConto, R. (2013). A 40-million-year history of atmospheric CO2. Philosophical Transactions of the Royal Society A: 112 Mathematical, Physical and Engineering Sciences, 371(2001), 20130096. https://doi.org/10.1098/rsta.2013.0096 Zhang, Y. G., Pagani, M., & Wang, Z. (2016). Ring Index: A new strategy to evaluate the integrity of TEX86 paleothermometry. Paleoceanography, 31(2), 220–232. https://doi.org/10.1002/2015PA002848 113 Chapter 4 Cenozoic Antarctic Peninsula temperatures and glacial erosion signals from a multi-proxy biomarker study of SHALDRIL sediments This chapter is in review at AGU Paleoceanography and Paleoclimatology: Tibbett, E. J., Warny, S., Tierney, J. E., & Feakins, S. J. (2022). Cenozoic Antarctic Peninsula temperatures and glacial erosion signlas from a multi-proxy biomarker study of SHALDRIL sediments. Paleoceanography and Paleoclimatology Abstract Climate records for Antarctica, beyond the age limit of ice cores, are restricted to the few unglaciated areas with exposed rock outcrops. Offshore there are also few marine cores recovered from the challenging seas around the Antarctic margins. The SHALDRIL II expedition in 2006 collected sediment from near the Antarctic Peninsula that captured time slices of the late Eocene, Oligocene, middle Miocene, and early Pliocene. Here we revisit the cores with biomarker approaches including analyses of hopanes and n-alkanes that indicate increased erosion of mature components as glaciation advanced. Branched glycerol dialkyl glycerol tetraethers (GDGTs) suggest similar temperatures for months above freezing for the segments of the Eocene, Oligocene, and Miocene of 11-12°C but much colder (and likely shorter) periods of thaw by the Pliocene (5.4°C). TEX86-based sea surface temperature estimates are unavailable for the Eocene. However, decreasing terrestrial fluxes allow for the application of TEX86-based sea surface temperature estimates which capture ocean cooling from 11°C in the Miocene to 2.8°C in the Pliocene. Across the Cenozoic, the SHALDRIL II sediments record declining air and sea surface temperatures and increasing glacial inputs from the late Eocene to the Pliocene. 114 4.1 Introduction The Cenozoic is divided into the Greenhouse and Icehouse world with the transition noted by the growth of high latitude ice sheets (Westerhold et al., 2020). The oxygen isotope transition (34.2- 33.5 Ma) spanning the Eocene-Oligocene boundary (35.9 Ma) marks the formation of a large ice sheet on Antarctica (Coxall et al., 2005; Coxall & Pearson, 2007). However, portions of the Antarctic Peninsula may have been among the last areas of the continent to be glaciated, given their lower latitude, providing refugia for vegetation (Anderson et al., 2011). Although there has been a continuous Antarctic ice sheet since 33.7 Ma, there have been fluctuations in its extent and volume throughout the Cenozoic (Westerhold et al., 2020). Soon after the initial formation in the early Oligocene, volumetric estimates place the Antarctic ice sheet at its modern size or perhaps 125% larger than modern (Katz et al., 2008; Miller et al., 2020; Wilson et al., 2013). During the Oligocene, although there was a significant ice sheet, warm temperatures persisted at the margins of Antarctica (O’Brien et al., 2020). The Antarctic ice sheet volume increased at the Oligocene/Miocene boundary (Mi1) and during other Mi events noted by increases in benthic δ 18 O (Flower & Kennett, 1994; Miller et al., 1991). A major glacial contraction occurred during the Miocene Climatic Optimum followed by cooling during the Middle Miocene Climatic Transition, with increases in West Antarctic Ice Sheet (WAIS) glacial extent during the Miocene (Marschalek et al., 2021). The Pliocene saw orbitally-paced advances and retreats of WAIS (McKay et al., 2022; Naish et al., 2009). On the Antarctic Peninsula ice sheet fluctuations across the Cenozoic are less constrained. Glaciation may have occurred in the Paleogene based on glaciogenic sediments on King Georges Island with evidence for Neogene glaciation coming from glaciovolcanic strata on King George and James Ross island as well as marine geophysical evidence (Davies et al., 2012). 115 Records generated from the Antarctic Peninsula capture shifts in climate and ice sheet growth. Due to uplift and glacial erosion of overlying sediments, older sediment is lying within tens of meters of the seafloor making these shallowly accessible (Anderson & Wellner, 2012). However, the presence of thick multi-year sea ice and iceberg hazards limits the possibility for most drilling of marine sediment proximal to the coast. To collect sediment from the Joinville Plateau adjacent to the tip of the Antarctic Peninsula, a drilling rig was added to the R.V. icebreaker Nathaniel B. Palmer for the SHALDRIL II (Shallow Drilling on the Antarctic Continental Margin) expedition (Anderson, 2006) that collected Eocene, Oligocene, Miocene, and Pliocene age sediments (Anderson et al, 2011; Bohaty et al., 2011). Pollen analyses indicated a decrease in the diversity of angiosperm-dominated vegetation on the Northern Peninsula as glaciation proceeded (Anderson et al., 2011; Warny & Askin, 2011a,b). The sediment collected by SHALDRIL II allows a comparison of climate conditions on the Antarctic Peninsula across the transition from ephemeral ice sheets and alpine glaciers in the Late Eocene to a grounded Antarctic Ice Sheet extending onto the continental shelf by the Pliocene (Anderson et al., 2011). The pollen results from SHALDRIL II Hole 3C indicates a cooling in the latest Eocene (Anderson et al., 2011). In the same samples, additional analyses revealed increased water stress from δ 13 C of pollen grains (Griener et al., 2013) and a 30‰ increase in the δD values of leaf wax (Feakins et al., 2014). Here we revisit Hole 3C with biomarker approaches (similar to Inglis et al., 2018; Duncan et al., 2019; Tibbett et al., 2021). In particular, we use n-alkanes and hopanes to assess terrestrial erosion and maturity signals, we generate soil temperatures using the MBT´5Me proxy recently calibrated with Bayesian approaches (BayMBT0) (Dearing Crampton- Flood et al., 2020), and investigate sea surface temperatures using TEX86 and the BAYSPAR calibration (Tierney & Tingley, 2014). In addition, we extend the multi-biomarker study to the 116 Oligocene, Miocene and Pliocene sediments to capture the long term erosional and cooling signals, similar to the approaches used in Prydz Bay, East Antarctica (Tibbett et al., 2021). This new record for the Antarctic Peninsula is compared to the demise of vegetation based on pollen concentration from the same sites (Anderson et al., 2011) and global signals of atmospheric carbon dioxide, deep ocean temperatures, and ice volume that describe the global Cenozoic cooling trend. 4.2 Study Location Cruise NBP0602A Hole 3C (63ºS, 54º39.21'W, and 340 m water depth) from the James Ross Basin in the Weddell Sea and Hole 5D and 12A from the Joinville Plateau (corresponding to 63ºS, 52º21.94'W, and 506 m water depth and 63ºS, 52º49.50'W, and 442 m water depth respectively) were used in this study (Fig. 4.1). The age model is primarily based on diatom biostratigraphy with nannofossil and dinoflagellate cyst biostratigraphy as well as strontium isotopes (Anderson et al., 2011; Bohaty et al., 2011). Strontium isotopes from bivalve fragments from Hole 3C had a 87 Sr/ 86 Sr ratio indicating middle Eocene (35.9±1.1 Ma) which is supported by diatom, nannofossil, and dinoflagellate assemblages (Bohaty et al., 2011). In situ pollen supports this interpretation (Warny & Askin, 2011a,b). Hole 12A was estimated to be late Oligocene with a 87 Sr/ 86 Sr ratio indicating an age of 27.2±0.6 Ma. Age estimates for this hole are placed between 28.4 and 23.3 Ma based on the strontium isotopes and biostratigraphy of diatoms and calcareous nannofossils. Hole 5D is estimated as Middle Miocene (12.8-11.7 Ma) with an unconformity separating the Middle Miocene and Pliocene (5.1 to 4.3 Ma) sediment with ages based on the presence of Thalassiosira complicata and the lack of Fragilariopsis barronii. Based on the sediment thickness of the core sections used in this paper, these sections only represent an upward limit of a few 100 kyrs of deposition (Bohaty et al., 2011). The sediment from Hole 3C 117 consists of muddy to very fine sand, Hole 12A consists of diatomaceous mud and muddy sand while Hole 5D consists of pebbly sandy mud and pebbly muddy sand (unsorted, delivered by icebergs) with layers of diatomaceous mud (Anderson et al., 2011). Figure 4.1. Map of the present day Antarctic Peninsula bed elevation (Fretwell et al., 2013; Greene et al., 2017), showing SHALDRIL II drilling locations (white stars). 4.3 Methods 4.3.1 Biomarker Extraction and Purification Sediment weighing from 10 to 22 g were freeze dried, homogenized and lipids were extracted using DIONEX Accelerated Solvent Extraction with 9:1 (v/v) dichloromethane (DCM) to methanol at 100°C and 1500 psi at the University of Southern California. The total lipid extract was separated into a neutral and acid fraction using a NH2 sepra column eluted with a 2:1 (v/v) ratio of DCM:isopropanol (neutral fraction) and 4% formic acid in diethyl ether (acid fraction). The neutral fraction was separated over silica gel eluting alkanes with hexanes and then the polar 118 fraction eluting first with DCM and then with methanol. The acid fraction was methylated overnight with methanol of a known isotopic concentration and hydrochloric acid (95:5) and dried using an anhydrous sodium sulfate column yielding the fatty acid methyl esters (FAMEs). The FAMEs were purified using a silica gel column eluting first with hexane, and then collecting the FAMEs with DCM. 4.3.2 Liquid Chromatography Analyses Sediment weighing from 10 to 22 g were freeze dried, homogenized and lipids were extracted using DIONEX Accelerated Solvent Extraction with 9:1 (v/v) dichloromethane (DCM) to methanol at 100°C and 1500 psi at the University of Southern California. The total lipid extract was separated into a neutral and acid fraction using a NH2 sepra column eluted with a 2:1 (v/v) ratio of DCM:isopropanol (neutral fraction) and 4% formic acid in diethyl ether (acid fraction). The neutral fraction was separated over silica gel eluting alkanes with hexanes and then the polar fraction eluting first with DCM and then with methanol. The acid fraction was methylated overnight with methanol of a known isotopic concentration and hydrochloric acid (95:5) and dried using an anhydrous sodium sulfate column yielding the fatty acid methyl esters (FAMEs). The FAMEs were purified using a silica gel column eluting first with hexane, and then collecting the FAMEs with DCM. An internal C46 GDGT standard (Huguet et al., 2006) was added to the polar fractions containing glycerol dialkyl glycerol tetraethers (GDGTs) for quantification. These fractions were then dissolved in hexane: isopropanol (99:1) and filtered (0.45 μm PTFE) prior to injection on an Agilent 1260 High-Performance Liquid Chromatography (HPLC) coupled to an Agilent 6120 mass spectrometer at the University of Arizona. GDGTs were analyzed using two BEH HILIC silica columns (2.1×150 mm, 1.7 μm; 90 Waters) and the methodology of Hopmans et al., 119 (2016). Single Ion Monitoring (SIM) of the protonated molecules (M + H + ions) was used to detect and quantify GDGTs with abundances determined by comparison to an internal standard at m/z 744. We report total concentrations of branched (brGDGTs) and isoprenoidal (isoGDGTs) GDGTs in the sediments as measures of terrestrial and marine inputs respectively, and calculate the Branched and Isoprenoidal Tetraether (BIT) index: 𝐵𝐵𝐵𝐵 𝐵𝐵 = 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐶𝐶𝐶𝐶𝐶𝐶 𝑛𝑛 (1) where Ia, IIa and IIIa represent the abundances of both the 5´ and 6´ methyl isomers of the non- cyclic terrestrial brGDGTs (from soil bacteria) and Cren represents the abundance of crenarchaeol (mostly produced by marine archaea) (Hopmans et al., 2004). In all samples, the 5-methyl brGDGT index, MBT´5Me (de Jonge et al., 2014, Hopmans et al., 2016) was calculated as 𝑀𝑀 𝐵𝐵𝐵𝐵 5 𝑀𝑀𝐶𝐶 ′ = 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 (2) and converted to mean annual air temperature for months above freezing (MAF) using the BayMBT0 calibration (Dearing Crampton-Flood et al., 2020). For the Eocene prior, a value of 10°C was used as previous studies suggest regional temperatures of 8-13°C during the middle Eocene based on weathering proxies (Passchier et al., 2013), fossil leaf assemblages (Francis et al., 2008), and bivalve oxygen isotopes (Judd et al., 2019). Temperature estimates for the Oligocene are ~7-9°C from CIROS-1 and CRP cores from the Ross Sea (Passchier et al., 2013); therefore, a prior of 8°C is used. For the Miocene, a prior of 7°C is used based on best estimates of summer temperatures from isotope-enabled climate model interpretations of plant wax δD from AND-2A from the Ross Sea (Feakins et al., 2012), summer temperatures of 5°C based on fossil data from the Dry Valleys at 14 Ma (Lewis et al., 2008), and similar temperatures from 120 IODP Site U1356 from Wilkes Land (Passchier et al., 2013). For the standard deviation a value of ±15°C was used. In addition, the cyclization of branched tetraether (CBT') index (De Jonge et al., 2014) was calculated and used to estimate soil pH. 𝐶𝐶 𝐵𝐵𝐵𝐵 ′ = 𝑙𝑙𝑙𝑙 𝑙𝑙 1 0 � 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 ′ 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 + 𝐼𝐼 𝐼𝐼 𝐼𝐼 𝐼𝐼 � (3) 𝑝𝑝𝑝𝑝 = 7.15 + 1.59𝐶𝐶 𝐵𝐵 𝐵𝐵 ′ (4) IsoGDGTs with 0–3 cyclopentane moieties (GDGT-0 to GDGT-3) and crenarchaeol (Cren) with an additional cyclohexane moiety and its regioisomer crenarchaeol´ (Cren´) are dominantly from marine archaeal production. Following (Schouten et al., 2007), TEX86 was calculated using the equation: 𝐵𝐵𝑇𝑇𝑇𝑇 8 6 = [ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐺𝐺 − 2] +[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐺𝐺 − 3] +[ 𝐶𝐶𝐶𝐶𝐶𝐶 𝑛𝑛 ′ ] [ 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 1] +[ 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 2] +[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐺𝐺 − 3] +[ 𝐶𝐶𝐶𝐶𝐶𝐶 𝑛𝑛 ′ ] (5) We converted the TEX86 record to mean annual sea surface temperatures (SSTs) using the BAYSPAR analogue calibration (Tierney & Tingley, 2014). The prior for the Eocene was set to 15°C based on TEX86 L and biostratigraphy from Seymour Island (Douglas et al., 2014) and BAYSPAR and U k’ 37 estimates from DSDP site 511 suggesting temperatures of 16-18°C (Houben et al., 2019; Lauretano et al., 2021). The prior used for sea surface temperature for the Oligocene was also 15°C based on TEX86 data from Site U1356, which place SST around 15°C during the middle Oligocene (Hartman et al., 2018) and U k’ 37 estimates of 10 to 15°C around 31.5 Ma (Plancq et al., 2014) from DSDP Site 511. The Miocene prior was 12°C based on BAYSPAR TEX86 estimates from the Weddell sea of 10-15°C at 11 Ma (Hartman et al., 2018) and estimates of 10-13°C from TEX86 and Δ47 from Site 1171 from the South Tasman Rise (Leutert et al., 2020). For the Pliocene the prior utilized was 5°C based on SST estimates around 121 4 Ma from TEX86 L from ANDRILL AND-1B core from the Ross Sea McKay et al., 2012) and the presence of Leiosphaeridia from James Ross Island suggesting SST of -2 to 5°C (Edwards et al., 1991; Salzmann et al., 2011). The standard deviation used in the reconstruction was ±10°C. Samples with a BIT>0.4 as well as samples with a delta ring index (ΔRI) >0.3 were excluded from TEX86 reconstruction as a high ΔRI implies non-analogue distributions (Zhang et al., 2016), where: 𝑅𝑅 𝐵𝐵 𝑠𝑠 𝐼𝐼 𝑠𝑠𝑠𝑠 𝑠𝑠 𝐶𝐶 = (0[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 0] + 1[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 1] + 2[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 2] + 3[ 𝐺𝐺𝐺𝐺 𝐺𝐺𝐵𝐵 − 3] + 4[ 𝑐𝑐𝑐𝑐𝑐𝑐 𝑛𝑛 ] + 4[ 𝑐𝑐𝑐𝑐𝑐𝑐 𝑛𝑛 ′ ]) (6) 𝑅𝑅 𝐵𝐵 𝐺𝐺 𝑇𝑇 𝑇𝑇 = −0.77 𝐵𝐵𝑇𝑇𝑇𝑇 8 6 + 3.32( 𝐵𝐵𝑇𝑇𝑇𝑇 8 6 ) 2 + 1.59 (7) ∆𝑅𝑅 𝐵𝐵 = 𝑅𝑅 𝐵𝐵 𝐺𝐺 𝑇𝑇 𝑇𝑇 − 𝑅𝑅 𝐵𝐵 𝑠𝑠 𝐼𝐼 𝑠𝑠𝑠𝑠 𝑠𝑠 𝐶𝐶 (8) Samples with a high methane index (MI)>0.5 are excluded from the TEX86 reconstruction as high MI indicate that GDGTs could be sourced from methanogenic producers which can lead to bias TEX86 values (Zhang et al., 2011). 𝑀𝑀 𝐵𝐵 = [ 𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 1] +[ 𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 2] +[𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 3] [ 𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺 𝐺𝐺𝐺𝐺 − 1] +[ 𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺 𝐺𝐺𝐺𝐺 − 2] +[ 𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 3] +[ 𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 5] +[𝑖𝑖 𝑠𝑠 𝑖𝑖 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 − 5 ′ ] (9) 4.3.3 Gas Chromatography Analyses FAMEs, alkanes and hopanes were identified by gas chromatography mass spectrometry (GC- MS) and were quantified by GC-FID (gas chromatography flame ionization detection) relative to in-house standards of alkanes and FAMEs of known concentration at the University of Southern California. We report the concentration of C20–C32 n-alkanoic acids and C16–C29 n-alkanes and calculate the summed C23 to C33 alkanes and summed C20 to C32 FAMEs in ng/gdw. The C16-C18 are not included in the total as they are potentially sourced from contaminants based on 122 processing blanks. We also calculate the average chain length (ACL) and carbon preference index (CPI) using the following formulae: ACL: ∑(𝑛𝑛 ∗ [ 𝐶𝐶 𝑛𝑛 ]) ∑ ⌊ 𝐶𝐶 𝑛𝑛 ⌋ � (10) CPI: 2 ∑ 𝐶𝐶 𝑛𝑛 (∑ 𝐶𝐶 𝑛𝑛 + 1 + ∑ 𝐶𝐶 𝑛𝑛 − 1 ) � (11) where n is 24–32 for n-alkanoic acids and 23–33 for n-alkanes. We also screened the alkanes fraction for the presence of hopanes, biomarkers that derive from membrane lipids in bacteria and undergo isomerization with increasing thermal maturity, using their diagnostic mass fragments 191 m/z ion trace and published spectra (Inglis et al., 2018; Sessions et al., 2013; Uemura & Ishiwatari, 1995), and noted where those samples included a detectable uncharacterized complex mixture above baseline. Each sample was assessed for the presence of C27 through C31 hopanes and their isomers. We calculated the hopane index for each hopane present as follows: ℎ𝑙𝑙𝑝𝑝 𝑜𝑜 𝑛𝑛𝑐𝑐 𝑝𝑝𝑛𝑛 𝑖𝑖 𝑐𝑐 𝑖𝑖 = 𝛽𝛽𝛽𝛽 (𝛼𝛼 𝛽𝛽 + 𝛽𝛽 𝛼𝛼 + 𝛽𝛽𝛽𝛽 ) (12) 4.4 Results 4.4.1 GDGTs 4.4.1.1 BIT BIT values decreased from the Eocene to Pliocene sections. The Eocene section has an average value of 0.65 with a range from 0.57 to 0.74. The Oligocene section has an average of 0.35 and ranges from 0.09 to 0.57. The Miocene section has an average BIT of 0.27 ranging from 0.09 to 0.40. The Pliocene section has average BIT values of 0.09 only varying between 0.08-0.10. 123 Decreasing values towards the present suggest increased marine production and/or decreased flux of terrestrial materials to the marine margins. 4.4.1.2 BrGDGTs ΣbrGDGT concentrations ranged from 0.1 to 10.3 to ng/gdw (mean 2.1 ng/gdw). The concentrations decrease from the Eocene to Pliocene. The maximum concentration occurs in the Eocene samples with an average concentration of 4.5 ng/gdw (2.2 to 10.3 ng/gdw). The Oligocene concentration ranged from 0.7 to 2.9 ng/gdw (mean 1.9 ng/gdw); the Miocene from 0.1 to 3.3 ng/gdw (mean 1.5 ng/gdw); and the Pliocene had the lowest concentrations ranging from 0.5-0.7 ng/gdw, likely implying reduced terrestrial soil inputs. Reconstructed temperatures for months above freezing (MAF) were consistent throughout the Eocene varying minimally between 12.3 to 10.9°C (mean 11.5°C). The Oligocene MAF range from 13.3 to 10.8°C (mean 12.5°C); the Miocene ranged from 12.7 to 8.7°C (mean 11.1°C); with considerably cooler temperatures in the Pliocene ranging from 6.5 to 3.8°C (mean 5.4°C). Soil pH calculated from CBT′ ranged from 5.6 to 7.3 across all sections. For the Eocene, Oligocene, and Miocene average values were similar with averages of pH 6.2, 6.0, and 6.4 respectively. The Pliocene sections have the highest pH ranging from 7.0 to 7.3. 4.4.1.3 IsoGDGTs ΣisoGDGT concentrations ranged from 1.1 to 16.0 ng/gdw (mean 7.7 ng/gdw) overall. Concentrations in the Eocene ranged from to 4.1 to 14.8 ng/gdw; in the Oligocene from 3.7 to 13.2 ng/gdw (mean 8.1 ng/gdw); and in the Miocene section ranges from 1.1 to 13.2 ng/gdw (mean 7.4 ng/gdw). The Pliocene has the maximum concentration of ΣisoGDGTs with a range 124 from 6.4 to 16.0 ng/gdw (mean 10.2 ng/gdw) despite sedimentary dilution by pebbles and sand (Fig. A.8), likely implying increased ocean productivity. The MI was below 0.5 for all samples except for two that also had high ΔRI and were excluded from the SST calculations. Eocene SSTs were not generated due to high BIT and ΔRI values in these samples, indicating terrestrial inputs and microbial communities other than those of the open ocean today. Limited Oligocene SSTs were generated due to high BIT and ΔRI but where BAYSPAR could be applied SST ranged between 11.6 and 6.4°C (mean 9.6°C). Miocene SSTs ranged from 14.3 to 3.9 °C (mean 10.2°C). Pliocene SSTs were considerably colder, ranging from 3.3 to 1.8°C (mean 2.8°C). 4.4.2 Alkyl lipids 4.4.2.1 Alkanoic acids n-Alkanoic acids were detected from C16 to C32; however, C16 to C19 were excluded as possible contaminants present in processed blanks and are therefore not reported or further discussed. The concentration of n-alkanoic acids (C20 to C32) ranged from 0 to 3650 ng/gdw with an average value of 1625 ng/gdw (Fig. A.8). The lowest concentrations 0 and 7 ng/gdw occur in the two youngest Pliocene samples. The additional two Pliocene samples have concentrations of 211 and 482 ng/gdw. Except for two Miocene samples from Hole 5D at 22.3 mbsf and 25.2 mbsf, all other samples exceed 500 ng/gdw with average values of 1894 and 1950 ng/gdw for the Miocene and Oligocene respectively. Only three samples from the Eocene were able to be quantified with an average n-alkanoic acid concentration of 1203 ng/gdw. The CPI ranged from 3.6 to 9.7 with an average value of 5.3 across the record. The average Pliocene CPI was 4.1 based on the two samples that contained long chain n-alkanoic acids, Miocene 6.0, Oligocene 6.0, and Eocene 4.6. 125 ACL ranged from 26.0 to 28.5 with an average of 27.8 with the highest average ACL from the Eocene at 28.2, the other three timeslices have average ACLs from 26.9-27.5. Molecular abundance distributions display C26 and C28 co-dominance (Fig. A.9). 4.4.2.2 Alkanes The n-alkanes were detected from C17 to C31 with concentrations ranging from 119 to 1904 ng/gdw (mean 768 ng/gdw). Long chain n-alkanes (C23 to C31) ranged from 87 to 1822 ng/gdw (mean 607 ng/gdw). The CPI varied from 1.3 (mature) to 4.6 (immature) (mean 2.6). The Pliocene samples have an average concentration of long chain n-alkanes of 140 ng/gdw with the Miocene and Oligocene averaging 542 and 414 ng/gdw respectively. CPI varied between timeslices, Eocene (3.6), the Oligocene (1.9), Miocene (2.1) and Pliocene (1.5). The ACL ranged from 25.9 to 28.1 with an average value of 27.2. The Eocene had the highest ACL of 27.9 with the Oligocene averaging 26.5, the Miocene 26.8 and the Pliocene 26.6. A hierarchical cluster analysis using Euclidean distance was performed on the n-alkane chain length distribution after normalizing to assess changes in inputs such as reworking. The clustering analysis divided the sample set into 3 groups (Fig. 4.2A), with clusters 2 and 3 showing presence of shorter chain compounds (<C23), without odd-over-even carbon preference, i.e., diagnostic of mature n- alkanes. The typical plant-like odd-over-even predominance can be seen (>C24) in all clusters. 126 Figure 4.2. a) Dendrogram of n-alkane chain length distributions with clusters labeled and showing b) molecular abundance distribution, showing mean (bars) and 1 standard deviation (error bars) for each cluster. Samples labelled (core_depth) from cores 3C (Eocene), 12A (Oligocene), 5D (middle Miocene and Pliocene). Eocene samples (3C, bold) are mostly in cluster 1 with plant-like distributions of n-alkanes, but three samples are in cluster 3 along with samples from 5D, with low CPI components, denoting mature inputs. 4.4.2.3 Hopanes Hopanes are derived from bacterial membrane lipids and their isomerization can be used to indicate thermal maturity (Inglis et al., 2020). Hopanes were detected in all samples; however, not all hopanes (C27-C31) were detected in the Pliocene samples. To be able to compare across all samples we report the average hopane ratio across all hopanes detected (hereafter hopane ratio). The hopane ratio suggests moderate maturity overall, with increasing thermal maturity in the younger sections. The Eocene section (Hole 3C) has an average hopane ratio of 0.43 (0.25 to 127 0.53) indicating moderate maturity, with the most mature values found at the top of the section. The Oligocene samples (Hole 12A) have an average hopane index of 0.26 (0.20 to 0.29). The Miocene samples have an average value of 0.28 (0.09 to 0.34). Hopanes were detected across a limited range in the Pliocene samples compared to the other SHALDRIL II samples. Pliocene samples with an average value of 0.20 (0.03 to 0.34) suggesting higher thermal maturity. 128 Figure 4.3. Multi-proxy reconstruction of terrestrial environmental change from the SHALDRIL II cores, a) Nothofagus fusca (Anderson et al., 2011) and reworked pollen percentage (newly reported for comparison to the biomarker reworking data in this study), b) hopane index, c) n- alkane CPI d) BIT index e) Both mean annual air temperature above freezing (BayMBT0) and sea surface temperature from BAYSPAR. 129 4.5 Discussion 4.5.1 Eocene SHALDRIL II Hole 3C provides a snapshot of the late Eocene. Our new brGDGT-based temperature reconstruction using BayMBT0 estimates MAF temperatures of 12°C during the late Eocene (Fig. 4.3E) presumably reflecting conditions on the adjacent tip of the Antarctic Peninsula when temperatures are above freezing. These values are similar to mean annual air temperature estimates based on climate leaf analysis multivariate program (CLAMP) analyses from nearby Seymour Island of 10.8°C (Francis et al., 2008), summer temperature estimates from the La Meseta Formation on Seymour Island of >10°C based on pollen (Warny et al., 2019), mean annual temperature estimates of 11-16°C from the Weddell Sea (site 696) (Thompson et al., 2022), and mean annual temperature estimates from King George Island of 12- 13°C from coexistence analysis of fossil floras (Hunt & Poole, 2003; Poole et al., 2005). Since winter temperatures are not recorded by microbial growth, we do not recommend the MAAT BayMBT calibration, however we do provide those estimates for comparison purposes (Fig. A.8). The seasonal cycle for nearby Seymour Island during the middle Eocene has been constrained to be 8°C using oxygen isotopes on shallow marine bivalves (Judd et al., 2019); if coastal land temperature seasonality is similar, this would imply winter temperatures of about 3°C. Based on the air temperatures from other proxies and the seasonal cycle for nearby Seymour Island, Eocene temperatures are likely above freezing year round (i.e., MAF = MAAT) on the Antarctic Peninsula. Pollen from two different locations on King George Island on the Antarctic Peninsula suggest MAAT of 11-15°C (mean of 12°C) and 9-27°C (mean 13.3°C) (Hunt & Poole, 2003; Poole et al., 2005). The SHALDRIL reconstructions seem reasonable as they fall between estimates from higher and lower latitudes: being warmer than MAAT estimates of 8-10°C from weathering proxies in the Ross Sea (Passchier et al., 2013) and within pollen 130 estimates (8.10-18.4°C) from the Weddell Sea sector (Houben et al., 2011; Mohr, 1990; Pound & Salzmann, 2017). SST estimates were not generated for the Eocene SHALDRIL sediments due to BIT>0.4 suggesting strong fluxes of terrestrial inputs to this site that would interfere with reliable SST estimates. However, TEX86 measurements from the La Meseta Formation on nearby Seymour Island (Douglas et al., 2014) are recalibrated using BAYSPAR here to yield SST estimates of ~13°C (13.5 to 11.3°C). Clumped isotope measurements on shallow marine bivalves likewise indicate temperatures of ~13°C (Douglas et al., 2014). In the uppermost 4 samples of Hole 3C, a decline in N. fusca pollen abundance (Fig. 4.3A) suggests the recovered sediments capture the first signs of decreasing temperature and deteriorating conditions for plants on the Antarctic Peninsula, although unfortunately the EOT was not recovered (Anderson et al., 2011). Compound specific isotopic analysis of that Eocene section similarly found lower δD values (Feakins et al., 2014), compared to the earlier samples within 3C, tentatively interpreted as drying associated with cooling. The increased proportion of reworked pollen, originating from nearby Campanian-Maastrichtian sections, is likely indicative of glacial erosion, also supporting a cooling and drying episode. In the same horizon, we report declines in the hopane index (Fig. 4.3B) and n-alkane CPI (Fig. 4.3C), which both suggest more thermally mature inputs. Given the co-occurrence of non-mature biomarkers (e.g., n-alkanoic acids and GDGTs) as well as pollen, we interpret the declining CPI and hopane index to indicate an input of more reworked mature components, rather than thermal heating in situ. The n-alkane chain length distribution for the uppermost samples in 3C are in alkane Cluster 3 (Fig. 4.2), showing more mature inputs compared to the earlier samples within 3C. The cluster analysis shows that the mature signature in the late Eocene samples is similar to the reworked mature 131 components found in the Miocene and Pliocene of Hole 5D and dissimilar to anything closer in time (Fig. 4.2). Other studies capture declining temperatures on Antarctica further afield. Based on the chemical weathering of minerals, the chemical index of alteration (CIA) <60 and increasing illite mineralogy suggest a cool and wet climate shifting to a cold relatively dry environment in the late Eocene from Seymour Island (Dingle et al., 1998). Similar methods are used to study the late Eocene from Prydz Bay where CIA values after 34.5 Ma are below 65 suggest cooler temperatures and glacial inputs (Passchier et al., 2017). 4.5.2 Oligocene We see signs that glacial erosion increased during the Oligocene, as we find biomarker evidence for more reworked mature inputs based on the reworked pollen percentage, the hopane index and the alkane CPI (Fig. 4.3A-C) relative to the earlier Eocene section. Although there is reworking, we do not expect contributions of brGDGTs from sediments of such maturity and assume iso- and brGDGTs are penecontemporary. In addition, we note the presence of immature material (including n-alkanoic acids and pollen) and no change in the concentration of GDGTs associated with changes in the hopane index and CPI. On land, brGDGT reconstructed air temperatures for MAF range from 13.3 to 10.8°C (mean 12.5°C) similar to Eocene MAF temperatures. Although during the Oligocene an ice sheet was present on Antarctica, our reconstructions suggest that there is little change in the months above freezing temperatures recorded by brGDGTs on the tip of the Antarctic Peninsula, at the northerly limit of the continent, thus registering less cooling than the glaciated continental interior. Cooling is registered at more southerly sites, for example land adjacent to the Ross Sea is colder (7-9°C) (Francis et al., 2008; Passchier et al., 2013). MAAT ranges from pollen from Seymour Island, near the tip of the Antarctic Peninsula, suggest 132 a temperatures between 8 and 18°C with warm month mean temperatures of 17-24°C (Pound & Salzmann, 2017;Warny et al., 2019). Diatom assemblages from SHALDRIL II show a less diverse assemblage and smaller taxa in the Oligocene core, also indicating surface cooling from the Eocene to Oligocene (Bohaty et al., 2011). High BIT and ΔRI meant we were not able to estimate any Eocene and limited Oligocene SSTs. Where BAYSPAR could be applied SST ranged between 11.6 and 6.4°C (mean 9.6°C; Fig. 4.3E). The limited comparable SST records in the region indicate temperatures of 12°C for Maud Rise Site 689 from clumped isotopes (Petersen & Schrag, 2015) and 13°C from BAYSPAR and 10°C from U k′ 37 from DSDP site 511 near the Falkland Plateau for 34-32 Ma (Houben et al., 2019; Lauretano et al., 2021). Based on clumped isotope paleothermometry and BAYSPAR recalibration of TEX86 SST data reported from Seymour Island (Douglas et al., 2014), temperatures decreased by 3°C from the Eocene to the Oligocene (reaching 10°C), matching the interpretation for the decrease in diatom diversity (Bohaty et al., 2011). 4.5.3 Miocene Penecontemporary pollen from SHALDRIL II suggest austral summer temperatures of 10°C on the tip of the Antarctic Peninsula during the Miocene (Anderson et al., 2011). The pollen assemblage likely corresponds to a low shrubby tundra environment (Anderson et al., 2011). We note an increase in reworked pollen from the Oligocene to Miocene time slices (Fig. 4.3A) that we attribute to enhanced terrestrial erosion, including glacial erosion associated with the increased prevalence of pebble-sized dropstones in these Miocene sediments (Fig. A.8). Similarly, we find biomarker evidence for reworked material being delivered to the Miocene horizons of Hole 5D, in addition to the continued presence of fresh pollen and n-alkanoic acids 133 indicating that the sedimentary organics contain a mixture of penecontemporary inputs and transported, mature, reworked materials. We find brGDGTs yield estimates for Miocene MAF ranging from 12.7 to 8.7°C (mean 11.1°C). While these are relatively warm temperatures they reflect the months above freezing (likely only summer months in the Miocene) when soils on unglaciated landscapes are thawed and microbial communities are active, and are thus not incompatible with the dropstone evidence for icebergs calving off of marine-terminating valley glaciers sourced likely from the WAIS and EAIS coastlines bordering the Weddell Sea (Carter et al., 2016). Although there are no local temperature estimates for the middle Miocene, there are Late Miocene records elsewhere on Antarctica that provide summer temperatures derived from animal and plant indicators. From the Dry Valleys (77°S, 1.35-1.5 km asl) in the Transantarctic Mountains – an area that remains unglaciated today - cypridoidean ostracodes and lathridiid beetle fossils yield summer temperatures of 5°C at 14 Ma (Lewis et al., 2008). Although not within the same catchment, plant wax δD and isotope-enabled climate model experiments suggested mid Miocene summer temperatures of 7°C for land adjacent to the Ross Sea (Feakins et al., 2012). Our TEX86 data indicate Miocene SSTs were still relatively warm with estimates ranging from 14.3 to 3.9°C (mean 10.2°C). This is similar to values from previous SST estimates using TEX86 from the Weddell Sea (10-15°C around 14 Ma) (Hartman et al., 2018) and from the South Tasman Rise (10-13°C for 12-14 Ma) (Leutert et al., 2020). In addition, TEX86 measurements from ANDRILL-2A from the Ross Sea (Levy et al., 2016) are recalibrated using BAYSPAR here, using the same prior and standard deviation as the SHALDRIL Miocene samples, yielding SST estimates of ~11°C (6 to 26°C). In contrast, diatom assemblages from the same core suggest the presence of seasonal sea ice (Bohaty et al., 2011) indicating cold sea-ice forming conditions 134 in the winter season, and warm, open ocean conditions in summer. We also note a surge in the number of iceberg-rafted dropstone pebbles in these Miocene sediments in Hole 5D (Fig. A.8) indicating a contrast between active calving from marine-terminating glaciers and warm summer water offshore. Sea ice diatoms indicate winter season ice formation. We also note prior reports of high organic content and no bioturbation in Hole 5D (Miocene and Pliocene sections) which indicates low oxygen conditions (Anderson, 2006), conducive to increased preservation of biomarkers (Fig. A.8). 4.5.4 Pliocene Our BAYSPAR estimates of Pliocene SSTs range from 3.3 to 1.8°C (mean 2.8°C). Diatom assemblages from the same core suggests the presence of seasonal sea ice (Bohaty et al., 2011) which is further corroborated by the presence of Leiosphaeridia from James Ross Island (Salzmann et al., 2011). There was no significant vegetation on the Antarctic Peninsula during the Pliocene (Anderson et al., 2011) reflecting the cold climate and glaciation. The high proportion of reworked pollen, biomarker evidence carrying signals of maturity (particularly the hopane index and alkane CPI), suggest glacial erosion of older strata whether locally on the Antarctic Peninsula or from the Weddell Sea margins of WAIS and EAIS (see Section 4.5.5). In the Pliocene horizons of Hole 5D the brGDGTs yield MAF estimates ranging from 6.5 to 3.8°C (mean 5.4°C), showing that they capture part of the expected cooling. However, they are not as cold as might be expected for a fully glaciated continent. This can be reconciled by understanding that soil microbial activity is limited to the months when the soil thaws, and as climate cools, this activity is likely restricted to a shorter summer. We see considerable cooling in the Pliocene age soil-derived proxy, relative to the earlier times, so part of the Antarctic 135 cooling is captured by the proxy, but we expect conditions to be colder than reconstructed in the Pliocene due to the inherent temperature limits (to MAF) of the microbial recorder. 4.5.5 Iceberg transport in the Weddell Sea The presence of iceberg-delivered pebbles in Eocene-Pliocene age sediments in the studied sections of SHALDRIL II sediments (Fig. A.8) and the location of the drill sites in “iceberg alley” in the Weddell Sea, implies pebbles could have been delivered by ice calved from virtually anywhere in the Weddell Sea, which includes both EAIS and WAIS (Anderson et al., 2011). Iceberg alley is a major route for WAIS and EAIS derived icebergs via the Weddell gyre circulation in the modern ocean, a circulation and iceberg transit route that has likely been in place since at least the late Eocene (Carter et al., 2017). Additional studies of this region have shown, that in the late Eocene, there was widespread ice that extended from the mountainous interiors to the coastal areas fringing the southern Weddell Sea and EAIS contributions have been identified in provenance studies for the Eocene (Carter et al., 2017). Iceberg transport via Weddell Sea circulation allows for distal sourcing of terrestrial biomarkers from both WAIS and EAIS via this route. In the upper few samples of the Eocene, in the Oligocene, Miocene and Pliocene, we see evidence for reworked mature biomarkers (based on the high CPI alkanes and the low hopane ratio) and reworked pollen in the same horizons as the pebbles (Fig. A.8). This means distal transport of mature sedimentary organics, rafted by ice. We do not however infer distal sourcing of the soil-derived brGDGTs or fatty acids. The maturity of the reworked material, means that these compounds would not be preserved in the eroding sediments, and thus we infer that the fatty acids and brGDGTs are not reworked from older strata, but are penecontemporary and more locally sourced from the peninsula. 136 When considering iceberg transport, the warm SSTs reconstructed in this study present a contrast. Paleotemperature reconstructions from this study and other studies (Douglas et al., 2014) suggest warm (above freezing) SSTs at the margins of the Antarctic Peninsula (Fig. 4.4). Given that warm SSTs would have melted icebergs rapidly, the presence of iceberg-rafted pebbles at the tip of the Antarctic Peninsula is an apparent paradox (Douglas et al., 2014; Carter et al., 2017). Those prior warm Eocene temperature estimates are corroborated with our new data with warm land temperatures in the Eocene and warm SSTs persisting across the Oligocene and Miocene, adding more confidence to the interpretation of icebergs calving into a relatively warm ocean, from the Eocene to the Miocene, with a vigorous Weddell Gyre current transporting the icebergs to the tip of the peninsula before they melt. This ocean circulation is supported by ocean modelling studies (Bijl et al., 2011; Goldner et al., 2015) and the paleotemperature proxies (Anderson et al., 2011; Douglas et al., 2014; Carter et al., 2017; and this study). The relatively warm ocean temperatures reconstructed from SHALDRIL II in Oligocene and Miocene times, with cooling only in the Pliocene (Fig. 4.4), provide a useful contribution to the oceanographic history of the Weddell Sea and the global climate. 4.5.6 Cenozoic Cooling The global Cenozoic cooling trend is often described by the record of benthic foraminiferal calcite (Westerhold et al., 2020), which records the glacial ice volume on the Antarctic climate as well as the cooling of the deep oceans. This general cooling trend is linked to the decline in atmospheric pCO2 (Fig. 4.4). Across the time windows studied, there is a transition from ~1000 ppmv in the Eocene to ~400 ppmv in the other time windows studied (Rae et al., 2021) reconstructed from alkenone (Badger et al., 2019; Pagani et al., 2005, 2010, 2011; Zhang et al., 2013) and boron isotope proxies (Anagnostou et al., 2016, 2020; Badger et al., 2013; Foster et 137 al., 2012; Greenop et al., 2014, 2019; Henehan et al., 2020; Pearson et al., 2009; Sosdian et al., 2018). Here we compare snapshots of conditions on and around Antarctica that capture the local cooling to the global trend (Fig. 4.4). 138 Figure 4.4. Compiled Cenozoic proxy records from the continent of Antarctica and the surrounding Southern Ocean including the SHALDRIL II data generated (yellow squares) a) Surface air temperature from around the Antarctic including temperatures generated form the S- index (Passchier et al., 2013, 2017), BayMBT from Prydz Bay (Tibbett et al., 2021), BayMBT0 (this study), pollen, MBT/CBT from Seymour Island (Douglas et al., 2014), climate estimates using probability density functions (PDFs) from sporomorphs (Thompson et al., 2022), and the 139 relationship between δD and temperature (Feakins et al., 2012) b) sea surface temperature from BAYSPAR using TEX86 (Douglas et al., 2014; Hartman et al., 2018; Lauretano et al., 2021; Leutert et al., 2020; Levy et al., 2016; Liu et al., 2009; Plancq et al., 2014), U k’ 37 from DSDP Site 511(Houben et al., 2019; Petersen & Schrag, 2015), and clumped isotopes (Δ47) (Douglas et al., 2014), c) δ 18 Obenthic record spline (Westerhold et al., 2020) and c) pCO2 record compiled from δ 11 B (blue) and alkenones (red) proxies (Rae et al., 2021). The colors for the proxy records correspond to different ocean sectors Indian Ocean (orange), Western Pacific (black), Ross Sea (blue), Weddell Sea (brown). The pollen record (green) comes from various locations including the Antarctic Peninsula (Francis et al., 2008; Poole et al., 2005), the Ross Sea region (Askin & Raine, 2000; Francis et al., 2008; Lewis et al., 2008; Prebble et al., 2006; Raine, 1998; Warny et al., 2009), and the Indian Ocean sector (Macphail & Truswell, 2004; Truswell & Macphail, 2009). On the inset map the symbols with a white outline correspond to surface air temperature while the black outlines are sea surface temperature records. Comparing before and after the Eocene-Oligocene transition, there is an approximate decrease pCO2 of 1000 – 600 ppmv, that coincides with the major glacial expansion of the “permanent” Antarctic ice sheet (Fig. 4.4C). Yet, at the SHALDRIL II sites in the Antarctic Peninsula sampled there is no detectable (<1°C) difference in the reconstructed MAF from brGDGTs between the Eocene and Oligocene sections (Fig. 4.4C). Likely the declining winter temperatures shortened the summer season, but the temperatures remained similar in summer months. We note that globally, the Oligocene has warm temperatures comparable to the Eocene (O’Brien et al., 2020). However, pollen indicate that plants started to detect a change in the climate with N. fusca declining (Anderson et al., 2011). However, the tip of the Antarctic Peninsula is far from the nucleation of initial glaciation on the Gamburstev Mountains (Rose et al., 2013) and it was likely late to respond to the continental cooling, as evidenced by the lack of cooling detected by the soil biomarker proxy recorder of MAF temperatures. We cannot generate SST estimates for the Eocene time periods from the SHALDRIL cores because of the high terrestrial flux (high BIT and high ΔRI). Another study from Seymour Island generated TEX86 data (Douglas et al., 2014) that we reanalyzed using BAYSPAR estimating 13°C for SST during the late Eocene oceans adjacent to the eastern Antarctic Peninsula on what is today above sea level on Seymour Island. 140 We were able to reconstruct some Oligocene SSTs with considerable variability between 11.6 and 6.4°C in just 3 samples with an average SST of 10°C. Based on the reassessment of previous SSTs from the Antarctic Peninsula, SST cooled 3°C from the Eocene to the Oligocene which is supported by the change in diatom assemblage from the same cores indicating a cooling and increase in sea ice (Bohaty et al., 2011). From the Oligocene to Miocene there was again no clear drop in temperatures detected in these snapshots of conditions from the SHALDRIL biomarker records. On land temperatures drop modestly with MAF of 12.5°C during the Oligocene and 11.1°C during the Miocene, within the variability in each segment. Between the Oligocene and Miocene time slices the brGDGTs indicate temperatures are largely unchanged on the Antarctic Peninsula, and this is either because the change is minimal from the sampled windows of the Oligocene and the Miocene or as the climate cools, the winter frozen months lengthen, and the soil proxy is restricted to a short summer season, thus the temperatures may appear to be relatively similar despite a shorter time of thawed soil activity. The largest cooling contrast is seen between the sampled windows of the late Miocene and the Pliocene. SSTs cool from 10.2 to 2.8°C on average for the recovered sections, a drop of 7°C (Fig. 4.4). Between the Miocene and Pliocene, the soil biomarkers record a commensurate cooling for MAF dropping from 11 to 5°C. By the late Miocene the Antarctic Peninsula Ice sheet (Bart et al., 2005; Davies et al., 2012) had expanded onto the northern Antarctic Peninsula continental shelf (Smith & Anderson, 2010). However, in the time period sampled, glaciation was still contained on the Peninsula nearby and the site was underwater as we have microfossil (diatom) and biomarker input from marine productivity. There is winter sea ice present in this region in the Pliocene as indicated by sea-ice proximal species from the same cores (Anderson et al, 2011, Bohaty et al., 2011). Increased glacial erosion is apparent from glacial deposits on land 141 (Marenssi et al., 2010) and glacial strata in the James Ross Basin (Smith & Anderson, 2010). Glacial erosion from land is clearly evident from the presence of dropstones (Anderson et al., 2011; Fig. A.8) and we find clear evidence for reworked deposits based on the pollen, low CPI alkanes and hopane index in the Pliocene samples compared to other time slices from SHALDRIL II. Importantly both time periods are found within the same Hole 5D so there are no concerns about comparison of site of deposition, with material expected to still be delivered from the proximal, glaciated peninsula. However, icebergs may also have been delivering material from around the Weddell Sea for all the time periods when pebbles are present (Anderson et al., 2011). 4.6 Conclusions In this multi-biomarker study of the SHALDRIL II cores adjacent to the tip of the Antarctic Peninsula, we capture signals of terrestrial erosion of increasingly glacially-reworked sediments and track the cooling on land based on brGDGTs. In the Eocene, relatively warm summer (months above freezing) conditions occurred on the Antarctic Peninsula, Oligocene and even Miocene with MAF estimates of 11-12ºC, demonstrating how the peninsula warmth provided growing conditions for plants as refugia from the cooler conditions further south. As the iceberg delivery of pebbles in the Miocene and Pliocene attests, glacial advance was accompanied by cooling on land on the peninsula, and there was distal transport of terrestrial material from under the WAIS and EAIS, both of which were fully glaciated at that time (Anderson et al., 2011; Naish et al., 2009). Ocean waters off the Antarctic Peninsula at 63ºS cooled from 12.5ºC in the Oligocene to 2.8ºC in the Pliocene based on BAYSPAR reconstructions. This evidence for ocean cooling near the Antarctic Peninsula adds to sparse regional data on the Southern Ocean. While the cooling 142 broadly agrees with the expectation of global cooling and ice volume expansion reconstructed from the deep sea carbonates and the declining carbon dioxide levels in the atmosphere (Fig. 4), there are some significant regional differences. We find warm ocean waters persist well into Neogene glaciation, confirming earlier tentative multi-proxy temperature data from Seymour Island (Douglas et al., 2014). Icebergs calving into warm oceans and persisting to the tip of the Antarctic Peninsula, imply strong Weddell Sea circulation in iceberg alley as previously noted (Carter et al., 2017). These temperature constraints from land and sea add to sparse information for the Cenozoic cooling trend at the lowest latitude extent of the Antarctic continent. Acknowledgements We declare no financial conflicts of interests for any author or their affiliations. This research was funded by the U.S. National Science Foundation NSF-OPP-1908548 to SJF. This research used samples and/or data collected by the Nathaniel B. Palmer from SHALDRIL II, supported by funding from the U.S. National Science Foundation and other member nations. Samples were provided by Oregon State University Marine and Geology Repository. Thanks to Miguel Rincon for laboratory assistance in 2012 at USC, to Patrick Murphy for laboratory assistance with GDGT analyses at the University of Arizona in 2021, and to John Anderson and Julia Wellner for information about SHALDRIL II drilling and core sedimentology from their prior work. Data files are archived at the NOAA paleoclimatology database at https://www.ncdc.noaa.gov/paleo/study/35613 (Tibbett et al., 2022). 143 References Anagnostou, E., John, E. H., Babila, T. L., Sexton, P. F., Ridgwell, A., Lunt, D. J., Pearson, P. N., Chalk, T. B., Pancost, R. D., & Foster, G. L. (2020). Proxy evidence for state- dependence of climate sensitivity in the Eocene greenhouse. Nature Communications, 11(1), 4436. https://doi.org/10.1038/s41467-020-17887-x Anagnostou, E., John, E. H., Edgar, K. M., Foster, G. L., Ridgwell, A., Inglis, G. N., Pancost, R. D., Lunt, D. J., & Pearson, P. N. (2016). Changing atmospheric CO2 concentration was the primary driver of early Cenozoic climate. Nature, 533, 380. Anderson, J. B. (2006). SHALDRIL II 2006 NBP0602A Cruise Report. Rice University, Department of Earth Science, Houston, Texas. Anderson, J. B., Warny, S., Askin, R. A., Wellner, J. S., Bohaty, S. M., Kirshner, A. E., Livsey, D. N., Simms, A. R., Smith, T. R., Ehrmann, W., Lawver, L. A., Barbeau, D., Wise, S. W., Kulhanek, D. K., Weaver, F. M., & Majewski, W. (2011). Progressive Cenozoic cooling and the demise of Antarctica’s last refugium. Proceedings of the National Academy of Sciences, 108(28), 11356. https://doi.org/10.1073/pnas.1014885108 Anderson, J. B., & Wellner, J. S. (2012). SHALDRIL I and II: Drilling from the Research Vessel Icebreaker Nathaniel B. Palmer. Oceanography. Askin, R., & Raine, J. (2000). Oligocene and Early Miocene terrestrial palynology of the Cape Roberts Drillhole CRP-2/2A, Victoria Land Basin, Antarctica. Terra Antartica, 7(4), 493–501. Badger, M. P., Chalk, T. B., Foster, G. L., Bown, P. R., Gibbs, S. J., Sexton, P. F., Schmidt, D. N., Pälike, H., Mackensen, A., & Pancost, R. D. (2019). Insensitivity of alkenone carbon isotopes to atmospheric CO 2 at low to moderate CO 2 levels. Climate of the Past, 15(2), 539–554. Badger, M. P., Lear, C. H., Pancost, R. D., Foster, G. L., Bailey, T. R., Leng, M. J., & Abels, H. A. (2013). CO2 drawdown following the middle Miocene expansion of the Antarctic Ice Sheet. Paleoceanography, 28(1), 42–53. Bart, P. J., Egan, D., & Warny, S. A. (2005). Direct constraints on Antarctic Peninsula Ice Sheet grounding events between 5.12 and 7.94 Ma. Journal of Geophysical Research: Earth Surface, 110(F4). Bijl, P. K., Pross, J., Warnaar, J., Stickley, C. E., Huber, M., Guerstein, R., Houben, A. J. P., Sluijs, A., Visscher, H., & Brinkhuis, H. (2011). Environmental forcings of Paleogene Southern Ocean dinoflagellate biogeography. Paleoceanography, 26(1). https://doi.org/10.1029/2009PA001905 Bohaty, S. M., Kulhanek, D. K., Wise Jr, S. W., Jemison, K., Warny, S., & Sjunneskog, C. (2011). Age assessment of Eocene–Pliocene drill cores recovered during the SHALDRIL 144 II expedition, Antarctic Peninsula. Tectonic, Climatic, and Cryospheric Evolution of the Antarctic Peninsula, 63–113. Carter, A., Riley, T. R., Hillenbrand, C.-D., & Rittner, M. (2017). Widespread Antarctic glaciation during the Late Eocene. Earth and Planetary Science Letters, 458, 49–57. https://doi.org/10.1016/j.epsl.2016.10.045 Coxall, H. K., & Pearson, P. N. (2007). The Eocene-Oligocene transition. Deep Time Perspectives on Climate Change: Marrying the Signal From Computer Models and Biological Proxies, 351–387. Coxall, H. K., Wilson, P. A., Palike, H., Lear, C. H., & Backman, J. (2005). Rapid stepwise onset of Antarctic glaciation and deeper calcite compensation in the Pacific Ocean. Nature, 433(7021), 53–57. https://doi.org/10.1038/nature03135 Davies, B. J., Hambrey, M. J., Smellie, J. L., Carrivick, J. L., & Glasser, N. F. (2012). Antarctic Peninsula Ice Sheet evolution during the Cenozoic Era. Quaternary Science Reviews, 31, 30–66. https://doi.org/10.1016/j.quascirev.2011.10.012 De Jonge, C., Hopmans, E. C., Zell, C. I., Kim, J.-H., Schouten, S., & Sinninghe Damsté, J. S. (2014). Occurrence and abundance of 6-methyl branched glycerol dialkyl glycerol tetraethers in soils: Implications for palaeoclimate reconstruction. Geochimica et Cosmochimica Acta, 141, 97–112. https://doi.org/10.1016/j.gca.2014.06.013 Dearing Crampton-Flood, E., Tierney, J. E., Peterse, F., Kirkels, F. M. S. A., & Sinninghe Damsté, J. S. (2020). BayMBT: A Bayesian calibration model for branched glycerol dialkyl glycerol tetraethers in soils and peats. Geochimica et Cosmochimica Acta, 268, 142–159. https://doi.org/10.1016/j.gca.2019.09.043 Dingle, R. V., Marenssi, S. A., & Lavelle, M. (1998). High latitude Eocene climate deterioration: Evidence from the northern Antarctic Peninsula. Journal of South American Earth Sciences, 11(6), 571–579. https://doi.org/10.1016/S0895- 9811(98)00035-2 Douglas, P. M. J., Affek, H. P., Ivany, L. C., Houben, A. J. P., Sijp, W. P., Sluijs, A., Schouten, S., & Pagani, M. (2014). Pronounced zonal heterogeneity in Eocene southern high-latitude sea surface temperatures. Proceedings of the National Academy of Sciences, 111(18), 6582. https://doi.org/10.1073/pnas.1321441111 Edwards, L. E., Mudie, P. J., & de Vernal, A. (1991). Pliocene paleoclimatic reconstruction using dinoflagellate cysts: Comparison of methods. Quaternary Science Reviews, 10(2), 259–274. https://doi.org/10.1016/0277-3791(91)90024-O Feakins, S. J., Warny, S., & DeConto, R. M. (2014). Snapshot of cooling and drying before onset of Antarctic Glaciation. Earth and Planetary Science Letters, 404, 154–166. https://doi.org/10.1016/j.epsl.2014.07.032 145 Feakins, S. J., Warny, S., & Lee, J.-E. (2012). Hydrologic cycling over Antarctica during the middle Miocene warming. Nature Geoscience, 5(8), 557–560. https://doi.org/10.1038/NGEO1498 Flower, B. P., & Kennett, J. P. (1994). The middle Miocene climatic transition: East Antarctic ice sheet development, deep ocean circulation and global carbon cycling. Cenozoic Climate and Paleogeographic Changes in the Pacific Region, 108(3), 537–555. https://doi.org/10.1016/0031-0182(94)90251-8 Foster, G. L., Lear, C. H., & Rae, J. W. (2012). The evolution of pCO2, ice volume and climate during the middle Miocene. Earth and Planetary Science Letters, 341, 243–254. Francis, J., Ashworth, A., Cantrill, D., Crame, J., Howe, J., Stephens, R., Tosolini, A., & Thorn, V. (2008). 100 million years of Antarctic climate evolution: Evidence from fossil plants. Antarctica: A Keystone in a Changing World, 19–28. Francis, J. E., Marenssi, S., Levy, R., Hambrey, M., Thorn, V. C., Mohr, B., Brinkhuis, H., Warnaar, J., Zachos, J., Bohaty, S., & DeConto, R. (2008). Chapter 8 From Greenhouse to Icehouse – The Eocene/Oligocene in Antarctica. In F. Florindo & M. Siegert (Eds.), Developments in Earth and Environmental Sciences (Vol. 8, pp. 309–368). Elsevier. https://doi.org/10.1016/S1571-9197(08)00008-6 Fretwell, P., Pritchard, H. D., Vaughan, D. G., Bamber, J. L., Barrand, N. E., Bell, R., Bianchi, C., Bingham, R. G., Blankenship, D. D., Casassa, G., Catania, G., Callens, D., Conway, H., Cook, A. J., Corr, H. F. J., Damaske, D., Damm, V., Ferraccioli, F., Forsberg, R., … Zirizzotti, A. (2013). Bedmap2: Improved ice bed, surface and thickness datasets for Antarctica. Cryosphere, 7(1), 375–393. https://doi.org/10.5194/tc-7-375-2013 Goldner, A., Herold, N., & Huber, M. (2015). Antarctic glaciation caused ocean circulation changes at the Eocene-Oligocene transition (vol 511, pg 574, 2014). Nature, 519(7543), 378–378. https://doi.org/10.1038/nature14220 Greene, C. A., Gwyther, D. E., & Blankenship, D. D. (2017). Antarctic mapping tools for MATLAB. Computers & Geosciences, 104, 151–157. Greenop, R., Foster, G. L., Wilson, P. A., & Lear, C. H. (2014). Middle Miocene climate instability associated with high‐amplitude CO2 variability. Paleoceanography, 29(9), 845–853. Greenop, R., Sosdian, S. M., Henehan, M. J., Wilson, P. A., Lear, C. H., & Foster, G. L. (2019). Orbital forcing, ice volume, and CO2 across the Oligocene‐Miocene transition. Paleoceanography and Paleoclimatology, 34(3), 316–328. Griener, K. W., Nelson, D. M., & Warny, S. (2013). Declining moisture availability on the Antarctic Peninsula during the Late Eocene. Palaeogeography, Palaeoclimatology, Palaeoecology, 383, 72–78. 146 Hartman, J. D., Sangiorgi, F., Salabarnada, A., Peterse, F., Houben, A. J. P., Schouten, S., Brinkhuis, H., Escutia, C., & Bijl, P. K. (2018). Paleoceanography and ice sheet variability offshore Wilkes Land, Antarctica – Part 3: Insights from Oligocene–Miocene TEX86-based sea surface temperature reconstructions. Climate of the Past, 14(9), 1275– 1297. https://doi.org/10.5194/cp-14-1275-2018 Henehan, M. J., Edgar, K. M., Foster, G. L., Penman, D. E., Hull, P. M., Greenop, R., Anagnostou, E., & Pearson, P. N. (2020). Revisiting the Middle Eocene Climatic Optimum “Carbon Cycle Conundrum” with new estimates of atmospheric pCO2 from boron isotopes. Paleoceanography and Paleoclimatology, 35(6), e2019PA003713. Hopmans, E. C., Weijers, J. W. H., Schefuß, E., Herfort, L., Sinninghe Damsté, J. S., & Schouten, S. (2004). A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraether lipids. Earth and Planetary Science Letters, 224(1), 107–116. https://doi.org/10.1016/j.epsl.2004.05.012 Houben, A. J. P., Bijl, P. K., Guerstein, G. R., Sluijs, A., & Brinkhuis, H. (2011). Malvinia escutiana, a new biostratigraphically important Oligocene dinoflagellate cyst from the Southern Ocean. Review of Palaeobotany and Palynology, 165(3), 175–182. https://doi.org/10.1016/j.revpalbo.2011.03.002 Houben, A. J. P., Bijl, P. K., Sluijs, A., Schouten, S., & Brinkhuis, H. (2019, May 1). Late Eocene Southern Ocean Cooling and Invigoration of Circulation Preconditioned Antarctica for Full‐Scale Glaciation. Geochemistry, Geophysics, Geosystems. https://doi.org/10.1029/2019GC008182 Huguet, C., Hopmans, E. C., Febo-Ayala, W., Thompson, D. H., Sinninghe Damsté, J. S., & Schouten, S. (2006). An improved method to determine the absolute abundance of glycerol dibiphytanyl glycerol tetraether lipids. Organic Geochemistry, 37(9), 1036– 1041. https://doi.org/10.1016/j.orggeochem.2006.05.008 Hunt, R. J., & Poole, I. (2003). Paleogene West Antarctic climate and vegetation history in light of new data from King George Island. Inglis, G. N., Naafs, B. D. A., Zheng, Y., McClymont, E. L., Evershed, R. P., & Pancost, R. D. (2018). Distributions of geohopanoids in peat: Implications for the use of hopanoid- based proxies in natural archives. Geochimica et Cosmochimica Acta, 224, 249–261. https://doi.org/10.1016/j.gca.2017.12.029 Judd, E. J., Ivany, L. C., DeConto, R. M., Halberstadt, A. R. W., Miklus, N. M., Junium, C. K., & Uveges, B. T. (2019). Seasonally Resolved Proxy Data From the Antarctic Peninsula Support a Heterogeneous Middle Eocene Southern Ocean. Paleoceanography and Paleoclimatology, 34(5), 787–799. https://doi.org/10.1029/2019PA003581 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, 1(5), 329–334. https://doi.org/10.1038/ngeo179 147 Lauretano, V., Kennedy-Asser, A. T., Korasidis, V. A., Wallace, M. W., Valdes, P. J., Lunt, D. J., Pancost, R. D., & Naafs, B. D. A. (2021). Eocene to Oligocene terrestrial Southern Hemisphere cooling caused by declining pCO2. Nature Geoscience, 14(9), 659–664. https://doi.org/10.1038/s41561-021-00788-z Leutert, T. J., Auderset, A., Martínez-García, A., Modestou, S., & Meckler, A. N. (2020). Coupled Southern Ocean cooling and Antarctic ice sheet expansion during the middle Miocene. Nature Geoscience, 13(9), 634–639. https://doi.org/10.1038/s41561-020-0623- 0 Levy, R., Harwood, D., Florindo, F., Sangiorgi, F., Tripati, R., Von Eynatten, H., Gasson, E., Kuhn, G., Tripati, A., & DeConto, R. (2016). Antarctic ice sheet sensitivity to atmospheric CO2 variations in the early to mid-Miocene. Proceedings of the National Academy of Sciences, 113(13), 3453–3458. Lewis, A. R., Marchant, D. R., Ashworth, A. C., Hedenäs, L., Hemming, S. R., Johnson, J. V., Leng, M. J., Machlus, M. L., Newton, A. E., Raine, J. I., Willenbring, J. K., Williams, M., & Wolfe, A. P. (2008). Mid-Miocene cooling and the extinction of tundra in continental Antarctica. Proceedings of the National Academy of Sciences, 105(31), 10676. https://doi.org/10.1073/pnas.0802501105 Liu, Z., Pagani, M., Zinniker, D., DeConto, R., Huber, M., Brinkhuis, H., Shah, S. R., Leckie, R. M., & Pearson, A. (2009). Global Cooling During the Eocene-Oligocene Climate Transition. Science, 323(5918), 1187. https://doi.org/10.1126/science.1166368 Macphail, M. K., & Truswell, E. M. (2004). Palynology of Neogene slope and rise deposits from ODP Sites 1165 and 1167, East Antarctica. Proceedings of the Ocean Drilling Program. Scientific Results, 188, 1–20. Marenssi, S. A., Casadío, S., & Santillana, S. N. (2010). Record of Late Miocene glacial deposits on Isla Marambio (Seymour Island), Antarctic Peninsula. Antarctic Science, 22(2), 193–198. Cambridge Core. https://doi.org/10.1017/S0954102009990629 Marschalek, J. W., Zurli, L., Talarico, F., van de Flierdt, T., Vermeesch, P., Carter, A., Beny, F., Bout-Roumazeilles, V., Sangiorgi, F., Hemming, S. R., Pérez, L. F., Colleoni, F., Prebble, J. G., van Peer, T. E., Perotti, M., Shevenell, A. E., Browne, I., Kulhanek, D. K., Levy, R., … IODP Expedition 374. (2021). A large West Antarctic Ice Sheet explains early Neogene sea-level amplitude. Nature, 600(7889), 450–455. https://doi.org/10.1038/s41586-021-04148-0 McKay, R. M., Escutia, C., Santis, L. D., Donda, F., Duncan, B., Gohl, K., Gulick, S., Hernández-Molina, J., Hillenbrand, C.-D., Hochmuth, K., Kim, S., Kuhn, G., Larter, R., Leitchenkov, G., Levy, R. H., Naish, T. R., O’Brien, P., Pérez, L. F., Shevenell, A. E., & Williams, T. (2022). Chapter 3—Cenozoic history of Antarctic glaciation and climate from onshore and offshore studies. In F. Florindo, M. Siegert, L. D. Santis, & T. Naish (Eds.), Antarctic Climate Evolution (Second Edition) (Second Edition, pp. 41–164). Elsevier. https://doi.org/10.1016/B978-0-12-819109-5.00008-6 148 McKay, R., Naish, T., Carter, L., Riesselman, C., Dunbar, R., Sjunneskog, C., Winter, D., Sangiorgi, F., Warren, C., Pagani, M., Schouten, S., Willmott, V., Levy, R., DeConto, R., & Powell, R. D. (2012). Antarctic and Southern Ocean influences on Late Pliocene global cooling. Proceedings of the National Academy of Sciences, 109(17), 6423. https://doi.org/10.1073/pnas.1112248109 Miller, K. G., Browning, J. V., Schmelz, W. J., Kopp, R. E., Mountain, G. S., & Wright, J. D. (2020). Cenozoic sea-level and cryospheric evolution from deep-sea geochemical and continental margin records. Science Advances, 6(20), eaaz1346. https://doi.org/10.1126/sciadv.aaz1346 Miller, K. G., Wright, J. D., & Fairbanks, R. G. (1991). Unlocking the Ice House: Oligocene- Miocene oxygen isotopes, eustasy, and margin erosion. Journal of Geophysical Research: Solid Earth, 96(B4), 6829–6848. https://doi.org/10.1029/90JB02015 Mohr, B. (1990). Eocene and Oligocene Sporomorphs and Dinoflagellate Cysts from Leg 113 Drill Sites|Weddell Sea|Antarctica. Proc., Scientific Results, ODP, Leg 113, Weddell Sea, Antarctica, 120. https://doi.org/10.2973/odp.proc.sr.113.140.1990 Naish, T., Powell, R., Levy, R., Wilson, G., Scherer, R., Talarico, F., Krissek, L., Niessen, F., Pompilio, M., Wilson, T., Carter, L., DeConto, R., Huybers, P., McKay, R., Pollard, D., Ross, J., Winter, D., Barrett, P., Browne, G., … Williams, T. (2009). Obliquity-paced Pliocene West Antarctic ice sheet oscillations. Nature, 458(7236), 322–328. https://doi.org/10.1038/nature07867 O’Brien, C. L., Huber, M., Thomas, E., Pagani, M., Super, J. R., Elder, L. E., & Hull, P. M. (2020). The enigma of Oligocene climate and global surface temperature evolution. Proceedings of the National Academy of Sciences, 117(41), 25302. https://doi.org/10.1073/pnas.2003914117 Pagani, M., Huber, M., Liu, Z., Bohaty, S. M., Henderiks, J., Sijp, W., Krishnan, S., & DeConto, R. M. (2011). The Role of Carbon Dioxide During the Onset of Antarctic Glaciation. Science, 334(6060), 1261. https://doi.org/10.1126/science.1203909 Pagani, M., Liu, Z., LaRiviere, J., & Ravelo, A. C. (2010). High Earth-system climate sensitivity determined from Pliocene carbon dioxide concentrations. Nature Geoscience, 3(1), 27–30. 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(5734), 600. https://doi.org/10.1126/science.1110063 Passchier, S., Bohaty, S. M., Jimenez-Espejo, F. J., Pross, J., Roehl, U., van de Flierdt, T., Escutia, C., & Brinkhuis, H. (2013). Early Eocene to middle Miocene cooling and aridification of East Antarctica. Geochemistry Geophysics Geosystems, 14(5), 1399– 1410. https://doi.org/10.1002/ggge.20106 149 Passchier, S., Ciarletta, D. J., Miriagos, T. E., Bijl, P. K., & Bohaty, S. M. (2017). An Antarctic stratigraphic record of stepwise ice growth through the Eocene-Oligocene transition. Geological Society of America Bulletin, 129(3–4), 318–330. https://doi.org/10.1130/B31482.1 Pearson, P. N., Foster, G. L., & Wade, B. S. (2009). Atmospheric carbon dioxide through the Eocene–Oligocene climate transition. Nature, 461(7267), 1110–1113. https://doi.org/10.1038/nature08447 Petersen, S. V., & Schrag, D. P. (2015). Antarctic ice growth before and after the Eocene- Oligocene transition: New estimates from clumped isotope paleothermometry: Antarctic ice growth at the E/O transition. Paleoceanography, 30(10), 1305–1317. https://doi.org/10.1002/2014PA002769 Plancq, J., Mattioli, E., Pittet, B., Simon, L., & Grossi, V. (2014). Productivity and sea- surface temperature changes recorded during the late Eocene–early Oligocene at DSDP Site 511 (South Atlantic). Palaeogeography, Palaeoclimatology, Palaeoecology, 407, 34–44. https://doi.org/10.1016/j.palaeo.2014.04.016 Poole, I., Cantrill, D., & Utescher, T. (2005). A multi-proxy approach to determine Antarctic terrestrial palaeoclimate during the Late Cretaceous and Early Tertiary. Palaeogeography, Palaeoclimatology, Palaeoecology, 222(1), 95–121. https://doi.org/10.1016/j.palaeo.2005.03.011 Pound, M. J., & Salzmann, U. (2017). Heterogeneity in global vegetation and terrestrial climate change during the late Eocene to early Oligocene transition. Scientific Reports, 7, 43386. https://doi.org/10.1038/srep43386 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) cored by the Cape Roberts Project, Victoria Land Basin, Antarctica. Antarctic Climate Evolution: Geological Records from the Margin and Modelling, 231(1), 41–57. https://doi.org/10.1016/j.palaeo.2005.07.025 Rae, J. W. B., Zhang, Y. G., Liu, X., Foster, G. L., Stoll, H. M., & Whiteford, R. D. M. (2021). Atmospheric CO2 over the Past 66 Million Years from Marine Archives. Annual Review of Earth and Planetary Sciences, 49(1), 609–641. https://doi.org/10.1146/annurev-earth-082420-063026 Raine, J. (1998). Terrestrial palynomorphs from Cape Roberts Project drillhole CRP-1, Ross Sea, Antarctica. Terra Antartica, 5(3), 539–548. Rose, K. C., Ferraccioli, F., Jamieson, S. S. R., Bell, R. E., Corr, H., Creyts, T. T., Braaten, D., Jordan, T. A., Fretwell, P. T., & Damaske, D. (2013). Early East Antarctic Ice Sheet growth recorded in the landscape of the Gamburtsev Subglacial Mountains. Earth and Planetary Science Letters, 375, 1–12. https://doi.org/10.1016/j.epsl.2013.03.053 150 Salzmann, U., Riding, J. B., Nelson, A. E., & Smellie, J. L. (2011). How likely was a green Antarctic Peninsula during warm Pliocene interglacials? A critical reassessment based on new palynofloras from James Ross Island. Special Issue: Climate and Seasonality in a Pliocene Warm World, 309(1), 73–82. https://doi.org/10.1016/j.palaeo.2011.01.028 Schouten, S., Huguet, C., Hopmans, E. C., Kienhuis, M. V. M., & Sinninghe Damsté, J. S. (2007). Analytical Methodology for TEX86 Paleothermometry by High-Performance Liquid Chromatography/Atmospheric Pressure Chemical Ionization-Mass Spectrometry. Analytical Chemistry, 79(7), 2940–2944. https://doi.org/10.1021/ac062339v Sessions, A. L., Zhang, L., Welander, P. V., Doughty, D., Summons, R. E., & Newman, D. K. (2013). Identification and quantification of polyfunctionalized hopanoids by high temperature gas chromatography-mass spectrometry. Organic Geochemistry, 56, 120– 130. PubMed. https://doi.org/10.1016/j.orggeochem.2012.12.009 Smith, R. T., & Anderson, J. B. (2010). Ice-sheet evolution in James Ross Basin, Weddell Sea margin of the Antarctic Peninsula: The seismic stratigraphic record. GSA Bulletin, 122(5–6), 830–842. https://doi.org/10.1130/B26486.1 Sosdian, S. M., Greenop, R., Hain, M., Foster, G. L., Pearson, P. N., & Lear, C. H. (2018). Constraining the evolution of Neogene ocean carbonate chemistry using the boron isotope pH proxy. Earth and Planetary Science Letters, 498, 362–376. Thompson, N., Salzmann, U., López-Quirós, A., Bijl, P. K., Hoem, F. S., Etourneau, J., Sicre, M.-A., Roignant, S., Hocking, E., Amoo, M., & Escutia, C. (2022). Vegetation change across the Drake Passage region linked to late Eocene cooling and glacial disturbance after the Eocene–Oligocene transition. Climate of the Past, 18(2), 209–232. https://doi.org/10.5194/cp-18-209-2022 Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., & Feakins, S. J. (2021). Late Eocene Record of Hydrology and Temperature From Prydz Bay, East Antarctica. Paleoceanography and Paleoclimatology, 36(4), e2020PA004204. https://doi.org/10.1029/2020PA004204 Tierney, J. E., & Tingley, M. P. (2014). A Bayesian, spatially-varying calibration model for the TEX86 proxy. Geochimica et Cosmochimica Acta, 127, 83–106. https://doi.org/10.1016/j.gca.2013.11.026 Truswell, E. M., & Macphail, M. K. (2009). Polar forests on the edge of extinction: What does the fossil spore and pollen evidence from East Antarctica say? Australian Systematic Botany, 22(2), 57–106. https://doi.org/10.1071/SB08046 Uemura, H., & Ishiwatari, R. (1995). Identification of unusual 17β(H)-moret-22(29)-ene in lake sediments. Organic Geochemistry, 23(7), 675–680. https://doi.org/10.1016/0146- 6380(95)00036-E Warny, S., & Askin, R. (2011). Last remnants of Cenozoic vegetation and organic-walled phytoplankton in the Antarctic Peninsula’s icehouse world. In Tectonic, climatic, and 151 cryospheric evolution of the Antarctic peninsula (Vol. 63, pp. 167–192). American Geophysical Union Washington, DC. Warny, S., Askin, R. A., Hannah, M. J., Mohr, B. A. R., Raine, J. I., Harwood, D. M., Florindo, F., & the SMS Science Team. (2009). Palynomorphs from a sediment core reveal a sudden remarkably warm Antarctica during the middle Miocene. Geology, 37(10), 955–958. https://doi.org/10.1130/G30139A.1 Warny, S., Kymes, C. M., Askin, R., Krajewski, K. P., & Tatur, A. (2019). Terrestrial and marine floral response to latest Eocene and Oligocene events on the Antarctic Peninsula. Palynology, 43(1), 4–21. https://doi.org/10.1080/01916122.2017.1418444 Warny, S. S., & Askin, R. R. (2011). Vegetation and Organic-Walled Phytoplankton at the End of the Antarctic Greenhouse World: Latest Eocene Cooling Events. In Tectonic, Climatic, and Cryospheric Evolution of the Antarctic Peninsula (pp. 193–210). American Geophysical Union (AGU). https://doi.org/10.1029/2010SP000965 Westerhold, T., Marwan, N., Drury, A. J., Liebrand, D., Agnini, C., Anagnostou, E., Barnet, J. S. K., Bohaty, S. M., Vleeschouwer, D. D., Florindo, F., Frederichs, T., Hodell, D. A., Holbourn, A. E., Kroon, D., Lauretano, V., Littler, K., Lourens, L. J., Lyle, M., Pälike, H., … Zachos, J. C. (2020). An astronomically dated record of Earth’s climate and its predictability over the last 66 million years. Science, 369(6509), 1383–1387. https://doi.org/10.1126/science.aba6853 Wilson, D. S., Pollard, D., DeConto, R. M., Jamieson, S. S. R., & Luyendyk, B. P. (2013). Initiation of the West Antarctic Ice Sheet and estimates of total Antarctic ice volume in the earliest Oligocene. Geophysical Research Letters, 40(16), 4305–4309. https://doi.org/10.1002/grl.50797 Zhang, Y. G., Pagani, M., Liu, Z., Bohaty, S. M., & DeConto, R. (2013). A 40-million-year history of atmospheric CO2. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 371(2001), 20130096. https://doi.org/10.1098/rsta.2013.0096 Zhang, Y. G., Pagani, M., & Wang, Z. (2016). Ring Index: A new strategy to evaluate the integrity of TEX86 paleothermometry. Paleoceanography, 31(2), 220–232. https://doi.org/10.1002/2015PA002848 Zhang, Y. G., Zhang, C. L., Liu, X.-L., Li, L., Hinrichs, K.-U., & Noakes, J. E. (2011). Methane Index: A tetraether archaeal lipid biomarker indicator for detecting the instability of marine gas hydrates. Earth and Planetary Science Letters, 307(3), 525–534. https://doi.org/10.1016/j.epsl.2011.05.031 152 Chapter 5 Proxy-Model Comparison for the Eocene-Oligocene Transition in Southern High Latitudes This chapter is in preparation for submission to AGU Paleoceanography and Paleoclimatology: Tibbett, E. J., Burls, N.J., Hutchinson, D.K., & Feakins, S. J. Proxy-model comparison for the Eocene-Oligocene Transition in southern high latitudes. Abstract The Eocene-Oligocene Transition (EOT) marks the shift from the greenhouse to icehouse conditions at 34 Ma, when a permanent ice sheet developed on the Antarctic continent. Here we compile published proxy records on and around Antarctica for time slices of the late Eocene and early Oligocene bracketing the transition. Compiled proxy records for sea surface temperatures (SST) from the Southern Ocean cool by 0 to 7°C and surface air temperatures (SAT) by 0 to 3°C. Some of the variance is likely due to proxy uncertainties, archive averaging and geographic differences in cooling across the transition. Proxy data were compared to climate model simulations for each time interval, and the difference between the two experiments to achieve evaluation of the magnitude of cooling across the EOT. While the EOT climate simulations were initially forced with a halving of pCO2, this is larger than the expected forcing and results in a larger cooling in the model experiments than supported by the proxy data. In this study we revisited the model outputs in comparison to the proxy data and we scaled the output to identify the magnitude of pCO2 change needed to drive a commensurate change in temperature. Assuming a post-EOT pCO2 level of 560 ppmv, the SAT proxies require on average a decrease of 199 ppmv (101 to 301 ppmv across individual models), and the SST proxies require a larger 153 drop (149 to 560 ppmv). The larger forcing required to reconcile with SSTs is in part a function of the larger cooling recorded by the ocean proxies and the heterogenous regional cooling throughout the Southern Ocean. Proxy-model intercomparisons show that pCO2 is the primary driver of declining temperatures across the EOT recorded by proxies on the continents and in the oceans in the southern high latitudes, and the magnitude of that forcing was likely consistent with the 200 ppmv estimate from the latest pCO2 proxy compilations. 5.1 Introduction The Eocene-Oligocene transition spans 34.4 to 33.7 Ma (Coxall & Pearson, 2007; Hutchinson et al., 2021; Katz et al., 2008) and marks the growth of permanent ice sheets on Antarctica. This transition includes a two-step increase in benthic foraminiferal δ 18 O by 1.5‰ (Coxall et al., 2005). The first step, the Earliest Oligocene Isotope Step (Hutchinson et al., 2021), increases by 0.7‰ denoting the expansion of the Antarctic ice sheet. Estimates for the size of the ice sheet based on the benthic δ 18 O signal suggest an ice sheet 60-130% of the modern East Antarctic Ice Sheet (Bohaty et al., 2012b; Lear et al., 2008). This transition is noted by a decrease in pCO2 (Rae et al., 2021), temperature (Coxall & Pearson, 2007; Hutchinson et al., 2021; Lear et al., 2008; Liu et al., 2009), and sea level (Houben et al., 2012; Miller et al., 2020). An early hypothesis for the growth of permanent ice sheets on Antarctica was that gateway openings at the Drake Passage and Tasman Gateway led to thermal isolation of Antarctica (Kennett, 1977). Several ocean-only or intermediate complexity climate models suggest that the opening or deepening of the Southern Ocean gateways could have a local cooling effect close to the Antarctic coast (Sauermilch et al., 2021; Sijp et al., 2009). However, the accumulating proxy records and coupled climate modelling experiments have indicated that the gateway hypothesis does not fully explain the global cooling experienced at the EOT (e.g., Hutchinson et al., 2021). 154 Ocean circulation proxy reconstructions indicate that the timing doesn’t match the proposed mechanism. Deep water currents through the Tasman Gateway were first established around 30 Ma (Scher et al., 2015), i.e., after the EOT. For the Drake Passage, full opening may have occurred even later, in the Miocene (Dalziel et al., 2013). A growing consensus is that a decrease in pCO2 across the EOT is the primary driver for the EOT and temperature decrease globally (DeConto & Pollard, 2003; Goldner et al., 2014; Hutchinson et al., 2021; Lauretano et al., 2021; Pagani et al., 2011). Previous model proxy comparisons indicating a decrease in pCO2 by 40% can explain the global temperature shift (Hutchinson et al., 2021). Recent pCO2 compilations (Rae et al., 2021) constrain a decrease in pCO2 from 980 to 830 ppmv from boron isotopes (16% decrease) (Anagnostou et al., 2016, 2020; Henehan et al., 2020; Pearson et al., 2009) and a decrease from 660 to 520 ppmv from alkenones (27% decreases) across the EOT (Pagani et al., 2005, 2011). Both proxies converge on the magnitude of the decrease being just 140-150 ppmv between the late Eocene and the early Oligocene, and when averaging across both proxies there roughly a 25% decrease across the Eocene-Oligocene transition (Rae et al., 2021). Although carbon dioxide has been established as the leading cause, additional feedbacks are invoked from both the ice-albedo feedback and gateway-induced changes to deep-water formation (Goldner et al., 2014). Several coupled climate model studies have found a shift from South Atlantic to South Pacific deep-water formation across the EOT due to Southern Ocean gateway opening (Kennedy et al., 2015; Toumoulin et al., 2020). Furthermore, deep water circulation proxies suggest that there was an expansion of North Atlantic Deep Water formation around the EOT (Coxall et al., 2018), supported by paleogeographic and modelling evidence of the Arctic becoming isolated from the North Atlantic (Hutchinson et al., 2019; Vahlenkamp et al., 2018). These studies suggest that 155 ocean gateway and ice sheet changes could be involved in driving the observed changes at the EOT, although declining pCO2 is the only mechanism proven to cause global cooling. Climate models allow the drivers of change to be tested. Inter-model differences in boundary conditions (e.g., continental configuration) and parameterization schemes can lead to different outcomes. Multi-model comparisons can test the robustness of hypotheses for the transition to these differences in model formulation. One surprising feature of climate model experiments, is the finding of a smaller decrease in surface air temperatures at higher latitudes in comparison to mid-latitudes across the EOT (Kennedy-Asser et al., 2020). Model experiments also indicate Southern Ocean SSTs cooled more than the land at the same latitude. SST proxies indicate a global average cooling of 2.5°C across the EOT and regional differences in cooling ranging from 0 to 8°C (Hutchinson et al., 2021). Compiled global SAT proxy records suggest a global mean cooling of 2.3°C with latitudinal and regional differences in cooling from 0 to 8°C (Hutchinson et al., 2021). However, proxy records are concentrated in northern mid-latitudes with limited records from the Southern Hemisphere and few from Antarctica. The sparse coverage of proxy records in the Southern Hemisphere and from Antarctica has hampered past efforts to evaluate model outputs. We now have more temperature records to assess the magnitude of the SST and SAT shift across the EOT surrounding the Southern Ocean. For example, there are now brGDGT-based temperature estimates on both sides of the Southern Ocean from Prydz Bay (Tibbett et al., 2021) and South Australia (Lauretano et al., 2021). We add these new records to compiled proxies and model experiments for the EOT (compiled by Hutchinson et al., 2021). That study compiled proxy data globally, whereas we take a more in-depth look at the southern high latitudes and Antarctica for this proxy-model comparison. While the individual model experiments generally 156 used a halving of carbon dioxide to force a large EOT response, we scale the model experiments to identify the CO2 forcing required to better reproduce the temperature anomaly across the transition observed in the proxy data in the high southern latitudes. The focused proxy-model comparison allows us to identify differences within the region in the proxies and in the climate model experiments to reach new understanding of the forcing and response during the Eocene to Oligocene Transition on Antarctica and in the Southern Ocean region. 5.2 Methods 5.2.1 Proxy data Proxy temperature records were collected south of 45°S based on paleolatitudes for the late Eocene (40 to 34 Ma) and early Oligocene (34 to 30 Ma) collating records from land and sea for SAT and SST (Tables 5.1 and 5.2). Proxy methods for the temperature reconstructions are noted and where appropriate the data was recalibrated to the latest methods for compatibility within the compilation. Averaging may attenuate the magnitude of the abrupt changes in some cases where a rebound in temperature occurs post-EOT (Bohaty et al., 2012a; Tibbett et al., 2021). At Prydz Bay, SST cooling at the EOT was 4°C and SAT cooled by 5°C (Tibbett et al., 2021), but by averaging across the record the decrease in temperature is 2.2°C and 3.2°C for SST and SAT respectively which gives a dampened impact of 30-45%. Land temperature proxy reconstructions of surface air temperatures (SAT, Table 5.1) come from pollen assemblages (Francis et al., 2008; Hunt & Poole, 2003; Macphail & Truswell, 2004; Poole et al., 2005; Truswell & Macphail, 2009), the bacterial biomarker proxy indices of the branched Glycerol Dialkyl Glycerol Tetraethers (brGDGTs) with the more recent BayMBT (Bayesian regression model of the Methylation of Branched Tetraethers index)(Lauretano et al., 2021; 157 Tibbett et al., 2021) and the earlier MBT/CBT (Cyclization of Branched Tetraethers) (Douglas et al., 2014), as well as mineral weathering via the S-index (Passchier et al., 2013, 2017). Both records generated from BayMBT are interpreted with the calibration to mean annual surface air temperature (Dearing Crampton-Flood et al., 2020) for the purposes of consistency with mean annual reporting conventions for this proxy-model comparison, however we note that this index can also be calibrated to months above freezing (MAF) based on the understanding that soil microbial communities are unlikely to be active below freezing (Deng et al., 2016; Weijers et al., 2007, 2011). However, for both the Eocene and the Oligocene cases MAF approximates mean annual air temperature in this case. The MBT/CBT record cannot be recalibrated due to the lack of separation of the 5 and 6 methyl isomers with the laboratory methods in use at that time. The SST records (Table 5.2) derive from the archaeal membrane lipid TEX86 index (Douglas et al., 2014; Lauretano et al., 2021; Tibbett et al., 2021), the haptophyte algal biomarker U k’ 37 index (Houben et al., 2019; Pagani et al., 2011) produced by the reticulofenestrids in the Eocene and Oligocene (Henderiks & Pagani, 2008) and carbonates with temperatures derived from clumped isotopes Δ47 values measured on shallow coastal bivalves (Douglas et al., 2014; Petersen & Schrag, 2015), and δ 18 O value of planktonic foraminifera (Zachos et al., 1994). For Douglas et al., 2014 the TEX86 SSTs were reevaluated using, BAYSPAR (Bayesian, Spatially-Varying Regression calibration for TEX86) (Tierney & Tingley, 2014) with a prior of 13°C and a standard deviation of 15°C. Other TEX86 records from the Southern Ocean were either originally calibrated by BAYSPAR (Lauretano et al., 2021; Tibbett et al., 2021), or were recently reevaluated using BAYSPAR (Lauretano et al., 2021) with priors ranging from 12 to 21°C and a standard deviation of 20°C. 158 Proxy uncertainty ranges vary by proxy and region. For temperature values generated from BAYSPAR and BayMBT, one standard deviation varies between 3 to 4°C (Lauretano et al., 2021; Tibbett et al., 2021). For SAT generated from S-index the proxy standard deviation is 3.6°C (Passchier et al., 2013, 2017). The U k’ 37 index has a standard error of 1.5°C (Houben et al., 2019; Müller et al., 1998). For clumped isotopes temperature uncertainty is 2.5°C (Douglas et al., 2014) and the uncertainty for MBT/CBT is 5.5°C (Weijers et al., 2011). The standard deviation for each time slice and site was calculated (Table 5.1 and 5.2) to account for the averaging across time. Multiple proxies were applied to the same site previously and can be compared at each location providing a proxy-proxy discrepancy. For Prydz Bay the difference between the two SAT proxies (BayMBT and the s-index) is 3-4°C for each time slice which is attributed to a difference in sourcing between the two proxies (Tibbett et al., 2021). Both DSDP Site 511 and Site 277 have a consistent SST proxy-proxy discrepancy between BAYSPAR and U k’ 37 of ~2°C. Although consistent their temperature difference between the Eocene and Oligocene varies with DSDP Site 511 cooling by ~3°C for BAYSPAR and ~7°C for U k’ 37 (Houben et al., 2019; Lauretano et al., 2021). At Site 277 the cooling is more comparable ranging from 0.4 to 1.0°C (Lauretano et al., 2021; Pagani et al., 2011). Although both are reported and are used in this study as SSTs, production of isoprenoid GDGTs (isoGDGTs) which are used for TEX86 can be produced in the subsurface (Lopes dos Santos et al., 2010; Schouten et al., 2013) while haptophyte algae which produce alkenones utilized for U k’ 37 are primary producers and are found in the photic zone (Popp et al., 2006; Volkman et al., 1980). Thaumarchaeota (producers of isoGDGTs) was found to be limited in the Antarctic summer surface water and are more abundant in the subsurface (Kalanetra et al., 2009) suggesting that proxy-proxy discrepancies may in part arise from different depth habitats in the water column. 159 Table 5.1. Surface air temperature proxy compilation for the Eocene and Oligocene with standard deviations from averaging across time. * Douglas et al., (2014) also reported MBT′/CBT, excluded as unrealistically warm**Pollen- based temperature estimates are reported here as SAT. 160 Table 5.2. Sea surface temperature proxy compilation for the Eocene and Oligocene with standard deviations from averaging across time * DSDP Site 511 (Houben et al., 2019) updated with ages from Lauretano et al., (2021) 5.2.2 Models We re-use the ensemble of model experiments gathered onto a uniform grid by Hutchinson et al., (2021). The compiled experiments include two broad groupings 4x CO2 (Eocene like) and 2x CO2 (Oligocene like) each run without ice sheets to isolate only the effects of changing pCO2 (Table 5.3). Additional model runs were included for a subset of models which included the paleogeography changes across the EOT (CESM_H, GFDL CM2.1, HadCM3BL, FOAM, UVic, NorESM-L), and the inclusion of an ice sheet (CESM_H, FOAM, HadCM3BL) (Table 5.3). While the NorESM-L model experiments run with a pCO2 drop from 980 to 560 ppmv, this was scaled by Hutchinson et al., (2021) to match the 4x/2x simulations in the other models and applied here for consistency to compare the model temperature difference (O-E). The summary of the model parameters can be found in Table 5.3 and detailed model run information for each 161 of the models in the ensemble can be found in the Hutchinson et al., (2021; and references therein). Here, we scale the model outputs to the proxy temperature differences to identify the pCO2 decrease across the EOT as will be explained further in Section 5.3.2.4. Where land temperature proxies (e.g., soil bacterial biomarkers or soil weathering indicators) were recovered from marine sedimentary archives, we inferred sourcing from the adjacent continent. Source regions were defined on the adjacent land mass, averaging the modelled surface temperatures within an area reflective of Antarctic drainage basins. For SST proxies we assume they capture temperatures in the overlying water column at the marine core site and thus compare to the nearest grid point within the model. In cases where proxy records from marine cores appeared to plot “on land” due to modelled coastline imprecision, we obtained comparison points from the nearest ocean grid cell for comparison. The proxy-model intercomparison differences are expressed as root mean square error (RMSE) (equation 1) with n as the number of proxies. This was assessed for each model scenario (Eocene and Oligocene) and the difference between the two. 𝑅𝑅 𝑀𝑀 𝑅𝑅 𝑇𝑇 = � ∑ ( 𝑠𝑠𝑖𝑖 𝑚𝑚 𝐶𝐶𝑠𝑠 − 𝑠𝑠 𝐶𝐶𝑖𝑖 𝑝𝑝𝑝𝑝 ) 2 𝑛𝑛 𝑖𝑖 = 1 𝑛𝑛 (1) 162 Table 5.3. Model run parameters used in this comparison. For models with more than one parameter all other factors held constant. NorESM-L was scaled to 1120 ppmv for this comparison. For the ice/no ice runs pCO2 is held constant at 560 ppmv. For the paleogeography runs the pCO2 for both pre- and post-EOT is 1120 ppmv for CESM_H, 560 ppmv for FOAM, HadCM3BL and NorESM-L, 800 ppmv for GFDL CM2.1, and 1600 ppmv for UVic. Drake = Drake Passage, Tasman = Tasman Gateway, WA = West Antarctica, Olig = Oligocene. 163 Figure 5.1. Southern hemisphere surface air model temperatures (SAT) for the a) Eocene (4x pCO2), b) Oligocene (2x pCO2 model runs), and c) the difference across the transition (2x-4x) showing results for the unscaled multi model ensemble mean. The circles correspond to proxy mean annual air temperature records while the dotted areas show the source area used to compare the model temperature to the proxy record. d) The RMSE for pCO2 model runs for SAT, for individual model mapped output see Fig. A.10. Red lines are the RMSE for each model after the pCO2 scaling for RMSE values see Table 5.3. 164 Figure 5.2. Southern Ocean sea surface temperatures (SST) for the a) Eocene (4x pCO2), b) Oligocene (2x pCO2 model runs), and c) the difference across the transition (2x-4x) showing results for the unscaled multi model ensemble mean. The circles correspond to proxy mean annual air temperature records while the dotted areas show the source area used to compare the model temperature to the proxy record. d) Summarizing the RMSE for pCO2 model runs for SAT, for individual model mapped output see Fig. A.11. Red lines are the RMSE for each model after the pCO2 scaling for RMSE values see Table 5.3. 165 Figure 5.3. Summary of average RMSE across the model experiments a) Eocene runs for CO2 using 4x pCO2 (Table 3), ice runs contain no ice with pCO2 of 560 ppmv, Eocene paleogeography runs (Table 5.3) b) Oligocene runs for pCO2 using 2x pCO2, ice runs containing the model prescribed ice sheet with pCO2 of 560 ppmv, and Oligocene paleogeography runs (Table 5.3). Mean is the orange line with outliers as dots. Dots are outliers with both from the GFDL CM2.1 model. Blue line is the ensemble mean from the pCO2 scaling. 5.3 Results 5.3.1 Proxy-model comparison 5.3.1.1 pCO2 For the pCO2 runs, the RMSE for the SAT ranges from 4.6 to 9.9°C for the Eocene, 7.5 to 11.6°C for the Oligocene, and 1.8 to 5.8 for the difference comparison (Fig. 5.1 and Fig. A.10). The RMSE for the SST ranges from 5.7 to 7.7°C for the Eocene, 6.1 to 8.6°C for the Oligocene, and 1.9 to 2.9°C for the temperature comparison (Fig. 5.2 and Fig. A.11). The best fit for the Eocene data is CESM_B for both SAT and SSTs (4.6°C and 5.7°C) as well as for the Oligocene is CESM_B for SATs (7.5°C) and for SSTs (6.1°C). The lowest RMSE for the 2x-4x pCO2 166 comparison comes from HadCM3BL of 1.8°C and 1.9°C for SAT and SSTs respectively. Although, Nor ESM-L has a similar RMSE value of 2.0°C for SST. Although they have the best fit to the data this would imply a higher pCO2 decrease given the difference between the Eocene and Oligocene runs is a halving of pCO2. The ensemble mean RMSE, for a halving of pCO2, is 3°C and 2°C for SATs and SSTs respectively. 5.3.1.2 Ice sheet For the ice sheet comparison only 3 models were used (CESM_H, HadCM3BL, and FOAM) as well as the ensemble mean. The RMSE for SAT ranged from 9.3 to 12.3°C for the Eocene, 21.5 to 29.3°C for the Oligocene, and 12.3 to 20.4°C for the temperature comparison (Fig. 5.3 and Fig. A.12). The RMSE for SST ranged from 7.4 to 9.2°C for the Eocene, 6.8 to 8.0°C for the Oligocene, and 2.4 to 2.8°C for the ice-no ice comparison (Fig. 5.3 and Fig. A.13). The lowest RMSE for SAT comes from the HadCM3BL model and for SST the CESM_H model for the difference comparison. FOAM has the highest RMSE across all three time slices for SST. The Eocene SAT no ice runs have a higher RMSE (average of ~10°C when excluding the ensemble mean) compared to the SAT Eocene pCO2 run RMSE (average of ~7°C excluding the ensemble mean) which are run at a higher pCO2 (800-1120 ppmv versus 560 ppmv for Eocene ice runs) indicating that a high pCO2 is needed to better reflect Eocene temperatures. For the Oligocene there are substantial proxy-model discrepancies (high RMSE). Proxies confidently identify SAT above freezing, whereas the climate models forced with a large difference between the ice and no ice runs yields too large a cooling compared to proxies. 167 5.3.1.3 Paleogeography Paleogeography was changed across all models except for CESM_B with pCO2 held constant between the Eocene and Oligocene runs. Both CESM_H, GDFLCM 2.1, and UVic reflect changes in ocean gateways while FOAM models a change in West Antarctic paleogeography where the region is above sea level (Eocene) and most of the land area is below sea level (Oligocene) (Table 5.3). NorESM-L and HadCM3BL model slight changes in continental positions from the Late Eocene to early Oligocene. The RMSE for SAT ranged from 5.4 to 13.7°C for the Eocene, 3.1 to 11.8°C for the Oligocene, and 2.3 to 5.0°C for the temperature comparison (Fig. 5.3 and Fig. A.14). The RMSE for SST ranged from 6.7 to 9.2°C for the Eocene, 6.1 to 10.2°C for the Oligocene, and 2.7 to 4.5°C for the difference comparison (Fig. 5.3 and Fig. A.15). The ensemble mean for SAT and SST are 2.9°C and 3.0°C respectively. For the Eocene and Oligocene SAT runs GFDL CM2.1 has the lowest RMSE of 5.4°C and 3.1°C respectively. For the Eocene and Oligocene SST runs GFDL CM2.1 had the lowest RMSE of 6.1°C for the Eocene and HadCM3BL had the lowest RMSE of 6.1°C for the Oligocene. For the difference between the paleogeography for each model run the lowest RMSE was 2.3°C and 2.7°C from FOAM for SAT and SST (Fig. A.14 and A.15). The best fit to the Eocene data is from the model with the lowest pCO2 of 800 ppmv compared to the other models with pCO2 as 1120 ppmv and 1600 ppmv for UVic. Most of the models suggest a warming in SAT with regional differences (Fig A.14). This in contrast to the proxy data which suggest temperature changes of 0 to 3°C. The model with the best fit for post/pre paleogeography is FOAM which has the most cooling regionally. To note additional regional differences UVic indicates more warming in the Pacific and Ross Sea sectors of the Southern Ocean while CESM_H suggest warming in the Atlantic and Indian Ocean sector with a cooling in the Pacific and Ross Sea 168 sectors. This difference cold be attributed to the prescribed modeled gateway opening in the Southern Ocean with CESM_H modeling the opening of Drake Passage and the Tasman Gateway and UVic modeling the opening of only Drake Passage. The overall warming trend suggests that paleogeography is not the primary driver of hemispheric cooling as previously noted (Hutchinson et al., 2021; Kennedy-Asser et al., 2020) but could impact regional differences in combination with pCO2. It also remains plausible that paleogeography changes could have indirectly triggered pCO2 changes. Two such mechanisms include a shift in the dominant basin of deep-water formation changing the ocean’s ability to store carbon (Fyke et al., 2015; Speelman et al., 2009), or through land-based CO2 weathering feedbacks triggered by the onset of the Atlantic meridional overturning circulation (Elsworth et al., 2017). 5.3.2 CO2 scaling As in the approach of Hutchinson et al., (2021), pCO2 was scaled by a constant to determine the forcing required to achieve the best fit between the proxies and the model for each model. Here we compare the updated Antarctic proxy record (rather than the global proxy data) with scaled pCO2 to both surface air temperature and sea surface temperature. For the pCO2 decrease calculations the assumed post-EOT pCO2 is set at 560 ppmv, since this matches most of the models. RMSE was calculated between the difference between the 2x-4x pCO2 runs and the Oligocene-Eocene proxies for both SAT and SST. By varying the scaling factor, we found the lowest RMSE and the best estimated decrease in pCO2 for each model for SAT and SST proxies separately (Table 5.3). The best fit for each model from the pCO2 scaling can be seen in Figure 5.4. Averaging across all models, the average pCO2 decrease is 199 ppmv for SAT and 336 ppmv for SST. 169 Table 5.4. Climate model RMSE for the SAT and SST proxy-model comparison for the best fit pCO2 forcing. The pCO2 is expressed for the Oligocene-Eocene runs and as a % of the Eocene runs. SAT proxy-model comparison SST proxy-model comparison Climate Model RMSE (°C) pCO2 decrease (ppmv) pCO2 decrease (%) RMSE (°C) pCO2 decrease (ppmv) pCO2 decrease (%) CESM_B 1.5 169 23.2 2.0 289 34.0 CESM_H 1.3 189 25.3 1.8 337 37.6 FOAM 1.3 179 24.2 2.0 210 27.3 GFDLCM 2.1 1.5 101 15.3 2.0 149 21.0 HadCM3BL 1.3 301 34.9 1.9 560 50.0 NorESM-L 1.4 254 31.2 2.0 471 45.7 Ensemble mean 1.4 189 25.3 1.9 289 34.0 In order to achieve best fit to the proxy data, the largest decrease in pCO2 is required in the HadCM3BL (560 ppmv) and NorESM-L (471 ppmv) climate models when fitting to the proxy SST data. When fitting to the smaller SAT difference, commensurately smaller changes are needed, with the largest changes needed in the HadCM3BL (301 ppmv) and NorESM-L (254 ppmv) when matching the shift in SAT. The lowest RMSE for SAT is 1.3°C and for SST it is 1.8°C for the FOAM and CESM_H model experiments respectively. Although the RMSE range is small ranging from 1.3 to 1.5°C for SAT and 1.8 to 2.0°C for SST. SSTs have the highest range in pCO2 percent decrease (21.0-50.0%) in comparison to the SATs (15.3 to 34.9°C) (Table 5.4). The ensemble mean indicates a decrease of pCO2 of 25.3 and 34.0% across the Eocene-Oligocene Transition for SAT and SST respectively. 170 Figure 5.4. a) Proxy-model ∆SAT comparison after scaling the pCO2 forcing to achieve best fit to the magnitude of cooling in the proxy SATs, b) Proxy-model ∆SST comparison after scaling the pCO2 forcing to achieve best fit to the magnitude of cooling in the proxy SSTs. 171 5.4 Discussion 5.4.1 Ice sheet extent Globally the main feature of the Eocene Oligocene Transition is the glaciation of Antarctica. The ice sheet expanded yet there were still unglaciated areas at the margins of the continent in the Oligocene. The model runs that were prescribed an ice sheet used in this comparison have various ice sheet sizes ranging from 17x10 6 , 20x10 6 , and 25 x10 6 km 3 for the HadCM3BL, CESM_H, and FOAM models respectively (Goldner et al., 2014; Kennedy et al., 2015; Ladant et al., 2014a; Ladant et al., 2014b). These ice volumes correspond to ~65%, 75% and 95% of the modern Antarctic ice sheet respectively, which fall within estimates from benthic δ 18 O that place the EAIS at 60-130% of the modern due to a large range of δ 18 Oice for the Oligocene ice sheet (Bohaty et al., 2012b; Lear et al., 2008). Sedimentary records from the western Ross Sea suggest an Antarctic ice sheet smaller than today until 32.8 Ma when it reached the coastline (Galeotti et al., 2016). The use of a full ice sheet for model outputs for the Oligocene post-EOT is not representative of available proxy evidence supporting the presence of vegetation including pollen records from the peninsula (Anderson et al., 2011). In this study, we defined source areas with basic polygons on the continent to capture the catchment area when terrestrial proxies (e.g., soil biomarkers and rock weathering proxies) are exported to marginal marine settings. With the presence of a large ice sheet, soil and plant derived temperature proxies would be limited to unglaciated areas. For the purposes of this model comparison, the source areas used in the model averaging were held constant for both the Eocene and Oligocene scenarios, a simplification of the likely range constriction. Proxies confidently identify SAT above freezing, whereas the climate models forced with a large difference between the ice and no ice runs yields too large a cooling compared to proxies due to 172 the inclusion of an ice sheet proximal to modern size for the Oligocene model simulations (Fig. A.12) leading to large RMSE (Fig 5.3). The presence of plants and soils warm enough for bacterial production suggests that an ice sheet did not cover the entire continent after the Eocene- Oligocene Transition. 5.4.2 Southern Ocean SST proxies indicate cooling by up to 7°C across the EOT (Table 5.2). In the study region, the warmest waters are found in the Atlantic Ocean sector of the Southern Ocean and persist into the Oligocene with <1°C cooling within proxy uncertainty (Fig. A.11). The largest drop is recorded at DSDP Site 511, in the Atlantic Ocean sector of the Southern Ocean, where BAYSPAR and U k’ 37, estimate cooling of 3 and 7°C (Houben et al., 2019; Lauretano et al., 2021) also corroborated by dinocyst assemblages changing from high latitude endemic to diverse autotrophic (Houben et al., 2019). This ocean cooling matches one coupled climate modelling scenario linked to enhanced ocean overturning as deep water formation intensifies with cooling (Goldner et al., 2014). In Prydz Bay, on the edge of the Antarctic Continent in the Indian Ocean sector of the Southern Ocean, SST cools by 4°C across the EOT (Tibbett et al., 2021) whereas further offshore at the Kerguelen Plateau temperatures drop by <1°C, although these changes are the same within uncertainties of the proxies and expected spatial variability. Cooling in the Indian Ocean sector could be attributed to sea ice development in Prydz Bay (Hutchinson et al., 2018) and has been predicted in climate models with the Antarctic ice sheet forcing (Kennedy et al., 2015). Given mean annual temperatures around 10°C there could be seasonal winter ice in this region. In the Pacific Ocean sector of the Southern Ocean, BAYSPAR and U k’ 37 detect at most a 1°C change in SSTs from ODP Site 1172 (Houben et al., 2019) and DSDP Site 277 (Lauretano et al., 173 2021; Pagani et al., 2011). This small change is perhaps not surprising as some models with only runs comparing no ice to an ice sheet have predicted a warming (Kennedy et al., 2015). The heterogenous proxy estimates give differing magnitudes of Southern Ocean cooling and make a model-proxy agreement difficult to achieve (Fig. 5.4). For example on the Falkand Plateau DSDP Site 511, the estimated cooling between the Eocene and Oligocene was ~3°C for BAYSPAR (Lauretano et al., 2021) and ~7°C for U k’ 37 (Houben et al., 2019). The two SST proxies differ in the magnitude of the cooling and this may be in part due to their water column distribution with the haptophyte algae producing alkenones likely living at shallower depths in the photic zone (Popp et al., 2006), whereas the Thaumarchaeota, the producers of isoGDGTs can be found at greater depths (Kim et al., 2012; Schouten et al., 2013). In addition, the large seasonality of light and temperature in the high latitudes may lead to varying seasonality in the proxy recorders. For instance, U k’ 37 is produced by algae requiring light (summer/fall) for production; whereas Thaumarchaeota are chemoautotrophs (Smittenberg et al., 2004; Wuchter et al., 2003) or primarily autotrophic (Ingalls et al., 2006) depending on their depth in the water column and thus do not have the same photic requirements (may have more year-round production). Other differences may result from advection of alkenones (Mollenhauer et al., 2003), in the latest Eocene the decrease in SSTs in the Southwest Atlantic could alternatively be attributed to the intensification of the westward circum-Antarctic circulation (Houben et al., 2019) bringing colder waters to DSDP Site 511. In addition, alkenone paleothermometry is based on the temperature relationship with modern species and the temperature relationship may differ in extinct species (Herbert, 2014). Evolutionary changes in size and extinctions within the reticulofenestrids in the Oligocene (Henderiks & Pagani, 2008) may affect the temperature reconstruction. Beyond the physical reasons for the offsets in the two proxies, that are 174 inadequately constrained for the ancient ocean, we can evaluate the numerical implications of the proxy uncertainty, by performing the calculations with and without each proxy. The experiment was rerun without the U k’ 37 and then without the BAYSPAR estimates to identify how this would impact the comparison (Table 5.5). The inclusion of both proxies leads to the highest pCO2 decrease across the transition from the scaling experiments (35.9%), and the removal of the larger magnitude cooling from the alkenones reduces the pCO2 scaling the most, to just 26.3% decrease needed to force the transition. The large difference in pCO2 decrease with the addition of the U k’ 37 temperatures is due to the higher temperature decrease of 7°C. This decrease exceeds the model estimates for temperature decrease of 2-5°C across the best fits from the pCO2 scaling. This larger cooling would require a greater change in pCO2 across the Eocene-Oligocene Transition. Due to low number of sites available (n=7) the CO2 scaling is not as robust. The discrepancy between these two proxies highlights the need for more proxy records with larger longitudinal variance to account for proxy-proxy discrepancy and the need to assess proxy data to identify outliers or data points that may not reflect sea surface temperature. Table 5.5. Experiment to compare the effects of proxy uncertainty on the pCO2 forcing estimation. Showing the RMSE for SST pCO2 runs and scaling averages across all six models when using both proxies from DSDP Site 511, or only using BAYSPAR or U k’ 37. Both proxies BAYSPAR only U k’ 37 only RMSE (°C) Oligocene-Eocene 2.5 2.2 2.7 pCO2 decrease (%) 35.9 26.3 33.5 pCO2 decrease (ppmv) 336 207 300 5.4.3 Declining pCO2 Based on the proxy-model comparison it is clear that the lowest RMSE for the Eocene occurs at higher pCO2s (>560 ppmv) which is in line with previous estimates of pCO2 suggesting a late Eocene pCO2 of 830 to 980ppmv from boron and alkenone isotopes (Rae et al., 2021). Previous 175 global proxy model comparison suggest a 40% decrease in pCO2 across the EOT (Hutchinson et al., 2021), attributed to the lack of dynamic ice sheets and under sensitivity to CO2 forcing (Hutchinson et al., 2021). The absolute pCO2 levels are uncertain in the past, due to factors such as boron isotope seawater uncertainties; however, the boron isotope is better at assessing relative change (Raitzsch & Hönisch, 2013). The boron and alkenone isotopes are well studied for the EOT and have a high amount of data relative to other pCO2 proxies. Current estimates of EOT pCO2 changes from alkenone δ 13 C and boron isotopes suggest a decrease of 140 to 150 ppmv, or roughly 25% (Rae et al., 2021). The best fits between the proxies and model runs for the change in temperature, both SAT and SST, and across all the models (Table 5.4 and Fig. 5.4) suggest a 15-35% decrease in pCO2 for SAT and 21-50% decrease for SST. The percent decrease is similar to previous pCO2 proxy estimates of 16% decrease from boron isotopes (Anagnostou et al., 2016, 2020; Henehan et al., 2020; Pearson et al., 2009) and a 27% decreases from alkenones (Pagani et al., 2005, 2011). The total amount (~200 ppmv) from the proxy-model comparison is within plausible range of proxy uncertainties. Given that the prescribed post-EOT level was 560 ppmv for the calculations the total amount may vary; however, the percent change is more comparable to pCO2 records. The pCO2 range falls within estimate from pCO2 proxies. The discrepancy between the scaling and pCO2 proxies could be due to additional forcing from ice- albedo feedbacks associated with from the presence of an ice sheet, sea ice (both likely) and/or changes in paleogeography. 5.5 Conclusions More records are accumulating from the Southern Hemisphere allowing a new perspective at Antarctic-proximal changes across the EOT that were previously best known from records in the mid-latitudes of the Northern Hemisphere. These new records in the southern high latitudes 176 highlight the spatially heterogeneous nature of cooling, with declines in sea surface temperature (ranging from 0 to 7°C) and in surface air temperature (0 to 3°C). However, these records are limited in their source regions with no records spanning the late Eocene and early Oligocene from Bellingshausen or Amundsen Sea for both sea surface and air temperature. Climate model experiments that prescribed ice sheets lead to extreme local cooling in the Oligocene which is cooler than the proxy records. This comparison supports higher pCO2 estimates for the late Eocene as higher concentrations are needed to match late Eocene temperature proxies. By comparing the proxy records to model outputs that assessed decline in pCO2, changes in paleogeography, and the addition of a near or above modern size ice sheet the decline in pCO2 provides the best fit to proxy records from the Antarctic across the Eocene-Oligocene Transition. The decline in surface air temperature suggested by the new proxy compilation was used to scale model runs suggesting a 25% decrease in pCO2 similar to recent pCO2 compilations (Rae et al., 2021). This is encouraging as it suggests that the proxy and climate model data on temperature, pCO2 and sensitivity may be converging on a coherent explanation for the EOT. The SST proxy record identifies the need for an increase in proxy data coverage, better spanning longitudes to further constrain the uncertainties on the magnitude of the change, to assess proxy-proxy discrepancies, and to identify if paleogeography could explain the SST heterogeneity in the Southern Ocean. Acknowledgements We declare no financial conflicts of interests for any author or their affiliations. This research was funded by the U.S. National Science Foundation AGS-1844380 to NJB and OPP-1908548 to SJF. DKH was supported by ARC grant DE220100279. Thanks to Scott Knapp for technical assistance to EJT with coding. 177 References Anagnostou, E., John, E. H., Babila, T. L., Sexton, P. F., Ridgwell, A., Lunt, D. J., Pearson, P. N., Chalk, T. B., Pancost, R. D., & Foster, G. L. (2020). Proxy evidence for state- dependence of climate sensitivity in the Eocene greenhouse. Nature Communications, 11(1), 4436. https://doi.org/10.1038/s41467-020-17887-x Anagnostou, E., John, E. H., Edgar, K. M., Foster, G. L., Ridgwell, A., Inglis, G. N., Pancost, R. D., Lunt, D. J., & Pearson, P. N. (2016). Changing atmospheric CO2 concentration was the primary driver of early Cenozoic climate. Nature, 533, 380. Baatsen, M., von der Heydt, A. S., Huber, M., Kliphuis, M. A., Bijl, P. K., Sluijs, A., & Dijkstra, H. A. (2020). The middle-to-late Eocene greenhouse climate, modelled using the CESM 1.0.5. Climate of the Past Discussions, 2020, 1–44. https://doi.org/10.5194/cp- 2020-29 Bohaty, S. M., Zachos, J. C., & Delaney, M. L. (2012). Foraminiferal Mg/Ca evidence for Southern Ocean cooling across the Eocene–Oligocene transition. Earth and Planetary Science Letters, 317–318, 251–261. https://doi.org/10.1016/j.epsl.2011.11.037 Coxall, H. K., Huck, C. E., Huber, M., Lear, C. H., Legarda-Lisarri, A., O’Regan, M., Sliwinska, K. K., van de Flierdt, T., de Boer, A. M., Zachos, J. C., & Backman, J. (2018). Export of nutrient rich Northern Component Water preceded early Oligocene Antarctic glaciation. Nature Geoscience, 11(3), 190–196. https://doi.org/10.1038/s41561-018- 0069-9 Coxall, H. K., & Pearson, P. N. (2007). The Eocene-Oligocene transition. Deep Time Perspectives on Climate Change: Marrying the Signal From Computer Models and Biological Proxies, 351–387. Coxall, H. K., Wilson, P. A., Palike, H., Lear, C. H., & Backman, J. (2005). Rapid stepwise onset of Antarctic glaciation and deeper calcite compensation in the Pacific Ocean. Nature, 433(7021), 53–57. https://doi.org/10.1038/nature03135 Dalziel, I. W. D., Lawver, L. A., Pearce, J. A., Barker, P. F., Hastie, A. R., Barfod, D. N., Schenke, H.-W., & Davis, M. B. (2013). A potential barrier to deep Antarctic circumpolar flow until the late Miocene? Geology, 41(9), 947–950. https://doi.org/10.1130/G34352.1 Dearing Crampton-Flood, E., Tierney, J. E., Peterse, F., Kirkels, F. M. S. A., & Sinninghe Damsté, J. S. (2020). BayMBT: A Bayesian calibration model for branched glycerol dialkyl glycerol tetraethers in soils and peats. Geochimica et Cosmochimica Acta, 268, 142–159. https://doi.org/10.1016/j.gca.2019.09.043 DeConto, R. M., & Pollard, D. (2003). Rapid Cenozoic glaciation of Antarctica induced by declining atmospheric CO2. Nature, 421(6920), 245–249. https://doi.org/10.1038/nature01290 178 Deng, L., Jia, G., Jin, C., & Li, S. (2016). Warm season bias of branched GDGT temperature estimates causes underestimation of altitudinal lapse rate. Organic Geochemistry, 96, 11– 17. https://doi.org/10.1016/j.orggeochem.2016.03.004 Douglas, P. M. J., Affek, H. P., Ivany, L. C., Houben, A. J. P., Sijp, W. P., Sluijs, A., Schouten, S., & Pagani, M. (2014). Pronounced zonal heterogeneity in Eocene southern high-latitude sea surface temperatures. Proceedings of the National Academy of Sciences, 111(18), 6582. https://doi.org/10.1073/pnas.1321441111 Elsworth, G., Galbraith, E., Halverson, G., & Yang, S. (2017). Enhanced weathering and CO2 drawdown caused by latest Eocene strengthening of the Atlantic meridional overturning circulation. Nature Geoscience, 10(3), 213–216. https://doi.org/10.1038/ngeo2888 Francis, J. E., Marenssi, S., Levy, R., Hambrey, M., Thorn, V. C., Mohr, B., Brinkhuis, H., Warnaar, J., Zachos, J., Bohaty, S., & DeConto, R. (2008). Chapter 8 From Greenhouse to Icehouse – The Eocene/Oligocene in Antarctica. In F. Florindo & M. Siegert (Eds.), Developments in Earth and Environmental Sciences (Vol. 8, pp. 309–368). Elsevier. https://doi.org/10.1016/S1571-9197(08)00008-6 Fyke, J. G., D’Orgeville, M., & Weaver, A. J. (2015). Drake Passage and Central American Seaway controls on the distribution of the oceanic carbon reservoir. Global and Planetary Change, 128, 72–82. https://doi.org/10.1016/j.gloplacha.2015.02.011 Galeotti, S., DeConto, R., Naish, T., Stocchi, P., Florindo, F., Pagani, M., Barrett, P., Bohaty, S. M., Lanci, L., Pollard, D., Sandroni, S., Talarico, F. M., & Zachos, J. C. (2016). Antarctic Ice Sheet variability across the Eocene-Oligocene boundary climate transition. Science, 352(6281), 76. https://doi.org/10.1126/science.aab0669 Goldner, A., Herold, N., & Huber, M. (2014). Antarctic glaciation caused ocean circulation changes at the Eocene–Oligocene transition. Nature, 511(7511), 574–577. https://doi.org/10.1038/nature13597 Henderiks, J., & Pagani, M. (2008). Coccolithophore cell size and the Paleogene decline in atmospheric CO2. Earth and Planetary Science Letters, 269(3), 576–584. https://doi.org/10.1016/j.epsl.2008.03.016 Henehan, M. J., Edgar, K. M., Foster, G. L., Penman, D. E., Hull, P. M., Greenop, R., Anagnostou, E., & Pearson, P. N. (2020). Revisiting the Middle Eocene Climatic Optimum “Carbon Cycle Conundrum” with new estimates of atmospheric pCO2 from boron isotopes. Paleoceanography and Paleoclimatology, 35(6), e2019PA003713. Herbert, T. D. (2014). Alkenone Paleotemperature Determinations. In Treatise on Geochemistry (pp. 399–433). Elsevier. https://doi.org/10.1016/B978-0-08-095975- 7.00615-X Houben, A. J. P., Bijl, P. K., Sluijs, A., Schouten, S., & Brinkhuis, H. (2019). Late Eocene Southern Ocean Cooling and Invigoration of Circulation Preconditioned Antarctica for 179 Full-Scale Glaciation. Geochemistry, Geophysics, Geosystems, 20(5), 2214–2234. https://doi.org/10.1029/2019GC008182 Houben, A. J. P., van Mourik, C. A., Montanari, A., Coccioni, R., & Brinkhuis, H. (2012). The Eocene–Oligocene transition: Changes in sea level, temperature or both? Cenozoic Evolution of Antarctic Climates, Oceans and Ice Sheets, 335–336, 75–83. https://doi.org/10.1016/j.palaeo.2011.04.008 Hunt, R. J., & Poole, I. (2003). Paleogene West Antarctic climate and vegetation history in light of new data from King George Island. Hutchinson, D. K., Coxall, H. K., Lunt, D. J., Steinthorsdottir, M., de Boer, A. M., Baatsen, M., von der Heydt, A., Huber, M., Kennedy-Asser, A. T., Kunzmann, L., Ladant, J.-B., Lear, C. H., Moraweck, K., Pearson, P. N., Piga, E., Pound, M. J., Salzmann, U., Scher, H. D., Sijp, W. P., … Zhang, Z. (2021). The Eocene–Oligocene transition: A review of marine and terrestrial proxy data, models and model–data comparisons. Climate of the Past, 17(1), 269–315. https://doi.org/10.5194/cp-17-269-2021 Hutchinson, D. K., Coxall, H. K., OʹRegan, M., Nilsson, J., Caballero, R., & de Boer, A. M. (2019). Arctic closure as a trigger for Atlantic overturning at the Eocene-Oligocene Transition. Nature Communications, 10(1), 3797. https://doi.org/10.1038/s41467-019- 11828-z Hutchinson, D. K., de Boer, A. M., Coxall, H. K., Caballero, R., Nilsson, J., & Baatsen, M. (2018). Climate sensitivity and meridional overturning circulation in the late Eocene using GFDL CM2.1. Clim. Past, 14(6), 789–810. https://doi.org/10.5194/cp-14-789-2018 Ingalls, A. E., Shah, S. R., Hansman, R. L., Aluwihare, L. I., Santos, G. M., Druffel, E. R., & Pearson, A. (2006). Quantifying archaeal community autotrophy in the mesopelagic ocean using natural radiocarbon. Proceedings of the National Academy of Sciences, 103(17), 6442–6447. Kalanetra, K. M., Bano, N., & Hollibaugh, J. T. (2009). Ammonia‐oxidizing Archaea in the Arctic Ocean and Antarctic coastal waters. Environmental Microbiology, 11(9), 2434– 2445. 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, 1(5), 329–334. https://doi.org/10.1038/ngeo179 Kennedy, A. T., Farnsworth, A., Lunt, D. J., Lear, C. H., & Markwick, P. J. (2015). Atmospheric and oceanic impacts of Antarctic glaciation across the Eocene–Oligocene transition. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 373(2054), 20140419. https://doi.org/10.1098/rsta.2014.0419 Kennedy-Asser, A. T., Lunt, D. J., Valdes, P. J., Ladant, J.-B., Frieling, J., & Lauretano, V. (2020). Changes in the high-latitude Southern Hemisphere through the Eocene– 180 Oligocene transition: A model–data comparison. Climate of the Past, 16(2), 555–573. https://doi.org/10.5194/cp-16-555-2020 Kennett, J. P. (1977). Cenozoic evolution of Antarctic glaciation, the circum-Antarctic Ocean, and their impact on global paleoceanography. Journal of Geophysical Research (1896-1977), 82(27), 3843–3860. https://doi.org/10.1029/JC082i027p03843 Kim, J., Crosta, X., Willmott, V., Renssen, H., Bonnin, J., Helmke, P., Schouten, S., & Sinninghe Damsté, J. S. (2012). Holocene subsurface temperature variability in the eastern Antarctic continental margin. Geophysical Research Letters, 39(6). Ladant, J.-B., Donnadieu, Y., & Dumas, C. (2014). Links between CO2, glaciation and water flow: Reconciling the Cenozoic history of the Antarctic Circumpolar Current. Clim. Past, 10(6), 1957–1966. https://doi.org/10.5194/cp-10-1957-2014 Ladant, J.-B., Donnadieu, Y., Lefebvre, V., & Dumas, C. (2014). The respective role of atmospheric carbon dioxide and orbital parameters on ice sheet evolution at the Eocene- Oligocene transition. Paleoceanography, 29(8), 810–823. https://doi.org/10.1002/2013PA002593 Lauretano, V., Kennedy-Asser, A. T., Korasidis, V. A., Wallace, M. W., Valdes, P. J., Lunt, D. J., Pancost, R. D., & Naafs, B. D. A. (2021). Eocene to Oligocene terrestrial Southern Hemisphere cooling caused by declining pCO2. Nature Geoscience, 14(9), 659–664. https://doi.org/10.1038/s41561-021-00788-z Lear, C. H., Bailey, T. R., Pearson, P. N., Coxall, H. K., & Rosenthal, Y. (2008). Cooling and ice growth across the Eocene-Oligocene transition. Geology, 36(3), 251–254. https://doi.org/10.1130/G24584A.1 Liu, Z., Pagani, M., Zinniker, D., DeConto, R., Huber, M., Brinkhuis, H., Shah, S. R., Leckie, R. M., & Pearson, A. (2009). Global Cooling During the Eocene-Oligocene Climate Transition. Science, 323(5918), 1187. https://doi.org/10.1126/science.1166368 Lopes dos Santos, R. A., Prange, M., Castañeda, I. S., Schefuß, E., Mulitza, S., Schulz, M., Niedermeyer, E. M., Sinninghe Damsté, J. S., & Schouten, S. (2010). Glacial–interglacial variability in Atlantic meridional overturning circulation and thermocline adjustments in the tropical North Atlantic. Earth and Planetary Science Letters, 300(3), 407–414. https://doi.org/10.1016/j.epsl.2010.10.030 Macphail, M. K., & Truswell, E. M. (2004). Palynology of Neogene slope and rise deposits from ODP Sites 1165 and 1167, East Antarctica. Proceedings of the Ocean Drilling Program. Scientific Results, 188, 1–20. Miller, K. G., Browning, J. V., Schmelz, W. J., Kopp, R. E., Mountain, G. S., & Wright, J. D. (2020). Cenozoic sea-level and cryospheric evolution from deep-sea geochemical and continental margin records. Science Advances, 6(20), eaaz1346. https://doi.org/10.1126/sciadv.aaz1346 181 Mollenhauer, G., Eglinton, T., Ohkouchi, N., Schneider, R. R., Müller, P. J., Grootes, P., & Rullkötter, J. (2003). Asynchronous alkenone and foraminfera records from the Benguela Upwelling System. Geochimica et Cosmochimica Acta, 67, 2157–2171. https://doi.org/10.1016/S0016-7037(03)00168-6 Müller, P. J., Kirst, G., Ruhland, G., Von Storch, I., & Rosell-Melé, A. (1998). Calibration of the alkenone paleotemperature index U37K′ based on core-tops from the eastern South Atlantic and the global ocean (60 N-60 S). Geochimica et Cosmochimica Acta, 62(10), 1757–1772. Pagani, M., Huber, M., Liu, Z., Bohaty, S. M., Henderiks, J., Sijp, W., Krishnan, S., & DeConto, R. M. (2011). The Role of Carbon Dioxide During the Onset of Antarctic Glaciation. Science, 334(6060), 1261. https://doi.org/10.1126/science.1203909 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(5734), 600. https://doi.org/10.1126/science.1110063 Passchier, S., Bohaty, S. M., Jimenez-Espejo, F. J., Pross, J., Roehl, U., van de Flierdt, T., Escutia, C., & Brinkhuis, H. (2013). Early Eocene to middle Miocene cooling and aridification of East Antarctica. Geochemistry Geophysics Geosystems, 14(5), 1399– 1410. https://doi.org/10.1002/ggge.20106 Passchier, S., Ciarletta, D. J., Miriagos, T. E., Bijl, P. K., & Bohaty, S. M. (2017). An Antarctic stratigraphic record of stepwise ice growth through the Eocene-Oligocene transition. Geological Society of America Bulletin, 129(3–4), 318–330. https://doi.org/10.1130/B31482.1 Pearson, P. N., Foster, G. L., & Wade, B. S. (2009). Atmospheric carbon dioxide through the Eocene–Oligocene climate transition. Nature, 461(7267), 1110–1113. https://doi.org/10.1038/nature08447 Petersen, S. V., & Schrag, D. P. (2015). Antarctic ice growth before and after the Eocene- Oligocene transition: New estimates from clumped isotope paleothermometry: Antarctic ice growth at the E/O transition. Paleoceanography, 30(10), 1305–1317. https://doi.org/10.1002/2014PA002769 Poole, I., Cantrill, D., & Utescher, T. (2005). A multi-proxy approach to determine Antarctic terrestrial palaeoclimate during the Late Cretaceous and Early Tertiary. Palaeogeography, Palaeoclimatology, Palaeoecology, 222(1), 95–121. https://doi.org/10.1016/j.palaeo.2005.03.011 Popp, B. N., Prahl, F. G., Wallsgrove, R. J., & Tanimoto, J. (2006). Seasonal patterns of alkenone production in the subtropical oligotrophic North Pacific. Paleoceanography, 21(1). Scopus. https://doi.org/10.1029/2005PA001165 Rae, J. W. B., Zhang, Y. G., Liu, X., Foster, G. L., Stoll, H. M., & Whiteford, R. D. M. (2021). Atmospheric CO2 over the Past 66 Million Years from Marine Archives. Annual 182 Review of Earth and Planetary Sciences, 49(1), 609–641. https://doi.org/10.1146/annurev-earth-082420-063026 Raitzsch, M., & Hönisch, B. (2013). Cenozoic boron isotope variations in benthic foraminifers. Geology, 41(5), 591–594. https://doi.org/10.1130/G34031.1 Sauermilch, I., Whittaker, J. M., Klocker, A., Munday, D. R., Hochmuth, K., Bijl, P. K., & LaCasce, J. H. (2021). Gateway-driven weakening of ocean gyres leads to Southern Ocean cooling. Nature Communications, 12(1), 6465. https://doi.org/10.1038/s41467- 021-26658-1 Scher, H. D., Whittaker, J. M., Williams, S. E., Latimer, J. C., Kordesch, W. E. C., & Delaney, M. L. (2015). Onset of Antarctic Circumpolar Current 30 million years ago as Tasmanian Gateway aligned with westerlies. Nature, 523(7562), 580–583. https://doi.org/10.1038/nature14598 Schouten, S., Hopmans, E. C., & Damsté, J. S. S. (2013). The organic geochemistry of glycerol dialkyl glycerol tetraether lipids: A review. Organic Geochemistry, 54, 19–61. Sijp, W. P., England, M. H., & Toggweiler, J. R. (2009). Effect of Ocean Gateway Changes under Greenhouse Warmth. Journal of Climate, 22(24), 6639–6652. https://doi.org/10.1175/2009JCLI3003.1 Smittenberg, R. H., Hopmans, E. C., Schouten, S., Hayes, J. M., Eglinton, T. I., & Sinninghe Damsté, J. (2004). Compound‐specific radiocarbon dating of the varved Holocene sedimentary record of Saanich Inlet, Canada. Paleoceanography, 19(2). Speelman, E. N., van Kempen, M. M., Barke, J., Brinkhuis, H., Reichart, G.-J., Smolders, A. J., Roelofs, J. G., Sangiorgi, F., de Leeuw, J. W., & Lotter, A. F. (2009). The Eocene Arctic Azolla bloom: Environmental conditions, productivity and carbon drawdown. Geobiology, 7(2), 155–170. Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., & Feakins, S. J. (2021). Late Eocene Record of Hydrology and Temperature From Prydz Bay, East Antarctica. Paleoceanography and Paleoclimatology, 36(4), e2020PA004204. https://doi.org/10.1029/2020PA004204 Tierney, J. E., & Tingley, M. P. (2014). A Bayesian, spatially-varying calibration model for the TEX86 proxy. Geochimica et Cosmochimica Acta, 127, 83–106. Toumoulin, A., Donnadieu, Y., Ladant, J.-B., Batenburg, S. J., Poblete, F., & Dupont-Nivet, G. (2020). Quantifying the Effect of the Drake Passage Opening on the Eocene Ocean. Paleoceanography and Paleoclimatology, 35(8), e2020PA003889. https://doi.org/10.1029/2020PA003889 Truswell, E. M., & Macphail, M. K. (2009). Polar forests on the edge of extinction: What does the fossil spore and pollen evidence from East Antarctica say? Australian Systematic Botany, 22(2), 57–106. https://doi.org/10.1071/SB08046 183 Vahlenkamp, M., Niezgodzki, I., De Vleeschouwer, D., Lohmann, G., Bickert, T., & Pälike, H. (2018). Ocean and climate response to North Atlantic seaway changes at the onset of long-term Eocene cooling. Earth and Planetary Science Letters, 498, 185–195. https://doi.org/10.1016/j.epsl.2018.06.031 Volkman, J. K., Eglinton, G., Corner, E. D. S., & Forsberg, T. E. V. (1980). Long-chain alkenes and alkenones in the marine coccolithophorid Emiliania huxleyi. Phytochemistry, 19(12), 2619–2622. https://doi.org/10.1016/S0031-9422(00)83930-8 Weijers, J. W. H., Bernhardt, B., Peterse, F., Werne, J. P., Dungait, J. A. J., Schouten, S., & Sinninghe Damsté, J. S. (2011). Absence of seasonal patterns in MBT–CBT indices in mid-latitude soils. Geochimica et Cosmochimica Acta, 75(11), 3179–3190. https://doi.org/10.1016/j.gca.2011.03.015 Weijers, J. W. H., Schouten, S., van den Donker, J. C., Hopmans, E. C., & Sinninghe Damsté, J. S. (2007). Environmental controls on bacterial tetraether membrane lipid distribution in soils. Geochimica et Cosmochimica Acta, 71(3), 703–713. https://doi.org/10.1016/j.gca.2006.10.003 Wuchter, C., Schouten, S., Boschker, H. T., & Sinninghe Damsté, J. S. (2003). Bicarbonate uptake by marine Crenarchaeota. FEMS Microbiology Letters, 219(2), 203–207. Zachos, J. C., Stott, L. D., & Lohmann, K. C. (1994). Evolution of Early Cenozoic marine temperatures. Paleoceanography, 9(2), 353–387. https://doi.org/10.1029/93PA03266 184 Chapter 6 Dissertation Conclusions There are few records of past climate on the Antarctic due to the difficult conditions when collecting both marine sediment and terrestrial material from outcrops. The continued advancement and development of biomarker techniques allow more information to be extracted from sediment that was previously collected. Chapter 2 applies these new techniques such as the hopane index to short cores from the Sabrina Coast. Plant wax δ 13 C suggests an open woodland or shrubby vegetation with precipitation isotopes from δD similar to today. Chapter 3 creates a record spanning across the Eocene-Oligocene Transition from Prydz Bay and identifies an abrupt drop in air temperature of 5°C before the transition as well as an increase in reworking inputs based on the hopane index. Chapter 4 applies the same suite of biomarker techniques to sediments off the tip of the Antarctic Peninsula with recovered sedimentary time slices from the Eocene, Oligocene, Miocene, and Pliocene. The temperature records indicate similar temperatures both air and sea surface temperatures between the Eocene and Oligocene. Temperature declines in the Miocene and Pliocene timeslices. There is increasing reworking as indicated by pollen and the hopane index. There is also increasing marine input from the Eocene to Pliocene. Chapter 5, compares the records generated in Chapters 3 and 4 and compiles published temperature records from across the southern continents surrounding the Southern Ocean including both air and sea surface temperature reconstructions and compares them to an ensemble of climate model experiments recently compiled for the Eocene Oligocene Transition. The proxy-model comparison suggests that the inclusion of a full ice sheet on Antarctica underestimates temperatures on the continent indicating the prescribed model ice sheets are too large. EOT model experiments wer forced with larger than expected pCO2 decreases. To 185 estimate the pCO2 decrease across the EOT the experiments were scaled by applying a best fit between the model and proxies. The temperature-scaled pCO2 drop used for the revised EOT model scenario, now better fits pCO2 proxy reconstructions indicating a dcrease of pCO2 of roughly 25% with no ice sheet present. In summary, the proxy reconstructions in this thesis adds new proxy data for the Antarctic, reinforces the value of multi-biomarker and multi-proxy comparisons, and demonstrates how screening for glacial reworking allows us to reconstruct past climate in polar regions across glacial transitions. In the final chapter, newly generated and compiled proxy data were compared to previous modeling of the EOT which illustrates the value of adding more proxy reconstructions. More mulit-proxy records are needed from the Antarctic as most records are limited to specific regions with narrow timeslices. New records from other regions can identify regional impacts of the Antarctic ice sheet as it expanded through the Eocene-Oligocene Transition and throughout the Cenozoic. In addition, more data is needed on the Oligocene which was the limiting factor for the proxy-model comparison. This time period is under researched and with future drilling expeditions and reassessment of legacy cores more information on the early stages of the Antarctic ice sheet can be collected. 186 References Aitken, A. R. A., J. L. Roberts, T. D. Van Ommen, D. A. Young, N. R. Golledge, J. S. Greenbaum, D. D. Blankenship, and M. J. Siegert. (2016). Repeated large-scale retreat and advance of Totten Glacier indicated by inland bed erosion. Nature, 533(7603), 385- 389. Anagnostou, E., John, E. H., Babila, T. L., Sexton, P. F., Ridgwell, A., Lunt, D. J., Pearson, P. N., Chalk, T. B., Pancost, R. D., & Foster, G. L. (2020). Proxy evidence for state- dependence of climate sensitivity in the Eocene greenhouse. Nature Communications, 11(1), 4436. https://doi.org/10.1038/s41467-020-17887-x Anagnostou, E., John, E. H., Edgar, K. M., Foster, G. L., Ridgwell, A., Inglis, G. N., et al. (2016). Changing atmospheric CO2 concentration was the primary driver of early Cenozoic climate. Nature, 533, 380. https://doi.org/10.1038/nature17423 Anderson, J. B. (2006). SHALDRIL II 2006 NBP0602A Cruise Report. Rice University, Department of Earth Science, Houston, Texas. Anderson, J. B., Warny, S., Askin, R. A., Wellner, J. S., Bohaty, S. M., Kirshner, A. E., Livsey, D. N., Simms, A. R., Smith, T. R., Ehrmann, W., Lawver, L. A., Barbeau, D., Wise, S. W., Kulhanek, D. K., Weaver, F. M., & Majewski, W. (2011). Progressive Cenozoic cooling and the demise of Antarctica’s last refugium. Proceedings of the National Academy of Sciences, 108(28), 11356. https://doi.org/10.1073/pnas.1014885108 Anderson, J. B., & Wellner, J. S. (2012). SHALDRIL I and II: Drilling from the Research Vessel Icebreaker Nathaniel B. Palmer. Oceanography. Ashworth, A., Lewis, A., Marchant, D., Askin, R., Cantrill, D., Francis, J., Leng, M., Newton, A., Raine, J., & Williams, M. (2007). The Neogene biota of the Transantarctic Mountains. Online Proceedings of the ISAES X, edited by A.K. Cooper and C.R. Raymond et al., USGS Open-File Report 2007–1047, Extended Abstract 071, 4 p. Askin, R. A. (1988). Campanian to Paleocene palynological succession of Seymour and adjacent islands, northeastern Antarctic Peninsula. In R. M. Feldman and M. O. Woodburne, eds., Geology and Paleontology of Seymour Island, Antarctic Peninsula. Geological Society of America Memoir 169, 131-153. Askin, R. A., Elliot, D. H., Stilwell, J. D., & Zinsmeister, W. J. (1991). Stratigraphy and paleontology of Campanian and Eocene sediments, Cockburn Island, Antarctic Peninsula. Journal of South American Earth Sciences, 4(1), 99–117. https://doi.org/10.1016/0895- 9811(91)90021-C Askin, R. A. (2000). Spores and pollen from the McMurdo Sound erratics, Antarctica. AGU Antarctic Research Series, 76:161-181. Baatsen, M., von der Heydt, A. S., Huber, M., Kliphuis, M. A., Bijl, P. K., Sluijs, A., & Dijkstra, H. A. (2020). The middle-to-late Eocene greenhouse climate, modelled using the CESM 1.0.5. Climate of the Past Discussions, 2020, 1–44. https://doi.org/10.5194/cp-2020-29 187 Badger, M. P., Chalk, T. B., Foster, G. L., Bown, P. R., Gibbs, S. J., Sexton, P. F., Schmidt, D. N., Pälike, H., Mackensen, A., & Pancost, R. D. (2019). Insensitivity of alkenone carbon isotopes to atmospheric CO 2 at low to moderate CO 2 levels. Climate of the Past, 15(2), 539–554. Badger, M. P., Lear, C. H., Pancost, R. D., Foster, G. L., Bailey, T. R., Leng, M. J., & Abels, H. A. (2013). CO2 drawdown following the middle Miocene expansion of the Antarctic Ice Sheet. Paleoceanography, 28(1), 42–53. Bakkelund, A., Porter, T.J., Froese, D.G., Feakins, S.J. (2018). Net fractionation of hydrogen isotopes in n-alkanoic acids from soils in the northern boreal forest. Organic Geochemistry 125, 1-13. Barron, J., Larsen, B., & et al. (Eds.). (1991). Proceedings of the Ocean Drilling Program, 119 Scientific Results (Vol. 119). Ocean Drilling Program. https://doi.org/10.2973/odp.proc.sr.119.1991 Bart, P. J., Egan, D., & Warny, S. A. (2005). Direct constraints on Antarctic Peninsula Ice Sheet grounding events between 5.12 and 7.94 Ma. Journal of Geophysical Research: Earth Surface, 110(F4). Baudoin, P. (2018). Reconstruction of Antarctic Cenozoic Paleoenvironments Through Palynological Analysis of Subglacial Lake and Ice Stream Sediments. LSU Master’s thesis. https://digitalcommons.lsu.edu/gradschool_theses/4721/. Belt, S., Smik, L., Brown, T., Kim, J.-H., Rowland, S., Allen, C., Gal, J.-K., Shin, K.-H., Lee, J., & Taylor, K. (2016). Source identification and distribution reveals the potential of the geochemical Antarctic sea ice proxy IPSO 25. Nature Communications, 7(1), 1–10. Belt, S. T. (2018). Source-specific biomarkers as proxies for Arctic and Antarctic sea ice. Organic Geochemistry, 125, 277–298. https://doi.org/10.1016/j.orggeochem.2018.10.002 Belt, S. T., Massé, G., Rowland, S. J., Poulin, M., Michel, C., & LeBlanc, B. (2007). A novel chemical fossil of palaeo sea ice: IP25. Organic Geochemistry, 38(1), 16–27. Berg, S., White, D. A., Jivcov, S., Melles, M., Leng, M. J., Rethemeyer, J., Allen, C., Perren, B., Bennike, O., & Viehberg, F. (2019). Holocene glacier fluctuations and environmental changes in subantarctic South Georgia inferred from a sediment record from a coastal inlet. Quaternary Research, 91(1), 132–148. Cambridge Core. https://doi.org/10.1017/qua.2018.85 Bijl, P. K., Pross, J., Warnaar, J., Stickley, C. E., Huber, M., Guerstein, R., Houben, A. J. P., Sluijs, A., Visscher, H., & Brinkhuis, H. (2011). Environmental forcings of Paleogene Southern Ocean dinoflagellate biogeography. Paleoceanography, 26(1). https://doi.org/10.1029/2009PA001905 Bohaty, S. M., & Zachos, J. C. (2003). Significant Southern Ocean warming event in the late middle Eocene. Geology, 31(11), 1017–1020. https://doi.org/10.1130/G19800.1 188 Bohaty, S. M., Zachos, J. C., Florindo, F., & Delaney, M. L. (2009). Coupled greenhouse warming and deep-sea acidification in the middle Eocene. Paleoceanography, 24(2). https://doi.org/10.1029/2008PA001676 Bohaty, S. M., Kulhanek, D. K., Wise Jr, S. W., Jemison, K., Warny, S., & Sjunneskog, C. (2011). Age assessment of Eocene–Pliocene drill cores recovered during the SHALDRIL II expedition, Antarctic Peninsula. Tectonic, Climatic, and Cryospheric Evolution of the Antarctic Peninsula, 63–113. Bohaty, S. M., Zachos, J. C., & Delaney, M. L. (2012). Foraminiferal Mg/Ca evidence for Southern Ocean cooling across the Eocene–Oligocene transition. Earth and Planetary Science Letters, 317–318, 251–261. https://doi.org/10.1016/j.epsl.2011.11.037 Bowman, V. C., Francis, J. E., Askin, R. A., Riding, J. B., Swindles, G. T. (2014). Latest Cretaceous-earliest Paleogene vegetation and climate change at the high southern latitudes: palynological evidence from Seymour Island, Antarctic Peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology 408, 26-47. Brown, C. A., Riding, J. B., & Warny, S. (2008). Palynological techniques. American Association of Stratigraphic Palynologists Foundation; 2nd ed., 137 pp., AASP Special Publications. Cande, S. C., Mutter, J. C. (1982), A Revised Identification of the Oldest Sea-Floor Spreading Anomalies between Australia and Antarctica. Earth and Planetary Science Letters, 58:151-160. Carter, A., Riley, T. R., Hillenbrand, C.-D., & Rittner, M. (2017). Widespread Antarctic glaciation during the Late Eocene. Earth and Planetary Science Letters, 458, 49–57. https://doi.org/10.1016/j.epsl.2016.10.045 Cerda-Peña, C., Contreras, S., & Rau, J. R. (2020). Molecular n-alkyl leaf waxes of three dominant plants from the temperate forest in South America. Organic Geochemistry, 149, 104105. Chen, X., Liu, X., Wei, Y., Huang, Y. (2019). Production of long-chain n-alkyl lipids by heterotrophic microbes: new evidence from Antarctic lakes. Organic Geochemistry, 103909. Clarke, K.R. (1993). Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18, 117–143. Coenen, J. J., Scherer, R., Baudoin, P., Warny, S., Castañeda, I. S., & Askin, R. (2019). Paleogene marine and terrestrial development of the West Antarctic Rift System Geophysical Research Letters, 47, 3. Collins, L. G., Allen, C. S., Pike, J., Hodgson, D. A., Weckström, K., & Massé, G. (2013). Evaluating highly branched isoprenoid (HBI) biomarkers as a novel Antarctic sea-ice proxy in deep ocean glacial age sediments. Quaternary Science Reviews, 79, 87–98. 189 Colwyn, D. A., & Hren, M. T. (2019). An abrupt decrease in Southern Hemisphere terrestrial temperature during the Eocene-Oligocene transition. Earth and Planetary Science Letters, 512, 227–235. https://doi.org/10.1016/j.epsl.2019.01.052 Contreras, L., Pross, J., Bijl, P. K., Koutsodendris, A., Raine, J. I., van de Schootbrugge, B., Brinkhuis, H. (2013). Early to middle Eocene vegetation dynamics at the Wilkes Land Margin (Antarctica). Review of Palaeobotany and Palynology, 197: 119-142. Contreras, L., Pross, J., Bijl, P. K., O’Hara, R. B., Raine, J. I., Sluijs, A., Brinkhuis, H. (2014). Southern high-latitude terrestrial climate change during the Palaeocene-Eocene derived from a marine pollen record (ODP 1172, East Tasman Plateau). Climate of the Past, 10(4):1401-1420. Cookson, I.C. (1950). Fossil pollen grains of Proteaceous type from Tertiary deposits in Australia. Australian Journal of Science, series B, 3: 166-177. Cookson, I.C.; Pike, K.M. (1954). Some dicotyledonous pollen types from Cainozoic deposits in the Australian region. Australian Journal of Botany 2: 197-219. Couper, R.A. (1960). New Zealand Mesozoic and Cainozoic plant microfossils. New Zealand Geological Survey Palaeontological Bulletin 32: 87. Coxall, H. K., Wilson, P. A., Pälike, H., Lear, C. H., & Backman, J. (2005). Rapid stepwise onset of Antarctic glaciation and deeper calcite compensation in the Pacific Ocean. Nature, 433(7021), 53-57. Coxall, H. K., & Pearson, P. N. (2007). The Eocene-Oligocene transition. In Deep time perspectives on climate change: marrying the signal from computer models and biological proxies, edited by M. Williams, A. M. Haywood, F. J. Gregory & D. N. Schmidt, Geological Society of London 351–387. https://doi.org/10.1144/TMS002.16 Cramer, B. S., Toggweiler, J. R., Wright, J. D., Katz, M. E., & Miller, K. G. (2009). Ocean overturning since the Late Cretaceous: Inferences from a new benthic foraminiferal isotope compilation. Paleoceanography, 24(4). https://doi.org/10.1029/2008PA001683 Curtin, L., D’Andrea, W. J., Balascio, N., Pugsley, G., de Wet, G., & Bradley, R. (2019). Holocene and Last Interglacial climate of the Faroe Islands from sedimentary plant wax hydrogen and carbon isotopes. Quaternary Science Reviews, 223, 105930. https://doi.org/10.1016/j.quascirev.2019.105930 Dalziel, I. W. D., Lawver, L. A., Pearce, J. A., Barker, P. F., Hastie, A. R., Barfod, D. N., Schenke, H.-W., & Davis, M. B. (2013). A potential barrier to deep Antarctic circumpolar flow until the late Miocene? Geology, 41(9), 947–950. https://doi.org/10.1130/G34352.1 Damsté, J. S. S., Rijpstra, W. I. C., Hopmans, E. C., Weijers, J. W., Foesel, B. U., Overmann, J., & Dedysh, S. N. (2011). 13, 16-Dimethyl octacosanedioic acid (iso-diabolic acid), a common membrane-spanning lipid of Acidobacteria subdivisions 1 and 3. Applied and Environmental Microbiology, 77(12), 4147–4154. 190 Davies, B. J., Hambrey, M. J., Smellie, J. L., Carrivick, J. L., & Glasser, N. F. (2012). Antarctic Peninsula Ice Sheet evolution during the Cenozoic Era. Quaternary Science Reviews, 31, 30–66. https://doi.org/10.1016/j.quascirev.2011.10.012 de Bar, M. W., Rampen, S. W., Hopmans, E. C., Sinninghe Damsté, J. S., & Schouten, S. (2019). Constraining the applicability of organic paleotemperature proxies for the last 90 Myrs. Organic Geochemistry, 128, 122–136. https://doi.org/10.1016/j.orggeochem.2018.12.005 De Jonge, C., Hopmans, E. C., Zell, C. I., Kim, J.-H., Schouten, S., & Sinninghe Damsté, J. S. (2014). Occurrence and abundance of 6-methyl branched glycerol dialkyl glycerol tetraethers in soils: Implications for palaeoclimate reconstruction. Geochimica et Cosmochimica Acta, 141, 97–112. https://doi.org/10.1016/j.gca.2014.06.013 Dearing Crampton-Flood, E., Tierney, J. E., Peterse, F., Kirkels, F. M. S. A., & Sinninghe Damsté, J. S. (2020). BayMBT: A Bayesian calibration model for branched glycerol dialkyl glycerol tetraethers in soils and peats. Geochimica et Cosmochimica Acta, 268, 142–159. https://doi.org/10.1016/j.gca.2019.09.043 DeConto, R. M., Pollard, D. (2016). Contribution of Antarctica to past and future sea-level rise. Nature, 531(591), doi: 10.1038/nature17145. DeConto, R. M., Pollard, D., Wilson, P. A., Palike, H., Lear, C. H., & Pagani, M. (2008). Thresholds for Cenozoic bipolar glaciation. Nature, 455(7213), 652–656. https://doi.org/10.1038/nature07337 DeConto, R. M., & Pollard, D. (2003). Rapid Cenozoic glaciation of Antarctica induced by declining atmospheric CO2. Nature, 421(6920), 245–249. https://doi.org/10.1038/nature01290 Delmotte, M. (1997). Enregistrements climatiques à Law Dome: Variabilité pour les périodes récentes et pour la déglaciation. PhD Thesis. Glaciologie. Université Joseph-Fourier - Grenoble I. Français. https://tel.archives-ouvertes.fr/tel-00759854/ Deng, L., Jia, G., Jin, C., & Li, S. (2016). Warm season bias of branched GDGT temperature estimates causes underestimation of altitudinal lapse rate. Organic Geochemistry, 96, 11–17. https://doi.org/10.1016/j.orggeochem.2016.03.004 Dettmann, M. E., Jarzen, D. M. (1988). Angiosperm pollen from uppermost Cretaceous strata of southeastern Australia and the Antarctic Peninsula. Memoirs of the Association of Australasian Palaentologists 5, 217-237. Diefendorf, A. F., Mueller, K. E., Wing, S. L., Koch, P. L., Freeman, K. H. (2010). Global patterns in leaf 13 C discrimination and implications for studies of past and future climate. Proceedings of the National Academy of Science, 107, 13. Diefendorf, A. F., Leslie, A. B., & Wing, S. L. (2015). Leaf wax composition and carbon isotopes vary among major conifer groups. Geochimica et Cosmochimica Acta, 170, 145– 156. https://doi.org/10.1016/j.gca.2015.08.018 191 Dingle, R. V., Marenssi, S. A., & Lavelle, M. (1998). High latitude Eocene climate deterioration: Evidence from the northern Antarctic Peninsula. Journal of South American Earth Sciences, 11(6), 571–579. https://doi.org/10.1016/S0895-9811(98)00035-2 Douglas, P. M. J., Affek, H. P., Ivany, L. C., Houben, A. J. P., Sijp, W. P., Sluijs, A., Schouten, S., & Pagani, M. (2014). Pronounced zonal heterogeneity in Eocene southern high- latitude sea surface temperatures. Proceedings of the National Academy of Sciences, 111(18), 6582. https://doi.org/10.1073/pnas.1321441111 Duncan, B., McKay, R., Bendle, J., Naish, T., Inglis, G. N., Moossen, H., Levy, R., Ventura, G. T., Lewis, A., Chamberlain, B., & Walker, C. (2019). Lipid biomarker distributions in Oligocene and Miocene sediments from the Ross Sea region, Antarctica: Implications for use of biomarker proxies in glacially-influenced settings. Palaeogeography, Palaeoclimatology, Palaeoecology, 516, 71–89. https://doi.org/10.1016/j.palaeo.2018.11.028 Ebisuzaki, W. (1997). A Method to Estimate the Statistical Significance of a Correlation When the Data Are Serially Correlated. Journal of Climate, 10(9), 2147–2153. https://doi.org/10.1175/1520-0442(1997)010<2147:AMTETS>2.0.CO;2 Edwards, L. E., Mudie, P. J., & de Vernal, A. (1991). Pliocene paleoclimatic reconstruction using dinoflagellate cysts: Comparison of methods. Quaternary Science Reviews, 10(2), 259–274. https://doi.org/10.1016/0277-3791(91)90024-O Elsworth, G., Galbraith, E., Halverson, G., & Yang, S. (2017). Enhanced weathering and CO2 drawdown caused by latest Eocene strengthening of the Atlantic meridional overturning circulation. Nature Geoscience, 10(3), 213–216. https://doi.org/10.1038/ngeo2888 Erohina, T., Cooper, A., Handwerger, D., & Dunbar, R. (2004). Seismic stratigraphic correlations between ODP Sites 742 and 1166: Implications for depositional paleoenvironments in Prydz Bay, Antarctica. Proceedings of the Ocean Drilling Program: Scientific Results, 188, 1–21. https://doi.org/10.2973/odp.proc.sr.188.2004 Escutia, C., Brinkhuis, H., & Klaus, A. (2011). IODP expedition 318: From greenhouse to icehouse at the Wilkes Land Antarctic Margin. Scientific Drilling, 12, 15-23. Etourneau, J., Sgubin, G., Crosta, X., Swingedouw, D., Willmott, V., Barbara, L., Houssais, M.- N., Schouten, S., Damsté, J. S. S., Goosse, H., Escutia, C., Crespin, J., Massé, G., & Kim, J.-H. (2019). Ocean temperature impact on ice shelf extent in the eastern Antarctic Peninsula. Nature Communications, 10(1), 304. https://doi.org/10.1038/s41467-018- 08195-6 Evangelinos, D., Escutia, C., Etourneau, J., Hoem, F., Bijl, P., Boterblom, W., van de Flierdt, T., Valero, L., Flores, J.-A., Rodriguez-Tovar, F. J., Jimenez-Espejo, F. J., Salabarnada, A., & López-Quirós, A. (2020). Late Oligocene-Miocene proto-Antarctic Circumpolar Current dynamics off the Wilkes Land margin, East Antarctica. Global and Planetary Change, 191, 103221. https://doi.org/10.1016/j.gloplacha.2020.103221 192 Feakins, S. J., Warny, S., & DeConto, R. M. (2014). Snapshot of cooling and drying before onset of Antarctic Glaciation. Earth and Planetary Science Letters, 404, 154–166. https://doi.org/10.1016/j.epsl.2014.07.032 Feakins, S. J., Warny, S., & Lee, J.-E. (2012). Hydrologic cycling over Antarctica during the middle Miocene warming. Nature Geoscience, 5(8), 557–560. https://doi.org/10.1038/NGEO1498 Feakins, S. J., Wu, M. S., Ponton, C., Galy, V., & West, A. J. (2018). Dual isotope evidence for sedimentary integration of plant wax biomarkers across an Andes-Amazon elevation transect. Geochimica et Cosmochimica Acta, 242, 64–81. https://doi.org/10.1016/j.gca.2018.09.007 Fietz, S., Martínez-Garcia, A., Rueda, G., Peck, V. L., Huguet, C., Escala, M., & Rosell-Melé, A. (2011). Crenarchaea and phytoplankton coupling in sedimentary archives: Common trigger or metabolic dependence? Limnology and Oceanography, 56(5), 1907–1916. https://doi.org/10.4319/lo.2011.56.5.1907 Flower, B. P., & Kennett, J. P. (1994). The middle Miocene climatic transition: East Antarctic ice sheet development, deep ocean circulation and global carbon cycling. Cenozoic Climate and Paleogeographic Changes in the Pacific Region, 108(3), 537–555. https://doi.org/10.1016/0031-0182(94)90251-8 Forsberg, C., Florindo, F., Gruetzner, J., Venuti, A., & Solheim, A. (2008). Sedimentation and aspects of glacial dynamics from physical properties, mineralogy and magnetic properties at ODP Sites 1166 and 1167, Prydz Bay, Antarctica. Palaeogeography, Palaeoclimatology, Palaeoecology, 260(1–2), 184–201. https://doi.org/10.1016/j.palaeo.2007.08.022 Foster, G. L., Lear, C. H., & Rae, J. W. (2012). The evolution of pCO2, ice volume and climate during the middle Miocene. Earth and Planetary Science Letters, 341, 243–254. Francis, J. E. (2000). Fossil wood from Eocene high latitude forests McMurdo Sound, Antarctica. AGU Antarctic Research Series. 76:253–260. Francis, J., Ashworth, A., Cantrill, D., Crame, J., Howe, J., Stephens, R., Tosolini, A., & Thorn, V. (2008). 100 million years of Antarctic climate evolution: Evidence from fossil plants. Antarctica: A Keystone in a Changing World, 19–28. Francis, J. E., Marenssi, S., Levy, R., Hambrey, M., Thorn, V. C., Mohr, B., Brinkhuis, H., Warnaar, J., Zachos, J., Bohaty, S., DeConto, R. (2009). From greenhouse to icehouse– the Eocene/Oligocene in Antarctica. Developments in Earth and Environmental Science, 8:309-368. Freeman, K. H., & Pancost, R. D. (2014). 12.15—Biomarkers for Terrestrial Plants and Climate. In H. D. Holland & K. K. Turekian (Eds.), Treatise on Geochemistry (Second Edition) (pp. 395–416). Elsevier. https://doi.org/10.1016/B978-0-08-095975-7.01028-7 Fretwell, P., Pritchard, H. D., Vaughan, D.G., Bamber, J. L., Barrand, N. E., Bell, R., Bianchi, C., Bingham, R. G., Blankenship, D. D., Casassa, G., Catania, G., Callens, D., Conway, 193 H., Cook, A. J., Corr, H. F. J., Damaske, D., Damm, V., Ferraccioli, F., Forsberg, R., Fujita, S., Gogineni, P., Griggs, J. A., Hindmarsh, R. C. A., Holmlund, P., Holt, J. W., Jacobel, R. W., Jenkins, A., Jokat, W., Jordan, T., King, E. C., Kohler, J., Krabill, W., Riger-Kusk, M., Langley, K. A., Leitchenkov, G., Leuschen, C., Luyendyk, B. P., Matsuoka, Mouginot, K. J., Nitsche, F. O., Nogi, Y., Nost, O. A., Popov, S. V., Rignot, E., Rippin, D. M., Rivera, A., Roberts, J., Ross, N., Siegert, M. J., Smith, A. M., Steinhage, D., Studinger, M., Sun, B., Tinto, B. K., Welch, B. C., Wilson, D., Young, D. A., Xiangbin, C., Zirizzotti, A. (2013). Bedmap2: improved ice bed, surface and thickness datasets for Antarctica. The Cryosphere, 7, 375–393, doi:10.5194/tc-7-375- 2013. Freymond, C. V., Kündig, N., Stark, C., Peterse, F., Buggle, B., Lupker, M., et al. (2018). Evolution of biomolecular loadings along a major river system. Geochimica et Cosmochimica Acta, 223, 389–404. https://doi.org/10.1016/j.gca.2017.12.010 Fyke, J. G., D’Orgeville, M., & Weaver, A. J. (2015). Drake Passage and Central American Seaway controls on the distribution of the oceanic carbon reservoir. Global and Planetary Change, 128, 72–82. https://doi.org/10.1016/j.gloplacha.2015.02.011 Galeotti, S., DeConto, R., Naish, T., Stocchi, P., Florindo, F., Pagani, M., Barrett, P., Bohaty, S. M., Lanci, L., Pollard, D., Sandroni, S., Talarico, F. M., & Zachos, J. C. (2016). Antarctic Ice Sheet variability across the Eocene-Oligocene boundary climate transition. Science, 352(6281), 76. https://doi.org/10.1126/science.aab0669 Galy, V., & Eglinton, T. (2011). Protracted storage of biospheric carbon in the Ganges– Brahmaputra basin. Nature Geoscience, 4(12), 843–847. https://doi.org/10.1038/ngeo1293 Galy, V., Eglinton, T., France-Lanord, C., & Sylva, S. (2011). The provenance of vegetation and environmental signatures encoded in vascular plant biomarkers carried by the Ganges– Brahmaputra rivers. Earth and Planetary Science Letters, 304(1–2), 1–12. Goldner, A., Herold, N., & Huber, M. (2015). Antarctic glaciation caused ocean circulation changes at the Eocene-Oligocene transition (vol 511, pg 574, 2014). Nature, 519(7543), 378–378. https://doi.org/10.1038/nature14220 Gradstein, F. M., Ogg, J. G., Schmitz, M., & Ogg, G. (2012). The Geologic Time Scale 2012. The Geologic Time Scale 2012, 1144 pp., Elsevier, Boston. Greenbaum, J. S., D. Blankenship, A. Young, T. G. Richter, J. L. Roberts, A. R. A. Aitken, B. Legresy, D. M. Schroeder, R. C. Warner, T. D. van Ommen, M. J. Siegert (2015). Ocean access to a cavity beneath Totten Glacier in East Antarctica. Nature, 8, 294-298, doi:10.1038/NGEO2388. Greene, C. A., Gwyther, D. E., & Blankenship, D. D. (2017). Antarctic mapping tools for MATLAB. Computers & Geosciences, 104, 151–157. 194 Greenop, R., Foster, G. L., Wilson, P. A., & Lear, C. H. (2014). Middle Miocene climate instability associated with high‐amplitude CO2 variability. Paleoceanography, 29(9), 845–853. Greenop, R., Sosdian, S. M., Henehan, M. J., Wilson, P. A., Lear, C. H., & Foster, G. L. (2019). Orbital forcing, ice volume, and CO2 across the Oligocene‐Miocene transition. Paleoceanography and Paleoclimatology, 34(3), 316–328. Griener, K. W., Nelson, D. M., Warny, S. (2013). Declining moisture availability on the Antarctic Peninsula during the late Eocene. Palaeogeography, Palaeoclimatology, Palaeoecology, 383:72-78. Griener, K. W., Warny, S., Askin, R., & Acton, G. (2015). Early to middle Miocene vegetation history of Antarctica supports eccentricity-paced warming intervals during the Antarctic icehouse phase. Global and Planetary Change, 127, 67–78. https://doi.org/10.1016/j.gloplacha.2015.01.006 Gulick, S. P. S., Shevenell, A. E., Montelli, A., Fernandez, R., Smith, C., Warny, S., Bohaty, A. M., Sjunneskog, C., Leventer, A., Frederick, B., Blankenship, D. D. (2017). Initiation and long-term instability of the East Antarctic Ice Sheet. Nature, (552)225, doi: 10.1038/nature25026. Hammer, Ø., Harper, D., Ryan, P. (2001). PAST: paleontological statistics software package for education and data analysis. Palaeontologia Electronica 4. Harrowfield, M., Holdgate, G. R., Wilson, C. J., & McLoughlin, S. (2005). Tectonic significance of the Lambert Graben, East Antarctica: Reconstructing the Gondwanan rift. Geology, 33(3), 197–200. https://doi.org/10.1130/G21081.1 Hartman, J. D., Sangiorgi, F., Salabarnada, A., Peterse, F., Houben, A. J. P., Schouten, S., Brinkhuis, H., Escutia, C., & Bijl, P. K. (2018). Paleoceanography and ice sheet variability offshore Wilkes Land, Antarctica – Part 3: Insights from Oligocene–Miocene TEX86-based sea surface temperature reconstructions. Climate of the Past, 14(9), 1275– 1297. https://doi.org/10.5194/cp-14-1275-2018 Hemingway, J.D., Schefuß, E., Dinga, B.J., Pryer, H. and Galy, V.V. (2016) Multiple plant-wax compounds record differential sources and ecosystem structure in large river catchments. Geochimica et Cosmochimica Acta, 184, 20-40. Henderiks, J., & Pagani, M. (2008). Coccolithophore cell size and the Paleogene decline in atmospheric CO2. Earth and Planetary Science Letters, 269(3), 576–584. https://doi.org/10.1016/j.epsl.2008.03.016 Henehan, M. J., Edgar, K. M., Foster, G. L., Penman, D. E., Hull, P. M., Greenop, R., Anagnostou, E., & Pearson, P. N. (2020). Revisiting the Middle Eocene Climatic Optimum “Carbon Cycle Conundrum” with new estimates of atmospheric pCO2 from boron isotopes. Paleoceanography and Paleoclimatology, 35(6), e2019PA003713. Herbert, T. D. (2014). Alkenone Paleotemperature Determinations. In Treatise on Geochemistry (pp. 399–433). Elsevier. https://doi.org/10.1016/B978-0-08-095975-7.00615-X 195 Hill, D. J., Haywood, A. M., Valdes, P. J., Francis, J. E., Lunt, D. J., Wade, B. S., & Bowman, V. C. (2013). Paleogeographic controls on the onset of the Antarctic circumpolar current. Geophysical Research Letters, 40(19), 5199–5204. https://doi.org/10.1002/grl.50941 Ho, S. L., Mollenhauer, G., Lamy, F., Martínez-Garcia, A., Mohtadi, M., Gersonde, R., Hebbeln, D., Nunez-Ricardo, S., Rosell-Melé, A., & Tiedemann, R. (2012). Sea surface temperature variability in the Pacific sector of the Southern Ocean over the past 700 kyr. Paleoceanography, 27(4). https://doi.org/10.1029/2012PA002317 Hochmuth, K., Gohl, K., Leitchenkov, G., Sauermilch, I., Whittaker, J., Uenzelmann-Neben, G., et al. (2020). The evolving paleobathymetry of the circum‐Antarctic Southern Ocean since 34 Ma – a key to understanding past cryosphere‐ocean developments. Geochemistry, Geophysics, Geosystems. https://doi.org/10.1029/2020GC009122 Hochmuth, K., Paxman, G. J. G., Gohl, K., Jamieson, S. S. R., Leitchenkov, G. L., Bentley, M. J., et al. (2020). Combined palaeotopography and palaeobathymetry of the Antarctic continent and the Southern Ocean since 34 Ma. PANGAEA. https://doi.org/10.1594/PANGAEA.923109 Hoem, F. S., Valero, L., Evangelinos, D., Escutia, C., Duncan, B., McKay, R. M., Brinkhuis, H., Sangiorgi, F., & Bijl, P. K. (2020). Temperate Oligocene surface ocean conditions offshore Cape Adare, Ross Sea, Antarctica. Climate of the Past Discussions, 2020, 1–32. https://doi.org/10.5194/cp-2020-139 Holbourn, A., Kuhnt, W., Clemens, S., Prell, W., & Andersen, N. (2013). Middle to late Miocene stepwise climate cooling: Evidence from a high-resolution deep water isotope curve spanning 8 million years. Paleoceanography, 28(4), 688–699. https://doi.org/10.1002/2013PA002538 Hopmans, E. C., Weijers, J. W. H., Schefuß, E., Herfort, L., Sinninghe Damsté, J. S., & Schouten, S. (2004). A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraether lipids. Earth and Planetary Science Letters, 224(1), 107–116. https://doi.org/10.1016/j.epsl.2004.05.012 Hou, B., Alley, N. F., Frakes, L. A., Stoian, L, Cowley, W. M. (2006). Eocene stratigraphic succession in the Eucla Basin of South Australia and correlation to major regional sea- level events. Sedimentary Geology, 183:297-583. Houben, A. J. P., Bijl, P. K., Guerstein, G. R., Sluijs, A., & Brinkhuis, H. (2011). Malvinia escutiana, a new biostratigraphically important Oligocene dinoflagellate cyst from the Southern Ocean. Review of Palaeobotany and Palynology, 165(3), 175–182. https://doi.org/10.1016/j.revpalbo.2011.03.002 Houben, A. J. P., van Mourik, C. A., Montanari, A., Coccioni, R., & Brinkhuis, H. (2012). The Eocene–Oligocene transition: Changes in sea level, temperature or both? Cenozoic Evolution of Antarctic Climates, Oceans and Ice Sheets, 335–336, 75–83. https://doi.org/10.1016/j.palaeo.2011.04.008 196 Houben, A. J. P., Bijl, P. K., Pross, J., Bohaty, S. M., Passchier, S., Stickley, C. E., et al. (2013). Reorganization of Southern Ocean Plankton Ecosystem at the Onset of Antarctic Glaciation. Science, 340(6130), 341. https://doi.org/10.1126/science.1223646 Houben, A. J. P., Bijl, P. K., Sluijs, A., Schouten, S., & Brinkhuis, H. (2019). Late Eocene Southern Ocean Cooling and Invigoration of Circulation Preconditioned Antarctica for Full‐Scale Glaciation. Geochemistry, Geophysics, Geosystems. https://doi.org/10.1029/2019GC008182 Huber, M., & Caballero, R. (2011). The early Eocene equable climate problem revisited. Climate of the Past. Huguet, C., Hopmans, E. C., Febo-Ayala, W., Thompson, D. H., Sinninghe Damsté, J. S., & Schouten, S. (2006). An improved method to determine the absolute abundance of glycerol dibiphytanyl glycerol tetraether lipids. Organic Geochemistry, 37(9), 1036–1041. https://doi.org/10.1016/j.orggeochem.2006.05.008 Hunt, R. J., & Poole, I. (2003). Paleogene West Antarctic climate and vegetation history in light of new data from King George Island. Hutchinson, D. K., Coxall, H. K., Lunt, D. J., Steinthorsdottir, M., de Boer, A. M., Baatsen, M., von der Heydt, A., Huber, M., Kennedy-Asser, A. T., Kunzmann, L., Ladant, J.-B., Lear, C. H., Moraweck, K., Pearson, P. N., Piga, E., Pound, M. J., Salzmann, U., Scher, H. D., Sijp, W. P., … Zhang, Z. (2021). The Eocene–Oligocene transition: A review of marine and terrestrial proxy data, models and model–data comparisons. Climate of the Past, 17(1), 269– 315. https://doi.org/10.5194/cp-17-269-2021 Hutchinson, D. K., Coxall, H. K., OʹRegan, M., Nilsson, J., Caballero, R., & de Boer, A. M. (2019). Arctic closure as a trigger for Atlantic overturning at the Eocene-Oligocene Transition. Nature Communications, 10(1), 3797. https://doi.org/10.1038/s41467-019-11828- z Hutchinson, D. K., de Boer, A. M., Coxall, H. K., Caballero, R., Nilsson, J., & Baatsen, M. (2018). Climate sensitivity and meridional overturning circulation in the late Eocene using GFDL CM2.1. Clim. Past, 14(6), 789–810. https://doi.org/10.5194/cp-14-789-2018 Ingalls, A. E., Shah, S. R., Hansman, R. L., Aluwihare, L. I., Santos, G. M., Druffel, E. R., & Pearson, A. (2006). Quantifying archaeal community autotrophy in the mesopelagic ocean using natural radiocarbon. Proceedings of the National Academy of Sciences, 103(17), 6442– 6447. Inglis, G. N., Farnsworth, A., Lunt, D., Foster, G. L., Hollis, C. J., Pagani, M., Jardine, P. E., Pearson, P. N., Markwick, P., Galsworthy, A. M. J., Raynham, L., Taylor, Kyle. W. R., & Pancost, R. D. (2015). Descent toward the Icehouse: Eocene sea surface cooling inferred from GDGT distributions. Paleoceanography, 30(7), 1000–1020. https://doi.org/10.1002/2014PA002723 197 Inglis, G. N., Naafs, B. D. A., Zheng, Y., McClymont, E. L., Evershed, R. P., & Pancost, R. D. (2018). Distributions of geohopanoids in peat: Implications for the use of hopanoid-based proxies in natural archives. Geochimica et Cosmochimica Acta, 224, 249–261. Inglis, G. N., Farnsworth, A., Collinson, M. E., Carmichael, M. J., Naafs, B. D. A., Lunt, D. J., Valdes, P. J., & Pancost, R. D. (2019). Terrestrial environmental change across the onset of the PETM and the associated impact on biomarker proxies: A cautionary tale. Global and Planetary Change, 181, 102991. https://doi.org/10.1016/j.gloplacha.2019.102991 Inglis, G. N., Carmichael, M. J., Farnsworth, A., Lunt, D. J., & Pancost, R. D. (2020). A long- term, high-latitude record of Eocene hydrological change in the Greenland region. Palaeogeography, Palaeoclimatology, Palaeoecology, 537, 109378. https://doi.org/10.1016/j.palaeo.2019.109378 Ivany, L. C., Brey, T., Huber, M., Buick, D. P., & Schöne, B. R. (2011). El Niño in the Eocene greenhouse recorded by fossil bivalves and wood from Antarctica. Geophysical Research Letters, 38(16). https://doi.org/10.1029/2011GL048635 Jarzen, D. M., Dettmannn, M. E. (1992). Structure and form of austral Cretaceous Normapolles- like pollen. Geobios, 25:569-583. Judd, E. J., Ivany, L. C., DeConto, R. M., Halberstadt, A. R. W., Miklus, N. M., Junium, C. K., & Uveges, B. T. (2019). Seasonally Resolved Proxy Data From the Antarctic Peninsula Support a Heterogeneous Middle Eocene Southern Ocean. Paleoceanography and Paleoclimatology, 34(5), 787–799. https://doi.org/10.1029/2019PA003581 Kalanetra, K. M., Bano, N., & Hollibaugh, J. T. (2009). Ammonia‐oxidizing Archaea in the Arctic Ocean and Antarctic coastal waters. Environmental Microbiology, 11(9), 2434– 2445. 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, 1(5), 329–334. https://doi.org/10.1038/ngeo179 Kemp, D. B., Robinson, S. A., Crame, J. A., Francis, J. E., Ineson, J., Whittle, R. J., Bowman, V., & O’Brien, C. (2014). A cool temperate climate on the Antarctic Peninsula through the latest Cretaceous to early Paleogene. Geology, 42(7), 583–586. https://doi.org/10.1130/G35512.1 Kennedy, A. T., Farnsworth, A., Lunt, D. J., Lear, C. H., & Markwick, P. J. (2015). Atmospheric and oceanic impacts of Antarctic glaciation across the Eocene–Oligocene transition. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 373(2054), 20140419. https://doi.org/10.1098/rsta.2014.0419 Kennedy-Asser, A. T., Lunt, D. J., Valdes, P. J., Ladant, J.-B., Frieling, J., & Lauretano, V. (2020). Changes in the high-latitude Southern Hemisphere through the Eocene– Oligocene transition: A model–data comparison. Climate of the Past, 16(2), 555–573. https://doi.org/10.5194/cp-16-555-2020 198 Kennett, J. P. (1977). Cenozoic evolution of Antarctic glaciation, the Circum Antarctic Ocean, and their impact on global paleoceanography. Journal of Geophysical Research, 82, 3843–3860. Kennett, J. P., & Stott, L. (1991). Abrupt deep-sea warming, palaeoceanographic changes and benthic extinctions at the end of the Palaeocene. Nature, 353(6341), 225–229. Kim, J.-H., Schouten, S., Rodrigo-Gámiz, M., Rampen, S., Marino, G., Huguet, C., Helmke, P., Buscail, R., Hopmans, E. C., Pross, J., Sangiorgi, F., Middelburg, J. B. M., & Sinninghe Damsté, J. S. (2015). Influence of deep-water derived isoprenoid tetraether lipids on the TEX86H paleothermometer in the Mediterranean Sea. Geochimica et Cosmochimica Acta, 150, 125–141. https://doi.org/10.1016/j.gca.2014.11.017 Kim, J., Crosta, X., Willmott, V., Renssen, H., Bonnin, J., Helmke, P., Schouten, S., & Sinninghe Damsté, J. S. (2012). Holocene subsurface temperature variability in the eastern Antarctic continental margin. Geophysical Research Letters, 39(6). Kulhanek, D. K., Levy, R. H., Clowes, C. D., Prebble, J. G., Rodelli, D., Jovane, L., Morgans, H. E. G., Kraus, C., Zwingmann, H., Griffith, E. M., Scher, H. D., McKay, R. M., & Naish, T. R. (2019). Revised chronostratigraphy of DSDP Site 270 and late Oligocene to early Miocene paleoecology of the Ross Sea sector of Antarctica. Global and Planetary Change, 178, 46–64. https://doi.org/10.1016/j.gloplacha.2019.04.002 Kvenvolden, K., D. Hostettler, F., B. Rapp, J., & Frank, T. (1991). Aliphatic hydrocarbons in sediments from Prydz Bay, Antarctica. Proceedings of Ocean Drilling Program: Part B (119), 417–423. Ladant, J.-B., Donnadieu, Y., & Dumas, C. (2014). Links between CO2, glaciation and water flow: Reconciling the Cenozoic history of the Antarctic Circumpolar Current. Clim. Past, 10(6), 1957–1966. https://doi.org/10.5194/cp-10-1957-2014 Ladant, J.-B., Donnadieu, Y., Lefebvre, V., & Dumas, C. (2014). The respective role of atmospheric carbon dioxide and orbital parameters on ice sheet evolution at the Eocene- Oligocene transition. Paleoceanography, 29(8), 810–823. https://doi.org/10.1002/2013PA002593 Lauretano, V., Kennedy-Asser, A. T., Korasidis, V. A., Wallace, M. W., Valdes, P. J., Lunt, D. J., Pancost, R. D., & Naafs, B. D. A. (2021). Eocene to Oligocene terrestrial Southern Hemisphere cooling caused by declining pCO2. Nature Geoscience, 14(9), 659–664. https://doi.org/10.1038/s41561-021-00788-z Lear, C. H., Bailey, T. R., Pearson, P. N., Coxall, H. K., & Rosenthal, Y. (2008). Cooling and ice growth across the Eocene-Oligocene transition. Geology, 36(3), 251–254. https://doi.org/10.1130/G24584A.1 Lear, C. H., Rosenthal, Y., Coxall, H. K., & Wilson, P. A. (2004). Late Eocene to early Miocene ice sheet dynamics and the global carbon cycle. Paleoceanography, 19(4). https://doi.org/10.1029/2004PA001039 199 Legendre, P., Legendre, L. (2012). Numerical ecology, 3rd Edition. Developments in Environmental Modelling 24, 1006. Leutert, T. J., Auderset, A., Martínez-García, A., Modestou, S., & Meckler, A. N. (2020). Coupled Southern Ocean cooling and Antarctic ice sheet expansion during the middle Miocene. Nature Geoscience, 13(9), 634–639. https://doi.org/10.1038/s41561-020-0623- 0 Levy, R., Harwood, D., Florindo, F., Sangiorgi, F., Tripati, R., Von Eynatten, H., Gasson, E., Kuhn, G., Tripati, A., & DeConto, R. (2016). Antarctic ice sheet sensitivity to atmospheric CO2 variations in the early to mid-Miocene. Proceedings of the National Academy of Sciences, 113(13), 3453–3458. Levy, R. H., Harwood, D. M. (2000). Tertiary marine palynomorphs from the McMurdo Sound erratics, Antarctica. AGU Antarctic Research Series. 76:183-242. Lewis, A. R., Marchant, D. R., Ashworth, A. C., Hedenäs, L., Hemming, S. R., Johnson, J. V., Leng, M. J., Machlus, M. L., Newton, A. E., Raine, I., Willenbring, J. K., Williams, M., Wolfe, A. P. (2008). Mid-Miocene cooling and the extinction of tundra in continental Antarctica. Proceedings of the National Academy of Science, 105, 31. Li, C., Sessions, A. L., Valentine, D. L., & Thiagarajan, N. (2011). D/H variation in terrestrial lipids from Santa Barbara Basin over the past 1400years: A preliminary assessment of paleoclimatic relevance. Organic Geochemistry, 42(1), 15–24. https://doi.org/10.1016/j.orggeochem.2010.09.011 Liu, Z., Pagani, M., Zinniker, D., DeConto, R., Huber, M., Brinkhuis, H., Shah, S. R., Leckie, R. M., & Pearson, A. (2009). Global Cooling During the Eocene-Oligocene Climate Transition. Science, 323(5918), 1187. https://doi.org/10.1126/science.1166368 Lopes dos Santos, R. A., Prange, M., Castañeda, I. S., Schefuß, E., Mulitza, S., Schulz, M., Niedermeyer, E. M., Sinninghe Damsté, J. S., & Schouten, S. (2010). Glacial–interglacial variability in Atlantic meridional overturning circulation and thermocline adjustments in the tropical North Atlantic. Earth and Planetary Science Letters, 300(3), 407–414. https://doi.org/10.1016/j.epsl.2010.10.030 MacPhail, M. K. (2021). The Sabrina Floras of East Antarctica: Late Cretaceous, Paleogene or reworked? Palynology, 45, doi: 10.1080/01916122.2021.1921070. MacPhail, M. K., Truswell, E. M. (2004). Palynology of Neogene slope and rise deposits from ODP Sites 1165 and 1167, East Antarctica. Pp.1-20 in: Cooper, A.K.; O'Brien, P.E. Richter, C. (eds). Proceedings of the Ocean Drilling Program, scientific results, volume 188. Marenssi, S. A., Casadío, S., & Santillana, S. N. (2010). Record of Late Miocene glacial deposits on Isla Marambio (Seymour Island), Antarctic Peninsula. Antarctic Science, 22(2), 193– 198. Cambridge Core. https://doi.org/10.1017/S0954102009990629 200 Marschalek, J. W., Zurli, L., Talarico, F., van de Flierdt, T., Vermeesch, P., Carter, A., Beny, F., Bout-Roumazeilles, V., Sangiorgi, F., Hemming, S. R., Pérez, L. F., Colleoni, F., Prebble, J. G., van Peer, T. E., Perotti, M., Shevenell, A. E., Browne, I., Kulhanek, D. K., Levy, R., … IODP Expedition 374. (2021). A large West Antarctic Ice Sheet explains early Neogene sea-level amplitude. Nature, 600(7889), 450–455. https://doi.org/10.1038/s41586-021-04148-0 Martínez-Garcia, A., Rosell-Melé, A., Jaccard, S. L., Geibert, W., Sigman, D. M., & Haug, G. H. (2011). Southern Ocean dust–climate coupling over the past four million years. Nature, 476(7360), 312–315. https://doi.org/10.1038/nature10310 Masson-Delmotte, V., Hou, S., Ekaykin, A., Jouzel, J., Aristarain, A., Bernardo, R. T., Bromwich, D., Cattani, O., Delmotte, M., Falourd, S., Frezzotti, M., Gallée, H., Genoni, L., Isaksson, E., Landais, A., Helsen, M. M., Hoffmann, G., Lopez, J., Morgan, V., Motoyama, N., Noone, D., Oerter, H., Petit, J. R., Royer, A., Uemura, R., Schmidt, G. A., Schlosser, E., Simöes, J. C., Steig, E. J., Stenni, B., Stievenard, M., van den Broeke, M. R., van de Wal, R. S. W., van de Berg, W. J., Vimeux, F., White, J. W. C. (2008). A review of Antarctic surface snow isotopic composition: Observations, atmospheric circulation, and isotopic modeling. Journal of Climate 21, 3359–3387. McFarlin, J.M., Axford, Y., Masterson, A.L., Osburn, M.R. (2019). Calibration of modern sedimentary δ2H plant wax-water relationships in Greenland lakes. Quaternary Science Reviews 225, 105978. McKay, R. M, Escutia, C., De Santis, L., Donda, F., Duncan, B., Gohl, K., Gulick, S., Hernández-Molina, J., Hillenbrand, C.-D., Hochmuth, K., Kim, S., Kuhn, G., Larter, R., Leitchenkov, G., H. Levy, R., R. Naish, T., O’Brien, P., F. Pérez, L., E. Shevenell, A., & Williams, T. (2022). Chapter 3—Cenozoic history of Antarctic glaciation and climate from onshore and offshore studies. In F. Florindo, M. Siegert, L. D. Santis, & T. Naish (Eds.), Antarctic Climate Evolution (Second Edition) (pp. 41–164). Elsevier. https://doi.org/10.1016/B978-0-12-819109-5.00008-6 McKay, R. M., De Santis, L., & Kulhanek, D. K. (2019). Ross Sea West Antarctic Ice Sheet History. Proceedings of the International Ocean Discovery Program, 374. McKay, R., Naish, T., Carter, L., Riesselman, C., Dunbar, R., Sjunneskog, C., Winter, D., Sangiorgi, F., Warren, C., Pagani, M., Schouten, S., Willmott, V., Levy, R., DeConto, R., & Powell, R. D. (2012). Antarctic and Southern Ocean influences on Late Pliocene global cooling. Proceedings of the National Academy of Sciences, 109(17), 6423. https://doi.org/10.1073/pnas.1112248109 McKelvey, B., Hambrey, M., Harwood, D., Mabin, M., Webb, P.-N., & Whitehead, J. (2001). The Pagodroma Group—A Cenozoic record of the East Antarctic ice sheet in the northern Prince Charles Mountains. Antarctic Science, 13(4), 455–468. https://doi.org/10.1017/S095410200100061X McLoughlin, S., Lindström, S., & Drinnan, A. N. (1997). Gondwanan floristic and sedimentological trends during the Permian–Triassic transition: New evidence from the 201 Amery Group, northern Prince Charles Mountains, East Antarctica. Antarctic Science, 9(3), 281–298. Meire, L., Mortensen, J., Meire, P., Juul-Pedersen, T., Sejr, M. K., Rysgaard, S., et al. (2017). Marine-terminating glaciers sustain high productivity in Greenland fjords. Global Change Biology, 23(12), 5344–5357. https://doi.org/10.1111/gcb.13801 Meissner, K. J., Bralower, T. J., Alexander, K., Jones, T. D., Sijp, W., & Ward, M. (2014). The Paleocene-Eocene Thermal Maximum: How much carbon is enough? Paleoceanography, 29(10), 946–963. https://doi.org/10.1002/2014PA002650 Miller, K. G., Browning, J. V., Schmelz, W. J., Kopp, R. E., Mountain, G. S., & Wright, J. D. (2020). Cenozoic sea-level and cryospheric evolution from deep-sea geochemical and continental margin records. Science Advances, 6(20), eaaz1346. https://doi.org/10.1126/sciadv.aaz1346 Miller, K. G., Wright, J. D., & Fairbanks, R. G. (1991). Unlocking the Ice House: Oligocene- Miocene oxygen isotopes, eustasy, and margin erosion. Journal of Geophysical Research: Solid Earth, 96(B4), 6829–6848. https://doi.org/10.1029/90JB02015 Mohr, B. (1990). Eocene and Oligocene Sporomorphs and Dinoflagellate Cysts from Leg 113 Drill Sites|Weddell Sea|Antarctica. Proc., Scientific Results, ODP, Leg 113, Weddell Sea, Antarctica, 120. https://doi.org/10.2973/odp.proc.sr.113.140.1990 Mollenhauer, G., Eglinton, T., Ohkouchi, N., Schneider, R. R., Müller, P. J., Grootes, P., & Rullkötter, J. (2003). Asynchronous alkenone and foraminfera records from the Benguela Upwelling System. Geochimica et Cosmochimica Acta, 67, 2157–2171. https://doi.org/10.1016/S0016-7037(03)00168-6 Montelli, A., Gulick, S., Fernandez-Vasquez, R., Frederick, B., Shevenell, A., Leventer, A., Blankenship, D. (2019). Seismic stratigraphy of the Sabrina Coast shelf, East Antarctica: history of early dynamic glaciations. GSA Bulletin, https://doi.org/10.1130/B35100.1. Morlighem, M., Rignot, E., Binder, T., Blankenship, D., Drews, R., Eagles, G. (2020). Deep glacial troughs and stabilizing ridges unveiled beneath the margins of the Antarctic ice sheet. Nature Geoscience, 13(2), 132-137. Müller, P. J., Kirst, G., Ruhland, G., Von Storch, I., & Rosell-Melé, A. (1998). Calibration of the alkenone paleotemperature index U37K′ based on core-tops from the eastern South Atlantic and the global ocean (60 N-60 S). Geochimica et Cosmochimica Acta, 62(10), 1757–1772. Naish, T., Powell, R., Levy, R., Wilson, G., Scherer, R., Talarico, F., Krissek, L., Niessen, F., Pompilio, M., Wilson, T., Carter, L., DeConto, R., Huybers, P., McKay, R., Pollard, D., Ross, J., Winter, D., Barrett, P., Browne, G., … Williams, T. (2009). Obliquity-paced Pliocene West Antarctic ice sheet oscillations. Nature, 458(7236), 322–328. https://doi.org/10.1038/nature07867 O’Brien, P. E., Cooper, A. K., & Richter, C. (2001). Leg 188 summary: Prydz Bay-Cooperation Sea, Antarctica. Proceedings of the Ocean Drilling Program. Initial Reports, 188, 1–65. 202 O’Brien, C. L., Huber, M., Thomas, E., Pagani, M., Super, J. R., Elder, L. E., & Hull, P. M. (2020). The enigma of Oligocene climate and global surface temperature evolution. Proceedings of the National Academy of Sciences, 117(41), 25302. https://doi.org/10.1073/pnas.2003914117 Pagani, M., Huber, M., Liu, Z., Bohaty, S. M., Henderiks, J., Sijp, W., Krishnan, S., & DeConto, R. M. (2011). The role of carbon dioxide during the onset of Antarctic glaciation. Science, 334(6060), 1261. https://doi.org/10.1126/science.1203909 Pagani, M., Liu, Z., LaRiviere, J., & Ravelo, A. C. (2010). High Earth-system climate sensitivity determined from Pliocene carbon dioxide concentrations. Nature Geoscience, 3(1), 27–30. 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(5734), 600. https://doi.org/10.1126/science.1110063 Pancost, R. D., & Boot, C. S. (2004). The palaeoclimatic utility of terrestrial biomarkers in marine sediments. New Approaches in Marine Organic Biogeochemistry: A Tribute to the Life and Science of John I. Hedges, 92(1), 239–261. https://doi.org/10.1016/j.marchem.2004.06.029 Partridge, A. D. (2006). Late Cretaceous–Cenozoic palynology zonations, Gippsland Basin. In: Monteil, E. (coord.), Australian Mesozoic and Cenozoic Palynology Zonations. Passchier, S., Bohaty, S. M., Jimenez-Espejo, F. J., Pross, J., Roehl, U., van de Flierdt, T., Escutia, C., & Brinkhuis, H. (2013). Early Eocene to middle Miocene cooling and aridification of East Antarctica. Geochemistry Geophysics Geosystems, 14(5), 1399–1410. https://doi.org/10.1002/ggge.20106 Passchier, S., Ciarletta, D. J., Miriagos, T. E., Bijl, P. K., & Bohaty, S. M. (2017). An Antarctic stratigraphic record of stepwise ice growth through the Eocene-Oligocene transition. Geological Society of America Bulletin, 129(3–4), 318–330. https://doi.org/10.1130/B31482.1 Paxman, G. J. G., Jamieson, S. S. R., Hochmuth, K., Gohl, K., Bentley, M. J., Leitchenkov, G., & Ferraccioli, F. (2019). Reconstructions of Antarctic topography since the Eocene– Oligocene boundary. Palaeogeography, Palaeoclimatology, Palaeoecology, 535, 109346. https://doi.org/10.1016/j.palaeo.2019.109346 Pearson, P. N., Foster, G. L., & Wade, B. S. (2009). Atmospheric carbon dioxide through the Eocene–Oligocene climate transition. Nature, 461(7267), 1110–1113. https://doi.org/10.1038/nature08447 Petersen, S. V., & Schrag, D. P. (2015). Antarctic ice growth before and after the Eocene- Oligocene transition: New estimates from clumped isotope paleothermometry: Antarctic ice growth at the E/O transition. Paleoceanography, 30(10), 1305–1317. https://doi.org/10.1002/2014PA002769 203 Plancq, J., Mattioli, E., Pittet, B., Simon, L., & Grossi, V. (2014). Productivity and sea-surface temperature changes recorded during the late Eocene–early Oligocene at DSDP Site 511 (South Atlantic). Palaeogeography, Palaeoclimatology, Palaeoecology, 407, 34–44. https://doi.org/10.1016/j.palaeo.2014.04.016 Poole, I., Cantrill, D., & Utescher, T. (2005). A multi-proxy approach to determine Antarctic terrestrial palaeoclimate during the Late Cretaceous and Early Tertiary. Palaeogeography, Palaeoclimatology, Palaeoecology, 222(1), 95–121. https://doi.org/10.1016/j.palaeo.2005.03.011 Popp, B. N., Prahl, F. G., Wallsgrove, R. J., & Tanimoto, J. (2006). Seasonal patterns of alkenone production in the subtropical oligotrophic North Pacific. Paleoceanography, 21(1). Scopus. https://doi.org/10.1029/2005PA001165 Pound, M. J., & Salzmann, U. (2017). Heterogeneity in global vegetation and terrestrial climate change during the late Eocene to early Oligocene transition. Scientific Reports, 7, 43386. https://doi.org/10.1038/srep43386 Pound, M. J., Haywood, A. M., Salzmann, U., & Riding, J. B. (2012). Global vegetation dynamics and latitudinal temperature gradients during the Mid to Late Miocene (15.97– 5.33Ma). Earth-Science Reviews, 112(1), 1–22. https://doi.org/10.1016/j.earscirev.2012.02.005 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) cored by the Cape Roberts Project, Victoria Land Basin, Antarctica. Antarctic Climate Evolution: Geological Records from the Margin and Modelling, 231(1), 41–57. https://doi.org/10.1016/j.palaeo.2005.07.025 Pross, J., Contreras, L., Bijl, P. K., Greenwood, D. R., Bohaty, S. M., Schouten, S., Bendle, J. A., Rohl, U., Trauze, L. (2012). Persistent near-tropical warmth on the Antarctic continent during the Early Eocene epoch. Nature, 488 (7409):73-77. Raine, J. I., Mildenhall, D. C., Kennedy, E. M., (2011). New Zealand fossil spores and pollen: an illustrated catalogue. 4 th edition. GNS Science miscellaneous series no. 4. http://data.gns.cri.nz/sporespollen/index.htm. Rae, J. W. B., Zhang, Y. G., Liu, X., Foster, G. L., Stoll, H. M., & Whiteford, R. D. M. (2021). Atmospheric CO2 over the Past 66 Million Years from Marine Archives. Annual Review of Earth and Planetary Sciences, 49(1), 609–641. https://doi.org/10.1146/annurev-earth- 082420-063026 Raine, J. I. (1984). Outline of a palynological zonation of Cretaceous to Paleogene terrestrial sediments in west coast region, South Island, New Zealand. Republic of New Zealand Geological Survey. 109:1-82. Raine, J. (1998). Terrestrial palynomorphs from Cape Roberts Project drillhole CRP-1, Ross Sea, Antarctica. Terra Antartica, 5(3), 539–548. 204 Raitzsch, M., & Hönisch, B. (2013). Cenozoic boron isotope variations in benthic foraminifers. Geology, 41(5), 591–594. https://doi.org/10.1130/G34031.1 Rau, D. (2017). In search of Antarctica’s last vegetation refugium within the McMurdo Dry Valleys. LSU Master’s Thesis. https://digitalcommons.lsu.edu/gradschool_theses/4468/. Rees-Owen, R. L., Gill, F. L., Newton, R. J., Ivanović, R. F., Francis, J. E., Riding, J. B., Vane, C. H., & Lopes dos Santos, R. A. (2018). The last forests on Antarctica: Reconstructing flora and temperature from the Neogene Sirius Group, Transantarctic Mountains. Organic Geochemistry, 118, 4–14. https://doi.org/10.1016/j.orggeochem.2018.01.001 Rose, K. C., Ferraccioli, F., Jamieson, S. S. R., Bell, R. E., Corr, H., Creyts, T. T., Braaten, D., Jordan, T. A., Fretwell, P. T., & Damaske, D. (2013). Early East Antarctic Ice Sheet growth recorded in the landscape of the Gamburtsev Subglacial Mountains. Earth and Planetary Science Letters, 375, 1–12. https://doi.org/10.1016/j.epsl.2013.03.053 Salzmann, U., Riding, J. B., Nelson, A. E., & Smellie, J. L. (2011). How likely was a green Antarctic Peninsula during warm Pliocene interglacials? A critical reassessment based on new palynofloras from James Ross Island. Special Issue: Climate and Seasonality in a Pliocene Warm World, 309(1), 73–82. https://doi.org/10.1016/j.palaeo.2011.01.028 Sangiorgi, F., Bijl, P. K., Passchier, S., Salzmann, U., Schouten, S., McKay, R., Cody, R. D., Pross, J., van de Flierdt, T., Bohaty, S. M., Levy, R., Williams, T., Escutia, C., & Brinkhuis, H. (2018). Southern Ocean warming and Wilkes Land ice sheet retreat during the mid-Miocene. Nature Communications, 9(1), 317. https://doi.org/10.1038/s41467- 017-02609-7 Sauermilch, I., Whittaker, J. M., Klocker, A., Munday, D. R., Hochmuth, K., Bijl, P. K., & LaCasce, J. H. (2021). Gateway-driven weakening of ocean gyres leads to Southern Ocean cooling. Nature Communications, 12(1), 6465. https://doi.org/10.1038/s41467- 021-26658-1 Scher, H. D., Bohaty, S. M., Smith, B. W., & Munn, G. H. (2014). Isotopic interrogation of a suspected late Eocene glaciation. Paleoceanography, 29(6), 628–644. https://doi.org/10.1002/2014PA002648 Scher, H. D., Bohaty, S. M., Zachos, J. C., & Delaney, M. L. (2011). Two-stepping into the icehouse: East Antarctic weathering during progressive ice-sheet expansion at the Eocene– Oligocene transition. Geology, 39(4), 383–386. https://doi.org/10.1130/G31726.1 Scher, H. D., Whittaker, J. M., Williams, S. E., Latimer, J. C., Kordesch, W. E. C., & Delaney, M. L. (2015). Onset of Antarctic Circumpolar Current 30 million years ago as Tasmanian Gateway aligned with westerlies. Nature, 523(7562), 580–583. https://doi.org/10.1038/nature14598 Schimmelmann, A., Sessions, A. L., & Mastalerz, M. (2006). Hydrogen isotopic (D/H) composition of organic matter during diagenesis and thermal maturation. Annual Review of 205 Earth and Planetary Sciences, 34(1), 501–533. https://doi.org/10.1146/annurev.earth.34.031405.125011 Schouten, S., Hopmans, E. C., Schefuß, E., & Damste, J. S. S. (2002). Distributional variations in marine crenarchaeotal membrane lipids: A new tool for reconstructing ancient sea water temperatures? Earth and Planetary Science Letters, 204(1–2), 265–274. Schouten, S., Huguet, C., Hopmans, E. C., Kienhuis, M. V. M., & Sinninghe Damsté, J. S. (2007). Analytical Methodology for TEX86 Paleothermometry by High-Performance Liquid Chromatography/Atmospheric Pressure Chemical Ionization-Mass Spectrometry. Analytical Chemistry, 79(7), 2940–2944. https://doi.org/10.1021/ac062339v Schouten, S., Hopmans, E. C., & Damsté, J. S. S. (2013). The organic geochemistry of glycerol dialkyl glycerol tetraether lipids: A review. Organic Geochemistry, 54, 19–61. Sessions, A. L., Zhang, L., Welander, P. V., Doughty, D., Summons, R. E., & Newman, D. K. (2013). Identification and quantification of polyfunctionalized hopanoids by high temperature gas chromatography-mass spectrometry. Organic Geochemistry, 56, 120– 130. PubMed. https://doi.org/10.1016/j.orggeochem.2012.12.009. Sheldon, N. D., Retallack, G. J., & Tanaka, S. (2002). Geochemical Climofunctions from North American Soils and Application to Paleosols across the Eocene‐Oligocene Boundary in Oregon. The Journal of Geology, 110(6), 687–696. https://doi.org/10.1086/342865 Shepherd, A., Ivins, E., Rignot, E., Smith, B., van den Broeke, M., Velicogna, I., et al. (2018). Mass balance of the Antarctic Ice Sheet from 1992 to 2017. Nature, 558(7709), 219–222. https://doi.org/10.1038/s41586-018-0179-y Shevenell, A. E., Kennett, J. P., & Lea, D. W. (2004). Middle Miocene southern ocean cooling and Antarctic cryosphere expansion. Science, 305(5691), 1766–1770. Sijp, W. P., England, M. H., & Toggweiler, J. R. (2009). Effect of Ocean Gateway Changes under Greenhouse Warmth. Journal of Climate, 22(24), 6639–6652. https://doi.org/10.1175/2009JCLI3003.1 Sikes, E. L., & Volkman, J. K. (1993). Calibration of alkenone unsaturation ratios (Uk’37) for paleotemperature estimation in cold polar waters. Geochimica et Cosmochimica Acta, 57(8), 1883–1889. https://doi.org/10.1016/0016-7037(93)90120-L Sikes, E. L., Volkman, J. K., Robertson, L. G., & Pichon, J. J. (1997). Alkenones and alkenes in surface waters and sediments of the Southern Ocean: Implications for paleotemperature estimation in polar regions. Geochimica Et Cosmochimica Acta, 61(7), 1495–1505. https://doi.org/10.1016/S0016-7037(97)00017-3 Sluijs, A., Bijl, P. K., Schouten, S., Röhl, U., Reichart, G.-J., & Brinkhuis, H. (2011). Southern ocean warming, sea level and hydrological change during the Paleocene-Eocene thermal maximum. Climate of the Past, 7(1), 47–61. https://doi.org/10.5194/cp-7-47-2011 206 Smith, C. (2016). An Early Paleogene Palynological Assemblage from the Sabrina Coast, East Antarctica: New Species and Implications for Depositional History, University of South Florida Master’s thesis. Smith, C., Warny, S., Shevenell, A. E., Gulick, S. P. S., Leventer, A. (2019). New species from the Sabrina Flora: an early Paleogene pollen and spore assemblage from the Sabrina Coast, East Antarctica. Palynology, 43, 650-659, doi: 10.1080/01916122.2018.1471422. Smith, R. T., & Anderson, J. B. (2010). Ice-sheet evolution in James Ross Basin, Weddell Sea margin of the Antarctic Peninsula: The seismic stratigraphic record. GSA Bulletin, 122(5– 6), 830–842. https://doi.org/10.1130/B26486.1 Smittenberg, R. H., Hopmans, E. C., Schouten, S., Hayes, J. M., Eglinton, T. I., & Sinninghe Damsté, J. (2004). Compound‐specific radiocarbon dating of the varved Holocene sedimentary record of Saanich Inlet, Canada. Paleoceanography, 19(2). Sophie Warny, C. Madison Kymes, Rosemary Askin, Krzysztof P. Krajewski, & Andrzej Tatur. (2019). Terrestrial and Marine Floral Response to Latest Eocene and Oligocene Events on the Antarctic Peninsula. Palynology, 43(1), 4–21. https://doi.org/10.1080/01916122.2017.1418444 Sosdian, S. M., Greenop, R., Hain, M., Foster, G. L., Pearson, P. N., & Lear, C. H. (2018). Constraining the evolution of Neogene ocean carbonate chemistry using the boron isotope pH proxy. Earth and Planetary Science Letters, 498, 362–376. Speelman, E. N., van Kempen, M. M., Barke, J., Brinkhuis, H., Reichart, G.-J., Smolders, A. J., Roelofs, J. G., Sangiorgi, F., de Leeuw, J. W., & Lotter, A. F. (2009). The Eocene Arctic Azolla bloom: Environmental conditions, productivity and carbon drawdown. Geobiology, 7(2), 155–170. Stocchi, P., Escutia, C., Houben, A. J. P., Vermeersen, B. L. A., Bijl, P. K., Brinkhuis, H., et al. (2013). Relative sea-level rise around East Antarctica during Oligocene glaciation. Nature Geoscience, 6(5), 380–384. https://doi.org/10.1038/ngeo1783 Stover, L.E., Partridge, A.D. (1973). Tertiary and Late Cretaceous spores and pollen from the Gippsland Basin, southeastern Australia. Proceedings of the Royal Society of Victoria 85: 237-286. Thompson, N., Salzmann, U., López-Quirós, A., Bijl, P. K., Hoem, F. S., Etourneau, J., Sicre, M.-A., Roignant, S., Hocking, E., Amoo, M., & Escutia, C. (2022). Vegetation change across the Drake Passage region linked to late Eocene cooling and glacial disturbance after the Eocene–Oligocene transition. Climate of the Past, 18(2), 209–232. https://doi.org/10.5194/cp-18-209-2022 Tibbett, E. J., Scher, H. D., Warny, S., Tierney, J. E., Passchier, S., & Feakins, S. J. (2021). Late Eocene Record of Hydrology and Temperature From Prydz Bay, East Antarctica. Paleoceanography and Paleoclimatology, 36(4), e2020PA004204. https://doi.org/10.1029/2020PA004204 207 Tierney, J. E., & Tingley, M. P. (2014). A Bayesian, spatially-varying calibration model for the TEX86 proxy. Geochimica et Cosmochimica Acta, 127, 83–106. https://doi.org/10.1016/j.gca.2013.11.026 Tierney, J. E., & Tingley, M. P. (2015). A TEX86 surface sediment database and extended Bayesian calibration. Scientific Data, 2(1), 150029. https://doi.org/10.1038/sdata.2015.29 Tipple, B. J., Meyers, S. R., & Pagani, M. (2010). Carbon isotope ratio of Cenozoic CO2: A comparative evaluation of available geochemical proxies. Paleoceanography, 25(3). https://doi.org/10.1029/2009PA001851 Tipple, B. J., Pagani, M., Krishnan, S., Dirghangi, S. S., Galeotti, S., Agnini, C., Giusberti, L., & Rio, D. (2011). Coupled high-resolution marine and terrestrial records of carbon and hydrologic cycles variations during the Paleocene–Eocene Thermal Maximum (PETM). Earth and Planetary Science Letters, 311(1), 82–92. https://doi.org/10.1016/j.epsl.2011.08.045 Tochilin, C. J., Reiners, P. W., Thomson, S. N., Gehrels, G. E., Hemming, S. R., & Pierce, E. L. (2012). Erosional history of the Prydz Bay sector of East Antarctica from detrital apatite and zircon geo- and thermochronology multidating. Geochemistry, Geophysics, Geosystems, 13(11). https://doi.org/10.1029/2012GC004364 Toumoulin, A., Donnadieu, Y., Ladant, J.-B., Batenburg, S. J., Poblete, F., & Dupont-Nivet, G. (2020). Quantifying the Effect of the Drake Passage Opening on the Eocene Ocean. Paleoceanography and Paleoclimatology, 35(8), e2020PA003889. https://doi.org/10.1029/2020PA003889 Truswell, E. M. (1983). Recycled Cretaceous and Tertiary pollen and spores in Antarctic marine sediments: a catalogue. Palaeontographica Abteilung B. 186:121-174. Truswell, E. M., & Macphail, M. K. (2009). Polar forests on the edge of extinction: What does the fossil spore and pollen evidence from East Antarctica say? Australian Systematic Botany, 22(2), 57–106. https://doi.org/10.1071/SB08046 Uemura, H., & Ishiwatari, R. (1995). Identification of unusual 17β(H)-moret-22(29)-ene in lake sediments. Organic Geochemistry, 23(7), 675–680, doi: 10.1016/0146-6380(95)00036-E. Vahlenkamp, M., Niezgodzki, I., De Vleeschouwer, D., Lohmann, G., Bickert, T., & Pälike, H. (2018). Ocean and climate response to North Atlantic seaway changes at the onset of long-term Eocene cooling. Earth and Planetary Science Letters, 498, 185–195. https://doi.org/10.1016/j.epsl.2018.06.031 van de Flierdt, T., Hemming, S. R., Goldstein, S. L., Gehrels, G. E., & Cox, S. E. (2008). Evidence against a young volcanic origin of the Gamburtsev Subglacial Mountains, Antarctica. Geophysical Research Letters, 35(21). https://doi.org/10.1029/2008GL035564 Villa, G., Fioroni, C., Persico, D., Roberts, A. P., & Florindo, F. (2014). Middle Eocene to Late Oligocene Antarctic glaciation/deglaciation and Southern Ocean productivity. Paleoceanography, 29(3), 223–237. https://doi.org/10.1002/2013PA002518 208 Volkman, J. K., Eglinton, G., Corner, E. D. S., & Forsberg, T. E. V. (1980). Long-chain alkenes and alkenones in the marine coccolithophorid Emiliania huxleyi. Phytochemistry, 19(12), 2619–2622. https://doi.org/10.1016/S0031-9422(00)83930-8 Warny, S., Askin, R. A., Hannah, M. J., Mohr, B. A. R., Raine, J. I., Harwood, D. M., Florindo, F., & the SMS Science Team. (2009). Palynomorphs from a sediment core reveal a sudden remarkably warm Antarctica during the middle Miocene. Geology, 37(10), 955– 958. https://doi.org/10.1130/G30139A.1 Warny, S. S., & Askin, R. R. (2011). Vegetation and Organic-Walled Phytoplankton at the End of the Antarctic Greenhouse World: Latest Eocene Cooling Events. In Tectonic, Climatic, and Cryospheric Evolution of the Antarctic Peninsula (pp. 193–210). American Geophysical Union (AGU). https://doi.org/10.1029/2010SP000965 Warny, S., & Askin, R. (2011). Last remnants of Cenozoic vegetation and organic-walled phytoplankton in the Antarctic Peninsula’s icehouse world. In Tectonic, climatic, and cryospheric evolution of the Antarctic peninsula (Vol. 63, pp. 167–192). American Geophysical Union Washington, DC. Warny, S., Kymes, C. M., Askin, R., Krajewski, K. P., & Tatur, A. (2019). Terrestrial and marine floral response to latest Eocene and Oligocene events on the Antarctic Peninsula. Palynology, 43(1), 4-21. Weijers, J. W. H., Bernhardt, B., Peterse, F., Werne, J. P., Dungait, J. A. J., Schouten, S., & Sinninghe Damsté, J. S. (2011). Absence of seasonal patterns in MBT–CBT indices in mid-latitude soils. Geochimica et Cosmochimica Acta, 75(11), 3179–3190. https://doi.org/10.1016/j.gca.2011.03.015 Weijers, J. W. H., Schouten, S., van den Donker, J. C., Hopmans, E. C., & Sinninghe Damsté, J. S. (2007). Environmental controls on bacterial tetraether membrane lipid distribution in soils. Geochimica et Cosmochimica Acta, 71(3), 703–713. https://doi.org/10.1016/j.gca.2006.10.003 Westerhold, T., Marwan, N., Drury, A. J., Liebrand, D., Agnini, C., Anagnostou, E., Barnet, J. S. K., Bohaty, S. M., Vleeschouwer, D. D., Florindo, F., Frederichs, T., Hodell, D. A., Holbourn, A. E., Kroon, D., Lauretano, V., Littler, K., Lourens, L. J., Lyle, M., Pälike, H., … Zachos, J. C. (2020). An astronomically dated record of Earth’s climate and its predictability over the last 66 million years. Science, 369(6509), 1383–1387. https://doi.org/10.1126/science.aba6853 Wilkie, K.M.K., Chapligin, B., Meyer, H., Burns, S., Petsch, S., Brigham-Grette, J. (2013). Modern isotope hydrology and controls on δD of plant leaf waxes at Lake El'gygytgyn, NE Russia. Climate of the Past 9, 335-352. Wilson, D. J., Bertram, R. A., Needham, E. F., van de Flierdt, T., Welsh, K. J., McKay, R. M., Mazumder, A., Riesselman, C. R., Jimenez-Espejo, F. J., & Escutia, C. (2018). Ice loss from the East Antarctic Ice Sheet during late Pleistocene interglacials. Nature, 561(7723), 383–386. https://doi.org/10.1038/s41586-018-0501-8 209 Wilson, D. S., Pollard, D., DeConto, R. M., Jamieson, S. S. R., & Luyendyk, B. P. (2013). Initiation of the West Antarctic Ice Sheet and estimates of total Antarctic ice volume in the earliest Oligocene. Geophysical Research Letters, 40(16), 4305–4309. https://doi.org/10.1002/grl.50797 Wright, A. P., Young, D. A., Roberts, J. L., Schroeder, D. M., Bamber, J. L., Dowdeswell, J. A., Young, N. W., Le Brocq, A. M., Warner, R. C., Payne, A. J., Blankenship, D. D., van Ommen, T. D., Siegert, M. J. (2012). Evidence of a hydrological connection between the ice divide and ice sheet margin in the Aurora Subglacial Basin, East Antarctica. Journal of Geophysical Research, 117, F01033, doi:10.1029/2011JF002066. Wright, N. M., Scher, H. D., Seton, M., Huck, C. E., & Duggan, B. D. (2018). No change in Southern Ocean circulation in the Indian Ocean from the Eocene through Late Oligocene. Paleoceanography and Paleoclimatology, 33(2), 152–167. https://doi.org/10.1002/2017PA003238 Wuchter, C., Schouten, S., Boschker, H. T., & Sinninghe Damsté, J. S. (2003). Bicarbonate uptake by marine Crenarchaeota. FEMS Microbiology Letters, 219(2), 203–207. Yamamoto, M., Ficken, K., Baas, M., Bosch, H.-J., & de Leeuw, J. W. (1996). Molecular palaeontology of the earliest Danian at Geulhemmerberg (the Netherlands). Geologie En Mijnbouw, 75(2), 255–267. Young, D.A., Wright, A. P., Roberts, J. L., Warner, R. C., Young, N. W., Greenbaum, J. S., Schroeder, D. M., Holt, J. W., Sugden, D. E., Blankenship, D.D., van Omen, T. D., Siegert, M. J. (2011). A dynamic early East Antarctic Ice Sheet suggested by ice-covered fjord landscapes. Nature, 474, 72-75. Zachos, J. C., Quinn, T. M., & Salamy, K. A. (1996). High‐resolution (104 years) deep‐sea foraminiferal stable isotope records of the Eocene‐Oligocene climate transition. Paleoceanography, 11(3), 251–266. Zachos, J. C., Shackleton, N. J., Revenaugh, J. S., Pälike, H., & Flower, B. P. (2001). Climate Response to Orbital Forcing Across the Oligocene-Miocene Boundary. Science, 292(5515), 274. https://doi.org/10.1126/science.1058288 Zachos, J. C., Stott, L. D., & Lohmann, K. C. (1994). Evolution of Early Cenozoic marine temperatures. Paleoceanography, 9(2), 353–387. https://doi.org/10.1029/93PA03266 Zhang, Y. G., Pagani, M., Liu, Z., Bohaty, S. M., & DeConto, R. (2013). A 40-million-year history of atmospheric CO2. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 371(2001), 20130096. https://doi.org/10.1098/rsta.2013.0096 Zhang, Y. G., Pagani, M., & Wang, Z. (2016). Ring Index: A new strategy to evaluate the integrity of TEX86 paleothermometry. Paleoceanography, 31(2), 220–232. https://doi.org/10.1002/2015PA002848 210 Zhang, Y. G., Zhang, C. L., Liu, X.-L., Li, L., Hinrichs, K.-U., & Noakes, J. E. (2011). Methane Index: A tetraether archaeal lipid biomarker indicator for detecting the instability of marine gas hydrates. Earth and Planetary Science Letters, 307(3), 525–534. https://doi.org/10.1016/j.epsl.2011.05.031 211 Appendix A Supplementary Information A.1 Supplementary Information for Chapter 3 A.1 Oblique view of the 3D landscape showing the catchment from the Gamburtsev Mountains delivering sediments to Prydz Bay. Basemap is the reconstructed topography and maximum bathymetry at 34 Ma (Hochmuth et al., 2020b). White circles indicate the Prydz Bay continental shelf locations of ODP Sites 739C (67°16.57'S, 75°04.91'E, 412 m), 742A (67°32.98'S, 75°24.27'E,416 m), 1166A (67°41.77´S, 74°47.22´E, 475 m) studied in chapter 3. 212 A.2 Lithostratigraphic summary for Ocean Drilling Program (ODP) Holes 739C, 742A, and 1166A and stacked area charts of particle size distribution and clay mineral abundance. Core recovery is indicated with black bars next to the depth scale. Cores were logged according to a common lithostratigraphic scheme. Lithological codes C, St, Sd, and G refer to clay, silt, sand, and gravel, respectively. Percentages of sand, silt, and clay were calculated from laser particle 213 size analysis. The clay mineralogy for ODP Sites 739 and 742 was taken from Hambrey et al. (1991). The clay mineral abundances for ODP Site 1166 were taken from Forsberg et al. (2008) and normalized to include only smectite (Sm), kaolinite (Kaol), illite, and chlorite (Chl). R.— Reticulofenestra; M.—Malvinia; H.—Hemiaulus; D.—Distephanosira. FO—First Occurrence; FCO—First Common Occurrence; LO—Last Occurrence; LCO—Last Common Occurrence. (Figure reproduced with permission from GSA; from Passchier et al., 2017). Purple line added to highlight the diamict to diatomite depositional transition in Prydz Bay at the EOT. Figure reproduced from Passchier et al., 2017, under fair use permission by GSA. Purple line added to highlight the diamict to diatomite depositional transition in Prydz Bay at the EOT. 214 A.3 Sedimentary sequence as in Fig. A.2, showing the new multiproxy biomarker and microfossil data from this study. Pollen concentration (counts/gram of dry sediment), differentiating reworked and penecontemporary pollen, and coal concentration in one sample (star). Plant wax n-alkanoic acid and n-alkane concentrations, as well as the δ13Cwax, δDwax, both showing 1σ uncertainties (error bars). Membrane lipid data showing the brGDGT and isoGDGT concentrations, BIT index, MAAT from MBT´5Me using BayMBT, and SSTs from 215 TEX86 using BAYSPAR. For comparison MAAT from the weathering proxy the S-index (Passchier et al., 2017) is also shown. All temperature estimates are shown with the 1σ error uncertainty as shading. We show MAP from the CIA-K index (Passchier et al., 2017), and soil pH from the CBT' index. We also show the downcore C31 hopane ββ/(αβ+βα+ββ) index. Purple line highlights the diamict to diatomite depositional transition in Prydz Bay at the EOT. 216 217 A.4 Example chromatograms from Prydz Bay sediments, showing: a) a FAME fraction containing abundant n-alkanoic acids; b) the alkane fraction from the same sample (739C-36R, 281.74 mcd) containing low abundances of n-alkanes representative of most of the record; and c) 191 m/z single ion chromatogram from the same sample (739C-36R, 281.74 mcd) with labeled hopanes, asterisks correspond to C30 hopanes d) alkane fraction from an anomalous sample (742A-20R, 484.05 mcd), with the greatest abundance of short and mid chain alkanes, along with many other compounds, including hopanes, whereas corresponding FAME and GDGT fractions were below detection levels. That sample was also found to be unusual via microscopy, with coal or charcoal fragments and little to no pollen. Regardless of concentration, almost all samples had n-alkanes dominated by short and mid chain lengths with a CPI~1 and a prominent uncharacterized complex mixture (UCM), likely from erosion of ancient sedimentary rocks, consistent with pollen evidence for a substantial reworked pollen component. In all samples, the n-alkanoic acids have molecular abundance distributions consistent with penecontemporary inputs without substantial alteration, and a distribution reflective of microbial and plant wax sources. Labels annotate the chain length for the homologous series (triangle symbols for n-alkyl lipids). 218 A.5 n-Alkanoic acid chain length distribution averaged across all samples with δ 13 C by chain length. Error bars (1σ). Plant wax n-alkanoic acid abundances and d13C values for the homologous series show expected patterns for terrestrial plants, with a trend toward 13 C- depletion for the longer chain lengths. Shorter chain lengths show 13C-enrichment that may indicate microbial sources (Chen et al., 2019). 219 A.6 n-Alkanoic acid δ 13 C and δD values downcore. Error bars (1σ). Downcore variations in the δ 13 C and δD of each of the even chain compounds in the homologous series are shown for reference. We use the C30 n-alkanoic acid for downcore reconstructions in the main text on the basis that it is most likely to reflect terrestrial vegetation. 220 A.2 Supplementary Information for Chapter 4 Figure A.7 Present day Antarctic bed elevation from bedmap2 (Fretwell et al., 2013) with the Eocene (34 Ma) coastline (Hochmuth et al., 2020) outlined in black. This shows the modest longitudinal rotation of the peninsula since the Eocene. 221 Figure A.8 Data from SHALDRIL II sedimentary summary showing facies and pebble count as well as the proportion of Nothofagus fusca pollen (Anderson et al., 2011) with new presentation of the proportion of reworked pollen, and new biomarker data including the hopane index, n- alkane CPI, BIT index, the MBT´5Me-based temperature reconstructions using two available soils calibrations to estimate MAAT (Dearing Crampton Flood et al., 2020) and estimate MAF (Martinez Sosa et al., 2021), as well as the TEX86-based BAYSPAR estimates of SST available for a subset of the samples. We also show the br- and isoGDGT concentrations (note low abundance of brGDGTs); the concentrations, CPI and ACL for both the n-alkanoic acid and n- alkane compound classes. Note the declining alkane CPI, declining hopane ββ/(αβ+βα+ββ) index, and increasing reworked pollen all tell a coherent story of more reworking up section in parallel to cooling temperatures. Concentration data for the n-alkanoic acid and n-alkanes from the Eocene section is limited due to missing data on sample masses extracted a decade ago (Feakins et al., 2014). Please note the middle Miocene datapoints after 30 mbsf are offset from the stratigraphic column due to core expansion post-recovery and should correspond to the depths below seafloor of the last pebble counts at 30 m. 222 Figure A.9 a) Dendrogram of n-alkanoic acid chain length distributions with clusters labeled and showing b) molecular abundance distribution, showing mean (bars) and 1 standard deviation (error bars) for each cluster. 223 A.3 Supplementary Information for Chapter 5 Figure A.10 Southern hemisphere high latitude surface air temperatures for the Eocene (4x pCO2 model runs), Oligocene (2x pCO2), and the difference across the transition (2x-4x) for the various climate models and the ensemble mean. The circles correspond to proxy mean annual air temperature records while the dotted areas show the source area used to compare the model temperature to the proxy record. 224 Figure A.11 Modelled Southern Ocean SSTs for the Eocene (4x pCO2 model runs), Oligocene (2x pCO2), and the difference across the transition (2x-4x). The colored circles show proxy SSTs estimates, when these marine core sites appear to plot “on land”, this is due to imprecision in modelled coastlines, and a black circle denotes the nearest marine location in the model used for proxy-model comparison. 225 Figure A.12 Climate model reconstructions for the Eocene (without ice), Oligocene (with ice) and Oligocene-Eocene showing the modelled surface air temperature (SAT) change associated with EOT glaciation. Circles show proxy evidence for SAT for comparison. 226 Figure A.13 Climate model reconstructions for the Eocene (without ice), Oligocene (with ice) and Oligocene-Eocene showing the modelled sea surface temperature (SST) change associated with EOT glaciation. Circles show proxy evidence for SSTs for comparison. 227 Figure A.14 Climate model SAT reconstructions of Eocene, Oligocene and Oligocene-Eocene difference associated with changes in geography. CESM_H contrasts Tasman Gateway closed in the Eocene and open in the Oligocene. FOAM compares the geography of the West Antarctic with the continent above sea level (Eocene) and below sea level (Oligocene). The UVIC experiments contrast a Drake Passage that is closed during the Eocene and open during the Oligocene. Circles show surface air temperature values reconstructed from proxies. 228 Figure A.15 SSTs comparison between Eocene and Oligocene climate model scenarios with different paleogeographies, and proxy-model comparison. Experiments, as in Fig. A.14. 229 Figure A.16 Proxy-model comparison for models scaled to a 25% reduction in pCO2 for Oligocene-Eocene scenarios showing a) SAT b) SST. The colored circles show proxy SSTs estimates, when these marine core sites appear to plot “on land”, this is due to imprecision in modelled coastlines, and a black circle denotes the nearest marine location in the model used for proxy-model comparison.

Abstract (if available)

Abstract The Cenozoic can be divided into a Greenhouse (Paleocene and Eocene) and an Icehouse world (Oligocene to present). This division is based on the presence of permanent ice sheets in Antarctica; however, the glacial history of Antarctica is more complex with periods of ice sheet growth and retreat of the East, West, and Antarctic Peninsula ice sheets. On the fully glaciated continent today, there are few records of past climate accessible, however the marine margins provide evidence for conditions on land. With the development of biomarker methodologies, these marginal sediments from around the Antarctic continent can now yield new proxy evidence for the fluctuating climate history of Antarctica. This thesis revisits legacy cores drilled over recent decades and finds new evidence for Antarctic climate across the Cenozoic. The first record, Chapter 2, captures a snapshot of the Paleocene/Eocene offshore of the Sabrina Coast, East Antarctica, where few records exist. Application of plant wax dual isotopes (δD and δ13C) in combination with pollen analysis suggest that the Sabrina Coast region consisted of an open canopy woodland or shrubby tundra with δD of precipitation similar to today. To evaluate how the terrestrial climate shifted from the Greenhouse to the Icehouse world, Chapter 3 generates a record spanning the Eocene-Oligocene Transition (EOT) from Prydz Bay, at the outflow of a major drainage basin for the East Antarctic ice sheet. Soil biomarkers revealed a 5°C cooling on land prior to the EOT and ocean paleothermometers revealed a decrease of 4°C at the EOT with additional changes in marine productivity. In addition, we found plant wax n-alkanoic acid whose δD and δ13C values identify increasing aridity across this transition. Whereas plant wax n-alkanoic acids are penecontemporary, other biomarkers show inputs of older material reworked by glacial erosion (n-alkanes and hopanes). To compare timeslices of the Cenozoic, Chapter 4, generates a biomarker record offshore of the Antarctic Peninsula to capture evidence for terrestrial and marine conditions for the late Eocene, late Oligocene, mid-Miocene, and Pliocene. This record captures declining vegetation with increasing glacial erosional inputs of older strata identified by hopanes. Chapter 5 compiles proxy temperature records from around the Antarctic for the late Eocene and Early Oligocene including those generated in this thesis. These records are compared to surface air and sea surface temperatures generated from previous model simulations across the EOT. As EOT model experiments were forced with larger than expected pCO2 decreases, the model experiments were scaled to a more realistic pCO2 drop, assessed by fitting to the temperature decreases found in the proxies. The temperature-scaled pCO2 drop used for the revised EOT model scenario, now better fits pCO2 proxy reconstructions. In summary, this thesis adds new proxy data for the Antarctic, reinforces the value of multi-biomarker and multi-proxy comparisons, and demonstrates how screening for glacial reworking allows us to reconstruct past climate in polar regions across glacial transitions. The added records help to provide a proxy comparison to climate model simulations and thus to test the mechanisms needed to drive the glaciation of Antarctica.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

From deep lakes to deserts: Plio-Pleistocene paleoenvironmental transitions

$Isotopic fractionations in plant biomarker molecules with application to paleoclimate$

PDF

Isotopic fractionations in plant biomarker molecules with application to paleoclimate

PDF

Flowstone ideograms: deciphering the climate messages of Asian speleothems

PDF

From tree tops to river runoff: tracing plant wax biomarkers across the Peruvian Andes and Amazon

PDF

Taking the temperature of the Common Era: statistics, patterns and dynamical insights

PDF

Evolution of the Indian Monsoon and rise of C₄ photosynthesis in the Miocene and Pliocene

PDF

Late Pleistocene changes in winter moisture source in the coastal southwest United States

PDF

Seeing the future through the lens of the past: fusing paleoclimate observations and models

PDF

Green River formation stromatolites as a paleoclimate indicator: an investigation of the early Eocene climatic optimum through mass spectrometry, micro-X-ray fluorescence spectroscopy, and petrography

PDF

Stromatolites as biosignatures and paleoenvironmental records: experiments with modern mats and examples from the Eocene Green River Formation

PDF

Recent variability in the hyrdological cycle of tropical Asia from oxygen isotopes of tree celulose

PDF

Abrupt climate variability during the last glacial and the late Holocene from the western tropical Pacific perspective

PDF

Exploring seasonal climate variability in East African trees: High-resolution oxygen and carbon isotope records of cellulose of Vitex payos from Chyulu Hills, Kenya

$Isotopic insights to human-impacted hydrologic systems: from urban watersheds to hydraulically fractured basins$

PDF

Isotopic insights to human-impacted hydrologic systems: from urban watersheds to hydraulically fractured basins

PDF

The geobiology of fluvial, lacustrine, and marginal marine carbonate microbialites (Pleistocene, Miocene, and Late Triassic) and their environmental significance

Asset Metadata

Creator Tibbett, Emily Jane (author)

Core Title Antarctic climate variability from greenhouse to icehouse world

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Geological Sciences

Degree Conferral Date 2022-05

Publication Date 04/16/2022

Defense Date 02/23/2022

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Antarctica,climate models,Eocene,GDGTs,OAI-PMH Harvest,Oligocene,organic geochemistry,paleoclimate,plant wax,pollen,proxies

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Feakins, Sarah J. (committee chair), Corsetti, Frank (committee member), Seiffert, Erik (committee member)

Creator Email ejtibbett@gmail.com,tibbett@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC110964848

Unique identifier UC110964848

Document Type Dissertation

Format application/pdf (imt)

Rights Tibbett, Emily Jane

Type texts

Source 20220417-usctheses-batch-927 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu