Title Page High latitude controls on the evolution of eastern equatorial Pacific oceanography and climate since the last glacial period By Samantha C. Bova A. B., Washington University in St. Louis Department of Earth and Planetary Sciences, 2011 M.Sc., Brown University Department of Earth, Environmental and Planetary Sciences, 2013 A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Department of Earth, Environmental and Planetary Sciences at Brown University Providence, Rhode Island May 2017 Copyright Page © Copyright by Samantha Claudia Bova Signature Page This dissertation by Samantha C. Bova is accepted in its present form by the Department of Earth, Environmental and Planetary Sciences as satisfying the dissertation requirement for the degree of Doctor of Philosophy. Date_______________ __________________________________________ Timothy D. Herbert, Advisor Recommended to the Graduate Council Date_______________ __________________________________________ Baylor Fox-Kemper, Reader Date_______________ __________________________________________ Steven Clemens, Reader Date_______________ __________________________________________ Karen Fischer, Reader Date_______________ __________________________________________ Jean Lynch-Stieglitz, Reader Approved by the Graduate Council Date_______________ __________________________________________ Andrew G. Campbell, Dean of the Graduate School iii Curriculum Vitae Samantha C. Bova Brown University, Department of Earth, Environmental and Planetary Sciences 324 Brook Street, Providence, RI 02912, samantha_bova@brown.edu Education Ph.D Candidate, Brown University, Enrolled Sept. 2011 Department of Earth, Environmental and Planetary Sciences Research Advisor: Dr. Timothy D. Herbert Dissertation: High latitude controls on the evolution of eastern equatorial Pacific oceanography and climate since the last glacial period M.S., Brown University, May 2013 Department of Geological Sciences Research Advisor: Dr. Timothy D. Herbert Thesis Title: High latitude controls on eastern equatorial Pacific intermediate water circulation during the last glacial termination B.A., Washington University in St. Louis, May 2011 Graduated with Honors: Magna Cum Laude Major: Earth and Planetary Sciences Minor: Environmental Studies Thesis title: Paleoclimatic significance of stable isotope, minor element, and trace element variations in spring-derived tufa deposits from Belize, Croatia, and Egypt Publications Bova, S.C., Herbert, T., Rosenthal, Y., Altabet, M., Kalansky, J., Chazen, C. Mojarro, A., and Zech, J (2015). Links between Eastern Equatorial Pacific Stratification and Atmospheric CO2 Rise during the last Deglaciation, Paleoceanography, 30(11), 1407- 1424, doi:10.1002/2015PA002816. iv Kalansky, J., Rosenthal, Y., Herbert, T., Altabet, M., Bova, S. C. (2015). Southern Ocean contributions to the Eastern Equatorial Pacific Heat content during the Holocene, Earth and Planetary Science Letters, 424, 158-167, doi:10.1016/j.epsl.2015.05.013. Lyle, M., Pockalny, R., Polissar, P., Lynch-Steiglitz, J., Bova, S.C., Dunlea, A., Ford, H., Hertzberg, J., Hovan, S., Jacobel, A., King Wertman, C., Maloney, A., Murray, R., Shackford Wilson, J., Wejnert, K., Xie, R., (2016). Dynamic carbonate sedimentation on the Northern Line Islands Ridge, Palmyra Basin, Marine Geology, 379, 194-207, doi:10.1016/j.margeo.2016.06.005. Lynch-Stieglitz, J., Polissar, P.J., Jacobel, A., Hovan, S. A., Pockalny, R., Lyle, M., Murray, R., Ravelo, A.C., Bova, S.C., Dunlea, A., Ford, H., Hertzbery, J., Wertman, C., Maloney, A., Shackford, J., Wejnerty, K., Xie, R., (2015). Glacial-interglacial changes in central tropical Pacific surface seawater property gradients, Paleoceanography, 30(5), 423-438, doi:10.1002/2014PA002746. Grants and Fellowships Dissertation Fellowship, Brown University, Spring 2016 Graduate Student Research Grant, Evolving Earth Foundation, May 2015 Graduate Student Fellowship, Institute at Brown for the Study of Environment and Society, Sept. 2014-Aug. 2015 Graduate Student Research Grant, Geological Society of America, May 2013 Professional Presentations: Invited Talk, Alfred Wegener Institute, S. Bova and T. D. Herbert. Untangling the mixed influences of high- and low-latitude climate forcing in the Eastern Equatorial Pacific, May 11, 2016. v AGU 2015 (poster) S. Bova, et al., (2015). Holocene Deep Ocean Variability Detected with Individual Benthic Foraminifera, Dec. 17, 2015, Abstract #68473. Invited Talk, Lamont-Doherty Earth Observatory, Lunch Hour Seminar Series, S. Bova et al., Deep ocean variability detected with individual benthic foraminifera, Nov. 2, 2015. CLIVAR-ICTP Workshop on Past and Future Climate Shifts: Decadal Climate Variability and Predictability (poster) S. Bova, et al., (2015). Detecting Deep Ocean variability with individual benthic foraminifera, University of Wisconsin, November 16- 24, 2015. High-Resolution Proxies of Paleoclimate (talk) S. Bova, et al., Assessing the natural variability of the deep ocean using individual benthic foraminifera, June 2, 2015. AGU 2014 (talk) S. Bova, et al., (2014) In pursuit of the mystery reservoir: marine radiocarbon evidence from the eastern tropical Pacific for a deglacial CO2 source, Dec. 15, 2014, Abstract # PP13D-05. Graduate Climate Conference 2014 (talk & poster) S. Bova, et al. (2014) Radiocarbon evidence for old carbon in the eastern equatorial Pacific subsurface during the last glacial termination, Nov. 1, 2014. Invited Talk, Woods Hole Oceanographic Institute, Lunch Hour Seminar Series, S. Bova et al. (2014), Time-transgressive changes in intermediate waters in the eastern equatorial Pacific during Termination 1, April 10, 2014. AGU 2013 (talk) S. Bova, et al., (2013) Variability in Eastern Equatorial Pacific intermediate water circulation during the last glacial termination: the impact of high latitude climate on equatorial stratification, Dec. 9-13 2013, Abstract # PP31D-1900. AGU 2012 (poster) S. Bova, et al., (2012) Centennial-scale variability in Peru Margin intermediate water circulation during the last deglaciation, Dec. 3-7 2012. vi Teaching Experience Teaching Assistant, Brown University, Providence RI GEOL 1240 – Sedimentology and Stratigraphy, Fall 2015 GEOL 0240 – Earth: Evolution of a Habitable Environment, Spring 2014 GEOL 0220 – Physical Processes in Geology, Fall 2013 GEOL 0070 – Introduction to Oceanography, Spring 2013 Teaching Certificate 1, Sheridan Center, Brown U., Aug. 2012-May 2013 Mentoring and Advising Sathya Annisetti, High School Student Summer Intern, June 2015 – Dec. 2015 Josh Zimmt, Undergraduate Summer Intern, Summer 2014 Katherine Hadley, Environmental Studies Undergraduate Senior Thesis, Spring 2013 Field Work R/V JOIDES Resolution Expedition 363 to the Western Pacific Warm Pool (sedimentologist), Oct. 6 – Dec. 8 2016 R/V Roger Revelle, Western Pacific Warm Pool Survey and Coring Cruise, R/V, Sept. 5 – Oct. 2 2013 R/V Marcus G. Langseth, Line Islands Survey and Coring Cruise, R/V Marcus G. Langseth, April 30-May 26 2012 Geology Field Camp, New Zealand, Spring 2010 vii Summer Training Courses Advanced Climate Dynamics Course, The Last Glacial Termination, University of Bergen, Nyksund, Norway, August 18-31 2013 Coccolith Identification and Sample Preparation Training, University of Oviedo with Dr. Clara Bolton, Oviedo, Spain, June 1-30 2013 Department and Community Service Leader (Fall 2014-Spring 2016) of the Vartan-Gregorian Elementary School Volunteer Science Teacher Program, Volunteer since Fall 2011. Narragansett Bay Dissolved Oxygen Survey Volunteer, Save the Bay, Summer Volunteer since July 2013. Geology Club President, Sept. 2012-May 2013 Graduate Student Council Member, Geology Representative, Sept. 2013-May 2014 Honors and Awards Elected to Sigma Xi: The Scientific Research Society, April 2014 viii PREFACE The eastern equatorial Pacific (EEP) amplifies changes to the global climate system through heat and carbon exchange with the atmosphere [Hastenrath, 1980; Takahashi et al., 2002; Philander and Fedorov, 2003; Boccaletti et al., 2004]. The physical oceanography and patterns of wind forcing across the region give rise to a uniquely strong link between cool, carbon-rich subsurface waters and warm, tropical surface waters [e.g. Fiedler and Talley, 2006]. Subsurface waters upwell to the surface ocean along the equator and the west coast of South America creating a tongue of cool water that stretches westward across the Pacific. The result is a net transfer of heat from the atmosphere into the ocean, elevated rates of primary production, and the release of CO2. The EEP yields 15-50% of global open ocean productivity [Chavez and Barber, 1987] and releases nearly two-thirds of the marine carbon dioxide flux to the atmosphere each year [Takahashi et al., 2002]. Upwelling strength and mixing between EEP subsurface and surface waters is not constant thru time. At present, temporal variability is dominated by the El Niño Southern Oscillation (ENSO), a tropical air-sea phenomenon that alters the heat distribution across the equator and the strength of upwelling in the EEP on a 2 to 7 year timescale [e.g. Wang and Fiedler, 2006]. The effects of ENSO are felt most strongly in the EEP, giving rise to changes in phytoplankton productivity, sea surface temperature and precipitation patterns. However, ENSO events also lead to global scale atmospheric reorganizations that affect weather patterns across the globe. In the United States, one may recall the 2015 strong El Niño event that provided a much-needed reprieve to drought-stricken California. ix On decadal and longer timescales we must rely on paleoceanographic records and models to reconstruct the dynamics of the EEP because the historical record of ocean observations is short (<135 years) and sparse [Hobbs and Willis, 2013]. Thus far, proxy and model evidence indicate that remotely forced mechanisms in the southern and northern high-latitudes may overwhelm locally driven variability through changes in the transport and physio-chemical properties of subsurface intermediate waters. These water masses form at the poleward boundaries of the subtropical gyres [Liu and Philander, 1995; Gu and Philander, 1997; Lu et al., 1998] and at even higher latitudes [Toggweiler et al., 1991; Andreasen et al., 2001]. They transport heat, salt, and nutrients from their formation region into the EEP subsurface where they drive oceanographic change from the bottom-up. This subsurface variability may then feedback to the surface ENSO-cycle in intriguing ways. My dissertation work investigates the relative role of high- and low-latitude forcings in shaping EEP oceanography and climate since the last glacial period. The transition from the Last Glacial Maximum (LGM) to the current interglacial is a natural experiment in rapid, global warming. Global temperatures rose by approximately 3.5°C between the LGM (21 ka) and the early Holocene (6-10 ka) [Denton et al., 2010; Shakun et al., 2012], a rise comparable to that projected for the coming century [IPCC, 2014]. The first three chapters reconstruct change within EEP subsurface intermediate waters, while the last two focus on the surface ocean response. The most significant finding of this dissertation is that the EEP is a passive responder to subsurface change driven from the bottom-up by the high southern and northern latitudes on millennial and longer timescales. The work therefore refutes ideas that locally forced ocean-atmosphere x dynamics in the EEP drove global climate shifts over the past 30 kyrs [e.g. Clement et al., 1999]. Chapter one examines the evolution of subsurface intermediate waters in the EEP upper water column since the last glacial period. Sediment cores from three intermediate depths in the EEP are used to reconstruct change in stratification over the top 1000 m of the water column. δ18O signatures measured on the carbonate shells of benthic foraminifera, zooplankton that live at the sediment-water interface, provide a combined measure of water temperature and salinity and therefore density at the depth of core recovery. Water density change arose first at depth, decreasing in phase with rising temperatures over Antarctica [Jouzel et al., 2007], and propagated up to the surface over the course of deglaciation. Early warming in the subsurface drove a reduction in stratification that increased mixing over the top 1000 m of the water column and CO2 degassing during the last deglaciation. These findings are significant because they suggest stratification and mixing between intermediate and surface waters in the EEP is regulated by climate in the southern high latitudes. Chapter two also focuses on the role of the subsurface in driving change in the EEP but on much shorter timescales (<200 years). This work was motivated by recent observations that expose unprecedented rates of warming in the deep ocean that track observed surface warming [Balmaseda et al., 2013]. We test a new tool, the δ18O signature of individual benthic foraminifera, to evaluate the natural variability of the deep ocean. A first use of the proxy at 1000 m water depth in the EEP reveals temperature variability of more than 2°C within 200 years, significantly larger than the modern range. xi Rates of modern deep water variability in the EEP are therefore not unprecedented in the Holocene. Chapter three investigates the role of Pacific deep and intermediate waters in glacial-interglacial atmospheric CO2 variations. The leading hypothesis suggests carbon is stored in the deep Southern Ocean during glacial periods and is released back to the atmosphere via Southern Ocean Intermediate Waters (SOIW) [Toggweiler et al., 2006; Marchitto et al., 2007; Anderson et al., 2009]. Initial tests of the hypothesis at intermediate depths in the Pacific were affirmative but recent proxy records and model runs have cast doubts [Hain et al., 2011; De Pol-Holz et al., 2010; Rose and Sikes, 2010]. In chapter three I propose a new mechanism for carbon release during the last deglaciation that reconciles all but one [Stott et al., 2009] previously published radiocarbon record under a single unified theory. The relative impact of remotely forced subsurface change versus local air-sea dynamics on surface ocean processes in the EEP is explored in Chapters four and five. We reconstruct sea surface temperature, thermocline depth, and primary productivity at four sites spread across the EEP. In Chapter four the relationships between these parameters are evaluated, for the first time in the paleo-record, to determine whether the mean state of the EEP can be regarded as “scaled-up” El Niño or La Niña, or a state similar to neither. Correlations between parameters indicate that ENSO was not the dominant source of variability in surface ocean properties on millennial timescales over the past 30 kyrs. Chapter 5 focuses on the response of phytoplankton in the EEP to global change over the past 30 kyrs. Productivity records from previous studies exhibit a regionally xii heterogeneous and at times conflicting response to global climate change since the Last Glacial Maximum, which makes it difficult to assess the role of the biological pump in atmospheric CO2 change. Carbon balance in the EEP is determined by the export ratio of organic carbon to calcium carbonate, which can be estimated as the relative abundance of diatoms (non-calcareous) versus coccolithophores (calcareous) [e.g. Kohfield et al., 2005]. Here, the productivity response of diatoms and coccolithophores is reconstructed at multiple sites in the EEP to show that diverging production histories are not caused by differing responses of diatoms and coccolithophores but reflect true spatial heterogeneity related to the proximity of local iron sources. Thus, the EEP biological pump was not an important carbon sink during the last glacial period and cannot be relied upon to mitigate anthropogenic climate change in the future. The work presented in this dissertation demonstrates a primary role for high- latitude climate in shaping the EEP since the last glacial period. Its fingerprint is observed at intermediate depths and the surface ocean using a variety of paleoclimate proxies. This finding is important for predicting the response of the EEP to future climate change. Based on the results presented in this disseration, I expect warm subsurface anomalies from the high latitudes will drive a reduction in upper ocean stratification in the EEP and a mean deepening of the equatorial Pacific thermocline. These changes in upper ocean struction will subsequently lead to weakened upwelling of cool subsurface waters and reduced heat uptake in the EEP over the coming centuries. xiii References Anderson, R. F., S. Ali, L. I. Bradtmiller, S. H. H. Nielson, M. Q. Fleisher, B. E. Anderson, and L. H. Burckle (2009), Wind-Driven Upwelling in the Southern Ocean and the Deglacial Rise in Atmospheric CO2, Science, 323, 1443-1448. Andreasen, D. H., A. C. Ravelo, and A. J. Broccoli (2001), Remote forcing at the last glacial maximum in the tropical Pacific Ocean, J Geophys Res-Oceans, 106(C1), 879- 897. Balmaseda, M. A., K. E. Trenberth, and E. Kallen (2013), Distinctive climate signals in reanalysis of global ocean heat content, Geophysical Research Letters, 40(9), 1754- 1759. Boccaletti, G., R. C. Pacanowski, S. G. H. Philander, and A. V. Fedorov (2004), The thermal structure of the upper ocean, J Phys Oceanogr, 34(4), 888-902. Chavez, F. P., and R. T. Barber (1987), An Estimate of New Production in the Equatorial Pacific, Deep-Sea Res, 34(7), 1229-1243. De Pol-Holz, R., L. Keigwin, J. Southon, D. Hebbeln, and M. Mohtadi (2010), No signature of abyssal carbon in intermediate waters off Chile during deglaciation, Nat Geosci, 3(3), 192-195. Denton, G. H., R. F. Anderson, J. R. Toggweiler, R. L. Edwards, J. M. Schaefer, and A. E. Putnam (2010), The Last Glacial Termination, Science, 328(1652), 1652-1656. Fiedler, P. C., and L. D. Talley (2006), Hydrography of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 143-180. the North Pacific abyss during the last deglaciation, Nature, 449(7164), 890-U899. Gu, D., and G. Philander (1997), Interdecadal Climate Fluctuations That Depend on xiv Exchanges Between the Tropics and Extratropics, Science, 275(5301), 805-807. Hain, M. P., Sigman, D. M., and G. H. Haug (2011), Shortcomings of the isolated abyssal reservoir model for deglacial radiocarbon changes in the mid-depth Indo-Pacific Ocean, Geophysical Research Letters, 38(4), L04605, doi: 10.1029/2010GL046158. Hastenrath, S. (1980), Heat-Budget of Tropical Ocean and Atmosphere, J Phys Oceanogr, 10(2), 159-170. Hobbs, W. R., and J. K. Willis (2013), Detection of an observed 135 year ocean temperature change from limited data, Geophysical Research Letters, 40(10), 2252- 2258. Jouzel, J., et al. (2007), Orbital and millennial Antarctic climate variability over the past 800,000 years, Science, 317(5839), 793-796. Kohfeld, K. E., C. Le Quere, S. P. Harrison, and R. F. Anderson (2005), Role of marine biology in glacial-interglacial CO2 cycles, Science, 308(5718), 74-78. Liu, Z. Y., and S. G. H. Philander (1995), How Different Wind Stress Patterns Affect the Tropical Subtropical Circulations of the Upper Ocean, J Phys Oceanogr, 25(4), 449- 462. Lu, P., J. P. McCreary, and B. A. Klinger (1998), Meridional circulation cells and the source waters of the Pacific Equatorial Undercurrent, J Phys Oceanogr, 28(1), 62-84. Marchitto, T. M., S. J. Lehman, J. D. Ortiz, J. Fluckiger, and A. van Geen (2007), Marine radiocarbon evidence for the mechanism of deglacial atmospheric CO2 rise, Science, 316(5830), 1456-1459. Philander, S. G., and A. V. Fedorov (2003), Role of tropics in changing the response to Milankovich forcing some three million years ago, Paleoceanography, 18(2). xv Rose, K. A., E. L. Sikes, T. P. Guilderson, P. Shane, T. M. Hill, R. Zahn, and H. J. Spero (2010), Upper-ocean-to-atmosphere radiocarbon offsets imply fast deglacial carbon dioxide release, Nature, 466(7310), 1093-1097. Shakun, J. D., P. U. Clark, F. He, S. A. Marcott, A. C. Mix, Z. Y. Liu, B. Otto-Bliesner, A. Schmittner, and E. Bard (2012), Global warming preceded by increasing carbon dioxide concentrations during the last deglaciation, Nature, 484(7392), 49-+. Stott, L., J. Southon, A. Timmermann, and A. Koutavas (2009), Radiocarbon age anomaly at intermediate water depth in the Pacific Ocean during the last deglaciation, Paleoceanography, 24(2). Takahashi, T., et al. (2002), Global sea-air CO2 flux based on climatological surface ocean pCO2, and seasonal biological and temperature effects, Deep-Sea Res Pt Ii, 49(9- 10), 1601-1622. Toggweiler, J. R., K. Dixon, and W. S. Broecker (1991), The Peru Upwelling and the Ventilation of the South Pacific Thermocline, Journal of Geophysical Research, 96(C11), 467-497. Toggweiler, J. R., J. L. Russell, and S. R. Carson (2006), Midlatitude westerlies, atmospheric CO2, and climate change during the ice ages, Paleoceanography, 21(2). Wang, C., and P. C. Fiedler (2006), ENSO variability and the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 239-266. xvi ACKNOWLEDGMENTS I would first like to thank my advisor, Dr. Timothy Herbert. I am grateful for the freedom you gave me to tackle problems that interest me and devise my own path to discovery, with, of course, many helpful nudges along the way. I am inspired by your ability to see the big picture and communicate why the science we do matters. I am also thankful for the helpful insight and instruction provided by other members of my committee. Dr. Baylor Fox-Kemper proved a critical resource when the physics and math of oceanographic problems exceeded the bounds of my understanding. Thank you to Dr. Steven Clemens for helping me push the limits of carbonate mass spectrometry. Lastly, many thanks go to Dr. Jim Russell and Dr. Karen Fischer for pushing me to define my niche in the scientific community. My graduate student colleagues at Brown have also contributed a great deal to my education over the past five years. I am grateful for the many enlivened discussions on scientific, and at times not so scientific topics, over lunch or a beer. I am particularly grateful for the insight and genuine interest in my work that often spiraled into friendly arguments with my officemate William Daniels. Discussions and edits from you have improved every chapter in this dissertation. I am also indebted to members of the TDH lab group. You taught me everything I know about running an organic geochemistry lab. Thanks especially to Alexa Tzanova for patiently passing on her troubleshooting know- how and taking one for the team by engaging with Thermo. Thank you also to April Martin (aka. our resident foram guru) for introducing me to the wonderful world of foraminifera. xvii I would also like to acknowledge the amazing group of friends who have seen me through the last five years. Both professionally and personally there have been many ups and downs over the course of my graduate tenure and Stephanie Spera has been there for all of them. Thank you for taking care of me in moments of need and for helping me celebrate moments of triumph (also you can stop worrying about the bananas ). Thank you to William Daniels for challenging me intellectually, while still making sure I made time to play outside. I will sorely miss your company. Additional thanks go to Chelsea Parker, William Longo, Rachel Lupien, and many others. Lastly, I would like to thank my family. My parents have always supported my goals and ambitions. You have instilled in me a sense that anything is achievable. Thank you for always being there for me. My sisters, Laura and Alex, have also been critical to my success and sanity. Thank your for entertaining me with stories about pastries and brains on my many walks home from the office. Completion of this dissertation is exhilarating, while simultaneously unsettling. It is exhilarating because it marks the end of many years of hard work and the beginning of a new adventure. I look forward to taking the skills I have acquired here to face new challenges and new scientific problems. Yet I am also apprehensive because it is hard to imagine a better place to do science or a better community of scholars and friends. Thank you to everyone at Brown for making my graduate career a truly wonderful experience. xviii TABLE OF CONTENTS TITLE PAGE .................................................................................................................................. I COPYRIGHT PAGE .................................................................................................................... II SIGNATURE PAGE .................................................................................................................... III CURRICULUM VITAE .............................................................................................................. IV PREFACE ..................................................................................................................................... IX ACKNOWLEDGMENTS ...................................................................................................... XVII LIST OF ABBREVIATIONS ................................................................................................ XXII LIST OF TABLES ................................................................................................................. XXIV LIST OF FIGURES ................................................................................................................. XXV CHAPTER ONE: LINKS BETWEEN EASTERN EQUATORIAL PACIFIC STRATIFICATION AND ATMOSPHERIC CO2 RISE DURING THE LAST DEGLACIATION .................................................. 1 ABSTRACT .................................................................................................................................... 2 1. INTRODUCTION ........................................................................................................................ 3 2. MODERN OCEANOGRAPHIC SETTING ...................................................................................... 5 3. SITE LOCATIONS ...................................................................................................................... 8 4. METHODS ................................................................................................................................. 8 4.1 Age Models ........................................................................................................................ 8 4.2 Alkenone SST estimates ................................................................................................... 10 4.3 Benthic Stable Isotope Records ....................................................................................... 11 4.4 Benthic Temperature Records ......................................................................................... 11 5. RESULTS ................................................................................................................................. 13 5.1 Stable Isotope Records .................................................................................................... 13 5.2 Temperature Records ...................................................................................................... 15 6. DISCUSSION ............................................................................................................................ 17 6.1 Depth dependent response to deglacial forcing in the EEP upper water column .......... 17 6.2 EEP intermediate water structure during HS1 and the YD ............................................ 22 7. SUMMARY AND CONCLUSIONS .............................................................................................. 29 REFERENCES .............................................................................................................................. 32 FIGURES ..................................................................................................................................... 48 xix SUPPLEMENTARY INFORMATION ............................................................................................... 58 CHAPTER TWO: RAPID VARIATIONS IN DEEP OCEAN TERMPERATURE NOT UNPRECEDENTED IN THE HOLOCENE ....................................................................................... 61 ABSTRACT .................................................................................................................................. 62 2. MATERIAL AND METHODS ..................................................................................................... 64 4. RESULTS ................................................................................................................................. 66 5. CAUSES OF FORAMINIFERAL δ18O VARIABILITY ................................................................... 67 6. SOURCES OF PAST AND PRESENT DEEP WATER VARIABILITY .............................................. 69 7. CONCLUSIONS ........................................................................................................................ 69 FIGURES ..................................................................................................................................... 79 SUPPLEMENTARY INFORMATION ............................................................................................... 86 CHAPTER THREE: RADIOCARBON EVIDENCE FOR STORAGE AND RELEASE OF THE GLACIAL CARBON RESERVOIR IN THE EQUATORIAL PACIFIC............................................. 100 ABSTRACT ................................................................................................................................ 101 1. INTRODUCTION .................................................................................................................... 101 2. RESULTS ............................................................................................................................... 104 2.1 Core Chronology and Benthic-Planktonic age offsets .................................................. 104 2.2 Benthic Foraminiferal Δ14C .......................................................................................... 106 3. DISCUSSION .......................................................................................................................... 107 4. METHODS ............................................................................................................................. 115 4.1 Area of study.................................................................................................................. 115 4.2 Radiocarbon Measurements .......................................................................................... 116 REFERENCES ............................................................................................................................ 117 FIGURES ................................................................................................................................... 126 SUPPLEMENTAL INFORMATION ............................................................................................... 139 CHAPTER FOUR: RELATIONSHIPS BETWEEN THERMOCLINE DEPTH, SEA SURFACE TEMPERATURE, AND PRODUCTIVITY IN THE EASTERN EQUATORIAL PACIFIC: NO ROLE FOR ENSO SINCE THE LAST GLACIAL PERIOD ........................................................................ 146 ABSTRACT ................................................................................................................................ 147 1. INTRODUCTION .................................................................................................................... 147 2. OCEANOGRAPHIC SETTING .................................................................................................. 150 2.1 Mean Hydrography ....................................................................................................... 150 2.2. Seasonal and Interannual Variability .......................................................................... 152 xx 3. SITE LOCATIONS .................................................................................................................. 153 4. METHODS ............................................................................................................................. 153 4.1 Age models .................................................................................................................... 153 4.2 Biomarkers .................................................................................................................... 155 4.3. Nannofossil Abundances .............................................................................................. 155 5. RESULTS ............................................................................................................................... 157 5.1 Alkenone Sea Surface Temperature Reconstructions.................................................... 157 5.2 Productivity Estimates................................................................................................... 158 5.3 Quantitative Thermocline Depth Calibration ............................................................... 159 5.4 Thermocline Depth Reconstructions ............................................................................. 160 6. DISCUSSION .......................................................................................................................... 161 6.1 No evidence for ENSO-like changes in the EEP mean state ......................................... 161 6.2.1 Extratropical controls on thermocline depth in the EEP ........................................... 163 7. CONCLUDING REMARKS ...................................................................................................... 165 REFERENCES ............................................................................................................................ 166 FIGURES ................................................................................................................................... 176 CHAPTER FIVE: NO ROLE FOR THE EASTERN EQUATORIAL PACIFIC BIOLOGICAL PUMP IN ATMOSPHERIC CO2 CHANGE SINCE THE LAST GLACIAL PERIOD .................................... 186 ABSTRACT ................................................................................................................................ 187 1. INTRODUCTION .................................................................................................................... 187 2. PRODUCTIVITY IN THE EEP ................................................................................................. 189 2.1. Spatial Gradients.......................................................................................................... 189 2.2 Temporal Variability ..................................................................................................... 191 3. SITE LOCATIONS .................................................................................................................. 191 4. METHODS ............................................................................................................................. 192 4.1. Age models ................................................................................................................... 192 4.2 Biomarkers .................................................................................................................... 193 5. RESULTS ............................................................................................................................... 195 5.1 Paleoproductivity estimates .......................................................................................... 195 6. DISCUSSION .......................................................................................................................... 196 6.1. Iron limitations in the open-ocean EEP and the efficiency of the biological carbon pump .................................................................................................................................... 196 6.3 Macro-nutrient limitations in the Peru and Galapagos Upwelling Systems ................ 199 6.2 Proxy Biases .................................................................................................................. 200 xxi CONCLUDING REMARKS .......................................................................................................... 202 REFERENCES ............................................................................................................................ 204 FIGURES ................................................................................................................................... 211 APPENDIX A: ORGANIC GEOCHEMICAL PROXY DATA ........................................... 223 APPENDIX B: CARBONATE STABLE ISOTOPE DATA ................................................. 256 APPENDIX C: RADIOCARBON DATA ................................................................................ 274 APPENDIX D: COCCOLITH ABUNDANCE DATA ........................................................... 275 xxii List of Abbreviations Currents Ecuador Peru Countercurrent EPCC Equatorial Undercurrent EUC North Equatorial Countercurrent NECC Peru Coastal Current PCC Peru-Chile Undercurrent PCUC South Equatorial Current SEC Oceanographic Zones Eastern Pacific Warm Pool EPWP Equatoiral Upwelling EU Equatorial Cold Tongue ECT Galapagos Upwelling GU Peru Coastal Upwelling PCU Water Masses Antarctic Intermediate Water AAIW Antarctic Surface Waters AASW Equatorial Pacific Intermediate Water EqPIW Equatorial Surface Water ESW North Pacific Deep Water NPDW North Pacific Intermediate Water NPIW Pacific Deep Water PDW South Pacific Deep Water SPDW Southern Ocean Intermediate Water SOIW Subantarctic Mode Water SAMW Subtropical Mode Water STMW Time Intervals Antarctic Cold Reversal ACR Bolling Allerod B/A Heinrich Stadial 1 HS1 Last Glacial Maximum LGM Younger Dryas YD Miscellaneous Eastern Equatorial Pacific EEP El Nino Southern Osciallation ENSO Gas Chromatograph Flame Ionization Detector GC-FID Oxygen Minimum Zone OMZ Sea Surface Temperature SST xxiii LIST OF TABLES 2.1 Age range and error for sampled intervals……………………………..….……...…85 2.2 Variance and range of foraminiferal δ18O distributions …………………….………85 4.1 Sediment core locations …………………….………………………………..……185 xxiv LIST OF FIGURES 1.1 Key formation regions of AAIW and SAMW and their paths to the EEP………..…48 1.2 Salinity of the EEP upper water column averaged along 87.5 °W…………………..49 1.3 Age models…………………………………………………………………………..50 1.4 Stable isotope records measured on Uvigerina Pergrina……………………………51 1.5 Comparison between temperature records from the EEP and high latitude records of global change during the last deglaciation……………………………………………….52 1.6 Stratification gradient anomalies in δ18O, δ13C, and temperature space …………….54 1.7 Comparison between strafication gradient anomalies and records of CO2…………..55 1.8 Schematic figure showing connections between the EEP and the Southern Ocean…57 1.S1 Statification gradient anomalies at 370 m, 600 m, and 1000 m in the EEP………..58 1.S2 Alkenone abundances and benthic foraminiferal carbon isotopes at site CDH 26...60 2.1 Modern observations of temperature and variability near the core sites…………….79 2.2 Benthic foraminiferal oxygen isotopes measured on individual U. peregrina……….80 2.3 δ18O distributions about the mean for each sampled interval………………………..81 2.4 X-radiograph images of core MC-18A and CDH-26………………………………..83 2.S1 Site map showing locations of the sediment cores, CDH-26 and MC-18A………..91 2.S2 Age model for core CDH-26 and MC-18A………………………………………...92 2.S3 Oxygen isotopes of a lab standard as a function of sample size……………………93 2.S4 Modern temperature and salinity observations before 2012 located between 95°W and 85°W and 10°S and 0°………………………………………………………………94 2.S5 Images of individual foraminifera………………………………………………….95 2.S6 Benthic foraminiferal carbon isotopes measured on individual U. peregrina….......97 xxv 2.S7 δ13C distributions about the mean for each sampled interval………………………98 3.1 Map of published intermediate and deep water radiocarbon records………………126 3.2 Modern Pacific Δ14C concentrations along 90°W …………………………………128 3.3 Paired planktonic and benthic foraminiferal 14C dates………………………...…...129 3.4 Radiocarbon data from 373, 600, and 1026 m water depth in the EEP…………….130 3.5 Δ14C evolution over the past 30 kyrs by water mass…………………….…………131 3.6 Ventilation records of north and south Pacific deep water and the rise of atmospheric CO2..................................................................................................................................133 3.7 Schematic representation of the major phases of storage and release of the glacial carbon reservoir…………………………………………………………………..…….135 3.8 Records of EEP stratification and CO2 outgassing during the deglaciation………..137 3.S1 Core chronologies…………………………………………………………..……..142 3.S2 Comparison of Galapagos Platform Δ14C records……………………….…….….143 4.1 Sediment core locations overlaid on a map of EEP SSTs……………….…………176 4.2 Age-depth models……………………………………………………………..……177 4.3 Alkenone SST reconstructions……………………………………………………..178 4.4 Organic biomarker abundances…………………………………………………….179 4.5 Thermocline depth core-top calibration…………………………………………….180 4.6 Thermocline depth histories from the EEP…………………………………………182 4.7 Relationships between thermocline depth and mixed layer properties……………..183 4.8 Relationships between SST and phytoplankton productivity………………………184 5.1 LGM productivity anomaly map…………………………………….......................212 5.2 Map of mean annual chlorophyll and oceanographic zones in the EEP……………214 xxvi 5.3 Maps of SST and pCO2 with core locations and major surface and subsurface currents………………………………………………………………………………….215 5.4 Organic biomarker abundances at open-ocean sites………………………………..216 5.5 C37-alkenone abundances in sediment cores from the Galapagos and the Peru Margin…………………………………………………………………………………..218 5.6 Cross-plot of alkenone and brassicasterol abundances……………………………..219 5.7 Ratio of brassicasterol to C37-alkenone abundances at EEP open-ocean sites……..220 5.8 LGM productivity anomalies with records based on % calcite, % opal, and benthic foraminiferal transfer functions removed………………………………………………222 xxvii CHAPTER ONE ________________________________________________________________________ Links between Eastern Equatorial Pacific stratification and atmospheric CO2 rise during the last deglaciation Samantha C. Bova1, Timothy D. Herbert1, Yair Rosenthal2, Julie Kalansky2,3, Mark Altabet4, Caitlin Chazen1, Angel Mojarro1, and Jana Zech1 1 Brown University, Department of Earth, Environmental, and Planetary Sciences, Providence, RI USA 2 Rutgers The State University of New Jersey, Department of Marine Science and Earth and Planetary Sciences, New Brunswick, NJ, USA 08901 3 Scripps Institution of Oceanography, University of California San Diego, La Jolla, CA, USA 4 University of Massachusetts, School for Marine Science and Technology, New Bedford, MA, USA Published in Paleoceanography, 30(11), 1407-1424 doi:10.1002/2015PA002816 1 Abstract It is difficult to untangle the mixed influences of high and low latitude climate forcing in the eastern equatorial Pacific (EEP). Here we test the hypothesis that the Southern Ocean drove change in the EEP via subsurface intermediate waters during the last deglaciation. We use the δ18O signature of benthic foraminifera to reconstruct water density changes during the last 25 kyrs at three intermediate water depths (370 m, 600 m, and 1000 m) in the EEP. Carbonate δ18O records a combined signature of temperature (0.25‰ δ18Oshell /°C) and salinity (0.24‰ δ18Oshell /psu) and is therefore more closely related to density than temperature or salinity alone. We find that benthic foraminiferal δ18O values decreased first in the subsurface, simultaneously with rising temperatures over Antarctica, and propagated up to the surface within ~3 kyrs. The early subsurface response initiated a rapid decrease in density stratification over the upper water column as indicated by reduced δ18O gradients between surface and intermediate depths. Stratification of the upper water column remained low through the termination, with stratification minima reached during Heinrich Stadial 1 and the Younger Dryas (YD), synchronous with the two-part deglacial rise in atmospheric CO2. Centennial-scale shifts towards heavier δ18O signatures at 370 and 600 m during the YD indicate short-lived shifts in the Subantarctic Mode Water/Antarctic Intermediate Water boundary to shallower intermediate depths. We suggest decreased density gradients during the deglaciation accelerated vertical mixing across the EEP, and potentially the entire South Pacific subtropical gyre, which enhanced CO2 delivery from depth to the surface ocean and atmosphere. 2 1. Introduction The last deglaciation represents earth’s most recent natural experiment in rapid global climate change [e.g. Denton et al., 2010]. Beginning at 18 ka, a relatively minor change in the amount of incoming solar radiation to the high northern latitudes triggered a series of climate feedbacks (such as ice-albedo and CO2) that amplified and distributed high latitude temperature anomalies through the climate system, resulting in a 4°C rise in global mean surface temperatures, large-scale changes in global patterns of ocean circulation, and a two-step rise in atmospheric CO2 [Imbrie et al., 1993; McManus et al., 2004; Denton et al., 2010; Jouzel et al, 2007; Lourantau et al., 2010; Shakun et al., 2012; Marcott et al., 2014]. Ice cores recovered from Greenland and Antarctica provide well- constrained records of the timing and nature of these transitions [Grootes et al., 1993; Jouzel et al., 2007; Lourantau et al., 2010; Marcott et al., 2014]; however, we still have few constraints on how high latitude change propagated spatially over time. One key unknown is how and to what extent high latitude forcing influenced low latitude climate, particularly in regions where surface and shallow subsurface properties have strong links to global climate, such as the equatorial Pacific [Clement and Cane, 1999; Rosenthal et al., 2003]. Dynamical shifts in the behavior of the EEP amplify changes to the global climate system through their control on heat and carbon exchange with the atmosphere. Upwelling from beneath a strong shallow thermocline regulates the exchange of cool, nutrient-rich intermediate waters with the surface ocean, exerting a first order control on the regional air-sea CO2 flux as well as the efficiency of Pacific meridional heat transport [Hastenrath, 1980; Takahashi et al., 2002 Philander and Fedorov, 2003; Boccaletti et al., 3 2004]. Heat and carbon exchange with the atmosphere is regulated by the efficiency of exchange between intermediate and surface waters and therefore depends strongly on oceanic circulation and stratification patterns. Subsurface variability, arising at the poleward boundary of the subtropical gyres [Liu and Philander, 1995; Gu and Philander, 1997; Lu et al., 1998] and even higher latitudes via intermediate waters [Toggweiler et al., 1991; Andreasen et al., 2001] may play an essential role in setting the background stratification state of the EEP. Both high and low latitude climate forcings are important to EEP dynamics, which makes the relative role of tropical versus high latitude influences and high latitude southern versus high latitude northern hemisphere influences difficult to parse during the last deglaciation. On short timescales, locally forced ocean-atmosphere interactions regulate the strength and direction of the prevailing winds and, in turn, patterns of upwelling in the region. There is a strong seasonal cycle in the EEP and the El Niño Southern Oscillation represents the primary source of interannual variability [Philander, 1985; Clement and Cane, 1999; Lavin et al., 2006]. On centennial to millennial timescales, however, remotely forced mechanisms may overwhelm locally driven variability. These mechanisms include: (1) changing interhemispheric temperature gradients that are capable of driving long-lived N-S shifts in the mean position of the easterlies and thus the zones of upwelling within the EEP via shifts in the Hadley circulation (i.e. the atmospheric bridge) [Pahnke et al., 2007; Koutavas and Sachs, 2008] and (2) changes in the transport and/or physio-chemical properties of subsurface waters via oceanic tunnels [Liu and Yang, 2003]. The efficiency and relative importance of these two pathways are currently not well-constrained. 4 Using a suite of well-dated, sediment cores from three intermediate water depths, we evaluate the response of the EEP upper water column to deglacial climate change to assess the role of the oceanic tunnel in driving oceanographic change in the EEP. Benthic foraminiferal δ18O records provide a combined signal of temperature and salinity, and therefore density at the surface, 370 m, 600 m, and 1000 m water depth in the equatorial cold tongue region throughout the last 25 kyrs. An alkenone record estimates sea surface temperature (SST) for the last 25 kyrs. Together, these records provide direct evidence for significant changes in EEP intermediate water structure and stratification during the last deglaciation at high temporal resolution with important implications for regional heat and carbon balance. We diagnose high versus low latitude driven change by comparing the timing and relative magnitude of warming across the EEP upper water column; locally forced coupled ocean-atmosphere interactions will produce the largest recorded signal of oceanographic change at the sea surface (i.e. top-down forcing), while remotely forced oceanic mechanisms are observed first, and most strongly in the subsurface (i.e. bottom-up forcing). In the EEP, we find the largest and earliest signals of deglacial warming in the subsurface, which provides support for the role of the Southern Ocean in driving deglacial change in the EEP water column from the bottom up. 2. Modern Oceanographic Setting The EEP is a region of complex currents forced primarily by the local wind field and intermediate water advection [Wyrtki, 1966; Fiedler and Talley, 2006; Kessler, 2006]. The southeast trade winds drive surface currents and Ekman divergence along the equator and the South American coast [Wyrtki, 1966; Kessler, 2006]. Upwelling velocities along the Peru Margin are strongest and maintain the equatorial cold tongue, a 5 region of anomalously cold sea surface temperatures just south of the equator [Kessler, 2006]. In the subsurface, the eastward flowing Equatorial Undercurrent (EUC), centered at ~80 m depth, flows from west to east across the Pacific Basin transporting modified Subtropical Mode Water (STMW). Below the EUC, lies the thermostad, a region of relatively uniform temperature (11-14°C) between ~150 and 300 m depth in the EEP [Tsuchiya, 1981]. More than 80% of the thermostad water is derived from mixing between waters above (STMW) and below the thermostad (Subantarctic Mode Water/Equatorial Pacific Intermediate Water), with horizontal advection contributing much less by volume (Qu et al., 2009). Some evidence suggests thermostad waters do reach the surface mixed layer in the EEP [Qu et al., 2009]; however, the influence of thermostad waters on regional SSTs appears negligible during at least the late Holocene [Kalansky et al., 2015] Subantarctic Mode Water (SAMW) and Equatorial Pacific Intermediate Water (EqPIW), a mixture of AAIW and Pacific Deep Water, dominate at water depths between 300 and 1200 m in the EEP [Sarmiento et al. 2004; Fiedler and Talley, 2006]. SAMW overlies EqPIW between about 300 and 600 m in the EEP where it experiences strong vertical mixing at the base of the thermostad [Fiedler and Talley, 2006; Kalansky et al., 2015]. SAMW and AAIW (a major component of EqPIW) are sourced from the southern high latitudes and transported into the EEP through an oceanic tunneling system (Figure 1) [Toggweiler et al., 1991; Bostock et al., 2010; 2013]. SAMW forms in the Subantarctic Zone, within the northernmost Antarctic Circumpolar current during late winter convective overturning and travels from the Southern Ocean northwestward to join the New Guinea Coastal Undercurrent, eventually flowing eastward across the equatorial 6 Pacific into the EEP subsurface (Figure 1) [McCartney, 1975; Toggweiler et al., 1991; Rintoul and England, 2002; Qu et al., 2009; Bostock et al., 2010; 2013]. Today, SAMW temperature and salinity signatures are not observed north of 30°S (Figure 1) [Herraiz- Borreguero and Rintoul, 2011]. However, water column radiocarbon (Δ14C) data from the Peru margin and neodymium and δ13C reconstructions from the EEP indicate that SAMW penetrates to the equator today and has done so for at least the last 30 kyrs [Spero and Lea, 2002; Toggweiler et al., 1991; Pena et al., 2013]. These data indicate that sufficiently strong SAMW temperature and/or salinity signals contribute to equatorial subsurface ocean heat content and stratification anomalies [Kalansky et al., 2015]. EqPIW dominates at intermediate depths in the EEP below ~600 m. Although AAIW is the main component of EqPIW, the geochemical signature of EqPIW is clearly distinct from that of AAIW. EqPIW has a higher minimum salinity (34.5-34.6 psu), higher nutrients, higher silicate, higher DIC, lower oxygen, and an older Δ14C signature, which indicate mixing between AAIW and Pacific Deep Water in the EEP [Bostock et al., 2010; 2013]. Despite this, AAIW is still a major conveyor of heat and salt to the intermediate depth EEP, bringing temperature and salinity signals from the Southern Ocean where the water mass forms in three hotspots: (1) southwest and east of New Zealand, (2) west of the East Pacific Rise, and (3) west of the Drake Passage (Figure 1) [Bostock et al., 2013; Herraiz-Borreguero et al., 2011]. The main AAIW formation region is the southeast Pacific site west of the Drake Passage where the coldest and densest forms of SAMW mix with Antarctic Surface Waters (AASW) south of the 7 subantarctic front (Figure 1) [McCartney, 1975; Sloyan and Rintoul, 2001; Bostock et al., 2010; 2013; Hartin et al.; 2011]. 3. Site Locations A series of sediment cores were collected during a cruise of the R/V Knorr 195-5 to the eastern equatorial Pacific cold tongue in 2009. We analyzed four cores representing three water depths to evaluate changes in the EEP subsurface structure through time. CDH 23 (03° 44.95 S, 81° 08.05 W) and CDH 26 (03°59.16 S, 81°18.52 W) were recovered along the Peru Margin at 373 and 1023 m depth, respectively (Figure 2). Cores CDH 41 (01° 15.94 S, 89° 41.88 W) and GGC 43 (01°15.13 S, 89° 41.07), were recovered at 595 and 617 m water depth from the Galapagos Platform, just north of Española Island. At present, the CDH 23 core depth (370 m) lies within SAMW, CDH 41/GGC 43 recovery depths lie near the boundary between SAMW and EqPIW, and CDH 26’s coring site is bathed by EqPIW (Figure 2). 4. Methods 4.1 Age Models Age models were constructed by applying a polynomial fit to a set of planktonic radiocarbon dates for each core (Figure 3). All AMS 14C dates were analyzed at the NOSAMS facility at Woods Hole Oceanographic Institute. Each radiocarbon date represents an average of 200-250 individuals, picked from the >150µm size fraction. The age model for the Galapagos composite (GGC 43 and CDH41) is based on 26 of 33 radiocarbon measurements made on the planktonic foraminifer Globigerinoides ruber. The cores are spliced together at 20.2 ka, which corresponds to 365 cm in core GGC 43 and 95.25 cm in core CDH 41. Seven radiocarbon dates were excluded from our 8 calculation of the CDH 41 age model. Two of these dates, at 215.25 and 235.25 cm, were thousands of years too young (Figure 3b) and require movement of younger foraminfera through more than 150 cm of sediment. Given the lack of obvious bioturbation features in the core, these anomalously young 14C dates may arise due to deformation of the sediments during coring. We applied a polynomial fit to the remaining 14C dates and calculated the average offset between the predicted age and the 14C calibrated calendar age at each age control point. Five dates, whose offset from the fit exceeded the average plus the standard deviation of the offsets (>1150 years), were removed from the final age model (Figure 3b). The age profiles for cores CDH 23 and CDH 26 are constrained with 24 and 25 AMS 14C measurements, respectively, made on the thermocline dwelling foraminifer, Neogloboquadrina dutertrei. Planktonic foraminifera are scarce in the Peru Margin cores with only N. dutertrei consistently abundant. AMS 14C ages were converted to calendar ages using the calibration curve of Fairbanks et al. [2005]. A constant reservoir age of 500 years is assumed for all measurements. Higher order polynomial fits were tested but only assigned when the R2 values were statistically different from that of the lower order using an F test. The age model for CDH 26 was further tuned to a nearby core, CDH 23, using the Match 2.3.1 software package to create a composite and internally consistent chronology for the two cores [Lisiecki and Lisiecki, 2002]. Our match is based on alkenone sea surface temperature and nitrogen isotope records from each core and thus explicitly assumes synchronicity between the two sites. This assumption is reasonable because CDH 23 and CDH 26 are located just 32 km apart. Based on our composite fit, we did not 9 identify any outliers among the CDH 26 AMS 14C dates and therefore revert to the original polynomial fit for age control. Our age models indicate higher sedimentation rates along the Peru Margin relative to the Galapagos Platform. The Galapagos cores, GGC 43 and CDH 41 have average sediment accumulation rates of circa 20 cm/kyr and 14 cm/kyr, respectively, while the Peru Margin cores each have sedimentation rates well over 100 cm/kyr. CDH23, the shallower core from the Peru Margin, exhibits the highest sedimentation rate, approaching 130 cm/kyr, which is presumably due to sediment trapping at shallower depths due to late deglacial and Holocene sea level rise. CDH 26 has a slightly lower average sedimentation rate (~120 cm/kyr) than the shallower core but its record extends to 25 ka. 4.2 Alkenone SST estimates A ~25 kyr high-resolution alkenone-inferred sea surface temperature record was produced from CDH 26 and CDH 23. Samples were taken every 4 cm from the top 2.5 m of CDH 26 and the top 1.7 m of CDH 23, providing a measurement approximately every 35 years. One to two grams of freeze-dried sediment were extracted in a Dionex 200 Accelerated Solvent Extractor and analyzed on a Gas Chromatograph Flame Ionization Detector, using a modified method of Herbert et al. [1998]. Temperatures were calculated using the UK’37 index and the calibration of Müller et al. [1998]. Lab analytical error is equivalent to ± 0.1°C based on 55 replicate analyses of a laboratory standard. Approximately 10% of our analyzed samples were run in duplicate with an average UK’37-based temperature reproducibility of ± 0.05°C. 10 4.3 Benthic Stable Isotope Records Benthic foraminiferal stable isotope measurements were taken at ~350-year time steps in cores CDH 26, CDH 41, and GGC 43. The higher sedimentation rate of CDH 23 allowed higher resolution, with measurements taken approximately every 120 years. Samples were freeze-dried, re-soaked in DI water, and wet-sieved through a series of mesh sieves to separate size fractions. Three to four specimens of the benthic foraminifer Uvigerina peregrina were picked from the 212-355 µm size fraction for all cores. Samples were analyzed at Brown University on a Finnigan MAT 252 isotope ratio mass spectrometer with a Kiel III carbonate device. Each sample was measured in duplicate, with an average standard deviation of ±0.15‰ for δ18O and ±0.13‰ for δ13C measurements. δ18O and δ13C values are expressed relative to the Vienna Pee Dee Belemnite standard. The isotopic time series discussed below reflects the average of the replicates. Three outliers were removed from the CDH 23 time series due to small sample size. 4.4 Benthic Temperature Records Benthic foraminiferal oxygen isotope compositions reflect both seawater δ18O (salinity and ice volume, ~0.24‰ δ18Oshell /psu) [Fairbanks et al., 1982; Conroy et al., 2014] and the ambient temperature of formation (0.25‰ δ18Oshell /°C), providing a record of seawater density at the depth of recovery [Lynch-Stieglitz et al., 1999a; 1999b]. We approximate the temperature component by estimating the δ18Osw evolution using modern δ18Osw profiles acquired from bottle samples recovered via CTD (Conductivity, Temperature and Depth) casts on station during coring and correcting these values for ice volume using the Barbados sea level record [Peltier and Fairbanks, 2006]. Bottle 11 samples were analyzed for water isotopes at Brown University on a Picarro L1102-i isotopic water liquid analyzer. This treatment of the data explicitly assumes no local hydrologic salinity imprint on the δ18O values through time. This is a reasonable interpretation because (1) modern vertical and lateral salinity gradients are much weaker than corresponding temperature gradients [WOA13] and (2) records from the Southern Ocean indicate little influence of local hydrologic conditions on subantarctic seawater δ18O records [Mashiotta et al., 1999]. Glacial-interglacial temperature change accounts for 40-60% of the δ18O signal in subantarctic planktonic foraminifera, with the remainder dominated by the global ice volume changes [Mashiotta et al., 1999]. We therefore converted benthic stable isotope values to temperature (°C) using the paleotemperature equation of Bemis et al. [2002] calibrated specifically for Uvigerina spp. in the Santa Barbara Basin. T (°C)= 21.6 - 5.50 ✕ (δ18Oc - δ18Osw) The calibration range for this relationship, 3 to 8°C, is appropriate for the temperature range observed at intermediate depths in the EEP [Bemis et al., 2002]. Duplicate Uvigerina (δ18Oc) values were averaged and a constant offset of -0.3‰ was applied to bring core top values in line with modern values. Corrected core top temperature estimates based on benthic δ18Osw agree within ± 0.8°C of directly measured modern values. Alternate calibrations were also applied [Shackleton, 1974; Marchitto et al., 2014] but ultimately rejected because calculated temperatures were unrealistically low. 12 5. Results 5.1 Stable Isotope Records 5.1.1 Oxygen isotopes Over the past 25 kyrs, benthic foraminiferal δ18O values decreased towards the present day at all water depths in the EEP; however, the timing and nature of the transition to interglacial values differs between records (Figure 4a). We determine the timing of the transition in each record using the Change-Point Analyzer software package, which uses mean square error estimates and bootstrapping to detect change [Taylor, 2000]. For each change point, we provide the most likely value as well as the 95% confidence interval in the main text. The record from 1000 m water depth exhibits the heaviest benthic δ18O values as well as the largest range of values (1.7‰). Glacial δ18O values reached a maximum at 21 ka before a rapid but steady shift to lighter values began at 17.6 ka (17.6-17.7, p<0.05). The transition from 17.6 to 14.9 ka accounts for nearly half of the total decrease in δ18O over the full record. Values stabilized from 14.9 to 13.6 ka, before decreasing towards present day. Minimum δ18O values are reached during the early and mid-Holocene together with a decrease in short-term variability. Shallower in the water column, at 600 m, benthic δ18O values began to decrease at 16.6 ka (16.4-17.0 ka, p<0.05), approximately 1000 years later than at 1000 m. The early deglacial period, (16.6 to 14.3 ka) produces just over a third of the total δ18O decrease. From 14.3 ka to 12.3 ka we observe centennial-scale oscillations, up to 0.3 ± 0.15‰. This time period is also characterized by a minimum offset in δ18O values between 600 and 1000 m. At the end of this period δ18O values shift rapidly, decreasing by 0.5 ± 0.15‰ in 13 just 500 years. This 500-year shift accounts for nearly half of the total δ18O decrease during the deglaciation. After, δ18O gradually decreases throughout the Holocene portion of the record. The benthic δ18O record from 370 m does not extend to the LGM, but does cover the late deglacial and Holocene time periods in great detail. As seen at 600 m, centennial- scale variability is observed from 13.9-12.3 ka at 370 m. Then, after a rapid 300 yr transition to lower values beginning 10.8 ka, δ18O values decreased towards the middle Holocene, reaching a minimum at 4.8 ka. 5.1.2 Carbon Isotopes The carbon isotope records from the Peru Margin show much larger variability over the last glacial-interglacial cycle than the records from the Galapagos platform (Figure 4b). The Peru Margin cores exhibit an increasing trend towards present day with centennial-scale fluctuations up to 0.72‰ throughout the record. In contrast, the Galapagos cores vary by just 0.35‰ over the entire 25 kyr record. At 1000 m water depth we observe two periods of prolonged δ13C depletion: (1) Late Glacial to Early Deglacial (~22 -16 ka) and (2) Late Deglacial (~14.2-12.7 ka). The record from 370 m suggests δ13C values continued to increase through the late Holocene until 8 ka when δ13C values stabilized at an average value of 0.13 ± 0.13‰. δ13C values at 370 m are consistently lighter than those observed at 600 and 1000 m, but records from 600 and 1000 m do not follow the expected heavy-light trend from the surface to depth. This is likely related to change in the strength and extent of the Peru Margin oxygen minimum, with the biggest impacts of metabolically-derived dissolved inorganic carbon (DIC) felt closest to the Peru Margin. 14 5.2 Temperature Records 5.2.1 Alkenone Sea Surface Temperatures Our alkenone SST record from the Peru Margin exhibits a 3.3°C increase over the last 25 kyrs (Figure 5). Our age models indicate that the cores used in this study do not contain modern sediments. However, the core-top temperature of CDH 23 is within 0.7°C of the mean annual SST (22.42°C) for the closest World Ocean Atlas [2013] 0.25 degree grid box [WOA 13]. We therefore consider the seasonal bias at our sites negligible [Kienast et al., 2012]. Additionally, a recent synthesis demonstrates that UK’37-estimated SST from the region are within 0.42-1.77°C of mean annual SST [Kienast et al., 2012]. Although many of the alkenone-inferred temperatures are biased towards overestimating observed temperatures in the near-coastal Peru upwelling zone, this bias may be due to large spatial interpolation of limited SST measurements. This is particularly problematic in the complex and spatially heterogeneous near-coastal region examined here. The timing of SST change across the termination is broadly similar to that suggested by Koutavas and Sachs [2008] but the higher resolution of our record provides a new look at the fine-scale structure of the SST evolution across the termination and Holocene. Cool temperatures reconstructed at the last glacial maximum evolved to reach the lowest observed temperature of the entire 25 kyr record, 19.4 ± 0.1°C, about 14.7 ka (14.5-14.9 ka, p<0.05). This temperature decrease is followed by a rapid rise through the late Holocene, peaking to 23.5°C at 6 ka before decreasing again towards present day. This alkenone SST record, and those produced by Koutavas and Sachs [2008], differ from EEP Mg/Ca SST records, which instead suggest an early SST warming beginning at least 18 ka [Lea et al., 2000; Pena et al., 2008]. 15 5.2.2 Benthic Temperatures The reconstructed temperature evolution in the EEP subsurface over the past 25 kyrs varies significantly with depth (Figure 5). Temperatures at 1000 m and 600 m increased by ~3°C and 1°C across the termination, respectively. At 1000 m, temperatures were stable at ~2°C through the glacial but begin to increase at 17.9 ka (17.8-18.6, p<0.05) by as much as 1°C/kyr. After a brief period of cooling (13-15 ka), warming continued until reaching a maximum temperature (6.7°C) at approximately 11.8 ka. A mid-Holocene maximum occurred at 7.4 ka and is followed by a cooling trend towards present day. The temperature history at 600 m water depth is characterized by centennial-scale temperature swings up to 2.7°C during the late deglacial (12-14 ka). Relative to temperatures at 1000 m, we do not observe significant warming during the deglaciation. The thermal evolution at the shallowest core site (370 m) along the Peru Margin reveals a similar temperature trend to that seen at 600 m water depth. However, since the 370 m record is truncated at 14.1 ka, we cannot assess the glacial-interglacial temperature change at this site. Instead, we observe a period of highly variable temperature from 12- 14 ka, with centennial-scale temperature fluctuations up to 2.6 ± 0.5°C, analogous to that seen at 600 m. We also note a near zero offset between temperatures at 600 and 370 m from 14 until 11 ka when a rapid transition at 370 m shifts temperatures to the warmest recorded throughout the record. After, temperatures decreased gradually towards the modern day at an average rate of 0.14°C/kyrs. 16 6. Discussion 6.1 Depth dependent response to deglacial forcing in the EEP upper water column Timing of the initial response to deglacial warming is depth dependent in the EEP upper water column (surface to 1000 m), with subsurface mode and intermediate waters responding thousands of years before the surface ocean. Our deepest site at 1000 m water depth, within the heart of EqPIW, responded first at 17.6 ka (17.6-17.8 ka, p<0.05), synchronously (within the limitations of radiocarbon dating) with rising temperatures over Antarctica. Timing of the initial δ18O decrease at 600 m water depth, within SAMW, is more uncertain due to high variability in δ18O values across the transition and a less well-constrained age-depth model. However, the best estimate for the timing of the transition is 16.6 ka (16.4-17 ka, p<0.05), approximately 1 kyrs later than the record at 1000 m. Water masses at both sites are sourced primarily from the Southern Ocean where southward shifts in Southern Ocean frontal boundaries, in addition to rising atmospheric temperatures over the Southern Ocean, were likely responsible for the early warming signal in these water masses [e.g. Denton et al., 2010]. At the onset of the deglaciation, SSTs in the circum-Antarctic region warmed in phase with glacier recession in the southern mid-latitudes and southward displacement of the southern westerly winds (SWWs) to at least the northern boundary of the SAMW formation region [Lamy et al., 2007; Caniupan et al., 2011; Ho et al., 2012; Barker et al., 2009; Putnam et al., 2013]. SST records from the SAMW and AAIW water mass source regions rise shortly after 19 ka, in phase with Antarctica and each other [Lamy et al., 2007; Caniupan et al., 2011; Ho et al., 2012; Barker et al., 2009]. We therefore suggest that SAMW and AAIW warmed synchronously and the delay recorded at 600 m is either a statistical fallacy or results 17 from a change in water mass contributions from southern (SAMW, AAIW), northern (NPIW), and/or subtropical waters (STMW) at the core site, with the subtropical and northern hemisphere sourced waters warming later than those sourced from the southern high latitudes. Timing of the initial decrease in benthic δ18O depends on a combination of factors, including global ice volume as well as the in-situ temperature and salinity properties of seawater. Therefore, the bottom-up timing likely reflects a depth dependence in (1) the timing of the influx of light δ18O from the North Atlantic to the Pacific subsurface as suggested by Stern and Lisiecki [2014], (2) the response of intermediate and mode water temperature and salinity properties, and/or (3) changes in the fractional water mass contributions to the core site. The spatiotemporal pattern of deglacial meltwater and propagation of its light δ18O signature into the ocean interior remains poorly constrained; model output [Friedrich and Timmermann, 2012] and observations [Stern and Lisiecki, 2014] from the intermediate and deep oceans are in direct opposition, particularly in the Pacific. We therefore assume a globally uniform response in seawater δ18O to ice volume changes over the past 25 ka when estimating intermediate depth temperature histories from our benthic foraminiferal δ18O records. After applying an ice volume correction [Fairbanks et al., 2005] we still observe the earliest response at depth, with temperatures at 1000 m (EqPIW/AAIW) mirroring those over Antarctica (Figure 5). Sea surface temperature records from SAMW and AAIW formation zones exhibit similar temperature histories to that of EqPIW in the EEP [Lamy et al., 2007; Caniupan et al., 2011; Ho et al., 2012]. The synchronous timing between Antarctic air recorded in EPICA Dome C ice core and 18 surface ocean temperatures throughout the Southern Ocean suggest SST changed rapidly and in phase with temperatures over the continent (Figure 5) [Jouzel et al, 2007; Lamy et al., 2007; Caniupan et al., 2011; Ho et al., 2012]. Warming at the EEP surface is delayed until 15 ka, approximately 3 kyrs later than subsurface intermediate waters (Figure 5). This vertical pattern of deglacial warming resembles the latitudinal one shown by Shakun et al. [2012] where Antarctic temperatures exhibit a 2 kyr lead over the global average. The 3 kyr delay in EEP surface warming, as recorded by alkenones, is thus supported by the upward (or south to north) propagating signature of deglaciation and demonstrates that EEP SSTs were not sensitive to the early rise in southern hemisphere (SH) temperatures or atmospheric CO2 at the start of the deglacial. This delay could be the result of a zonal adjustment of SST across the Pacific Basin and/or increased upwelling of cool waters from beneath the thermocline in the EEP. Delayed warming at the surface is consistent with previous alkenone-SST reconstructions from the region [Kienast et al., 2006; Koutavas and Sachs, 2008] but in opposition to Mg/Ca-based reconstructions of SST, which are synchronous with Antarctic air temperatures [Lea et al., 2000; Pena et al., 2008]. The disagreement may be the result of differences in the depth habitat or seasonality of alkenone producers and planktonic foraminifera [Shaari et al., 2014; Timmermann et al., 2014]. Here, we interpret EEP alkenone UK’37 measurements as a recorder of mean annual sea surface temperatures along the northern Peru Margin (see more detailed discussions in Kienast et al., [2012] and Timmermann et al., [2014] in support of this interpretation). Coupled alkenone SST and Mg/Ca records over the last 16 kyrs from site M772-059, just a few 19 km from our core location, corroborate this interpretation [Nurnberg et al., 2015]. Absolute temperature estimates and the amplitude of SST change reconstructed by shell Mg/Ca and by alkenones are essentially identical during the deglaciation [Nurnberg et al., 2015]. Furthermore, the progression of rising temperatures towards the surface as recorded by our δ18O depth series provides independent support for the lag in SST inferred from alkenone paleothermometry (Figure 4). An important consequence of a delay in surface warming relative to the subsurface is a decrease in the thermal gradient across the upper water column during the glacial termination and thus, assuming constant salinity, a decrease in density stratification between surface and intermediate water masses. However, given the inherent uncertainties in our temperature proxies outlined above, we assess a change in stratification over the top 1000 m of the water column in temperature, δ18O, and δ13C space (Figure 6). Because these records provide similar stratification histories for the EEP (Figure S1), we normalized all records to a late Holocene value and calculated an average offset between the surface and intermediate water records to produce one index for the strength of stratification (Figure 7). Two previously published planktonic stable isotope records measured on G. sacculifer from cores V19-28 (2°22’S, 84°39’W) and V21-30 (1°13’S, 89°41’W) are used to represent surface δ18O and δ13C values [Koutavas and Lynch-Stieglitz, 2003]. Measurements from all records were linearly interpolated to ages where there are measurements at site V19-28, the lowest resolution record, and normalized to a late Holocene value (2785 yrs ago). δ18O and δ13C gradient anomalies represent an average of the normalized offset between the surface and subsurface records (Figure 6). The thermal gradient anomaly is calculated in the same manner; however, we 20 use alkenone SST records from CDH 23 and CDH 26 to represent the SST evolution and calculated benthic temperatures from 370, 600, and 1000 m (Figure 6). The stratification gradient anomaly plotted in Figure 7 represents the average of the temperature, δ18O, and δ13C gradient anomaly calculations. Each of the three gradient anomaly calculations can be viewed separately in Figure 6. Values of the stratification gradient anomaly greater than zero represent times when stratification between surface and intermediate waters was stronger than the late Holocene, while values less than zero are characterized by lower upper water column stratification. Stratification indices based on our temperature (alkenone SST record and foraminiferal δ18O derived temperature calculations) and stable oxygen and carbon isotope records (planktonic records from Koutavas and Lynch-Stieglitz [2003] and our benthic isotope records) all indicate that stratification was strongest at the end of the last glacial period, decreased rapidly at the onset of Heinrich Stadial 1 (HS1) and the Younger Dryas (YD) and increased towards present day (Figure 6, 7, see also Figure S1 in the supplemental information). These records support a latitudinal response by water mass to deglacial warming, with Southern Ocean intermediate and mode waters responding first. The physiochemical properties of Southern Ocean intermediate waters (SOIWs) in the EEP were in phase with temperatures in the SE Pacific and over Antarctica; this implies rapid adjustments of the heat content of intermediate waters with atmospheric temperatures surrounding the Antarctic continent and rapid transport into the EEP subsurface. Any deglacial trends in the EEP upper water column driven by changing air-sea interactions were thus overwhelmed by variability in the transport and/or physio-chemical properties of the SOIWs as demonstrated by the delay in surface warming. We therefore suggest that 21 during the last glacial termination, a period of rapid change in the global radiation balance, remotely forced “bottom-up” rather than local “top-down” forcing drove oceanographic change in the EEP cold tongue. 6.2 EEP intermediate water structure during HS1 and the YD The last deglaciation was not a smooth transition from glacial to interglacial conditions but was punctuated by three abrupt millennial-scale events: HS1, the Antarctic Cold Reversal (ACR), and the YD. HS1 (14.6-17.6 ka) and the YD (11.7-12.9 ka) are Northern Hemisphere stadial (SH interstadial) events likely caused by NH ice sheet collapse and the subsequent weakening of the Atlantic Meridional Overturning Circulation [McManus et al., 2004]. Despite NH cooling, global temperatures rose steadily throughout these periods in phase with atmospheric CO2 levels [Lourantau et al., 2010; Shakun et al., 2012]. Several recent studies suggest intermediate water chemistry, formation rate, and volume export to the low latitudes responded to these high latitude events [e.g. Pahnke and Zahn, 2005; Anderson et al., 2009; Pena et al., 2013). The benthic isotope records presented here from 370 m, 600 m, and 1000 m, sites that straddle the boundary between SAMW and EqPIW/AAIW, provide a rare opportunity to directly assess the response of EEP intermediate water structure to these events. The results indicate that during HS1 and the YD 1) the boundary between these water masses shoaled by at least 230 m during NH stadials and 2) density stratification in the EEP weakened by approximately 10% relative to the last glacial period. 6.2.1 Shoaling of the SAMW/AAIW Boundary In the EEP, the YD and HS1 are characterized by anomalously small carbon and oxygen isotope gradients across the EEP upper water column. During the YD, benthic 22 foraminiferal δ18O values at 370 m and 600 m are approximately equivalent and are closer to values seen at 1000 m water depth, while carbon isotope values exhibit local minima in each sediment record (Figure 4). We suggest these periods of isotopic convergence across intermediate depths in the EEP reflect a shoaling of the SAMW/AAIW boundary from its modern day position at ~600 m to at least 370 m depth, with numerous centennial-scale shifts. At these times AAIW (or at the very least a more AAIW-like water mass) bathed all core sites, which contributed to a decrease in density stratification between 370 and 1000 m depth. The modern boundary between SAMW and AAIW lies at approximately 600 m but may shift in response to climatic changes in their source regions. Physical models show that southward displacement of the westerly winds force deepening and thickening of the thermocline, the depth interval containing SAMW and AAIW, in the Southern Hemisphere, [Toggweiler et al., 2006; Russell et al., 2006; Downes et al., 2011; Hain et al., 2014]. During the YD and HS1, reduced SH pole to equator temperature gradients likely forced southward migration of the SWWs, increasing upwelling in the Drake Passage, forcing stronger Ekman upwelling, and increasing formation of Antarctic Surface Water (AASW) that feeds AAIW production [Oke and England 2004; Toggweiler et al., 2006, Russell et al., 2006; Downes et al., 2011]. Radiocarbon, stable isotope, and carbonate ion concentrations from the Southern Ocean support a downward displacement of the AAIW lower boundary during the deglaciation, consistent with model results for a thickening of the AAIW layer [Pahnke and Zahn, 2005; Allen et al., 2015]. 23 Benthic foraminiferal carbon isotope records from our core sites provide evidence for enhanced AAIW contributions to the EEP subsurface during HS1 and the YD, consistent with a thickening of the thermocline layer as observed further south. Today, the δ13C content of AAIW in the EEP subsurface is determined by multiple factors, including the carbon chemistry of upwelled Circumpolar Deep Water (CDW), air-sea exchange in the subantarctic zone, and local rates of primary production. During the deglaciation, enhanced upwelling of light δ13C in CDW and transport of this signal to the EEP thermocline via SOIWs was the primary cause of the deglacial carbon isotope minimum events [Spero and Lea, 2002; Martínez-Boti et al., 2015]. Our records corroborate this assertion. Although overprinting by respiratory CO2 is a concern, alkenone abundances (C37 total), a proxy for the rain rate of organic matter, does not show contemporary maxima with carbon isotope minima. In fact, inferred rates of surface primary production reach a broad minimum during the deglaciation at our core site (Figure S2). Furthermore, the timing of carbon isotope minima in the δ13C records from the Peru Margin are consistent with that observed in the Pacific sector of the Southern Ocean [Ninnemann and Charles, 1997], the influx of δ13C depleted SOIW to the EEP thermocline [Spero and Lea, 2002], and resurgence of Upper Circumpolar Deep Water (UCDW) upwelling in the Southern Ocean (Figure 4b) [e.g. Anderson et al., 2009] and suggest that advection of isotopically light DIC from the Southern Ocean was the primary driver of carbon isotopic change over the past 25 kyrs. Our data further suggest AAIW shoaled in the EEP water column during HS1 and the YD. Strengthened upwelling, rising atmospheric temperatures, and enhanced freshwater inputs to the AAIW source region may be responsible for the observed 24 vertical shifts in EEP subsurface water masses during these intervals [Mashiotta et al., 1999; Moreno et al., 2012]. Of these factors, surface salinity values in the Southern Ocean are the least well-constrained. Today, the freshwater balance over the Southern Ocean is regulated primarily by the seasonal position of the SWWs, which are the main source of freshwater to the Southern Ocean [Toggweiler et al., 2006; Moreno et al., 2012]. During the last glacial termination, poleward shifts in the SWWs likely altered the latitudinal distribution of precipitation, with zones of maximum precipitation tracking poleward movement of the SWWs during the YD and HS1 [Lamy et al., 2010; Moreno et al., 2012]. Sea ice also plays an important role in the fresh water balance over the Southern Ocean. During the deglaciation, sea ice retreat began in phase with SH warm events and was amplified by poleward shifts in the position of the SWWs and the subsequent break- up of sea ice via northeastward Ekman transport [Gersonde et al., 2005; Levermann et al., 2007]. Without sea ice capping its surface, AASWs began to directly exchange heat and fresh water with the atmosphere, which led to greater interaction between the regional hydrologic cycle and the surface ocean and thus enhanced fresh water inputs to AASWs and eventually SOIWs [Keeling and Stephens, 2001]. Consequently, during HS1 and the YD, net precipitation to the surface ocean in addition to melting sea ice led to a net increase in the freshwater inputs to the surface ocean and positive buoyancy forcing for AAIW and SAMW [Keeling and Stephens, 2001]. Increased freshwater inputs likely contributed to greater AAIW buoyancy at these times forcing AAIW, and therefore EqPIW, to shoal within the water column. Thus some of the observed isotopic depletion observed in our records during the YD and HS1 likely records freshening as well as 25 warming of intermediate water masses. Records of seawater δ18O from the Pacific sector of the Southern Ocean, however, indicate minimal influence from local hydrologic conditions [Mashiotta et al., 1999]. Regardless, even if freshening did have a significant effect on δ18O values at our site, the interpretation is the same: sub-surface density stratification was weak throughout the deglaciation, with pronounced minima during the YD and HS1. 6.2.2 EEP de-stratification events and the global carbon cycle Stratification across the upper water column was reduced during the last deglaciation relative to the end of the last glacial and late Holocene but reached minimum observed values during HS1 and the YD (Figure 7). At these times, the δ18O, δ13C, and temperature gradients across the upper water column decreased by approximately 10% relative to glacial conditions. Although a 10% reduction is not enough for AAIW or even SAMW to mix completely with surface waters, such a reduction in the upper ocean density gradient must have allowed for more vertical mixing and therefore greater penetration of subsurface mode and intermediate waters to the surface mixed layer. This notion is corroborated by a recent study: Nurnberg et al., [2015] demonstrate that temperature gradients were reduced between the sea surface and ~200 m (±~100m), a key depth interval not sampled by our cores, during the deglaciation relative to the Holocene at a site within a few kilometers of sites CDH 23 and CDH 26, presented here. In addition, three proxy records of surface ocean conditions—the alkenone C37total, planktonic foraminiferal δ13C values, and alkenone SSTs—support an increase in communication between intermediate and surface waters during these millennial events. 26 Productivity in the EEP depends primarily on the supply of cold, nutrient-rich intermediate waters from beneath the thermocline and iron limitations [Sarmiento et al., 2004; Pennington et al., 2006]. Therefore, greater productivity suggests enhanced nutrient delivery from the subsurface to the surface, either via stronger vertical mixing or higher nutrient concentrations in the upwelled water. During the deglaciation, productivity, as inferred from our alkenone C37total record of coccolithophorid productivity, was lower than during the glacial and Holocene at our site, but we find evidence for small increases in primary productivity during both HS1 and the YD (Figure 6). We observe local maxima in coccolithophore (C37 alkenone) [Figure 7] and diatom (brassicasterol) [Calvo et al., 2011; Pena et al., 2013] productivity in addition to planktonic isotope minima [Spero and Lea, 2002]; all three are independent indicators of enhanced primary production and greater nutrient delivery to the surface ocean. Reductions in the vertical density gradient over the upper water column observed during these intervals suggest strengthened vertical mixing was responsible for the small increases in surface primary production. During these intervals SOIWs composed as much as 25% of EUC water, roughly a 20% increase from the Holocene and present day average [Pena et al., 2013]. Finally, greater penetration of cool intermediate waters to the sea surface could help explain how SSTs remained cool despite rising atmospheric CO2 concentrations. Reduced upper ocean stratification during the YD and HS1 was not limited to the EEP but is also observed in the southwest and southeast Pacific (Figure 6) [Bostock et al., 2004; Siani et al., 2013]. The analogous timing and nature of the transition from strong glacial stratification to weak deglacial stratification in all records is consistent with a 27 change in upper water column stratification forced by the same mechanism at each site: an early response of subsurface intermediate waters to SH climate at the end of the last ice age [Bostock et al., 2004; Pena et al., 2013; Siani et al., 2013]. Together these records suggest a South Pacific-wide decrease in water column density stratification during HS1 and the YD. Given the vastness of the South Pacific, these de-stratification events may be responsible for ventilating large amounts of CO2 from a deep Southern Ocean reservoir [Ninnemann and Charles, 2002; Lourantou et al., 2010]. Ice core records from Antarctica show a two-step increase in atmospheric CO2 during the YD and HS1, in phase with periods of decreased stratification in the EEP and the South Pacific (Figure 7) [Lourantou et al., 2010, Marcott et al., 2014]. These findings are consistent with a rapid decrease in stratification in the Southern Ocean, enhanced upwelling of carbon-rich UCDW and degassing of isotopically light remineralized organic carbon to the atmosphere. A south Pacific-wide “carbon window”, however, requires that upwelled UCDW was resubducted as SOIWs faster than the air-sea carbon equilibration time or became entrained into SAMW and/or AAIW before it reached the surface [Holte et al., 2012; Holte et al., 2013]. This notion is corroborated by marine radiocarbon evidence from the southwestern Pacific Ocean, which showed that —at least at the onset of HS1—upwelled carbon-rich UCDW did not fully equilibrate with the atmosphere for at least 1000 years [Rose et al., 2010]. Instead, the non-equilibrated CO2 was likely resubducted or entrained in AAIW and released to the atmosphere somewhere along the flow path between the Southern Ocean and the EEP; decreased density gradients over the S. Pacific upper water column 28 allowed the resequestered CO2 to mix upwards, resurface and exchange with the atmosphere. Boron isotopes provide direct evidence for CO2 outgassing far afield from the Southern Ocean in the sub-Antarctic Atlantic and at least 3 sites across the equatorial Pacific, including the EEP [Palmer and Pearson, 2003; Douville et al., 2010; Kubota et al., 2014; Martínez-Boti et al., 2015]. At site ODP 1238, ~300 km northwest of our site, Martínez -Boti et al. [2015] record CO2 outgassing synchronous with periods of reduced upper ocean stratification. Although correlation does not imply causation, the remarkable similarity between records suggests reduced stratification enhanced delivery of CO2-rich subsurface waters to the surface mixed layer, providing a pathway for oceanic carbon to the atmosphere (Figure 7). Finally radiocarbon and δ13C depleted benthic and planktonic foraminifera from the Galapagos platform and the Peru Margin during HS1 and the YD provide further support for an inflow of carbon-rich intermediate waters to the EEP [Spero and Lea, 2002; Stott et al., 2009; Chapter 3]. 7. Summary and Conclusions Benthic foraminiferal isotope and temperature records recovered from three intermediate water depths in the EEP provide evidence for large changes in upper water column structure during the deglaciation forced by climate change over the Southern Ocean [Figure 8]. Rising temperatures over Antarctica led to poleward shifts in the westerly wind belt that together altered the buoyancy and water mass boundaries of southern ocean intermediate waters. SOIWs responded first, synchronous with atmospheric temperatures over the Antarctic continent, and transported the early SH deglacial warming response along its flow path to the base of the equatorial thermocline 29 and the EEP. Surface warming in the EEP was delayed by 3 kyrs, while records from the shallow subsurface at 600 m, document an averaged time response of the two, ~1 kyrs after Antarctica. The observed upward-propagation in our benthic δ18O and temperature records does not favor the tropical surface ocean as an early responder and promoter of deglaciation. Instead, the EEP surface ocean was likely a passive responder to subsurface change driven from the bottom-up by the southern high latitudes. During SH warm events (YD/HS1), the early warming response of AAIW drove rapid reductions in vertical temperature and salinity gradients in the EEP upper water column, which led to reduced stratification, shoaling of water mass boundaries, and enhanced vertical mixing. Thus despite rising atmospheric CO2 concentrations, greater penetration of cool subsurface waters to the sea surface during these events helped maintain a strong cold tongue. Reduced upper ocean stratification during the YD and HS1 was not limited to the EEP but is also observed in records from the southwest and the southeast Pacific [Bostock et al., 2004; Siani et al., 2012]. Comparable timing and structure of these records suggest an early rise in SOIW heat content, synchronous with SH temperatures, drove change in upper water column density gradients across the entire south Pacific [Bostock et al., 2004; Siani et al., 2013]. These gyre-wide de-stratification events are in phase with rising atmospheric CO2 concentrations and increasingly depleted atmospheric carbon isotope signatures [Monnin et al., 2001; Lourantau et al., 2010]. We suggest a decrease in stratification across the upper water column led to enhanced vertical mixing, which provided a pathway for carbon-rich intermediate waters to invade surface waters and exchange with the atmosphere during the YD and HS1. 30 The response of intermediate and mode waters to rapid global warming during the last deglaciation may provide a good basis for predicting their response to modern climate change. Both periods of rapid global warming are associated with shifts in the position and strength of the southern westerly winds. Today movement of the SWWs is associated with the polarity state of the Southern Annular Mode, which has shifted in response to ozone depletion and warming global temperatures over the last few decades. These shifts had significant effects on intermediate water formation rates and properties, perhaps comparable to those observed in the past [Naveira-Garabato et al., 2009; Thompson et al., 2011, Schmidtko and Johnson, 2012]; since 1925 AAIW core densities decreased resulting in an overall shoaling of the water mass by 30-50 dbar decade-1, with the larger shifts occurring towards the southernmost extent of AAIW [Schmidtko and Johnson, 2012]. These rates of warming within AAIW are unprecedented in the historical record [Schmidtko and Johnson, 2012]. Based on our observations from the deglaciation, we therefore predict reduced stratification and enhanced vertical mixing to occur, with the potential to alter the accumulation rate of carbon in SOIWs, key sinks for anthropogenic carbon today [Sabine et al., 2004; Schmidtko and Johnson, 2012]. 31 References Allen, K.A., Sikes, E.L., Hönisch, B., Elmore, A.C., Guilderson, T.P., Rosenthal, Y. and Anderson, R.F., (2015), Southwest Pacific deep water carbonate chemistry linked to high southern latitude climate and atmospheric CO2 during the last glacial termination, Quaternary Science Reviews (122), 180-191, doi:10.1016/j.quascirev.2015.05.007. Anderson, R. F., S. Ali, L. I. Bradtmiller, S. H. H. Nielson, M. Q. Fleisher, B. E. Anderson, and L. H. Burckle (2009), Wind-Driven Upwelling in the Southern Ocean and the Deglacial Rise in Atmospheric CO2, Science, 323 (5920), 1443-1448, doi:10.1126/science.1167441. Andreasen, D. H., A. C. Ravelo, and A. J. Broccoli (2001), Remote forcing at the last glacial maximum in the tropical Pacific Ocean, J Geophys Res-Oceans, 106(C1), 879- 897, doi:10.1029/1999JC000087. Barker, S., Diz, P., Vautravers, M.J., Pike, J., Knorr, G., Hall, I.R. and Broecker, W.S., (2009), Interhemispheric Atlantic seesaw response during the last deglaciation, Nature, 457(7233), 1097-1102, doi:10.1038/nature07770. Bemis, B. E., H. J. Spero, and R. C. Thunell (2002), Using species-specific paleotemperature equations with foraminifera: a case study in the Southern California Bight, Mar Micropaleontol, 46(3-4), 405-430, doi:10.1016/S0377-8398(02)00083-X. Boccaletti, G., Pacanowski, R.C., Philander, S.G.H., Fedorov, A.V., (2004). The thermal structure of the upper ocean. J Phys Oceanogr 34, 888-902, doi:10.1175/1520- 0485(2004)034<0888:TTSOTU>2.0.CO;2. 32 Bostock, H. C., B. N. Opdyke, M. K. Gagan, and L. K. Fifield (2004), Carbon isotope evidence for changes in Antarctic Intermediate Water circulation and ocean ventilation in the southwest Pacific during the last deglaciation, Paleoceanography, 19(4), doi:10.1029/2004PA001047. Bostock, H. C., B. N. Opdyke, and M. J. M. Williams (2010), Characterising the intermediate depth waters of the Pacific Ocean using δ13C and other geochemical tracers, Deep-Sea Res Pt I, 57(7), 847-859, doi:10.1016/j.dsr.2010.04.005. Bostock, H. C., P. J. Sutton, M. J. M. Williams, and B. N. Opdyke (2013), Reviewing the circulation and mixing of Antarctic Intermediate Water in the South Pacific using evidence from geochemical tracers and Argo float trajectories, Deep-Sea Res Pt I, 73, 84-98, doi:10.1016/j.dsr.2012.11.007. Caniupán, M., Lamy, F., Lange, C. B., Kaiser, J., Kilian, R., Baeza Urrea, O., Aracena, C., Hebbeln, D., Kissel, C., Laj, C., Mollenhauer, G., and Tiedemann, R., (2011), Millenial-scale sea surface temperature and Patagonian Ice Sheet changes off southernmost Chile (53°S) over the past ~60 kyr, Paleoceanography, 26, PA3221, doi:10.1029/2010PA002049. Calvo, E., Pelejero, C., Pena, L.D., Cacho, I. and Logan, G.A., (2011), Eastern Equatorial Pacific productivity and related-CO2 changes since the last glacial period, Proceedings of the National Academy of Sciences of the United States of America, 108(14), 5537-5541, doi:10.1073/pnas.1009761108. 33 Chaigneau, A., N. Dominguez, G. Eldin, L. Vasquez, R. Flores, C. Grados, and V. Echevin (2013), Near-coastal circulation in the Northern Humboldt Current System from shipboard ADCP data, J Geophys Res-Oceans, 118(10), 5251-5266, DOI: 10.1002/jgrc.20328. Clement, A.C., & Cane, M.A. (1999). A role for the tropical Pacific coupled ocean- atmosphere system on Milankovitch and millennial timescales part I: A modeling study of tropical Pacific Variability. In P.U. Clark, R.S. Webb, & L.D. Keigwin (Eds.), Mechanisms of Global Climate Change at Millennial Time Scales. Washington D. C.: American Geophysical Union, doi: 10.1029/GM112p0363. Conroy, J. L., K. M. Cobb, J. Lynch-Stieglitz, and P. J. Polissar (2014), Constraints on the salinity-oxygen isotope relationship in the central tropical Pacific Ocean, Mar Chem, 161, 26-33. Denton, G. H., R. F. Anderson, J. R. Toggweiler, R. L. Edwards, J. M. Schaefer, and A. E. Putnam (2010), The Last Glacial Termination, Science, 328(5986), 1652-1656, doi: 10.1126/science.1184119. Douville, E., Paterne, M., Cabioch, G., Louvat, P., Gaillardet, J., Juillet-Leclerc, A., and Ayliffe, L., (2010), Abrupt sea surface pH change at the end of the Younger Dryas in the central sub-equatorial Pacific inferred from boron isotope abundance in corals (Porites), Biogeosciences, 7, 2445-2459, doi:10.5194/bg-7-2445-2010. Downes, S. M., A. S. Budnick, J. L. Sarmiento, and R. Farneti (2011), Impacts of wind stress on the Antarctic Circumpolar Current fronts and associated subduction, Geophysical Research Letters, 38(11), L11605, doi:10.1029/2011GL047668. 34 Fairbanks, R. G., R. A. Mortlock, T. Chiu, L. Cao, A. Kaplan, T. P. Guilderson, T. W. Fairbanks, and A. L. Bloom (2005), Marine Radiocarbon Calibration Curve Spanning 0 to 50,000 Years B.P. Based on Paired 230Th/234U/238U and 14C Dates on Pristine Corals, Quaternary Science Reviews, 24, 1781-1796, doi:10.1016/j.quascirev.2005.04.007. Fairbanks, R. G., M. Sverdlove, R. Free, P. H. Wiebe, and A. W. H. Be (1982), Vertical- Distribution and Isotopic Fractionation of Living Planktonic-Foraminifera from the Panama Basin, Nature, 298(5877), 841-844. Fiedler, P. C., and L. D. Talley (2006), Hydrography of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 143-180, doi:10.1016/j.pocean.2006.03.008. Friedrich, T., and A. Timmermann (2012), Millennial-scale glacial meltwater pulses and their effect on the spatiotemporal benthic d18O variability, Paleoceanography, 27(3), PA3215, doi:10.1029/2012PA002330. Gersonde, R., X. Crosta, A. Abelmann, and L. Armand (2005), Sea-surface temperature and sea ice distribution of the Southern Ocean at the EPILOG Last Glacial Maximum - A circum-Antarctic view based on siliceous microfossil records, Quaternary Science Reviews, 24(7-9), 869-896, doi:10.1016/j.quascirev.2004.07.015. Grootes, P. M., M. Stuiver, J. W. C. White, S. Johnsen, and J. Jouzel (1993), Comparison of Oxygen-Isotope Records from the GISP2 and GRIP Greenland Ice Cores, Nature, 366(6455), 552-554, doi:10.1038/366552a0. 35 Gu, D., Philander, G., (1997). Interdecadal Climate Fluctuations That Depend on Exchanges Between the Tropics and Extratropics. Science 275(5301), 805-807, DOI:10.1126/science.275.5301.805. Hain, M.P., Sigman, D.M. and Haug, G.H., (2014), Distinct roles of the Southern Ocean and North Atlantic in the deglacial atmospheric radiocarbon decline, Earth and Planetary Science Letters, 394(0), 198-208, doi:10.1016/j.epsl.2014.03.020. Hartin, C. A., R. A. Fine, B. M. Sloyan, L. D. Talley, T. K. Chereskin, and J. Happell (2011), Formation rates of Subantarctic mode water and Antarctic intermediate water within the South Pacific, Deep-Sea Res Pt I, 58(5), 524-534, doi:10.1016/j.dsr.2011.02.010. Hastenrath, S. (1980), Heat Budget of Tropical Ocean and Atmosphere, J Phys Oceanogr, 10(2), 159-170, doi:10.1175/1520- 0485(1980)010<0159:HBOTOZ>2.0.CO;2. Herbert, T. D., J. D. Schuffert, D. Thomas, C. Lange, A. Weinheimer, A. Peleo-Alampay, and J. C. Herguera (1998), Depth and seasonality of alkenone production along the California margin inferred from a core top transect, Paleoceanography, 13(3), 263- 271, doi:10.1029/98PA00069. Herraiz-Borreguero, L., and S. R. Rintoul (2011), Subantarctic mode water: distribution and circulation, Ocean Dynam, 61(1), 103-126, doi:10.1007/s10236-010-0352-9. Ho, S. L., Mollenhauer, G., Lamy, F., Martinez-Garcia, A., Mohtadi, M., Gersonde, R., Hebbeln, D., Nunez-Ricardo, S., Rosell-Mele, A., and Tiedemann, R., (2012), Sea surface temperature variability in the Pacific sector of the Southern Ocean over the past 700 kyr, Paleoceanography, 27, PA4202, doi:10.1029/2012PA002317. 36 Holte, J.W., Talley, L.D., Chereskin, T.K. and Sloyan, B.M., (2012), The role of air-sea fluxes in Subantarctic Mode Water formation, Journal of Geophysical Research, 117(C3), C03040 doi:10.1029/2011JC007798. Holte, J.W., Talley, L.D., Chereskin, T.K. and Sloyan, B.M., (2013), Subantarctic mode water in the southeast Pacific: Effect of exchange across the Subantarctic Front. Journal of Geophysical Research: Oceans, 118(4), 2052-2066, doi:10.1002/jgrc.20144. Imbrie, J., et al. (1993), On the Structure and Origin of Major Glaciation Cycles 2. The 100,000-Year Cycle, Paleoceanography, 8(6), 699-735, doi:10.1029/93PA02751. Jouzel, J., et al. (2007), Orbital and millennial Antarctic climate variability over the past 800,000 years, Science, 317(5839), 793-796, doi:10.1126/science.1141038. Kalansky, J., Rosenthal, Y., Herbert, T., Bova, S., Altabet, M. (2015), Southern Ocean contributions to the Eastern Equatorial Pacific heat content during the Holocene, Earth and Planetary Science Letters, 424, 158-167, doi:10.1016/j.epsl.2015.05.013. Keeling, R. F., and B. B. Stephens (2001), Antarctic sea ice and the control of Pleistocene climate instability, Paleoceanography, 16(1), 112-131, doi:10.1029/2000PA000529. Kessler, W. S. (2006), The circulation of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 181-217, doi:10.106/j.pocean.2006.03.009. Kienast, M., S. S. Kienast, S. E. Calvert, T. I. Eglinton, G. Mollenhauer, R. Francois, and A. C. Mix (2006), Eastern Pacific cooling and Atlantic overturning circulation during the last deglaciation, Nature, 443(7113), 846-849, doi:10.1038/nature05222. 37 Kienast, M., G. MacIntyre, N. Dubois, S. Higginson, C. Normandeau, C. Chazen, and T. D. Herbert (2012), Alkenone unsaturation in surface sediments from the eastern equatorial Pacific: Implications for SST reconstructions, Paleoceanography, 27(1), PA1210, doi:10.1029/2011PA002254. Koutavas, A., and J. Lynch-Stieglitz (2003), Glacial-interglacial dynamics of the eastern equatorial Pacific cold tongue Intertropical Convergence Zone system reconstructed from oxygen isotope records, Paleoceanography, 18(4), 1089, doi:10.1029/2003PA000894. Koutavas, A., and J. P. Sachs (2008), Northern timing of deglaciation in the eastern equatorial Pacific from alkenone paleothermometry, Paleoceanography, 23(4), PA4205, doi: 10.1029/2008PA001593 Kubota, K., Yokoyama, Y., Ishikawa, T., Obrochta, S., and Suzuki, A., (2014), Larger CO2 source at the equatorial Pacific during the last deglaciation, Nature Scientific Reports, 4, doi:10.1038/srep05261. Lamy, F., Kaiser, J., Arz, H. W., Hebbeln, D., Ninnemann, U., Timm, O., Timmermann, A., Toggweiler, J. R., (2007), Modulation of the bipolar see saw in the Southeast Pacific during Termination 1, Earth and Planetary Science Letters, 259, 400-413, doi: 10.1016/j.epsl.2007.04.040. Lamy, F., Kilian, R., Arz, H. W., Francois, J., Kaiser, J., Matthias, P., Steinke, T., (2010), Holocene changes in the position and intensity of the southern westerly wind belt, Nature Geoscience, 3, 695-699, doi:10.1038/ngeo959. 38 Lavín, M. F., P. C. Fiedler, J. A. Amador, L. T. Ballance, J. Färber-Lorda, and A. M. Mestas-Nuñez (2006), A review of eastern tropical Pacific oceanography: Summary, Progress In Oceanography, 69(2-4), 391-398, doi:10.1016/j.pocean.2006.03.005. Lea, D. W., D. K. Pak, and H. J. Spero (2000), Climate impact of late quaternary equatorial Pacific sea surface temperature variations, Science, 289(5485), 1719-1724, doi: 10.1126/science.289.5485.1719. Levermann, A., J. Schewe, and M. Montoya (2007), Lack of bipolar see-saw in response to Southern Ocean wind reduction, Geophysical Research Letters, 34(12), L12711, doi: 10.1029/2007GL030255. Lisiecki, L. E., and P. A. Lisiecki (2002), Application of dynamic programming to the correlation of paleoclimate records, Paleoceanography, 17(4), 1049, doi:10.1029/2001PA000733. Liu, Z.Y., Philander, S.G.H., (1995). How Different Wind Stress Patterns Affect the Tropical Subtropical Circulations of the Upper Ocean, J Phys Oceanogr, 25, 449-462, doi.org/10.1175/1520-0485(1995)025<0449:HDWSPA>2.0.CO;2. Liu, Z. Y., and H. J. Yang (2003), Extratropical control of tropical climate, the atmospheric bridge and oceanic tunnel, Geophysical Research Letters, 30(5), 1230, doi:10.1029/2002GL016492. Lourantou, A., J. V. Lavric, P. Kohler, J. M. Barnola, D. Paillard, E. Michel, D. Raynaud, and J. Chappellaz (2010), Constraint of the CO2 rise by new atmospheric carbon isotopic measurements during the last deglaciation, Global Biogeochemical Cycles, 24(2), GB2015, doi:10.1029/2009GB003545. 39 Lu, P., McCreary, J.P., Klinger, B.A., (1998). Meridional circulation cells and the source waters of the Pacific Equatorial Undercurrent. J Phys Oceanogr 28, 62-84, doi:10.1175/1520-0485(1998)028<0062:MCCATS>2.0CO;2. Luyten, J. R., J. Pedlosky, and H. Stommel (1983), The Ventilated Thermocline, J Phys Oceanogr, 13(2), 292-309, doi.org/10.1175/1520- 0485(1983)013<0292:TVT>2.0.CO;2. Lynch-Stieglitz, J., W. B. Curry, and N. Slowey (1999a), Weaker Gulf Stream in the Florida straits during the last glacial maximum, Nature, 402(6762), 644-648, doi:10.1038/45204. Lynch-Stieglitz, J., W. B. Curry, and N. Slowey (1999b), A geostrophic transport estimate for the Florida Current from the oxygen isotope composition of benthic foraminifera, Paleoceanography, 14(3), 360-373, doi: 10.1029/1999PA900001. Marchitto, T. M., W. B. Curry, J. Lynch-Stieglitz, S. P. Bryan, K. M. Cobb, and D. C. Lund (2014), Improved oxygen isotope temperature calibrations for cosmopolitan benthic foraminifera, Geochimica et Cosmochimica Acta, 130(1), 1-11, doi:10.1016/j.gca.2013.12.034. Marcott, S. A., et al. (2014), Centennial-scale changes in the global carbon cycle during the last deglaciation, Nature, 514(7524), 616-619, doi:10.1038/nature13799. Martinez-Boti, M.A., Marino, G., Foster, G.L., Ziveri, P., Henehan, M.J., Rae, J.W.B., Mortyn, P.G. and Vance, D., (2015), Boron isotope evidence for oceanic carbon dioxide leakage during the last deglaciation, Nature, 518(7538), 219-222, doi:10.1038/nature14155. 40 Mashiotta, T. A., D. W. Lea, and H. J. Spero (1999), Glacial-interglacial changes in Subantarctic sea surface temperature and d18O-water using foraminiferal Mg, Earth and Planetary Science Letters, 170(4), 417-432, doi:10.1016/S0012-821X(99)00116- 8. McCartney, M. S. (1975), Subantarctic Mode Water, Eos T Am Geophys Un, 56(12), 1011-1011. McManus, J. F., R. Francois, J. M. Gherardi, L. D. Keigwin, and S. Brown-Leger (2004), Collapse and rapid resumption of Atlantic meridional circulation linked to deglacial climate changes, Nature, 428(6985), 834-837, doi:10.1038/nature02494. Monnin, E., A. Indermuhle, A. Dallenbach, J. Fluckiger, B. Stauffer, T. F. Stocker, D. Raynaud, and J. M. Barnola (2001), Atmospheric CO2 concentrations over the last glacial termination, Science, 291(5501), 112-114, DOI: 10.1126/science.291.5501.112. Moreno, P. I., R. Villa-Martinez, M. L. Cardenas, and E. A. Sagredo (2012), Deglacial changes of the southern margin of the southern westerly winds revealed by terrestrial records from SW Patagonia (52°S), Quaternary Science Reviews, 41, 1-21, doi:10.1016/j.quascirev.2012.02.002. Muller, P. J., G. Kirst, G. Ruhland, I. von Storch, and A. Rosell-Mele (1998), Calibration of the alkenone paleotemperature index UK’37 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, doi:10.1016/S0016-7037(98)00097-0. Naveira-Garabato, A. C. N., L. Jullion, D. P. Stevens, K. J. Heywood, and B. A. King 41 (2009), Variability of Subantarctic Mode Water and Antarctic Intermediate Water in the Drake Passage during the Late-Twentieth and Early-Twenty-First Centuries, Journal of Climate, 22, 3661-3688, doi:10.1175/2009JCLI2621.1. Ninnemann, U. S., and C. D. Charles (2002), Changes in the mode of Southern Ocean circulation over the last glacial cycle revealed by foraminiferal stable isotopic variability, Earth and Planetary Science Letters, 201(2), 383-396, doi:10.1016/S0012- 821X(02)00708-2. Oke, P. R., and M. H. England (2004), Oceanic response to changes in the latitude of the Southern Hemisphere subpolar westerly winds, Journal of Climate, 17(5), 1040-1054, doi:10.1175/1520-0442(2004)017. Pahnke, K., J. P. Sachs, L. Keigwin, A. Timmermann, and S. P. Xie (2007), Eastern tropical Pacific hydrologic changes during the past 27,000 years from D/H ratios in alkenones, Paleoceanography, 22(4), PA4214, doi:10.1029/2007PA001468. Pahnke, K., and R. Zahn (2005), Southern Hemisphere Water Mass Conversion Linked with North Atlantic Climate Variability, Science, 307(5716), 1741-1746, DOI: 10.1126/science.1102163. Palmer, M. R. and P. N. Pearson (2003), A 23,000-year record of Surface Water pH and pCO2 in the Western Equatorial Pacific Ocean, Science, 300(5618), 480-482, doi:10.1126/science.1080796. Peltier, W. R., and R. G. Fairbanks (2006), Global glacial ice volume and Last Glacial Maximum duration from an extended Barbados sea level record, Quaternary Science Reviews, 25(23-24), 3322-3337, doi:10.1016/j.quascirev.2006.04.010. Pena, L. D., I. Cacho, P. Ferretti, and M. A. Hall (2008), El Nino-Southern Oscillation- 42 like variability during glacial terminations and interlatitudinal teleconnections, Paleoceanography, 23(3). DOI: 10.1029/2008PA001620. Pena, L. D., S. L. Goldstein, S. R. Hemming, K. M. Jones, E. Calvo, C. Pelejero, and I. Cacho (2013), Rapid changes in meridional advection of Southern Ocean intermediate waters to the tropical Pacific during the last 30 kyr, Earth and Planetary Science Letters, 368, (20-32), doi:10.1016/j.epsl.2013.02.028. Pennington, J. T., K. L. Mahoney, V. S. Kuwahara, D. D. Kolber, R. Calienes, and F. P. Chavez (2006), Primary production in the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 285-317, doi:10.1016/j.pocean.2006.03.012. Philander, S. G. H. (1985), El Niño and La Niña, J Atmos Sci, 42(23), 2652–2662, doi:10.1175/1520-0469(1985)042<2652:ENALN>2.0.CO;2. Philander, S. G., and A. V. Fedorov (2003), Role of tropics in changing the response to Milankovich forcing some three million years ago, Paleoceanography, 18(2), 1045, doi:10.1029/2002PA000837. Putnam, A.E., Schaefer, J.M., Denton, G.H., Barrell, D.J.A., Andersen, B.r.G., Koffman, T.N.B., Rowan, A.V., Finkel, R.C., Rood, D.H., Schwartz, R., Vandergoes, M.J., Plummer, M.A., Brocklehurst, S.H., Kelley, S.E. and Ladig, K.L., (2013), Warming and glacier recession in the Rakaia valley, Southern Alps of New Zealand, during Heinrich Stadial 1. Earth and Planetary Science Letters, 382(0), 98-110, doi:10.1016/j.epsl.2013.09.005. Qu, T. D., et al. (2009), Origin and Pathway of Equatorial 13 degrees C Water in the Pacific Identified by a Simulated Passive Tracer and Its Adjoint, Journal of Physical Oceanography, 39(8), 1836-1853. 43 Rella, S. F., R. Tada, K. Nagashima, M. Ikehara, T. Itaki, K. Ohkushi, T. Sakamoto, N. Harada, and M. Uchida (2012), Abrupt changes of intermediate water properties on the northeastern slope of the Bering Sea during the last glacial and deglacial period, Paleoceanography, 27. PA3203, doi:10.1029/2011PA002205. Rintoul, S. R., and M. H. England (2002), Ekman transport dominates local air-sea fluxes in driving variability of subantarctic mode water, J Phys Oceanogr, 32(5), 1308-1321, doi:10.1175/1520-0485(2002)032<1308:ETDLAS>2.0.CO;2. Rose, K. A., E. L. Sikes, T. P. Guilderson, P. Shane, T. M. Hill, R. Zahn, and H. J. Spero (2010), Upper-ocean-to-atmosphere radiocarbon offsets imply fast deglacial carbon dioxide release, Nature, 466(7310), 1093-1097, doi:10.1038/nature09288. Rosenthal, Y., D. W. Oppo, and B. K. Linsley (2003), The amplitude and phasing of climate change during the last deglaciation in the Sulu Sea, western equatorial Pacific, Geophysical Research Letters, 30(8), 1428, DOI: 10.1029/2002GL016612. Russell, J. L., K. W. Dixon, A. Gnanadesikan, R. J. Stouffer, and J. R. Toggweiler (2006), The Southern Hemisphere westerlies in a warming world: Propping open the door to the deep ocean, Journal of Climate, 19(24), 6382-6390, doi:10.1175/JCLI3984.1. Sabine, C. L., et al. (2004), The oceanic sink for anthropogenic CO2, Science, 305(5682), 367-371, doi: 10.1126/science.1097403. Sarmiento, J. L., N. Gruber, M. A. Brzezinski, and J. P. Dunne (2004), High-latitude controls of thermocline nutrients and low latitude biological productivity, Nature, 427(6969), 56-60, doi:10.1038/nature02127. 44 Schmidtko, S., and G. C. Johnson (2012), Multidecadal Warming and Shoaling of Antarctic Intermediate Water, Journal of Climate, 25(1), 207-221, doi:10.1175/JCLI- D-11-00021.1. Schmitt, J., Schneider, R., Elsig, J., Leuenberger, D., Lourantou, A., Chappellaz, J., from ice cores, Science, 336(6082), 711-714, doi:10.1126/science.1277161. Shackleton, N. J. (1974), Attainment of isotopic equilibrium between ocean water and the benthonic foraminifera genus Uvigerina: Isotopic changes in the ocean during the last glacial, Centre National de la Recherche Scientifique Colloques Internationaux, 219, 203-209. Shakun, J. D., P. U. Clark, F. He, S. A. Marcott, A. C. Mix, Z. Y. Liu, B. Otto-Bliesner, A. Schmittner, and E. Bard (2012), Global warming preceded by increasing carbon dioxide concentrations during the last deglaciation, Nature, 484(7392), 49-54, doi:10.1038/nature10915. Siani, G., E. Michel, R. De Pol-Holz, T. DeVries, F. Lamy, M. Carel, G. Isguder, F. Dewilde, and A. Lourantou (2013), Carbon isotope records reveal precise timing of enhanced Southern Ocean upwelling during the last deglaciation, Nat Commun, 4, doi:10.1038/ncomms3758. Sloyan, B. M., and S. R. Rintoul (2001), Circulation, renewal, and modification of Antarctic mode and intermediate water, J Phys Oceanogr, 31(4), 1005-1030, doi.org/10.1175/1520-0485(2001)031<1005:CRAMOA>2.0.CO;2. Spero, H. J., and D. W. Lea (2002), The cause of carbon isotope minimum events on glacial terminations, Science, 296(5567), 522-525, DOI: 10.1126/science.1069401. 45 Stern, J. V., and L. E. Lisiecki (2014), Termination 1 timing in radiocarbon-dated regional benthic d18O stacks, Paleoceanography, 29, 1127–1142, doi:10.1002/2014PA002700. Stott, L., J. Southon, A. Timmermann, and A. Koutavas (2009), Radiocarbon age anomaly at intermediate water depth in the Pacific Ocean during the last deglaciation, Paleoceanography, 24(2), PA2223, doi:10.1029/2008PA001690. Takahashi, T., et al. (2002), Global sea-air CO2 flux based on climatological surface ocean pCO2, and seasonal biological and temperature effects, Deep-Sea Res Pt Ii, 49(9-10), 1601-1622, doi:10.1016/S0967-0645(02)00003-6. Taylor, Wayne A. (2000), Change-Point Analysis: A Powerful New Tool for Detecting Change, WEB: http://www.variation.com/cpa/tech/changepoint.html. Thompson, D. W. J., S. Solomon, P. J. Kushner, M. H. England, K. M. Grise, and D. J. Karoly (2011), Signatures of the Antarctic ozone hole in Southern Hemisphere surface climate change, Nat Geosci, 4(11), 741-749. doi:10.1038/ngeo1296. Timmermann, A., J. Sachs, and O. E. Timm (2014), Assessing divergent SST behavior during the last 21 ka derived from alkenones and G. ruber-Mg/Ca in the equatorial Pacific, Paleoceanography, 29(6), 680-696, doi:10.1002/2013PA002598. Toggweiler, J. R. (2009), Shifting Westerlies, Science, 323(5920), 1434-1435, doi: 10.1126/science.1169823. Toggweiler, J. R., K. Dixon, and W. S. Broecker (1991), The Peru Upwelling and the Ventilation of the South Pacific Thermocline, Journal of Geophysical Research, 96(C11), 20467–20497, doi:10.1029/91JC02063. 46 Toggweiler, J. R., J. L. Russell, and S. R. Carson (2006), Midlatitude westerlies, atmospheric CO2, and climate change during the ice ages, Paleoceanography, 21(2), PA2005, doi:10.1029/2005PA001154. Wyrtki, K. (1966), Oceanography of the eastern equatorial Pacific Ocean, Oceanography and Marine Biology, An Annual Review, 4, 33-68. Zahn, R., K. Winn, and M. Sarnthein (1986), Benthic Foraminiferal δ13C and Accumulation Rates of Organic Carbon: Uvigerina Peregrina Group and Cibicidoides Wuellerstorfi, Paleoceanography, 1(1), 27-42, doi:10.1029/PA001i001p00027. 47 Figures: Figure 1. (a.) Key formation regions of AAIW and SAMW and their subsurface paths to the EEP. In the Pacific, AAIW forms primarily in the SE Pacific along the tip of South America, just northwest of the Drake Passage, with small contributions from west and east of New Zealand and the East Pacific Rise [Bostock et al., 2013]. SAMW forms along the Subantarctic Front (SAF) south of the Subtropical Convergence (STC). AAIW, SAMW, and small contributions from NPIW contribute to Equatorial thermostad waters. 48 Figure 2. Salinity of the EEP upper water column from 65°S to 10° N averaged along 87.5 °W [WOA 13]. AAIW is characterized by a core of low salinity water at intermediate depth, with psu values between 34.3 and 34.5. SAMW is saltier but warmer than AAIW. EqPIW is composed of a mixture of AAIW and Pacific Deep Water [Bostock et al., 2010, 2013]. Cores used in this study are plotted at their depth and latitude of recovery. The deepest core, CDH 26, at 1000 m lies within EqPIW, GGC43 and CDH41 lie along the boundary between the water masses, and CDH23 is bathed by SAMW. 49 Figure 3. Age models for the four cores used in this study. The age model for GGC 43 is based on 9 radiocarbon measurements of the planktonic foraminifera, Globigerinoides ruber, (filled squares) and 17 of 24 measurements from CDH41 (open circles). CDH 23 and CDH 26 are better constrained with 24 and 25 AMS 14C measurements, respectively, made on the thermocline dwelling foraminifer, Neogloboquadrina dutertrei. AMS 14C ages were converted to calendar ages using the calibration curve of Fairbanks et al. [2005]. A constant reservoir age of 500 years is assumed for all measurements. Seven AMS 14C measurements from CDH 41 were removed from age model calculations (filled circles). 50 1.5 a. 2 2.5 δ18O (‰, VPDB) 3 3.5 4 4.5 b. 0 δ13C (‰, VPDB) -0.5 -1 370 m -1.5 600 m 1000 m -2 0 5 10 15 20 25 Age (kyrs BP) Figure 4. Stable isotope records from the benthic foraminifer, Uvigerina peregrina, from three water depths, 370 m, 600 m, and 1000 m, in the EEP. (a.) Oxygen isotope values converge between 12.3 and 14.3 ka at 370 and 600 m. (b.) Carbon isotope values at 370, 600, and 1000 m also converge at this time. Carbon isotope minima occur between 12.7-14.2 and 15.6-22 ka. Because Uvigerina peregrina is an infaunal species the recorded values may in part reflect porewater chemistry and thus the accumulation rate of organic carbon [Zahn et al., 1986]. We therefore interpret these data with caution, noting that the excursions to large negative values likely reflect an influx of 13C depleted carbon to the EEP subsurface in intermediate water as well as pore water production of depleted 13C via high productivity and organic matter export to the sediments. 51 Figure 5. Comparison between temperature records from the EEP and high latitude records of global change during the last deglaciation. Ice core records from Greenland (top, GISP2) and Antarctica (bottom, EPICA Dome C) document the temperature evolution in the Northern and Southern Hemisphere, respectively [Grootes et al. 1993; 52 Jouzel et al. 2007]. The subsurface temperature record from 1000 m appears to have SH timing for deglacial warming. At 600 m, we do not observe a deglacial warming trend, which is likely explained by a mean shoaling of the SAMW/AAIW boundary to at least 600 m during SH warm events. At the sea surface, warming is delayed until 14.7 ka. 53 Figure 6. The stratification gradient anomaly was calculated as the offset between records from the sea surface, 370 m, 600 m, and 1000 m in δ18O, δ13C and temperature space. Subsurface temperature records were linearly interpolated to ages where there are δ18O and δ13C measurements in the lower resolution surface record from site V19-28. All records were subsequently normalized to a late Holocene value (2785 yrs ago). δ18O, δ13C, and thermal gradient anomalies represent an average of the normalized offset between the surface and subsurface records. The stratification gradient anomaly represents the average of these three gradient anomaly calculations. 54 55 Figure 7. Measures of stratification in (a) the EEP, (d), the northwest Tasman Sea, southwest Pacific [Bostock et al., 2004], and (e.) the southeast Pacific [Siani et al., 2013]. (b.) Surface ocean ΔpCO2 at ODP site 1238 in the EEP [Martinez-Boti et al., 2015]. (c.) Coccolithophore productivity inferred from alkenone abundances (C37total) at sites CDH 23 and CDH 26 along the northern Peru Margin [this study]. (e.) Atmospheric δ13C record from Antarctic ice cores [Schmitt et al., 2012] (f.) Atmospheric CO2 concentrations from Epica Dome C [Lourantau et al., 2010] and the WAIS Divide core [Marcott et al., 2014]. Stratification in the EEP decreased at the onset of the last glacial termination, 18 ka, synchronous with oceanic CO2 outgassing to the atmosphere in the EEP [Martinez-Boti et al., 2015], increasing coccolithophore productivity along the Peru Margin [this study] and rising atmospheric CO2 levels [Lourantau et al., 2010; Marcott et al., 2014]. Periods of reduced stratification at all Pacific sites are synchronous with each other, which suggests that reduced stratification created a pathway for isotopically light carbon to escape from the deep ocean to the atmosphere during HS1 and the YD. 56 Figure 8. Connections between the southern high latitudes and the EEP during (a.) the Holocene, (b.) the YD/HS1, and (c.) the LGM. AAIW formation regions are shown with red x’s. For each time slice, a 2-d north-south transect shows structural change in intermediate and mode water geometry. AAIW is thinner and deeper in the water column during the LGM and shoals and thickens during the YD/HS1 in response to changing sea ice extent, shifts in the position and strength of the SWWs, and rising atmospheric temperatures. 57 Supplementary Information a. Surface to 370 m 0 5 10 15 20 25 3 12 13 3.3 2.8 C Gradient Thermal Gradient 3.2 Gradient (‰, VPDB) 18O Thermal Gradient (°C) 11.5 18 O Gradient 2.6 Gradient (‰, VPDB) 3.1 11 2.4 3 10.5 2.9 2.2 13C 10 2.8 2 9.5 2.7 1.8 9 2.6 0 b. Surface to 600 m 3.2 16 3.8 15 3 3.6 Gradient (‰, VPDB) Thermal Gradient (°C) 14 18O 2.8 Gradient (‰, VPDB) 3.4 13 2.6 12 3.2 2.4 13C 13C 11 Gradient Thermal Gradient 3 2.2 10 18O Gradient 2 9 2.8 c. Surface to 1000 m 3.4 19 5 4.8 3.2 18 Gradient (‰, VPDB) 4.6 Thermal Gradient (°C) 18O 3 Gradient (‰, VPDB) 17 4.4 2.8 4.2 16 2.6 13 13C C Gradient 4 15 Thermal Gradient 2.4 3.8 18 O Gradient 2.2 14 3.6 0 5 10 15 20 25 Age (ka) Figure S1. The stratification gradient in δ18O, δ13C and temperature space was calculated as the offset between records from the sea surface, 370 m, 600 m, and 1000 m. Here we show the δ18O, δ13C, and temperature gradient between the surface records and each 58 subsurface record individually. (a.) δ18O, δ13C and thermal gradient between the surface and 370 m, (b.) δ18O, δ13C and thermal gradient between the surface and 600 m, (c.) δ18O, δ13C and thermal gradient between the surface and 1000 m. Calculated gradients to 370, 600, and 1000 m each exhibit reduced stratification during the deglaciation, with the smallest gradients occurring during the YD and HS1 (gray bars). 59 10 A. 8 C37total (nmol/g) 6 4 2 B. -0.5 δ13C (‰, VPDB) -1 -1.5 -2 0 5 10 15 20 25 Age (kyrs BP) Figure S2. (A.) Alkenone abundances (C37total) at sites CDH 23 and CDH 26 along the northern Peru Margin. (B.) Carbon isotopes measured on the benthic foraminifer Uvigerina peregrina at 1000 m water depth at site CDH 26. The C37total, a proxy for the rain rate of organic matter, is not correlated to subsurface δ13C values, which suggests overprinting by respiratory CO2 is not the primary control on Uvigerina δ13C values. Instead, advection of isotopically light DIC from the Southern Ocean was likely the primary driver of carbon isotopic change at this site during the last deglaciation. 60 CHAPTER TWO ________________________________________________________________________ Rapid variations in deep ocean temperature not unprecedented in the Holocene Samantha C. Bova, Timothy D. Herbert, Baylor Fox-Kemper Brown University, Department of Earth, Environmental, and Planetary Sciences, Institute at Brown for the Study of Environment and Society, Providence, RI USA In preparation for submission to Geophysical Research Letters 61 Abstract The observational record of deep-ocean variability is short, which makes it difficult to attribute the recent rise in deep-ocean temperatures to anthropogenic forcing. Here, we test a new proxy – the oxygen isotopic signature of individual benthic foraminifera – to detect rapid (i.e. monthly to decadal) variations in deep-ocean temperature and salinity in the sedimentary record. We apply this technique at 1000 m water depth in the Eastern Equatorial Pacific during seven 200-year Holocene intervals. Variability in foraminifer δ18O over the past 200 years is below the detection limit, but δ18O signatures from two mid-Holocene intervals indicate temperature swings >2°C within 200 years. Transport between the surface and deep ocean operating on human timescales or natural unforced variability, not active during the historical record, are potential explanations. Distinguishing externally forced climate trends in deep ocean properties from unforced variability should be possible with systematic analysis of suitable deep-sea cores. 1. Introduction On centennial and shorter timescales the deep ocean (>1000 m water depth) is often considered a passive observer of global change. Velocities in the deep ocean are slow, orders of magnitude lower than in the surface ocean, and eddies are also thought to be weaker at depth [Lozier, 2010]. As a result, most regions below 1000 m have not been ventilated, i.e., have not visited the sea surface to exchange heat, carbon, or other conserved properties with the atmosphere, for hundreds of years [Gebbie and Huybers, 2012]. Variability in stratification can be communicated more rapidly by waves or occur via hydrodynamic instabilities, neither of which require direct ventilation [Masuda et al., 62 2010; LaCasce and Pedlosky, 2004]. Recent hydrographic sections taken under the GO- SHIP program provide evidence for quick transport of surface anomalies to the deep ocean, with rising temperatures below 700 m on the order of 0.1°C/decade and accelerated warming over the past 15 years consistent with the recent slowdown in rising mean surface air temperatures [Purkey and Johnson, 2010]. Because the temperature rise in the deep ocean parallels the observed surface warming in many places, this suggests the it is capable of exchanging heat and freshwater on decadal to centennial timescales. Rates of warming in the deep ocean since 1999 are unprecedented in the historical record [Balmaseda et al., 2013], but the record is short (<135 yrs) [Hobbs and Willis, 2013] and sparse, and therefore provides potentially insufficient constraints on the natural variability of the deep ocean. Repeat hydrographic sections below 700 m are rare [Roemmich et al., 2012; Arbic et al., 2014], which limits the degree to which variability and trends can be associated with particular processes. As a result, spatial variability is often used to estimate temporal variability, but many processes that contribute to deep ocean variability prevent simple conversion of observed spatial variability to temporal variability [Arbic et al., 2014]. Sampling has improved since 2004 with the Argo float program, but the record is short and limited in depth. Here, we test a new tool to examine the natural variability of intermediate and deep ocean temperature change on short time scales beyond the historical record. Specifically, we measured the stable oxygen isotopic signature of individual benthic foraminifera from ocean sediments to reconstruct monthly to decadal-scale variability in deep ocean temperature and salinity. We apply this technique to two sediment cores, CDH-26 (03°59.16 S, 81°18.52 W) and MC-18A (03°59.20 S, 81°18.60 W), collected in 63 2009 from the eastern equatorial Pacific (EEP) at 1000 m water depth, approximately the boundary between deep and intermediate water masses in the eastern tropical Pacific (Fig. 1, supplemental information, Figure S1). Benthic foraminiferal shell chemistry therefore reflects the mixed influence of intermediate and deep ocean water masses; Equatorial Pacific Intermediate Water (EqPIW), a mixture of Antarctic Intermediate Water (AAIW) and Pacific Deep Water, bathes the core sites today. Flow is generally westward between 500 and 1500 m at 2 to 4°S and eastward at greater depths [Firing et al., 1998]. AAIW is the dominant component of EqPIW and is the main conveyor of heat and salt to the core site, ageing 750 years before reaching the EEP subsurface [McCartney, 1977; Sloyan and Rintoul, 2001; Bostock et al., 2010; 2013; Gebbie and Huybers, 2012; Trenberth and Fasullo, 2014]. Waters at the site therefore have not been at the sea surface to exchange heat and gas with the atmosphere since well before industrialization. Figure 1 indicates the level of variability near the study site as a function of depth and location. Interestingly, in this location the whole span of observed modern variability is too weak to detect, but two episodes of variability within the Holocene are sufficiently strong to detect. 2. Material and Methods The average oxygen isotopic signature measured on multiple, typically 2-4, benthic foraminiferal shells is an established proxy for the mean temperature and salinity of deep water masses through time [e.g. Shackleton, 1974]; the δ18O signature of shell carbonate is determined by (1) the δ18O of ambient seawater (ice volume and salinity) and (2) seawater temperature during shell growth. By measuring individuals rather than bulk samples, we exploit the short lifetime of benthic foraminifera, ~ 1 month, to 64 reconstruct the variability rather than the mean of deep ocean physical properties [Murray, 1991; Koutavas et al., 2006]. The benthic foraminifer, Uvigerina Peregrina, is abundant in both cores and is therefore used exclusively for this study. U. peregrina live in the sediment, inhabiting the top 2 cm of the sediment. Studies show U. peregrina calcifies in approximate equilibrium with the overlying seawater and provides reliable records of seawater δ18O and temperature through time [e.g. Shackleton, 1974; Fontaneir et al., 2006]. Stable isotopic signatures of live infaunal benthic foraminifers, including U. peregrina, from the Bay of Biscay indicate that microhabitat does not have a systematic affect on test δ18O but does affect test δ13C signatures [Fontaneir et al., 2006]. We therefore examine the relationship between test δ18O and δ13C; we find that foraminifers with anomalous δ13C values are not more likely to have anomalous δ18O values (Supplemental Information, Text S2). Seven approximately 200-yr intervals were chosen to assess deep water variability. These intervals span -56-153 yrs BP, 3190-3343 yrs BP, 3406-3580 yrs BP, 4008-4151 yrs BP, 6104-6284 yrs BP, 6910-7095 yrs BP and 7912-8095 yrs BP but will be named by midpoint age (Table 1). The corresponding core depth ranges span up to 20 cm in the multicore to just 6 cm in the late Holocene portion of CDH-26 based on calculated sedimentation rates from AMS 14C constrained chronologies (Supplemental Information, Figure S2) [Bova et al., 2015]. CDH-26 does not contain modern sediment. We therefore estimate variability over the past 200 years using sediment from MC-18A. Both cores exhibit high sedimentation rates, ~100 cm/kyrs, likely due to the nearby outflow of the Guayas River. Each interval was sampled at 1 cm resolution and wet- sieved to separate size fractions. Forty individual U. Peregrina were picked from the 65 300-355 µm size fraction uniformly across each interval and measured on a Finnigan MAT 252 isotope ratio mass spectrometer with a Kiel III carbonate device. Prior to analysis, all individuals were sonicated for 10 seconds in 50 µL of ethanol to remove unwanted material from shells and weighed individually. All foraminifera were photographed at 60x magnification except 20 individuals from the intervals spanning 3190-3343 yrs BP and 4008-4251 yrs BP (Supplemental Information, Figure S5). In addition to the sampled foraminifera, 102 standards of known oxygen and carbon isotopic values ranging in weights from 4 µg to 111 µg were run to assess machine linearity and sample size dependent precision (Supplemental Information, Figure S3). 4. Results The small size of individual foraminifers (16 µg - 87 µg) is problematic and reduces analytical precision relative to conventional analyses. As a consequence, the average analytical error for small samples is nearly 0.15‰, over double that for standard sized samples and gets larger (smaller) with decreasing (increasing) sample size. This is apparent in the small standard dataset (supplemental information, Figure S3). In total we discard 21 standards and 52 samples. For a discussion of the controls on data exclusion please see the Supplemental Information (Text S1). Once data quality controls are enforced we calculate the variance and range in each time interval and perform a Kolmogorov-Smirnov Test to determine whether δ18O distributions about the mean within each sampling interval are significantly more variable than a distribution of lab standards. Failure to reject the null hypothesis indicates that variability is explained by analytical noise alone. Each sample distribution was compared to 5000 randomly sampled distributions of standard measurements (with replacement). 66 Sample size was taken into account by only comparing sets of standards that are indistinguishable based on sample size (Supplemental Information, Text S1). Results from the size-distribution matched K-S tests indicate that only the sampling intervals at 4080 (p<0.01) and 7003 yrs BP (p<0.10) are distinguishable from the standard with confidence (Fig. 3). They are also distinguishable from the δ18O distribution from the past 200 years (p<0.01). The statistics for each interval and a lab standard appear in Table 2. 5. Causes of Foraminiferal δ 18O Variability High variability intervals are distinguished by a handful of anomalous δ18O values that create tails in their distributions towards either positive or negative values (Figure 2, 3). These anomalous δ18O signatures can be explained in one of three ways. First, individual benthic foraminifera may be inconsistent recorders of deep water temperature and salinity. If true, this challenges the reliability of countless benthic δ18O records used to infer paleo-temperatures and past global ice volume. However, good reproducibility of benthic δ18O records across the global oceans suggests random variability in individuals is unlikely [Lisiecki and Raymo, 2005]. Furthermore, five of the seven sampled intervals exhibit no statistically significant variability among foraminifer tests. In contrast, in the high variability sample windows 10 to 20% of the individuals lie outside of the distributions of the standard or quiescent 200 year interval measurements (Figure 2, 3). Secondly, anomalous values could also result from sediment mixing via downslope reworking or bioturbation. However, the sedimentology across high variability intervals does not support a role for either process as there is no observable evidence for turbidity flows or burrows from x-radiograph scans of the cores (Figure 4). 67 Carbon isotopes measured on anomalous individuals are not systematically offset from the rest of the population, which further suggests no downslope movement of foraminifers (Supplemental Information, Text S2; Figure S6, S7). Bioturbation is most likely to introduce spurious variability within low sedimentation rate intervals, but again we observe no correlation between sedimentation rate and foraminifer δ18O variability. Finally, the relatively high sedimentation rate, approximately 100 cm/kyrs, throughout the core inhibits bioturbation impacts; the anomalously heavy δ18O values observed 7003 yrs BP require organisms to mix deglacial foraminifers up thru more than 300 cm of sediment. The third and most likely explanation is that outlying δ18O values record episodes of anomalously warm or cold temperatures or influxes of fresh or saline water at 1000 m water depth. Interpreted under this framework, our data indicate greater temperature and salinity variability at 1000 m depth in the EEP 4080 yrs BP and 7003 yrs BP relative to today (Supplemental Information, Figure S4). δ18O fluctuations must exceed ~0.5‰ to give an attribution of variability by our method. Because our time intervals are short and sea level has remained relatively stable during the Holocene, we discount the effects of changing global ice volume on the δ18O signature of water and suggest that change in shell δ18O dominantly reflects temperature, with a 0.25‰ change in δ18O corresponding to ~1°C change in EqPIW [Shackleton, 1974; Fontaneir et al., 2006]. Thus, when attributed fully to temperature, variability within these two intervals provides evidence for temperature variations exceeding 2°C that last for a month or longer. These data therefore imply that modern warming at the site near 1000 m water depth (estimated to be >0.1°C/decade [Purkey and Johnson, 2010], or 2°C over 200 yr) is not unprecedented and 68 falls well within the range of natural variability. 6. Sources of Past and Present Deep Water Variability Periods of high natural deep ocean variability during the Holocene support strong, rapid transport of signals, whether natural or anthropogenic, by advection, wave or other mechanism, from the surface ocean to at least 1000 m water depth in the EEP. Our data thus provide evidence that the deep ocean is not isolated from surface forcing on short timescales, or that significantly more vigorous modes of internal variability are occasionally active. Although the mechanism responsible remains uncertain, our data help constrain the timescale (months to decades) and the magnitude (>2°C) of this deep ocean variability, which in turn limits the possible mechanisms. It is difficult to imagine that a shift of local watermasses would be sufficient to explain this signal. In the modern ocean (based on the World Ocean Atlas version 2, Locarnini et al., 2013), seawater with a mean temperature 2°C (roughly 0.5‰ δ18O equivalent) or more warmer at 1000 m depth occurs only in the Red Sea, Mediterranean Sea, and the outflows from these seas into the Eastern Indian Ocean and North Atlantic Ocean. Seawater with a mean temperature 2°C or more colder at 1000 m depth occurs only in the Southern Ocean south of 40°S, the Arctic Ocean, or the Bering, Japan, or Okhotsk Seas. Salinity deviations (based on the World Ocean Atlas version 2, Zweng et al. 2013) of 2 ppt (rough 0.5‰ δ18O equivalent) at this depth occur only in the Black Sea (fresher) or the Mediterranean and Red Seas (saltier). Significant changes to the global circulation can yield this mangitude of variations [Toggweiler et al., 1991; Kalansky et al., 2015; Bova et al., 2015], but not on such short timescales (centennial and faster). It is perhaps easier to imagine a vertical relocation of watermasses, but 2°C warmer water occurs near 660 m 69 depth, and 2°C colder water occurs near 1750m depth (indicated by the grids in Figure 1). Heaving density surfaces this distance would require internal waves nearly double the world's largest (Alford et al., 2015). Additional processes linked to deep water variability include: advection, the El Niño Southern Oscillation (ENSO) [Meehl et al., 2011; England et al., 2014], planetary waves and related instabilities [Johnson et al., 2007; Kawano et al., 2006], and mesoscale eddies [Chaigneau and Pizarro, 2005; Chaigneau et al., 2011; Le Bars et al., 2016]. However, the timescale and magnitude of intermediate and deep water variability forced by most of these mechanisms are also difficult to reconcile with modern and past observations. The advective timescale is too long [Masuda et al., 2010] and anomalies forced by planetary and internal waves are typically too small (<1°C) [Nakano and Suginohara, 2002]. Coastal variability associated with El Niño [Kessler, 2006], upwelling anomalies [Huyer et al., 1990], and their propagation as Kelvin waves [Brink et al., 1983] are of sufficient magnitude, but this variability occurs primarily within the shallowest 500 m in modern times. Numerous strong ENSO events and shifts in the PDO over the last six decades fail to produce sufficient anomalies at 1000 m depth (Figure 1). Over the past 62 years, the temperature at our study location varies by less than 1°C or just 0.25‰ when converted to shell carbonate δ18O (95°W-85°W, 0°-10°S, 1000-1050 m depth, Supplemental Information, Figure S4). Finally, though short-lived (<120 days), mesoscale eddies are abundant from the coast to about 600-800 km offshore [Hormazabal et al., 2004] and influence subsurface temperature and salinity by roughly the magnitude observed near Peru, but farther to the south [Chaigneau and Pizarro, 2005; Chaigneau et al., 2011; Stramma et al., 2013; 70 Colas et al., 2012]. Perhaps the anomalies observed here represent a relocation of such eddies to our core site. High-resolution modeling of seawater δ18O demonstrates that eddies can produce anomalies up to 0.25‰ that persist for months, but does not indicate why some centuries exhibit this variability and others do not at our location [Stevenson et al., 2015]. At 500 m water depth mesoscale eddy activity along the coast of Chile is linked to the ENSO cycle; eddy transport is strongest during La Niña or normal years when horizontal density gradients are at a maximum [Hormazabal et al., 2004]. Although quite a bit shallower in the water column than our study site, it is nevertheless interesting that maximum variability at 1000 m depth in the EEP is documented when ENSO activity was at a minimum at the surface [Koutavas et al., 2012; Carré et al., 2014]. Deep isotope measurements along the coast of Chile and in other regions of strong eddy heat and salt transport, such as the Antarctic Circumpolar Current or the Kuroshio and Gulf Stream extensions may help assess eddy-driven deep ocean variability. 7. Conclusions Detection of high variability intervals in the Holocene suggests the modern ocean is not inherently more variable than the pre-anthropogenic one. We find two 200-yr intervals during the Holocene with temperature swings greater than about 2°C (~0.5‰ in δ18Oforam space) that persist for a month or longer at intermediate depths in the EEP. Whether the detected variability in the temperature of EEP intermediate and deep water masses is merely a local phenomenon, reflecting only mesoscale flow over the rough local bathymetry, or more widespread, affecting regions far afield across both intermediate and deep water depths, remains unknown. Systematic studies of individual benthic foraminifera variability at locations throughout the global oceans will constrain 71 the geographic extent of variability and provide better statistics of deep ocean variability on sub-centennial timescales that may lead to an enhanced ability to understand, model and predict the role of the deep and intermediate ocean in future global change. Given significant, and widespread variability of the deep ocean on human timescales, new estimates of the global energy budget may include a more important deep-ocean reservoir for excess atmospheric CO2 and heat with potential consequences for surface temperatures over the coming decades. 72 References Alford, Matthew H., et al. "The formation and fate of internal waves in the South China Sea." Nature 521.7550 (2015): 65-69. Arbic, B. K., M. Müller, J. G. Richman, J. F. Shriver, A. J. Morten, R. B. Scott, G. Sérazin, and T. Penduff (2014), Geostrophic turbulence in the frequency–wavenumber domain: Eddy-driven low-frequency variability. Journal of Physical Oceanography, 44(8):2050–2069. Balmaseda, M. A., K. E. Trenberth, and E. Kallen (2013), Distinctive climate signals in reanalysis of global ocean heat content, Geophysical Research Letters, 40(9), 1754- 1759. Bostock, H. C., B. N. Opdyke, and M. J. M. Williams (2010), Characterising the intermediate depth waters of the Pacific Ocean using delta C-13 and other geochemical tracers, Deep-Sea Res Pt I, 57(7), 847-859. Bostock, H. C., P. J. Sutton, M. J. M. Williams, and B. N. Opdyke (2013), Reviewing the circulation and mixing of Antarctic Intermediate Water in the South Pacific using evidence from geochemical tracers and Argo float trajectories, Deep-Sea Res Pt I, 73, 84-98. Bova, S. C., T. Herbert, Y. Rosenthal, J. Kalansky, M. Altabet, C. Chazen, A. Mojarro, and J. Zech (2015), Links between eastern equatorial Pacific stratification and atmospheric CO2 rise during the last deglaciation, Paleoceanography, 30, doi:10.1002/2015PA002816. Brink, K. H., D. Halpern, A. Huyer, and R. L. Smith (1983), The Physical-Environment of the Peruvian Upwelling System, Progress In Oceanography, 12(3), 285-305. 73 Carré, M., J. P. Sachs, S. Purca, A. J. Schauer, P. Braconnot, R. A. Falcon, M. Julien, and D. Lavallee (2014), Holocene history of ENSO variance and asymmetry in the eastern tropical Pacific, Science, 345(6200), 1045-1048. Chaigneau, A., and O. Pizarro (2005), Eddy characteristics in the eastern South pacific, J Geophys Res-Oceans, 110(C6). Chaigneau, A., M. Le Texier, G. Eldin, C. Grados, and O. Pizarro (2011), Vertical structure of mesoscale eddies in the eastern South Pacific Ocean: A composite analysis from altimetry and Argo profiling floats, J Geophys Res-Oceans, 116. Colas, F., J. C. McWilliams, X. Capet, and J. Kurian (2012), Heat balance and eddies in the Peru-Chile current system, Clim Dynam, 39(1-2), 509-529. England, M. H., S. McGregor, P. Spence, G. A. Meehl, A. Timmermann, W. J. Cai, A. Sen Gupta, M. J. McPhaden, A. Purich, and A. Santoso (2014), Recent intensification of wind-driven circulation in the Pacific and the ongoing warming hiatus, Nat Clim Change, 4(3), 222-227. Firing, E., S. E. Wijffels, and P. Hacker (1998), Equatorial subthermocline currents across the Pacific, J Geophys Res-Oceans, 103(C10), 21413-21423. Fontanier, C., Machensen, A., Jorissen, F. J., Anschutz, P., Licari, L., and C. Griveaud, (2006), Stable oxygen and carbon isotopes of live benthic foraminifera from the Bay of Biscay: Microhatbitat and seasonal variability, Marine Micropaleontology, 58, 159- 183. Gebbie, G., and P. Huybers (2012), The Mean Age of Ocean Waters Inferred from Radiocarbon Observations: Sensitivity to Surface Sources and Accounting for Mixing Histories, J Phys Oceanogr, 42(2), 291-305. 74 Hobbs, W. R., and J. K. Willis (2013), Detection of an observed 135 year ocean temperature change from limited data, Geophysical Research Letters, 40(10), 2252- 2258. Hormazabal, S., G. Shaffer, and O. Leth (2004), Coastal transition zone off Chile, J Geophys Res-Oceans, 109(C1). Huyer, A., M. Knoll, T. Paluszkiewicz, and R. L. Smith (1991), The Peru Undercurrent - a Study in Variability, Deep-Sea Res, 38, S247-S271. Johnson, G. C., S. Mecking, B. M. Sloyan, and S. E. Wijffels (2007), Recent bottom water warming in the Pacific Ocean, Journal of Climate, 20(21), 5365-5375. Kalansky, J., Y. Rosenthal, T. Herbert, S. Bova, and M. Altabet (2015), Southern Ocean contributions to the Eastern Equatorial Pacific heat content during the Holocene, Earth and Planetary Science Letters, 424, 158-167. Kawano, T., M. Fukasawa, S. Kouketsu, H. Uchida, T. Doi, I. Kaneko, M. Aoyama, and W. Schneider (2006), Bottom water warming along the pathway of lower circumpolar deep water in the Pacific Ocean, Geophysical Research Letters, 33(23). Kessler, W. S. (2006), The circulation of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 181-217. Koutavas, A., and S. Joanides (2012), El Niño-Southern Oscillation extrema in the Holocene and Last Glacial Maximum, Paleoceanography, 27. Koutavas, A., P. B. deMenocal, G. C. Olive, and J. Lynch-Stieglitz (2006), Mid- Holocene El Niño–Southern Oscillation (ENSO) attenuation revealed by individual foraminifera in eastern tropical Pacific sediments, Geology, 34(12), 993. 75 LaCasce, J. H. and J. Pedlosky, (2004), The instability of Rossby basin modes and the oceanic eddy field. J. Phys. Oceanogr. (34), 743-769. Le Bars, D., J. P. Viebahn, and H. A. Dijkstra, (2016), A Southern Ocean mode of multidecadal variability, Geophys. Res. Lett., 43, 2101-2110, doi:10.1002/2016GL068177. Lisiecki, L. E., and M. E. Raymo (2005), A Pliocene-Pleistocene stack of 57 globally distributed benthic delta O-18 records, Paleoceanography, 20(1). Lozier, M.S. (2010), Deconstructing the conveyor belt. Science, 328(5985):1507–1511. Masuda, S., et al. (2010), Simulated Rapid Warming of Abyssal North Pacific Waters, Science, 329(5989), 319-322. McCartney, M. S. "Subantarctic mode water." A Voyage of Discovery: George Deacon 70th Anniversary Volume (1977): 103-119. Meehl, G. A., J. M. Arblaster, J. T. Fasullo, A. X. Hu, and K. E. Trenberth (2011), Model-based evidence of deep-ocean heat uptake during surface-temperature hiatus periods, Nat Clim Change, 1(7), 360-364. Murray, J. W. "Ecology and paleoecology of benthonic foraminifera.-397 p." (1991). Nakano, H., and N. Suginohara (2002), Importance of the eastern Indian Ocean for the abyssal Pacific, Journal of Geophysical Research, 107(C12). Purkey, S. G., and G. C. Johnson (2010), Warming of Global Abyssal and Deep Southern Ocean Waters between the 1990s and 2000s: Contributions to Global Heat and Sea Level Rise Budgets, Journal of Climate, 23(23), 6336-6351. 76 Roemmich, D., W. J. Gould, and J. Gilson (2012), 135 years of global ocean warming between the Challenger expedition and the Argo Programme, Nat Clim Change, 2(6), 425-428. Shackleton, N. J. (1974), Attainment of isotopic equilibrium between ocean water and the benthonic foraminifera genus Uvigerina: Isotopic changes in the ocean during the last glacial, Centre National de la Recherche Scientifique Colloques Internationaux(219), 203-209. Sloyan, B. M., and S. R. Rintoul (2001), Circulation, renewal, and modification of Antarctic mode and intermediate water, J Phys Oceanogr, 31(4), 1005-1030. Stevenson, S., B. S. Powell, M. A. Merrifield, K. M. Cobb, J. Nusbaumer, and D. Noone (2015), Characterizing seawater oxygen isotopic variability in a regional ocean modeling framework: Implications for coral proxy records, Paleoceanography, 30, doi:10.1002/2015PA002824. Stramma, Lothar, et al. "On the role of mesoscale eddies for the biological productivity and biogeochemistry in the eastern tropical Pacific Ocean off Peru." Biogeosciences (BG) 10.11 (2013): 7293-7306. Toggweiler, J. R., K. Dixon, and W. S. Broecker (1991), The Peru Upwelling and the Ventilation of the South Pacific Thermocline, Journal of Geophysical Research, 96(C11), 467-497. Trenberth, K. E., and Fasullo, J. T., (2014), Earth's Energy Imbalance, Journal of Climate, 27, 3129-3144. 77 Zweng, M.M, J.R. Reagan, J.I. Antonov, R.A. Locarnini, A.V. Mishonov, T.P. Boyer, H.E. Garcia, O.K. Baranova, D.R. Johnson, D.Seidov, M.M. Biddle, 2013. World Ocean Atlas 2013, Volume 2: Salinity. S. Levitus, Ed., A. Mishonov Technical Ed.; NOAA Atlas NESDIS 74, 39 pp. 78 Figures: Figure 1. Modern observations of temperature and variability near the CDH-26 and MC- 18A core site (yellow ball). Ocean mean surface temperature (white contours, 1°C intervals) and twice the standard deviation away from climatology of all observations (i.e., 95% confidence distance from mean) in °C (shading of curtains and mesh); on this scale isotopic variability detectable using individual foraminifers corresponds to values greater than 1°C, while null episodes are <1°C. The meshes lie on surfaces of potential temperature 2°C above, at 660 m, and below, at 1750 m, the mean temperature at the core site, indicating distance in depth required to achieve a 2°C warming or cooling. Details in the Supplemental Information (Text S2). 79 Figure 2. Benthic foraminiferal oxygen isotopes measured on Uvigerina peregrina. The mean δ18O for each 200 year interval is plotted in black. Individual δ18O data are plotted in blue/green with the highest density region of points colored blue. 80 Figure 3. δ18O distributions about the mean for each sample interval plotted with 5000 randomly sampled standard distributions (light blue). Distributions from 49, 3267, 3493, 4080, 6194, 7003, and 8004 yrs before present are not distinguishable from the standard distributions. Samples from 4080 and 7003 yrs BP have distributions that have larger 81 variability than the standards. In the 4080 yr interval the anomaly is a tail consisting of an excess of extreme light isotopic values when compared to the standard. An excess of heavy isotopic values distinguish variability during the interval centered around 7003 yrs BP from the standard. 82 MC 18A CDH 26 300 10 150 10 Depth (cm) 330 40 180 40 360 70 210 390 100 240 420 130 270 Figure 4. X-radiograph images of core MC-18A and CDH-26. Cores CDH-26 and MC18A were scanned on an ITRAX XRF Core Scanner at the UMASS X-Ray 83 Fluorescence Laboratory. Light grey indicates low density intervals while the dark grey intervals represent comparatively high density regions. High density intervals tend to have high sand content and may be formed by turbidite flows. Scans were completed at 1 mm resolution. Images were imported into Adobe Photoshop to visually enhance the grayscale density differences throughout the cores. 84 Table 1. Age range and error for each sampled interval Core Core Depth Range Upper Age Error Lower Error Age span Mean Age (cm) (yrs BP) (1 sd) Age (yrs BP) (1 sd) (Best estimate, yrs) (yrs BP) MC-18A 0 – 20 -56 2.6 153 25.6 208 49 CDH-26 70-76 3190 182 3343 175 153 3267 CDH-26 78.5-86.5 3406 172 3580 165 174 3493 CDH-26 103-110 4008 148 4151 142 143 4080 CDH-26 198-208 6104 76 6284 70 180 6194 CDH-26 239.5-250.5 6910 52 7095 48 185 7003 CDH-26 296-308 7912 36 8095 36 183 8004 * Years BP is referenced to the year 1950. Negative values therefore indicate years after 1950. Table 2. Variance and range of foraminiferal δ18O distributions during each of the seven 200 year intervals sampled and results of K-S tests indicating whether each interval distribution is distinguishable from the standard distribution with and without taking sample sizes into account. Interval Number of Std. Range (‰) K-S Test K-S Test Result (yrs BP) Individuals Deviation (result, confidence level) (result, confidence level) (‰) all standards bootstrapped and size weighted standards -56 – 153 38 0.076 0.41 Null Null 3190 – 3343 35 0.073 0.47 Null Null 3406 – 3580 29 0.096 0.43 Null Null 4008 – 4151 33 0.19 0.95 Distinguishable (99%) Distinguishable (99%) 6104 – 6384 33 0.053 0.22 Null Null 6910 – 7095 29 0.11 0.49 Distinguishable (90%) Distinguishable (90%) 7912 – 8095 37 0.11 0.74 Null Null Standard 83 0.010 0.45 85 Supplementary Information Text S1. Data Quality Controls Individual benthic foraminifera can be difficult to measure to the same level of precision as conventionally sized samples because they are smaller by mass and therefore produce less CO2. Problems arise because 18O is present at much lower concentrations relative to 16O and can therefore be difficult to measure accurately for small samples. As a result, mass spectrometers may exhibit a “linearity effect”, essentially a drift with sample size. Exceedingly small samples may also produce so little CO2 that gas flow in the capillary lines transitions from laminar to turbulent flow, resulting in isotopic fractionation. At small sample sizes it is difficult to balance the sample and reference gas pressures. During analyses of small samples the sample gas pressure was systematically lower than that of the standard, leading to a systematic offset towards light δ18O values when the sample gas signal fell below 800 mV on cycle 1 (Fig. S3). Determining the sample size transition point into turbulent flow and the linearity of your instrument is therefore important, particularly when the target signal is small. Prior to making any individual benthic measurements we therefore ran 102 standards ranging from 4 to 111 µg of calcium carbonate, each with a known oxygen isotopic value of -6.52 ± 0.06 ‰ to detect these biases. Results from this test are shown in Figure S3. Controlling for Turbulent Flow Symptoms of turbulent flow are detected in measurements made on less than ~800 mV of mass 44 CO2 (Supplementary Figure 3). We therefore eliminate all measurements made below this threshold. All samples analyzed on the IR-MS are measured 8 times. For a conventionally sized sample, all 8 cycles are averaged to 86 minimize random errors. For small samples, however, it may be advantageous to remove some measurement cycles because the abundance of mass 44 CO2 decreases with each cycle. In some cases, the initial measurement cycles are made when the mass 44 CO2 signal is greater than 800 mV and later measurements on less. As long as 5 of the 8 cycles have a CO2 mass 44 signal greater than 800 mV we remove just the cycles less than 800 mV and keep the remainder in the dataset. When more than 5 of the 8 cycles measured are less than 800 mV, however, we eliminate the sample entirely. Controlling for Random Errors After controlling for variability induced by turbulent flow in the IR-MS, we evaluate the data for random error during analysis. We control for random error by assessing the standard deviation of δ18O values measured across one sample’s 8 measurement cycles (or as few as 5 if we removed some of the cycles). We remove samples whose δ18O values exhibit a standard deviation greater than 0.08‰ from the dataset. Controlling for Linearity Effects A small linearity effect is detected in samples that produce between 800 and 2000 mV of mass 44 CO2, with smaller samples in this range shifted systematically towards heavier δ18O values. We do not explicitly correct the individual data for linearity effects when comparing variability between foraminiferal δ18O assemblages and the small standard distributions (Figure 3). Instead, K-S tests are performed between sample and standard δ18O distributions only if they are indistinguishable based on size. First, we randomly select a 29-37 member set of the 81 small standards with replacement to match the sample being considered in number (e.g. the 3300 yr sample of 34 individual 87 foraminifer measurements is compared to 34 standard measurements). A K-S test between the sample and standard size distributions must be passed at the 95% confidence level to proceed. If so, a second K-S test is performed to compare the standard and sample δ18O distributions (Figure 3). This procedure was repeated 5000 times. Text S2. Supplementary Results Carbon Isotopic Variability of Individual U. peregrina U. peregrina live in the sediment, inhabiting the top 2 cm of the sediment. Studies show U. peregrina calcify in approximate equilibrium with the overlying seawater and provide reliable records of seawater δ18O and temperature through time [e.g. Shackleton, 1974], but their δ13C signature is prone to overprinting by porewater processes (i.e. respiration) [McCorkle et al., 1990, 1997; Fontanier et al., 2006]. To evaluate the impact of porewater chemistry on shell stable isotope signatures we include the carbon isotopic values from the same shells used for oxygen isotope analysis here (Supplementary Figures 7 and 8). Carbon isotopic variance and oxygen isotopic variance have no significant correlation, and shells with anomalous δ18O values were not more likely to have anomalous δ13C values. All intervals exhibit δ13C distributions that are more variable than the lab standards (Supplementary Figure 8). Modern Observations The modern observations used to create Figure 1 and Supplementary Figure S5 were derived from the averaged decadal averages from 1955 to 2012 of the World Ocean Atlas [Locarnini et al., 2013] and all high-quality in situ temperature observations from the World Ocean Database [Boyer et al., 2013] respectively. The near-surface temperature and depth corresponding to surface-referenced potential temperatures 2°C 88 above (6.629°C) and 2°C below (2.629°C) were derived from the average of the decadal averages of the World Ocean Atlas. The standard deviations used for shading in Figure 1 are the standard deviations from all observations at each site away from the climatology. Supplementary Figure 5 is formed from binning all World Ocean Database observations within 25 m of the target depth from the National Ocean Database Center collection of hydrographic stations (n=177,152 observations at 0-50m, n=152,791 at 225- 275m, n=154,689 at 475-525m, n=25,296 at 1000-1050m, and n=23,875 at 1200- 1250m). Directly observed variations of surface values in δ18O from the LeGrande & Schmidt (2006) database (http://data.giss.nasa.gov/o18data) are similarly spread in magnitude to the measured spread of 100 yr sediment foraminifera ensembles. In this location, only surface observations of δ18O have been collected. Supplemental References Boyer, T.P., J. I. Antonov, O. K. Baranova, C. Coleman, H. E. Garcia, A. Grodsky, D. R. Johnson, R. A. Locarnini, A. V. Mishonov, T.D. O'Brien, C.R. Paver, J.R. Reagan, D. Seidov, I. V. Smolyar, and M. M. Zweng, (2013), World Ocean Database 2013, NOAA Atlas NESDIS 72, S. Levitus, Ed., A. Mishonov, Technical Ed.; Silver Spring, MD, 209 pp., doi.org/10.7289/V5NZ85MT. LeGrande, A. N. and G. A. Schmidt (2006), Global gridded data set of the oxygen isotopic composition in seawater, Geophysical Research Letters, 33, L12604, doi:10.1029/2006/GL026011. Locarnini, R. A., A. V. Mishonov, J. I. Antonov, T. P. Boyer, H. E. Garcia, O. K. 89 Baranova, M. M. Zweng, C. R. Paver, J. R. Reagan, D. R. Johnson, M. Hamilton, and D. Seidov (2013), World Ocean Atlas 2013, Volume 1: Temperature. S. Levitus, Ed., A. Mishonov Technical Ed.; NOAA Atlas NESDIS 73, 40 pp. McCorkle, D. C., B. H. Corliss, and C. A. Farnham (1997), Vertical distributions and stable isotopic compositions of live (stained) benthic foraminifera from the North Carolina and California continental margins, Deep-Sea Res Pt I, 44(6), 983-1024. McCorkle, D. C., L. D. Keigwin, B. H. Corliss, and S. R. Emerson (1990), The Influence of Microhabitats on the Carbon Isotopic Composition of Deep-Sea Benthic Foraminifera (Vol 5, Pg 161, 1990), Paleoceanography, 5(3), 295-295. 90 Figure S1. Site map showing locations of the sediment cores, CDH-26 and MC-18A, used in this study. 91 Figure S2. Age model for core CDH-26 and MC-18A based on 27 and 2 AMS 14C dates, respectively. 92 −6 −6.5 −7 −7.5 (‰, VPDB) −8 18O −8.5 −9 −9.5 −10 0 1000 2000 3000 4000 5000 6000 Mass 44 (mV) Figure S3. Oxygen isotopes of a lab standard as a function of the mass 44 CO2 voltage signature on cycle one of eight. Sample size increases with mass 44 CO2 voltage signals. The orange dashed lines indicates the true standard δ18O signature of -6.52‰. Between 800 and 3000 mV, δ18O signatures trend towards anomalously heavy values with decreasing mass 44 cycle 1 voltages as a result of linearity effects, which leads to an increase in δ18O signatures decrease by 0.06‰ per volt. Below 800 mV on cycle 1, the sample and reference gas pressures are not balanced and the sample gas transitions to turbulent flow before the reference gas. As a result, δ18O signatures become increasingly light at low mass 44 voltages. To avoid biases resulting from poor balance between the sample and reference gas at low sample size we eliminate any measurement on <800 mV at the mass 44 detector on cycle 1. 93 0 50th 25 th 5 th th Pe rce 75 ntil th e 95 200 400 Depth (m) 600 800 1000 1200 −0.8 −0.4 0 0.4 0.8 18 O Anomaly (‰) Supplementary S4. Modern observations [all dates up to 2012, from Boyer et al., 2013] located between 95°W and 85°W and 10°S and 0° and 100 m vertical bins are used to categorize variability in this region. Solid lines indicate the depth dependence in estimated δ18O deviation from the mean based on observed temperatures. Blue crosses and red circles indicate the same percentiles from the 4080 yr and 49 yr sediment foraminifera ensembles, respectively. 94 Figure S5. Images of individual foraminifera whose measured δ18O signatures were >±1σ from the mean δ18O value calculated for each sampling interval. *Images are unavailable for 1 foraminifer with a measured δ18O value more than 2σ from the mean δ18O value 4080 years BP and 5 images are unavailable for foraminifera more than 1σ 95 from the mean δ18O value 3267 years BP. These intervals were the first to be measured and we only began photographing the foraminifers after the first round of results. 96 Figure S6. Benthic foraminiferal carbon isotopic values measured on Uvigerina peregrina. The mean δ13C for each ~200 year interval is plotted in black. Individual δ13C data is plotted in blue/green with the highest density region of points colored blue. 97 Figure S7. δ13C distributions about the mean for each sample interval plotted with 5000 randomly sampled standard distributions (light blue). Distributions from all intervals are distinguishable from the standard distributions at high confidence, likely due to 98 overprinting by porewater processes that affect shell δ13C but not δ18O signatures [McCorkle et al., 1990, 1997; Fontanier et al., 2006]. 99 CHAPTER THREE ________________________________________________________________________ Radiocarbon evidence for storage and release of the glacial carbon reservoir in the equatorial Pacific Samantha C. Bova1, Timothy D. Herbert1, and Marc Altabet2 1 Brown University, Department of Earth, Environmental, and Planetary Sciences, Institute at Brown for the Study of Environment and Society, Providence, RI USA 2 University of Massachusetts, School for Marine Science and Technology, New Bedford, MA, USA In preparation for submission to Nature Communications 100 Abstract Atmospheric carbon dioxide rose by ~80 ppmv during the last deglaciation, which helped propel Earth’s climate into the present interglacial. Identifying the source and release pathway of this carbon is critical for understanding glacial-interglacial feedbacks within the climate system. The prevailing hypothesis suggests carbon accumulated in the deep Southern Ocean, becoming increasing 14C-depleted during the last glacial period, and was released into southern ocean intermediates waters (SOIWs) and, subsequently, the atmosphere during deglaciation. In this paper, we present new radiocarbon records from three intermediate depths in the Eastern Equatorial Pacific that challenge aspects of this paradigm. We find that SOIWs were not the primary conduit for transporting upwelled old carbon from the Southern Ocean to the atmosphere. Instead, aged carbon was isolated in both the deep north and deep south Pacific and injected directly into the intermediate depth equatorial Pacific. 1. Introduction Atmospheric carbon dioxide (CO2) levels are an excellent indicator of Earth’s climate state. Ice core records from Antarctica demonstrate a tight coupling between global temperatures and atmospheric CO2 over the past 420,000 years, with shifts in atmospheric CO2 accounting for close to half of the radiative forcing required to move between glacial and interglacial states [Weaver et al., 1998; Yoshimori et al., 2001]. Since these changes were first observed nearly 40 years ago [Berner et al., 1979], studies of oceanic radiocarbon have demonstrated that the deep ocean is capable of storing and releasing enough carbon to explain glacial-interglacial CO2 changes [Sarnthein et al., 2013; Skinner et al., 2015]. The requirements for carbon storage within the deep ocean 101 are twofold: (1) an influx of organic matter from the surface ocean and (2) isolation of deep ocean waters from the surface ocean to prevent gas exchange with the atmosphere. Evidence for these conditions should be evident in the proxy record as large chemical gradients between surface and deep waters and the build-up of old, 14C-depleted, nutrient- rich, and oxygen poor waters in the subsurface. Aged deep water masses have been detected during the Last Glaical Maximum (LGM) at numerous sites throughout the global ocean [Keigwin, 2004; Robinson et al., 2005; Galbraith et al., 2007; Skinner et al., 2004; 2010; 2014; 2015; Lund et al., 2011; Okazaki et al., 2012; Cook and Keigwin, 2015; Ronge et al., 2016; Sikes et al., 2016]. LGM radiocarbon signatures from intermediate and deep water masses in the southwestern and south central Pacific provide evidence for aged carbon within Pacific Deep Water (PDW) [Ronge et al., 2016]. Radiocarbon signatures within PDW in the North Pacific are depleted as well, suggesting widespread reduction in the ventilation of PDW relative to today [Galbraith et al., 2007; Lund et al., 2011; Cook and Keigwin, 2015]. Occurrences of old glacial deep waters are also observed in the South Atlantic and North Atlantic [Skinner et al., 2004; 2010; 2014], with moderately aged deep water in the Drake Passage [Burke and Robinson, 2012; Chen et al., 2015]. Although the magnitude of the 14C anomaly in the deep ocean is not consistent at all sites, glacial water mass ventilation ages are consistently older at depth in each basin than those observed today, supporting the presence of a deep glacial carbon reservoir. During the last deglaciation, the aged deep carbon pool disappeared, likely fueling the two-part rise in atmospheric CO2 [Monnin et al., 2004; Skinner et al., 2010; 2014; Marcott et al., 2014; Ronge et al., 2016; Sikes et al., 2016]. Enhanced upwelling around 102 the Antarctic continent triggered by poleward shifts in the position of the westerly winds, reduced sea-ice extent, salinity driven buoyancy differences in Southern Ocean, or some combination of the three [e.g. Keeling et al., 2001; Toggweiler et al., 2006; Watson et al., 2015] reinvigorated deep ocean circulation, potentially ventilating aged, 14C-depleted carbon to the Southern Ocean surface. Because carbon equilibration with the atmosphere is slow, on the order of a year, the majority of upwelled carbon would have been re- sequestered in Antarctic Intermediate Water (AAIW) and spread throughout the shallow sub-surface oceans. Strongly depleted 14C-concentrations have been documented at four locations bathed by AAIW coincident with atmospheric CO2 rise during deglaciation, suggesting the water mass was an important conduit for the release of CO2 to the atmosphere [Marchitto et al., 2007; Lindsey et al., 2015; Bryan et al., 2010; Stott et al., 2009; Basak et al., 2010]. Efforts to corroborate these findings at sites close to the Antarctic Intermediate Water formation region, however, find little to no change in the water mass ventilation ages, with all those that exhibit signatures of old carbon located at the furthest reaches of AAIW, within the tropical Pacific and Indian Oceans (Fig. 1) [Kennett and Ingram, 1995; Marchitto et al., 2007; Stott et al., 2009; Bryan et al., 2010; De Pol-Holz et al., 2010; 2012; Magini et al., 2010; Rose and Sikes, 2010; Sorter et al., 2011; Burke and Robinson, 2012; Siani et al., 2013; Romahn et al., 2014]. If AAIW was the conduit for old carbon the most 14C-depleted signatures should be observed closest to the Southern Ocean source region. In this study, we evaluate the radiocarbon history of intermediate and surface water masses in the eastern equatorial Pacific over the past 25 kyrs. Three new paired 103 planktonic and benthic foraminiferal radiocarbon records from the Northern Peru Margin, at 373 and 1023 m water depth (CDH 23: 03°44.95’ S, 81°08.05’ W and CDH 26: 03°59.16’ S, 81°18.52’ W, respectively), and the Galapagos Platform, at ~600 m water depth (GGC 43: 01°15.13’S, 89°41.07’W, 617 m and CDH41: 01°15.94’S, 89°41.88’W, 595 m), provide a regional profile of water column radiocarbon variability from the tropical Pacific. We find that the radiocarbon signature of waters at 600 m depth on the Galapagos platform were strongly influenced by the deep Pacific carbon reservoir during both the LGM and the deglaciation. During the LGM, deep water 14C-concentrations, as recorded by benthic foraminifera, were strongly depleted and are comparable to signatures typically observed much deeper in the water column, while deglacial 14C trends align with intermediate water radiocarbon records from the Baja Peninsula [Marchitto et al., 2007; Lindsay et al., 2015] and the Arabian Sea [Bryan et al., 2010]. This dataset therefore supports a strong link between the cold, dense waters of the deep ocean and the warm, tropical waters of the Pacific and a new mechanism for the release of the glacial carbon reservoir outside the Southern Ocean. 2. Results 2.1 Core Chronology and Benthic-Planktonic age offsets We measured a total of fifty-seven paired planktonic and benthic 14C dates between the three EEP sites (Appendix C, Fig. 3). Today, all cores are bathed by Equatorial Pacific Intermediate Water (EqPIW), a mixture of Antarctic Intermediate Water (AAIW) with contributions from Pacific Deep Water (PDW) and Subantarctic Mode Water (Fig. 2) [Bostock et al., 2010; 2013]. Planktonic 14C dates, measured on Neogloboquadrina dutertrei (CDH 23 and CDH 26) and Globogerinoides Ruber (GGC 104 43 and CDH41), were collected as part of a previous study [Bova et al., 2015] and paired here with benthic 14C ages to calculate age offsets between surface and deep water masses. Benthic 14C dates were measured on mixed benthics (Bolivina spp., Hanzawaia concentrica, and Uvigerina spp.) at the Peru Margin sites and Uvigerina spp. at the Galapagos. Core chronologies were constructed using a polynomial fit to sets of planktonic AMS 14C dates using the Clam version 2.2 age-modeling software package for R (Supplementary Fig. 1) [Blaauw, 2010]. All 14C dates were converted to calendar ages using the marine13 radiocarbon age calibration curve [Reimer et al., 2013]. We modify all age models from the published chronologies of Bova et al. [2015] in order to characterize age model uncertainty using the Clam software. In addition, we remove 7 planktonic 14C dates from the chronology for cores GGC 43 and CDH 41 (617 and 595 m, respectively). In the Galapagos Platform cores, paired benthic and planktonic foraminiferal 14C ages converge approximately during the Younger Dryas (YD) and Heinrich Stadial 1 (HS1), possibly indicating enhanced mixing of 14C-depleted deep waters to the surface ocean and shifting surface 14C ages to artificially old ages (Fig. 3). We remove planktonic 14 C dates from these susceptible intervals (i.e. ~YD and HS1) and assume a constant 500 year reservoir age correction based on modern observations from the marine reservoir age database [Taylor and Berger, 1967; Druffel, 1981; Etayo-Cadvid et al., 2013]. The difference between benthic and planktonic 14C ages (B-P) typically reflects age since water mass ventilation. The B-P 14C ages from the Galapagos Platform indicate larger offsets during the last glacial period than during the Holocene by 2000 to 3500 105 years (Fig. 4b). B-P offsets decline during the deglaciation in two phases. They first decrease by 2200 years 18.4 ka. Then, after a brief rebound to near glacial values from 14.9 to 12.4 ka, they shift abruptly to just 150 yrs at the end of the deglaciation within 1.5 kyrs. Holocene B-P offsets at 600 m are relatively stable, averaging ~315 years. The records from 370 and 1000 m water depth along the northern Peru margin exhibit markedly different B-P ventilation age histories. At 1000 m water depth B-P ages are largest during the LGM, with an average offset of 1125 yrs, about half that observed between surface waters and water masses at 600 m water depth near the Galapagos, approximately 1000 km to the west. At the end of the deglaciation, B-P offsets decrease to 465 years during the early Holocene. 14C offsets between the surface and 373 m are stable, averaging 260 years over the entirety of the record. 2.2 Benthic Foraminiferal Δ14C Benthic foraminiferal Δ14C estimates the radiocarbon activity of the water mass in which the individuals grew. We calculate this value from the foraminiferal AMS 14C date and its estimated calendar age using the equation of Adkins and Boyle [1997], which accounts for decay since the foraminifera were alive. The calendar age is the largest source of uncertainty in this calculation. Patterns in Δ14C at all water depths are similar, but opposite in sign to trends in B- P, with reduced Δ14C values generally corresponding to larger B-P offsets (Fig. 4a). Holocene Δ14C values from all water depths track atmospheric Δ14C values, exhibiting a near constant 100‰ offset. Δ14C signatures at 600 and 1026 m water depth exhibit significantly larger offsets from the atmosphere during the deglaciation and LGM, on average ~402 and 254‰, respectively. The Galapagos record (600 m) exhibits the lowest 106 Δ14C values of the three sites, with minor enrichments occurring 16.8 ka before plunging back towards glacial values during the YD. At the end of the YD, Δ14C values at 600 m shift from a 518‰ offset from atmospheric Δ14C values to 91‰ offset in just over 1.5 kyrs. ΔΔ14C offsets at 600 and 1000 m water depth are largest during the LGM, shifting to modern values abruptly at the end of the YD interval. The record from 373 m differs from the deeper records in that its Δ14C values remain constant during the YD. 3. Discussion Radiocarbon records from within AAIW are valuable for constraining the mechanism responsible for glacial-interglacial atmospheric CO2 variability because the water mass is the hypothesized conduit between a deep ocean glacial carbon reservoir and the atmosphere. In this study, we present radiocarbon records of EqPIW, a mixture of predominantly AAIW with contributions from PDW, in the EEP. Two records from the far eastern equatorial Pacific, along the N. Peru Margin, from 370 and 1000 m exhibit little or no evidence for aged waters at the core sites during the LGM or deglaciation. Our Δ14C record from 600 m water depth on the Galapagos platform, however, is tightly coupled to intermediate depth Δ14C record from off the western coast of Baja California (23.5°N, 111.6°W) during the last deglaciation (Fig. 5) [Marchitto et al., 2007; Lindsay et al., 2015]. Like our sediment cores, EqPIW baths the Baja sediment cores used in the 2007 study [Bostock et al., 2010]. Δ14C records from Baja and the Galapagos are strongly depleted during the last deglaciation, with minimum Δ14C values between -200 and - 300‰ observed at both sites during the YD interval. During the LGM, however, the records diverge. In fact, our record from the intermediate depth Galapagos region is unlike any other published intermediate depth radiocarbon record, with Δ14C values in the 107 LGM comparable to some of the most 14C-depleted deep water masses documented at the time (Fig. 5). A synthesis of the presently available radiocarbon records spanning the past 30 kyrs, including those presented here, reveals characteristic radiocarbon histories by water mass in the Pacific Ocean (Fig. 5). Benthic foraminiferal radiocarbon records recovered from AAIW depths maintain a relatively constant 14C offset from the atmosphere over the past 30 kyrs [De Pol-Holz et al., 2010; Rose and Sikes, 2010; Sorter et al., 2011; Burke and Robinson, 2012; Siani et al., 2013]. The Peru Margin records (this study), from 373 and 1026 m, compare well with these records. In contrast, intermediate depth sites bathed by EqPIW, a mixture of primarily AAIW with contributions from PDW, exhibit Δ14C depletions of -200 to -300‰ during the last deglaciation [Marchitto et al., 2007; Bryan et al., 2010; Romahn et al., 2014]. North PDW (NPDW) Δ14C values were as low as -181‰ and 8‰ during the last deglaciation and LGM, respectively, while the 14C signature of South PDW (SPDW), as revealed in the Atlantic [Skinner et al., 2010] and Pacific [Skinner et al., 2015; Ronge et al., 2016; Sikes et al., 2016] sectors of the Southern Ocean, is still more depleted during the LGM. In fact, 14C concentrations are so low at some south Pacific locations (Δ14C reached -600‰ in the south Pacific) that hydrothermal CO2 inputs of 14C-free carbon to an already isolated reservoir are likely required given the apparent absence of widespread anoxia [Ronge et al., 2016]. Regardless of the ultimate CO2 source, these records indicate that 14C-depleted carbon accumulated within PDW during the last glacial period and was released into EqPIW during the last deglaciation. The 600 m core site on the Galapagos Platform appears to be uniquely positioned such that the dominant water mass at the core site switched from 108 PDW during the LGM to EqPIW during deglaciation, recording both the PDW glacial reservoir and its deglacial release. PDW is the dominant deep water mass in the EEP at depths greater than ~1200 m today but its signature can be traced much shallower in the water column [Bostock et al., 2010]. AAIW is the main component of EEP intermediate waters but the high salinity, nutrient concentrations, alkalinity, and DIC as well as the low oxygen and Δ14C signature of these waters require contributions from depth, most likely PDW [Bostock et al., 2010]. Diapycnal mixing of PDW to intermediate depths is not complete and a large Δ14C gradient (~100‰) is maintained between the water masses today [Key et al., 2015]. During the LGM, however, Δ14C signatures at 600 m water depth near the Galapagos Islands are within error of those observed in the heart of PDW, at 2921 m water depth, at site ODP 1240, and at five other PDW sites from the Atlantic and Pacific sectors of the Southern Ocean [Skinner et al., 2010; de la Fuente et al., 2015; Sikes et al., 2016] and the North Pacific [Galbraith et al., 2007; Lund et al., 2011; Cook and Keigwin, 2015], suggesting enhanced mixing of PDW to at least 600 m water depth near the Galapagos Islands and challenging the local presence of a bathyal front at 2000 m during glacial times [Herguera et al., 1992]. We suggest contributions from aged PDW can explain the sustained presence of a 14C-depleted water mass to at least 600 m on the Galapagos Platform during the LGM. Penetration of PDW to shallow subsurface depths was not ubiquitous in the EEP. Along the Peru Margin the modern Δ14C gradient between intermediate waters and PDW was maintained during the LGM (Fig. 5), which suggests that the aged water mass identified at 600 m on the Galapagos platform was localized to either a limited vertical 109 range, between 370 and 1000 m, and/or longitudinally constrained west of the continental margin. Water mass ages at 370 and 1000 m water depth along the margin are comparable to modern 14C based ventilation age calculations from the nearest GLODAP station (85.833°W, 1.999°S) over the duration of both records [Key et al., 2015]. The high regional heterogeneity is surprising given the proximity of the core sites but is likely related to the complexity of ocean currents in the region. Today there is no North Pacific source to thermocline waters east of the Galapagos, and a sharp boundary at the equator between South and North Pacific contributions is identified based on tritium and 3He sections from WOCE [Jenkins, 1996; Fiedler and Talley, 2006]. In addition, interaction of currents with the shallow, rough bathymetry surrounding the Galapagos Islands may enhance mixing, and can induce leeside formation of eddies, fronts, and localized upwelling possibly increasing the influence of deep water masses, specifically PDW, at relatively shallow depths in the vicinity of the islands [Heywood et al., 1990; Caldeira et al., 2002]. We therefore suggest that pulses of old carbon observed at intermediate depths near the Galapagos, Baja, and Oman coasts during the last deglaciation represent the release of aged, 14C-depleted carbon from PDW. Strongly reduced Δ14C signatures characteristic of PDW in the Atlantic and Pacific sectors of the Southern Ocean during the LGM increase beginning 18 ka and reach modern values by the end of HS1 [Skinner et al., 2010; 2015; Sikes et al., 2016; Ronge et al., 2016]. Some of the exceptionally aged carbon stored within PDW in the Southern Ocean may have escaped via Southern Ocean upwelling and transport by AAIW during HS1 as suggested by Marchitto et al., [2007] and others but the exceptionally aged signatures in the tropical oceans and the lack of an 110 aged signature along the flow path of AAIW out of the Southern Ocean necessitate direct release of a significant portion of the old carbon pool to intermediate depths in the tropics during HS1, fully depleting the Southern Ocean glacial carbon pool. Based on Δ14C records from the deep south Pacific, which constrain the deep carbon reservoir between approximately 2000 and 3500 m water depth [Ronge et al., 2016], we calculate that an increase in Δ14C activity of 400‰ over this portion of the water column corresponds to the release of 339 GT of carbon from SPDW during HS1 [Skinner et al., 2015, Supplementary Information Test S3]. Furthermore, in at least the vicinity of the Galapagos, 14C depleted carbon likely reached the surface ocean and atmosphere during HS1; B-P 14C offsets decrease by 2200 years during HS1, which supports enhanced mixing between surface and intermediate depth water masses relative to the LGM. By the end of HS1 the Southern Ocean carbon reservoir was fully depleted (Fig. 5) and a second, distinct carbon source for the YD rise in atmospheric CO2 is therefore required. Although no deep water radiocarbon signatures from regions outside the Southern Ocean are as exceptionally aged, PDW in the North Pacific was less well ventilated during the LGM than it is today; minimum Δ14C values are observed between 2 and 3.6 km in the North Pacific reaching values as low as -350‰ [Cook and Keigwin, 2015]. Furthermore, and of key interest here, isolation of NPDW persists through HS1 [Cook and Keigwin, 2015; Jaccard and Galbraith, 2013; Galbraith et al., 2007; Jaccard et al., 2009]. Authigenic uranium concentrations in Gulf of Alaska sediments from 3.2 and 3.6 km reveal sustained suboxic conditions until at least 14.6 ka [Galbraith et al., 2007; Jaccard et al., 2009; Jaccard and Galbraith, 2013], and B-P ventilation ages are largest during the LGM and HS1 (Fig. 6). 111 Proxy evidence suggests that ventilation of NPDW was delayed 3.4 kyrs relative to SPDW [Galbraith et al., 2007; Ronge et al., 2016] (Fig. 6); however, given that NPDW is the source of SPDW today both PDW endmembers have undergone synchronous ventilation changes if the modern circulation pattern was maintained. To explain the delay in NPDW ventilation we therefore invoke an equatorward shift in the formation region of PDW. Today, “new” PDW forms north of 40°N; Circumpolar Deep Water flows north at abyssal depths, upwells to the base of the lower thermocline via vertical diffusion, and flows back to the Southern Ocean. During the last glacial period we suggest PDW formation occurred further south, thereby isolating a deep water mass in the far north Pacific until 14.6 ka when the formation region shifted poleward, ventilating the deep N. Pacific poleward of 40°N (Fig 7). Although this hypothesis is inline with a large body of proxy data [Galbraith et al., 2007; Jaccard et al., 2009; Jaccard and Galbraith, 2013; Cook and Keigwin, 2015; Ronge et al., 2016], it still requires model simulations to test feasibility of the proposed circulation changes. Experiments that investigate the role of stratification and changing water mass geometries on PDW ventilation rates must therefore be the subject of future research. Galbraith et al., [2007] link an abrupt change in North Pacific ventilation to a similarly abrupt resumption of the North Atlantic overturning circulation, although the dynamical link between the two basins remains unclear Regardless of mechanism, the shift to enhanced deep North Pacific ventilation 14.6 ka occurs in phase with a sharp rise in atmospheric CO2 concentrations by 10 p.p.m.v. [Grootes et al., 1993; Marcott et al., 2014]. Spikes in subarctic productivity and export production co-occur with the onset of enhanced deep water ventilation, which 112 indicates surface productivity was able to nearly keep pace with enhanced upwelling of nutrients and CO2 [Galbraith et al., 2007]. As a result, three-fourths of the upwelled 14C- depleted carbon was re-sequestered at intermediate depths [Galbraith et al., 2007]. Intermediate depth radiocarbon records downstream, near Baja and within the EEP, reflect the influx of the re-sequestered but still 14C-depleted waters from the north; radiocarbon signatures trend towards more depleted values at the onset of the Bolling Allerod/ACR at both sites but was not able to reach the surface ocean or exchange with the atmosphere, at least in the EEP (Fig. 6) [Marchitto et al., 2007; Lindsay et al., 2015]. Galapagos B-P offsets indicate a return to near glacial levels of stratification and substantially reduced vertical mixing. The 30 p.p.m.v. rise in atmospheric CO2 during the YD interval reflects the subsequent release of re-sequestered North Pacific carbon under reduced stratification, and therefore, upwelling-favorable conditions. Between 17.2 and 14.1 ka the offset between NPDW Δ14C and atmospheric Δ14C dropped by 130‰, indicating the release of ~110 GT of carbon from NPDW at this time [Galbraith et al., 2007; Skinner et al., 2015, Supplementary Information Test S3]. Intermediate depth records from the Gulf of Alaska, close to the source, as well as the records from the tropical Pacific further afield (Baja and the Galapagos), find large increases in B-P ages coeval with the YD rise in CO2 suggestive of an influx of older waters to the shallow subsurface 12.5 to 12.0 ka [Davies-Walczak et al., 2014]. In the EEP, B-P offsets at the Galapagos site exhibit an abrupt shift from 2750 years to 150 in 1.5 kyrs, simultaneous to the rise in atmospheric CO2 [Monnin et al., 2004]. Although much smaller, B-P 14C offsets from 1000 m water depth along northern Peru Margin drop by 570 years. Stratification indices based on 113 δ18O, δ13C, and temperature gradients from the three sites also support a regional stratification minima during the YD [Bova et al., 2015], while boron isotopes provide direct evidence for CO2 outgassing (Fig. 8) [Martinez-Boti et al., 2015]. We therefore suggest that during the YD and HS1 reduced stratification in the EEP provided a “carbon window” for old carbon to reach the surface ocean and atmosphere. The radiocarbon records presented here cast doubt on the conventional view that AAIW was the primary conduit for the return of 14C depleted carbon to the atmosphere during the last deglaciation [e.g. Marchitto et al., 2007; Anderson et al., 2009]. Δ14C signatures at 600 m water depth on the Galapagos platform during the deglaciation are equivalent to those documented within PDW, the source of 14C-deplete carbon, which suggests minimal dilution and short transport to the EEP. Preservation of such low Δ14C signatures within AAIW along its circuitous route to the EEP, however, is not possible in model simulations with realistic circulations [Hain et al., 2011]. In addition, intermediate depth radiocarbon records from along the flow path of AAIW to the EEP find no evidence for an aged water mass during either phase of atmospheric CO2 rise during the last deglaciation [Pol-Holz et al., 2010; Burke and Robinson, 2012]. If sourced from the Southern Ocean, those records closest to the AAIW formation region should document the most extreme 14C depletions, much larger than what we observe further afield in the tropical Indo-Pacifc, but the opposite is observed. We therefore suggest that direct injection of old carbon from its source, PDW, directly into the tropical Pacific subsurface is the best explanation for the intermediate depth 14C anomalies detected along the Galapagos, Baja, and the Oman coasts during the two-part rise in atmospheric CO2 during the last deglaciation (Fig. 7). 114 4. Methods 4.1 Area of study Four sediment cores recovered from 3 intermediate water depths (373, 595-617, and 1023 m) in the EEP were recovered on the 2009 cruise of the R/V Knorr. These cores sample Equatorial Pacific Intermediate Water (EqPIW) a mixture of AAIW with contributions from PDW and SAMW [Fiedler and Talley, 2006; Bostock et al., 2010; 2013]. CDH 23 (03°44.95’ S, 81°08.05’ W, 373 m) and CDH 26 (03°59.16’ S, 81°18.52’ W, 1023 m), recovered along the Northern Peru Margin, record SAMW and EqPIW, respectively. At 600 m on the Galapagos Platform, GGC 43 (01°15.13’S, 89°41.07’W, 617 m) and CDH 41 (01°15.94’S, 89°41.88’W, 595 m) lie near the boundary between the two water masses. SAMW and AAIW, the main components of EqPIW, are sourced primarily from the Southern Ocean [Bostock et al., 2010; 2013]. SAMW forms within the Subantarctic Zone during late winter convective overturning and travels a circuitous route from the Southern Ocean into the western equatorial Pacific before finally flowing eastward across the Pacific into the EEP [McCartney, 1975; Toggweiler et al., 1991; Rintoul and England, 2002; Qu et al., 2009; Bostock et al., 2010; 2013]. EqPIW is composed primarily of Antarctic Intermediate water (AAIW) with some contributions from PDW. AAIW forms closer to the continent than SAMW, south of the subantarctic front, in three primary formation regions: (1) southwest and east of New Zealand, (2) west of the East Pacific Rise, and (3) west of the Drake Passage [Bostock et al., 2013; Herraiz- Borreguero and Rintoul, 2011]. Intermediate waters in the EEP subsurface are sourced 115 primarily from the southeast Pacific site west of the Drake Passage, but its geochemical signature is distinct. EqPIW is significantly more saline, nutrient-rich, lower in oxygen and exhibits an older Δ14C signature than Drake Passage AAIW, which requires mixing between AAIW and an older, more nutrient-rich waters mass, most likely PDW. North Pacific Intermediate Water is not an important source of intermediate water to the EEP today but some studies suggest a greater north Pacific influence in the past [e.g. Max et al., 2014] 4.2 Radiocarbon Measurements Paired planktonic and benthic radiocarbon measurements were made on four cores from the eastern equatorial Pacific. We use a splice between two cores from the Galapagos platform to extend the record from 600 m into the last glacial period [Bova et al., 2015]. In total, we made twenty-six paired measurements from the Galapagos cores, fifteen from CDH 23, and twelve from CDH 26. All radiocarbon dates were analyzed at the National Ocean Sciences Accelerator Mass Spectrometry facility at Woods Hole Oceanographic Institute. Planktonic dates were published previously by Bova et al., [2015] for age control. In this study, 1.4 to 7.9 mg of benthic foraminifers were picked from the >150 um size fraction (Appendix C). Particles and other debris were removed from each the sample with a fine hair brush before sonicating repeatedly in deionized water. 116 References Adkins, J. F., and E. A. Boyle (1997), Changing atmospheric Δ14C and the record of deep water paleoventilation ages, Paleoceanography, 12(3), 337-344. Anderson, R. F., S. Ali, L. I. Bradtmiller, S. H. H. Nielson, M. Q. Fleisher, B. E. Anderson, and L. H. Burckle (2009), Wind-Driven Upwelling in the Southern Ocean and the Deglacial Rise in Atmospheric CO2, Science, 323, 1443-1448. Basak, C., E. E. Martin, K. Horikawa, and T. M. Marchitto (2010), Southern Ocean source of superset of 14C-depleted carbon in the North Pacific Ocean during the last deglaciation, Nat Geosci, 3(11), 770-773. Berner, W., B. Stauffer, and H. Oeschger (1978), Past Atmospheric Composition and Climate, Gas Parameters Measured on Ice Cores, Nature, 276(5683), 53-55. Bostock, H. C., B. N. Opdyke, and M. J. M. Williams (2010), Characterising the intermediate depth waters of the Pacific Ocean using δ13C and other geochemical tracers, Deep-Sea Res Pt I, 57(7), 847-859. Bostock, H. C., P. J. Sutton, M. J. M. Williams, and B. N. Opdyke (2013), Reviewing the circulation and mixing of Antarctic Intermediate Water in the South Pacific using evidence from geochemical tracers and Argo float trajectories, Deep-Sea Res Pt I, 73, 84-98. Bova, S. C., T. Herbert, Y. Rosenthal, J. Kalansky, M. Altabet, C. Chazen, A. Mojarro, and J. Zech (2015), Links between eastern equatorial Pacific stratification and atmospheric CO2 rise during the last deglaciation, Paleoceanography, 30(11), 1407- 1424. Broecker, W. (2009), The Mysterious 14C Decline, Radiocarbon, 51, 1090119. 117 Bryan, S. P., T. M. Marchitto, and S. J. Lehman (2010), The release of 14C-depleted carbon from the deep ocean during the last deglaciation: Evidence from the Arabian Sea, Earth and Planetary Science Letters, 298(1-2), 244-254. Burke, A., and L. F. Robinson (2012), The Southern Ocean's role in carbon exchange during the last deglaciation, Science, 335(6068), 557-561. Butzin, M., M. Prange, and G. Lohmann (2005), Radiocarbon simulations for the glacial ocean: The effects of wind stress, Southern Ocean sea ice and Heinrich events, Earth and Planetary Science Letters, 235(1-2), 45-61. Caldeira, R. M. A., S. Groom, P. Miller, D. Pilgrim, and N. P. Nezlin (2002), Sea-surface signatures of the island mass effect phenomena around Madeira Island, Northeast Atlantic, Remote Sens Environ, 80(2), 336-360. Calvo, E., C. Pelejero, L. D. Pena, I. Cacho, and G. A. Logan (2011), Eastern Equatorial Pacific productivity and related-CO2 changes since the last glacial period, P Natl Acad Sci USA, 108(14), 5537-5541. Chen, T. Y., L. F. Robinson, A. Burke, J. Southon, P. Spooner, P. J. Morris, and H. C. Ng (2015), Synchronous centennial abrupt events in the ocean and atmosphere during the last deglaciation, Science, 349(6255), 1537-1541. Cleroux, C., P. Demenocal, and T. Guilderson (2011), Deglacial radiocarbon history of tropical Atlantic thermocline waters: absence of CO2 reservoir purging signal, Quaternary Science Reviews, 30(15-16), 1875-1882. Cook, M. S., and L. D. Keigwin (2015), Radiocarbon profiles of the NW Pacific from the LGM and deglaciation: Evaluating ventilation metrics and the effect of uncertain surface reservoir ages, Paleoceanography, 30(3), 174-195. 118 de la Fuente, M., L. Skinner, E. Calvo, C. Pelejero, and I. Cacho (2015), Increased reservoir ages and poorly ventilated deep waters inferred in the glacial Eastern Equatorial Pacific, Nat Commun, 6. De Pol-Holz, R., M. Mohtadi, and J. Southon (2012), Intermediate water radiocarbon off west Sumatra during the last 45,000 years, EGU General Assembly 2012, April 22-27, 2012, Vienna, Austria, 11829. De Pol-Holz, R., L. Keigwin, J. Southon, D. Hebbeln, and M. Mohtadi (2010), No signature of abyssal carbon in intermediate waters off Chile during deglaciation, Nat Geosci, 3(3), 192-195. DeVries, T., and F. Primeau (2011), Dynamically and Observationally Constrained Estimates of Water-Mass Distributions and Ages in the Global Ocean, J Phys Oceanogr, 41(12), 2381-2401. Druffel, E. M. (1981), Radiocarbon in Annual Coral Rings from the Eastern Tropical Pacific-Ocean, Geophysical Research Letters, 8(1), 59-62. Etayo-Cadavid, M. F., C. F. T. Andrus, K. B. Jones, G. W. L. Hodgins, D. H. Sandweiss, S. Uceda-Castillo, and J. Quilter (2013), Marine radiocarbon reservoir age variation in Donax obesulus shells from northern Peru: Late Holocene evidence for extended El Niño, Geology, 41(5), 599-602. Falkowski, P., et al. (2000), The global carbon cycle: A test of our knowledge of earth as a system, Science, 290(5490), 291-296. Fiedler, P. C., and L. D. Talley (2006), Hydrography of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 143-180. 119 Galbraith, E. D., S. L. Jaccard, T. F. Pedersen, D. M. Sigman, G. H. Haug, M. Cook, J. R. Southon, and R. Francois (2007), Carbon dioxide release from the North Pacific abyss during the last deglaciation, Nature, 449(7164), 890-U899. Grootes, P. M., M. Stuiver, J. W. C. White, S. Johnsen, and J. Jouzel (1993), Comparison of Oxygen-Isotope Records from the Gisp2 and Grip Greenland Ice Cores, Nature, 366(6455), 552-554. Hain, M. P., D. M. Sigman, and G. H. Haug (2011), Shortcomings of the isolated abyssal reservoir model for deglacial radiocarbon changes in the mid-depth Indo-Pacific Ocean, Geophysical Research Letters, 38. Herguera, J. C., E. Jansen, and W. H. Berger (1992), Evidence for a Bathyal Front at 2000-M Depth in the Glacial Pacific, Based on a Depth Transect on Ontong Java Plateau, Paleoceanography, 7(3), 273-288. Herraiz-Borreguero, L., and S. R. Rintoul (2011), Subantarctic mode water: distribution and circulation, Ocean Dynam, 61(1), 103-126. Heywood, K. J., E. D. Barton, and J. H. Simpson (1990), The Effects of Flow Disturbance by an Oceanic Island, J Mar Res, 48(1), 55-73. Jaccard, S., and E. D. Galbraith (2013), Direct ventilation of the North Pacific did not reach the deep ocean during the last deglaciation, Geophysical Research Letters, 40(1), 199-203. Jaccard, S. L., E. D. Galbraith, D. M. Sigman, G. H. Haug, R. Francois, T. F. Pedersen, P. Dulski, and H. R. Thierstein (2009), Subarctic Pacific evidence for a glacial deepening of the oceanic respired carbon pool, Earth and Planetary Science Letters, 277(1-2), 156-165. 120 Jenkins, W. J. (1996), Tritium and 3He in the WOCE Pacific Programme, International WOCE Newsletter(23), 6-9. Keeling, R. F., and B. B. Stephens (2001), Antarctic sea ice and the control of Pleistocene climate instability, Paleoceanography, 16(1), 112-131. Keigwin, L. D. (2004), Radiocarbon and stable isotope constraints on Last Glacial Maximum and Younger Dryas ventilation in the western North Atlantic, Paleoceanography, 19(4). Key, R. M. e. a. (2015), Global Ocean Data Analysis Project, Version 2. Koutavas, A., J. Lynch-Stieglitz, T. M. Marchitto, and J. P. Sachs (2002), El Nino-like pattern in ice age tropical Pacific sea surface temperature, Science, 297(5579), 226- 230. Lindsay, C. M., S. J. Lehman, T. M. Marchitto, and J. D. Ortiz (2015), The surface expression of radiocarbon anomalies near Baja California during deglaciation, Earth and Planetary Science Letters, 422, 67-74. Lund, D. C., A. C. Mix, and J. Southon (2011), Increased ventilation age of the deep northeast Pacific Ocean during the last deglaciation, Nat Geosci, 4(11), 771-774. Mangini, A., J. M. Godoy, M. L. Godoy, R. Kowsmann, G. M. Santos, M. Ruckelshausen, A. Schroeder-Ritzrau, and L. Wacker (2010), Deep sea corals off Brazil verify a poorly ventilated Southern Pacific Ocean during H2, H1 and the Younger Dryas, Earth and Planetary Science Letters, 293(3-4), 269-276. Marchitto, T. M., S. J. Lehman, J. D. Ortiz, J. Fluckiger, and A. van Geen (2007), Marine radiocarbon evidence for the mechanism of deglacial atmospheric CO2 rise, Science, 316(5830), 1456-1459. 121 Marcott, S. A., et al. (2014), Centennial-scale changes in the global carbon cycle during the last deglaciation, Nature, 514(7524), 616-+. Mariotti, V., D. Paillard, L. Bopp, D. M. Roche, and N. Bouttes (2016), A coupled model for carbon and radiocarbon evolution during the last deglaciation, Geophysical Research Letters, 43(3), 1306-1313. Martinez-Boti, M. A., G. Marino, G. L. Foster, P. Ziveri, M. J. Henehan, J. W. B. Rae, P. G. Mortyn, and D. Vance (2015), Boron isotope evidence for oceanic carbon dioxide leakage during the last deglaciation, Nature, 518(7538), 219-U154. Max, L., L. Lembke-Jene, J. R. Riethdorf, R. Tiedemann, D. Nurnberg, H. Kuhn, and A. Mackensen (2014), Pulses of enhanced North Pacific Intermediate Water ventilation from the Okhotsk Sea and Bering Sea during the last deglaciation, Clim Past, 10(2), 591-605. Mccartney, M. S. (1975), Subantarctic Mode Water, Eos T Am Geophys Un, 56(12), 1011-1011. Okazaki, Y., T. Sagawa, H. Asahi, K. Horikawa, and J. Onodera (2012), Ventilation changes in the western North Pacific since the last glacial period, Clim Past, 8(1), 17- 24. Qu, T., S. Gao, I. Fukumori, R. A. Fine, and E. J. Lindstrom (2008), Subduction of South Pacific waters, Geophysical Research Letters, 35(2). Rintoul, S. R., and M. H. England (2002), Ekman transport dominates local air-sea fluxes in driving variability of subantarctic mode water, J Phys Oceanogr, 32(5), 1308-1321. Robinson, L. F., J. F. Adkins, L. D. Keigwin, J. Southon, D. P. Fernandez, S. L. Wang, 122 and D. S. Scheirer (2005), Radiocarbon variability in the western North Atlantic during the last deglaciation, Science, 310(5753), 1469-1473. Romahn, S., A. Mackensen, J. Groeneveld, and J. Patzold (2014), Deglacial intermediate water reorganization: new evidence from the Indian Ocean, Clim Past, 10(1), 293- 303. Ronge, T. A., R. Tiedemann, F. Lamy, P. Kohler, B. V. Alloway, R. De Pol-Holz, K. Pahnke, J. Southon, and L. Wacker (2016), Radiocarbon constraints on the extent and evolution of the South Pacific glacial carbon pool, Nat Commun, 7. Rose, K. A., E. L. Sikes, T. P. Guilderson, P. Shane, T. M. Hill, R. Zahn, and H. J. Spero (2010), Upper-ocean-to-atmosphere radiocarbon offsets imply fast deglacial carbon dioxide release, Nature, 466(7310), 1093-1097. Siani, G., E. Michel, R. De Pol-Holz, T. DeVries, F. Lamy, M. Carel, G. Isguder, F. Dewilde, and A. Lourantou (2013), Carbon isotope records reveal precise timing of enhanced Southern Ocean upwelling during the last deglaciation, Nat Commun, 4. Sikes, E. L., M. S. Cook, and T. P. Guilderson (2016), Reduced deep ocean ventilation in the Southern Pacific Ocean during the last glaciation persisted into the deglaciation, Earth and Planetary Science Letters, 438, 130-138. Skinner, L., I. N. McCave, L. Carter, S. Fallon, A. E. Scrivner, and F. Primeau (2015), Reduced ventilation and enhanced magnitude of the deep Pacific carbon pool during the last glacial period, Earth and Planetary Science Letters, 411, 45-52. Skinner, L. C., and N. J. Shackleton (2004), Rapid transient changes in northeast Atlantic deep water ventilation age across Termination I, Paleoceanography, 19(2). 123 Skinner, L. C., C. Waelbroeck, A. E. Scrivner, and S. J. Fallon (2014), Radiocarbon evidence for alternating northern and southern sources of ventilation of the deep Atlantic carbon pool during the last deglaciation, P Natl Acad Sci USA, 111(15), 5480-5484. Skinner, L. C., S. Fallon, C. Waelbroeck, E. Michel, and S. Barker (2010), Ventilation of the Deep Southern Ocean and Deglacial CO2 Rise, Science, 328(5982), 1147-1151. Sortor, R. N., and D. C. Lund (2011), No evidence for a deglacial intermediate water Delta 14C anomaly in the SW Atlantic, Earth and Planetary Science Letters, 310(1-2), 65-72. Stott, L., J. Southon, A. Timmermann, and A. Koutavas (2009), Radiocarbon age anomaly at intermediate water depth in the Pacific Ocean during the last deglaciation, Paleoceanography, 24(2). Stuiver, M., and T. F. Braziunas (1993), Modeling Atmospheric 14C Influences and 14C Ages of Marine Samples to 10,000 Bc, Radiocarbon, 35(1), 137-189. Taylor, R. E., and R. Berger (1967), Radiocarbon content of marine shells from the Pacific coasts of Central and South America, Science, 158, 1180-1182. Toggweiler, J. R., K. Dixon, and W. S. Broecker (1991), The Peru Upwelling and the Ventilation of the South Pacific Thermocline, Journal of Geophysical Research, 96(C11), 467-497. Toggweiler, J. R., J. L. Russell, and S. R. Carson (2006), Midlatitude westerlies, atmospheric CO2, and climate change during the ice ages, Paleoceanography, 21(2). Watson, A. J., G. K. Vallis, and M. Nikurashin (2015), Southern Ocean buoyancy forcing of ocean ventilation and glacial atmospheric CO2, Nat Geosci, 8(11), 861-+. 124 Weaver, A. J., M. Eby, F. F. Augustus, and E. C. Wiebe (1998), Simulated influence of carbon dioxide, orbital forcing and ice sheets on the climate of the Last Glacial Maximum, Nature, 394(6696), 847-853. Yoshimori, M., A. J. Weaver, S. J. Marshall, and G. K. C. Clarke (2001), Glacial termination: sensitvity to orbital and CO2 forcing in a coupled climate system model, Clim Dynam, 17, 571-588. 125 Figures: Figure 1. Overview of published intermediate and deep water radiocarbon records. Locations of previously published benthic foraminiferal and coral radiocarbon records from intermediate water depths (circles) [Bryan et al., 2010; Marchitto et al., 2007; Lindsey et al., 2015; Siani et al., 2013; De Pol-Holz et al., 2010; Davies-Walczak et al., 2014; Sorter et al., 2011; Romahn et al., 2014; De Pol-Holz et al., 2012; Burke and Robinson, 2012; Magini et al., 2010; Rose and Sikes, 2010; Stott et al., 2009; Kennett and Ingram, 1995], deep/abyssal water depths (squares) [Sikes et al., 2016; Ronge et al., 2016; Galbraith et al., 2007; Lund et al., 2011; Skinner et al., 2010; Skinner et al., 2014; Okazaki et al., 2012; Robinson et al., 2005; Keigwin, 2004] and three new records from intermediate depths in the EEP (stars, this study). Intermediate depth records that do (do not) exhibit excursions to depleted Δ14C values during the deglaciation are plotted in yellow (blue). Deep and intermediate Δ14C records that exhibit low Δ14C values during 126 the last glacial period, indicative of the presence of a glacial carbon pool, are plotted in red. Orange dotted lines outline the extent of AAIW in each ocean basin [Talley et al., 2011]. Intermediate depth records that document an influx of old carbon during the last deglaciation are localized to the tropical Pacific, while those closer to the Pacific AAIW formation regions (pink ovals) do not. The record from 600 m on the Galapagos platform (this study) is the only record to exhibit evidence for old carbon during both the last glacial period and deglaciation. 127 Figure 2. Modern Pacific Δ 14C concentrations along 90°W. [WOCE P19]. Locations of sediment cores CDH 23, CDH 26, and GGC43/CDH41 projected onto WOCE line P19. Core sites are bathed by a combination of SAMW and AAIW from the south with contributions from PDW. PDW is the oldest water mass in the global oceans today, with Δ14C values as low as ~-250‰ at its core. Figure generated using the ODV software. 128 Figure 3. Paired planktonic and benthic foraminiferal 14C dates. B-P offsets plotted versus adjusted core depth, a splice of GGC 43 and CDH 41. B-P offsets are largest during the last glacial period, which suggests strong stratification between surface and deep waters. B-P offsets decrease through the deglaciation towards a Holocene average of ~315 years. Seven planktonic 14C dates are identified as outliers during the deglaciation likely due to reduced stratification conditions and increased upwelling of 14 C-depleted deep waters. 129 Figure 4. Radiocarbon data from 373, 600, and 1026 m water depth in the EEP. (a.) Δ14C concentrations at 600 m on the Galapagos Platform (purple) are exceptionally depleted during the last glacial period and deglaciation. The records from the N. Peru Margin (373 m = Green, 1026 m = Blue) maintain a near constant offset with atmospheric Δ14C concentrations. Error bars reflect 2 sd in the age-depth model and are angled because the calculation for Δ14C is contingent on calendar age [Blaauw, 2010]. (b.) B-P offsets exhibit similar trends to the Δ14C signatures at the core sites (opposite in sign), with large offsets at 600 m during the LGM and deglaciation. At 1026 m water depth along the N. Peru Margin B-P 14C age offsets are also larger during the LGM and deglaciation than during the early Holocene. Larger than modern B-P offsets imply reduced ventilation of intermediate water masses and/or enhanced stratification over the upper water column in the EEP relative to today. Error bars reflect 2 sd in the age-depth model in the x-direction and the sum of the measurement errors for both the planktonic and benthic 14C dates in the y-direction. 130 Figure 5. Δ 14C evolution over the past 30 kyrs by water mass. Radiocarbon records from AAIW depths in the Pacific Ocean maintain a relatively constant offset with the atmospheric Δ14C concentrations over the past 30 kyrs [De Pol-Holz et al., 2010; Burke and Robinson, 2012; Siani et al., 2013]. The N. Peru Margin records from 373 and 1026 m water depth compare well to the AAIW Δ14C history. Sites bathed by EqPIW exhibit negative Δ14C anomalies during the deglaciation, similar to what we observed at 600 m 131 water depth on the Galapagos platform [Marchitto et al., 2007; Lindsay et al., 2015; Bryan et al., 2010]. We exclude a record from marine sediment core VM21-30 recovered from 617 m, also on the Galapagos platform (Supplementary Text 2 and Supplementary Fig. 2) [Stott et al., 2009]. Δ14C values are exceptionally depleted in this record during the deglaciation, reaching nearly -600‰, and are not reconcilable with our record or any others from intermediate depths (Supplemental information Fig. 2). Finally, the radiocarbon history of PDW is characterized by low Δ14C concentrations during the last glacial period, which recover to near modern values by the end of HS1 [Skinner et al., 2010; Skinner et al., 2015; Ronge et al., 2016; Sikes et al., 2016; Galbraith et al., 2007; Lund et al., 2011]. 132 Figure 6. Release of the glacial carbon pool and the rise of atmospheric CO2. Radiocarbon and authigenic uranium records from the mid-depth Pacific ocean (2000- 3500 m) indicate low levels of oxygen and radiocarbon in glacial PDW indicative of poor ventilation [Galbraith et al., 2007; Ronge et al., 2016]. 14C-depleted carbon stored in 133 South PDW was released during HS1, fueling the first rise of atmospheric CO2 during HS1 [Monnin et al., 2004; Ronge et al., 2016]. In the EEP we observe a shift towards reduced B-P offsets, which suggests reduced stratification around the Galapagos Islands and enhanced vertical mixing of old carbon from South PDW to intermediate depths and the surface ocean. North PDW is ventilated ~14.6 ka in time with a 10 p.p.m.v. rise in atmospheric CO2 levels, with the rest of the upwelled carbon re-squestered at depth by surface primary productivity and export to depth. Stratification breakdowns again in the EEP during the YD, which allows the North PDW carbon pool to reach the surface ocean atmosphere. 134 Figure 7. Schematic representation of the major phases of storage and release of the glacial carbon reservoir. (LGM, top left) Deep water circulation and upwelling in the Southern Ocean was reduced in the LGM Pacific, which allowed carbon to accumulate within PDW. Over time the 14C-concentration of the carbon was reduced due to radioactive decay and isolation from the atmospheric source of 14C. (HS1, top right) Enhanced upwelling around the Antarctic continent triggered the release of the PDW glacial carbon reservoir to the atmosphere. Old carbon was directly injected into the intermediate depth EEP and may have also upwelled to the surface ocean around the Antarctic continent. However, either only a small portion of the stored carbon upwelled in the Southern Ocean surface ocean or it quickly dissipated along the flow path of AAIW. (YD, bottom) During the YD the formation region of “new” PDW was shifted poleward to its present day position at 40°N, thereby ventilating 14C-depleted carbon 135 stored in the far north deep Pacific during the last glacial period and HS1. Aged carbon was again injected into the EEP subsurface. 136 Figure 8. EEP stratification and CO2 outgassing. 14C age offsets between 600 m depth on the Galapagos platform and the surface ocean (a.) compare well with other EEP 137 stratification records (c.) based on δ18O, δ13C, and temperature [Bova et al., 2015]. Both stratification histories identify HS1 and the YD as periods of reduced stratification. (b.) Boron isotopes from nearby site ODP 1238 document CO2 outgassing [Martinez-Botí et al., 2015]. Periods of reduced stratification as documented here occur simultaneously with enhanced CO2 outgassing at site ODP 1238, which suggests enhanced upwelling of PDW to the surface ocean during periods of reduced stratification in the EEP helped fuel the deglacial rise in atmospheric CO2. 138 Supplemental Information Text S1. Reservoir Ages Deep ocean Δ14C calculations depend on knowing benthic foraminiferal 14C ages and their corresponding calendar age. Implicit in this calculation is the assumption that the surface reservoir age does not change substantially over the past 25 kyrs. However, it is a practical certainty that surface reservoir ages have varied over the past 25 kyrs. There is strong evidence that upwelling rates and water column stratification have changed substantially over this time interval [e.g. Koutavas et al., 2002; Calvo et al., 2011; Bova et al., 2015], potentially resulting in variable amounts of aged subsurface waters mixing with surface water and substantially altering surface reservoir ages [de la Fuente et al., 2015]. Our data indicate that the largest reservoir ages (and thus the largest offset from the assumed constant reservoir age) occurred during the deglaciation. Enhanced upwelling of aged intermediate and deep waters are most likely to occur during intervals of reduced stratification. At the Galapagos site we observe reduced B-P offsets, indicative of reduced stratification, during the YD and HS1, coincident with the two-step rise in atmospheric CO2. Stratification indices based on δ18O, δ13C, and temperature records from the three sites confirm these intervals as periods of reduced stratification and enhanced vertical mixing (Fig. 7) [Bova et al., 2015]. In contrast, de la Fuente et al. [2015] found that reservoir ages varied by 2000 years during the LGM at nearby site ODP 1240 (0° 01.31’N, 86° 27.76 W). We find these reservoir age shifts counterintuitive; the largest surface reservoir ages, which the authors suggest reflect an enhanced influx of 14 C-depleted Pacific Deep Water to the surface ocean, occur during the LGM when B-P offsets are the largest. Large B-P offsets indicate strong stratification between surface and 139 deep water masses and reduced surface to deep mixing. Given the tentative and sparse nature of the tie points to the Greenland ice cores, the calculated reservoir ages are therefore suspect. We therefore suggest that, without independent age constraints, a constant 500 year reservoir age is the best approximation and is sufficient for interpretation and comparison to other subsurface 14C records. Text S2. Comparison to the Δ 14C record from VM21-30 Core VM21-30 was recovered just 5 km north of cores CDH 41 and GGC 43 but exhibits Δ14C-concentrations as low as -600‰, approximately 300‰ lower than the minimum observed value in our record (Supplementary Fig. 2) [Stott et al., 2009]. Stott and Timmermann [2011] suggest the anomalously old water masses detected at site VM21-30 can be explained by inputs of 14C free geologic carbon via the destabilization of CO2- hydrates. Given the strong similarity between our 14C record from intermediate depths on the Galapagos Platform with the record from off the coast of Baja, CA, however, it seems unlikely that CO2-hydrate destabilization would result in the same pattern at both locations simultaneously. We therefore do not consider the VM21-30 record in our analyses. Text S3. Carbon storage in PDW We base our estimates for the amount of carbon stored and released from within SPDW and NPDW on the empirical approach of Sarnthein et al. [2013] and the subsequent application by Skinner et al. [2015]. Skinner et al. [2015] demonstrate that seawater DIC increases by ~1.22 µmol/kg per ‰ decrease in radiocarbon concentration. We 140 approximate the volume of water in the Pacific ocean between 2000 and 3500 m as 1.13 x 1017 m3 and halve this number to approximate the Northern and Southern Pacific Deep Water volumes. During HS1, SPDW and NPDW Δ14C concentrations rose by approximately 400‰ [Ronge et al., 2016] and 130‰, respectively. An increase of 400‰ corresponds to an increase in deep water DIC by 488 µmol/kg and thus the release of 339 GT of carbon from SPDW. A 130‰ increase, as observed within NPDW during the YD, corresponds to a roughtly 158.6 µmol/kg increase in DIC and the release of 110 GT of carbon. Approximately two-thirds of the deglacial rise in atmospheric CO2 occurred during HS1 and one-third during the YD. Our calculations demonstrate that roughly two- thirds of the carbon released from the deep Pacific came from SPDW (339 GT carbon) during HS1 and one-third from NPDW (110 GT carbon) during the YD and therefore correspond well to the observed changes in atmospheric CO2 concentrations. 141 Supplementary Figure 1. Core chronologies. 14C dates using the Clam version 2.2 age- modeling software package for R [Blaauw, 2010]. All 14C dates were converted to calendar ages using the Marine13 radiocarbon age calibration curve [Reimer et al., 2013]. The shaded region represents the 95% confidence window. Because B-P offsets at the Galapagos site vary wildly over the last 35 kyrs, implying signifcant change in vertical stratification near the core site and variable rates of deep/intermediate water upwelling to the surface ocean, we remove five 14C dates from the original core chronologies of Bova et al. [2015] during intervals of decreasing B-P offsets to avoid possible changes in the surface reservoir age. Together with dates removed in the original chronology a total of ten 14C dates from the Galapagos cores were removed [Appendix C and Fig. 3 in the main text]. 142 Supplementary Figure 2. Comparison to previously published benthic Δ 14C record from the Galapagos Platform. A paired benthic and planktonic 14C record from core VM21-30 recovered from 617 m water depth on the the Galpagos platform (green symbols) [Stott et al., 2009] exhibits a radically different benthic foramiferal Δ14C history than that observed at site GGC 43/CDH 41 (this study, blue symbols) and off the Baja Peninsula [Marchitto et al., 2007]. Core VM21-30 was recovered just 5 km north of cores CDH 41 and GGC 43 but exhibits Δ14C-concentrations as low as almost -600‰, approximately 300‰ lower than the minimum observed value in our record. 143 Supplemental References: Bova, S. C., T. Herbert, Y. Rosenthal, J. Kalansky, M. Altabet, C. Chazen, A. Mojarro, and J. Zech (2015), Links between eastern equatorial Pacific stratification and atmospheric CO2 rise during the last deglaciation, Paleoceanography, 30(11), 1407- 1424. Calvo, E., C. Pelejero, L. D. Pena, I. Cacho, and G. A. Logan (2011), Eastern Equatorial Pacific productivity and related CO2 changes since the last glacial period, Proceedings of the National Academy of Sciences, 108(14), 5. de la Fuente, M., L. Skinner, E. Calvo, C. Pelejero, and I. Cacho (2015), Increased reservoir ages and poorly ventilated deep waters inferred in the glacial Eastern Equatorial Pacific, Nat Commun, 6. Koutavas, A., J. Lynch-Stieglitz, T. M. Marchitto, and J. P. Sachs (2002), El Nino-like pattern in ice age tropical Pacific sea surface temperature, Science, 297(5579), 226- 230. Lindsay, C. M., S. J. Lehman, T. M. Marchitto, and J. D. Ortiz (2015), The surface expression of radiocarbon anomalies near Baja California during deglaciation, Earth and Planetary Science Letters, 422, 67-74. Marchitto, T. M., S. J. Lehman, J. D. Ortiz, J. Fluckiger, and A. van Geen (2007), Marine radiocarbon evidence for the mechanism of deglacial atmospheric CO2 rise, Science, 316(5830), 1456-1459. Ronge, T. A., R. Tiedemann, F. Lamy, P. Kohler, B. V. Alloway, R. De Pol-Holz, K. Pahnke, J. Southon, and L. Wacker (2016), Radiocarbon constraints on the extent and evolution of the South Pacific glacial carbon pool, Nat Commun, 7. 144 Sarnthein, M., Grootes, P. M., Kennett, J. P., and M.J. Nadeau 2007. 14C reservoir ages show deglacial changes in ocean currents and carbon cycle. In: Schmittner, A., Chiang, C.H., Hemming s. R., (Eds.), Ocean Circulation: Mechanisms and Impacts. AGU, Washington, DC. Skinner, L., McCave, I.N., Carter, L., Fallon, S. Scrivner, A.E., and F. Primeau (2015). Reduced ventilation and enhanced magnitude of the deep Pacific carbon pool during the last glacial period, Earth and Planetary Science Letters, 411, 45-52. Stott, L., and A. Timmermann (2011), Hypothesized Link Between Glacial/Interglacial Atmospheric CO2 Cycles and Storage/Release of CO2-Rich Fluids From Deep-Sea Sediments, Geophysical Monograph, Abrupt Climate Change: Mechanisms, Patterns, and Impacts, 193. Stott, L., J. Southon, A. Timmermann, and A. Koutavas (2009), Radiocarbon age anomaly at intermediate water depth in the Pacific Ocean during the last deglaciation, Paleoceanography, 24(2). 145 CHAPTER FOUR ________________________________________________________________________ Relationships between thermocline depth, sea surface temperature, and productivity in the Eastern Equatorial Pacific: no role for ENSO since the last glacial period Samantha C. Bova1, Timothy D. Herbert1, Sathya Annisetti2, Caitlin Chazen1, Angel Mojarro1, and Jana Zech1 1 Brown University, Department of Earth, Environmental, and Planetary Sciences, Providence, RI USA 2 Massachusetts Academy of Math and Science, Worcester, MA USA 146 Abstract The Eastern Equatorial Pacific (EEP) is a complex environment, with important feedbacks to the global carbon cycle and ocean heat content. The shallow thermocline, characteristic of the EEP, connects the EEP subsurface with the surface ocean and atmosphere, stimulating high rates of primary production as well as rapid heat and gas exchange with the atmosphere. Today, the El Niño Southern Oscillation (ENSO), an ocean-atmosphere interaction that originates over a small latitudinal band of the tropical Pacific, exerts primary control on the depth of the EEP thermocline and is the second largest source of variability in the region on human timescales, after the seasonal cycle. Predicting the response of the EEP thermocline and ENSO to alternate climate background states is therefore of primary concern. In this study, we assess the response of the EEP to climate change since the last glacial period. We reconstruct a 30 kyr history of SST, thermocline depth, and phytoplankton productivity at four sites in the EEP. We find that the modern relationships between sea surface parameters, characteristic of ENSO, are not maintained over the past 30 kyrs. Mixed layer properties were likely affected primarily by the local wind-driven forcing, while remote forcing regulated thermocline depth. Thus, though ENSO may have been active over the past 30 kyrs, its influence on EEP thermocline depth was overshadowed by the influx of subsurface temperature anomalies from the extratropics. 1. Introduction El Niño, and its counterpart, La Niña, are the most important source of interannual climate variability on the planet. The phenomena, termed the El Niño Southern Oscillation (ENSO), lead to large changes in the heat distribution across 147 equatorial Pacific with important consequences for global atmospheric circulation patterns [e.g. Philander, 1985]. Predicting the response of ENSO to future global warming is therefore of primary concern to policy makers. Thus far, simulations with coupled general circulation models project conflicting ENSO responses to warm scenarios: some model results indicate change in the zonal mean thermocline depth of the equatorial Pacific [DiNezio et al., 2009] while others predict sustained “ENSO-like” variation in its tilt [Vecchi and Soden, 2007]. Given the poor constraints available, records from the past are necessary to test model estimates of the EEP thermocline response under alternate climate background states. The ENSO cycle is characterized by specific relationships between thermocline depth, sea surface temperature, and primary productivity, each of which can be independently quantified in paleoceanographic records [e.g. Cane et al., 1997; Pennington et al., 2006]. A shallow thermocline allows the surface wind to more efficiently entrain and upwell cold, nutrient-rich sub-thermocline water masses. Consequently, during El Niño when the thermocline is deep, warm SSTs are coupled with reduced primary production, and during La Niña when it is shallow, SSTs cool and primary production increases [e.g. Philander et al., 1995; Barber et al., 1996]. This is the classic El Niño Southern Oscillation (ENSO) paradigm- one paleoceanographers extend to timescales of 102-106 years [Beaufort et al., 2001; Koutavas et al., 2002; Stott et al., 2002; Wara et al., 2005; Pena et al., 2008]. A recent synthesis of Holocene SSTs, using both Mg/Ca and alkenone paleo-SST proxies, finds that equatorial Pacific SSTs were in a La Niña-like distribution and therefore infer reduced ENSO variability prior to 2 ka [Gill et al., 2016]. But do SST patterns adequately capture dynamics in the eastern equatorial 148 Pacific that play out on timescales of decades to millennia? That is, can ENSO serve as a model for shifts in the EEP mean state under differing boundary conditions? A number of theoretical paleoceanographic observations [Seager and Murtugedde,1997; DiNezio et al., 2011] suggest that the answer may be “no”. On centennial and longer scales, a primary forcing on thermocline depth, SST, and paleoproductivity may come from the extratropics. Subsurface variability reflecting anomalies sourced from the mid- [Liu and Philander, 1995; Gu and Philander, 1997; Lu et al., 1998; Boccaletti et al., 2004] and high-latitudes [Toggweiler et al., 1991; Andreasen et al., 2001] might play an essential role in setting the background stratification of the EEP on centennial and longer scales. These signals emerge as rapid changes in denitrification [Chazen et al., 2009] and subsurface water density along the Peru margin [Bova et al., 2015]. In the case of positive nitrogen isotope anomalies (increased strength of the oxygen minimum zone (OMZ)), these coincide with minima in inferred paleoproductivity, indicating that increased residence time of “shadow-zone” waters, not a wind-driven upwelling increase in carbon flux and oxygen demand, trigger centennial pulses in the OMZ. The observed subsurface anomalies occur too deep to be forced from the surface by mechanisms associated with ENSO. In this study, we apply independent proxies to determine the thermocline depth, SST, and productivity at four EEP locations over the past 30 kyrs to provide an essential test of the hypothesis that millennial to orbital scale changes in the oceanography of the region can be regarded as “scaled-up” El Niño or La Niña conditions. The transition from the most recent glacial period into the current interglacial provides an excellent opportunity to test the EEP upper ocean response to glacial boundary conditions as well 149 as a rapidly warming climate. We provide SST and productivity estimates along with the first quantitative regional reconstruction of the depth of the thermocline based on the relative abundance of the coccolithophore Florispheara profunda. Based on the observed correlations between SST, productivity, and thermocline depth we reject ENSO as the dominant source of variability in the EEP since the last glacial period. In addition, we find that SST is not a straightforward indicator of thermocline depth and is therefore a necessary, but not sufficient, parameter for determining whether on millennial and longer timescales the EEP can shift between states resembling dominantly El Niño, La Niña, or a state similar to neither. 2. Oceanographic Setting 2.1 Mean Hydrography Circulation patterns in the EEP upper water column are highly complex and result in sharp chemical and physical gradients. The region’s most prominent feature, the Equatorial Cold Tongue (ECT), centered at 1°S, is composed of three intersecting but dynamically distinct upwelling systems: the Peru Chile Upwelling (PCU), the Equatorial Upwelling (EU), and the Galapagos Upwelling (GU) systems [e.g. Fiedler and Talley, 2006]. Along-shore winds drive surface waters away from the South American coast in the PCU system, bringing cool, nutrient-rich subsurface waters from depth to the surface ocean [Pennington et al., 2006]. The EU system is sustained by easterly winds that drive waters away from the equator as the Coriolis force changes sign across the equator. Westward advection of upwelled waters from the Peru Current connect the two upwelling systems creating a tongue of cool water just south of the equator in the EEP, known as Equatorial Surface Water (ESW). Finally, collision of eastward flowing subsurface 150 Equatorial Undercurrent (EUC) with the Galapagos Islands drives perennial upwelling west of the islands [Kessler, 2006]. Surface water properties in the ECT are thus determined by advection of cool waters from the Peru Current and the strength of upwelling within each system [Wyrtki, 1966; 1981; Fiedler and Talley, 2006]. A sharp thermal and salinity gradient along 2°N separates the ECT from the eastern Pacific warm pool (EPWP) [Fiedler and Talley, 2006]. The EPWP is the second largest tropical warm pool on the planet and arises through a combination of weak wind- driven mixing and a seasonally large net heat flux [Wang and Enfield, 2001]. The sharp SST gradient between the ECT and the warm pool is maintained by the influx of cool waters primarily via local upwelling along the equator with smaller contributions from zonal advection of waters from the PCU system to the east [Kessler et al., 1998; Kessler, 2006]. The eastward flowing EUC, centered at 80 m depth in the EEP, supplies the PCU, the GU, and the EU systems. It is composed primarily of modified subtropical mode water with contributions from southern ocean mode and intermediate waters (SOIW) sourced from the southeast Pacific and northeast of New Zealand [Toggweiler et al., 1991; Qu et al., 2009; Bostock et al., 2010] and small additions from North Pacific Intermediate Water (NPIW). Tracer experiments [Qu et al., 2010] demonstrate that intermediate waters do eventually upwell to the surface ocean in the EEP but have minimal impacts on regional SSTs through at least the Holocene [Kalansky et al., 2015]. Nevertheless, SOIW and NPIW are the primary macronutrient sources to the EEP; SOIW and NPIW are both nitrate- and phosphate-rich but silicate is sourced almost exclusively from the north [Sarmiento et al., 2004]. Competition between northern and southern 151 intermediate waters can be diagnosed as a dichotomy of silica-rich thermocline waters from the north Pacific versus silica poor waters from the south Pacific [Dubois et al., 2010; Kienast et al., 2006; Sarmiento et al., 2004] and north-south shifts in the relative abundance of siliceous organisms in the EEP. 2.2. Seasonal and Interannual Variability Seasonal migration of the Intertropical Convergence Zone (ITCZ) is the primary driver of variability in the ECT in the historical record [e.g. Fiedler and Talley, 2006]. The seasonal amplitude of SST variability in the ECT is ±1-3°C, with the coldest temperatures observed in September and October, when the Intertropical Convergence Zone reaches its most northerly position and strong, easterly winds over the region drive a mean shoaling of the thermocline south of the ITCZ by about 20 m [Fiedler and Talley, 2006]. A shallow thermocline acts as positive feedback to surface cooling, allowing winds to more efficiently mix cool, nutrient-rich subsurface waters to the surface ocean [Kessler, 2006]. Because the delivery of macronutrients exerts primary control on rates of production in the EEP today, phytoplankton are expected to respond in phase with the seasonal cycle. However, phytoplankton in the EEP do not exhibit strong seasonality and, in fact, in the northern PCU system phytoplankton exhibit some of the highest rates of production during boreal winter when upwelling rates are at a minimum [Chavez, 1995; Pennington et al., 2006]. Light limitations resulting from stratus cloud formation and fog over cool surface waters or dilution by advection and mixing may explain this surprising relationship. ENSO is the second leading cause of variation in upwelling strength, thermocline depth, and SST in the EEP over the time period of historical observations, and the 152 primary source of variability in phytoplankton growth [Pennington et al., 2006; Wang and Fiedler, 2006]. The influence of ENSO is strongest in in the EU and coastal boundary regions (±1-2°C) and much less so in the warm EPWP [Fiedler and Talley, 2006]. Phytoplankton production decreases substantially during El Niño events across the entire EEP and in coastal regions ecological shifts from dominantly diatoms and larger phytoplankton communities to smaller phytoplankton, typically observed in the open- ocean, have been observed during particularly strong events [e.g. Chavez, 1989; Chavez et al., 1991]. 3. Site Locations Processes that might affect the evolution of the EEP over the past 30 kyrs include local wind-driven mechanisms, such as ENSO, and remote forcing from the extratropics. The spatial response of the EEP to each mechanism is different and can help distinguish the relative role of local versus remote controls on EEP oceanography. We therefore reconstruct thermocline depth, SST, and primary production at multiple sites spread across the region. Sediment cores used in this study were recovered from four distinct hydrographic regimes in the EEP: (1) the EPWP (ME0005A-15MC/17JC), (2) the open- ocean ECT (VNTR01-13GC), (3) the PCU system (CDH 23/CDH 26), and (4) the GU system (GGC 43/CDH 41) (Figure 1, Table 1). 4. Methods 4.1 Age models Core chronologies were constructed by applying polynomial fits to AMS 14C dates of planktonic foraminifera using the Clam age model software of Blaauw [2010] (Figure 2). Radiocarbon measurements were made at the National Ocean Sciences 153 Accelerator Mass Spectrometry facility at Woods Hole Oceanographic Institute. Planktonic foraminifera were picked from the >150 µm size fraction. All 14C dates were converted to calendar ages using the marine13 radiocarbon age calibration curve [Reimer et al., 2013]. We assume constant reservoir ages at each of the sites based on the nearest modern measurements from the Marine Reservoir Age database (Table 1) [Taylor and Berger, 1967; Druffel, 1981; 1995; 2004; Jones et al., 2007; Etayo-Cadvid et al., 2013]. Age models for sites CDH 23 (373 m) and CDH 26 (1026 m) are based on 24 and 25 AMS 14C measurements, respectively, made on the thermocline dwelling foraminifer, Neogloboquadrina dutertrei. Cores CDH 23 and CDH 26 are spliced together at 13.7 ka to provide a complete record spanning the last 30 ka. Planktonic 14C dates at the Galapagos sites were measured on Globogerinoides Ruber. We modify the published chronologies of Bova et al., [2015] for the Peru Margin and Galapagos cores by using the marine13 calibration curve and the Clam Software package to characterize age model uncertainty instead of the Fairbanks calibration program. In addition, we remove seven planktonic 14C dates as outliers from the chronology for cores GGC 43 and CDH 41 (617 and 595 m, respectively) [see Chapter 3, Results]. Age models for ME0005-15MC/17JC and VNTR01-13GC are based on 5 (G. Ruber) and 9 (N. dutertrei) AMS 14C measurements, respectively. In addition, oxygen and carbon isotopic signatures of benthic foraminifera from each site were used to confirm the initial 14C stratigraphy by comparison to LR04 benthic stack [Lisiecki and Raymo, 2005]. Isotopes were measured on Uvigerina peregrina and Cibicoides wuellerstorfi at site VNTR01-13GC. C. wuellerstorfi was not abundant at site ME0005A- 17JC so only U. peregrina species were measured. Based on the radiocarbon and oxygen 154 isotope data, we identify two identical Holocene sections in core VNTR01-13GC that must represent either slumping or duplication in coring. Because the top Holocene section appears more complete, we splice the top 52 cm of the core, spanning 1.35 to 16.6 ka, to early deglacial sediment beginning at 90 cm, and remove measurements from the middle 36 cm of the core from the dataset. 4.2 Biomarkers Organic biomarker records were analyzed in cores GGC 43/CDH 41, VNTR01- 13GC, and ME0005A-15MC/17JC on an Agilent 6890 Gas Chromatograph with a Flame Ionization Detector [GC-FID]. Samples from the Galapagos cores, GGC 43 and CDH 41, were analyzed only for alkenone biomarkers. Alkenone records from the N. Peru Margin sites (CDH 23 and CDH 26) were published previously by Bova et al., [2015]. We use the calibration equation of Müller et al., [1998] to calculate paleo-SSTs from alkenone distributions. Lab analytical error is equivalent to ±0.1°C based on 55 replicate analyses of a laboratory standard. Approximately 10% of our analyzed samples were run in duplicate with an average UK’37-based temperature analytical reproducibility of ±0.05°C. Brassicasterol abundances were also characterized by GC-FID analysis. Thirty-nine brassicasterol samples were run in duplicate with an average reproducibility of ±6.5 ng/g. For a more detailed description of our methods please refer to Chapter 5. 4.3. Nannofossil Abundances Twenty-seven core-top samples located between 15°N and 15°S and east of 110°W were analyzed for coccolith abundances (Figure 5). Smear slides were prepared using Norland Optical Adhesive 81. Coccolith identification was carried out using a 155 Leica DMLSP polarized microscope at 1250x magnification. All slides were analyzed using simple relative abundance counting. A minimum of 300 species were identified per slide, which assures identification of any taxon whose abundance is greater than 1% of the total population [Thierstein et al., 1977]. Molfino and McIntrye [1990] and Bolton et al. [2010] previously identified a negative relationship between the abundance of F. Profunda and thermocline depth but a quantitative relationship has not yet been established for the EEP. We calculate the relative abundance of F. profunda according to the following formula adapted from Flores et al., [2000]. N ratio= %G. oceania + %E. huxleyi/(%G. oceanica+%E. huxleyi+ %F. profunda). This index computes the abundance of F. profunda relative to the most abundant surface dwelling coccolithophores, G. oceanica and E. huxleyi. The N ratio is highly sensitive to dissolution biases because of the differential dissolution behavior of Emiliania huxleyi and Gephyrocapsa oceanica relative to F. profunda. E. huxleyi and G. oceanica, are delicate and therefore more prone to dissolution than F. profunda. To avoid dissolution biases we therefore only use core-tops that lie above the modern lysocline (3700 to 4000 m) [Weber et al., 1995]. Even above the nominal lysocline, however, dissolution can be still a problem. We therefore assess coccolith preservation using the CEX’ index following Boeckel and Baumann [2004]: CEX’= (% E. huxleyi + % G. oceanica)/(% E. huxleyi + % G. oceanica +% C. leptoporus). The index compares the preservation state of Emiliania huxleyi and Gephyrocapsa oceanica, relative to the robust species, Calcidiscus leptoporus). Any sample with a CEX’ index less than 0.63 is removed from the N ratio calibration. 156 Coccolithophorid relative abundances were calculated down-core at three EEP locations: ME0005A-15MC/17JC, GGC 43/CDH 41, and VNTR01-13GC. Slides were prepared and analyzed in the same manner as the core-top samples. Counts were completed approximately every 2 cm over the top 60 cm in core ME0005A-15MC/17JC, every 10 cm over the top 280 cm in core GGC 43/CDH 41 and every 6 cm over the top 200 cm in core VNTR01-13GC to provide a measurement approximately every 1 to 1.5 kyrs over the past 30 kyrs. The average standard deviation in the N ratio based on five duplicate counts is 0.033. Alkenone and brassicasterol biomarker concentrations were analyzed for all counted intervals. 5. Results 5.1 Alkenone Sea Surface Temperature Reconstructions The SST history of the EEP across the past 25 kyrs, as reconstructed by alkenones is geographically heterogeneous. The timing of minimum SSTs at the end of the last glacial period and the amplitude of deglacial warming differs by location. At the northernmost site (ME0005A-15MC/17JC, 4.61°N) SSTs decrease during the last glacial period, reach a minimum at 20.0 ± 0.4 ka, and then increase through the deglaciation and into the early Holocene by 1.15 ±0.05 °C. Early Holocene SSTs at this site are comparable to those observed during the last glacial period between 33 and 35 ka. At the Galapagos (1.25°S) we observe a broadly similar pattern, with minimum SSTs occurring slightly later at 18.0 ± 0.3 ka and the amplitude of LGM to Holocene warming slightly larger, at 2.04 ± 0.05 °C. The higher resolution of the record enables observation of a two-part warming during the last deglaciation during both periods of SH warming (11.2 157 to 12.3 & 16.4 to 18 ka). Holocene SSTs stabilize during the late Holocene at 24.01 ± 0.05 °C. The two southernmost sites exhibit markedly different SST histories both from each other and from the Galapagos and northernmost site. At 3.09°S (VNTR01-13GC) we observe a minimum in SSTs (20.88 ± 0.05°C) during the last glacial period at 25.8 ± 0.4 ka. SSTs rise by 2.75 ± 0.05°C towards the onset of deglaciation at 19.8 ±0.3 ka and exhibit high variability, with multiple 1 to 1.5 ± 0.05°C temperatures swings in less than 1 kyrs. Temperatures stabilized during HS1 at 22.70 ± 0.05°C before rising another 2.23 ± 0.05°C between 15.5 ± 0.4 and 13.2 ± 0.4 ka. Unlike the Galapagos record, SSTs decrease at this location during the early Holocene. SSTs reach a minimum temperature of 23.48°C ± 0.05°C by the late Holocene. The record from our southernmost site along the Peru Margin (3.99°S) exhibits the latest onset for warming during the deglaciation, 14.9 ± 0.1 ka. SSTs rise 4.15°C from 19.81 ± 0.05 to 23.96 ± 0.05°C beginning at 14.9 ± 0.1 ka. Holocene SSTs here decrease towards present day but exhibit millennial-scale variability. 5.2 Productivity Estimates We analyzed C37-alkenones and brassicasterol concentrations to characterize coccolithophorid and diatom production at four and two sites, respectively, in the EEP: ME0005A-15MC/17JC (alkenones and brassicasterol), CDH 41/GGC 43 (alkenones only), VNTR01-13GC (alkenones and brassicasterol), and CDH 23/CDH 26 (alkenones only). Alkenone concentrations are a proxy for haptophye or coccolithophorid production and brassicasterol concentrations have been shown to trace diatom production rates [e.g. Higginson et al., 2004; Calvo et al., 2011]. 158 Biomarker records from across the EEP exhibit strong spatial heterogeneity. Open ocean sites (ME0005A-15MC/17JC and VNTR01-13GC) are most similar with maxima in both C37-alkenone and brassicasterol abundances at the end of the last glacial period between approximately 15 and 31 ka. C37-alkenone and brassicasterol abundances at sites ME0005A-15MC/17JC and VNTR01-13GC are positively correlated, with r-squared values of 0.83 (p<0.01) and 0.38 (p<0.01), respectively [Chapter 5, Figure 6]. Productivity records from the Galapagos and Peru Margin sites are unlike those from the open ocean regions of the EEP. High alkenone abundances are observed along the Peru Margin during the last glacial period, similar to the open ocean sites, but are also high during the Holocene. C37-alkenone abundances increase at the onset of the Holocene until 6.4 ka and then decrease towards present day. In addition, alkenone abundances along the Peru Margin exhibit increased variance in the mid- to late-Holocene. At the Galapagos, alkenone abundances decrease between 27.9 and 22.0 ka from 6300 to 3200 ng/g and remain stable at 3900 ng/g until 11 ka. C37-alkenone abundances increase from 11 ka to present day. 5.3 Quantitative Thermocline Depth Calibration The N ratio is negatively correlated (R2=0.92, p<0.001, Figure 5) with modern thermocline depth [Locarini et al., 2013], defined here as the midpoint of the depth interval with the maximum temperature gradient where ΔT>2°C or Δz>10 m [Fiedler and Talley, 2006]. A low N ratio suggests a high relative abundance of F. profunda and consequently a deep thermocline (Figure 5). Conversely, a large N index indicates a low relative abundance of F. profunda and a shallow thermocline. 159 Six core top samples were removed from the dataset because coccolith abundances were either too low for accurate counting (< 20 individuals per field of view) or species known to have gone extinct prior to the Quaternary were identified in the sample, suggesting significant reworking of nannofossils. An additional three core-top samples recovered from the vicinity of the Galapagos Islands record lower than expected N ratios for their modern day thermocline depths relative to the rest of the dataset (Figure 5), and we therefore also remove these data from the calibration. The preservation of coccoliths in all core-tops samples is excellent as determined by the CEX’ index. All samples exhibit high CEX’ values above the critical value of 0.63 [Boeckel and Baumann, 2004]. 5.4 Thermocline Depth Reconstructions We reconstruct thermocline depth at sites ME0005A-15MC/17JC (4.62°N), GGC 43/CDH 41 (1.25°S), and VNTR01-13GC (3.09°S) over the past 30 kyrs using the established N ratio calibration (Figure 6). All locations exhibit a shallower thermocline during the LGM and deglaciation relative to present day. The thermocline depth at each site converges towards comparable values, between 20 and 30 m depth, during the last deglaciation and diverges at the onset of the Holocene towards present day. Site VNTR01-13GC exhibits the largest range in thermocline depth over the past 30 kyrs with a minimum depth of 20.9 m observed 15.5 ka and maximum values observed during the late Holocene. Because core-tops in the vicinity of the Galapagos fall off the calibration (Figure 5), the N ratio inferred thermocline depth may be offset from the true thermocline depth at the Galapagos site. However, we expect that the trend is accurately reconstructed. 160 6. Discussion 6.1 No evidence for ENSO-like changes in the EEP mean state Deepening (shoaling) of the EEP thermocline is a characteristic feature of El Niño (La Niña) events and leads to a number of correlations between parameters that leave proxy traces in the sedimentary record. ENSO dynamics lead to a strong positive relationship between thermocline depth and SST and a strong negative relationship between primary production and thermocline depth (and SST) on interannual timescales [e.g. Cane et al., 1997; Barber et al., 1996]. Over the past 30 kyrs we find that these characteristic ENSO relationships between thermocline depth, SST, and productivity breakdown. Open-ocean sites (ME0005A-15MC/17JC and VNTR01-13GC) north and south of the equator exhibit no statistically significant relationship between thermocline depth and SST or productivity (Figure 7). Near the Galapagos Islands thermocline depth is weakly correlated with SST (R2=0.50, p<0.01) and coccolithophorid production (R2=0.34, p<0.05), but productivity exhibits a positive correlation with both parameters (SST: R2=0.35, p<0.01), exactly the opposite expected from ENSO dynamics. The relationship between SST and primary production was examined at three additional sites in the EEP, spanning 3.99°S to 4.61°N, and include both open-ocean and coastal upwelling sites. All sites exhibit statistically significant relationships between SST and productivity but not always the inverse correlation expected from ENSO-related dynamics (Figure 8a). Open-ocean sites in the EPWP and the southern edge of the ECT maintain the modern ENSO-like relationship between production and SST, with cool SSTs leading to greater coccolithophorid production (R2=0.72, p<0.01 and R2=0.36, p<0.01, respectively) and diatom productivity (R2=0., 0.74, p<0.01 and R2=0.84, p<0.01, 161 respectively). Sites closer to perennial upwelling zones, however, near the Galapagos and N. Peru Margin exhibit a more complicated history. During the LGM and deglaciation, coccolithophorid production and SSTs are not significantly correlated, while the Peru Margin site exhibits a significant negative relationship between the two parameters (R2=0.19, p<0.05). At the onset of the Holocene 11.8 ka, both sites show a significant rise in the production when SSTs rise (Galapagos: R2=0.55, p<0.01; N. Peru Margin: R2=0.47, p<0.01), again the opposite relationship expected for ENSO forcing. Although, we cannot eliminate the possibility that diatoms responded differently to SST variations due to an absence of diatom biomarker record from these sites, given the tight correlation between brassicasterol concentrations and C37-alkenones at sites VNTR01-13GC, ME0005A-15MC/JC [Chapter 5, Figure 7] and ODP site 1240 [Calvo et al., 2011] this scenario seems unlikely. Based on the relationships between SST, thermocline depth, and paleoproductivity we suggest that ENSO-like changes in the east-west tilt of the equatorial thermocline do not characterize the paleooceanography of the EEP on millennial timescales since at least the last glacial period. ENSO variability should be strongest in the cold tongue but, even sites in the heart of it, do not exhibit the characteristic ENSO-relationships between thermocline depth and SST or productivity. These results support model results of DiNezio et al., [2011], which do not simulate clear El Niño or La Niña patterns of SST change at the LGM. They find that on centennial and longer timescales the Bjerknes Feedback, which drives ENSO variability, is weak. Instead, zonal advection, evaporation, and cloud cover influence SST locally, leading to patterns of SST that do not resemble the modern ENSO pattern [Ma et al., 1996; DiNezio 162 et al., 2009]. Future work in the western equatorial Pacific is required to confirm the lack of ENSO-like dynamics in the tropical Pacific over the past 30 ka; variability in EEP SST, thermocline depth, and paleoproductivity reconstructed here is just one part of the E-W contrast that characterize ENSO dynamics. Nevertheless, observed relationships between parameters in the east caution against interpreting trends in EEP SST and paleoproductivity under the ENSO framework on millennial and longer timescales. 6.2.1 Extratropical controls on thermocline depth in the EEP On decadal and longer timescales remote forcing from the mid- and high-latitudes is likely more effective in determining the depth of the EEP thermocline than locally forced, ENSO dynamics [Andreasen et al., 2001; Boccaletti et al., 2004]. Climate change in the high latitudes can drive long-lived shifts in the position of the Hadley cells, and consequently the ITCZ, through the atmospheric bridge, with direct consequences for wind patterns and upwelling in the EEP. In addition, the oceanic tunnel connects the EEP subsurface to the mid- and high-latitudes by transporting waters from the extratropics directly into the EEP subsurface [Toggweiler et al., 1991; Liu and Philander, 1995; Gu and Philander, 1997; Lu et al., 1998; Andreasen et al., 2001; Boccaletti et al., 2004]. Thermal and/or salinity anomalies sourced from these regions significantly alter stratification in the EEP [Bova et al., 2015] and interact with the surface-driven ENSO cycle in intriguing ways. The relative impact of the atmospheric bridge and the oceanic tunnel on the EEP thermocline can be diagnosed by evaluating thermocline depth trends along a meridional transect. North-south movement of the Hadley cells will lead to disparate thermocline depth trends north and south of the ITCZ; a southward migration of the ITCZ as 163 reconstructed for the LGM [Peterson et al., 2000; Koutavas et al., 2003; Koutavas and Lynch-Stieglitz, 2004] would lead to weaker winds over the ECT and reduced equatorial upwelling and stronger winds in the EPWP. In contrast, subsurface anomalies transported via the oceanic tunnel will drive a uniform shoaling or deepening of the thermocline across the EEP. Subsurface waters are derived from the same, predominantly southern hemisphere sources [Toggweiler et al., 1991; Qu et al., 2009; Bostock et al., 2010] across the entire EEP. Subsurface waters were cooler in the LGM relative to today [Bova et al., 2015] and should therefore promote greater stratification between surface and thermocline waters and a shallow EEP thermocline. Our thermocline depth histories from three sites along a N-S transect, spanning 3.09°S to 4.61°N, in the EEP exhibit approximately synchronous changes in thermocline depth over the past 30 kyrs, and are consistent with the oceanic tunneling system exerting primary control on the EEP thermocline depth since the LGM. The thermocline shoals to the shallowest observed depths relative to today, by 20 to 30 m, between 21 and 13 ka and becomes progressively deeper at all sites towards present day, synchronous with a rise in SOIW temperatures during the Younger Dryas [Bova et al., 2015]. In addition, we observe a disconnect between thermocline depth and surface ocean SSTs, which suggests change in the thermocline depth was driven from below. A regionally synchronous change in thermocline depth by about 20 m is supported by two previous reconstructions based on planktonic foraminiferal abundance transfer functions [Andreasen and Ravelo, 1997] and δ18O gradients between surface and thermocline dwelling planktonic foraminifera [Faul et al., 2000]. It is therefore likely that trends in the depth of the EEP 164 thermocline since the LGM were set by the temperature and salinity of waters delivered to the base of the equatorial thermocline from the extratropics. Although trends in thermocline depth variability since the LGM are synchronous across the region, the N-S structure of thermocline depth was different at the LGM in the EEP relative to today. Today the thermocline is deeper north of the equator than south of the equator but thermocline depth at the LGM was uniform across the equator. A more southerly position of the ITCZ, as reconstructed for the LGM, could explain this pattern; southward migration of the ITCZ leads to weaker winds over the ECT and a deeper thermocline south of the equator, while stronger winds in the EPWP drive a shoaling of the thermocline north of the equator. Shifts in the mean position of the trades winds may therefore alter the N-S thermocline depth gradient simultaneous to regional changes in thermocline depth driven from the subsurface. 7. Concluding Remarks Reconstruction of the change in thermocline depth, SST, and primary productivity at four sites across the eastern equatorial Pacific demonstrates that ENSO is not the dominant source of variability in the EEP during the LGM to Holocene transition. Correlations between SST, thermocline depth, and primary productivity do not exhibit the characteristic ENSO relationships observed today. Changes in mixed layer SST and productivity do not accompany changes in thermocline depth, perhaps indicating a greater influence of local wind forcing on the mixed layer and remote forcing on thermocline depth [Fiedler and Talley, 2006]. We therefore conclude that transmission of high latitude climate signals via the oceanic tunneling system exerts primary influence on the depth of the EEP thermocline over the timespan of our records. If we are correct, 165 inferring paleo-ENSO state from a single near-surface proxy, or the variance in a single near-surface proxy may be misleading. The conclusions of this study have important implications for predicting the future response of the equatorial Pacific to impending climate change. Temperatures are projected to rise globally but polar amplification means that regions closer to the poles will warm fastest [Masson-Delmotte et al., 2006]. We therefore expect subsurface waters, sourced from the extratropics, to warm faster than those at the EEP surface, leading to a reduction in upper ocean stratification in response to anthropogenic climate change. Reduced stratification will in turn produce a mean deepening of the equatorial Pacific thermocline and reduced heat uptake in the tropical Pacific. The response of ENSO to such a change remains uncertain [DiNezio et al., 2009; Vecchi and Soden, 2007], but we can predict that a deeper EEP thermocline will lead to a reduction in CO2 evasion from the EEP over the coming centuries [Le Borgne et al., 2002]. 166 References Andreasen, D. H., A. C. Ravelo, and A. J. Broccoli (2001), Remote forcing at the last glacial maximum in the tropical Pacific Ocean, J Geophys Res-Oceans, 106(C1), 879- 897. Andreasen, D. J., and A. C. Ravelo (1997), Tropical Pacific Ocean thermocline depth reconstructions for the Last Glacial Maximum, Paleoceanography, 12(3), 395. Barber, R. T., M. P. Sanderson, S. T. Lindley, F. Chai, J. Newton, C. C. Trees, D. G. Foley, and F. P. Chavez (1996), Primary productivity and its regulation in the equatorial Pacific during and following the 1991-1992 El Nino, Deep-Sea Res Pt Ii, 43(4-6), 933-969. Beaufort, L., T. de Garidel-Thoron, A. C. Mix, and N. G. Pisias (2001), ENSO-like forcing on oceanic primary production during the Late Pleistocene, Science, 293(5539), 2440-2444. Blaauw, M. (2010), Methods and code for 'classical' age-modelling of radiocarbon sequences, Quat Geochronol, 5(5), 512-518. Boccaletti, G., R. C. Pacanowski, S. G. H. Philander, and A. V. Fedorov (2004), The thermal structure of the upper ocean, J Phys Oceanogr, 34(4), 888-902. Boeckel, B., and K. H. Baumann (2004), Distribution of coccoliths in surface sediments of the south-eastern South Atlantic Ocean: ecology, preservation and carbonate contribution, Mar Micropaleontol, 51(3-4), 301-320. Bolton, C. T., S. J. Gibbs, and P. A. Wilson (2010), Evolution of nutricline dynamics in the equatorial Pacific during the late Pliocene, Paleoceanography, 25. 167 Bostock, H. C., B. N. Opdyke, and M. J. M. Williams (2010), Characterising the intermediate depth waters of the Pacific Ocean using delta C-13 and other geochemical tracers, Deep-Sea Res Pt I, 57(7), 847-859. Bova, S. C., T. Herbert, Y. Rosenthal, J. Kalansky, M. Altabet, C. Chazen, A. Mojarro, and J. Zech (2015), Links between eastern equatorial Pacific stratification and atmospheric CO2 rise during the last deglaciation, Paleoceanography, 30(11), 1407- 1424. Calvo, E., C. Pelejero, L. D. Pena, I. Cacho, and G. A. Logan (2011), Eastern Equatorial Pacific productivity and related CO2 changes since the last glacial period, Proceedings of the National Academy of Sciences, 108(14), 5. Cane, M. A., A. C. Clement, A. Kaplan, Y. Kushnir, D. Pozdnyakov, R. Seager, S. E. Zebiak, and R. Murtugudde (1997), Twentieth-century sea surface temperature trends, Science, 275(5302), 957-960. Chavez, F. P. (1989), Size distribution of phytoplankton in the central and eastern tropical pacific, Global Biogeochemical Cycles, 3(1), 27-35. Chavez, F. P. (1995), A comparison of ship and satellite chlorophyll from California and Peru, J Geophys Res-Oceans, 100(C12), 24855-24862. Chavez, F. P., K. R. Buck, K. H. Coale, J. H. Martin, G. R. Ditullio, N. A. Welschmeyer, A. C. Jacobson, and R. T. Barber (1991), Growth-Rates, Grazing, Sinking, and Iron Limitation of Equatorial Pacific Phytoplankton, Limnol Oceanogr, 36(8), 1816-1833. Chazen, C. R., M. A. Altabet, and T. D. Herbert (2009), Abrupt mid-Holocene onset of centennial-scale climate variability on the Peru-Chile Margin, Geophysical Research Letters, 36. 168 DiNezio, P. N., A. C. Clement, G. A. Vecchi, B. J. Soden, and B. P. Kirtman (2009), Climate Response of the Equatorial Pacific to Global Warming, Journal of Climate, 22(18), 4873-4892. DiNezio, P. N., A. Clement, G. A. Vecchi, B. Soden, A. J. Broccoli, B. L. Otto-Bliesner, and P. Braconnot (2011), The response of the Walker circulation to Last Glacial Maximum forcing: Implications for detection in proxies, Paleoceanography, 26. Druffel, E. M. (1981), Radiocarbon in Annual Coral Rings from the Eastern Tropical Pacific-Ocean, Geophysical Research Letters, 8(1), 59-62. Druffel, E. M. (1995), Pacific Bomb Radiocarbon Coral Data, in IGBP PAGES/World Data Center-A for Paleoclimatology Data Contribution Series # 1995-011, NOAA/NGDC Paleoclimatology Program, Boulder, CO, USA. Druffel, E. R. M., S. Griffin, J. Hwang, T. Komada, S. R. Beaupre, K. C. Druffel- Rodriguez, G. M. Santos, and J. Southon (2004), Variability of monthly radiocarbon during the 1760s in corals from the Galapagos Islands, Radiocarbon, 46(2), 627-631. Dubois, N., M. Kienast, S. Kienast, S. E. Calvert, R. Francois, and R. F. Anderson (2010), Sedimentary opal records in the eastern equatorial Pacific: It is not all about leakage, Global Biogeochemical Cycles, 24. Etayo-Cadavid, M. F., C. F. T. Andrus, K. B. Jones, G. W. L. Hodgins, D. H. Sandweiss, S. Uceda-Castillo, and J. Quilter (2013), Marine radiocarbon reservoir age variation in Donax obesulus shells from northern Peru: Late Holocene evidence for extended El Nino, Geology, 41(5), 599-602. 169 Faul, K. L., A. C. Ravelo, and M. L. Delaney (2000), Reconstructions of upwelling, productivity, and photic zone depth in the eastern equatorial Pacific Ocean using planktonic foraminiferal stable isotopes and abundances, J Foramin Res, 30(2), 110- 125. Fiedler, P. C., and L. D. Talley (2006), Hydrography of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 143-180. Flores, J. A., M. A. Barcena, and F. J. Sierro (2000), Ocean-surface and wind dynamics in the Atlantic Ocean off Northwest Africa during the last 140 000 years, Palaeogeogr Palaeocl, 161(3-4), 459-478. Gill, E. C., Rajagopalan, B., Molnar, P., and T. M., Marchitto (2016), Reduced- dimension reconstruction of the equatorial Pacific SST and zonal wind fields over the past 10,000 years using Mg/Ca and alkenone records, Paleoceanography, doi:10.1002/2016PA002948. Gu, D., and G. Philander (1997), Interdecadal Climate Fluctuations That Depend on Exchanges Between the Tropics and Extratropics, Science, 275(5301), 805-807. Herbert, T. D., J. D. Schuffert, D. Thomas, C. Lange, A. Weinheimer, A. Peleo-Alampay, and J. C. Herguera (1998), Depth and seasonality of alkenone production along the California margin inferred from a core top transect, Paleoceanography, 13(3), 263- 271. Higginson, M. J., and M. A. Altabet (2004), Initial test of the silicic acid leakage hypothesis using sedimentary biomarkers, Geophysical Research Letters, 31(18). 170 Jones, K. B., G. W. L. Hodgins, D. L. Dettman, C. F. T. Andrus, A. Nelson, and M. F. Etayo-Cadavid (2007), Seasonal variations in Peruvian marine reservoir age from pre- bomb Argopecten purpuratus shell carbonate, Radiocarbon, 49(2), 877-888. Kalansky, J., Y. Rosenthal, T. Herbert, S. Bova, and M. Altabet (2015), Southern Ocean contributions to the Eastern Equatorial Pacific heat content during the Holocene, Earth and Planetary Science Letters, 424, 158-167. Kessler, W. S. (2006), The circulation of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 181-217. Kessler, W. S., L. M. Rothstein, and D. K. Chen (1998), The annual cycle of SST in the eastern tropical Pacific, diagnosed in an ocean GCM, Journal of Climate, 11(5), 777- 799. Kienast, S. S., M. Kienast, S. Jaccard, S. E. Calvert, and R. Francois (2006), Testing the silica leakage hypothesis with sedimentary opal records from the eastern equatorial Pacific over the last 150 kyrs, Geophysical Research Letters, 33(15). Koutavas, A., J. Lynch-Stieglitz, T. M. Marchitto, and J. P. Sachs (2002), El Nino-like pattern in ice age tropical Pacific sea surface temperature, Science, 297(5579), 226- 230. Koutavas, A., and J. Lynch-Stieglitz (2003), Glacial-interglacial dynamics of the eastern equatorial Pacific cold tongue Intertropical Convergence Zone system reconstructed from oxygen isotope records, Paleoceanography, 18(4). 171 Koutavas, A., and J. Lynch-Stieglitz (2004), Variability of the marine ITCZ over the eastern Pacific during the past 30,000 years: Regional perspective and global context, in The Hadley Circulation: Present, Past and Future, edited by H. F. Diaz and R. Bradley, pp. 347-369, Springer Netherlands. Le Borgne, R., Feely, R. A., and D. J. Mackey (2002), Carbon fluxes in the equatorial Pacific: a synthesis of the JGOFS programme, Deep-Sea Research II, 49(13-14), 2425-2442, doi:10.1016/S0967-0645(02)00043-7. Lisiecki, L. E., and M. E. Raymo [2005], A Pliocene-Pleistocene stack of 57 globally distributed benthic δ18O records, Paleoceanography, 20, PA1003, doi:10.1029/2004PA001071. Liu, Z. Y., and S. G. H. Philander (1995), How Different Wind Stress Patterns Affect the Tropical Subtropical Circulations of the Upper Ocean, J Phys Oceanogr, 25(4), 449- 462. Locarnini, R. A., A. V. Mishonov, J. I. Antonov, T. P. Boyer, H. E. Garcia, O. K. Baranova, M. M. Zweng, C. R. Paver, J. R. Reagan, D. R. Johnson, M. Hamilton, and D. Seidov, 2013. World Ocean Atlas 2013, Volume 1: Temperature. S. Levitus, Ed., A. Mishonov Technical Ed.; NOAA Atlas NESDIS 73, 40 pp. Lu, P., J. P. McCreary, and B. A. Klinger (1998), Meridional circulation cells and the source waters of the Pacific Equatorial Undercurrent, J Phys Oceanogr, 28(1), 62-84. Masson-Delmotte, V., et al. (2006), Past and future polar amplification of climate change: climate model intercomparisons and ice-core constraints (vol 26, pg 513, 2006), Clim Dynam, 27(4), 437-440. 172 Ma, C. C., C. R. Mechoso, A. W. Robertson, and A. Arakawa (1996), Peruvian stratus clouds and the tropical Pacific circulation: A coupled ocean-atmosphere GCM study, Journal of Climate, 9(7), 1635-1645. Molfino, B., and A. Mcintyre (1990), Precessional Forcing of Nutricline Dynamics in the Equatorial Atlantic, Science, 249(4970), 766-769. Müller, P.J., Kirst, G., Ruhland, G., von Storch, I., and A. Rosell-Melé (1998), Calibration of the alkenone paleotemperature index Uk'37 based on core-tops from the eastern South Atlantic and global ocean (60°N-60°S), Gechim. Cosmochim. Acta, 62(10), 1757-1772. Pena, L. D., I. Cacho, P. Ferretti, and M. A. Hall (2008), El Nino-Southern Oscillation- like variability during glacial terminations and interlatitudinal teleconnections, Paleoceanography, 23(3). Pennington, J. T., K. L. Mahoney, V. S. Kuwahara, D. D. Kolber, R. Calienes, and F. P. Chavez (2006), Primary production in the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 285-317. Peterson, L. C., G. H. Haug, K. A. Hughen, and U. Rohl (2000), Rapid changes in the hydrologic cycle of the tropical Atlantic during the last glacial, Science, 290(5498), 1947-1951. Philander, S. G. H. (1985), El-Nino and La-Nina, J Atmos Sci, 42(23), 2652-2662. Prahl F. G. and Wakeham S. G. (1987) Calibration of unsaturation patterns in long- chain ketone compositions for paleotemperature assessment. Nature 330, 367–369. 173 Qu, T. D., S. Gao, I. Fukumori, R. A. Fine, and E. J. Lindstrom (2009), Origin and Pathway of Equatorial 13 degrees C Water in the Pacific Identified by a Simulated Passive Tracer and Its Adjoint, J Phys Oceanogr, 39(8), 1836-1853. Qu, T. D., S. Gao, I. Fukumori, R. A. Fine, and E. J. Lindstrom (2010), The Obduction of Equatorial 13 degrees C Water in the Pacific Identified by a Simulated Passive Tracer, J Phys Oceanogr, 40(10), 2282-2297. Reimer, P. J., et al. (2013), Intcal13 and Marine13 Radiocarbon Age Calibration Curves 0-50,000 Years Cal Bp, Radiocarbon, 55(4), 1869-1887. Sarmiento, J. L., N. Gruber, M. A. Brzezinski, and J. P. Dunne (2004), High-latitude controls of thermocline nutrients and low latitude biological productivity, Nature, 427(6969), 56-60. Seager, R., and R. Murtugudde (1997), Ocean dynamics, thermocline adjustment, and regulation of tropical SST, Journal of Climate, 10(3), 521-534. Stott, L., C. Poulsen, S. Lund, and R. Thunell (2002), Super ENSO and global climate oscillations at millennial time scales, Science, 297(5579), 222-226. Taylor, R. E., and R. Berger (1967), Radiocarbon content of marine shells from the Pacific coasts of Central and South America, Science, 158, 1180-1182. Thierstein, H. R., K. R. Geitzenauer, and B. Molfino (1977), Global Synchroneity of Late Quaternary Coccolith Datum Levels - Validation by Oxygen Isotopes, Geology, 5(7), 400-404. Toggweiler, J. R., K. Dixon, and W. S. Broecker (1991), The Peru Upwelling and the Ventilation of the South Pacific Thermocline, Journal of Geophysical Research, 96(C11), 467-497. 174 Vecchi, G. A., and B. J. Soden (2007), Global warming and the weakening of the tropical circulation, Journal of Climate, 20(17), 4316-4340. Wang, C., and P. C. Fiedler (2006), ENSO variability and the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 239-266. Wang, C. Z., and D. B. Enfield (2001), The tropical Western Hemisphere warm pool, Geophysical Research Letters, 28(8), 1635-1638. Wara, M. W. (2006), Permanent El Niño-like conditions during the Pliocene warm period (29 July, pg 758, 2005), Science, 313(5794), 1739-1739. Weber, M. E., M. Wiedicke, V. Riech, and H. Erlenkeuser (1995), Carbonate Preservation History in the Peru Basin - Paleoceanographic Implications, Paleoceanography, 10(4), 775-800. Wyrtki, K. (1966), Oceanography of the eastern equatorial Pacific Ocean, Oceanography and Marine Biology, An Annual Review, 4, 33-68. 175 Figures: Figure 1. Locations of sediment cores used in this study overlaid on a map of mean annual surface ocean temperatures [Locarnini et al., 2013]. Black circle: ME0005A- 15MC/17JC, Green Circle: GGC 43/CDH 41, Pink circle: VNTR01-13GC, Blue circle: CDH 23/CDH 26. Map constructed using Ocean Data View software. 176 Figure 2. Age-depth models for sediment cores used in this study. All age models were constructed using the Clam age modeling software package for R [Blaauw, 2010]. Grey error envelopes represent one standard deviation. 177 Figure 3. Alkenone SST reconstructions from four sites in the EEP. The magnitude of warming during the deglaciation (11.8 to 18 ka) is larger and delayed at the two southernmost locations (pink: VNTR01-13GC, blue: CDH 23/CDH 26) relative to the GU (green: GGC 43/CDH 41) and EPWP (black: ME0005A-15MC/17JC) sites. 178 Figure 4. Organic biomarker abundances at five sites in the eastern equatorial Pacific spanning 4°S to 4.5°N, including a previously published record from ODP site 1240 (purple) [Calvo et al., 2011]. Alkenone abundances (dark) reflect coccolithophorid productivity and brassicasterol concentrations (light) reflect diatom production rates. 179 Figure 5. (top) Map of thermocline depth in the EEP between 20°S and 20°N with locations of multicore core top samples used in the N ratio-thermocline depth calibration plotted as black circles. White regions represent areas with low data coverage and thus bad estimates of thermocline depth. (bottom) Relationship between the N ratio of core 180 top samples and the modern thermocline depth at the core locations as estimated from the WOA 13 database [Locarini et al., 2013]. We find a significant relationship between the N ratio of core top samples and observed thermocline depth (R2=0.918, p<0.01). Samples collected near the Galapagos Islands (red) exhibit lower than expected N ratios given the modern thermocline depth at their locations. We exclude these data points from the calibration. 181 Figure 6. Thermocline depth histories at sites ME0005A-15MC/17JC (4.61°N), GGC 43/CDH 41 (1.25°S), and VNTR01-13GC (3.09°S). Thermocline depth is uniformly lower during the last glacial period and deglaciation and increases towards present day at all sites. Modern estimates, plotted on the y-axis, are from WOA13 database [Locarini et al., 2013]. Error bars represent one standard deviation. 182 Figure 7. Relationships between thermocline depth and mixed layer properties (SST and phytoplankton production). Thermocline depth and SST at the Galapagos site (GGC 43/CDH 41) are significantly correlated. However, the relationship between thermocline depth and coccolithphorid productivity is the inverse of that expected from ENSO dynamics. 183 Figure 8. Relationships between SST and phytoplankton productivity at four sites in the EEP. Open-ocean sites exhibit the expected relationship between SST and coccolithphorid and diatom production (a, b), with cool temperatures leading to higher rates of primary production for both groups. (c,d) The relationship between SST and coccolithophorid production near the Galapagos and off the coast of northern Peru exhibit different relationships during the LGM, deglaciation, and Holocene (0-11.8 ka). Both sites exhibit a statistically significant positive relationship between SST and alkenone abundances during the Holocene. At the Peru Margin we observe the inverse relationship during the LGM (older than 18 ka) and deglaciation (11.8-18 ka). Within the GU there is no statistically significant relationship between productivity and SST during the last deglaciation and LGM. 184 Table 1. Detailed information on sediment cores used in this study. 14 Core ID Latitude Longitude Water C Average depth(m) Reservoir Sedimentation Age (yrs) Rate (cm/kyr) ME0005A- 4.61°N 86.70°W 905 440 2 15MC/17MC KNR 195-5 GGC 1.25°S 89.69°W 617 500 19 43 KNR 195-5 CDH 41 1.27°S 89.70°W 595 500 21 KNR 195-5 CDH 23 3.75°S 81.13°W 373 500 127 KNR 195-5 CDH 26 3.99°S 81.31°W 1026 500 133 VNTR01-13GC 3.09°S 90.82°W 3304 535 4.5 185 CHAPTER FIVE ________________________________________________________________________ No role for the eastern equatorial Pacific biological pump in atmospheric CO2 change since the last glacial period Samantha C. Bova, Timothy D. Herbert, Caitlin Chazen, Angel Mojarro, and Jana Zech 1 Brown University, Department of Earth, Environmental, and Planetary Sciences, Providence, RI USA 186 Abstract Changes in the rates of phytoplankton production in the Equatorial Eastern Pacific (EEP) may play a critical role in modulating orbital-scale climate change, yet a clear picture of their productivity history remains elusive. Reconstructions of the change in productivity based on indicators of diatom (% opal, δ30Si, brassicasterol), coccolithophore (C37-alkenones, % calcite), and total algal production (Corg, carotenoid flux, δ15N) during the Last Glacial Maximum (LGM, 21 ka) and last deglaciation (18-11 ka) are inconsistent. It is unclear whether the disagreement between records arises from strong spatial heterogeneity in paleoproductivity, differing responses of algal groups to oceanographic and climatic drivers, or proxy biases. In this study we reconstruct coccolithophorid (C37-alkenones) and diatom paleoproductivity (brassicasterol) at multiple sites spread across the EEP. We find that total production increased during the last glacial period at open-ocean sites, likely due to aeolian iron fertilization, but we do not observe the expected shift to greater diatom relative to coccolithophorid productivity. Diatoms and coccolithophores change similarly during the last glacial period and deglaciation, indicating that shifts between calcareous and non-calcareous algal groups had a small impact on the biological carbon pump and, consequently, little effect on the evasion of CO2 from the EEP. Disagreement among paleoproductivity records arises from a combination of proxy bias and true regional heterogeneity. 1. Introduction Biological production changes in the eastern equatorial Pacific (EEP) are often implicated in mechanisms for glacial-interglacial variations in atmospheric carbon dioxide (CO2) [e.g. Matsumoto et al., 2002; 2008; Cermeño et al., 2008; Pichevin et al., 187 2009]. Today, the region is responsible for nearly two-thirds of the global marine CO2 flux to the atmosphere each year, despite composing less than 0.1% of the surface ocean [Takahashi et al., 2002]. High CO2 fluxes are the result of physical and biological processes; easterly and alongshore winds drive upwelling along the equator and the Peru Margin bringing cool, nutrient- and CO2-rich sub-thermocline waters to the surface ocean [Pennington et al., 2006; Brzezinski et al., 2011; Moore et al., 2013]. Currently, silicate and iron limitations restrict complete carbon uptake by phytoplankton, but, under (micro)nutrient replete conditions, such as those hypothesized during the last glacial period, an increase in the efficiency of phytoplankton production may have reversed the role of the EEP in the global carbon cycle and driven a reduction in atmospheric CO2 [e.g. Brezezinski et al., 2002; Matsumoto et al., 2002; 2008; Cermeño et al., 2008; Pichevin et al., 2009; Murray et al., 2012]. Importantly, the strength of the biological carbon pump is determined not by the total production but the relative export of organic carbon versus calcium carbonate from the surface ocean [e.g. Kohfeld et al., 2005]. The relative contribution of calcareous (coccolithophores) versus non-calcareous (diatoms) phytoplankton to export production alters surface ocean pCO2 [e.g. Calvo et al., 2011]. Diatoms and coccolithophores both take up CO2 during photosynthesis but coccolihophores also produce a carbonate skeleton that releases CO2 back into surface waters [e.g. Calvo et al., 2011]. A decrease in coccolithophores relative to diatoms therefore reduces the efficiency of the carbonate pump causing a decrease in surface ocean pCO2 and a reduction in atmospheric CO2 for the same production of organic carbon. Evidence suggests the relative abundance of diatoms versus coccolithophores is regulated by the supply of biologically important 188 nutrients, particularly silicate and iron, to the surface ocean, because nutrient replete conditions favor diatoms at the expense of coccolithophores (e.g. Matsumoto et al., 2002). Constraining the change in calcareous and non-calcareous phytoplankton productivity at the LGM compared to today is thus the subject of a large body of recent paleoceanographic work [e.g. Lyle et al., 1988; Farrell et al., 1995; Loubere et al., 2003; Loubere et al., 2004; Bradtmiller et al., 2006; Koutavas and Sachs, 2008; Chazen et al., 2009; Pichevin et al., 2009; Arellano-Torres et al., 2011; Dubois et al., 2010; Calvo et al., 2011]. Published paleoproductivity records show regionally heterogeneous changes during the deglaciation and LGM (Figure 1) [Loubere et al., 2003; Loubere et al., 2004; Bradtmiller et al., 2006; Koutavas and Sachs, 2008; Pichevin et al., 2009; Dubois et al., 2010; Arellano-Torres et al., 2011; Calvo et al., 2011], which makes it difficult to assess the role of the EEP in glacial-interglacial CO2 change. We propose that by incorporating spatial analysis of LGM and deglacial paleoproductivity we can provide a framework for explaining the diversity of changes observed in the region and for identifying the mechanisms for paleoproductivity changes. In this study, we analyze organic biomarkers of diatoms and coccolithophores from multiple sites, spanning 4°S to 4.6°N, and begin to reconcile these seemingly conflicting productivity records from the EEP. 2. Productivity in the EEP 2.1. Spatial Gradients Primary production in the EEP is a function of macronutrient and iron supply to the surface mixed layer [Pennington et al., 2006]. Macronutrients (nitrate, phosphate, and silicate), sourced primarily from subsurface intermediate waters formed in the northern 189 and southern high latitudes, are delivered to the surface ocean fueling high rates of productivity in coastal and equatorial upwelling centers [Sarmiento et al., 2004; Fiedler and Talley, 2006]. Despite the high rates of production, nitrate and phosphate are never fully depleted in EEP surface waters and the region is classified as a High-Nutrient Low- Chlorophyll region. It has been hypothesized that low iron concentrations limit growth [Brzezinski et al., 2011], particularly in EEP open ocean locations where aeolian dust is the only significant source of iron [Behrenfeld and Kolber, 1999]. Iron is less limiting along the continental margin and the Galapagos Islands, due to interactions of currents with margin sediments, riverine inputs, and upwelling from the relatively iron-rich Equatorial Undercurrent (EUC) [Coale et al., 1996; Gordon et al., 1997; Johnson et al., is 1999]. Variation in upwelling strength and iron availability subsequently lead to strong regional variations in productivity across the region. We identify four zones within the EEP based on physical oceanographic regimes and nutrient restrictions [Pennington et al., 2006]. These include: (1) the Peruvian Coastal Upwelling (PCU), (2) the Galapagos Upwelling (GU), (3) the Equatorial Upwelling (EU), and (4) the eastern Pacific warm pool (EPWP) (Figure 2). The PCU is the most productive zone in the EEP, yielding nearly 1.3 kg C m-3 yr-1. Here alongshore winds drive surface waters away from the South American coast in the PCU system, bringing cool, macronutrient- and iron-rich subsurface waters from depth to the surface ocean [Pennington et al., 2006]. Waters in the GU system are also highly productive, generating 0.7 kg C m-3 yr-1. Collision of the eastward flowing EUC with the islands drives perennial upwelling of modified subtropical mode water with contributions from southern ocean mode and intermediate waters (SOIW) [Kessler, 2006]. The EU zone and 190 the EPWP support lower levels of production, 0.23 and 0.18 kg C m-3 yr-1, respectively, because macronutrients and iron are more limited [Pennington et al., 2006]. Equatorial upwelling relies on easterly winds to drive divergence but, east of the Galapagos, the trades are primarily meridional. The EPWP lies beneath the ITCZ where low wind speeds, excess precipitation, and a seasonally large heat flux drive a strongly stratified surface layer [Kessler et al., 1998; Kessler, 2006]. 2.2 Temporal Variability ENSO is the primary source of temporal variation in productivity over the historical record, larger even than the seasonal cycle [Chavez, 1995; Pennington et al., 2006; Wang and Fiedler, 2006]. Its effects are felt most dramatically in the EU and coastal boundary regions (±1-2°C) and much less so in the EPWP [Fiedler and Talley, 2006]. ENSO drives strong and coherent changes in phytoplankton production. During an El Niño event, primary production rates and chlorophyll levels across the entire EEP decrease by several-fold. In coastal regions, the decline in phytoplankton abundance is accompanied by a shift in the community composition as well, with diatoms and other large-celled phytoplankton being replaced by picophytoplankton that are characteristic of the nutrient-limited open ocean [e.g. Chavez, 1989; Chavez et al., 1991]. Although more work has been done to characterize the impacts of El Niño, strong La Niña events also impact the regional ecology and trigger large phytoplankton blooms within the EU system [Behrenfeld et al., 2001]. 3. Site Locations In this study, we reconstruct the productivity history of each of the four key oceanographic zones of the EEP outlined above (Figure 2, 3). Core ME0005A- 191 15MC/17JC was recovered from the EPWP. Here, surface waters are sourced from the North Equatorial Countercurrent, which brings warm, nutrient poor surface waters to the core site [Kessler et al., 1998; Kessler, 2006]. Cores GGC 43 and CDH 41 were recovered on the Galapagos Platform just north of Española Island and records surface water change in the GU system. Surface waters around the islands reflect the mixed influence of local upwelling around the islands and the eastward flowing South Equatorial Current [Kessler, 2006]. Evolution of the PCU system is characterized by cores CDH 23/CDH 26, retrieved from the Gulf of Guayaquil. Coastal upwelling, as well as competition between north and south flowing surface currents affects the core sites [Kessler, 2006; Chaigneau et al., 2013]. Lastly, sites VNTR01-13GC together with previously published records from Calvo et al., [2011] at ODP site 1240 record the evolution of the EU system in the EEP. ODP 1240 sits within the heart of the EU, while site VNTR01-13GC tracks its southern margin and is also under the influence of the South Equatorial Current, which advects Peruvian coastal waters offshore [Kessler, 2006; Chaigneau et al., 2013]. 4. Methods 4.1. Age models Core chronologies were constructed by applying polynomial fits to sets of AMS 14 C dates on planktonic foraminifera using the Clam age model software [Blaauw 2010]. Radiocarbon measurements were made at the National Ocean Sciences Accelerator Mass Spectrometry facility at Woods Hole Oceanographic Institute. Planktonic foraminifera were picked from the >150 µm size fraction. All 14C dates were converted to calendar ages using the marine13 radiocarbon age calibration curve [Reimer et al., 2013]. We 192 assume constant reservoir ages at each of the sites based on the nearest modern measurements from the marine reservoir age database. For more details on the construction of the age-depth models used in this study see Chapter 4. 4.2 Biomarkers We analyzed C37-alkenones and brassicasterol concentrations at four new sites in the EEP: ME0005A-15MC/17JC, CDH 41/GGC 43, VNTR01-13GC, and CDH 23/CDH 26 (Figure 4). Alkenones are straight-chain hydrocarbons produced by the haptophyte algae Emiliania huxleyi and Gephyrocapsa oceanica, which represent the two dominant surface dwelling coccolithophores in the EEP [Okada and Honjo, 1973]. The concentration of C37-alkenones in the sediment tracks coccolithophorid production [e.g. Lee and Schneider, 2005] and perhaps even total production, including siliceous diatoms, as demonstrated by Bolton et al. [2010] in the late Pliocene EEP. Brassicasterol concentrations are used to trace diatom production rates [e.g. Higginson et al., 2004; Calvo et al., 2011]. Brassicasterol is the major sterol produced by some diatoms, but may also be generated by haptophyte algae and cryptophytes [Volkman, 1986; Rampen et al., 2010]. Therefore, the ratio of brassicasterol to alkenone concentrations must be used to evaluate shifts between coccolithophore versus diatom dominated production regimes. Deposition and preservation of alkenone and sterol biomarkers depend on a number of factors including zooplankton herbivory, sediment accumulation, and water column oxicity [Herbert et al., 2003]. Sterols are more labile than alkenones and are observed to decrease in relative abundance in the water column and into the sediments [Wakeham et al., 1997]. 193 One to two grams of freeze-dried sediment was extracted in a Dionex 200 Accelerated Solvent Extractor with dichloromethane. Samples from sites CDH 26, CDH 23, GGC 43 and CDH 41 were analyzed only for alkenone biomarkers. After extraction samples were evaporated to dryness under a stream of nitrogen gas and spiked with an internal standard before analysis on a Gas Chromatograph with a Flame Ionization Detector (GC-FID). Alkenone records from the northern Peru Margin sites (CDH 23 and CDH 26) were published previously by Bova et al., [2015] and provide a measurement approximately every 35 years. Samples from sites VNTR01-13GC and ME0005A-15MC/17JC were analyzed for both alkenones and brassicasterol (24-methylcholesta-5,22-dien-3β-ol) concentrations. After extraction these samples were reacted in 2 N potassium hydroxide in methanol and water (95:5) in order to hydrolyze esters and reduce interferences during gas chromatograph analysis. Samples were then further separated using silica gel chromatography with hexane, dichloromethane, and methanol as eluents. The methanol fraction was derivatized with bis(trimethlsilyl)trifluoroacetamide. The methanol and DCM fractions were then evaporated under a stream of nitrogen gas and reconstituted in toluene containing known concentrations of two internal alkane standards (n- hexatriacontane and n-heptatriacontane) before analysis on a GC-FID. Brassicasterol was not measured in cores from the Peru Margin and the Galapagos. Alkenones in the DCM fraction were analyzed using the method of Herbert et al., [1998]. Lab analytical error is equivalent to ±0.1°C based on 55 replicate analyses of a laboratory standard. Approximately 10% of our analyzed samples were run in duplicate with an average UK’37-based temperature reproducibility of ±0.05°C. Brassicasterol was 194 analyzed on an HP-1 column with an oven temperature programmed from 60°C (holding time of 0.8 minutes) to 180°C at 40°C/min, 180°C to 220°C at 10°C/min, 220°C to 315°C at 2°C/min, and 315°C to 325°C at 10°C/min. The brassicasterol compound was identified by analysis of a select number of samples on a gas chromatograph mass spectrometer. Analysis of synthetic brassicasterol confirmed peak identification. Thirty- nine brassicasterol samples were run in duplicate with an average reproducibility of ±6.5 ng/g. 5. Results 5.1 Paleoproductivity estimates Productivity records from across the EEP are regionally heterogenous. Open ocean sites (ME0005A-15MC/17JC and VNTR01-13GC) exhibit maxima in both C37- alkenone and brassicasterol abundances at the end of the last glacial period between approximately 15 and 31 ka, with a subsequent decline towards present day. The ratio of brassicasterol to alkenones remains less than 0.25 for the duration of both records, comparable to background levels at site ODP 1240 previously published by Calvo et al., 2011. Both sites, however, exhibit maximum brassicasterol to alkenone ratios between 12 and 8 ka, synchronous with maximum values in the record from ODP 1240. The Galapagos and Peru Margin sites exhibit markedly different productivity records than those from open ocean regions of the EEP (Figure 5). The Peru Margin site exhibits high alkenone abundances during the last glacial period, similar to the open ocean sites, but increases rather than decreases at the onset of the Holocene until 6.4 ka. Additionally, alkenone abundances are highly variable during the Holocene. At the Galapagos, alkenone abundances decrease between 27.9 and 22.0 ka from 6300 to 3200 195 ng/g and remain stable at 3900 ng/g from 22 to 11 ka. After 11 ka, alkenone abundances increase towards present day. 6. Discussion The modern EEP is characterized by sharp gradients in macro- and micro-nutrient availability determined largely by the wind-driven circulation and the proximity of coastal upwelling zones [Pennington et al., 2006]. Macro- and key micro-nutrients, such as iron, are supplied to the surface ocean via wind-driven upwelling of subsurface waters from beneath the thermocline. Iron is also delivered via aeolian dust deposition, but the supply is orders of magnitude lower than the subsurface supply [Coale et al., 1996; Gordon et al., 1997]. As a result, the highest rates of production in the modern EEP are centered in zones of strong coastal upwelling, such as the Peru Margin and the Galapagos (Figure 2) [Pennington et al., 2006]. Because productivity is so variable across the region today it should come as no surprise that glacial to Holocene changes in productivity are not uniform across the region (Figures 4, 5). Here, we demonstrate a dichotomy in paleoproductivity trends between open-ocean sites, more than 100 km offshore, and coastal upwelling systems. 6.1. Iron limitations in the open-ocean EEP and the efficiency of the biological carbon pump Paleoproductivity records from the EPWP (ME0005A-15MC/17JC) and within the EU system (VNTR01-13GC and ODP site 1240), all more than 100 km from the nearest coastline, exhibit maxima in diatom and coccolithophorid production during the last glacial period and early deglaciation between 15 and 33 ka (Figure 4). The near uniform response across a north to south transect (3°S to 4.6°N) in the open-ocean EEP 196 limits the possible causes of productivity change. We reject north-south competition of intermediate water masses and north-south shifts in wind-driven upwelling; both mechanisms should produce a north-south contrast in nutrient-availability, which we do not observe [Dubois et al., 2010; Kienast et al., 2006; Sarmiento et al., 2004]. Instead, a change in the macronutrient content of upwelled waters sourced from the Southern Ocean and/or variation in iron availability are possible explanations. Iron fertilization experiments demonstrate severe iron limitations in the modern open-ocean EEP [Hutchins and Bruland, 1998; Landry et al., 2000; Pennington et al., 2006]. After iron additions, the phytoplankton standing stock increases dramatically, with diatoms composing the majority of the bloom biomass [Hutchins and Bruland, 1998; Landry et al., 2000; Barber and Hiscock, 2006; Pennington et al., 2006]. The LGM represents a natural iron fertilization experiment. Aeolian iron inputs to the EEP increased 2 to 3-fold during the last glacial period in response to stronger glacial winds, a less robust hydrological cycle, and the expansion of desert source regions [Winckler et al., 2008]. Productivity records presented here are consistent with a rise in total production during the last glacial period. Organic biomarker concentrations recorded at open ocean sites ME0005-15MC/17JC, VNTR01-13GC, and ODP 1240 suggest production of both coccolithophores and diatoms reached a maximum during the last glacial period and early deglaciation synchronous with an increase in dust delivery to the EEP, 33 to 15 ka [Winckler et al., 2008; Calvo et al., 2011] (Figure 4). The tight correlation between C37-alkenone and brassicasterol abundances (ME0005A-15MC/17JC: R2=0.83, p<0.01, VNTR01-13GC: R2=0.38, p<0.01) demonstrates that diatoms and coccolithophores responded equivalently to iron additions, 197 leading to no significant change in the strength of the biological carbon pump and CO2 evasion in the EPWP and the southern margin of the EU system (Figure 6). This observation is in opposition to bottle experiments that show an increase in diatoms relative to coccolithophores when iron is more available [Hutchins and Bruland, 1998]. However, observations from the IronEx II mesoscale experiment suggest that, though iron additions do lead to greater diatom productivity, they do not lead to an increase in organic matter export relative to carbonate export. Efficient grazing of diatoms by protists recycles carbon in the surface mixed layer and maintains a low carbon export ratio [Landry et al., 2000]. Thus, despite greater iron inputs to the EEP during the LGM, the EEP biological carbon pump did not strengthen and therefore did not aid in glacial atmospheric CO2 drawdown. The early Holocene is instead the most likely interval for the EEP biological pump to sequester CO2, despite no apparent rise in iron availability [Winckler et al., 2008; Calvo et al., 2011]. Closer examination of the ratio of brassicasterol to C37- alkenones at sites ME0005A-15MC/17JC, VNTR01-13GC and ODP site 1240 [Calvo et al., 2011] reveals a coherent shift to enhanced diatom relative to coccolithophorid production between 12 and 8 ka (Figure 7). During this interval, silica leakage from the Southern Ocean at the end of the last deglaciation may have fueled the rise in diatom production [Anderson et al., 2009; Calvo et al., 2011]. Opal records from the Southern Ocean suggest surface waters around Antarctica, which feed the EEP upwelling systems, were more silica rich at this time [Anderson et al., 2009]. Although the relative increase at sites ME0005A-15MC/17JC and VNTR01-13GC is small, it is consistent with the much larger shift observed within the heart of the EU system at ODP site 1240. 198 Atmospheric CO2 decreased by 10 ppmv between 10.4 and 8 ka [Petit et al., 1999; Monnin et al., 2001] synchronous with the rise in the ratio of diatoms to coccolithophores and thus the strength of the EEP biological carbon pump (Figure 7). 6.3 Macro-nutrient limitations in the Peru and Galapagos Upwelling Systems In contrast to the open-ocean sites, phytoplankton living along the northern Peru Margin and the Galapagos Islands do not respond to changes in aeolian iron delivery over the past 30 kyrs (Figure 5). Paleoproduction rates in these locations instead reflect variations in macronutrient delivery via upwelling, zonal advection and/or variation in intermediate water nutrient content. Furthermore, because productivity in the Galapagos and Peru Margin upwelling systems has evolved differently, the nutrient supply at each site is likely controlled by different forcing mechanisms. Upwelling immediately to the west of the Galapagos results from the collision of the eastward flowing EUC with the islands and is a perennial feature of the EEP [Fiedler and Talley, 2006; Eden and Timmermann, 2004]. Compared to the rest of the EEP, we posit that delivery of subsurface waters to the surface ocean was relatively stable at this location through time. We therefore use the record of GU productivity as an indicator of intermediate water nutrient content and infer no change in source water nutrient content until 11 ka, when coccolithophorid productivity increased at the site (Figure 5). The increase in productivity after 11 ka likely reflects greater nutrient concentrations in EEP sub-thermocline waters due to nutrient leakage from the Southern Ocean [Anderson et al., 2009]. This is consistent with the observed rise in diatom to coccolithophorid productivity 12 ka at three open-ocean sites (Figure 7). Enhanced upwelling of sub- thermocline waters is not a good explanation for the observed increase in productivity 199 because the mean thermocline depth in the GU, the EPWP and EU increased beginning 13 ka and continued to increase towards present day [Chapter 4]. Paleoproductivity in the PCU system is influenced by a combination of source water nutrient change and local changes in the strength of wind-driven upwelling. Seasonal and interannual changes in the strength and direction of alongshore winds drive significant variations in regional productivity today [Pennington et al., 2006]. The paleoproductivity record from this site exhibits two major shifts over the past 30 ka: (1) at 18 ka, productivity decreases and (2) at 11 ka, productivity increases. The second shift aligns well with the rise in productivity observed in the GU system and an increase in diatoms relative to coccolithophores in the open-ocean EEP. Because the Galapagos and Peru Margin upwelling systems are fed by the same intermediate water sources, this change likely reflects the same increase in source water nutrient content. However, the first shift, 18 ka, does not align with any significant shift in productivity in the GU system. Thus, reduced production likely results from a decrease in local wind-driven upwelling and weaker along-shore winds. A significant negative relationship between SST and coccolithophorid production supports the role of local upwelling as it brings both cool and nutrient-rich waters to the surface ocean prior to 11 ka [Chapter 4]. We do not observe a similar relationship at the Galapagos. 6.2 Proxy Biases We began this paper by evaluating a map of LGM productivity anomalies that showed a regionally heterogeneous response to glacial boundary conditions, unlike that observed today (Figure 1). We determined that phytoplankton in open-ocean versus coastal margin sites experienced different nutrient conditions, which led to diverse 200 productivity trends over the past 30 ka. These differences explain some, but not all of the observed differences in paleoproduction records from the EEP. Inorganic proxy records of paleoproductivity in the open-ocean EEP based on the accumulation of calcite or opal [Loubere et al., 2004; Bradtmiller et al., 2006; Kienast et al., 2006] and benthic foraminiferal assemblages in the sediments [Loubere et al., 2003] are still at odds with organic biomarker and isotopic proxy records of production and nutrient utilization [Dubois et al., 2010; Pichevin et al., 2009; Calvo et al., 2011]. 230Th-normalized calcite and opal accumulation rates are 30 to 50% lower during the last glacial period relative to today and benthic transfer functions from the cold tongue indicate a 10 to 45% reduction in productivity [Loubere et al., 2004; Bradtmiller et al., 2006]. Preservation of proxy indicators may bias paleoproductivity records. For example, the accumulation of calcite and opal in the sediment is sensitive to production, but also to dissolution during descent and preservation in the sediments. Opal is particularly susceptible to dissolution, as it is everywhere undersaturated with respect to seawater. Dubois et al. [2010] re-evaluated Th-normalized opal flux records from the EEP and found that preservation was, in fact, more important than production. Of course, organic biomarkers are also subject to preservational biases [Wakeham et al., 1997; Herbert et al., 2003]. However, they are typically well-preserved in low-oxygen environments such as the EEP [Herbert et al., 2003] and also compare well to isotopic measures of nutrient utilization over the past 30 ka [Pichevin et al., 2009]. We therefore suggest that biomarker records provide more reliable records of paleoproduction in the EEP over the past 30 ka than inorganic indices. Aside from paleoproductivity records within coastal upwelling zones, once we remove inorganic proxy records from the map in Figure 1 we 201 find that productivity records consistently record greater total production at the LGM relative to the Holocene and today (Figure 8). Concluding Remarks Untangling the role of spatial variability, proxy bias, and ecological shifts in EEP paleoproductivity records has been a long-standing challenge for paleo-oceanographers. Paired diatom and coccolithophore biomarker records from multiple sites across the EEP suggest that differences between previously published proxy records reflect a combination of inter-proxy differences and spatial heterogeneity. Inorganic proxies are subject to dissolution and preservation biases in the sediment and confound regional interpretation of productivity change [Pichevin et al., 2009; Dubois et al., 2010]. Once removed, we find a regionally coherent productivity change in the LGM EEP. Total production, including coccolithophores and diatoms, increased during the LGM in the open-ocean in response to greater dust fluxes, while production within coastal upwelling regions did not (Figure 8). North-south shifts in upwelling zones and north-south competition between subsurface intermediate water masses were not the main drivers of paleoproductivity change in the open-ocean EEP, but may influence production in the comparatively iron-rich coastal upwelling zones. Our records further suggest that the EEP was not a sink for atmospheric CO2 during the last glacial period. CO2 evasion depends on the export ratio of calcareous (coccolithophore) to non-calcareous (diatom) phytoplankton; an increase in diatom abundances relative to coccolithophores will drive a decrease in carbon outgassing from the EEP and perhaps turn the region into a carbon sink. Despite experimental evidence for an increase in diatoms relative to coccolithophores upon iron addition [Landry et al., 202 2000; Brzezinski et al., 2011], biomarker records indicate that export of alkenone- producing coccolithophores from the surface mixed layer increased in phase with diatoms during the last glacial period despite greater iron availability. Our results therefore indicate that iron-fertilization of the oceans will not be an effective method for mitigating the ongoing rise in atmospheric CO2. 203 References Anderson, R. F., S. Ali, L. I. Bradtmiller, S. H. H. Nielson, M. Q. Fleisher, B. E. Anderson, and L. H. Burckle (2009), Wind-Driven Upwelling in the Southern Ocean and the Deglacial Rise in Atmospheric CO2, Science, 323, 1443-1448. Archer, D., and E. Maierreimer (1994), Effect of Deep-Sea Sedimentary Calcite Preservation on Atmospheric CO2 Concentration, Nature, 367(6460), 260-263. Arellano-Torres, E., L. E. Pichevin, and R. S. Ganeshram (2011), High-resolution opal records from the eastern tropical Pacific provide evidence for silicic acid leakage from HNLC regions during glacial periods, Quaternary Science Reviews, 30(9-10), 1112- 1121. Barber, R. T., and M. R. Hiscock (2006), A rising tide lifts all phytoplankton: Growth response of other phytoplankton taxa in diatom-dominated blooms, Global Biogeochemical Cycles, 20(4). Behrenfeld, M. J., and Z. S. Kolber (1999), Widespread iron limitation of phytoplankton in the South Pacific Ocean, Science, 283(5403), 840-843. Behrenfeld, M. J., et al. (2001), Biospheric primary production during an ENSO transition, Science, 291(5513), 2594-2597. Blaauw, M. (2010), Methods and code for 'classical' age-modelling of radiocarbon sequences, Quat Geochronol, 5(5), 512-518. Bolton, C. T., K. T. Lawrence, S. J. Gibbs, P. A. Wilson, L. C. Cleaveland, and T. D. Herbert (2010), Glacial-interglacial productivity changes recorded by alkenones and microfossils in late Pliocene eastern equatorial Pacific and Atlantic upwelling zones, Earth and Planetary Science Letters, 295(3-4), 401-411. 204 Bova, S. C., T. Herbert, Y. Rosenthal, J. Kalansky, M. Altabet, C. Chazen, A. Mojarro, and J. Zech (2015), Links between eastern equatorial Pacific stratification and atmospheric CO2 rise during the last deglaciation, Paleoceanography, 30(11), 1407- 1424. Bradtmiller, L. I., R. F. Anderson, M. Q. Fleisher, and L. H. Burckle (2006), Diatom productivity in the equatorial Pacific Ocean from the last glacial period to the present: A test of the silicic acid leakage hypothesis, Paleoceanography, 21(4). Brzezinski, M. A., C. J. Pride, V. M. Franck, D. M. Sigman, J. L. Sarmiento, K. Matsumoto, N. Gruber, G. H. Rau, and K. H. Coale (2002), A switch from Si(OH)4 to NO3- depletion in the glacial Southern Ocean, Geophysical Research Letters, 29(12). Brzezinski, M. A., et al. (2011), Co-limitation of diatoms by iron and silicic acid in the equatorial Pacific, Deep-Sea Res Pt Ii, 58(3-4), 493-511. Calvo, E., C. Pelejero, L. D. Pena, I. Cacho, and G. A. Logan (2011), Eastern Equatorial Pacific productivity and related CO2 changes since the last glacial period, Proceedings of the National Academy of Sciences, 108(14), 5. Cermeño, P., S. Dutkiewicz, R. P. Harris, M. Follows, O. Schofield, and P. G. Falkowski (2008), The role of nutricline depth in regulating the ocean carbon cycle, P Natl Acad Sci USA, 105(51), 20344-20349. Chaigneau, A., N. Dominguez, G. Eldin, L. Vasquez, R. Flores, C. Grados, and V. Echevin (2013), Near-coastal circulation in the Northern Humboldt Current System from shipboard ADCP data, J Geophys Res-Oceans, 118(10), 5251-5266. Chavez, F. P. (1989), Size distribution of phytoplankton in the central and eastern tropical Pacific, Global Biogeochemical Cycles, 3(1), 27-35. 205 Chavez, F. P. (1995), A comparison of ship and satellite chlorophyll from California and Peru, J Geophys Res-Oceans, 100(C12), 24855-24862. Chavez, F. P., K. R. Buck, K. H. Coale, J. H. Martin, G. R. Ditullio, N. A. Welschmeyer, A. C. Jacobson, and R. T. Barber (1991), Growth-Rates, Grazing, Sinking, and Iron Limitation of Equatorial Pacific Phytoplankton, Limnol Oceanogr, 36(8), 1816-1833. Chazen, C. R., M. A. Altabet, and T. D. Herbert (2009), Abrupt mid-Holocene onset of centennial-scale climate variability on the Peru-Chile Margin, Geophysical Research Letters, 36. Coale, K. H., S. E. Fitzwater, R. M. Gordon, K. S. Johnson, and R. T. Barber (1996), Control of community growth and export production by upwelled iron in the equatorial Pacific Ocean, Nature, 379(6566), 621-624. Dubois, N., M. Kienast, S. Kienast, S. E. Calvert, R. Francois, and R. F. Anderson (2010), Sedimentary opal records in the eastern equatorial Pacific: It is not all about leakage, Global Biogeochemical Cycles, 24. Eden, C., and A. Timmermann (2004), The influence of the Galapagos Islands on tropical temperatures, currents and the generation of tropical instability waves, Geophysical Research Letters, 31(15). Farrell, J. W., T. F. Pedersen, S. E. Calvert, and B. Nielson (1995), Glacial-interglacial changes in nutrient utilization in the equatorial Pacific Ocean, Nature, 377, 514-517. Fiedler, P. C., and L. D. Talley (2006), Hydrography of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 143-180. Gordon, R. M., K. H. Coale, and K. S. Johnson (1997), Iron distributions in the equatorial Pacific: Implications for new production, Limnol Oceanogr, 42(3), 419-431. 206 Herbert, T. D. (2003), Alkenone Paleotemperature Determinations, in Treatise on Geochemistry, edited by H. Elderfield, pp. 391-432, Elsevier. Herbert, T. D., J. D. Schuffert, D. Thomas, C. Lange, A. Weinheimer, A. Peleo-Alampay, and J. C. Herguera (1998), Depth and seasonality of alkenone production along the California margin inferred from a core top transect, Paleoceanography, 13(3), 263- 271. Higginson, M. J., and M. A. Altabet (2004), Initial test of the silicic acid leakage hypothesis using sedimentary biomarkers, Geophysical Research Letters, 31(18). Hutchins, D. A., and K. W. Bruland (1998), Iron-limited diatom growth and Si : N uptake ratios in a coastal upwelling regime, Nature, 393(6685), 561-564. Johnson, K. S., F. P. Chavez, and G. E. Friederich (1999), Continental-shelf sediment as a primary source of iron for coastal phytoplankton, Nature, 398(6729), 697-700. Kessler, W. S. (2006), The circulation of the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 181-217. Kessler, W. S., L. M. Rothstein, and D. K. Chen (1998), The annual cycle of SST in the eastern tropical Pacific, diagnosed in an ocean GCM, Journal of Climate, 11(5), 777- 799. Kienast, S. S., M. Kienast, S. Jaccard, S. E. Calvert, and R. Francois (2006), Testing the silica leakage hypothesis with sedimentary opal records from the eastern equatorial Pacific over the last 150 kyrs, Geophysical Research Letters, 33(15). Kohfeld, K. E., C. Le Quere, S. P. Harrison, and R. F. Anderson (2005), Role of marine biology in glacial-interglacial CO2 cycles, Science, 308(5718), 74-78. 207 Koutavas, A., and J. P. Sachs (2008), Northern timing of deglaciation in the eastern equatorial Pacific from alkenone paleothermometry, Paleoceanography, 23(4). Landry, M. R., M. E. Ondrusek, S. J. Tanner, S. L. Brown, J. Constantinou, R. R. Bidigare, K. H. Coale, and S. Fitzwater (2000), Biological response to iron fertilization in the eastern equatorial Pacific (IronEx II). I. Microplankton community abundances and biomass, Mar Ecol Prog Ser, 201, 27-42. Lee, K. D., and R. Schneider (2005), Alkenone production in the upper 200 m of the Pacific Ocean, Deep-Sea Research Part 1: Oceanographic Research Papers, 52(3), 443-456. Locarnini, R. A., A. V. Mishonov, J. I. Antonov, T. P. Boyer, H. E. Garcia, O. K. Baranova, M. M. Zweng, C. R. Paver, J. R. Reagan, D. R. Johnson, M. Hamilton, and D. Seidov, 2013. World Ocean Atlas 2013, Volume 1: Temperature. S. Levitus, Ed., A. Mishonov Technical Ed.; NOAA Atlas NESDIS 73, 40 pp. Loubere, P. (2003), Remote vs. local control of changes in eastern equatorial Pacific bioproductivity from the Last Glacial Maximum to the Present, Global and Planetary Change, 35(1-2), 113-126. Loubere, P., F. Mekik, R. Francois, and S. Pichat (2004), Export fluxes of calcite in the eastern equatorial Pacific from the Last Glacial Maximum to present, Paleoceanography, 19(2). Lyle, M. (1988), Climatically Forced Organic-Carbon Burial in Equatorial Atlantic and Pacific Oceans, Nature, 335(6190), 529-532. Matsumoto, K., and J. L. Sarmiento (2008), A corollary to the silicic acid leakage hypothesis, Paleoceanography, 23(2). 208 Matsumoto, K., J. L. Sarmiento, and M. A. Brzezinski (2002), Silicic acid leakage from the Southern Ocean: A possible explanation for glacial atmospheric pCO2, Global Biogeochemical Cycles, 16(3). Moore, C. M., et al. (2013), Processes and patterns of oceanic nutrient limitation, Nat Geosci, 6(9), 701-710. Monnin, E., A. Indermuhle, A. Dallenbach, J. Fluckiger, B. Stauffer, T.F. Stocker, D. Raynaud, and J.-M. Barnola. 2001. Atmospheric CO2 concentrations over the last glacial termination, Science, Vol. 291, pp. 112-114. Murray, R. W., M. Leinen, and C. W. Knowlton (2012), Links between iron input and opal deposition in the Pleistocene equatorial Pacific Ocean, Nat Geosci, 5(4), 270-274. Okada, H., and S. Honjo (1973), The distribution of oceanic coccolithophorids in the Pacific, Deep-Sea Research, 20, 355-374. Petit, J.R., J. Jouzel, D. Raynaud, N.I. Barkov, J.-M. Barnola, I. Basile, M. Benders, J. Chappellaz, M. Davis, G. Delayque, M. Delmotte, V.M. Kotlyakov, M. Legrand, V.Y. Lipenkov, C. Lorius, L. Pepin, C. Ritz, E. Saltzman, and M. Stievenard (1999), Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica, Nature 399, 429-436. Pennington, J. T., K. L. Mahoney, V. S. Kuwahara, D. D. Kolber, R. Calienes, and F. P. Chavez (2006), Primary production in the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 285-317. Pichevin, L. E., B. C. Reynolds, R. S. Ganeshram, I. Cacho, L. Pena, K. Keefe, and R. M. Ellam (2009), Enhanced carbon pump inferred from relaxation of nutrient limitation in the glacial ocean, Nature, 459(7250), 1114-1117. 209 Rampen, S. W., B. A. Abbas, S. Schouten, and J. S. S. Damste (2010), A comprehensive study of sterols in marine diatoms (Bacillariophyta): Implications for their use as tracers for diatom productivity, Limnol Oceanogr, 55(1), 91-105. Reimer, P. J., et al. (2013), Intcal13 and Marine13 Radiocarbon Age Calibration Curves 0-50,000 Years Cal Bp, Radiocarbon, 55(4), 1869-1887. Romero, O. E., G. Leduc, L. Vidal, and G. Fischer (2011), Millennial variability and long-term changes of the diatom production in the eastern equatorial Pacific during the last glacial cycle, Paleoceanography, 26. Sarmiento, J. L., N. Gruber, M. A. Brzezinski, and J. P. Dunne (2004), High-latitude controls of thermocline nutrients and low latitude biological productivity, Nature, 427(6969), 56-60. Takahashi, T., et al. (2002), Global sea-air CO2 flux based on climatological surface ocean pCO2, and seasonal biological and temperature effects, Deep-Sea Res Pt Ii, 49(9-10), 1601-1622. Takahashi, T., S.C. Sutherland and A. Kozyr, 2016. Global Ocean Surface Water Partial Pressure of CO2 Database: Measurements Performed During 1957-2015 (Version 2015). ORNL/CDIAC-161, NDP-088(V2015). Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tennessee, doi: 10.3334/CDIAC/OTG.NDP088(V2015). Volkman, J. K. (1986), A Review of Sterol Markers for Marine and Terrigenous Organic- Matter, Org Geochem, 9(2), 83-99. 210 Wakeham, S. G., J. I. Hedges, C. Lee, M. L. Peterson, and P. J. Hernes (1997), Compositions and transport of lipid biomarkers through the water column and surficial sediments of the equatorial Pacific Ocean, Deep-Sea Res Pt Ii, 44(9-10), 2131-2162. Wang, C., and P. C. Fiedler (2006), ENSO variability and the eastern tropical Pacific: A review, Progress In Oceanography, 69(2-4), 239-266. Winckler, G., R. F. Anderson, M. Q. Fleisher, D. Mcgee, and N. Mahowald (2008), Covariant glacial-interglacial dust fluxes in the equatorial Pacific and Antarctica, Science, 320(5872), 93-96. 211 Figures: Figure 1. Locations of records of past productivity in the EEP spanning at least the past 25 ka. Red symbols indicate a decrease in productivity in the LGM relative to Holocene conditions. Green symbols indicate an increase in productivity in the LGM relative to Holocene conditions. Yellow symbols indicate no difference. Paleoproductivity proxies included in this figure include: the C37-alkenones [Koutavas and Sachs, 2008; Calvo, et al., 2011; this study), brassicasterol (diatom biomarker) [Calvo et al., 2011; this study], 230 Th normalized opal and calcite fluxes [Loubere et al., 2004; Bradtmiller et al., 2006], opal mass accumulation rates [Arellano-Torres et al., 2011], benthic foraminifera assemblages [Loubere et al., 2003], diatom assemblages [Romero et al., 2011], silicon isotopes [Pichevin et al., 2009], and nitrogen isotopes [Pichevin et al., 2009]. Stars indicate locations that use opal or carbonate fluxes to infer paleoproduction. Re- 212 evaluation of these proxy records suggests they reflect preservation rather than production [Pichevin et al., 2009; Dubois et al., 2010]. 213 Figure 2. Map of mean annual chlorophyll concentrations in EEP surface waters (1 month – Aqua/MODIS). Black lines separate the region into four dynamically distinct oceanographic zones [Pennington et al., 2006]. Filled circles indicate locations of sediment cores used in this study. Black circle: ME0005A-15MC/17JC, Green Circle: GGC 43/CDH 41, Pink circle: VNTR01-13GC, Blue circle: CDH 23/CDH 26. Purple circle: ODP 1240 Map constructed using the GeoMapApp software package. 214 Figure 3. (left) Map of mean annual sea surface temperature [WOA13, Locarnini et al., 2013] in the EEP overlaid with the major surface currents that affect our study sites [Chaigneau et al., 2013]. (right) Map of the surface water partial pressure of CO2 in the EEP [Takahashi et al., 2016] overlaid with the major subsurface currents [Chaigneau et al., 2013]. Core sites are shown as filled circles. Black circle: ME0005A-15MC/17JC, Green Circle: GGC 43/CDH 41, Pink circle: VNTR01-13GC, Blue circle: CDH 23/CDH 26. Purple circle: ODP 1240 [Calvo et al., 2011]. Maps constructed using Ocean Data View software. EPCC: Ecuador-Peru Countercurrent; SEC: South Equatorial Current; NECC: North Equatorial Countercurrent; PCC: Peru Coastal Current; EUC: Equatorial Undercurrent; PCUC: Peru-Chile Undercurrent 215 Figure 4. (top three panels) Organic biomarker abundances at three open-ocean sites in the EEP spanning 3.1°S to 4.6°N, including a previously published record from ODP site 1240 (purple) [Calvo et al., 2011]. Alkenone abundances (dark) reflect coccolithophorid productivity and brassicasterol concentrations (light) reflect diatom production rates. 216 Black: ME0005-15MC/17JC; Pink: VNTR01-13GC (bottom panel) EEP dust records from the ODP site 1240 (C26-ol) [Calvo et al., 2011] and nearby site TTN013-PC72 (232Th flux) [Winckler et al., 2008]. Dust-borne iron inputs are several-fold higher during the last glacial period and may be the cause of increased productivity in the open-ocean EEP between 15 and 33 ka. 217 Figure 5. C37-alkenone abundances, a proxy for coccolithohorid productivity, at the Galapagos (top, green) and the Peru Margin (bottom, blue) study sites. Coccolithophorid production increases at the onset of the Holocene in both records, indicating an increase in the nutrient content of the source water for both upwelling systems. The earlier decrease in production 18 ka along the Peru Margin, however, cannot be explained by a change in source water nutrients and, instead, likely reflects a reduction in wind-driven upwelling at the site. 218 Figure 6. Relationship between coccolithophorid (alkenones) and diatom (brassicasterol) productivity at two sites north and south of the equator in the EEP. We observe a significant relationship between these two parameters at both sites, which suggests coccolithophores have kept pace with diatoms over the past 30 ka. The good correspondence also argues against preservational biases between biomarkers. The correlation is particularly strong at site ME0005A-15MC/17JC in the Eastern Pacific Warm Pool. Nutrients are much lower at this site and diatoms are consequently less abundant relative to other primary producers. It is therefore possible that haptophyte algae, rather than diatoms, are the dominant producer of brassicasterol at this site. 219 Figure 7. (top panel) The ratio of brassicasterol to C37-alkenone abundances at three EEP open-ocean sites (black: ME0005A-15MC/17JC; purple: ODP 1240 [Calvo et al., 2011]; pink: VNTR01-13GC) plotted with (bottom panel) the record of atmospheric CO2 over the past 30 kyrs from Epica Dome C in Antarctica [Monnin et al., 2001; Petit et al., 1999]. Higher values indicate shifts towards increasingly diatom dominated production regimes. Diatoms compose a greater proportion of the total production at all sites 220 between 8 and 12 ka, though the increase is modest at sites ME0005A-15MC/17JC and VNTR01-13GC. 221 Figure 8. Same as Figure 1 but locations for inorganic proxy paleoproductivity records, including % calcite, % opal, and benthic foraminiferal abundances are removed. These inorganic proxy indicators likely reflect preservation rather than production [Pichevin et al., 2009; Dubois et al., 2010]. Red symbols indicate a decrease in productivity in the LGM relative to Holocene conditions. Green symbols indicate an increase in productivity in the LGM relative to Holocene conditions. Yellow symbols indicate no significant difference. 222 APPENDIX A: Organic Geochemical Proxy Data CDH 23: Age (Fairbanks Age Depth Temp. Temp. C37total K Calibration) (Marine13, Clam) (cm) U 37’ (°C, Prahl) (°C, Muller) (nmol/g) 1273 932 28 0.814 22.78 23.32 2.19 1273 932 28 0.815 22.82 23.36 2.35 1326 983 32 0.816 22.86 23.41 2.82 1326 983 32 0.822 23.02 23.56 6.74 1379 1034 36 0.817 22.87 23.42 3.57 1379 1034 36 0.818 22.92 23.46 5.77 1431 1085 40 0.825 23.11 23.66 3.07 1483 1136 44 0.822 23.02 23.56 2.60 1483 1136 44 0.828 23.19 23.74 6.83 1534 1186 48 0.818 22.92 23.46 2.81 1585 1236 52 0.813 22.77 23.31 2.80 1585 1236 52 0.820 22.98 23.53 7.62 1636 1286 56 0.809 22.66 23.19 2.87 1687 1335 60 0.808 22.61 23.14 2.02 1687 1335 60 0.814 22.79 23.33 6.23 1737 1384 64 0.809 22.63 23.17 2.29 1786 1433 68 0.821 22.99 23.53 7.71 1786 1433 68 0.819 22.94 23.48 7.71 1786 1433 68 0.811 22.71 23.25 2.84 1786 1433 68 0.814 22.78 23.32 5.00 1836 1481 72 0.815 22.82 23.36 3.46 1836 1481 72 0.811 22.71 23.25 2.69 1836 1481 72 0.814 22.79 23.33 2.31 1885 1530 76 0.807 22.59 23.12 2.64 1885 1530 76 0.818 22.90 23.44 6.69 1934 1578 80 0.805 22.53 23.06 4.07 1934 1578 80 0.812 22.73 23.27 6.31 1983 1625 84 0.804 22.50 23.03 4.11 1983 1625 84 0.808 22.63 23.16 5.21 2031 1672 88 0.802 22.45 22.98 4.02 2079 1719 92 0.800 22.37 22.90 3.04 2079 1719 92 0.806 22.55 23.08 7.33 2126 1766 96 0.800 22.37 22.90 3.12 2174 1813 100 0.804 22.50 23.03 5.35 2174 1813 100 0.801 22.42 22.95 3.59 2221 1859 104 0.802 22.44 22.96 3.37 2267 1905 108 0.803 22.48 23.01 3.67 2314 1950 112 0.802 22.44 22.97 3.68 2314 1950 112 0.808 22.62 23.15 4.43 2360 1996 116 0.803 22.48 23.01 3.63 2360 1996 116 0.809 22.66 23.19 4.36 2383 2018 118 0.810 22.66 23.20 7.08 2406 2041 120 0.803 22.47 23.00 5.62 223 2406 2041 120 0.804 22.51 23.04 4.63 2452 2086 124 0.807 22.58 23.11 4.45 2497 2131 128 0.808 22.61 23.14 5.41 2542 2175 132 0.805 22.52 23.05 4.78 2587 2219 136 0.802 22.45 22.98 6.11 2631 2263 140 0.805 22.54 23.08 5.37 2676 2307 144 0.808 22.62 23.15 4.03 2720 2350 148 0.811 22.70 23.24 5.97 2763 2393 152 0.810 22.67 23.20 4.46 2763 2393 152 0.813 22.77 23.31 4.38 2807 2436 156 0.813 22.75 23.29 3.89 2850 2479 160 0.813 22.77 23.31 3.59 2893 2521 164 0.814 22.81 23.35 3.93 2936 2564 168 0.813 22.78 23.32 4.40 2979 2606 172 0.812 22.74 23.28 3.51 3021 2648 176 0.812 22.74 23.28 3.93 3063 2689 180 0.815 22.82 23.36 3.83 3063 2689 180 0.822 23.03 23.57 6.03 3105 2731 184 0.820 22.97 23.51 6.48 3105 2731 184 0.816 22.85 23.39 6.60 3105 2731 184 0.821 23.01 23.56 5.52 3105 2731 184 0.823 23.05 23.60 5.61 3147 2772 188 0.817 22.88 23.42 7.26 3188 2813 192 0.818 22.92 23.46 6.56 3229 2854 196 0.820 22.97 23.52 6.38 3270 2894 200 0.820 22.96 23.51 7.18 3291 2915 202 0.817 22.87 23.42 6.87 3291 2915 202 0.817 22.89 23.43 7.05 3311 2935 204 0.824 23.10 23.65 4.31 3352 2975 208 0.823 23.05 23.60 6.97 3392 3015 212 0.820 22.97 23.51 6.26 3432 3055 216 0.818 22.90 23.44 4.97 3472 3095 220 0.819 22.94 23.49 6.39 3472 3095 220 0.819 22.93 23.47 6.31 3482 3104 221 0.818 22.90 23.44 6.27 3512 3134 224 0.819 22.93 23.47 6.79 3552 3173 228 0.817 22.88 23.42 6.52 3591 3212 232 0.825 23.12 23.67 6.07 3630 3251 236 0.822 23.03 23.58 6.29 3669 3290 240 0.821 23.00 23.55 5.95 3708 3329 244 0.822 23.02 23.57 5.23 3747 3367 248 0.823 23.05 23.60 3756 3377 249 0.821 22.99 23.53 7.35 3785 3405 252 0.821 23.01 23.56 6.68 3824 3444 256 0.824 23.07 23.62 6.24 3900 3519 264 0.821 22.99 23.54 6.40 3900 3519 264 0.814 22.80 23.34 4.36 3900 3519 264 0.816 22.85 23.39 5.00 3938 3557 268 0.815 22.83 23.37 4.61 224 3975 3594 272 0.807 22.60 23.13 4.20 4013 3632 276 0.810 22.66 23.20 5.06 4050 3669 280 0.817 22.88 23.42 5.00 4087 3706 284 0.815 22.82 23.36 4.52 4124 3743 288 0.814 22.79 23.33 4.17 4161 3780 292 0.809 22.66 23.19 3.47 4198 3816 296 0.812 22.75 23.29 3.91 4198 3816 296 0.818 22.92 23.46 7.18 4235 3853 300 0.817 22.87 23.41 4.32 4235 3853 300 0.815 22.82 23.36 4.65 4271 3889 304 0.814 22.80 23.34 4.16 4308 3925 308 0.816 22.84 23.38 4.27 4344 3961 312 0.815 22.81 23.35 4.88 4380 3997 316 0.815 22.82 23.36 4.05 4416 4033 320 0.811 22.71 23.25 4.45 4416 4033 320 0.812 22.73 23.26 7.17 4452 4069 324 0.814 22.79 23.33 4.14 4461 4078 325 0.805 22.53 23.06 7.20 4487 4104 328 0.807 22.58 23.11 5.40 4523 4140 332 0.808 22.62 23.16 4.48 4541 4158 334 0.814 22.78 23.32 5.22 4541 4158 334 0.807 22.58 23.12 4.11 4576 4193 338 0.815 22.82 23.36 5.43 4611 4228 342 0.811 22.71 23.24 5.70 4647 4263 346 0.808 22.60 23.14 5.35 4682 4298 350 0.813 22.75 23.29 5.68 4682 4298 350 0.805 22.54 23.07 6.16 4682 4298 350 0.805 22.54 23.07 5.68 4708 4324 353 0.805 22.53 23.06 6.87 4717 4333 354 0.814 22.80 23.34 5.87 4751 4368 358 0.813 22.76 23.30 5.86 4786 4402 362 0.813 22.75 23.29 5.78 4786 4402 362 0.812 22.72 23.26 5.85 4821 4437 366 0.813 22.76 23.30 5.56 4855 4471 370 0.811 22.71 23.24 6.24 4890 4506 374 0.814 22.79 23.33 5.81 4924 4540 378 0.813 22.77 23.31 4.47 4958 4574 382 0.815 22.81 23.35 5.43 4993 4608 386 0.816 22.86 23.41 5.63 5027 4642 390 0.815 22.82 23.36 5.16 5044 4659 392 0.814 22.80 23.34 4.99 5044 4659 392 0.811 22.70 23.24 4.97 5052 4668 393 0.815 22.81 23.35 7.50 5078 4693 396 0.817 22.89 23.43 5.03 5112 4727 400 0.822 23.02 23.57 4.77 5145 4761 404 0.820 22.97 23.52 5.28 5179 4794 408 0.821 22.99 23.53 5.08 5213 4828 412 0.819 22.95 23.50 5.27 5238 4853 415 0.816 22.86 23.40 4.85 225 5246 4861 416 0.819 22.95 23.50 4.42 5280 4895 420 0.819 22.94 23.48 4.46 5288 4903 421 0.821 23.00 23.54 6.15 5288 4903 421 0.819 22.94 23.49 4.56 5313 4928 424 0.824 23.08 23.63 6.45 5313 4928 424 0.823 23.06 23.61 6.85 5346 4961 428 0.825 23.10 23.65 5.86 5380 4994 432 0.824 23.09 23.64 5.62 5413 5027 436 0.825 23.13 23.68 6.04 5446 5061 440 0.827 23.18 23.73 5.61 5463 5077 442 0.823 23.05 23.60 7.14 5479 5094 444 0.821 23.00 23.55 7.57 5479 5094 444 0.829 23.22 23.78 5.65 5512 5126 448 0.829 23.23 23.78 5.60 5545 5159 452 0.827 23.18 23.73 6.36 5578 5192 456 0.829 23.23 23.78 5.25 5578 5192 456 0.826 23.16 23.71 5.17 5611 5225 460 0.829 23.24 23.80 6.69 5644 5258 464 0.833 23.34 23.90 5.84 5676 5290 468 0.830 23.25 23.80 6.85 5709 5323 472 0.831 23.29 23.85 5.46 5742 5355 476 0.827 23.17 23.72 6.21 5774 5388 480 0.829 23.23 23.78 6.16 5807 5420 484 0.831 23.30 23.85 5.58 5840 5453 488 0.826 23.15 23.70 5.85 5872 5485 492 0.833 23.35 23.91 6.03 5872 5485 492 0.831 23.29 23.84 6.03 5904 5517 496 0.831 23.28 23.84 5.10 5937 5550 500 0.836 23.43 23.99 5.51 5969 5582 504 0.833 23.37 23.92 5.21 6002 5614 508 0.832 23.31 23.87 6.10 6034 5646 512 0.832 23.32 23.87 6.39 6066 5678 516 0.835 23.41 23.97 5.94 6098 5710 520 0.833 23.36 23.92 6.30 6131 5742 524 0.830 23.26 23.81 5.50 6163 5774 528 0.833 23.34 23.90 5.87 6195 5806 532 0.833 23.35 23.91 5.40 6211 5822 534 0.831 23.29 23.85 5.31 6211 5822 534 0.831 23.29 23.84 5.28 6243 5854 538 0.837 23.47 24.03 4.80 6275 5886 542 0.834 23.38 23.93 5.62 6307 5918 546 0.834 23.37 23.93 5.08 6339 5950 550 0.832 23.32 23.88 4.37 6372 5982 554 0.827 23.18 23.73 5.05 6404 6013 558 0.832 23.32 23.88 4.78 6436 6045 562 0.831 23.29 23.85 5.85 6468 6077 566 0.832 23.31 23.87 5.60 6500 6109 570 0.832 23.33 23.89 5.65 6508 6117 571 0.827 23.18 23.73 8.28 226 6532 6140 574 0.832 23.33 23.89 5.31 6564 6172 578 0.826 23.16 23.71 4.02 6564 6172 578 0.830 23.25 23.81 5.86 6595 6204 582 0.829 23.23 23.78 5.22 6627 6235 586 0.829 23.22 23.77 5.76 6659 6267 590 0.827 23.18 23.73 4.71 6691 6299 594 0.831 23.30 23.85 7.28 6691 6299 594 0.828 23.22 23.77 7.39 6723 6330 598 0.828 23.20 23.75 6.93 6755 6362 602 0.828 23.22 23.77 7.30 6763 6370 603 0.822 23.02 23.57 8.89 6763 6370 603 0.821 22.99 23.54 8.48 6787 6394 606 0.818 22.90 23.45 6.88 6819 6425 610 0.823 23.07 23.62 6.96 6851 6457 614 0.823 23.07 23.62 7.63 6883 6488 618 0.824 23.10 23.65 6.55 6883 6488 618 0.823 23.07 23.62 7.06 6915 6520 622 0.823 23.07 23.62 7.50 6947 6552 626 0.821 23.01 23.55 7.12 6979 6583 630 0.823 23.06 23.61 7.85 7011 6615 634 0.822 23.03 23.58 6.86 7043 6646 638 0.820 22.97 23.51 6.28 7074 6678 642 0.820 22.98 23.53 6.03 7106 6709 646 0.819 22.94 23.49 5.31 7106 6709 646 0.820 22.96 23.50 5.52 7138 6741 650 0.818 22.90 23.44 2.88 7170 6773 654 0.819 22.93 23.48 3.96 7202 6804 658 0.819 22.95 23.49 5.13 7234 6836 662 0.819 22.95 23.50 6.52 7266 6867 666 0.819 22.95 23.50 5.59 7298 6899 670 0.822 23.04 23.59 5.68 7330 6930 674 0.824 23.08 23.62 4.56 7362 6962 678 0.824 23.10 23.65 5.47 7394 6994 682 0.820 22.97 23.51 3.71 7426 7025 686 0.821 23.01 23.55 2.77 7442 7041 688 0.820 22.98 23.53 5.13 7475 7072 692 0.811 22.71 23.24 6.22 7475 7072 692 0.810 22.67 23.20 6.49 7507 7104 696 0.815 22.81 23.35 5.86 7539 7136 700 0.816 22.85 23.40 5.19 7555 7152 702 0.805 22.52 23.05 6.20 7571 7167 704 0.813 22.76 23.29 5.23 7603 7199 708 0.808 22.62 23.15 5.63 7635 7231 712 0.807 22.58 23.12 6.81 7667 7262 716 0.807 22.58 23.12 6.41 7700 7294 720 0.804 22.50 23.03 5.79 7732 7326 724 0.803 22.46 22.99 6.17 7764 7357 728 0.805 22.52 23.05 3.64 7796 7389 732 0.801 22.41 22.94 6.22 227 7829 7421 736 0.800 22.37 22.90 3.65 7829 7421 736 0.800 22.40 22.92 3.76 7861 7452 740 0.801 22.41 22.93 4.56 7893 7484 744 0.797 22.28 22.81 4.23 7925 7516 748 0.797 22.28 22.80 3.99 7958 7548 752 0.801 22.41 22.94 3.94 7990 7579 756 0.803 22.48 23.01 4.00 8023 7611 760 0.799 22.36 22.88 4.86 8023 7611 760 0.796 22.25 22.78 4.88 8055 7643 764 0.802 22.44 22.97 3.74 8087 7675 768 0.802 22.44 22.96 3.50 8120 7707 772 0.794 22.20 22.72 3.54 8152 7739 776 0.794 22.21 22.73 3.46 8217 7802 784 0.790 22.08 22.59 4.64 8274 7858 791 0.795 22.25 22.77 6.24 8283 7866 792 0.789 22.07 22.59 4.85 8315 7898 796 0.790 22.10 22.61 4.58 8348 7930 800 0.794 22.21 22.73 4.61 8380 7962 804 0.795 22.23 22.76 4.71 8413 7994 808 0.796 22.28 22.80 4.44 8446 8026 812 0.791 22.11 22.63 4.49 8479 8058 816 0.797 22.29 22.81 4.69 8511 8090 820 0.801 22.40 22.92 4.26 8511 8090 820 0.801 22.41 22.93 4.35 8536 8114 823 0.791 22.12 22.64 3.54 8544 8122 824 0.794 22.22 22.74 5.06 8552 8130 825 0.795 22.24 22.76 5.15 8577 8154 828 0.794 22.20 22.72 4.78 8610 8186 832 0.790 22.10 22.61 4.81 8642 8218 836 0.785 21.93 22.44 5.63 8675 8250 840 0.785 21.95 22.46 3.96 8675 8250 840 0.784 21.92 22.43 3.88 8708 8283 844 0.786 21.98 22.49 4.26 8741 8315 848 0.782 21.85 22.36 4.21 8774 8347 852 0.778 21.74 22.25 3.92 8807 8379 856 0.779 21.77 22.28 4.26 8840 8411 860 0.780 21.78 22.29 3.70 8873 8443 864 0.792 22.14 22.66 3.80 8906 8476 868 0.781 21.82 22.33 3.63 8939 8508 872 0.777 21.71 22.22 4.14 8947 8516 873 0.790 22.10 22.62 5.02 8972 8540 876 0.780 21.80 22.31 3.66 9005 8572 880 0.781 21.83 22.34 3.75 9038 8605 884 0.781 21.82 22.33 3.85 9038 8605 884 0.781 21.83 22.34 3.74 9071 8637 888 0.782 21.87 22.38 3.12 9104 8669 892 0.779 21.76 22.26 4.17 9137 8702 896 0.779 21.77 22.27 3.97 9170 8734 900 0.780 21.80 22.31 3.54 228 9204 8766 904 0.785 21.94 22.45 3.33 9237 8799 908 0.784 21.91 22.43 3.45 9270 8831 912 0.778 21.72 22.23 3.91 9303 8864 916 0.779 21.77 22.28 3.71 9303 8864 916 0.778 21.74 22.25 3.74 9336 8896 920 0.783 21.89 22.40 3.11 9370 8928 924 0.782 21.84 22.35 2.82 9403 8961 928 0.779 21.76 22.27 4.42 9403 8961 928 0.777 21.71 22.21 4.29 9436 8993 932 0.784 21.90 22.41 3.35 9469 9026 936 0.782 21.87 22.38 3.81 9503 9058 940 0.780 21.80 22.31 3.37 9536 9091 944 0.778 21.75 22.26 4.43 9569 9123 948 0.785 21.94 22.46 3.74 9603 9156 952 0.785 21.94 22.46 3.89 9636 9188 956 0.778 21.74 22.25 4.38 9669 9221 960 0.777 21.71 22.21 3.47 9703 9253 964 0.778 21.74 22.25 4.45 9703 9253 964 0.776 21.68 22.19 4.29 9736 9286 968 0.774 21.62 22.12 3.99 9769 9318 972 0.777 21.69 22.20 4.00 9803 9351 976 0.777 21.70 22.20 3.44 9836 9383 980 0.774 21.62 22.13 3.94 9869 9416 984 0.778 21.74 22.25 3.85 9903 9448 988 0.776 21.68 22.19 2.94 9936 9481 992 0.772 21.57 22.07 3.84 9953 9497 994 0.767 21.40 21.90 2.87 9953 9497 994 0.765 21.35 21.85 2.84 9986 9530 998 0.773 21.59 22.09 3.50 10020 9562 1002 0.773 21.59 22.09 3.76 10020 9562 1002 0.772 21.57 22.07 3.80 10086 9627 1010 0.774 21.61 22.12 3.34 10120 9660 1014 0.769 21.48 21.98 3.74 10153 9693 1018 0.765 21.37 21.86 3.32 10220 9758 1026 0.771 21.53 22.03 3.11 10253 9790 1030 0.771 21.52 22.02 3.20 10286 9823 1034 0.778 21.74 22.24 3.51 10320 9855 1038 0.780 21.79 22.29 2.99 10320 9855 1038 0.778 21.74 22.25 3.09 10353 9888 1042 0.788 22.03 22.55 3.05 10386 9920 1046 0.786 21.97 22.49 2.89 10386 9920 1046 0.789 22.05 22.57 3.31 10419 9953 1050 0.789 22.05 22.57 2.92 10419 9953 1050 0.785 21.93 22.44 3.10 10453 9985 1054 0.783 21.87 22.39 3.24 10453 9985 1054 0.780 21.79 22.30 3.40 10486 10018 1058 0.786 21.97 22.48 3.17 10519 10050 1062 0.788 22.02 22.53 2.79 10519 10050 1062 0.786 21.98 22.50 2.99 229 10585 10115 1070 0.779 21.76 22.27 3.19 10585 10115 1070 0.785 21.95 22.46 3.01 10619 10147 1074 0.779 21.78 22.29 2.95 10652 10180 1078 0.783 21.88 22.39 3.23 10685 10212 1082 0.785 21.93 22.45 2.98 10718 10244 1086 0.781 21.82 22.33 2.82 10751 10277 1090 0.784 21.91 22.42 2.94 10784 10309 1094 0.787 22.00 22.51 2.73 10817 10341 1098 0.786 21.96 22.48 3.07 10850 10374 1102 0.788 22.02 22.54 2.61 10883 10406 1106 0.788 22.02 22.53 1.95 10916 10438 1110 0.784 21.90 22.42 2.69 10948 10470 1114 0.788 22.04 22.56 2.62 10981 10502 1118 0.786 21.98 22.49 2.29 11014 10534 1122 0.782 21.85 22.36 2.12 11047 10566 1126 0.789 22.07 22.59 1.80 11079 10599 1130 0.782 21.86 22.37 2.18 11144 10662 1138 0.784 21.92 22.43 2.18 11209 10726 1146 0.780 21.79 22.30 2.38 11218 10734 1147 0.781 21.84 22.35 2.27 11250 10766 1151 0.783 21.89 22.40 2.29 11282 10798 1155 0.780 21.79 22.30 2.25 11347 10861 1163 0.773 21.59 22.09 2.98 11347 10861 1163 0.771 21.52 22.02 2.94 11379 10893 1167 0.773 21.60 22.10 2.82 11411 10925 1171 0.769 21.48 21.97 2.39 11411 10925 1171 0.771 21.54 22.04 2.45 11507 11019 1183 0.770 21.50 22.00 3.11 11539 11051 1187 0.760 21.21 21.70 2.67 11570 11082 1191 0.763 21.30 21.80 1.76 11602 11113 1195 0.763 21.29 21.78 2.43 11634 11145 1199 0.763 21.31 21.80 2.97 11665 11176 1203 0.763 21.30 21.80 3.07 11665 11176 1203 0.765 21.36 21.85 3.05 11665 11176 1203 0.765 21.36 21.86 3.00 11697 11207 1207 0.759 21.17 21.66 3.50 11759 11269 1215 0.761 21.24 21.74 3.50 11791 11300 1219 0.766 21.39 21.89 3.09 11791 11300 1219 0.763 21.29 21.78 3.12 11822 11331 1223 0.756 21.08 21.57 2.92 11853 11362 1227 0.755 21.05 21.54 2.98 11884 11392 1231 0.750 20.91 21.40 3.21 11884 11392 1231 0.752 20.96 21.45 3.48 11914 11423 1235 0.757 21.13 21.61 3.09 11914 11423 1235 0.761 21.23 21.72 3.06 11914 11423 1235 0.757 21.11 21.60 3.06 11945 11453 1239 0.755 21.07 21.55 2.92 11945 11453 1239 0.763 21.29 21.78 2.82 11976 11484 1243 0.752 20.98 21.46 2.88 230 12006 11514 1247 0.755 21.07 21.56 1.95 12006 11514 1247 0.759 21.18 21.67 2.80 12037 11545 1251 0.759 21.18 21.67 2.31 12037 11545 1251 0.750 20.91 21.39 3.57 12060 11567 1254 0.752 20.98 21.47 3.58 12097 11605 1259 0.755 21.04 21.53 2.68 12127 11635 1263 0.761 21.25 21.74 2.73 12127 11635 1263 0.756 21.09 21.58 2.69 12127 11635 1263 0.754 21.02 21.50 2.73 12157 11665 1267 0.754 21.04 21.53 2.53 12210 11717 1274 0.752 20.96 21.45 3.84 12217 11725 1275 0.747 20.83 21.31 2.76 12276 11784 1283 0.747 20.83 21.31 2.96 12276 11784 1283 0.751 20.95 21.44 2.95 12305 11814 1287 0.750 20.90 21.38 3.07 12335 11843 1291 0.753 20.99 21.47 3.95 12364 11872 1295 0.749 20.89 21.38 12393 11901 1299 0.753 21.01 21.50 3.07 12429 11938 1304 0.755 21.06 21.55 3.36 12486 11996 1312 0.753 21.01 21.49 3.40 12514 12024 1316 0.754 21.03 21.51 3.17 12543 12053 1320 0.752 20.98 21.46 2.41 12571 12081 1324 0.755 21.05 21.54 3.36 12626 12138 1332 0.736 20.51 20.98 2.91 12654 12166 1336 0.738 20.56 21.03 2.61 12682 12194 1340 0.729 20.30 20.76 3.47 12709 12222 1344 0.731 20.35 20.82 3.05 12736 12250 1348 0.738 20.56 21.03 2.93 12763 12277 1352 0.748 20.84 21.32 2.28 12790 12305 1356 0.745 20.78 21.25 2.04 12816 12332 1360 0.752 20.96 21.44 2.23 12843 12359 1364 0.754 21.04 21.52 2.02 12869 12386 1368 0.752 20.97 21.46 2.33 12895 12413 1372 0.746 20.79 21.27 2.24 12921 12440 1376 0.744 20.74 21.21 3.02 12947 12466 1380 0.750 20.90 21.38 2.66 12972 12493 1384 0.739 20.58 21.05 12998 12519 1388 0.739 20.57 21.05 3.15 13023 12545 1392 0.742 20.68 21.16 2.56 13023 12545 1392 0.738 20.56 21.03 2.58 13048 12571 1396 0.744 20.73 21.20 2.47 13072 12597 1400 0.746 20.79 21.27 2.26 13097 12623 1404 0.746 20.78 21.26 2.97 13097 12623 1404 0.746 20.78 21.26 2.92 13121 12648 1408 0.749 20.88 21.36 2.90 13145 12673 1412 0.746 20.78 21.26 2.60 13145 12673 1412 0.747 20.83 21.31 2.51 13169 12698 1416 0.742 20.67 21.14 1.88 13193 12723 1420 0.747 20.81 21.29 2.05 231 13193 12723 1420 0.747 20.82 21.30 2.38 13216 12748 1424 0.747 20.83 21.31 1.66 13239 12773 1428 0.743 20.70 21.18 2.50 13262 12797 1432 0.748 20.86 21.34 2.48 13285 12821 1436 0.744 20.73 21.21 2.35 13308 12845 1440 0.750 20.91 21.40 13324 12863 1443 0.744 20.72 21.20 2.12 13352 12893 1448 0.751 20.94 21.42 13374 12916 1452 0.744 20.73 21.20 1.83 13395 12939 1456 0.745 20.77 21.25 1.91 13417 12963 1460 0.746 20.81 21.29 2.14 13443 12991 1465 0.739 20.59 21.06 1.76 13463 13014 1469 0.742 20.69 21.16 2.00 13463 13014 1469 0.743 20.72 21.20 2.00 13484 13036 1473 0.730 20.33 20.79 1.71 13499 13053 1476 0.737 20.54 21.01 2.20 13524 13080 1481 0.741 20.66 21.13 1.83 13544 13102 1485 0.741 20.66 21.13 2.11 13563 13123 1489 0.744 20.74 21.22 1.98 13582 13144 1493 0.742 20.68 21.16 1.92 13601 13165 1497 0.733 20.40 20.87 2.15 13619 13186 1501 0.734 20.44 20.90 1.86 13637 13207 1505 0.732 20.39 20.85 1.99 13655 13227 1509 0.747 20.81 21.29 2.58 13655 13227 1509 0.745 20.77 21.25 2.15 13669 13242 1512 0.751 20.93 21.41 2.35 13686 13262 1516 0.743 20.69 21.17 1.79 13690 13267 1517 0.743 20.71 21.19 1.35 13690 13267 1517 0.742 20.69 21.17 1.27 13703 13282 1520 0.735 20.47 20.93 2.26 13720 13301 1524 0.743 20.72 21.20 1.99 13724 13306 1525 0.748 20.86 21.34 1.13 13740 13325 1529 0.749 20.88 21.36 2.32 13752 13339 1532 0.739 20.58 21.05 2.15 13772 13362 1537 0.750 20.90 21.39 1.89 13784 13376 1540 0.739 20.59 21.07 2.13 13784 13376 1540 0.737 20.52 20.99 2.17 13803 13399 1545 0.741 20.65 21.13 2.40 13814 13412 1548 0.746 20.79 21.27 1.90 13832 13434 1553 0.745 20.77 21.25 2.15 13846 13451 1557 0.740 20.62 21.10 1.77 13860 13468 1561 0.733 20.41 20.88 1.84 13870 13481 1564 0.742 20.68 21.15 1.97 13873 13485 1565 0.737 20.52 20.99 2.01 13873 13485 1565 0.738 20.55 21.03 1.94 13886 13501 1569 0.751 20.95 21.44 0.63 13896 13513 1572 0.747 20.81 21.29 1.96 13899 13517 1573 0.747 20.83 21.31 2.30 13911 13533 1577 0.756 21.08 21.57 1.81 232 13920 13545 1580 0.750 20.90 21.39 1.76 13935 13564 1585 0.750 20.90 21.38 13946 13578 1589 0.752 20.97 21.45 1.98 13957 13593 1593 0.741 20.66 21.13 1.49 13965 13604 1596 0.732 20.38 20.85 2.03 13977 13621 1601 0.738 20.56 21.03 1.91 13980 13625 1602 0.737 20.52 20.99 1.75 13982 13628 1603 0.749 20.87 21.35 2.28 13985 13632 1604 0.736 20.51 20.98 2.18 13994 13645 1608 0.742 20.68 21.16 1.81 14003 13658 1612 0.727 20.23 20.69 2.31 14012 13671 1616 0.730 20.33 20.80 14020 13683 1620 0.731 20.35 20.81 2.01 14035 13707 1628 0.730 20.33 20.79 2.07 14042 13719 1632 0.749 20.88 21.36 1.57 14045 13724 1634 0.731 20.34 20.81 2.34 14045 13724 1634 0.732 20.38 20.84 2.41 14048 13730 1636 0.748 20.85 21.33 1.68 14055 13740 1640 0.756 21.10 21.59 1.55 14060 13751 1644 0.751 20.95 21.43 1.86 14060 13751 1644 0.751 20.95 21.43 1.71 14066 13761 1648 0.754 21.02 21.51 1.98 14070 13770 1652 0.754 21.04 21.53 1.72 14079 13789 1660 0.770 21.49 21.99 1.63 14082 13797 1664 0.761 21.22 21.72 1.83 14085 13806 1668 0.760 21.21 21.70 1.82 14088 13813 1672 0.766 21.38 21.88 1.60 14090 13821 1676 0.760 21.19 21.68 1.88 14091 13828 1680 0.765 21.35 21.84 1.60 14093 13835 1684 0.769 21.46 21.96 1.50 14093 13841 1688 0.775 21.66 22.16 1.31 14094 13847 1692 0.776 21.68 22.19 1.35 14093 13852 1696 0.768 21.43 21.93 1.40 14092 13857 1700 0.770 21.50 22.00 1.63 14089 13866 1708 0.768 21.43 21.93 1.52 14087 13870 1712 0.771 21.52 22.02 1.64 14086 13872 1714 0.770 21.51 22.01 1.58 CDH 26: Age (yrs, Age (yrs, Depth SST Temp C37total k Fairbanks calibration) Marine 13, Clam) (cm) U 37’ (°C, Prahl) (°C, Muller) (nmol/g) 1366 1732 16 0.808 22.62 23.15 3.69 1924 2175 32 0.794 22.20 22.72 2.79 2460 2605 48 0.786 21.96 22.47 3.06 2976 3024 64 0.792 22.14 22.66 3.71 3472 3430 80 0.791 22.11 22.63 2.97 3948 3826 96 0.794 22.20 22.72 2.71 4406 4210 112 0.797 22.29 22.82 3.28 4846 4583 128 0.787 22.01 22.53 3.47 233 5562 5199 155.5 0.805 22.53 23.06 4.52 5956 5544 171.5 0.809 22.65 23.18 4.40 6334 5878 187.5 0.810 22.68 23.22 4.58 6698 6203 203.5 0.811 22.71 23.24 5.25 7047 6518 219.5 0.807 22.60 23.14 6.09 7382 6825 235.5 0.805 22.52 23.06 4.18 7703 7122 251.5 0.802 22.44 22.96 5.21 8012 7410 267.5 0.790 22.10 22.62 4.67 8308 7690 283.5 0.786 21.98 22.49 4.28 8478 7852 293 0.828 23.19 23.75 2.82 8755 8118 309 0.789 22.07 22.59 2.04 8840 8200 314 0.776 21.69 22.19 5.37 8840 8200 314 0.776 21.67 22.17 4.93 8906 8265 318 0.781 21.84 22.35 5.56 8973 8329 322 0.779 21.75 22.26 4.91 9103 8456 330 0.779 21.75 22.26 5.13 9167 8519 334 0.774 21.62 22.12 4.50 9167 8519 334 0.774 21.61 22.11 4.61 9230 8581 338 0.773 21.58 22.09 5.09 9355 8705 346 0.773 21.58 22.08 4.89 9417 8766 350 0.770 21.51 22.01 6.45 9477 8826 354 0.774 21.62 22.13 4.00 9597 8946 362 0.775 21.64 22.14 4.06 9715 9064 370 0.772 21.55 22.05 3.73 9715 9064 370 0.771 21.52 22.03 3.81 9830 9180 378 0.766 21.37 21.86 3.12 9887 9237 382 0.767 21.41 21.91 3.34 9943 9294 386 0.765 21.36 21.86 1.98 10054 9407 394 0.764 21.33 21.82 3.40 10108 9463 398 0.771 21.54 22.04 3.09 10108 9463 398 0.773 21.59 22.10 2.92 10162 9518 402 0.770 21.50 22.00 2.68 10268 9627 410 0.778 21.75 22.26 3.08 10321 9681 414 0.783 21.88 22.39 3.14 10373 9735 418 0.778 21.72 22.23 3.30 10475 9841 426 0.779 21.77 22.28 3.27 10525 9893 430 0.788 22.04 22.56 3.08 10575 9945 434 0.784 21.91 22.42 3.10 10575 9945 434 0.780 21.81 22.32 3.14 10673 10048 442 0.784 21.91 22.42 2.85 10770 10149 450 0.790 22.09 22.60 2.95 10817 10200 454 0.783 21.88 22.39 3.26 10864 10249 458 0.783 21.88 22.39 2.31 10864 10249 458 0.785 21.95 22.46 2.83 10968 10360 467 0.771 21.53 22.03 2.49 11014 10408 471 0.776 21.68 22.18 2.77 11059 10457 475 0.774 21.60 22.11 2.61 11148 10552 483 0.771 21.54 22.04 2.90 11192 10599 487 0.778 21.74 22.24 2.91 234 11235 10646 491 0.773 21.58 22.09 2.93 11321 10738 499 0.777 21.70 22.20 2.81 11321 10738 499 0.774 21.63 22.13 2.95 11364 10784 503 0.766 21.38 21.88 2.37 11405 10829 507 0.772 21.56 22.07 3.04 11488 10919 515 0.769 21.46 21.96 2.77 11529 10964 519 0.766 21.38 21.88 2.84 11569 11008 523 0.767 21.41 21.90 2.80 11569 11008 523 0.766 21.38 21.87 2.55 11649 11095 531 0.759 21.17 21.66 2.92 11688 11138 535 0.762 21.27 21.77 2.87 11727 11181 539 0.757 21.10 21.59 3.22 11804 11266 547 0.763 21.30 21.80 3.44 11842 11308 551 0.755 21.06 21.54 2.54 11880 11349 555 0.757 21.11 21.60 3.00 11954 11431 563 0.749 20.89 21.38 4.23 11954 11431 563 0.748 20.85 21.33 4.22 11991 11472 567 0.760 21.20 21.69 3.50 12027 11513 571 0.749 20.90 21.38 4.31 12099 11593 579 0.741 20.65 21.12 3.32 12170 11671 587 0.736 20.49 20.96 4.65 12239 11749 595 0.738 20.55 21.02 4.30 12316 11835 604 0.753 20.99 21.47 4.72 12350 11873 608 0.752 20.96 21.44 4.55 12383 11911 612 0.741 20.65 21.12 4.68 12449 11985 620 0.735 20.48 20.95 4.11 12482 12022 624 0.735 20.48 20.95 3.91 12482 12022 624 0.735 20.48 20.95 3.89 12514 12059 628 0.725 20.17 20.63 4.39 12579 12131 636 0.734 20.46 20.92 4.55 12579 12131 636 0.734 20.44 20.91 4.64 12610 12167 640 0.744 20.74 21.22 4.70 12642 12203 644 0.750 20.90 21.38 4.79 12704 12273 652 0.725 20.17 20.63 3.97 12735 12308 656 0.727 20.22 20.68 3.78 12735 12308 656 0.725 20.19 20.65 3.94 12766 12343 660 0.723 20.13 20.58 4.01 12826 12412 668 0.730 20.33 20.80 3.88 12856 12446 672 0.724 20.14 20.60 3.86 12856 12446 672 0.724 20.14 20.60 3.81 12886 12480 676 0.720 20.02 20.48 3.81 12945 12546 684 0.729 20.30 20.77 3.14 12974 12580 688 0.733 20.42 20.89 3.32 12974 12580 688 0.730 20.34 20.80 3.23 13003 12613 692 0.733 20.41 20.88 3.12 13061 12678 700 0.726 20.20 20.66 2.96 13089 12710 704 0.731 20.36 20.82 2.74 13118 12742 708 0.736 20.49 20.96 2.68 13174 12806 716 0.730 20.33 20.79 2.52 235 13202 12838 720 0.736 20.49 20.96 2.60 13229 12869 724 0.732 20.37 20.84 2.60 13284 12931 732 0.721 20.06 20.52 2.35 13311 12962 736 0.719 19.99 20.44 2.75 13339 12993 740 0.714 19.86 20.31 2.64 13392 13053 748 0.734 20.43 20.90 2.61 13392 13053 748 0.733 20.41 20.88 2.63 13419 13084 752 0.724 20.15 20.61 2.34 13432 13099 754 0.727 20.23 20.70 2.59 13465 13136 759 0.729 20.29 20.75 2.59 13492 13166 763 0.734 20.44 20.90 2.67 13511 13188 766 0.730 20.34 20.80 2.03 13550 13232 772 0.735 20.47 20.94 2.12 13583 13268 777 0.727 20.24 20.70 2.68 13608 13297 781 0.726 20.21 20.67 2.75 13660 13354 789 0.732 20.38 20.84 3.02 13685 13383 793 0.730 20.33 20.80 2.77 13710 13411 797 0.721 20.07 20.52 2.66 13761 13467 805 0.727 20.25 20.71 3.00 13786 13495 809 0.732 20.37 20.83 2.82 13811 13523 813 0.727 20.24 20.70 2.90 13860 13578 821 0.733 20.40 20.87 2.81 13860 13578 821 0.731 20.37 20.83 2.84 13885 13606 825 0.731 20.35 20.82 2.80 13909 13633 829 0.720 20.02 20.48 2.96 13909 13633 829 0.729 20.30 20.77 2.90 13909 13633 829 0.724 20.15 20.61 2.52 13934 13660 833 0.724 20.15 20.61 3.12 13958 13687 837 0.731 20.35 20.82 2.97 13983 13714 841 0.745 20.76 21.24 2.49 14007 13740 845 0.724 20.15 20.61 3.17 14031 13767 849 0.723 20.11 20.57 2.40 14055 13794 853 0.734 20.43 20.90 2.48 14079 13820 857 0.724 20.16 20.62 2.45 14103 13846 861 0.737 20.52 20.99 2.76 14127 13872 865 0.738 20.56 21.03 2.70 14151 13898 869 0.757 21.11 21.60 2.52 14175 13924 873 0.737 20.53 21.00 2.02 14199 13950 877 0.736 20.49 20.96 2.51 14223 13976 881 0.724 20.15 20.61 2.95 14246 14002 885 0.743 20.71 21.19 1.53 14270 14027 889 0.739 20.60 21.07 3.05 14294 14053 893 0.755 21.07 21.56 2.45 14317 14078 897 0.751 20.94 21.43 2.63 14317 14078 897 0.756 21.09 21.58 2.69 14341 14103 901 0.752 20.96 21.45 2.17 14364 14128 905 0.754 21.02 21.50 1.56 14364 14128 905 0.754 21.03 21.51 1.57 14387 14153 909 0.743 20.70 21.17 2.45 236 14411 14178 913 0.740 20.62 21.09 2.76 14434 14203 917 0.743 20.70 21.18 2.64 14457 14228 921 0.747 20.81 21.29 2.74 14481 14253 925 0.753 20.99 21.47 2.25 14504 14277 929 0.750 20.91 21.39 2.63 14539 14314 935 0.742 20.68 21.15 2.98 14550 14326 937 0.737 20.53 21.00 2.96 14573 14351 941 0.741 20.65 21.12 2.41 14597 14375 945 0.743 20.72 21.19 2.49 14620 14399 949 0.730 20.31 20.78 2.66 14643 14423 953 0.712 19.80 20.25 2.09 14666 14448 957 0.731 20.36 20.82 2.22 14689 14472 961 0.724 20.16 20.62 1.96 14689 14472 961 0.721 20.06 20.52 1.92 14712 14496 965 0.728 20.27 20.74 2.54 14735 14520 969 0.717 19.95 20.41 2.42 14758 14543 973 0.726 20.22 20.68 2.24 14781 14567 977 0.716 19.92 20.38 2.83 14804 14591 981 0.719 20.01 20.47 2.63 14827 14615 985 0.724 20.15 20.61 2.98 14850 14638 989 0.716 19.91 20.36 1.92 14873 14662 993 0.714 19.86 20.31 2.94 14896 14685 997 0.711 19.77 20.22 3.30 14919 14709 1001 0.715 19.90 20.35 3.86 14942 14732 1005 0.723 20.13 20.59 2.58 14965 14756 1009 0.706 19.63 20.07 2.74 14988 14779 1013 0.703 19.54 19.98 2.56 15011 14802 1017 0.711 19.75 20.20 2.70 15011 14802 1017 0.710 19.74 20.18 2.74 15034 14826 1021 0.716 19.90 20.36 3.02 15057 14849 1025 0.704 19.55 19.99 2.80 15080 14872 1029 0.701 19.48 19.92 2.93 15103 14895 1033 0.702 19.49 19.93 3.05 15126 14918 1037 0.705 19.58 20.02 3.18 15149 14941 1041 0.698 19.37 19.81 3.50 15172 14964 1045 0.701 19.47 19.91 3.31 15195 14987 1049 0.717 19.95 20.41 3.11 15218 15010 1053 0.710 19.73 20.17 2.94 15241 15033 1057 0.715 19.88 20.33 3.02 15241 15033 1057 0.713 19.82 20.27 3.07 15264 15056 1061 0.718 19.98 20.43 2.46 15287 15079 1065 0.722 20.08 20.54 3.23 15310 15102 1069 0.710 19.74 20.19 3.04 15333 15125 1073 0.717 19.93 20.38 3.53 15356 15147 1077 0.710 19.74 20.19 3.26 15356 15147 1077 0.711 19.77 20.22 3.42 15356 15147 1077 0.720 20.04 20.49 3.66 15380 15170 1081 0.726 20.22 20.68 3.30 15403 15193 1085 0.722 20.09 20.55 3.51 237 15426 15216 1089 0.720 20.02 20.48 3.82 15426 15216 1089 0.719 20.01 20.46 3.94 15449 15239 1093 0.732 20.39 20.86 3.67 15472 15261 1097 0.725 20.16 20.62 3.45 15496 15284 1101 0.714 19.85 20.30 3.52 15519 15307 1105 0.713 19.84 20.29 3.29 15542 15329 1109 0.723 20.11 20.57 3.05 15566 15352 1113 0.717 19.94 20.39 3.66 15589 15375 1117 0.726 20.19 20.65 3.28 15613 15397 1121 0.726 20.20 20.66 3.14 15636 15420 1125 0.713 19.82 20.27 3.79 15659 15443 1129 0.726 20.22 20.68 3.62 15683 15465 1133 0.716 19.91 20.36 3.45 15707 15488 1137 0.719 20.01 20.46 3.47 15707 15488 1137 0.720 20.03 20.48 3.48 15730 15511 1141 0.711 19.77 20.22 3.53 15754 15533 1145 0.717 19.94 20.39 4.24 15777 15556 1149 0.725 20.18 20.64 3.06 15801 15579 1153 0.728 20.27 20.73 3.30 15825 15601 1157 0.738 20.55 21.02 3.21 15848 15624 1161 0.732 20.38 20.84 3.13 15872 15647 1165 0.728 20.27 20.73 3.15 15896 15669 1169 0.728 20.26 20.72 2.96 15920 15692 1173 0.735 20.48 20.95 3.29 15968 15738 1181 0.731 20.34 20.81 3.18 15992 15760 1185 0.742 20.67 21.15 2.36 16016 15783 1189 0.720 20.02 20.47 3.29 16040 15806 1193 0.719 19.99 20.45 3.09 16064 15829 1197 0.721 20.07 20.52 2.42 16124 15886 1207 0.734 20.43 20.90 2.12 16160 15920 1213 0.708 19.67 20.11 2.92 16179 15937 1216 0.708 19.68 20.12 2.91 16203 15960 1220 0.705 19.58 20.02 1.97 16227 15983 1224 0.703 19.54 19.98 3.38 16252 16006 1228 0.708 19.66 20.11 3.30 16276 16029 1232 0.716 19.90 20.35 3.01 16301 16052 1236 0.715 19.88 20.33 2.42 16325 16075 1240 0.730 20.34 20.80 2.71 16374 16121 1248 0.712 19.79 20.24 3.34 16374 16121 1248 0.711 19.75 20.20 3.26 16399 16144 1252 0.717 19.95 20.40 2.97 16424 16167 1256 0.716 19.93 20.38 2.87 16448 16190 1260 0.720 20.03 20.49 0.85 16448 16190 1260 0.718 19.98 20.44 0.85 16473 16213 1264 0.721 20.06 20.51 3.26 16498 16237 1268 0.734 20.43 20.90 3.20 16523 16260 1272 0.737 20.53 21.00 2.48 16548 16283 1276 0.727 20.23 20.69 2.45 16573 16307 1280 0.716 19.90 20.35 3.01 238 16598 16330 1284 0.734 20.43 20.90 2.90 16623 16353 1288 0.736 20.50 20.97 2.95 16673 16400 1296 0.731 20.36 20.83 2.19 16723 16447 1304 0.727 20.22 20.68 2.68 16723 16447 1304 0.728 20.25 20.71 2.77 16749 16471 1308 0.719 20.00 20.46 3.08 16774 16494 1312 0.720 20.03 20.48 3.53 16799 16518 1316 0.720 20.02 20.47 2.95 16799 16518 1316 0.722 20.08 20.54 2.76 16825 16542 1320 0.734 20.46 20.92 2.44 16850 16566 1324 0.732 20.38 20.85 2.61 16876 16589 1328 0.736 20.49 20.95 3.30 16901 16613 1332 0.735 20.46 20.93 3.13 16978 16685 1344 0.733 20.40 20.87 3.71 17004 16709 1348 0.710 19.74 20.18 3.83 17030 16733 1352 0.733 20.41 20.88 3.27 17055 16757 1356 0.724 20.14 20.60 4.02 17081 16782 1360 0.720 20.04 20.50 4.27 17107 16806 1364 0.722 20.08 20.53 3.91 17140 16836 1369 0.720 20.04 20.49 4.67 17140 16836 1369 0.725 20.17 20.63 4.57 17166 16861 1373 0.714 19.87 20.32 3.35 17296 16983 1393 0.723 20.12 20.58 3.34 17322 17008 1397 0.720 20.04 20.49 2.41 17349 17033 1401 0.706 19.63 20.07 4.31 17375 17057 1405 0.726 20.19 20.65 2.60 17454 17132 1417 0.710 19.73 20.17 2.90 17481 17157 1421 0.714 19.85 20.30 3.88 17507 17182 1425 0.715 19.87 20.32 3.99 17534 17207 1429 0.720 20.04 20.49 4.07 17560 17232 1433 0.718 19.97 20.42 2.51 17560 17232 1433 0.718 19.96 20.41 2.49 17587 17258 1437 0.713 19.82 20.27 2.34 17614 17283 1441 0.733 20.43 20.89 4.03 17614 17283 1441 0.734 20.45 20.91 4.54 17647 17315 1446 0.710 19.73 20.17 2.64 17674 17340 1450 0.725 20.17 20.63 2.31 17755 17417 1462 0.706 19.61 20.05 4.17 17782 17442 1466 0.730 20.34 20.80 4.12 17809 17468 1470 0.735 20.47 20.94 1.81 17836 17494 1474 0.722 20.08 20.53 4.64 17863 17520 1478 0.724 20.16 20.62 2.54 17890 17546 1482 0.737 20.52 20.99 2.78 17917 17572 1486 0.728 20.26 20.72 4.67 17971 17624 1494 0.737 20.53 21.00 4.45 17998 17650 1498 0.727 20.24 20.71 4.70 18080 17729 1510 0.715 19.89 20.34 6.31 18108 17755 1514 0.713 19.83 20.28 5.93 18135 17782 1518 0.727 20.24 20.70 3.26 239 18163 17809 1522 0.732 20.37 20.84 4.07 18176 17822 1524 0.730 20.33 20.79 4.93 18176 17822 1524 0.729 20.30 20.76 5.13 18204 17848 1528 0.735 20.47 20.94 2.59 18204 17848 1528 0.735 20.47 20.93 2.58 18231 17875 1532 0.731 20.34 20.81 5.26 18259 17902 1536 0.722 20.09 20.55 2.80 18286 17929 1540 0.725 20.19 20.65 1.78 18314 17956 1544 0.722 20.10 20.55 4.41 18397 18037 1556 0.711 19.75 20.20 5.26 18425 18064 1560 0.709 19.70 20.15 5.81 18453 18091 1564 0.726 20.20 20.66 7.48 18480 18119 1568 0.728 20.27 20.74 7.85 18508 18146 1572 0.714 19.85 20.30 6.19 18536 18174 1576 0.719 20.01 20.46 5.75 18564 18201 1580 0.724 20.15 20.60 5.96 18592 18229 1584 0.728 20.26 20.72 6.59 18620 18256 1588 0.727 20.24 20.70 4.89 18648 18284 1592 0.715 19.89 20.34 5.80 18648 18284 1592 0.715 19.88 20.33 5.68 18676 18312 1596 0.711 19.77 20.22 6.77 18704 18340 1600 0.704 19.56 20.00 7.81 18760 18395 1608 0.712 19.78 20.23 6.17 18788 18423 1612 0.715 19.87 20.32 6.13 18788 18423 1612 0.715 19.87 20.32 6.05 18816 18452 1616 0.713 19.83 20.28 6.03 18872 18508 1624 0.724 20.16 20.62 5.75 18900 18536 1628 0.716 19.90 20.35 5.80 18929 18565 1632 0.707 19.64 20.08 5.50 18957 18593 1636 0.719 20.01 20.46 6.16 18985 18621 1640 0.707 19.64 20.09 5.72 19013 18650 1644 0.705 19.59 20.03 6.22 19070 18707 1652 0.721 20.07 20.52 5.67 19098 18736 1656 0.718 19.98 20.44 5.89 19126 18765 1660 0.730 20.32 20.78 5.52 19183 18823 1668 0.708 19.69 20.13 6.26 19212 18852 1672 0.715 19.89 20.34 7.20 19254 18895 1678 0.736 20.49 20.96 6.93 19254 18895 1678 0.741 20.65 21.12 7.43 19283 18924 1682 0.745 20.75 21.23 6.58 19283 18924 1682 0.742 20.67 21.14 6.30 19311 18954 1686 0.742 20.68 21.16 7.59 19339 18983 1690 0.737 20.54 21.01 8.22 19396 19041 1698 0.741 20.65 21.13 8.05 19425 19071 1702 0.740 20.61 21.08 7.02 19453 19100 1706 0.736 20.49 20.96 7.61 19482 19130 1710 0.730 20.32 20.79 7.87 19482 19130 1710 0.731 20.36 20.83 8.23 19510 19160 1714 0.742 20.69 21.16 8.35 240 19539 19189 1718 0.732 20.39 20.85 7.05 19568 19219 1722 0.742 20.68 21.16 5.56 19596 19249 1726 0.742 20.69 21.17 5.57 19625 19279 1730 0.743 20.69 21.17 5.25 19653 19308 1734 0.742 20.67 21.14 5.25 19710 19368 1742 0.735 20.47 20.93 5.12 19739 19398 1746 0.745 20.77 21.24 4.49 19768 19429 1750 0.724 20.14 20.60 5.94 19839 19504 1760 0.710 19.75 20.19 6.22 19868 19534 1764 0.707 19.64 20.09 5.74 19897 19565 1768 0.726 20.21 20.67 6.65 19897 19565 1768 0.719 20.00 20.45 6.35 19925 19595 1772 0.712 19.81 20.26 6.68 19925 19595 1772 0.709 19.69 20.14 6.26 19954 19626 1776 0.722 20.09 20.55 6.48 19983 19656 1780 0.724 20.15 20.61 6.83 20011 19687 1784 0.721 20.06 20.51 6.24 20040 19717 1788 0.735 20.47 20.94 5.29 20098 19779 1796 0.725 20.18 20.64 5.74 20126 19809 1800 0.720 20.03 20.48 6.22 20155 19840 1804 0.719 20.00 20.46 5.79 20184 19871 1808 0.719 20.00 20.46 3.68 20212 19902 1812 0.714 19.86 20.31 6.53 20241 19933 1816 0.717 19.95 20.40 6.71 20270 19964 1820 0.718 19.97 20.43 6.16 20299 19995 1824 0.733 20.42 20.89 6.65 20349 20049 1831 0.733 20.42 20.89 2.00 20349 20049 1831 0.733 20.41 20.88 1.96 20406 20111 1839 0.722 20.08 20.54 5.52 20435 20143 1843 0.727 20.24 20.70 5.68 20464 20174 1847 0.737 20.53 21.00 6.04 20464 20174 1847 0.732 20.37 20.84 5.36 20521 20236 1855 0.735 20.46 20.93 4.68 20550 20268 1859 0.725 20.16 20.62 5.36 20579 20299 1863 0.732 20.38 20.84 5.14 20608 20331 1867 0.718 19.96 20.41 4.82 20636 20362 1871 0.717 19.95 20.41 5.34 20723 20457 1883 0.736 20.51 20.98 5.82 20752 20488 1887 0.725 20.18 20.64 6.85 20752 20488 1887 0.728 20.25 20.72 7.10 20780 20520 1891 0.738 20.55 21.02 6.51 20809 20551 1895 0.740 20.62 21.09 6.78 20838 20583 1899 0.736 20.50 20.97 4.75 20867 20615 1903 0.738 20.56 21.03 6.06 20895 20647 1907 0.735 20.46 20.93 6.18 20982 20742 1919 0.725 20.19 20.65 5.45 21039 20806 1927 0.728 20.27 20.73 5.66 21039 20806 1927 0.728 20.25 20.72 5.59 21068 20837 1931 0.725 20.17 20.63 5.36 241 21097 20869 1935 0.729 20.28 20.74 5.42 21125 20901 1939 0.716 19.92 20.37 6.51 21154 20933 1943 0.721 20.06 20.52 6.06 21183 20965 1947 0.717 19.95 20.40 5.58 21212 20997 1951 0.735 20.47 20.94 6.01 21240 21029 1955 0.731 20.36 20.83 4.74 21269 21061 1959 0.728 20.27 20.74 4.63 21298 21093 1963 0.730 20.33 20.79 7.04 21327 21125 1967 0.730 20.32 20.78 6.78 21356 21157 1971 0.735 20.48 20.95 5.67 21384 21189 1975 0.732 20.37 20.84 5.57 21413 21221 1979 0.731 20.35 20.82 4.87 21456 21269 1985 0.724 20.16 20.62 4.21 21456 21269 1985 0.716 19.91 20.37 3.50 21485 21301 1989 0.724 20.15 20.61 5.24 21514 21333 1993 0.718 19.98 20.43 5.52 21542 21365 1997 0.733 20.42 20.88 5.38 21571 21397 2001 0.731 20.34 20.81 5.52 21571 21397 2001 0.733 20.42 20.89 5.27 21600 21429 2005 0.731 20.35 20.81 5.03 21686 21525 2017 0.712 19.80 20.25 4.09 21715 21557 2021 0.728 20.27 20.74 5.86 21744 21589 2025 0.726 20.22 20.68 5.36 21773 21621 2029 0.724 20.16 20.62 5.22 21801 21653 2033 0.725 20.18 20.64 5.24 21830 21685 2037 0.730 20.34 20.80 5.19 21859 21717 2041 0.733 20.42 20.89 5.46 21888 21749 2045 0.729 20.30 20.77 3.08 21917 21781 2049 0.716 19.92 20.37 4.20 21938 21804 2052 0.725 20.16 20.62 5.23 21974 21844 2057 0.730 20.32 20.78 7.09 22003 21876 2061 0.726 20.21 20.67 4.87 22032 21908 2065 0.743 20.70 21.17 6.11 22032 21908 2065 0.740 20.62 21.09 5.88 22061 21940 2069 0.742 20.68 21.15 4.57 22089 21972 2073 0.732 20.39 20.85 5.39 22089 21972 2073 0.727 20.22 20.68 4.77 22118 22003 2077 0.736 20.49 20.96 5.24 22147 22035 2081 0.729 20.31 20.77 4.89 22176 22067 2085 0.739 20.59 21.06 5.60 22205 22099 2089 0.734 20.45 20.92 4.44 22234 22130 2093 0.736 20.50 20.97 3.91 22263 22162 2097 0.723 20.12 20.58 5.12 22292 22193 2101 0.729 20.28 20.75 2.81 22320 22225 2105 0.736 20.50 20.97 5.47 22349 22257 2109 0.744 20.74 21.22 4.55 22378 22288 2113 0.736 20.50 20.97 4.66 22378 22288 2113 0.735 20.46 20.93 4.64 22407 22319 2117 0.739 20.59 21.06 5.16 242 22436 22351 2121 0.742 20.67 21.15 5.12 22465 22382 2125 0.733 20.41 20.87 5.30 22494 22414 2129 0.736 20.50 20.97 5.79 22523 22445 2133 0.742 20.69 21.17 3.90 22545 22468 2136 0.733 20.40 20.86 2.83 22574 22499 2140 0.727 20.23 20.69 6.02 22574 22499 2140 0.727 20.24 20.71 5.97 22603 22530 2144 0.728 20.27 20.73 5.35 22632 22562 2148 0.721 20.06 20.52 5.48 22661 22593 2152 0.719 20.00 20.45 5.95 22690 22624 2156 0.731 20.35 20.82 5.35 22719 22654 2160 0.727 20.24 20.71 5.53 22749 22685 2164 0.726 20.22 20.68 5.05 22778 22716 2168 0.740 20.60 21.08 5.27 22807 22747 2172 0.744 20.74 21.22 4.33 22836 22777 2176 0.743 20.71 21.18 4.39 22836 22777 2176 0.743 20.72 21.20 4.21 22866 22808 2180 0.733 20.42 20.89 5.61 22895 22839 2184 0.728 20.26 20.72 5.49 22924 22869 2188 0.735 20.48 20.95 5.66 22953 22899 2192 0.739 20.58 21.05 5.07 22953 22899 2192 0.739 20.59 21.06 5.01 22983 22930 2196 0.741 20.65 21.12 4.33 23012 22960 2200 0.744 20.72 21.20 2.70 23042 22990 2204 0.737 20.53 21.00 5.43 23071 23020 2208 0.735 20.48 20.95 6.17 23101 23050 2212 0.743 20.70 21.18 5.78 23130 23080 2216 0.745 20.77 21.25 6.14 23160 23110 2220 0.746 20.79 21.26 5.70 23189 23140 2224 0.744 20.73 21.21 4.94 23219 23169 2228 0.746 20.79 21.27 3.35 23249 23199 2232 0.742 20.68 21.16 5.32 23278 23229 2236 0.742 20.68 21.16 4.96 23308 23258 2240 0.738 20.55 21.02 5.71 23308 23258 2240 0.737 20.54 21.01 5.62 23338 23287 2244 0.757 21.13 21.62 4.49 23368 23316 2248 0.752 20.96 21.44 5.08 23398 23345 2252 0.745 20.76 21.24 5.09 23428 23374 2256 0.753 20.99 21.48 4.08 23458 23403 2260 0.749 20.87 21.35 4.45 23488 23432 2264 0.747 20.81 21.29 4.46 23518 23461 2268 0.743 20.70 21.17 4.34 23548 23489 2272 0.745 20.76 21.23 4.98 23578 23518 2276 0.746 20.78 21.26 5.15 23609 23546 2280 0.754 21.02 21.51 4.76 23639 23575 2284 0.760 21.21 21.70 4.36 23669 23603 2288 0.764 21.34 21.83 3.16 23685 23617 2290 0.759 21.19 21.68 2.46 23685 23617 2290 0.759 21.17 21.66 2.55 243 23715 23645 2294 0.758 21.13 21.62 4.21 23746 23672 2298 0.761 21.25 21.74 3.91 23746 23672 2298 0.760 21.22 21.71 4.02 23776 23700 2302 0.760 21.21 21.70 4.02 23807 23728 2306 0.747 20.81 21.29 4.27 23838 23755 2310 0.740 20.62 21.09 4.47 23869 23782 2314 0.745 20.78 21.26 4.32 23869 23782 2314 0.746 20.80 21.28 4.20 23900 23810 2318 0.752 20.97 21.45 3.66 23931 23837 2322 0.742 20.67 21.15 3.98 23962 23864 2326 0.735 20.47 20.94 4.64 23993 23890 2330 0.743 20.70 21.18 4.66 24024 23917 2334 0.745 20.76 21.24 3.98 24055 23944 2338 0.751 20.94 21.42 3.99 24087 23970 2342 0.752 20.96 21.44 3.92 24118 23996 2346 0.759 21.17 21.66 3.97 24150 24022 2350 0.750 20.92 21.41 4.11 24181 24048 2354 0.742 20.68 21.16 4.53 24213 24074 2358 0.740 20.63 21.10 4.86 24245 24099 2362 0.741 20.64 21.11 4.43 24277 24125 2366 0.749 20.87 21.35 4.20 24309 24150 2370 0.741 20.63 21.11 4.78 24341 24175 2374 0.747 20.83 21.31 4.44 24373 24200 2378 0.743 20.70 21.18 4.99 24405 24225 2382 0.741 20.66 21.13 5.12 24438 24250 2386 0.741 20.63 21.11 5.29 24470 24274 2390 0.736 20.49 20.96 4.86 24503 24298 2394 0.724 20.16 20.62 5.78 24536 24322 2398 0.736 20.51 20.98 6.17 24569 24346 2402 0.743 20.70 21.18 5.21 24602 24370 2406 0.730 20.34 20.80 5.27 24635 24393 2410 0.733 20.41 20.87 4.75 24668 24417 2414 0.742 20.67 21.15 4.94 24701 24440 2418 0.733 20.42 20.88 5.07 24735 24463 2422 0.735 20.48 20.95 4.55 24769 24486 2426 0.735 20.48 20.95 4.93 24802 24508 2430 0.741 20.66 21.14 4.60 24802 24508 2430 0.742 20.69 21.16 4.52 24836 24531 2434 0.742 20.68 21.16 4.82 24870 24553 2438 0.738 20.57 21.04 3.77 24904 24575 2442 0.734 20.44 20.90 5.42 24921 24586 2444 0.733 20.40 20.86 4.64 24956 24607 2448 0.735 20.47 20.94 5.18 24990 24629 2452 0.728 20.26 20.72 5.40 25025 24650 2456 0.730 20.32 20.78 4.71 25060 24671 2460 0.736 20.51 20.98 5.06 25095 24692 2464 0.739 20.60 21.07 4.72 25130 24713 2468 0.735 20.46 20.93 5.91 25165 24733 2472 0.726 20.21 20.68 5.25 244 25201 24753 2476 0.716 19.93 20.38 6.10 25272 24793 2484 0.714 19.85 20.30 5.03 25308 24812 2488 0.713 19.82 20.27 5.37 25308 24812 2488 0.713 19.83 20.28 5.12 25344 24831 2492 0.726 20.21 20.67 5.25 25381 24850 2496 0.723 20.12 20.58 5.21 25417 24869 2500 0.729 20.29 20.75 6.13 25454 24888 2504 0.732 20.38 20.85 6.17 25491 24906 2508 0.722 20.10 20.56 7.70 25528 24924 2512 0.716 19.92 20.38 5.10 25565 24942 2516 0.715 19.89 20.34 4.49 25602 24959 2520 0.712 19.78 20.23 4.91 25640 24976 2524 0.715 19.90 20.35 5.32 25678 24993 2528 0.720 20.02 20.47 4.32 25716 25010 2532 0.711 19.76 20.20 4.29 25754 25027 2536 0.715 19.88 20.33 5.17 25793 25043 2540 0.722 20.10 20.55 5.33 25831 25059 2544 0.722 20.10 20.56 5.17 GGC 43/CDH 41: Age (yrs, Marine13, C37total k’ Core ID Depth (cm) Clam) U 37 SST (°C, Prahl) SST (°C, Muller) (nmol/g) GGC 43 0 841 0.837 23.47 24.03 14.19 GGC 43 2 1032 0.836 23.44 24.00 13.93 GGC 43 2 1032 GGC 43 4 1220 0.836 23.44 24.00 13.94 GGC 43 4 1220 GGC 43 6 1405 0.836 23.44 24.00 13.02 GGC 43 8 1587 0.836 23.44 24.00 12.91 GGC 43 10 1766 0.837 23.48 24.04 13.89 GGC 43 10 1766 GGC 43 12 1943 0.835 23.41 23.97 11.91 GGC 43 14 2117 0.837 23.47 24.03 12.85 GGC 43 16 2289 0.833 23.35 23.91 12.74 GGC 43 18 2458 0.840 23.56 24.12 11.42 GGC 43 20 2624 0.837 23.47 24.03 13.75 GGC 43 22 2788 0.835 23.42 23.97 11.94 GGC 43 24 2950 0.835 23.42 23.98 11.78 GGC 43 26 3109 0.833 23.36 23.91 11.78 GGC 43 28 3266 0.839 23.53 24.09 12.08 GGC 43 28 3266 GGC 43 30 3420 0.836 23.43 23.99 12.00 GGC 43 32 3572 0.835 23.41 23.97 12.64 245 GGC 43 34 3722 0.837 23.47 24.03 10.78 GGC 43 36 3870 0.835 23.41 23.97 10.94 GGC 43 38 4015 0.835 23.42 23.98 12.84 GGC 43 40 4158 0.836 23.44 24.00 11.99 GGC 43 42 4299 0.837 23.46 24.02 5.70 GGC 43 42 4299 GGC 43 42 4299 0.838 23.49 24.05 10.93 GGC 43 44 4438 0.838 23.51 24.07 11.47 GGC 43 44 4438 GGC 43 46 4575 0.836 23.45 24.01 10.86 GGC 43 48 4710 0.836 23.43 23.99 11.12 GGC 43 50 4843 0.835 23.40 23.96 10.64 GGC 43 52 4974 0.836 23.45 24.01 11.76 GGC 43 52 4974 0.838 23.49 24.06 10.80 GGC 43 54 5104 0.834 23.39 23.94 11.29 GGC 43 54 5104 0.838 23.50 24.06 10.46 GGC 43 56 5231 0.835 23.42 23.98 10.56 GGC 43 58 5356 0.832 23.33 23.89 10.97 GGC 43 60 5480 0.836 23.45 24.01 11.14 GGC 43 62 5601 0.835 23.40 23.96 10.26 GGC 43 64 5721 0.836 23.45 24.01 10.97 GGC 43 64 5721 0.838 23.49 24.05 10.19 GGC 43 66 5839 0.834 23.39 23.95 11.14 GGC 43 68 5956 0.835 23.41 23.96 10.76 GGC 43 70 6070 0.835 23.40 23.96 11.04 GGC 43 72 6184 0.836 23.43 23.98 11.22 GGC 43 74 6295 0.836 23.43 23.99 9.80 GGC 43 76 6405 0.834 23.37 23.92 9.82 GGC 43 78 6513 0.835 23.42 23.98 7.93 GGC 43 80 6620 0.836 23.43 23.99 10.51 GGC 43 82 6725 0.835 23.40 23.96 10.79 GGC 43 82 6725 GGC 43 84 6828 0.832 23.32 23.88 10.12 GGC 43 86 6930 0.835 23.42 23.97 8.66 GGC 43 88 7031 0.835 23.41 23.97 9.55 GGC 43 90 7130 0.835 23.42 23.98 11.19 GGC 43 92 7228 0.833 23.36 23.91 9.67 GGC 43 94 7324 0.834 23.38 23.93 10.20 GGC 43 96 7419 0.834 23.39 23.94 10.02 GGC 43 98 7513 0.834 23.37 23.93 9.87 246 GGC 43 100 7605 0.833 23.36 23.92 10.72 GGC 43 102 7696 0.833 23.35 23.91 10.42 GGC 43 104 7786 0.832 23.33 23.88 10.37 GGC 43 106 7874 0.834 23.38 23.93 9.56 GGC 43 106 7874 GGC 43 108 7962 0.831 23.30 23.86 9.05 GGC 43 110 8048 0.828 23.20 23.75 9.69 GGC 43 112 8133 0.815 22.83 23.37 8.03 GGC 43 112 8133 0.816 22.84 23.39 7.78 GGC 43 114 8216 0.832 23.31 23.87 9.96 GGC 43 116 8299 0.829 23.24 23.79 8.92 GGC 43 118 8380 0.831 23.29 23.85 9.42 GGC 43 118 8380 GGC 43 120 8461 0.828 23.21 23.76 6.96 GGC 43 122 8540 0.832 23.33 23.89 9.09 GGC 43 124 8618 0.828 23.20 23.75 8.15 GGC 43 126 8695 0.831 23.28 23.84 9.72 GGC 43 128 8771 0.829 23.22 23.78 8.81 GGC 43 GGC 43 130 8846 0.827 23.18 23.73 9.62 GGC 43 132 8921 0.828 23.20 23.76 9.27 GGC 43 134 8994 0.826 23.15 23.70 8.98 GGC 43 136 9066 0.828 23.21 23.76 8.75 GGC 43 GGC 43 138 9137 0.831 23.30 23.86 9.53 GGC 43 138 9137 GGC 43 140 9208 0.829 23.24 23.79 9.60 GGC 43 142 9278 0.829 23.23 23.78 9.51 GGC 43 144 9346 0.829 23.23 23.78 9.55 GGC 43 144 9346 GGC 43 146 9414 0.828 23.21 23.76 8.37 GGC 43 148 9481 0.823 23.07 23.62 7.19 GGC 43 150 9548 0.829 23.23 23.79 8.39 GGC 43 150 9548 GGC 43 152 9613 0.825 23.10 23.65 7.56 GGC 43 154 9678 0.827 23.18 23.73 8.39 GGC 43 156 9743 0.830 23.27 23.82 8.68 GGC 43 158 9806 0.827 23.19 23.74 7.92 GGC 43 158 9806 GGC 43 160 9869 0.826 23.14 23.69 7.50 247 GGC 43 162 9931 0.832 23.31 23.87 4.78 GGC 43 164 9993 0.826 23.13 23.68 8.15 GGC 43 166 10054 0.825 23.13 23.68 8.02 GGC 43 168 10114 0.829 23.23 23.79 6.76 GGC 43 170 10174 0.823 23.06 23.60 8.07 GGC 43 172 10234 0.823 23.05 23.59 7.60 GGC 43 174 10292 0.822 23.03 23.58 8.00 GGC 43 176 10351 0.808 22.63 23.16 7.31 GGC 43 176 10351 GGC 43 178 10409 0.811 22.69 23.23 7.76 GGC 43 180 10466 0.824 23.09 23.64 7.32 GGC 43 180 10466 GGC 43 182 10524 0.820 22.97 23.51 6.68 GGC 43 GGC 43 184 10580 0.823 23.07 23.62 7.68 GGC 43 186 10637 0.819 22.95 23.49 6.29 GGC 43 188 10693 0.820 22.97 23.52 7.66 GGC 43 190 10749 0.820 22.97 23.51 6.85 GGC 43 192 10804 0.826 23.15 23.70 7.57 GGC 43 194 10860 0.833 23.35 23.91 9.38 GGC 43 194 10860 0.825 23.13 23.68 5.06 GGC 43 196 10915 0.822 23.02 23.57 6.80 GGC 43 196 10915 GGC 43 198 10970 0.824 23.10 23.65 7.34 GGC 43 200 11025 0.821 22.99 23.54 7.37 GGC 43 202 11079 0.821 23.01 23.56 7.44 GGC 43 204 11134 0.818 22.90 23.44 6.49 GGC 43 206 11189 0.819 22.95 23.49 5.73 GGC 43 208 11243 0.820 22.98 23.52 6.00 GGC 43 210 11298 0.816 22.86 23.40 6.31 GGC 43 212 11352 0.819 22.93 23.48 6.59 GGC 43 214 11407 0.821 22.99 23.53 9.78 GGC 43 216 11462 0.818 22.92 23.46 7.90 GGC 43 218 11517 0.816 22.86 23.40 8.43 GGC 43 220 11572 0.785 21.94 22.46 8.87 GGC 43 222 11627 0.820 22.96 23.50 7.67 GGC 43 224 11682 0.809 22.65 23.18 7.52 GGC 43 226 11738 0.811 22.72 23.26 7.49 GGC 43 226 11738 0.811 22.70 23.24 7.44 GGC 43 228 11794 0.814 22.79 23.33 6.17 248 GGC 43 230 11851 0.812 22.73 23.27 5.53 GGC 43 232 11908 0.815 22.81 23.35 7.02 GGC 43 234 11965 0.805 22.53 23.06 7.09 GGC 43 236 12022 0.812 22.73 23.27 6.93 GGC 43 236 12022 GGC 43 238 12081 0.811 22.72 23.25 7.06 GGC 43 240 12139 0.815 22.82 23.36 7.49 GGC 43 242 12199 0.812 22.73 23.27 7.16 GGC 43 244 12259 0.809 22.64 23.18 7.12 GGC 43 246 12319 0.799 22.35 22.87 7.22 GGC 43 248 12381 0.800 22.38 22.91 7.86 GGC 43 248 12381 GGC 43 250 12443 0.805 22.52 23.06 7.01 GGC 43 252 12506 0.801 22.42 22.95 7.57 GGC 43 254 12570 0.798 22.31 22.84 6.70 GGC 43 256 12634 0.805 22.53 23.06 8.03 GGC 43 258 12700 0.799 22.36 22.89 7.13 GGC 43 260 12766 0.802 22.44 22.97 6.34 GGC 43 262 12834 0.810 22.67 23.20 7.45 GGC 43 264 12903 0.796 22.26 22.78 8.12 GGC 43 266 12973 0.800 22.39 22.92 7.52 GGC 43 268 13044 0.807 22.58 23.11 8.44 GGC 43 270 13116 0.798 22.32 22.84 7.53 GGC 43 272 13189 0.797 22.31 22.83 8.40 GGC 43 274 13264 0.795 22.24 22.76 7.47 GGC 43 276 13340 0.799 22.35 22.88 8.51 GGC 43 278 13418 0.796 22.27 22.80 8.67 GGC 43 280 13497 0.807 22.59 23.12 6.89 GGC 43 282 13578 0.800 22.38 22.90 7.72 GGC 43 284 13661 0.795 22.23 22.76 7.45 GGC 43 286 13745 0.804 22.51 23.04 8.20 GGC 43 286 13745 GGC 43 288 13831 0.798 22.34 22.86 8.29 GGC 43 290 13918 0.805 22.52 23.05 7.93 GGC 43 292 14008 0.804 22.51 23.04 6.61 GGC 43 294 14099 0.800 22.39 22.91 6.67 GGC 43 296 14193 0.798 22.31 22.84 7.38 GGC 43 298 14288 0.798 22.32 22.85 7.68 GGC 43 300 14386 0.802 22.43 22.96 6.39 GGC 43 302 14486 0.804 22.50 23.03 8.38 249 GGC 43 304 14588 0.794 22.21 22.73 6.61 GGC 43 306 14692 0.798 22.33 22.85 8.76 GGC 43 308 14799 0.797 22.30 22.83 7.93 GGC 43 310 14908 0.798 22.31 22.84 6.51 GGC 43 GGC 43 312 15020 0.798 22.33 22.85 7.85 GGC 43 314 15135 0.797 22.30 22.82 8.12 GGC 43 314 15135 GGC 43 316 15252 0.797 22.29 22.81 8.28 GGC 43 GGC 43 320 15495 0.802 22.43 22.96 7.03 GGC 43 322 15620 0.800 22.39 22.91 7.91 GGC 43 324 15749 0.797 22.29 22.82 7.17 GGC 43 326 15881 0.797 22.29 22.81 7.49 GGC 43 328 16016 0.799 22.34 22.87 6.11 GGC 43 330 16154 0.795 22.22 22.75 8.08 GGC 43 332 16296 0.789 22.07 22.59 8.61 GGC 43 334 16441 0.800 22.37 22.90 7.15 GGC 43 336 16589 0.792 22.15 22.67 7.78 GGC 43 338 16741 0.795 22.24 22.76 7.85 GGC 43 340 16897 0.789 22.07 22.59 8.28 GGC 43 342 17057 0.779 21.77 22.28 7.44 GGC 43 344 17220 0.787 22.00 22.52 6.57 GGC 43 346 17388 0.777 21.71 22.21 6.52 GGC 43 348 17559 0.780 21.80 22.31 7.78 GGC 43 350 17735 0.776 21.68 22.19 6.40 GGC 43 352 17915 0.786 21.99 22.50 7.77 GGC 43 354 18099 0.769 21.46 21.96 7.92 GGC 43 356 18288 0.785 21.93 22.44 8.39 GGC 43 358 18481 0.779 21.78 22.28 7.12 GGC 43 360 18679 0.795 22.23 22.75 7.95 GGC 43 362 18882 0.790 22.10 22.62 8.00 CDH 41 66 18935 0.795 22.24 22.76 7.56 CDH 41 68 19099 0.799 22.36 22.88 11.12 CDH 41 70 19260 0.794 22.21 22.73 7.73 CDH 41 70 19260 CDH 41 72 19416 0.795 22.23 22.75 6.96 CDH 41 74 19569 0.797 22.29 22.82 6.65 CDH 41 76 19718 0.793 22.16 22.68 7.46 CDH 41 78 19863 0.795 22.23 22.76 6.31 250 CDH 41 80 20005 0.798 22.32 22.85 8.03 CDH 41 82 20143 0.792 22.14 22.66 7.74 CDH 41 84 20278 0.794 22.20 22.72 7.58 CDH 41 86 20410 0.788 22.03 22.55 6.69 CDH 41 88 20539 0.790 22.09 22.61 7.04 CDH 41 90 20665 0.791 22.10 22.62 5.83 CDH 41 92 20788 0.797 22.30 22.83 6.39 CDH 41 94 20908 0.797 22.28 22.81 7.74 CDH 41 96 21026 0.799 22.34 22.87 7.06 CDH 41 98 21142 0.800 22.37 22.90 7.19 CDH 41 100 21255 0.795 22.22 22.74 6.69 CDH 41 100 21255 CDH 41 102 21365 0.794 22.20 22.72 7.21 CDH 41 104 21474 0.794 22.21 22.73 7.17 CDH 41 106 21580 0.796 22.25 22.78 6.13 CDH 41 108 21685 0.793 22.16 22.68 6.61 CDH 41 110 21787 0.788 22.04 22.55 6.94 CDH 41 112 21888 0.790 22.10 22.62 6.05 CDH 41 114 21987 0.780 21.79 22.29 5.84 CDH 41 116 22085 0.791 22.12 22.64 6.02 CDH 41 118 22181 0.798 22.33 22.85 5.72 CDH 41 118 22181 CDH 41 120 22275 0.791 22.12 22.63 7.18 CDH 41 122 22368 0.798 22.32 22.84 7.45 CDH 41 124 22460 0.797 22.28 22.81 7.62 CDH 41 126 22550 0.799 22.35 22.88 6.96 CDH 41 128 22639 0.794 22.22 22.74 7.44 CDH 41 128 22639 CDH 41 130 22728 0.793 22.18 22.71 8.43 CDH 41 132 22815 0.796 22.26 22.79 8.16 CDH 41 134 22901 0.798 22.34 22.86 8.37 CDH 41 136 22986 0.798 22.34 22.86 8.14 CDH 41 138 23071 0.796 22.27 22.80 9.21 CDH 41 140 23155 0.790 22.09 22.61 10.55 CDH 41 140 23155 CDH 41 142 23238 0.791 22.11 22.63 7.32 CDH 41 144 23320 0.789 22.06 22.57 7.79 CDH 41 146 23402 0.794 22.21 22.73 9.25 CDH 41 148 23483 0.801 22.42 22.95 5.45 CDH 41 150 23564 0.800 22.38 22.91 7.45 251 CDH 41 152 23645 0.800 22.37 22.90 8.27 CDH 41 152 23645 CDH 41 154 23725 0.799 22.36 22.89 7.63 CDH 41 156 23805 0.803 22.47 23.00 7.19 CDH 41 158 23884 0.804 22.49 23.02 7.73 CDH 41 160 23963 0.802 22.45 22.98 8.08 CDH 41 162 24042 0.801 22.40 22.93 7.72 CDH 41 164 24121 0.800 22.38 22.91 9.02 CDH 41 166 24200 0.804 22.51 23.04 8.39 CDH 41 168 24279 0.801 22.40 22.93 7.27 CDH 41 170 24357 0.804 22.49 23.02 7.99 CDH 41 172 24436 0.799 22.35 22.87 8.44 CDH 41 174 24514 0.797 22.30 22.82 9.97 CDH 41 176 24593 0.796 22.26 22.78 9.90 CDH 41 178 24671 0.796 22.27 22.79 11.41 CDH 41 178 24671 CDH 41 180 24750 0.801 22.41 22.94 10.84 CDH 41 182 24829 0.801 22.40 22.93 8.91 CDH 41 184 24907 0.808 22.63 23.16 6.16 CDH 41 186 24986 0.796 22.26 22.78 10.08 CDH 41 188 25066 0.798 22.32 22.84 6.81 CDH 41 190 25145 0.803 22.46 22.99 8.56 CDH 41 192 25224 0.805 22.52 23.06 7.46 CDH 41 194 25304 0.803 22.47 23.00 9.88 CDH 41 196 25384 0.804 22.49 23.03 10.17 CDH 41 198 25463 0.807 22.60 23.13 8.29 CDH 41 200 25544 0.805 22.52 23.05 10.76 CDH 41 220 26353 0.800 22.38 22.91 6.90 CDH 41 240 27168 0.805 22.52 23.05 7.84 CDH 41 260 27956 0.809 22.65 23.19 11.98 CDH 41 280 28667 0.810 22.69 23.22 10.05 CDH 41 300 29228 0.802 22.43 22.96 1.60 CDH 41 320 29548 0.802 22.43 22.96 1.40 252 ME0005A-15MC/17JC: Age Depth (yrs, Marine13, SST (°C, SST (°C, C37total Brassicasterol k' Core (cm) Clam) U 37 Prahl) Muller) (nmol/g) (nmol/g) ME0005-15MC 2.25 6624 0.951 26.84 27.50 0.42 0.11 ME0005-15MC 4.25 7607 0.948 26.74 27.39 0.66 0.15 ME0005-15MC 5.25 8099 0.950 26.80 27.46 0.84 0.17 ME0005-15MC 6.25 8590 0.951 26.83 27.49 0.67 0.13 ME0005-15MC 8.25 9573 0.949 26.76 27.42 0.67 0.18 ME0005-15MC 10.25 10556 0.947 26.71 27.37 0.77 0.15 ME0005-15MC 14.25 12522 0.942 26.57 27.22 0.85 0.14 ME0005-15MC 15.25 13013 0.940 26.49 27.14 0.90 0.14 ME0005-17JC 2.25 7360 0.946 26.67 27.33 0.67 0.12 ME0005-17JC 4.25 8480 0.950 26.79 27.45 0.62 0.09 ME0005-17JC 6.25 9589 0.945 26.64 27.30 0.75 0.12 ME0005-17JC 8.25 10687 0.947 26.70 27.36 0.76 0.18 ME0005-17JC 10.25 11774 0.942 26.57 27.23 0.84 0.12 ME0005-17JC 12.25 12849 0.935 26.35 26.99 1.02 0.16 ME0005-17JC 14.25 13913 0.935 26.36 27.01 1.02 0.14 ME0005-17JC 16.25 14966 0.936 26.37 27.02 0.93 0.14 ME0005-17JC 18.25 16007 0.934 26.33 26.98 1.46 0.18 ME0005-17JC 20.25 17038 0.928 26.16 26.80 1.73 0.26 ME0005-17JC 22.25 18057 0.927 26.11 26.75 2.33 0.28 ME0005-17JC 24.25 19064 0.921 25.96 26.59 2.91 0.31 ME0005-17JC 26.25 20061 0.913 25.72 26.35 3.74 0.52 ME0005-17JC 28.25 21046 0.916 25.80 26.43 3.38 0.44 ME0005-17JC 30.25 22020 0.927 26.12 26.76 4.13 0.46 ME0005-17JC 32.25 22982 0.920 25.91 26.55 4.17 0.42 ME0005-17JC 34.25 23934 0.932 26.27 26.91 4.19 0.35 ME0005-17JC 36.25 24874 0.924 26.02 26.65 3.71 0.38 ME0005-17JC 38.25 25803 0.922 25.97 26.61 3.43 0.39 ME0005-17JC 40.25 26720 0.924 26.02 26.66 3.31 0.32 ME0005-17JC 42.25 27627 0.921 25.94 26.57 2.85 0.28 ME0005-17JC 44.25 28522 0.920 25.92 26.55 2.58 0.33 ME0005-17JC 46.25 29406 0.934 26.33 26.98 2.32 0.25 ME0005-17JC 48.25 30278 0.932 26.26 26.91 1.99 0.23 ME0005-17JC 50.25 31139 0.938 26.45 27.10 1.57 0.18 ME0005-17JC 52.25 31989 0.934 26.33 26.97 1.78 0.17 ME0005-17JC 54.25 32828 0.948 26.75 27.41 1.88 0.11 ME0005-17JC 56.25 33656 0.935 26.36 27.01 1.85 0.15 ME0005-17JC 58.25 34472 0.948 26.73 27.39 1.86 0.11 ME0005-17JC 60.25 35277 0.944 26.61 27.14 1.31 0.10 ME0005-17JC 62.25 36070 0.943 26.58 27.23 1.82 0.16 ME0005-17JC 64.25 36853 0.940 26.49 27.15 1.72 0.13 ME0005-17JC 66.25 37624 0.949 26.77 27.43 2.02 0.12 ME0005-17JC 68.25 38384 0.943 26.60 27.25 2.37 0.13 253 ME0005-17JC 70.25 39133 0.946 26.68 27.34 2.00 0.11 VNTR01-13GC: Adjusted Age (yrs, k SST (°C, SST (°C, C37total Brassicasterol Depth (cm) Depth Marine13, U '37 Prahl) Muller) (nmol/g) (nmol/g) (cm) Clam) 2.25 2.25 764 0.836 23.46 24.01 0.30 0.08 5.25 5.25 2883 0.04 5.75 5.75 3204 0.839 23.53 24.09 0.27 0.09 8.25 8.25 4682 0.10 8.75 8.75 4954 0.841 23.58 24.14 0.25 0.07 10.25 10.25 5727 0.06 10.75 10.75 5971 0.832 23.33 23.89 0.21 0.06 12.25 12.25 6664 0.839 23.52 24.00 0.35 0.07 15.25 15.25 7895 0.07 15.75 15.75 8082 0.838 23.50 24.06 0.22 0.08 17.25 17.25 8615 0.838 23.49 24.06 0.34 0.07 17.75 17.75 8784 0.841 23.60 23.97 0.22 0.08 19.75 19.75 9420 0.829 23.22 23.77 0.08 20.25 20.25 9570 0.855 24.01 24.59 1.26 0.08 23.25 23.25 10401 0.859 24.11 24.69 0.34 0.09 28.25 28.25 11589 0.851 23.88 24.45 0.95 0.10 30.25 30.25 12014 0.845 23.72 24.29 1.37 0.11 32.25 32.25 12419 0.846 23.74 24.31 0.90 0.07 36.25 36.25 13192 0.867 24.34 24.93 1.12 0.06 40.25 40.25 13940 0.824 23.10 23.65 2.09 0.16 44.25 44.25 14690 0.804 22.49 23.02 2.93 0.29 48.25 48.25 15458 0.794 22.20 22.72 3.56 0.36 50.25 15852 22.83 3.86 52.25 52.25 16255 0.797 22.31 22.70 3.29 0.33 90.25 54.25 16666 0.793 22.18 22.67 4.48 0.41 92.25 56.25 17086 0.799 22.35 22.69 4.69 0.21 96.25 60.25 17951 0.869 24.40 22.65 5.27 0.24 100.25 64.25 18843 0.797 22.28 22.80 3.96 0.33 102.25 66.25 19298 0.824 23.08 23.63 2.15 0.13 104.25 68.25 19755 0.813 22.77 23.30 2.66 0.23 106.25 70.25 20215 0.816 22.85 23.39 4.10 0.33 108.25 72.25 20675 0.812 22.73 23.27 3.82 0.38 110.25 74.25 21133 0.802 22.43 22.96 4.41 0.33 112.25 76.25 21587 0.815 22.82 23.36 4.01 0.26 254 114.25 78.25 22036 0.772 21.56 22.06 2.62 0.23 116.25 80.25 22478 0.819 22.95 23.49 5.39 0.19 118.25 82.25 22910 0.810 22.68 23.22 3.15 0.27 120.25 84.25 23330 0.806 22.56 23.10 3.90 0.36 122.25 86.25 23738 0.764 21.32 21.82 3.38 0.54 124.25 88.25 24130 0.754 21.03 21.51 3.93 0.76 126.25 90.25 24505 0.756 21.10 21.58 3.21 0.64 128.25 92.25 24861 0.785 21.95 22.47 4.40 0.47 130.25 94.25 25197 0.799 22.37 22.89 4.80 0.27 132.25 96.25 25512 0.765 21.35 21.85 3.50 0.52 134.25 98.25 25804 0.733 20.41 20.88 3.38 0.70 136.25 100.25 26072 0.758 21.15 21.64 3.57 0.64 138.25 102.25 26316 0.752 20.97 21.46 3.67 0.68 140.25 104.25 26535 0.755 21.07 21.56 4.04 0.61 142.25 106.25 26729 0.745 20.76 21.24 3.72 0.60 144.25 108.25 26898 0.757 21.11 21.60 4.08 0.56 148.25 112.25 27165 0.777 21.71 22.22 1.43 0.44 150.25 114.25 27264 0.799 22.36 22.89 6.56 0.36 154.25 118.25 27403 0.785 21.95 22.47 1.71 0.38 158.25 122.25 27479 0.788 22.03 22.55 0.98 0.21 162.25 126.25 27518 0.795 22.25 22.77 1.85 0.22 166.25 130.25 27556 0.813 22.77 23.30 1.33 0.16 168.25 132.25 27589 0.810 22.69 23.23 0.88 0.17 170.25 134.25 27638 0.836 23.43 23.99 2.02 0.00 190.25 154.25 31016 0.800 22.40 22.92 3.38 0.02 255 APPENDIX B: Stable Isotope Data CDH-23: 13 18 Age δ C (‰, VPDB) δ O (‰, VPDB) Depth (cm) (yrs BP, Fairbanks) U. Peregrina U.Peregrina 15.5 1105 0.07 1.89 26 1247 0.06 1.90 30 1300 0.17 1.82 42 1457 0.07 1.88 58 1661 0.22 1.78 68.5 1793 0.16 1.75 74 1861 0.14 1.86 89 2043 0.23 1.80 90 2055 0.15 1.85 92.5 2085 0.18 1.81 106 2244 0.15 1.80 118.5 2389 0.09 1.78 138 2609 0.00 1.85 154 2785 0.17 1.80 160 2850 0.12 1.81 180 3063 -0.06 1.88 184.5 3110 0.17 1.87 202.5 3296 0.15 1.79 221.75 3490 0.21 1.83 240 3669 0.12 1.84 249.5 3761 0.19 1.83 260.5 3867 0.24 1.81 280 4050 0.08 1.86 296.5 4203 0.13 1.76 301 4244 0.09 1.84 320.5 4420 0.06 1.76 325.5 4465 0.20 1.80 350.5 4686 0.19 1.76 350.50 4686 0.16 1.80 353.5 4712 0.11 1.75 366 4821 0.10 1.67 382 4958 0.15 1.76 393.5 5056 0.09 1.73 400 5112 0.20 1.83 415.5 5242 0.17 1.85 256 421.5 5292 0.16 1.76 442.5 5467 -0.01 1.82 444.5 5483 0.02 1.81 460.5 5615 0.17 1.76 500 5937 0.29 1.91 520 6098 0.05 1.78 550 6339 0.28 1.89 571.5 6512 -0.01 1.83 600 6739 0.11 1.85 603.5 6767 0.12 1.79 660.5 7222 0.02 1.77 675 7338 0.05 1.84 702.5 7559 0.08 1.95 718 7683 0.23 1.99 739 7853 -0.10 1.82 760.5 8027 -0.27 1.99 781 8193 0.03 1.96 791.5 8278 -0.34 1.95 809 8421 -0.12 1.98 823.5 8540 -0.19 2.01 825.5 8556 -0.11 1.95 850.25 8760 -0.36 1.89 873.5 8951 -0.10 1.86 893.5 9117 -0.22 1.84 900 9170 -0.36 1.96 916.5 9307 -0.42 1.87 925 9378 -0.24 1.91 940.5 9507 -0.38 1.92 967.5 9732 -0.28 1.89 975 9794 -0.30 2.07 1000.25 10005 -0.34 1.91 1006.5 10057 -0.28 1.92 1025 10211 -0.33 2.09 1050 10419 -0.39 2.11 1062.5 10523 -0.29 2.01 1073.5 10614 -0.41 2.03 1075 10627 -0.41 1.96 1090 10751 -0.28 2.04 1100 10833 -0.29 2.36 1120 10998 -0.39 2.23 257 1125 11038 -0.49 2.27 1135 11120 -0.47 2.21 1150 11242 -0.55 2.52 1160 11323 -0.59 2.29 1175 11443 -0.68 2.41 1190 11563 -0.56 2.46 1200 11642 -0.41 2.24 1220.5 11802 -0.55 2.36 1230 11876 -0.59 2.46 1250 12029 -0.46 2.50 1275 12217 -0.37 2.32 1285 12291 -0.68 2.62 1297 12378 -0.79 2.51 1302 12414 -0.62 2.69 1325 12578 -0.49 2.68 1340 12682 -0.67 2.71 1350 12750 -0.43 2.80 1355 12783 -0.46 2.48 1358.5 12806 -0.73 2.77 1360.5 12820 -0.51 2.35 1375 12915 -0.65 2.48 1400 13072 -0.68 2.86 1435.5 13282 -0.56 2.76 1440 13308 -0.49 2.42 1470.5 13471 -0.48 2.48 1475 13494 -0.59 2.50 1486.5 13551 -0.77 2.44 1489.5 13565 -0.80 2.63 1500 13615 -0.60 2.69 1525 13724 -0.58 2.59 1550 13821 -0.89 2.76 1560 13856 -1.03 2.78 1575 13905 -0.86 2.93 1585 13935 -0.88 2.68 1600 13975 -0.68 2.56 1625 14029 -0.97 3.02 1645.5 14062 -0.80 2.38 1651 14069 -0.85 2.49 1675 14089 -0.56 2.43 1700 14092 -0.90 2.40 258 1705.5 14091 -0.71 2.41 1712.5 14087 -0.70 2.38 1713 14087 -0.63 2.54 GGC 43/CDH 41: 13 18 Age δ C (‰, VPDB) δ O (‰, VPDB) Core Depth (cm) (yrs BP, Fairbanks) U. Peregrina U. Peregrina GGC43 2 1340 -0.59 2.134 GGC43 20 3178 -0.67 2.461 GGC43 40 4725 -0.70 2.159 GGC43 60 5942 -0.58 2.136 GGC43 80 6969 -0.60 2.104 GGC43 100 7896 -0.63 2.099 GGC43 120 8771 -0.77 2.250 GGC43 131 9216 -0.78 2.205 GGC43 140 9611 -0.70 2.224 GGC43 146 9836 -0.66 2.272 GGC43 160 10413 -0.71 2.353 GGC43 171 10814 -0.79 2.130 GGC43 176 10998 -0.77 2.190 GGC43 180 11161 -0.69 2.398 GGC43 195 11675 -0.82 2.448 GGC43 200 11836 -0.83 2.367 GGC43 203 11931 -0.65 2.421 GGC43 205 11993 -0.87 2.494 GGC43 210 12143 -0.75 2.581 GGC43 215 12289 -0.81 2.805 GGC43 220 12429 -0.75 2.879 GGC43 225 12565 -0.83 2.778 GGC43 230 12696 -0.82 2.708 GGC43 235 12823 -0.83 2.945 GGC43 240 12946 -0.73 2.866 GGC43 245 13066 -0.80 2.939 GGC43 250 13185 -0.89 2.762 GGC43 255 13302 -0.83 2.632 GGC43 260 13420 -0.76 2.880 GGC43 265 13539 -0.88 2.761 GGC43 270 13661 -0.64 3.137 GGC43 275 13787 -0.81 2.863 259 GGC43 280 13920 -0.83 2.682 GGC43 285 14062 -0.82 2.930 GGC43 290 14215 -0.80 2.904 GGC43 295 14381 -0.80 2.996 GGC43 300 14564 -0.73 3.034 GGC43 305 14766 -0.80 3.014 GGC43 310 14991 -0.78 3.047 GGC43 313 15138 -0.61 3.144 GGC43 315 15241 -0.77 3.078 GGC43 318 15406 -0.75 3.023 GGC43 320 15522 -0.65 3.249 GGC43 323 15707 -0.60 3.038 GGC43 325 15837 -0.80 2.863 GGC43 328 16044 -0.66 3.030 GGC43 330 16191 -0.71 3.040 GGC43 333 16423 -0.64 2.980 GGC43 335 16587 -0.65 3.195 GGC43 338 16848 -0.57 3.206 GGC43 340 17032 -0.82 3.379 GGC43 343 17324 -0.55 3.172 GGC43 345 17531 -0.83 3.156 GGC43 348 17858 -0.65 3.188 GGC43 350 18088 -0.75 3.334 GGC43 353 18454 -0.44 3.252 GGC43 355 18711 -0.84 3.218 GGC43 358 19119 -0.48 3.205 GGC43 360 19406 -0.71 3.132 GGC43 363 19859 -0.58 3.088 GGC43 365 20178 -0.71 3.318 CDH 41 85 20265 -0.67 3.454 CDH 41 87 20431 -0.45 3.431 CDH 41 90 20654 -0.75 3.164 CDH 41 92 20830 -0.41 3.266 CDH 41 95 21059 -0.74 3.254 CDH 41 97 21207 -0.58 3.195 CDH 41 100 21422 -0.70 3.128 CDH 41 105 21762 -0.67 3.404 CDH 41 110 22064 -0.74 3.337 CDH 41 115 22376 -0.77 3.453 CDH 41 120 22652 -0.68 3.123 260 CDH 41 125 22897 -0.72 3.327 CDH 41 130 23137 -0.76 3.265 CDH 41 135 23362 -0.60 3.348 CDH 41 140 23583 -0.67 3.316 CDH 41 145 23772 -0.69 3.248 CDH 41 150 23962 -0.66 3.143 CDH 41 155 24144 -0.54 3.237 CDH 41 160 24321 -0.60 3.073 CDH 41 165 24494 -0.69 3.266 CDH 41 170 24666 -0.61 3.041 CDH 41 175 24840 -0.72 3.280 CDH 41 180 25026 -0.58 3.233 CDH 41 185 25198 -0.63 3.290 CDH 41 190 25387 -0.62 3.237 CDH 41 195 25584 -0.69 3.283 CDH 41 200 25801 -0.64 3.200 CDH-26 13 18 δ C (‰, VPDB) δ O (‰, VPDB) Depth (cm) Age (yrs BP) U. Peregrina U. Peregrina 32 1924 -0.29 2.96 64 2976 -0.39 3.06 96 3948 -0.29 2.95 128 4846 -0.52 2.89 171.5 5956 -0.37 2.88 203.5 6698 -0.41 2.91 235.5 7382 -0.49 2.92 267.5 8012 -0.57 2.99 325 9022 -0.57 3.10 357 9523 -0.38 3.21 389 9985 -0.63 3.12 410 10268 -0.79 3.27 421 10411 -0.60 3.13 478 11093 -0.64 3.29 510 11437 -0.61 3.36 542 11756 -0.78 3.14 574 12054 -0.77 3.40 631 12539 -0.63 3.58 663 12788 -0.86 3.71 261 695 13025 -0.99 3.70 727 13250 -1.07 3.76 784.5 13631 -1.04 3.78 816.5 13832 -1.10 3.77 848.5 14028 -1.35 3.82 880.5 14220 -0.82 3.81 959.5 14680 -0.77 3.83 971.5 14749 -0.80 3.81 1003.5 14933 -0.86 3.88 1035.5 15117 -0.88 3.94 1092.5 15446 -0.87 3.94 1124.5 15633 -0.66 3.99 1156.5 15822 -0.92 4.05 1188.5 16013 -1.26 4.13 1247.5 16371 -0.89 4.19 1263.5 16470 -1.24 4.19 1295.5 16670 -1.46 4.25 1343.5 16975 -0.96 4.38 1400.5 17345 -0.89 4.34 1432.5 17557 -0.97 4.40 1464.5 17771 -1.43 4.43 1496.5 17988 -1.21 4.50 1555.5 18394 -1.47 4.45 1587.5 18616 -1.29 4.51 1619.5 18841 -1.46 4.50 1651.5 19066 -0.98 4.40 1709 19475 -1.13 4.46 1741 19703 -1.17 4.48 1773 19933 -0.89 4.51 1805 20162 -1.61 4.56 1863 20579 -0.97 4.39 1895 20809 -1.08 4.56 1927 21039 -0.90 4.56 1959 21269 -1.03 4.42 2016 21679 -0.96 4.44 2048 21909 -0.71 4.44 2080 22140 -1.21 4.44 2120 22429 -0.45 4.40 2169 22785 -0.98 4.50 2201 23020 -0.51 4.36 262 2233 23256 -0.63 4.36 2265 23495 -0.70 4.42 2322 23931 -0.75 4.32 2354 24181 -0.75 4.31 2386 24438 -0.69 4.40 2418 24701 -0.67 4.34 2475 25192 -0.89 4.38 2507 25481 -0.59 4.49 2539 25783 -0.61 4.42 Individual Benthic Stable Isotope Data 13 18 δ C (‰, VPDB) δ O (‰, VPDB) Core ID Depth Midpoint Age (yrs) Weight (mg) U. Peregrina U. Peregrina MC-18A 0.5 49 40 -0.368 2.901 MC-18A 0.5 49 40 -0.458 2.837 MC-18A 1.5 49 30 -0.360 2.847 MC-18A 1.5 49 42 -0.453 2.943 MC-18A 2.5 49 32 -0.299 2.921 MC-18A 2.5 49 28 -0.417 2.757 MC-18A 3.5 49 34 -0.232 2.954 MC-18A 3.5 49 44 -0.214 2.833 MC-18A 4.5 49 34 -0.517 2.692 MC-18A 4.5 49 42 -0.377 2.838 MC-18A 5.5 49 24 -0.106 3.090 MC-18A 5.5 49 36 -0.287 2.898 MC-18A 6.5 49 30 -0.289 2.829 MC-18A 6.5 49 36 -0.450 2.819 MC-18A 7.5 49 28 -0.636 2.954 MC-18A 7.5 49 46 -0.131 2.893 MC-18A 8.5 49 32 -0.304 2.943 MC-18A 8.5 49 38 -0.522 2.827 MC-18A 9.5 49 30 -0.389 2.788 MC-18A 9.5 49 34 -0.233 2.678 MC-18A 10.5 49 24 -0.431 3.108 MC-18A 10.5 49 36 -0.244 2.930 MC-18A 10.5 49 34 -0.423 2.936 MC-18A 11.5 49 32 -0.329 3.070 MC-18A 11.5 49 36 -0.457 2.815 MC-18A 12.5 49 26 -0.630 2.863 263 MC-18A 12.5 49 42 -0.331 2.866 MC-18A 13.5 49 30 -0.304 2.853 MC-18A 13.5 49 28 -0.256 2.682 MC-18A 14.5 49 32 -0.262 2.676 MC-18A 14.5 49 32 -0.246 2.813 MC-18A 15.5 49 40 -0.374 2.799 MC-18A 15.5 49 34 -7.634 -5.099 MC-18A 16.5 49 34 -0.205 2.831 MC-18A 16.5 49 42 -0.296 2.895 MC-18A 17.5 49 28 -0.475 2.882 MC-18A 17.5 49 32 -0.081 2.829 MC-18A 18.5 49 32 -0.380 2.829 MC-18A 18.5 49 38 -0.295 2.852 MC-18A 19.5 49 40 -0.181 2.822 MC-18A 19.5 49 42 -0.314 2.879 CDH-26 71 3267 34 0.082 2.916 CDH-26 71 3267 30 -0.424 2.820 CDH-26 71 3267 28 -0.159 2.857 CDH-26 71 3267 36 -0.283 2.809 CDH-26 71 3267 34 -0.455 2.847 CDH-26 71 3267 36 -0.417 2.847 CDH-26 72 3267 20 -0.629 2.938 CDH-26 72 3267 36 -0.374 2.834 CDH-26 72 3267 38 -0.565 2.849 CDH-26 72 3267 30 -0.347 2.904 CDH-26 72 3267 36 -0.395 2.784 CDH-26 72 3267 48 -0.273 2.842 CDH-26 73 3267 24 -0.077 2.945 CDH-26 73 3267 34 -0.342 3.101 CDH-26 73 3267 34 -0.173 2.979 CDH-26 73 3267 38 -0.412 2.761 CDH-26 73 3267 28 -0.430 2.788 CDH-26 73 3267 32 -0.575 2.735 CDH-26 73 3267 40 -0.223 2.729 CDH-26 74 3267 24 -0.598 3.081 CDH-26 74 3267 28 -0.787 2.849 CDH-26 74 3267 24 -0.422 2.684 CDH-26 74 3267 42 -0.213 2.874 CDH-26 74 3267 32 -0.197 2.696 CDH-26 74 3267 52 -0.186 2.886 264 CDH-26 74 3267 62 -0.116 2.742 CDH-26 75 3267 24 -0.075 2.626 CDH-26 75 3267 26 -0.617 2.852 CDH-26 75 3267 38 -0.331 2.832 CDH-26 75 3267 34 -0.307 2.929 CDH-26 75 3267 36 -0.196 2.748 CDH-26 75 3267 46 -0.208 2.814 CDH-26 76 3267 26 -0.040 3.062 CDH-26 76 3267 26 -0.276 2.889 CDH-26 76 3267 24 -0.397 2.978 CDH-26 76 3267 28 -0.351 2.933 CDH-26 76 3267 24 -0.886 2.762 CDH-26 76 3267 56 -0.511 2.886 CDH-26 76 3267 30 -0.469 2.848 CDH-26 76 3267 46 -0.304 2.773 CDH-26 103.5 4080 36 -0.103 2.917 CDH-26 103.5 4080 22 -0.167 2.503 CDH-26 103.5 4080 38 -0.217 2.961 CDH-26 103.5 4080 28 -0.462 2.905 CDH-26 103.5 4080 36 -0.289 2.675 CDH-26 103.5 4080 42 -0.351 2.826 CDH-26 104.5 4080 22 -0.188 3.109 CDH-26 104.5 4080 26 -0.631 2.939 CDH-26 104.5 4080 38 -0.120 2.648 CDH-26 104.5 4080 34 -0.076 2.609 CDH-26 104.5 4080 32 -0.194 2.825 CDH-26 104.5 4080 56 -0.267 2.786 CDH-26 105.5 4080 24 -0.254 2.976 CDH-26 105.5 4080 32 -0.443 2.865 CDH-26 105.5 4080 28 -0.390 2.802 CDH-26 105.5 4080 34 -0.364 2.906 CDH-26 105.5 4080 34 -0.154 2.844 CDH-26 106.5 4080 34 -0.759 2.895 CDH-26 106.5 4080 24 0.027 3.070 CDH-26 106.5 4080 34 -0.201 3.043 CDH-26 106.5 4080 26 -0.091 2.752 CDH-26 106.5 4080 38 -0.174 2.877 CDH-26 106.5 4080 34 -0.396 2.771 CDH-26 106.5 4080 32 -0.472 2.788 CDH-26 106.5 4080 32 -0.471 2.778 265 CDH-26 107.5 4080 34 -0.253 3.067 CDH-26 107.5 4080 16 -0.665 2.889 CDH-26 107.5 4080 28 -0.102 2.563 CDH-26 107.5 4080 32 -0.265 2.650 CDH-26 107.5 4080 40 -0.482 2.840 CDH-26 107.5 4080 50 -0.480 2.831 CDH-26 108.5 4080 32 -0.169 2.849 CDH-26 108.5 4080 36 -0.124 2.691 CDH-26 108.5 4080 46 -0.307 2.697 CDH-26 108.5 4080 42 -0.430 2.831 CDH-26 108.5 4080 50 -0.334 2.836 CDH-26 109.5 4080 42 -0.484 2.123 CDH-26 109.5 4080 42 -0.765 2.285 CDH-26 109.5 4080 38 -0.512 2.914 CDH-26 109.5 4080 36 -0.482 2.883 CDH-26 109.5 4080 50 -0.304 2.812 CDH-26 198.5 6194 44 -0.495 2.811 CDH-26 198.5 6194 38 -0.353 2.915 CDH-26 198.5 6194 56 -0.419 2.811 CDH-26 198.5 6194 36 -0.537 2.820 CDH-26 199.5 6194 38 -0.275 2.792 CDH-26 199.5 6194 34 -0.563 2.745 CDH-26 199.5 6194 38 -0.429 2.788 CDH-26 199.5 6194 44 -0.384 2.768 CDH-26 200.5 6194 24 -0.392 3.023 CDH-26 200.5 6194 30 -0.567 2.799 CDH-26 200.5 6194 38 -0.433 2.808 CDH-26 200.5 6194 40 -0.571 2.692 CDH-26 201.5 6194 34 -0.570 2.615 CDH-26 201.5 6194 40 -0.491 2.746 CDH-26 201.5 6194 58 -0.532 2.722 CDH-26 201.5 6194 48 -0.220 2.725 CDH-26 202.5 6194 48 -0.078 3.215 CDH-26 202.5 6194 18 -0.461 2.876 CDH-26 202.5 6194 34 -0.708 2.749 CDH-26 202.5 6194 40 -0.644 2.707 CDH-26 202.5 6194 34 -0.358 2.747 CDH-26 203.5 6194 32 -0.474 2.763 CDH-26 203.5 6194 44 -0.397 2.710 CDH-26 203.5 6194 40 -0.192 2.813 266 CDH-26 203.5 6194 50 -0.470 2.692 CDH-26 204.5 6194 28 -0.524 2.806 CDH-26 204.5 6194 34 -0.406 2.832 CDH-26 204.5 6194 34 -0.597 2.795 CDH-26 204.5 6194 50 -0.288 2.612 CDH-26 205.5 6194 34 -0.443 2.787 CDH-26 205.5 6194 38 -0.670 2.697 CDH-26 205.5 6194 34 -0.476 2.811 CDH-26 205.5 6194 44 -0.390 2.747 CDH-26 206.5 6194 22 -0.492 2.726 CDH-26 206.5 6194 28 -0.412 2.836 CDH-26 206.5 6194 38 -0.461 2.716 CDH-26 206.5 6194 46 -0.553 2.691 CDH-26 207.5 6194 26 -0.608 2.858 CDH-26 207.5 6194 26 -0.563 2.797 CDH-26 207.5 6194 30 -0.624 2.823 CDH-26 207.5 6194 34 -0.308 2.773 CDH-26 296.5 8004 30 -0.370 3.072 CDH-26 296.5 8004 34 -0.575 2.845 CDH-26 296.5 8004 32 -0.544 2.652 CDH-26 297.5 8004 28 -0.855 2.927 CDH-26 297.5 8004 28 -0.531 2.909 CDH-26 297.5 8004 40 -0.554 2.821 CDH-26 298.5 8004 26 -0.624 2.878 CDH-26 298.5 8004 32 -0.742 2.836 CDH-26 298.5 8004 28 -0.543 2.864 CDH-26 299.5 8004 20 -0.562 3.082 CDH-26 299.5 8004 24 -0.860 2.837 CDH-26 299.5 8004 36 -0.311 2.855 CDH-26 300.5 8004 34 -0.720 2.861 CDH-26 300.5 8004 48 -0.666 2.932 CDH-26 300.5 8004 44 -0.892 2.922 CDH-26 300.5 8004 74 -0.772 2.802 CDH-26 301.5 8004 32 -0.527 2.969 CDH-26 301.5 8004 26 -0.589 2.968 CDH-26 301.5 8004 32 -0.827 2.792 CDH-26 301.5 8004 44 -0.513 2.849 CDH-26 302.5 8004 30 -1.010 2.333 CDH-26 302.5 8004 42 -0.429 2.970 CDH-26 302.5 8004 46 -0.658 2.832 267 CDH-26 302.5 8004 36 -0.541 2.839 CDH-26 303.5 8004 34 -0.724 2.796 CDH-26 303.5 8004 30 -0.584 2.846 CDH-26 303.5 8004 38 -0.483 2.782 CDH-26 303.5 8004 44 -0.511 2.703 CDH-26 304.5 8004 24 -0.545 2.893 CDH-26 304.5 8004 38 -0.694 2.985 CDH-26 304.5 8004 30 -0.669 2.969 CDH-26 304.5 8004 38 -0.283 2.925 CDH-26 305.5 8004 32 -0.757 3.023 CDH-26 305.5 8004 32 -0.642 2.977 CDH-26 305.5 8004 26 -0.923 2.456 CDH-26 306.5 8004 32 -0.628 2.828 CDH-26 306.5 8004 76 -0.527 3.121 CDH-26 306.5 8004 36 -0.443 2.861 CDH-26 307.5 8004 30 -0.835 2.783 CDH-26 307.5 8004 36 -0.499 2.802 CDH-26 307.5 8004 40 -0.566 2.860 CDH-26 79.5 3493 44 -0.462 2.863 CDH-26 79.5 3493 47 -0.183 2.910 CDH-26 79.5 3493 23 0.020 2.602 CDH-26 79.5 3493 26 -0.339 2.846 CDH-26 81.5 3493 50 -0.306 2.841 CDH-26 79.5 3493 44 -0.130 2.793 CDH-26 83.5 3493 25 -0.164 2.888 CDH-26 81.5 3493 45 0.013 2.833 CDH-26 81.5 3493 29 0.043 2.874 CDH-26 79.5 3493 26 -0.140 2.838 CDH-26 81.5 3493 29 -0.155 3.094 CDH-26 81.5 3493 30 -0.256 2.951 CDH-26 84.5 3493 61 -0.317 2.856 CDH-26 84.5 3493 23 -0.427 2.775 CDH-26 84.5 3493 39 -0.269 2.799 CDH-26 84.5 3493 48 -0.203 2.906 CDH-26 85.5 3493 46 -0.216 2.829 CDH-26 84.5 3493 28 0.101 3.355 CDH-26 85.5 3493 43 -0.446 2.858 CDH-26 84.5 3493 25 -0.334 3.087 CDH-26 85.5 3493 36 -0.534 2.777 CDH-26 85.5 3493 59 -0.256 2.811 268 CDH-26 85.5 3493 32 -0.391 2.905 CDH-26 80.5 3493 35 -0.496 2.699 CDH-26 80.5 3493 28 -0.409 2.894 CDH-26 80.5 3493 40 -0.349 2.666 CDH-26 80.5 3493 37 -0.148 3.054 CDH-26 80.5 3493 58 -0.522 2.824 CDH-26 82.5 3493 38 -0.621 2.873 CDH-26 82.5 3493 25 -0.373 2.861 CDH-26 82.5 3493 17 0.234 3.090 CDH-26 82.5 3493 38 -0.238 2.934 CDH-26 82.5 3493 27 -0.433 3.027 CDH-26 82.5 3493 28 -0.118 2.941 CDH-26 83.5 3493 38 -0.347 2.753 CDH-26 83.5 3493 36 -0.336 2.674 CDH-26 83.5 3493 31 -0.340 2.829 CDH-26 81.5 3493 87 -0.078 3.047 CDH-26 83.5 3493 41 -0.367 2.840 CDH-26 83.5 3493 33 -0.476 2.818 CDH-26 240.5 7003 26 -0.349 2.899 CDH-26 240.5 7003 45 -0.579 2.876 CDH-26 240.5 7003 41 -0.457 2.836 CDH-26 240.5 7003 24 -0.353 2.843 CDH-26 241.5 7003 26 -0.538 2.972 CDH-26 241.5 7003 37 -0.423 2.907 CDH-26 241.5 7003 30 -0.394 2.826 CDH-26 241.5 7003 23 -0.366 2.930 CDH-26 243.5 7003 28 -0.643 2.811 CDH-26 243.5 7003 30 -0.544 2.838 CDH-26 243.5 7003 36 -0.269 2.816 CDH-26 243.5 7003 24 -0.558 2.874 CDH-26 242.5 7003 30 -0.400 2.823 CDH-26 242.5 7003 24 -0.685 2.837 CDH-26 242.5 7003 34 -0.828 2.735 CDH-26 248.5 7003 38 -0.865 3.172 CDH-26 248.5 7003 46 -0.556 2.774 CDH-26 248.5 7003 31 -1.012 1.227 CDH-26 249.5 7003 51 -0.665 3.065 CDH-26 249.5 7003 36 -0.415 2.717 CDH-26 249.5 7003 24 -0.426 2.891 CDH-26 249.5 7003 30 -0.409 3.062 269 CDH-26 244.5 7003 32 -0.355 2.916 CDH-26 244.5 7003 42 -0.283 2.863 CDH-26 244.5 7003 35 -0.362 2.852 CDH-26 244.5 7003 32 -0.356 2.894 CDH-26 247.5 7003 41 -0.559 2.947 CDH-26 247.5 7003 40 -0.749 2.937 CDH-26 247.5 7003 39 -0.591 2.818 CDH-26 247.5 7003 34 -0.292 2.803 CDH-26 245.5 7003 35 -0.388 2.889 CDH-26 245.5 7003 33 -0.534 2.812 CDH-26 245.5 7003 37 -0.397 3.186 CDH-26 245.5 7003 37 -0.388 3.148 CDH-26 245.5 7003 40 -0.524 2.878 CDH-26 246.5 7003 55 -0.633 2.787 CDH-26 246.5 7003 33 -0.424 3.225 CDH-26 246.5 7003 36 -0.820 2.779 CDH-26 246.5 7003 58 -0.598 2.845 Small Standards Standard: BYM63150 (δ13C= -2.26‰, δ18O= -6.52‰) 13 18 δ C (‰, VPDB) δ O (‰, VPDB) Weight (mg) U. Peregrina U. Peregrina 36 -2.670 -7.131 12 -2.399 -6.734 12 -2.487 -6.488 16 -2.335 -6.519 28 -2.326 -6.365 35 -2.252 -6.184 18 -2.285 -6.389 16 -2.277 -6.365 18 -2.388 -6.502 ? -2.331 -6.613 26 -2.346 -6.350 22 -2.311 -6.400 22 -2.305 -6.452 33 -2.204 -6.314 14 -2.281 -6.350 24 -2.274 -6.491 22 -2.269 -6.398 270 22 -2.311 -6.420 22 -2.352 -6.348 22 -2.266 -6.317 28 -2.293 -6.363 24 -2.249 -6.232 35 -2.338 -6.390 30 -2.291 -6.312 22 -2.294 -6.292 31 -2.266 -6.243 36 -2.245 -6.271 26 -2.330 -6.393 70 -2.314 -6.485 70 -2.314 -6.485 32 -2.233 -6.251 28 -2.321 -6.396 50 -2.263 -6.445 50 -2.263 -6.445 26 -2.304 -6.436 28 -2.272 -6.347 30 -2.253 -6.268 34 -2.287 -6.396 33 -2.299 -6.331 36 -2.473 -6.553 30 -2.304 -6.241 38 -2.351 -6.490 38 -2.342 -6.449 26 -2.320 -6.445 28 -2.324 -6.438 30 -2.296 -6.493 32 -2.299 -6.504 33 -2.330 -6.456 26 -2.279 -6.352 43 -2.338 -6.486 35 -2.288 -6.406 33 -2.359 -6.504 35 -2.342 -6.482 35 -2.372 -6.455 35 -2.472 -6.684 32 -2.326 -6.397 34 -2.306 -6.481 271 70 -2.303 -6.605 29 -2.341 -6.514 33 -2.305 -6.446 35 -2.304 -6.380 35 -2.323 -6.540 40 -2.368 -6.576 33 -2.342 -6.501 38 -2.349 -6.367 32 -2.323 -6.465 36 -2.354 -6.588 36 -2.365 -6.534 70 -2.328 -6.508 37 -2.344 -6.474 37 -2.274 -6.473 70 -2.311 -6.473 80 -2.292 -6.503 34 -2.321 -6.428 32 -2.329 -6.483 38 -2.353 -6.502 70 -2.303 -6.605 70 -2.303 -6.605 36 -2.328 -6.513 32 -2.362 -6.654 110 -2.291 -6.429 33 -2.322 -6.489 50 -2.263 -6.445 ? -2.331 -6.613 ? -2.331 -6.613 90 -2.302 -6.451 40 -2.314 -6.537 70 -2.297 -6.510 54 -2.351 -6.512 57 -2.058 -6.465 70 -2.312 -6.368 83 -2.362 -6.626 59 -2.341 -6.512 70 -2.328 -6.508 70 -2.328 -6.508 40 -2.314 -6.537 35 -2.369 -6.570 272 64 -2.332 -6.635 70 -2.333 -6.499 70 -2.312 -6.368 31 -2.351 -6.602 58 -2.314 -6.473 273 APPENDIX C: Radiocarbon data Depth Weight Accession 14C age 14C Error Core in Core (mg) Benthics Planktonics Number (yrs B.P.) (years) Outlier* CDH 23 30 2.8 Uvigerina 1960 20 CDH 23 30 4 N. dutertrei OS-77623 1680 40 CDH 23 160 2.3 Uvigerina 3330 25 CDH 23 160 3 N. dutertrei OS-77579 3420 35 CDH 23 260.5 2.6 mixed benthics 4290 25 CDH 23 260.5 4 N.dutertrei OS-77576 3980 35 CDH 23 350.5 4.2 Uvigerina 4730 20 CDH 23 350.5 3.3 H. concentrica 4660 25 CDH 23 350.5 3 N.dutertrei OS-77575 4540 40 CDH 23 400 5 N. dutertrei OS-84123 4800 35 CDH 23 460.5 5 N. dutertrei OS- 77546 5290 50 CDH 23 500 2.6 mixed benthics 6370 50 CDH 23 500 5 N. dutertrei OS-84122 5720 30 CDH 23 550 3.1 mixed benthics 6380 50 CDH 23 550 5 N. dutertrei OS-84108 6230 35 CDH 23 600 2.2 mixed benthics 6850 25 CDH 23 600 4 N. dutertrei OS-84168 6510 45 CDH 23 660.5 4 N. dutertrei OS- 77574 6900 40 CDH 23 781 3 N. dutertrei OS-84121 7950 40 CDH 23 850.25 4 N. dutertrei OS- 77580 8640 45 CDH 23 900 7.5 mixed benthics 8990 30 CDH 23 900 4 N. dutertrei OS-84120 8600 45 CDH 23 1000.25 6.6 mixed benthics 9510 30 CDH 23 1000.25 7 N. dutertrei OS- 77547 9220 60 CDH 23 1100 5.2 mixed benthics 10150 35 CDH 23 1100 7 N. dutertrei OS-84115 9920 40 CDH 23 1230 7.6 H. concentrica 10950 45 CDH 23 1230 4 N. dutertrei OS-84130 10750 45 CDH 23 1285 4 N. dutertrei OS-84124 10900 50 CDH 23 1360.5 5.4 Bolivina 11500 40 CDH 23 1360.5 6.9 H. concentrica 11650 35 CDH 23 1360.5 6 N. dutertrei OS- 77573 11650 55 CDH 23 1440 4 Bolivina 12200 45 B>P CDH 23 1440 10 N. dutertrei OS-84155 12000 55 CDH 23 1500 5.6 mixed benthics 12350 40 CDH 23 1500 5 N. dutertrei OS- 77544 12200 70 CDH 23 1600 10 N. dutertrei OS-84116 12600 55 274 CDH 23 1700 6.4 Bolivina 12950 45 CDH 23 1700 4 N. dutertrei OS- 77543 12800 95 CDH 23 118-119 5.8 N. dutertrei OS-101834 3000 20 CDH 23 42-43 6.5 N. dutertrei OS-101833 2110 20 GGC43 2 2.4 Uvigerina OS-109398 2080 20 GGC43 2 3.1 G. ruber OS-99077 1990 50 GGC43 40 3.5 Uvigerina OS-109328 5420 30 GGC43 40 3.9 G. ruber OS-109329 4630 40 GGC43 70.5 1.9 Uvigerina OS-109456 6520 30 GGC43 70.5 4.5 G. ruber OS-109330 6050 35 GGC43 100 3.8 Uvigerina OS-109331 8180 40 GGC43 100 3 G. ruber OS-99078 8030 50 GGC43 140 3.5 Uvigerina OS-109332 9040 40 GGC43 140 4.3 G. ruber OS-109333 8800 40 GGC43 180 3 Uvigerina OS-119052 10,300 30 GGC43 180 6.2 G. ruber OS-119355 10,150 35 GGC43 220 3.2 Uvigerina OS-109364 12200 55 GGC43 220 3.9 G. ruber OS-99079 11150 70 DWI GGC43 280 3.6 G. ruber OS-99080 12600 75 GGC43 320 4.1 G. ruber OS-99081 13800 60 GGC43 350 2.9 G. ruber OS-99082 15500 70 CDH41 0.25 1.9 Uvigerina OS-101662 9560 50 P>B CDH41 0.25 4.1 G. ruber OS-99072 9580 50 CDH41 4.5 4.8 Uvigerina OS-113941 11900 50 CDH41 4.5 4.4 G. ruber OS-113942 10500 45 CDH41 10 3 Uvigerina OS-102464 14900 70 CDH41 10 2.9 G. ruber OS-102469 12150 100 DWI CDH41 15.25 2.2 Uvigerina OS-102712 15300 65 CDH41 15.25 2.1 G. ruber OS-102711 12550 45 CDH41 17.25 2.6 Uvigerina OS-113943 14600 70 CDH41 17.25 6.6 G. ruber OS-113944 12750 55 CDH41 20.25 2.5 Uvigerina OS-102467 14150 70 CDH41 20.25 2.8 G. ruber OS-102466 11850 65 CDH41 25.25 6.9 Uvigerina OS-113945 15500 75 CDH41 25.25 4.5 G. ruber OS-113946 13800 55 CDH41 30 1.6 Uvigerina OS-101663 15400 95 Bova et CDH41 30 3.6 G. ruber OS-99073 14350 60 al., 2015 CDH41 35.25 6.7 Uvigerina OS-113947 16500 80 CDH41 35.25 5.2 G. ruber OS-113948 13950 60 DWI CDH41 40.25 2 Uvigerina OS-102714 14950 60 275 Bova et CDH41 40.25 2.4 G. ruber OS-102713 12700 50 al., 2015 CDH41 47.25 2.5 Uvigerina OS-113949 16700 80 CDH41 47.25 4.5 G. ruber OS-113950 14300 70 DWI CDH41 55.25 2 Uvigerina OS-102716 17700 90 CDH41 55.25 2.2 G. ruber OS-102715 14450 60 DWI CDH41 70 2.1 Uvigerina OS-101664 18850 150 CDH41 70 5.5 G. ruber OS-99074 15650 80 CDH41 90 2.1 Uvigerina OS-102718 21000 130 CDH41 90 2.6 G. ruber OS-102717 18150 90 CDH41 100.25 7.9 Uvigerina OS-119356 22,300 160 CDH41 100.25 5.8 G. ruber OS-119357 18,750 100 CDH41 115.25 2 Uvigerina OS-101665 21600 210 CDH41 115.25 5.1 G. ruber OS-99075 19700 140 CDH41 135 3.4 Uvigerina OS-102413 23100 220 CDH41 135 2.5 G. ruber OS-102719 20400 120 CDH41 170 2.2 Uvigerina OS-101667 23500 240 CDH41 170 3.2 G. ruber OS-99076 20600 120 CDH41 215.25 2.2 Uvigerina OS-118822 24,200 350 Bova et CDH41 215.25 4.7 G. ruber OS-119358 14200 60 al., 2015 CDH41 235.25 3.1 Uvigerina OS-109457 25500 140 Bova et CDH41 235.25 4.6 G. ruber OS-109334 17950 70 al., 2015 CDH41 255.25 2.5 Uvigerina OS-118823 29,800 710 CDH41 255.25 3.5 G. ruber OS-119359 24700 210 CDH41 275.25 5.4 Uvigerina OS-109287 28100 180 Bova et CDH41 275.25 3.3 G. ruber OS-109363 24000 120 al., 2015 CDH41 290.25 4.4 Uvigerina OS-109362 27000 140 Bova et CDH41 290.25 2.9 G. ruber OS-109458 27200 190 al., 2015 CDH41 310.25 2.8 Uvigerina OS-119362 28,100 320 CDH41 310.25 4.5 G. ruber OS-119360 26200 260 CDH 26 96 5 N. dutertrei OS-84131 3970 30 CDH 26 171.5 6 N. dutertrei OS-84109 5980 35 CDH 26 203.5 3 N. dutertrei OS-84125 6320 35 CDH 26 235.5 5.4 uvigerina sp. OS-103000 7620 40 CDH 26 235.5 6 N. dutertrei OS-84164 7070 35 CDH 26 267.5 8 N. dutertrei OS-84113 7440 35 CDH 26 325 6 N. dutertrei OS-84111 8550 40 CDH 26 421 1.8 uvigerina sp. OS-103151 9840 45 CDH 26 421 6 N. dutertrei OS-84110 9460 40 CDH 26 478 8 N. dutertrei OS-84159 10150 55 276 CDH 26 574 1.9 bolivina sp. OS-103153 11700 55 CDH 26 574 9 N. dutertrei OS-84112 10750 45 CDH 26 663 2.7 uvigerina sp. OS-103181 12400 40 CDH 26 663 8 N. dutertrei OS-84161 11650 45 CDH 26 784.5 1.8 mixed benthic OS-103150 13500 75 CDH 26 784.5 7 N. dutertrei OS-84160 12350 50 CDH 26 848.5 9 N. dutertrei OS-84165 12750 50 CDH 26 939.5 1.4 mixed benthic OS-104319 14,200 90 CDH 26 939.5 9 N. dutertrei OS-84156 13100 60 CDH 26 971.5 3.8 N. dutertrei OS-101831 13050 55 CDH 26 1156.5 1.7 mixed benthic OS-103154 14800 65 CDH 26 1156.5 4 N. dutertrei OS-84128 13900 60 CDH 26 1299 6.3 N. dutertrei OS-101830 14750 65 CDH 26 1400.5 4 uvigerina sp. OS-103182 16000 60 CDH 26 1400.5 4 N. dutertrei OS-84132 15150 65 CDH 26 1496.5 8.4 N. dutertrei OS-101835 15550 75 CDH 26 1619.5 9 N. dutertrei OS-84117 16250 65 CDH 26 1741 3.8 uvigerina sp. OS-103183 18050 70 CDH 26 1741 7 N. dutertrei OS-84114 17050 70 CDH 26 1895 2.2 uvigerina sp. OS-103155 18950 100 CDH 26 1895 6 N. dutertrei OS-84118 17700 70 CDH 26 2048 5.5 uvigerina sp. OS-103001 19700 90 CDH 26 2048 9 N. dutertrei OS-84158 18850 110 CDH 26 2169 6 N. dutertrei OS-84119 20200 85 CDH 26 2265 3 N. dutertrei OS-84129 19700 85 CDH 26 2386 2.5 mixed benthic OS-103152 22300 160 CDH 26 2386 5 N. dutertrei OS-84163 20900 80 CDH 26 2418 5.4 N. dutertrei OS-101832 21500 150 CDH 26 2539 3 N. dutertrei OS-84126 21900 95 *DWI = Deep water influence 277 APPENDIX D: Coccolith Relative Abundances Core-top Calibration Dataset: Mean Annual Core-top ID N ratio Latitude Longitude Thermocline Depth (m) ME0005A-38MC 0.494339623 7.3167 -84.113 36.25 ME0005A-14MC 0.355978261 5.8465 -86.449 42.5 046Tri-RC1111 0.512195122 -4.8333 -86.825 41.25 ME0005A-15MC 0.4 4.6137 -86.704 45 ME0005A-20MC 0.431818182 3.2123 -86.486 40 EQP-3a 0.409937888 -0.0448 -105.426 40 ME0005A-29MC 0.663082437 -0.5134 -81.995 27.5 KNR195-05-04-09C-MC 0.474226804 -0.826 -87.90817 27.5 ME0005A-25MC 0.568106312 -1.8534 -82.787 23.75 KNR195-05-13-33D-MC 0.407407407 -3.22358 -82.91417 28.75 KNR195-05-14-34C-MC 0.474666667 -3.59753 -83.96317 31.25 KNR195-5 22MC 0.684745763 -3.8504 -81.25 13.75 KNR195-05-08-18A-MC 0.667808219 -3.98662 -81.3099 16.25 043Tri-RC1111 0.30125523 -11.3333 -106.9667 75 045Tri-RC1111 0.284482759 -11.6 -95.6333 78.75 044Tri-RC1111 0.279411765 -13.3167 -100.95 98.75 KNR195-05 MC 12 0.636666667 -3.71057 -81.10383 13.75 KNR 195-5 25MC 0.75 -3.58253 -81.1645 13.75 Galapagos Core-tops KNR195-05-17-42A-MC 0.288951841 -1.25302 -89.68555 25 KNR195-05-02-04B-MC 0.325342466 -0.21613 -89.50422 26.25 KNR195-05-05-11A-MC 0.37458194 -1.53258 -86.78517 25 Down-core Thermocline Depth Records: ME0005A-15MC/17JC: Age (yrs BP, Thermocline Depth Core ID Core Depth (cm) Marine13) N ratio (m) ME5-15MC 5.25 8099 0.492 28.0 ME5-15MC 5.25 8099 0.519 25.9 ME5-15MC 8.25 9573 0.514 26.3 ME5-15MC 10.25 10556 0.501 27.2 ME5-17JC 8.25 10687 0.485 28.6 ME5-17JC 10.25 11774 0.480 29.1 278 ME5-15MC 15.25 12522 0.550 24.0 ME5-17JC 12.25 12849 0.519 25.9 ME5-17JC 14.25 13913 0.536 24.9 ME5-17JC 16.25 14966 0.539 24.7 ME5-17JC 18.25 16007 0.588 22.3 ME5-17JC 20.25 17038 0.667 19.8 ME5-17JC 22.25 18057 0.618 21.2 ME5-17JC 24.25 19064 0.689 19.3 ME5-17JC 26.25 20061 0.444 33.1 ME5-17JC 26.25 20061 0.571 23.0 ME5-17JC 28.25 21046 0.586 22.4 ME5-17JC 30.25 22020 0.549 24.1 ME5-17JC 32.25 22982 0.592 22.1 ME5-17JC 34.25 23934 0.472 29.9 ME5-17JC 36.25 24874 0.434 34.4 ME5-17JC 40.25 26720 0.456 31.6 ME5-17JC 42.25 27627 0.456 31.6 ME5-17JC 44.25 28522 0.502 27.2 ME5-17JC 46.25 29406 0.458 31.4 ME5-17JC 48.25 30278 0.505 26.9 ME5-17JC 52.25 31989 0.531 25.2 GGC 43/CDH 41: Age (yrs BP, Thermocline Depth Core ID Core Depth (cm) Marine13) N ratio (m) GGC 43 2.25 1055 0.411 41.3 GGC 43 26.25 3124 0.353 54.2 GGC 43 42.25 4313 0.461 33.6 GGC 43 60.25 5493 0.383 46.9 GGC 43 74.25 6308 0.434 37.4 GGC 43 90.25 7143 0.414 40.7 GGC 43 120.25 8472 0.431 37.8 GGC 43 138.25 9147 0.359 52.7 GGC 43 160.250 9875 0.435 37.3 GGC 43 202.25 11078 0.418 39.9 GGC 43 218.25 11514 0.502 28.8 GGC 43 234.25 11962 0.540 25.2 GGC 43 280.25 13501 0.582 22.0 GGC 43 298.25 14297 0.435 37.2 279 GGC 43 325.25 15833 0.436 37.0 GGC 43 338.25 16763 0.545 24.7 GGC 43 340.25 16920 0.549 24.4 GGC 43 346.25 17411 0.556 23.9 GGC 43 360.25 18704 0.543 25.0 CDH 41 82.25 20151 0.475 31.8 CDH 41 98.25 21162 0.506 28.4 CDH 41 122.25 22403 0.488 30.2 CDH 41 146.25 23442 0.596 21.1 CDH 41 170.25 24389 0.470 32.3 CDH 41 194.25 25315 0.582 22.0 CDH 41 238.25 27054 0.611 20.1 CDH 41 270.25 28290 0.645 18.3 CDH 41 286.25 28842 0.678 16.7 VNTR01-13GC: Adjusted Core Age (yrs BP, Thermocline Depth Core ID Core Depth (cm) Depth (cm) Marine13) N ratio (m) VNT 13GC 2.25 2.25 764 0.323 65.3 VNT 13GC 8.25 8.25 4682 0.290 86.6 VNT 13GC 12.25 12.25 6664 0.309 73.0 VNT 13GC 17.25 17.25 8615 0.406 39.1 VNT 13GC 20.25 20.25 9570 0.404 39.5 VNT 13GC 23.25 23.25 10401 0.442 33.3 VNT 13GC 28.25 28.25 11589 0.363 49.5 VNT 13GC 32.25 32.25 12419 0.490 28.2 VNT 13GC 36.35 36.35 13211 0.612 21.4 VNT 13GC 44.25 44.25 14690 0.566 23.2 VNT 13GC 48.25 48.25 15458 0.628 20.9 VNT 13GC 52.25 52.25 16255 0.509 26.6 VNT 13GC 92.25 56.25 17086 0.570 23.1 VNT 13GC 96.25 60.25 17951 0.522 25.7 VNT 13GC 100.25 64.25 18843 0.462 30.9 VNT 13GC 110.25 74.25 21133 0.491 28.0 VNT 13GC 122.25 86.25 23738 0.433 34.6 VNT 13GC 124.25 88.25 24130 0.399 40.5 VNT 13GC 130.25 94.25 25197 0.384 43.9 VNT 13GC 132.25 96.25 25512 0.462 30.9 VNT 13GC 134.25 98.25 25804 0.399 40.6 280 VNT 13GC 138.25 102.25 26316 0.443 33.2 VNT 13GC 148.25 112.25 27165 0.556 23.7 VNT 13GC 150.25 114.25 27264 0.475 29.5 VNT 13GC 154.25 118.25 27403 0.533 25.0 VNT 13GC 158.25 122.25 27479 0.580 22.6 VNT 13GC 166.25 130.25 27556 0.609 21.5 VNT 13GC 170.25 134.25 27638 0.507 26.8 281