|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 MULTI SPECIES STABLE ISOTOPE ANALYSIS OF FORAMINIFERA AT SITE U1304 (GARDAR DRIFT, NORTH ATLANTIC) BEFORE, DURING, AND AFTER THE LAST INTERGLACIAL By EMILY KAY MINTH A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIV ERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009
2 2009 Emily Kay Minth
3 ACKNOWLEDGEMENTS I thank my advisor for his insight, guidance and support throughout my career at the University of Florida. I thank the members of my supervisory committee for their unique expertise and helpful participation in this project. The Integrated Ocean Drilling Program contributed greatly to the succes s of my studies and I thank all the people involved in Expedition 303, as well as the international funding partners that make IODP research possible. Many other professors, fellow students, and laboratory staff have been a source of knowledge, encouragem ent, and assistance, and the sum of their contributions has improved the quality of this work. I would like to thank my family for their generous patience and constant optimism, which continues to motivate me personally and professionally.
4 TA BLE OF CONTENTS page ACKNOWLEDGEMENTS .3 LIST OF FIGURES .6 ACADEMIC ABSTRACT ..8 CHAPTER 1 INTRODUCTION .10 2 OBJECTIVES AND QUESTIONS ...12 Termination II and Heinrich Event 11...12 Substage 5e and Transition into Substage 5d13 Diatom Blooms..15 3 BACKGROUND ...16 Regional Location, Hydrog raphy, and Sedimentation...16 4 PROXIES ...22 Stable Isotopes of Foraminifera and Isotope Stratigraphy.22 Comparison of benthic Carbon isotopes to Sortable Silt...22 Foraminifera Species .23 Cibicidoides wuellerstorfi ..23 Globigerina bulloides 24 Globorotalia inflata ...25 Neogloboquadrina pachyderma and Neogloboquadrina incompta ..25 Glo borotalia truncatulinoides ..26 Diatom Blooms..27 Ice Rafted Detritus Amount and Composition...27 Age Model.28 5 MATERIALS AND METHODS ..29 Sampling and Processing...29 Stable Isotope Pretreatment and Analysis..30 6 RESULTS ..32 Stable Isotopes: 18 O..32 Marine Isotope Stage 6 and Termination II ..32 Substage 5e and transition into Substage 5d..32 Stable Isotopes: 13 C..35 Marine Isotope Stage 6 and Termination II...35 Substage 5e and transition into Substage 5d ..35 Diatom Mats...36
5 Ice Rafted Detritus Amount and Composition...36 7 DISCUSSION 43 Marine Isotope Stage 6, Termination II, and Heinrich Event 11...43 Neogloboqua drina pachyderma (s) Isotope Record..47 Substage 5e49 Substage 5d55 Surface Productivity .55 Deep Water Fluctuations...58 8 CON CLUSIONS ...81 REFERENCE LIST ...83 BIOGRAPHICAL SKETCH .92
6 LIST OF FIGURES Figure Page 1 1 Site Map 11 3 1 Physical features of the Iceland Basin ..19 3 2 Deep water masses of the North Atlantic .20 3 3 Surface currents of the modern North Atlantic .21 6 1 Oxygen isotope results for site U1304 .38 6 2 Oxygen isotope results for Globorotalia truncatulinoides ...39 6 3 Carbon isotope results for site U1304 ...40 6 4 Location of Diatom Mats at site U1304 41 6 5 Amount and compositio n of Ice Rafted Detritus .42 7 1 Oxygen isotope results with isotope stages and major events ...61 7 2 Number of foraminifera picked by species ...62 7 3 Planktonic oxygen isotopes during MIS 6 versus depth ...63 7 4 Comparison of planktonic isotope records 64 7 5 Coarse fraction proxy for IRD mass accumulation 65 7 6 Ca/Sr elememtal ratio with Percent CaCO 3 data ..66 7 7 Oxygen isotopes of G. bulloides and G inflata 67 7 8 Oxygen isotopes of G. bulloides and G. inflata and N. incompta ...68 7 9 Oxygen isotopes of G. bulloides N. incompta and G. truncatulinoides ....69 7 10 Population distribution of G. truncatulinoides by s ize ..70 7 11 Subdivision of Substage 5e using planktonic foraminifera ...71
7 7 12 Oxygen isotopes of G. bulloides at sites NEAP18K and U1304 ..72 7 13 Oxygen isotopes of N. incompta at sites 980 and U1304 .....73 7 14 Oxygen isotopes at site U1304 during Substage 5d ..74 7 15 Diatom mat formation at convergence fronts ...75 7 16 Movement of the NSAF from MIS 6 to Substage 5e ...76 7 17 Planktonic carbon isotopes during Subs tage 5d with LDM deposition highlighted 77 7 18 Benthic carbon isotopes at sites U1304 and 980 versus age ..78 7 19 Comparison of Carbon isotopes at site U1304 and Sortable silt at NEAP 18K 79 7 20 Comparison of benthic, planktonic, an d insolation records ...80
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Sciences MULTI SPECIES STABLE ISOT OPE ANALYSIS OF FORAMINIFERA AT SITE U1304 (GARDAR DRIFT, NORTH ATLANTIC) BEFORE, DURING, AND AFTER THE LAST INTERGLACIAL By Emily Kay Minth April 2009 Chair: David A. Hodell Major: Geology This study of North Atlantic site U1304 focuses on oceanogra phic changes before, during, and after Marine Isotope Substage 5e, from ~145 to 105 ka. Site U1304 records a high sedimentation rate due to the intermittent occurrence of laminated diatom mats, which hinder benthic bioturbation and increase temporal resol ution. A unique feature of this work is the multi species planktonic foraminiferal analysis, with each species representing different seasonal and/or depth habitats. The offsets in isotopic composition between species permit an estimate of seasonal chang es with depth. Comparison of this site with others in the North Atlantic reveals a regional pattern of season specific trends. Spring season warming occurred at the onset and end of Marine Isotope Substage 5e, and summer warmth persisted throughout the i nterglacial and into the next glacial period. Comparison of the trends in chemical ( 13 C, from sites U1304 and NEAP 18K) and physical (sortable silt, from NEAP 18K) proxies show deep water trends were approximately coincident with changes in surface clima te. Planktonic 18 O indicates peak warming and/or reduced salinity was concurrent with a peak in boreal summer insolation from 128 to 124.5 ka.
9 At this time, benthic 13 C and sortable silt indicate that Iceland Scotland Overflow Water (ISOW) that was weake r and less dense compared to present day ISOW because of high temperatures and/or enhanced freshwater fluxes. This may have relevance for how Atlantic Meridional Overturning Circulation might respond to a future warmer, fresher subpolar North Atlantic as a consequence of global warming. A strengthening ISOW began to penetrate depths of 3000 m by ~124 ka and remained strong until the end of the Substage 5e at ~116 ka. ISOW then underwent a temporary shoaling, which peaked at ~113 ka.
10 CHAPTER 1 INTRODUCTION North Atlantic site U1304 is located in the Iceland Basin on the southern edge of the Gardar Drift at 53¡ 3.401'N, 33¡ 31.781'W in a water depth of 3064.5 m (Figure 1 1) [ Channell et al. 2006]. Expedition 303 scientists chose t his site, located 970 km from the SE end of Greenland, as a monitor of deep water formation and sea surface dynamics [ Channell et al. 2004]. This study of site U1304 focuses on the interval spanning Marine Isotope Stages (MIS) 6 through Substage 5d (~14 5 to 105 ka) including the onset, duration and end of the last interglacial period (MIS 5e) [ Mller and Kukla 2004] The high sedimentation rate at site U1304 is due to the intermittent occurrence of laminated diatom mats/laminated diatom oozes (LDM/LDO ) dominated by needle shaped Thalassiothrix/Lioloma diatom assemblages [ Channell et al. 2006] The high tensi le strength of the mat sediment suppresses benthic activity and hinders bioturbation permitting the study of climate variability at high temporal resolution I measured stable oxygen and carbon isotopes on 1 benthic and 6 planktonic foraminiferal species each with a known depth habitat and peak season of growth. Planktonic 18 O of foraminifera record the nature of surface inflow to the Nordic Se as via the North Atlantic Current, whereas benthic 13 C monitors the ventilation of deep water. By comparing planktonic and benthic isotope records I examine the timing and linkages between surface water inflow and deep water outflow over the Iceland Scot land Ridge during the LIG.
11 Figure 1 1. Site Map. Site U1304 is located near the center of the figure. Figure generated using Google Earth and the IODP borehole database from: http://campanian.iodp mi sapporo.org/google/data/i odp.kml U1304 NEAP18K T90 9P EW9303 17 EW9302 8JPC 980 MD95 2042: Sou th at 37 ¡ N, 10 ¡ W M23352: North at 70 ¡ N, 12 ¡ W
12 CHAPTER 2 OBJECTIVES AND QUESTIONS Termination II and Heinrich Event 11 Terminations, rapid climate shifts from glacial to interglacial conditions, preceded both the Holocene and MIS 5 (Termination I and Termination II, respectively). Lototskaya and Ganssen  suggested that TII occurred in two steps, TIIa and TIIb, separated by a pause'; an interval of stable conditions identified by unchanging benthic and planktonic isotope values. Other workers have acknowledged the TII pause throughout the North Atlantic and Nordic Seas and assigned ages ranging from 131.5 to 128 ka [ Lototskaya and Ganssen 1999; Esat et al. 1999; Gallup et al. 2002; Shackleton et al. 2003; Gouzy et al. 2004; Risebrobakken et al. 2006; Skinner and Shackl eton 2006; Van Nieuwenhove 2008]. An objective of this work was to determine the structure of TII at site U1304. What does the presence or absence of a TII pause at site U1304 indicate for ocean dynamics during deglaciations ? What is the relative phas ing between planktonic and benthic stable isotope signals ? Do the sequence of events occurring over TII show leads and lags and what environmental factors caused them? How does deep water circulation change over TII as monitored by the benthic 13 C and s ortable silt proxies? Melting ice sheets surrounding the North Atlantic basin caused a major ice rafting event (Heinrich Event 11) during Termination II [ Heinrich 1988; Lototskaya and Ganssen 1999; Oppo et al. 2006; Skinner and Shackleton 2006; Carlso n et al. 2008] Initially identified using a high ratio of lithics to total grains in the 180 !m to 3 mm fraction [ Heinrich 1988], Bond et al.  and Broecker et al.  noted the abundance of detrital carbonate in Heinrich layers, specifically s ourced from Hudson Strait region [ Hemming 2004] Site U1304 provides a record of detrital layer stratigraphy from the central Atlantic, and lies just north of the Ruddiman Ice
13 Rafted Detritus (IRD) Belt' [ Ruddiman 1977; Channell et al. 2004]. The loc ation of this site is influenced by detrital input from Canada, Greenland, the British Isles, Fennoscandia, and Iceland [ Steenfelt 2001; Moros et al. 2002]. How does the amount of coarse fraction material change throughout the section and is there a pea k in detrital carbonate on TII? What proxies can be used to narrow down the source region for detrital carbonate grains? Substage 5e and Transition into Substage 5d The climate of the last interglacial period was warmer and sea level was several meters higher than today [ Overpeck et al. 2006]. Although orbital parameters differ, Substage 5e is commonly compared to modern, anthropogenically influenced conditions, potentially providing insight into climate change in the next few centuries [ Duplessy et al 2007; Van Nieuwenhove et al. 2008]. Ice cores have often provided annual scale resolution, though it is only now that a core is being drilled in Northwestern Greenland that will likely penetrate the entire Last Interglacial section ( NEEM North Green land Eemian Ice Drilling; http://neem.ku.dk/ ). A future challenge will be to compare the NEEM core with high resolution marine sediment cores. A major hindrance to previous paleoceanographic climate studies is the level of resolution attainable by studyin g deep sea sediment cores. Even in areas of high accumulation rates, sediment mixing by bioturbation can significantly reduce the maximum temporal resolution. At site U1304, a 5 m thick section of early MIS 5 contains laminated diatom ooze dominated by a ssemblages of needle shaped species of the Thalassiothrix/Lioloma complex [ Channell et al. 2004]. Fine laminae of interbedded sediment are preserved as a result of the tight meshwork of diatom frustules, which suppressed benthic activity and hindered biot urbation thereby providing the potential to produce high resolution records of past oceanic change [ Bodn and Backman 1996]. I calculate 40 cm kyr 1 using tie points at the start and end of the MIS 5e plateau.
14 Because of the reduced depth of bioturba tion, a 5 cm sampling interval provides an average sample spacing of 125 years for the Last Interglacial period. Many studies highlight the potential for millennial scale leads and lags in the climate system in the North Atlantic region [e.g. Chapman and Shackleton 1999a; Bauch and Kandiano 2007; Van Nieuwenhove 2008], and illustrate the importance of phasing between surface and deep water records. How do the planktonic and benthic records change relative to each other during Substage 5e at site U130 4? Are there leads or lags evident in the surface and/or deep water records? Are there inter species changes evident within the planktonic records that reveal seasonal variation in the upper water column? How did deep water production vary at site U1304 r elative to other locations and depths in the North Atlantic? Several small scale cooling events' (C events: C24 C27) mark the transition into Substage 5d [ Chapman and Shackleton 1999a; Oppo et al. 2006; Bauch and Kandiano 2007]. These times of rapid i ce growth and iceberg discharge may be detected by the presence of IRD, which can be estimated by an increase of the coarse fraction in each sample [ Hemming 2004]. Though the proxy signal is often subtle, high resolution sampling permits detection of the se brief cold events, which are used to provide age constraints on the younger portion of this section. How did the penultimate interglacial end? What was the phasing of stable isotope and IRD proxies throughout the demise of Substage 5e? Diatom Blooms The discovery of diatom blooms at site U1304 indicates distinctive surface ocean circulation patterns during Substage 5e. Diatoms are a group of photosynthetic algae that form silicate tests, with many species adapted to stratified water conditions, which commonly occur in regions of surface current convergence or during times of meltwater pulses [ Sarthou et al. 2005;
15 Kemp et al., 2006]. Leg 138 of the Ocean Drilling Program (ODP) extracted cores directly beneath a modern Rhizosolenia spp. bloom near the equatorial Pacific convergence front. LDM within cores from this region led Kemp et al  to hypothesize a surficial convergence zone and the associated physical mechanics as the principal method of LDM formation. Consequently, LDM deposits worldwi de are used as a proxy to indicate paleocurrent convergence zones [ Kemp et al ., 1995] In the North Atlantic, the R/V Ewing recovered a continuous 3 m sequence of LDO at site EW9303 17, ~490 km northwest of site U1304 [ Bodn and Backman 1996]. Bodn and Ba ckman  suggested LDO formation during Substage 5e resulted from an oscillating sub Arctic convergence zone over site EW9303 17. The discovery of LDM/LDO in Substage 5e at site U1304 highlights the regional nature of these deposits. Where do the dia tom mats found at site U1304 occur within the stratigraphy of MIS 6 to Substage 5d? What does the presence of LDM indicate for changes in surface currents? How did the Northern Sub Arctic Front move over time, and how did related diatom productivity affe ct foraminiferal populations?
16 CHAPTER 3 BACKGROUND Regional Location, Hydrography, and Sedimentation Site U1304 is located near the southern extent of the partially enclosed Iceland Basin (Figure 3 1). Deep currents transporting sediment s from Iceland and NW European landmasses are responsible for the formation of the Bjrn and Gardar Drifts, elongate sedimentary bodies on the eastern flank of the Reykjanes Ridge. The Gardar Drift extends for ~1000 km from its northeastern end (61.5 ¡ N, ~ 1400 m depth) toward the southwest, ending near the Charlie Gibbs Fracture Zone (52.5 ¡ N, >3000 m depth). Site U1304 is located at the southern extent of the Gardar drift, just north of the Charlie Gibbs Fracture Zone. The Bjrn and Gardar Drifts result f rom the interaction of deep western boundary currents (DWBC) with deep sea sediments along the margins of the North Atlantic [ Ruddiman 1972; Kidd and Hill 1987; McCartney 1992; Revel et al. 1996; Bianchi and McCave 2000]. The dominant DWBC in the Ice land Basin is Iceland Scotland Overflow Water (ISOW), a precursor of North Atlantic Deep Water (NADW) [ McCartney 1992; Bianchi and McCave 2000]. ISOW forms in the Norwegian Greenland Sea through cooling of warm surface waters, such as the North Atlantic Current (NAC), analogous to the Gulf Stream. The deep and intermediate parts of the central and western Iceland Basin lie along the main pathway of ISOW [ Broecker 1991; Saunders 1996; Bianchi and McCave 2000, and references therein; Duplessy et al. 2 007]. Site U1304 is located at a depth of 3064 m and is bathed by a combination of ISOW and Lower Deep Water (LDW) (Figure 3 2). LDW is derived from the south and penetrates into the deep North Atlantic during times of reduced NADW production Between ~ 1300 and 2000 m, the volumetrically dominant water mass in the Iceland Basin is Labrador Sea Water (LSW), which is a main contributor to NADW and develops as a
17 consequence of vertical convection in the western Labrador Sea [ Bianchi and McCave, 2000; de Ver nal et al. 2002; Thierry et al. 2008, and references therein]. At about the same water depth of site U1304, site NEAP18K (52 ¡ 46.02'N, 30 ¡ 20.68'W) lies just north of the Charlie Gibbs Fracture Zone, and under the path of ISOW leaving the Iceland Basin [ Ha ll et al. 1998]. Combined flow of ISOW and LDW through the Charlie Gibbs Fracture Zone is ~6.6 sverdrups (1 Sv = 10 6 m 3 /s), a significant contribution to the NADW mixture [ Dickson and Brown 1994; Hall et al. 1998]. More dynamic than the deep water cu rrents are the surface water currents in the North Atlantic Ocean system, which are identifiable based on location and/or physical and chemical properties. Site U1304 frequently lies beneath the northern branch of the NAC (Figure 3 3). The NAC traverses NE across the western North Atlantic with portions recirculating into the North Atlantic Gyre, and bifurcates into Northern and Southern branches within 48 ¡ N to 51 ¡ N, 30 ¡ to 35 ¡ W [ Arhan 1990; Sy et al. 1992; Belkin and Levitus 1996; Schiebel et al. 20 01; Reverdin et al. 2003; Pollard et al. 2004; Thierry et al. 2008]. These branches have distinct temperature and salinity characteristics that make them identifiable, but the northern branch is typically the most persistent and is associated with the Northern Sub Arctic Front (NSAF) [ Arhan 1990; Sy et al. 1992; Pollard et al. 2004]. Site U1304 is proximal to the modern NSAF where warm, low productivity waters from the south meet cool, high productivity waters from the north [ Bodn and Backman 19 96]. Several groups have shown that algal blooms, such as the LDM/LDO seen in site U1304, tend to form at frontal convergence zone s Yoder et al.  reported a modern bloom of Rhizosolenia spp. at the equatorial convergence front in the Pacific Ocean Correlative LDM in Neogene age sediments in the region were used to hypothesize frontal zones as a source for massive diatom
18 blooms [ Kemp and Bauldauf 1993; Kemp et al. 1995; King et al. 1998]. Diatoms also occur at other times in sediments of the N orth Pacific [ Dickens and Barron 1997 ], Southern Ocean [ Kemp et al. 2006], and specifically, in the North Atlantic subarctic during the last interglacial period [ Bodn and Backman 1996]. Fine laminae are preserved within LDM/LDO, bulk sedimentation rat es are usually exceptionally high and bioturbation is reduced [ King et al., 1998]. As hypothesized by Bodn and Backman , the LDM/LDO preserved at site U1304 is likely related to the location of the NSAF throughout MIS 6 to Substage 5d. Site U1304 s ediments also contain ice rafted detritus. Terrigenous sediment becomes entrained in glaciers through basal erosion, and when land based glaciers meet the ocean they calve into icebergs, carrying the sediments seaward [ Ruddiman 1977; Andrews and Tedesco 1992]. However, terrigenous sedimentation can also be wind blown [e.g. Holz et al. 2007], but this input is likely very small at site U1304. The principle source of volcanic input is from eruptions on Iceland [ Hskuldsson et al. 2007]. Abundant plankt onic foraminifera tests are well preserved at site U1304, which settle to the bottom in pulses as a response to changing surface conditions or reproduction [ Schiebel et al. 1995; Schiebel et al. 1997]. As passive inhabitants of their environment, plankt onic foraminifera are susceptible to changes in hydrodynamics and their deposition is dependent upon regional currents and local turbulence [ Schiebel et al ., 1997]. Despite these fluxes, the isotopic composition of their calcite shells remains a reliable proxy. Many species of planktonic foraminifera are found at site U1304, and chemical analysis of their shells provides the foundation for this study.
19 Figure 3 1. Physical features of the Iceland Basin. (W to E): Charlie Gibbs Fracture Zone, Reykjanes Ridge, Gardar Drift, Bjrn Drift, Maury channel, and Rockall Plateau. Background image modified from Google Earth .
20 Figure 3 2. Deep water masses of the North Atlantic. Yellow arrows are Labrador Sea Water (~1300 to 2000 m, LSW) [ Bianchi and McCave, 2000; de Vernal et al. 2002; Thierry et al. 2008] Blue arrows indicate Iceland Scotland Overflow Water (~1800 to 4000 m, ISOW) [ Broecker 1991; Saunders 1996; Bianchi and McCave 2000, and references therein; Duplessy et al. 2007] which bathes site U1304, marked by a red circle. Green arrows indicate Lower Deep Water (below ~4000 m, LDW) [ Broecker 1991; Saunders 1996; Bianchi and McCave 2000, and references therein; Duplessy et al. 2007] Background image modif ied from Google Earth .
21 Figure 3 3. Surface currents of the modern North Atlantic. Orange arrows mark the North Atlantic Current (NAC) and its branches, including the offshoots that are re circulated into the North Atlantic Gyre [ A rhan 1990; Sy et al. 1992; Belkin and Levitus 1996; Schiebel et al. 2001; Reverdin et al. 2003; Pollard et al. 2004; Thierry et al. 2008] Each branch of the NAC eventually turns anticlockwise to feed into the Irminger Current (yellow arrows). Pol ar surface currents including the East Greenland Current, West Greenland Current, and the Labrador Current are all indicated by blue arrows. Background image modified from Google Earth .
22 CHAPTER 4 PROXIES Stable Isotopes of Foraminifera an d Isotope Stratigraphy A principle objective of this work is to obtain, analyze, and compare high resolution isotope stratigraphies for Substage 5e to determine the substructure of the water column above site U1304. Isotopic ratios were measured on one s pecies of benthic foraminifer, Cibicidoides wuellerstofi and 4 species of planktonic foraminifera: Globigerina bulloides, Globorotalia inflata, Globorotalia truncatulinoides (both dextral and sinistral coiling varieties), and Neogloboquadrina pachyderma ( both dextral and sinistral coiling varieties). Darling et al.  showed the coiling direction of N. pachyderma to be genetic and not a morphological feature reflecting ecophenotypic variation, indicating these are separate species. The right coiling v ariety of N. pachyderma is hereby referred to as N. incompta, widely recognized as equivalent nomenclature [ Darling et al. 2006]. Each planktonic species was chosen due to depth habitat and seasonal growth preferences, each of which may affect the calcit e isotopic composition. Isotope stratigraphy is a fundamental practice in paleoclimatology, and is based on fundamental principles. The two most common isotopes of oxygen found in water are 16 O (abundance ~98%) and 18 O (abundance ~2%) [ Sharp 2006]. The ratio of these two isotopes is governed by thermodynamic properties, and small variations in the isotopic ratio occur during physical and chemical processes as a consequence of both kinetic and equilibrium isotope effects [ Sharp 2006]. When foraminifera build their shells of CaCO 3 the 18 O/ 16 O ratio of shell calcite is dependent upon the 18 O/ 16 O ratio of the water and the temperature of shell secretion. When Earth's climate is in a glacial phase, the 18 O/ 16 O ratio of shell carbonate precipitated from sea water is greater due both to lowered temperature and increased continental ice volume. In contrast, warmer temperatures or reduced continental ice volume result in lower 18 O/ 16 O ratios.
23 Foraminifera also incorporate carbon isotopes when building their s hells. The ratio of the two dominant stable isotopes, 12 C and 13 C, varies with changes in primary productivity and atmosphere ocean surface exchange. Thus, water that is nutrient depleted due to enhanced primary productivity has a higher 13 C/ 12 C ratio th an nutrient enriched water. NADW, originating in a nutrient poor region, tends to have high 13 C/ 12 C compared to AABW derived from the nutrient rich Southern Ocean [ Curry et al. 1988; Bigg et al. 2000]. Comparison of benthic Carbon stable isotopes to Sor table Silt Throughout Substage 5e, changes in the bottom water record are of interest, particularly the strength of ISOW production. Here I compare the 13 C record of site U1304 to site NEAP 18K, ~200 m away. At site NEAP18K, Hall et al.  estimated relative current speed via the sortable silt proxy, a distribution of grain size in the 10 to 63 m fraction, where stronger bottom currents yield a coarser mean grain size [ McCave et al. 1995; Hall et al. 1998]. Unlike nutrient proxies such as benthic 13 C, sortable silt is a physical proxy that is unaffected by biological processes Foraminifera Species The planktonic data presented here reflects values that are not corrected for vital effects, because I compare the relative interspecies changes, and n ot absolute changes. C. wuellerstorfi 18 O values have been corrected using the +0.64 correction factor estimated by Shackleton and Opdyke , which is widely used and accepted for the North Atlantic. Given that foraminifera build their shells as the y grow, the whole shell chemistry is the weighted average of chambers added at different growth stages [ Lohmann 1995]. Unfortunately, few foraminiferal species secrete calcite tests in equilibrium with surrounding waters [ Sharp 2006]. The difference bet ween measured and calculated isotope values is called the "vital effect", and
24 represents each species' deviation from equilibrium calcification, which for well studied species is a known offset [ Hemleben et al. 1989; Volkman and Mensch 2001]. Though dis solution may also affect isotope values, there is little evidence for it throughout this section, as the last interglacial was a time favorable to carbonate preservation [ Balsam 1983]. Cibicidoides wuellerstofi This epibenthic species has been shown to m ost accurately record bottom water carbon chemistry compared with other bottom dwelling foraminifera [ Curry et al., 1988; Weinelt et al. 2001]. Because of its epibenthic habitat, pore waters have little influence on the chemistry of C. wuellerstofi tests though Mackensen et al.  proposed that times of high organic matter flux can cause 13 C values to be anomalously low due to the presence of a phytodetritus layer. Similarities between the site U1304 benthic 13 C record and a physical proxy record of sortable silt at site NEAP18K favors the interpretation the Site U1304 has recorded changes in deep water ventilation [ Hodell et al. submitted, 2008 ]. Here, the benthic 18 O record is used to estimate long term glacioeustatic change such as Marine Iso tope Stage transitions, and 13 C is used to examine local and regional changes in deep water production [ Skinner and Shackleton, 2006]. Comparisons are made with the planktonic species described below and with other sites containing C. wuellerstorfi data. This species is reliable for carbon isotopes due to a constant offset between test calcite and equilibrium calcite for bottom water dissolved inorganic carbon (DIC), so no adjustment is necessary [ McCorkle and Keigwin 1994]. Globigerina bulloides Glo bigerina bulloides a surface dwelling planktonic foraminifer, typically calcifies in the uppermost portion of the oceanic mixed layer [ Ravelo and Fairbanks 1992]. Growth and reproduction takes place in the upper 60 m [ Hemleben et al. 1989; Schiebel et al ., 1997], and the
25 habitat of G. bulloides shows strong coupling with phytoplankton, it's dominant food source [ Mortyn and Charles 2003]. The flux median from sediment trap studies by Deuser and Ross  suggests peak reproduction occurs in early Mar ch; the period most likely represented by a set of individuals of G. bulloides The 250 to 300!m size fraction was used in this study, as it results in the most reliable isotope calculations for G. bulloides [ Spero and Lea 1996] Globorotalia inflata Th is species lives in subpolar to transitional waters (~7 ¡ to 25 ¡ C), and typically calcifies slightly deeper than G. bulloides [ Hemleben et al. 1989] Sedimentary accumulation of G. inflata shells indicates peak reproduction occurs in late March [ Deuser an d Ross 1989]. Measured 18 O values indicate shell growth occurs at different depths throughout the life cycle, but is completed in the upper 150 to 200 m [ Ravelo and Fairbanks 1992; Wilke et al. 2006]. Neogloboquadrina pachyderma (s) and N. incompta Neogloboquadrina pachyder ma (sinistral coiling) is a polar to sub polar planktonic foraminiferal species, and can tolerate salinities up to 82 [ Hemleben et al ., 1989]. N. pachyderma (s) peak abundance is primarily found between 0.5 and 8.5 ¡ C in the North Atlantic Ocean [ Tolderl und and B 1972]. If surface water temperatures are above the tolerance limit for this species, it is capable of vertically migrating several hundred meters deeper in the water column during its life cycle [ Wu and Hillaire Marcel 1994]. N. pachyderma ( s) is commonly found at water depths from ~100 to 250 m in the polar to subpolar environment [ Rashid and Boyle 2007]. In the subarctic region of the North Atlantic Ocean, tests are mainly produced during summer months, however N. pachyderma (s) may chang e its growth season to secrete its shell in temperatures < 8 ¡ C [ Tolderlund and B, 1972; Wu and Hillaire Marcel 1994].
26 N. incompta is present almost throughout the entire oceanic sea surface temperature (SST) range, yet exhibits a clear preference for in termediate temperatures (i.e. above 8 ¡ C) [ Fraile et al. 2008]. N. incompta lives within the upper 100 m of the mixed layer, and peak growth occurs during the summer months of July to September [ Johannesen and Jansen 1992; Ravelo and Fairbanks 1992; Gan ssen and Kroon 2000; Salgueiro et al. 2008] Globorotalia truncatulinoides Globorotalia truncatulinoides (left and right coiling varieties) is a transitional to subtropical planktonic foraminifer with an annual life cycle and large vertical migration during ontogeny. Juvenile G. truncatulinoides ( 125 to 177 !m) begin their life cycle above 120 to 180 m during winter [ Lohmann and Schweitzer 1990]. During growth they sink to deeper waters, adding primary shell chambers, increasing the mass and size o f their shells. Upon reaching ~425 to 500 !m in size, G. truncatulinoides reproduces, and a population of shells most likely represents January, the peak month of reproduction [ Deuser and Ross 1989; Lohmann and Schweitzer 1990]. Work by LeGrande et al.  indicated most G. truncatulinoides shells greater than 355 !m have a significant 18 O enriched secondary calcite crust, resulting from secondary calcification at depth [ B 1977; McKenna and Prell 2004]. Isotope analyses for site U1304 were done on individuals >351 !m in an effort to capture shells with a high amount of secondary calcite [ Lohmann 1995]. I also measured left and right coiling specimens separately to verify the isotopic range of each morphotype. The mean 18 O of G. truncatulinoi des shells recovered from plankton tows corresponded to calculated 18 O equilibrium for ~200 m depth, suggesting the average depth of growth for a shell [ Deuser and Ross 1989; Lohmann 1995; McKenna and Prell 2004 ].
27 Diatom Blooms Thal a s siothrix longiss ima dominates the diatom population in the LDM at site U1304 creating easily identifiable layers that resemble tissue paper [ Bodn and Backman 1996; Channell et al. 2004] The presence of undisturbed laminae indicates minimal bioturbation, enhancing th e resolution of site U1304. Determi ning the stratigraphic location of the LDM/LDO relative to planktonic isotopic events aids in understanding the nutrient conditions of the upper water column and the location of the NSAF. The biogenic silica component i s estimated through comparison of elemental XRF data. Here I compare Si to Zr, as Si has both a biogenic and detrital source and Zr is only derived from detrital sources (i.e. minerals), effectively normalizing total silica by detrital silica to obtain th e biogenic Si component. The magnetic susceptibility measurements are also compared to the XRF record; samples with thick LDM/LDO will dilute magnetic minerals in the sediment, lowering the overall magnetic susceptibility. LDM also increase the color inde x, which is used as another method for diatom mat identification. Areas where these proxies overlap are interpreted to be times of high diatom mat deposition. Ice Rafted Detritus Amount and Composition Generally, the presence of detrital carbonate in IRD peaks suggests a Hudson Strait source for certain Heinrich events [ Bond et al. 1992; Broecker et al ., 1992; Andrews 1998; Moros et al. 2002; Hemming 2004 ]. Here, I use the elemental ratio Ca/Sr measured via XRF to estimate changes in detrital carbona te input throughout the section [ Hodell et al. 2008]. I also measured bulk sediment for percent carbonate to detect whether the IRD event at TII shows a rapid increase in carbonate, suggesting a large detrital input. Hodell and Curtis  measured ind ividual detrital carbonate grains within Heinrich events, and their results showed detrital carbonate has 18 O values that are 6 to 7 lower than biogenic carbonate. IRD events with high
28 detrital carbonate should have a very low 18 O signal compared to th e biogenic signal, so bulk carbonate 18 O provides an estimate for the overall change across TII. During the Substage 5e to 5d transition there were several episodes of iceberg discharge, recorded by an increase in IRD [ Chapman and Shackleton 1999a; Mller and Kukla 2004; Oppo et al. 200 6; Bauch and Kandiano 2007]. Specifically, C24 (107 ka) is believed to represent a southward shift of the northern branch of the NAC, triggering European climate change and marking the onset of the subsequent glacial period [ Mller and Kukla 2004]. Bec ause IRD events are recorded in the >150 !m fraction, I measured the mass of the coarse fraction for each sample to approximate the relative changes in IRD [ Bond et al. 1992; Revel et al. 1996]. These data are used to identify periods of iceberg dischar ge to site U1304, and serve to locate C events during the decline of Substage 5e into Substage 5d. Age Model The age model for this study is based on the radiometric scale developed by Shackleton et al. , who estimated an age of 128 ka for the start of the MIS 5e plateau and an age of 116 ka for the end of the plateau. To constrain age in the top of the section I use the location of cooling event 24, a peak in abundance of coarse fraction material at 107 ka at MIS 5d [Oppo et al., 2007].
29 CHAPTER 5 MATERIALS AND METHODS Sampling and Processing This study used samples from a spliced composite core section at site U1304 from 24.52 to 15.47 mcd. This corresponds to core interval from 1304B 3H 4 145 to 1304A 2H 1 95, with a single splice at 2 2.77 mcd. Four samples on either side of the splice were included in analyses, to ensure the splice was placed properly. At the Bremen Core Repository (Bremen, Germany), sediment cores belonging to the spliced composite section were sampled at 5 cm inter vals, using (two) 10 cc plastic scoops. A knife was often needed to sample diatom mats because of the high tensile strength of the sediment. Samples were placed into plastic bags, sealed, and shipped to the University of Florida in Gainesville, Florida, U.S.A. The sediment was removed from sample bags and placed into 250 mL plastic beakers. The beakers were placed into a low temperature oven (~50 o C) and dried for a minimum of 24 hours. With the exception of a small volume of sediment saved in the sampl e bag as an archive and a portion retained in a snap cap vial for bulk carbonate analyses, each sample was scraped into a metal tray and weighed. After the samples were returned to their respective beakers, the beakers were filled about halfway with deion ized (DI) water to disaggregate the sediment. Sediments were soaked overnight then washed with DI water through a set of sieves comprised of a 150 !m sieve on top and a 63 !m sieve below. Only the sand sized fraction (>63 !m) was retained for analysis. Each size fraction was captured on filter paper inserted into a plastic funnel and rinsed with DI water. The filters were placed inside beakers and dried in a ~50 o C oven for a minimum of twenty four hours. Once dry, the sediment was poured from the fil ter paper into a metal tray, and the mass was recorded for each fraction of the sample. The sediment was then transferred into 7.4 mL
30 glass vials, which were labeled, capped, and placed into a box in stratigraphic order. The >150 !m fraction was dry siev ed through another set of sieves; 425 !m on the top to remove any diatom cotton, followed by 300 !m, 250 !m, and 212 !m sieve on the bottom. First, the 250 to 300 !m fraction was spread onto a tray and examined using a binocular microscope to identify the species of planktonic foraminifera G. bulloides as this size fraction is most reliable for isotope studies [ Spero and Lea 1996] A maximum of 10 individuals of G. bulloides were picked using a water dampened paintbrush and placed into a 1.8 mL glass vi al. The remaining size fractions were scattered on the tray to pick up to 10 individuals of G. inflata 20 individuals each of N. pachyderma (s) and N. incompta 5 individuals of C. wuellerstorfi and all of the G. truncatulinoides (s and d) present in ea ch sample. Each species and subspecies were kept in separate 1.8 mL vials, and labeled accordingly. G. truncatulinoides were also sorted into the following size fractions: <180 !m, 180 to 212 !m, 212 to 351 !m, 351 to 425 !m, and >425 !m. Stable Isotope Pretreatment, Preparation, and Analysis To clean the foraminiferal tests and remove organic matter, each vial containing foraminifera was filled with 15% H 2 O 2 sonicated, and dried in methanol. For G. inflata the shells were crushed prior to sonication to ensure all debris was removed from the interior of the shell. After cleaning, the 351 to 425 !m and >425 !m size fractions of G. truncatulinoides were separated into (s) and (d) coiling directions, and some shells that were missing the bottom or last c hambers were determined to be unfit for isotopic analysis. G. truncatulinoides tests were also cracked to maximize surface area and ensure full reaction in the mass spectrometer. Foraminiferal oxygen and carbon isotopic ratios were measured for each s ample using a Finnigan MAT 252 isotope ratio mass spectrometer coupled with a Kiel III carbonate preparation device. Samples were reacted in 100% orthophosphoric acid at 70¡C and evolved
31 CO 2 gas was measured with the mass spectrometer. Isotopic results a re reported in standard delta notation 1 relative to Vienna Pee Dee Belemnite ( VPDB) using National Bureau of Standards #19 (NBS 19) for calibration. This is an international standard with an accepted 18 O value of 2.20 and 13 C of 1.94. The measured e rror associated with 18 O values is 0.07 and 13 C is 0.04 indicating the reported values are statistically sound. To determine the amount of total inorganic carbon in each sample, I used the UIC (Coulometrics) 5011 CO 2 coulometer coupled with an Auto Mate carbonate prep system (AutoMateFX.com). Each sample was homogenized to a powder, then ~15 mg was weighed and placed into a septum top tube, and loaded into the AutoMate carousel. Using a CO 2 free nitrogen carrier gas, a double needle assembly purged the vial of atmospheric gas. Acid was injected into each vial and evolved CO 2 was carried through a silver nitrate scrubber to the coulometer where total C was measured. Oxygen and carbon isotopes on bulk carbonate were measured using a VG/Micromass (n ow GV Instruments) PRISM Series II isotope ratio mass spectrometer. Powdered samples were loaded into stainless steel boats and loaded into a 44 position Isocarb prep system. Each sample was reacted in a common orthophosphoric acid bath at 90¡C, and wa ter was cryogenically removed in a methanol slush. Isotope values were measured by a PRISM mass spectrometer and all isotope values are reported in standard delta notation relative to VPDB. 1 Standard delta notation formula from Sharp . spl' denotes sample rat io and std' denotes NBS standard ratio. 18 O= ( 18 O/ 16 O) spl ( 18 O/ 16 O) std x 10 3 ( 18 O/ 16 O) std
32 CHAPTER 6 RESULTS Stable Isotopes: 18 O MIS 6 and Termination II In the benthic 18 O record, MIS 6 extends from the base of the section studied through ~22.87 mcd At the bottom of the section, values begin to increase through the 18 O maximum of 5.48 at 23.02 mcd (Figure 6 1). During M IS 6, planktonic G. bulloides and G. inflata show similar 18 O values of ~2.6 from 24.52 to 23.42 mcd. After this, G. inflata data are scarce and G. bulloides records a rapid increase to a maximum 18 O of 3.9, approximately coincident with the maximum r ecorded in the benthic record. N. pachyderma (s) displays a similar pattern to C. wuellerstorfi with nearly identical 18 O values throughout much of MIS 6. N. incompta records the most millennial scale variability through MIS 6, with changes in 18 O as great as 2 occurring within 5 cm (i.e. 23.92 to 23.87 mcd). G. truncatulinoides are also scarce during MIS 6, but a generally increasing upsection trend is distinguishable from the 8 individuals measured. At TII, all species show a rapid decrease in 18 O values, though each is different in magnitude. Benthic values decrease by ~2.3 from maximum MIS 6 values. G. bulloides increases ~3.4 across the termination, though minimum 18 O is not reached until further into Substage 5e. There is a distinct dec rease in 18 O at TII of ~1.5 for G. inflata but a lack of specimens during the coldest periods means these conditions may not be represented. For all of the species measured, N. incompta records the greatest fluctuation of ~3.6 across TII. G. truncatu linoides is absent during the peak of glacial conditions, but returns during Substage 5e. Substage 5e and transition into Substage 5d Cibicidoides wuellerstorfi 18 O values define Substage 5e as occurring from 22.47 to 17.77 mcd. Benthic values are nearl y unchanging during this plateau, with a mean of 3.33
33 ( n =88) and a standard deviation of 0.08 (1 # ) Planktonic species record surface variability, and the N. pachyderma (s) record shows a particularly unique pattern. Globogerina bulloides record a dou ble minima from 22.47 to 21.92 mcd (Figure 6 1). After this, G. bulloides values increase to reach a plateau with a mean of 1.6 (n=63) and a standard deviation of 0.22 (1 # ). At 18.42 mcd, near the end of Substage 5e, G. bulloides values decrease for ~7 0 cm to reach 0.8 Coinciding with the end of Substage 5e in the benthic record, G. bulloides 18 O values transition to Substage 5d. Globorotalia inflata show a similar decrease at the onset of Substage 5e, and a minimum is recorded with 18 O values of ~ 1.6 from 22.47 to 21.12 mcd. The G. inflata minimum lasts slightly longer than the G. bull oides plateau, and after 21.07 mcd, G. inflata values increase and reach a broad plateau through 18.52 mcd. Globorotalia inflata also shows a decrease at the end of Substage 5e similar, though shorter in duration, to G. bulloides G. inflata then begins t he transition to Substage 5d at approximately the same time as the benthic species. At the onset of Substage 5e, N. incompta remains low, slowly increasing through ~18.97 mcd. Neogloboqudrina incompta finally reaches a plateau of ~1.4 near the end of the G. bulloides and G. inflata plateaus. Stability continues beyond the benthic transition to Substage 5d, and N. incompta begins its increase to Substage 5d, ~30 cm later than the benthic transition. Globorotalia truncatulinoides' 18 O record is an av erage of multiple individuals for most samples (Figure 6 2). Up to 15 individuals were measured for each sample, and the sample average was taken when two or more were present, as suggested by Lohmann . Also, several samples only had one individual >351 !m, which skew the average line at their respective depths. Individual measurements show a wide range of scatter, likely due to the variation of secondary calcite, but left and right coiling specimens show no distinct isotopic
34 trend within single s amples. During Substage 5e, G. truncatulinoides begins with a plateau from 22.42 to 18.82 mcd, and the mean of the average 18 O values is 1.68 (n=53), with a standard deviation of 0.38 (1 # ). 18 O values begin to increase after this plateau, signaling the decline into Substage 5d, though this transition occurs much earlier for G. truncatulinoides than it does for any other species, at ~18.77 mcd. The most unusual pattern of these data is the N. pachyderma (s) record (Figure 6 1). Compared with all ot her foraminiferal species, this 18 O record shows the highest degree of variability, and 18 O values are much greater than expected for this interval. During the beginning of Substage 5e, values oscillate up to 2 within 5 cm. In the middle of Substage 5 e, from 21.27 to 18.72 mcd, there is a moderate decreasing trend from ~4.9 to ~4.1, with perturbations of up to 1. The latter part of Substage 5e is the most variable, with a general sawtooth' pattern, and adjacent measurements offset by up to 2 (i.e. 18.42 to 18.37 mcd). Benthic 18 O shows a generally steady increase into Substage 5d, beginning at 17.82 mcd and continuing through the top of the section. Planktonic 18 O data has a similar overall trend. G. bulloides and G. inflata follow a similar s lope of increasing 18 O values, and N. incompta begins to increase later and with a steeper slope than other planktonic species. The G. truncatulinoides values also increase at this time, though a slope is difficult to determine due to an absence of speci mens for ~1 m during Substage 5d. Variability of N. pachyderma (s) continues through about 16.57 mcd, after which time the structure of the 18 O record looks similar to other planktonic species. Within the records of G. bulloides G. inflata and N. inco mpta 18 O increases in the upper 50 cm of the 18 O record. This increase is not measured in the benthic record, so it appears to be a solely planktonic signal.
35 Stable Isotopes: 13 C MIS 6 and Termination II Benthic carbon isotopes for the studied sectio n range from 0.42 to 1.15, and overall, 13 C increases from the bottom up (Figure 6 3). Isotope values for C. wuellerstorfi G. inflata N. pachyderma (s), and N. incompta decrease from the bottom of the record through ~23.27 mcd. From here, 13 C value s for each of these species begin to increase. Cibicidoides wuellerstorfi 13 C increases briefly to the onset of TII, and decreases into Substage 5e. Globigerina bulloides 13 C increases from the bottom of the section into Substage 5e. Globorotalia infl ata and N. incompta continue to increase through TII into Substage 5e. Globorotalia truncatulinoides are too scarce during MIS 6 to determine a trend. Unlike the 18 O records, there are no large changes in 13 C observed across TII in any planktonic or ben thic species. Substage 5e and transition into Substage 5d Benthic C. wuellerstorfi 13 C values peak during the middle of Substage 5e, with all planktonic species record an increasing trend throughout Substage 5e. Globogerina bulloides, G. inflata, and G. truncatulinoides have the steepest rates of increase, whereas both N. pachyderma and N. incompta show only a moderate increase in 13 C through Substage 5e. Planktonic species peak near the end of Substage 5e and decrease towards the onset of Substage 5d (Figure 6 3). The onset of Substage 5d marks a low in benthic 13 C, which increases through the top of the section. After an initial decrease during the onset of Substage 5d, 13 C values of G. bulloides level out towards the top of the section. Neoglobo quadrina incompta also decrease initially into Substage 5d, but N. pachyderma (s) remain nearly steady throughout Substage 5d. Both G. inflata and G. truncatulinoides reach a peak just before Substage 5d and maintain high 13 C values upsection.
36 Diatom Mat s Shipboard measurements of color lightness reveal intermediate lightness values for MIS 6, declining into TII where they reach a minimum for the section (Figure 6 4). During Substage 5e, color lightness increases, reaching a broad maximum in the latter half of Substage 5e. Magnetic susceptibility measurements also record an interval of low susceptibility during the second half of Substage 5e, concurrent with the peak in color lightness. Scanning XRF data show multiple peaks in Si/Zr between ~20.1 and 1 6.5 mcd. Maximum peaks in Si/Zr occur within the same zone as the magnetic susceptibility minimum and color lightness maximum. The gap in XRF data between ~19.0 and 19.5 mcd represents a sample slab removed from the archive half of the core, and consists of a thick LDM section. Visual identification of pennate diatoms in samples indicates the second half of Substage 5e is represented by near continuous LDM/LDO deposition. IRD Amount and Composition During MIS 6, the sediment coarse fraction (i.e., we ight percent > 150 microns) reveals a broad region with a high foraminifera content (Figure 6 5). This proxy also shows two distinct peaks that are a result of IRD deposition within the section; one occurs at ~23.02 mcd at TII, and the other from 16.02 to 15.72 mcd during Substage 5d. A peak at 22.37 mcd is attributed to a single large IRD grain. The coarse fraction reaches 18.8% of the total mass in the older IRD peak, versus the younger IRD peak, which has a coarse fraction content of ~6%. The IRD rec ord within Substage 5e is fairly uniform and low, with coarse fractions typically representing between 1 and 3% of the total mass of each sample. Bulk carbonate measurements were done on a subset of the section, focused on TII from 23.37 to 21.92 mcd (F igure 6 5). The overlapped core splice at 22.77 mcd allowed dual
37 measurements of percent carbonate and isotope values to be made on either side of the splice tie point. Percent carbonate results show a steady decrease from the end of MIS 6 into TII, with the lowest percent CaCO 3 occurring at ~22.5 mcd, the maximum of IRD deposition. After TII, the % CaCO 3 rapidly increases into Substage 5e. Bulk oxygen isotope results for this subsection show values that are overall lower than the benthic isotope reco rd (Figure 6 5). Additionally, there is a ~5 decrease in 18 O values between 22.87 and 22.57 mcd, approximately coincident with the decrease in the benthic 18 O record at TII. Bulk 18 O values reach a minimum of 2.80 at 22.57 mcd then quickly retur n to nearly 1 in the beginning of Substage 5.
38 Figure 6 1. Oxygen isotope results for site U1304. Foraminifera species are as follows: C. wuellerstorfi green triangles; G. bulloides blue diamonds; G. inflata pink crosses; N. incompata y ellow circles; N. pachyderma (s), red squares; G. truncatulinoides, gray circles. Data for individual G. truncatulinoides based on coiling direction can be found in figure 6 2.
39 Figure 6 2. Oxygen isotope results for Globorotalia trun catulinoides The line with gray filled circles represents the average of up to 15 individuals for each sample. Right coiling individuals are marked by open circles, and left coiling individuals are marked by black filled circles.
40 F igure 6 3. Carbon isotope results for site U1304. Foraminifera species are as follows: C. wuellerstorfi green triangles, G. bulloides open diamonds; G. inflata pink crosses; N. incompata yellow circles; N. pachyderma (s), red squares; G. truncatul inoides gray open circles.
41 Figure 6 4. Diatom Mats at site U1304. A) Color lightness, where higher values indicate lighter colored sediment. B) Magnetic Susceptibility vs. mcd. C) Scanning XRF data of Si:Zr. Dark gray bar marks where all three datasets indicate persistent LDM/LDO deposition, generally in the latter half of Substage 5e. Light gray bars represent periods of intermittent or thi n LDM/LDO deposition. A Color Lightness (% light) Magnetic Susceptibility B C Persistent, Thick LDM/LDO Intermittent LDM/LDO 63 43 53 160 40 80 120 200 600 Si:Zr Ratio 1000 15 16 17 18 19 20 21 22 23 24 25 Meters Composite Depth (mcd)
42 Figure 6 5. Amount and composition of Ice Rafted Detritus. A) Percent of sample by mass > 150 !m, B) Percent Bulk CaCO 3 across TII, C) Oxygen isotope results for the bulk fine fraction during TII (blue line with squares), with benthic 18 O (dark green line with triangles) for comparison. 0 5 10 15 20 25 15 16 17 18 19 20 21 22 23 24 25 mcd wei ght % >15 0 m C24: 107 ka H11 High Biogenic Input % CaCO 3 18 O (per mil) A Percent of sample (by mass) greater than 150 !m B C. Single Erratic 21.6 22 22.4 22.8 23.2 23.6 Meters Composite Depth (mcd) 2 0 2 4 40 30 20 10 0
43 CHAPTER 7 DISCUSSION Marine Isotope Stage 6, Termination II, an d Heinrich Event 11 The benthic oxygen isotope signal at site U1304 was correlated to the marine oxygen isotope record according to the criteria outlined by Shackleton et al. . At the base of the section, benthic 18 O values >4.5 correspond to MIS 6 (Figure 7 1). Most planktonic species have low abundances during this time due to the much cooler SST during MIS 6, and it was often difficult to find enough individuals for isotope analysis. Figure 7 2 shows the number of foraminifera picked for each sample by species and although it does not reflect a true census, the data do indicate when a species was present or nearly absent. Neogloboquadrina incompta 18 O varies significantly throughout much of MIS 6, perhaps reflecting Dansgaard Oeschger type events that were common during the last glaciation (Figure 7 3, Figure 7 4). The 18 O of C. wuellerstorfi G. bulloides G. inflata and N. incompta all increase toward maximum values at ~23.5 mcd (~140 ka), indicating the greatest ice volume and/or cold est glacial conditions. Changes in 18 O of planktonic species slightly lead those of C. wuellerstorfi suggesting surface cooling occurred before deep water cooling. North Hemisphere ice sheets reached their maximum thickness by the end of MIS 6, which m ay have led to basal melting (i.e., by exceeding the pressure melting point) and instability [ Clark et al. 1999; Bintanja and van de Wal 2008, and references there in]. Once the ice sheet reached a critical size, the next insolation maximum triggered de glaciation after an estimated lag of ~5 kyrs, leading to the termination [ Alley and Clark 1999; Bintanja and van de Wal 2008]. This large change in global ice volume is evident in all of the foraminiferal 18 O records, except N. pachyderma (s). The tran sition from MIS 6 to Substage 5e is placed at 22.6 mcd, corresponding to the midpoint between the maximum benthic 18 O values (+4.84), and minimum 18 O values
44 marking the start of the 5e oxygen isotope plateau (+2.58). The midpoint of TII is assigned an age of 130 ka [ Lisiecki and Raymo 2005]. Oxygen isotope signals of all species show a rapid, near synchronous decrease from glacial MIS 6 to Substage 5e. A brief reversal of ~0.1 was measured for both C. wuellerstorfi and G. bulloides within the begin ning of TII. Increases in 18 O are not o f the same magnitude as the "pauses" reported at other sites in the North Atlantic and Nordic Seas [ Chapman and Shackleton 1999a; Lototskaya and Ganssen ,1999; Snchez Goi et al. 1999; Bauch et al. 2000; Oppo et al. 2001; Shackleton et al. 2003; Gouzy et al. 2004; Skinner and Shackleton 2006]. Approximately 1250 km east of site U1304, the TII pause at ODP site 980 was associated with a ~1.0 increase in planktonic 18 O and a coeval decrease in benthic 13 C [ O ppo et al. 2001]. Neither of these features is observed in the isotopic records at site U1304. Skinner and Shackleton  reported synchronous decreases in planktonic and benthic 18 O occurring prior to the TII ice rafting event from Iberian Margin core MD01 2444, further illustrating the non monotonic character of the penultimate deglaciation. One objective of this work was to determine if the TII pause, seen in other North Atla ntic cores prior to H11, was recorded at site U1304. The absence of the TII pause and the rapid decline of 18 O at site U1304 may be due to a hiatus or condensed section When IRD was the principal sediment source, associated meltwater and reduction of MO C may have slowed sediment delivery from other sources to the Gardar Drift. Foraminifera content is also greatly reduced during TII, which could be a response to unfavorable surface conditions associated with the delivery of meltwater during H11. Alter natively, the TII pause may be regional in extent and not prevalent at site U1304. The formation of intermediate waters in the Nordic Seas during the deglaciation may have
45 caused the event [ Lohmann 1998; Bauch et al. 2000; Oppo et al. 2001]. Meltwater produced during the deglaciation would have lowered the salinity of uppermost surface waters. The lack of perennial ice cover permitted the denser southern sourced NAC to penetrate into the Norwegian/Greenland Sea underneath the meltwater lid [ Bauch et a l. 2000; Oppo et al. 2001]. Despite stratification, numerical model results suggested deep water formation north of the Greenland Scotland Sill was still possible [ Lohmann 1998]. Outflow over the sills entrained ambient NAC water, mixing and sinking t o intermediate depths, and likely influenced sites above ~2000 m [ Lohmann 1998; Oppo et al. 2001]. Intermediate water carried a low 13 C signature during a time of increasing temperatures and 18 O values [ Oppo et al. 2001]. Because Site U1304 is locat ed in a water depth of 3064 m, the absence of the TII pause may support the hypothesis that the pause was restricted to intermediate water depths, and affected deeper sites to a lesser extent [ Oppo et al. 2001]. Occurring before the TII pause and the deg laciation is Heinrich Event 11, delivering IRD to the North Atlantic from ~136 to 130 ka [ Skinner and Shackleton 2006]. IRD layers typically show an increase in the wgt. % coarse fraction [ Rashid et al. 2003], and thus this proxy was used to identify H1 1 and estimate relative changes in IRD throughout the section. Wgt.% Coarse Fraction = (mass of 150 !m fraction) / (total mass) *100 (7 1) Because lithic fragments greater than 150 !m were undoubtedly transported by ice [ Ruddiman 1977; Hemming 2004], this proxy provides a relative estimate of the change in IRD. The major drawback to using coarse fraction as an IRD proxy, however, is that it does not distinguish between IRD and foraminifera, which frequently make up a large portion of the > 150 !m size fraction. Peaks in coarse fraction were examined visually at site U1304 to determine if the increase was a result of IRD of foraminifera. With the exception of the peaks at TII, the
46 increases noted during MIS 6 are all related to high foraminife ral input (Figure 7 5). The large IRD flux during TII diluted the number of foraminifera and at 25.52 mcd, resulting in fewer than 25 specimens in each sample. This high lithic to foraminifer ratio is also indicative of H11 [ Heinrich 1988] The peak bet ween 23.17 and 22.67 mcd is interpreted as H11, which has been dated between 136 and 130 ka [ Skinner and Shackleton 2006]. Although detrital carbonate from the Paleozoic basins of Canada and northwestern Greenland typifies most Heinrich events, other ar eas around the North Atlantic contain carbonate lithologies that may have contributed to IRD input. For example, erosion by the British and Fennoscandian ice sheets can transport Cretaceous aged chalk southward of 59 ¡ N [ Hebblen et al. 1997]. Large parts of Ireland are covered by Carboniferous carbonates, representing another possible source of detrital carbonate IRD [ Richter et al. 2001]. The ratio of Ca to Sr (Ca/Sr), measured via XRF, has been shown to be a sensitive indicator of detrital carbonate because biogenic calcite has a greater Sr concentration than inorganic calcite [ Hodell et al. 2008]. Heinrich events with high amounts of allochtonous carbonate show peaks in Ca/Sr during H1, H2, H4, and H5 at IODP site U1308, but not during H3 and H6 [ H odell et al. 2008]. Site U1304 shows no Ca/Sr maximum during H11 and, together with %CaCO 3 data, indicates little detrital carbonate input during the IRD event (Figure 7 6). Because Heinrich Events sourced from Hudson Strait contain between 50 and 60% detrital carbonate by weight [ Rashid et al. 2003], these data suggest H11 was not derived from Hudson Strait at the location of site U1304. The final analysis for determining if the IRD at site U1304 was sourced from the Hudson Straits was to measure sta ble isotopes on the bulk sediment. The mean and standard deviation of oxygen isotope values of detrital carbonate grains sourced from the Hudson Strait region is 5.35
47 1.26 [ Hodell and Curtis 2008]. Bulk carbonate oxygen isotope measurements at site U1 304 show a value of 2.80 at TII (Figure 6 9). Although this is lower than biogenic oxygen isotope values, it does not approach the 5.35 suggested by Hodell and Curtis  to be indicative of a Hudson Strait source. Site U1308 (49 ¡ 53'N, 24 ¡ 14'W) co ntains little detrital carbonate over TII [ Hodell et al. 2008]. Within the IRD belt, Lototskaya and Ganssen  measured individual detrital carbonate grains from H11 at site T90 9P at 45 ¡ N, 25 ¡ W. Their mean 18 O values were 5.4, consistent with a Hudson Strait source. The difference in detrital input at these sites suggests the deposition of Hudson Strait detrital carbonate occurred during H11 but may have been limited to the southern part of the IRD belt. Neogloboquadrina pachyderma (s) Isotope Record Planktonic 18 O signals typically parallel benthic 18 O for foraminifera in marine sediment cores over glacial to interglacial cycles of the Pleistocene. One of the most obvious incongruities of the data from site U1304 are the high 18 O values expressed by N. pachyderma (s) throughout Substage 5e, with 18 O values up to 1.5 greater than corresponding benthic values (Figure 7 1). Other North Atlantic cores do not show this offset or variability, and the unusual results cannot be due to instrumental erro r because the analyses were completed over several months and standard results for this time period were within analytical precision. Portions of the section were picked for N. pachyderma (s) by two people, and double checked prior to analysis. Species m isidentification can therefore be ruled out as a possible explanation for the inconsistencies. Most likely, specimens are reworked and not representative of surface water conditions under which the foraminifera lived. Within the Iceland Basin, Bianchi and McCave  showed that strong ISOW flow is able to erode and transport sediment along the
48 northern Gardar Drift. It is therefore possible that ISOW resuspended material from glacial MIS 6 and redeposited it at site U1304 during Substage 5e. During g lacial periods, including MIS 6, N. pachyderma (s) dominated the foraminiferal population in the North Atlantic, and glacial age sediments are composed almost exclusively of the species [ Bauch and Kandiano 2007; Van Nieuwenhove et al. 2008]. Even without true census data, there is a noticeable decrease in the abundance of N. pachyderma (s) at TII and into Substage 5e, likely owing to SST exceeding temperature tolerances for this species (Figure 7 2). During Substage 5e, the foraminiferal assemblage was d ominated by G. bulloides G. inflata G. truncatulinoides and N. incompta. Resuspended and transported glacially derived sediments would be dominated by cold water N. pachyderma (s). Pelagic rain diluted any reworked interglacial warm water fauna, reduc ing the likelihood of picking a reworked interglacial individual. For example, if a transported, glacially derived G. bulloides were picked during Substage 5e, its isotope signal would have been masked by the measurement of up to 9 other interglacial indiv iduals from the pelagic rain. The major drawback of this hypothesis is that I picked each foraminifer from the >212 !m sieve fraction. Transport of this size fraction would have required very strong deep water currents on Gardar Drift. At nearby site NEAP18K, Hall et al.  estimated relative current speed via the sortable silt proxy, a distribution of grain size in the 10 to 63 m fraction, where stronger bottom currents yield a coarser mean grain size [ McCave et al. 1995; Hall et al. 1998]. Th eir study suggested a reduction in ISOW production during the first ~4 ka of the Last Interglacial (128 to 124 ka), with ISOW production increasing from ~124 to 117ka [Hall et al., 1998]. Bottom water speeds decreased after ~117 ka and remained low into S ubstage 5d, with the slowest speeds at ~113 ka [ Hall et al. 1998]. Details of the sortable silt proxy of Hall et al.
49  and how it relates to inferred bottom water changes at site U1304 are discussed in a later section; however, high flow speeds requ ired to transport foraminifera >212 !m were not reported for NEAP18K for the majority of Substage 5e and therefore unlikely for site U1304. Despite this drawback, the reworking hypothesis remains the most plausible explanation for the anomalous 18 O value s of N. pachyderma (s) Substage 5e The most significant aspect of this work is the high resolution, multi species approach used for isotopic analysis, with each foraminiferal species representing different depth habitats and/or seasons. Most other stud ies of MIS 5e in the North Atlantic measured a single species of planktonic foraminifera only [e.g. Chapman and Shackleton 1999a; Shackleton et al. 2002; Oppo et al. 2006; Bauch and Erienkeuser 2008; Van Nieuwenhove et al. 2008]. Various workers have subdivided Substage 5e into a number of intervals based on variations in planktonic 18 O [e.g. Chapman and Shackleton 1999a; Van Nieuwenhove et al. 2008]. In this study of Substage 5e, I used m ultiple species of planktonic foraminifera to compare isoto pic responses among species with different depth habitats and seasonal abundance peaks I found that the pattern of 18 O variation during the last interglacial is highly dependent upon which foraminiferal species is used for isotopic analysis. With the e xception of N. pachyderma (s), the overall pattern of each oxygen isotope signal is slightly different, with differences attributed to the preferred depth habitat and/or median growth season for each species. Globogerina bulloides calcifies in the upper 6 0 m of the water column [ Schiebel et al. 1997] whereas G. inflata completes most of its shell growth in the 25 to 75 m range, although it may continue adding calcite as deep as 150 to 200 m [ Wilke et al. 2006] Both species represent spring conditions, with peak abundances in March [ Deuser and
50 Ross 1989]. Neogloboquadrina incompta lives and calcifies in the upper 100 m of the water column during the late summer [ Ravelo and Fairbanks 1992; Ganssen and Kroon 2000; Salgueiro et al. 2008] Shell growt h for G. truncatulinoides occurs in the winter between the surface and 800 m depth, with average shell composition thought to represent ~200 m depth [ Deuser and Ross 1989; Lohmann 1995] The onset of Substage 5e is assigned the age of 128 ka [ Shackleton et al., 2003 ], based on the attainment of minimum benthic 18 O values at the start of the oxygen isotope plateau. Shortly after the onset of interglacial conditions, G. bulloides, G. inflata and N. incompta show a minimum in 18 O values at 21.97 mcd, cor responding to an age of ~123 ka (Figure 7 4). This time coincides with the sea level highstand inferred from coral terraces [ de Diego Forbis and Douglas 2002]. The minimum in the 18 O record of G. inflata lasts until 122 ka, approximately 1 ka longer th an the minimum in the G. bulloides record. Neogloboquadrina incompta shows the longest period of minimum 18 O values, lasting between 127 and 119.7 ka. In contrast, the 18 O record of G. truncatulinoides does not show a similar period of minimum 18 O val ues as do the other planktonic species. Low 18 O values recorded in G. bulloides and G. inflata represent a peak in spring temperature during early Substage 5e from 127 to 122 ka. Summer SST remained high for the longest period of Substage 5e with modera te cooling beginning at ~121.5 ka. Although warming may have affected G. truncatulinoides at depths of ~200 m during the winter months, low abundance of large individuals and variability of the oxygen isotope values makes it difficult to determine whether a minimum in 18 O may have existed. As spring season waters cooled at ~122 ka, the entire upper water column established a period of year round uniformity through ~119 ka as G. bulloides and G. inflata 18 O values become nearly equal.
51 After 123 ka, 18 O values of G. bulloides and G. inflata begin to increase to a plateau that is maintained for the majority of Substage 5e, through ~119 ka. The highest 18 O values for Substage 5e are recorded at 118.7 ka, immediately prior to a final peak in warmth befor e the glacial inception. Globorotalia inflata minima at the start and end of 5e differ in both magnitude and duration from the minima in the G. bulloides record, which reflects their slightly different depth habitats (Figure 7 7). The pattern of the N. in compta oxygen isotope record during Substage 5e is markedly different than both G. bulloides and G. inflata probably because of its different growth season (Figure 7 8) After 123 ka, the 18 O of N. incompta shows a slow increase and maintains low values through the end of Substage 5e. Similar to N. incompta G. truncatulinoides oxygen isotopes show a plateau from the onset of Substage 5e through the decline into Substage 5d. However, the 18 O of this deep dweller begins to increase into Substage 5d by ~119 ka, the earliest of all the planktonic species (Figure 7 9). Substage 5e stability was marked by warm summers and an intermediate depth spring mixed layer. Deep vertical mixing up to 600 m persisted during the winter, as seen by the increase of yo ung G. truncatulinoides (Figure 7 10) [ Lohmann and Schweitzer 1990; Lohmann 1992]. From 119.7 to 119.1 ka, summer intermediate depth temperatures underwent cooling of ~1 ¡ C, assuming 0.23 change in 18 O is equivalent to 1 ¡ C [ Shackleton 1974; Bemis et a l. 1998]. By ~119.5 ka, deep winter waters began to cool and the population of G. truncatulinoides became dominated by intermediate to moderate sized adults, which suggests a decrease in the scale of vertical mixing [ Lohmann and Schweitzer 1990; Lohmann 1992]. The end of Substage 5e was marked by spring surface warmth that deepened to the depth range of G. inflata 's habitat just prior to the transition to Substage 5d.
52 Near the end of the Substage 5e, G. bulloides reached another low in 18 O values dur ing a warming at the end of Substage 5e from 118.5 to 116.3 ka. This warming was also recorded in the G. inflata record though spring warmth at depth lagged the surface record, and occurred from 118.1 to 116.6 ka. Neither N. incompta nor G. truncatulinoid es recorded the last warm event, so it was likely restricted to the spring season, or to waters shallow than the living habitats of N. incompta and G. truncatulinoides The oxygen isotope transition into Substage 5d is marked by each species at a slightl y different time at site U1304 (Figure 7 1). The first species to transition to Substage 5d is G. truncatulinoides which indicates interglacial warmth ended by ~119.5 ka in deep winter waters. The benthic transition from Substage 5e to Substage 5d occurs at 118 ka, defined by the C26 cooling event [ Oppo et al. 2006]. This also correlates to the end of the benthic oxygen isotope plateau at 17.82 mcd. During the transition to Substage 5d, G. bulloides and G. inflata records diverge, suggesting more prono unced spring season stratification during and beyond the end of Substage 5e. Neogloboquadrina incompta is the last species to begin the transition to Substage 5d at 111 ka, 7 ka after the benthic record, indicating warm summer conditions persisted during the onset of ice growth at the beginning of Substage 5d [ Oppo et al., 2006], likely owing to a strong NAC. This scenario holds as long as the abundance of N. incompta remains high through the record of summer warmth. The oxygen isotope record of each sp ecies at site 1304 shows a different pattern during MIS 5e, suggesting that analysis of a single species results in an incomplete, biased interpretation of climate change during the last interglacial. Intraseasonal changes with depth are observed between G. bulloides and G. inflata indicating each species only represents oceanographic changes confined to its unique depth habitat and season of peak growth. At high resolution,
53 each species leads to a slightly different interpretation of surface water cond itions during Substage 5e (Figure 7 11), emphasizing the need to compare multiple species with different depth and/or seasonal habitats at a single site to assess patterns of climate change. C omparison of oxygen isotope records of the same species at mul tiple sites is useful for understanding the geographic extent of climate variability associated with a specific season and/or depth. Chapman and Shackleton [1999a] sampled site NEAP18K every 2 cm and measured stable isotope of G. bulloides beginning just after TII and through MIS 5 (Figure 7 12). Their record shows similar warmings at the onset and decline of Substage 5e, marked by 18 O minima. However, site NEAP18K lacks a plateau between the minima, and 18 O increases step wise throughout Substage 5e. The change in 18 O from the first to second minima at site NEAP18K is 0.34, and at site U1304 is 0.37, suggesting the overall change throughout Substage 5e was very similar for both sites. However, 18 O values at site U1304 are less variable than thos e at site NEAP18K during the majority of Substage 5e. Site T90 9P, a low resolution multi species site, also shows the two warm periods framing MIS 5e in the G. bulloides record, though the final warmth is represented by a single measurement [ Lototskaya and Ganssen 1999]. Site T90 9P is the only other study of Substage 5e in the North Atlantic to include four or more species of planktonic foraminifera [ Lototskaya and Ganssen 1999]. Despite the same multi species approach as my study, the low resolutio n sampling (21 samples within the 45 cm interval of Substage 5e) at T90 9P makes it difficult to directly compare the high resolution record of site U1304. On the Iberian Margin, Gouzy et al.  investigated site MD99 2331, and the 18 O record of G bulloides shows a dissimilar pattern to those in the central North Atlantic. Oxygen isotopes of G. bulloides at MD99 2331 do not show the same minima at either end of Substage
54 5e, but rather a broad peak occurs during approximately the first half of the interglacial, with no final warming. These findings indicate the warmings at the onset and end of Substage 5e may have been restricted to the central North Atlantic, although the period between the warming events varied with location, and was most pronou nced at site U1304. Oppo et al.  provided a high resolution record of site 980, including a planktonic isotope analysis of N. pachyderma (d), the equivalent of N. incompta (Figure 7 13). Neogloboquadrina incompta 18 O records for sites 980 and U1 304 are similar, but peak warmth occurs at 123 ka at site U1304, and near 120 ka for site 980 [ Oppo et al., 2007]. Neither site records the spring warming seen near the end of Substage 5e. Similar shapes of the N. incompta records at sites 980 and U1304 during Substage 5e show summer conditions were fairly uniform between these sites, indicating persistent, regional summer warmth. However, summer warmth declined more rapidly at site U1304 after 111 ka, coincident with C25 at site 980. Though C25 had li ttle effect on the N. incompta record at site 980, Oppo et al.  hypothesized greater cooling occurred in the central subpolar North Atlantic, within the region of site U1304. The correlation of these events suggests C25 caused the final surface wate r cooling ending the marine interglacial. Work by Kukla et al.  on a terrestrial site in Le Grande Pile, France suggested the interglacial climate lasted twice as long on land versus in the marine realm. Conditions at Le Grande Pile began to deter iorate in response to a climate shift at 111 ka [ Kukla et al. 1997], coincident with the decline in summer warmth recorded by N. incompta and C25 in site 980. Eemian warmth lasted approximately 4 ka longer, until 107 ka, when cooling event 24 of Substage 5d brought interglacial conditions to an end.
55 Substage 5d The onset of Substage 5d was marked by ice sheet growth and instability, and t he benthic isotope record shows a steady decline into cooler conditions after 118 ka. Both G. bulloides and G. infla ta show similar increases in 18 O through 107 ka, and their offset values suggest the development of surface water stratification during the spring season (Figure 7 14). Neogloboquadrina incompta also increases through 107 ka, though with a much steeper s lope By 110.4 ka, the N. pachyderma (s) 18 O signal returns to a shape reminiscent of planktonic trends. Cooling event C24 was identified at 107 ka, and associated IRD deposition occurred in the North Atlantic as far south as 40 ¡ N [ Cortijo et al. 1999; McManus et al. 2002; Mller and Kukla 2004; Oppo et al. 2006 ]. Cooling event 24 was likely a response to ice sheet instability, versus earlier cooling events such as C25 and C26, which were related to surface water circulation changes [ Oppo et al. 20 06]. Low abundance of G. truncatulinoides during the C24 event makes it difficult to assess the impact of the melt water event on this species. After 107 ka, the 18 O values for G. bulloides G. inflata N. incompta and N. pachyderma (s) decrease rapidl y, which may be either millennial scale variability or caused by surficial melt water during C24. Benthic 18 O values continue the increasing trend, suggesting the melt water during C24 was not pervasive enough to modify deep water isotopes. Surface Prod uctivity Carbon stable isotope analyses on planktonic foraminifera are used in conjunction with the presence of LDM/LDO to estimate changes in surface productivity. Generally, higher 13 C values reflect an increase in biological pumping, a process that d epletes the amount of 12 C in the photic zone. Low 13 C values for G. bulloides reflects pseudo symbiosis involving the utilization of metabolic CO 2 derived from algae, its dominant food source [ Hemleben et al.
56 1989; Cooke and Rohling 2001]. All plankto nic species record a minimum in 13 C, suggesting a low in surface productivity within MIS 6 at approximately 143 ka. After this, surface productivity increases for all planktonic species through TII (Figure 6 3), with 13 C data showing very little change over the termination. Lototskaya and Ganssen  reported a surface productivity trend at site T90 9P for a surface (30 to 60 m) summer dwelling species, G. ruber [ Deuser and Ross 1989; Kawahata et al. 2005]. At site T90 9P, G. ruber recorded the min imum in surface productivity during MIS 6, and increased into TII, reaching the peak in 13 C values at the onset of Substage 5e. The G. ruber 13 C values decline through Substage 5d, with low 13 C values related to enriched nutrients and DIC in the surface water [ Lototskaya and Ganssen 1999]. The findings at site T90 9P contrast the results for site U1304. At site U1304, Substage 5e became increasingly favorable for phytoplankton blooms, evidenced by the LDM. Diatoms caused an increase in surface produ ctivity, raising the 13 C in foraminiferal shell calcite for planktonic species at site U1304. Because the extent of LDM are limited across the North Atlantic, the surface productivity record at site U1304 are only interpreted to represent conditions with in a frontal convergence boundary. Site T90 9P was not under the NSAF and Lototskaya and Ganssen  did not report LDM within the core, so T90 9P is likely more representative of productivity conditions in the North Atlantic. Without the influence of large diatom blooms, peak surface productivity occurred during the onset of Substage 5e. However, surface productivity is closely tied to phytoplankton blooms, and site U1304 records peak surface productivity approximately coincident with peak LDM deposi tion. LDM are identified at site U1304 by peaks in color lightness, a broad magnetic susceptibility minimum, and peaks in the Si/Zr record (Figure 6 4). Correlating these proxies
57 provides an estimate for LDM deposition, which occurs from ~121 to 109.5 ka Regular, thick LDM layers occur in the latter half of Substage 5e from ~121 to 117 ka, with intermittent deposits both before and after the peak of LDM deposition. Though diatoms typically bloom during episodes of high nutrient flux, physical processes cause massive accumulation along convergence fronts [ Sarthou et al. 2005; Kemp et al., 2006]. At the Northern Sub Arctic Front, dense, cool Labrador Sea currents subduct beneath the warmer NAC (Figure 7 15). Diatoms entrained in the cool, more nutrient rich waters also begin to subduct, but buoyancy regulation causes them to rise again [ Yoder et al. 1994; Kemp et al. 1995]. They rise on the warm side of the front, become concentrated, strip the water of nutrients, increase the 13 C seawater and die, s inking in massive colonies at rates of up to 100 m/day [ Sarthou et al. 2005]. Surface accumulation of diatoms is restricted to a narrow zone at the convergence front, and deposition is sensitive to frontal movements [ Bodn and Backman 1996]. Several gr oups have hypothesized that LDM/LDO sequences in oceanic sediments can be used as proxies for paleoceanic convergence fronts [ Yoder et al. 1994; Kemp et al. 1995; Bodn and Backman 1996]. The NSAF position fluctuated throughout Substage 5e, leading to v ariable LDM/LDO deposition. Bodn and Backman  reported a LDO sequence throughout Substage 5e from site EW9303 17, NW of site U1304. However, Hall et al.  did not report any LDM/LDO during Substage 5e or 5d at site NEAP18K From these sites, it is reasonable to hypothesize that the front remained close site U1304 throughout much of Substage 5e but never reached far enough southeast to affect site NEAP18K. Comparing the LDM record to the planktonic 13 C values supports an increasing influenc e of highly productive waters at site U1304 through ~118 ka. Each species' 13 C maximum occurs during Substage 5e between ~118.8 and 117.5 ka, within the peak production
58 of LDM (Figure 7 17). For G. bulloides, 13 C values begin to decrease at ~118.8 ka, suggesting a decrease in phytoplankton populations in the surface most water at this time. Globorotalia inflata and N. incompta 13 C values remain high (through 118.3 and 117.7 ka, respectively) longer than G. bulloides which suggests intermediate produc tivity lasted slightly longer than surface productivity. The 13 C of G. truncatulinoides is derived entirely from it's secondary crust, and represents the carbon cycle and productivity at ~800 m [ Lohmann 1995]. Globorotalia truncatulinoides exhibits the latest peak in 13 C at 116.5 ka, which may reflect the persistence of a Deep Chlorophyll Maximum. Substage 5d is characterized by lower 13 C and associated productivity for all of the planktonic species This is also coincident with a decrease in LDM de position, reinforcing the relationship between planktonic foraminiferal 13 C and diatom production. Deep Water Fluctuation As a monitor of deep water circulation benthic 13 C decreases just before MIS 6 and following H11, suggesting stronger influences o f southern sourced bottom water. Sites SU90 03 and T90 9P also record low 13 C values during MIS 6, supporting an incursion of LDW [ Chapman and Shackleton 1998; Lototskaya and Ganssen 1999]. Following H11, 13 C minima at sites 980 and T90 9P suggest fu rther reduction in ISOW formation [ Lototskaya and Ganssen 1999; Oppo et al. 2006]. In contrast, 13 C values show little change over Termination II for site U1304. Other workers have inferred a nutrient rich water mass in the North Atlantic that was ass ociated with MIS 6 termed Glacial North Atlantic Intermediate Water (GNAIW) [ Adkins et al. 1997; Oppo et al. 1997]. The GNAIW, present at intermediate depths, was replaced by a low 13 C water mass during TII [Oppo et al., 1997]. This low 13 C water mass could have originated either via brine rejection in the Nordic Seas or from the Southern Ocean (as LDW).
59 Oppo et al.  suggested low 13 C values at site 980 (2169 m, Feni Drift) during TII was in response to an intermediate water mass formed in the Nordic Seas, which likely affected other shallower water sites. Fluctuations in 13 C at deeper sites, such as site U 1304, could be related to a reduction in MOC and incursions of LDW. The 13 C values measured at site U1304 are generally lower than site 980, illustrating the greater LDW influence at site U1304 (Figure 7 18), which continues into early Substage 5e. Inc reasing 13 C values at site 980 results in the development of a gradient between sites U1304 and 980, suggesting shallower ISOW production and that site U1304 was influenced by LDW. Based on a constant sedimentation rate within the Substage 5e plateau, th ese conditions persisted for ~3 ka until ~124.5 ka. This is consistent with the sortable silt record from Hall et al. , which suggests weak ISOW flow at site NEAP 18K during early Substage 5e (Figure 7 19) [ Oppo et al. 1997]. Further comparison of these records shows the low benthic 13 C at site U1304 and the weak ISOW flow from the sortable silt record coincide with minima in the 18 O records of G. bulloides and G. inflata The surface and deep records coincide with a strong warm summer insolatio n anomaly in the Northern Hemisphere during early Substage 5e (Figure 7 20). This insolation anomaly may have resulted in higher temperatures, melting continental ice sheets (i.e. Greenland, see Otto Bleisner et al ., 2006) and lowering surface salinity. At 124 ka, benthic 13 C increased and remained high until 117 ka. The high mean sortable silt values for the same time period suggest ISOW production was strong enough to penetrate to the depths of sites U1304 and NEAP 18K (Figure 7 19) [ Hall et al. 19 98]. Higher 18 O values for G. bulloides and G. inflata suggest slightly cooler and/or more saline surface waters. By ~116.5 ka, the benthic 13 C values decreased, concurrent with the lowest mean sortable silt values at NEAP 18K [ Hall et al. 1998], indi cative of a shoaling of ISOW during the
60 decline of Substage 5e into Substage 5d. In contrast to this, McManus et al.  suggested the subpolar North Atlantic experienced increased heat transport and enhanced MOC during the transition from Substage 5e to Substage 5d. This conflict can be resolved by a strong ISOW at intermediate depths on the Gardar Drift, but a shoaling of the boundary between intermediate and deep waters, which is consistent with the data. Work by Oppo et al.  also yielded a detailed examination of climate progression into Substage 5d. At 118 ka, C26 marks the end of interglacial warmth at sites 980 and U1304 [ Oppo et al. 2001; Oppo et al. 2006]. Several sites around the North Atlantic, including 980, 8JPC, SU90 03, NEAP18 K, T90 9P, and U1304 record C26 as a sharp decrease in 13 C near the end of Substage 5e [ Chapman and Shackleton 1998; Chapman and Shackleton 1999b; Lototskaya and Ganssen 1999; Oppo et al. 2001; Shackleton et al. 2003; Oppo et al. 2006]. This sugges ts C26 was a large magnitude surface freshening related to iceberg discharge that reduced NADW production and marked the end of interglacial conditions in the benthic record at site U1304, as well as sites 980, 8JPC, SU90 03, NEAP18K, and T90 9P. Followin g the minimum in ISOW production at ~116.5 ka, benthic 13 C and sortable silt increased from 113 to 105 ka [ Hall et al. 1998], indicating a strengthening and deepening of ISOW to below 3000 m. Other workers noted an association between the 13 C records a nd the cooling events that occurred during the transition into Substage 5d [ Chapman and Shackleton 1999b; Oppo et al. 2007]. C24 was a large magnitude surface freshening event with correlative negative 13 C values in sites 980 and NEAP18K at ~107 ka [ Ch apman and Shackleton 1999b; Oppo et al. 2007]. C24 is the first cooling event related to ice sheet instability during the transition to Substage 5d [ Oppo et al. 2007], and is detected at site U1304 as a 0.7 decrease in 13 C values, supporting other st udies that estimated the extent of C24.
61 Figure 7 1. Oxygen isotope results with isotope stages and major events. Foraminifera species are as follows: C. wuellerstorfi green triangles; G. bulloides blue diamonds; G. inflata pink crosses; N. incom pata yellow circles; N. pachyderma (s), red squares. G. truncatulinoides, gray circles.
62 Figure 7 2. Number of foraminifera picked by species. A) C. wuellerstorfi marked by green triangles, G. bulloides m arked by blue diamonds, G. inflata marked by pink crosses, N. pachyderma (d) marked by gold circles, N. pachyderma (s) marked by red squares. B) Total G. truncatulinoides abundance by sample A B
63 Figure 7 3. Planktonic oxygen isotopes during MIS 6 vers us depth. Foraminifera species are as follows: C. wuellerstorfi green triangles; G. bulloides blue diamonds; G. inflata pink crosses; N. incompata yellow circles; N. pachyderma (s), red squares; G. truncatulinoides, gray circles.
64 Figure 7 4. Comparison of planktonic records. Oxygen isotope records of G. bulloides (blue diamonds), N. incompta (yellow circles), and G. inflata (pink triangles) from site U1304. Vertical dashed line designates the Substage 5e/5d boundary.
65 Figure 7 5. Coarse Fraction proxy for IRD mass accumulation. Peaks of high biogenic input are noted during MIS 6. The peak at 22.37 is attributable to a single erratic within the sample. Another sample at 24.47 mcd had a large erratic, a nd the coarse fraction mass was ~ 36 % > 150 !m for that sample, so the erratic was removed as not to skew the data. The sample weight changed to 9.6 % > 150!m by mass, which is reflected in the data above. Percent of sample (by mass) greater than 150 !m
66 Figure 7 6. Ca/Sr Elemental ratio with percent CaCO 3. This was measured over the TII interval to illustrate the absence of detrital carbonate at H11. A) Ca/Sr ratio. B) Percent CaCO 3 TII is approximately between 22.5 and 23 mcd. A B
67 Figure 7 7. Oxygen isotopes of G. bulloides (blue diamonds) and G. inflata (pink crosses). Shown within Substage 5e, with C. wuellerstorfi (green triangles) for reference.
68 Figure 7 8. Oxygen isotopes of G. bulloides (blue diam onds), G. inflata (pink crosses), and N. incompata (yellow circles). Shown within Substage 5e, with C. wuellerstorfi (green triangles) for reference.
69 Figure 7 9. Oxygen isotopes of G. bulloides (blue diamonds), N. incompata (yellow c ircles), and G. truncatulinoides (gray circles). Shown within Substage 5e, with C. wuellerstorfi (green triangles) for reference.
70 Figure 7 10. Population distribution of G. truncatulinoides by size From top to bottom: Large adults > 425 !m are purple; Moderate sized adults 355 425 !m are blue; Intermediate sized individuals 212 355 !m are yellow; Juveniles 180 212 !m are green; Infants < 180 !m are pink.
71 Figure 7 11. Subdivision of Substage 5e using planktonic foraminifera. A) G. bulloides blue diamonds. B) G. inflata pink crosses. C) G. truncatulinoides, gray circles. D) N. incompata yellow circles. A C B D
72 Figure 7 12. Oxygen isotopes of G. bulloides at sites NEAP18K (bottom ax is) and U1304 (top axis). Shown with respect to sediment depth. Note the length of Substage 5e at site NEAP18k is ~70 cm, versus site U1304, which has ~6 m of sediment during Substage 5e. NEAP18K data from Chapman and Shackleton [1999b].
73 Fi gure 7 13. Oxygen isotopes of N. incompata at sites 980 and U1304. Site 980 (black circles) is on an age scale, and site U1304 (yellow circles) is on a depth scale.
74 Figure 7 14. Oxygen isotopes at site U1304 during Substage 5d. C. wuellerstorfi (green triangles), G. bulloides (blue diamonds), G. inflata (pink crosses), N, incompata (yellow circles), N. pachyderma (s) (red squares), and G. truncatulinoides (gray circles), with for reference.
75 Figure 7 15. Diatom mat formation at convergence fronts. General illustration of how diatoms become physically concentrated at surface convergence fronts, modified from Yoder et al. . As cold waters begin to descend below the warme r water mass, diatoms entrained in the cool water regulate their buoyancy and rise to the surface on the warm side of the front. Over time, surface currents concentrate the diatoms and they eventually die and sink to the seafloor in large colonial mats. Warm current indicated by dashed arrows, cold current by solid arrows. Diatoms are illustrated by solid circles, and their movement upward by squiggled arrows. Depth (Changes seasonally) LDM formation Distance (km) 0 ~ 100 NAC NSAF Polar Currents
76 Figure 7 16. Movement of the NSAF from MIS 6 through Su bstage 5e. Site EW9303 17 from Bodn and Backman , site NEAP18K from Hall et al.  and Cortijo et al  and site U1304 are indicated by red circles. Double ended arrows indicate oscillation of the NSAF during Substage 5e. Image modified from Google Earth . Site U1304 EW9303 17 NEAP18K NSAF during 5e NSAF during MIS 6
77 Figure 7 17. Planktonic carbon isotopes during Substage 5d with LDM deposition highlighted. G. bulloides are marked by blue diamonds, G. inflata by pink crosses, N. incompata by yellow circles, N. pachyderma (s) b y red squares, and G. truncatulinoides by gray circles.
78 Figure 7 18. Benthic carbon isotopes at sites U1304 (orange circles) and 980 (green triangles) versus age. Site 980 data from Oppo et al. .
79 Figure 7 19. Compa rison of Carbon isotopes at site U1304 and sortable silt at NEAP18 K. A) Benthic 13 C at sites U1304 (blue data points and orange 5 point mean) and NEAP 18K (red data points and black 5 point mean) [ Chapman and Shackleton 1999]; B) Benthic 13 C at site U1 304 (blue) and sortable silt (red) at NEAP 18K [ Hall et al. 1998]; C) Benthic 13 C at site U1304 (blue) and G. bulloides 18 O at site U1304 (black) and NEAP 18K (red).
80 Figure 7 20. Comparison of benthic, planktonic, and insolation records. Benthic 13 C of C. wuellerstorfi (blue circles) at Site U1304 compared with oxygen isotope records of G. bulloides (green diamonds) and C.wuellerstorfi (red circles) from the same core. Bold orange line is a 5 point running mean of benthic 13 C. Vertical dashed li nes designates the Substage 5e/5d boundary. Gray shading indicates intervals of low benthic 13 C. Top panel shows insolation anomaly relative to today at 65 o N at boreal summer solstice.
81 CHAPTER 8 CONCLUSIONS The isotopic records at sit e U1304 revealed no evidence of a pause or structure to Termination II. The absence of a pause at site U1304 may be due to either a condensed TII section, or an intermediate water mass formed in the Nordic Seas that was not recorded at site U1304. I show ed that H11 lacked a high carbonate influx, and that bulk carbonate isotope measurements did not conclusively show the Hudson Straits as the source region. Future work to determine source area for the small amount of detrital carbonate at Site U1304 shoul d include measuring stable isotopes on individual carbonate grains (as in Hodell and Curtis 2008). Site U1304 is unusual in that it contains LDM/LDO sediments that were predominantly deposited during the latter half Substage 5e. The high sedimentation rate and reduced bioturbation afforded by the LDM/LDO permitted a high resolution study of changes in upper water column conditions by isotopic analysis of multiple species of planktonic foraminifera. Each species has different depth and seasonal habitats that are reflected in the different patterns of 18 O variation before, during, and after Substage 5e. Here I showed that interpretation of surface water conditions and subdivision of the last interglacial is dependent upon which species is analyzed C arbon isotope records from site U1304 and sortable silt records from NEAP 18K revealed minima in deep water production from 128 to 124.5 ka, suggesting a shoaling of ISOW production and an increased influence of southern sourced bottom waters in deep regio ns of the Gardar Drift. This event coincides with peak summer boreal insolation and minima in planktonic isotope records, suggesting relationships between continental ice sheet melting, sea surface salinity, and deep water production characteristics for S ubstage 5e. Although forcing regimes will be different for future climate change compared to the Last Interglacial, this work
82 suggests ISOW production is reduced when the climate is warmer than at present. A major contribution is the recognition for usin g multiple species of planktonic foraminifera to characterize surface conditions within habitat and seasonal frameworks. Finally, this study was completed using a 5 cm sampling interval, equivalent to ~125 yrs. Much of the section contains LDM, inhibitin g bioturbation and affording the opportunity to study this location at a much finer resolution in future work.
83 LIST OF REFERENCES Adkins, J. F., E.A. Boyle, L. Keigwin, and E. Cortijo (1997), Variability of the North Atlantic thermohal ine circulation during the last interglacial period, Nature 390 154 156, doi: 10.1038/36540, Letter. Andrews, J. T. (1998), Abrupt changes (Heinrich events) in late Quaternary North Atlantic marine environments: a history and review of data and concep ts, J. Quaternary Sci. 13 3 16. Andrews, J. T., and K. Tedesco (1992), Detrital carbonate rich sediments, northwestern Labrador Sea: Implications for ice sheet dynamics and iceberg rafting (Heinrich) events in the North Atlantic, Geology 20 1087 1090. Arhan, M. (1990), The North Atlantic Current and subarctic intermediate water, J. Mar. Res., 48 109 144. Bauch, H. A., and E. Kandiano (2007), Evidence for early warming and cooling in North Atlantic surface waters during the last interglacial, Paleoce anography 22 PA1201. Bauch, H. A., and H. Erlenkeuser (2008), A "critical" climatic evaluation of last interglacial (MIS 5e) records from the Norwegian Sea, Polar Res. 27 135 151, doi:10.1111/j.1751 8369.2008.00059.x. Bauch, H. A., H. Erlenkeuser, S. J. A. Jung, and J. Thiede (2000), Surface and deep water changes in the subpolar North Atlantic during Termination II and the last inerglaciation, Paleoceanography 15 (1) 76 84, doi:1998PA000343. B, A. W. H. (1977), An ecological, zoogeographic and taxo nomic review of Recent planktonic foraminifera, Oceanic Micropaleontology; Volume 1 and 2 edited by A.T.S. Ramsay, pp1 100, Acad. Press, London, United Kingdom (GBR). Belkin, I. M., and S. Levitus (1996), Temporal variability of the Subarctic Front near the Charlie Gibbs Fracture Zone, Journal of Geophysical Research 101 (C12), 317 328. Bemis, B. E., H. J. Spero, J. Bijma, and D. W. Lea (1998) Reevaluation of the oxygen isotopic composition of planktonic foraminifera: experimental results and revised pal eotemperature equations, Paleoceanography 13 (2), 150 160. Bianchi, G. G., and I. N. McCave (2000), Hydrography and sedimentation under the deep western boundary current on Bjrn and Gardar Drifts, Iceland Basin, Mar. Geol. 165 137 169. Bintanja, R., a nd R. S. W. van de Wal (2008), North American ice sheet dynamics and the onset of 100,000 year glacial cycles, Nature 454, 869 872, doi:10.1038/nature07158.
84 Bigg, G. R., M. R. Wadley, D. P. Stevens, and J. A. Johnson (2000), Glacial thermohaline circulat ion states of the northern Atlantic: the compatibility of modeling and observations, J. Geol. Soc. London 157 655 665. Bodn, P., and J. Backman (1996), A laminated sediment sequence from the northern North Atlantic Ocean and its climatic record, Geolog y(Boulder) 24 (6), 507 510. Bond, G., et al. (1992), Evidence for massive discharges of icebergs into the North Atlantic ocean during the last glacial period, Nature 360 245 249. Broecker, W. S. (1991), The great ocean conveyor, Oceanography, 4 79 89 Broecker, W. S., G. C. Bond, M. Klas, E. Clark, and J. F. McManus (1992), Origin of the northern Atlantic's Heinrich events, Clim. Dyn. 6 265 273. Brauer, A., J. R. M. Allen, J. Mingram, P. Dulski, S. Wulf, and B. Huntley (2007), Evidence for last in terglacial chronology and environmental change from Southern Europe, Proc. Natl. Acad. Sci. USA 104 (2), 450 455, doi:10.1073/pnas.0603321104. Carlson, A. S., J. S. Stoner, J. P. Donnelly, and C. Hillaire Marcel (2008), Response of the southern Greenland Ice Sheet during the last two deglaciations, Geology 36 (5), 359 362. Channell, J. E. T., T. Sato, T. Kanamatsu, R. Stein, M. J. Malone, and the Expedition 303/306 Project Team (2004), North Atlantic Climate, IODP Sci. Prosp. 303 306. http://iodp.tamu.ed u/publications/SP/303306SP.PDF. Channell, J. E. T., T. Sato, T. Kanamatsu, R. Stein, M. J. Malone, and the Expedition 303/306 Scientists (2006), Site U1304 summary, Proceedings of the Integrated Ocean Drilling Program Vol. 303/306, doi:10.2204/iodp.proc. 303306.104.2006. Chapman, M. R., and N. J. Shackleton (1998), Millennial scale fluctuations in North Atlantic heat flux during the last 150,000 years, Earth Planet. Sc. Lett. 159 57 70. Chapman, M.R. and N.J. Shackleton (1999a), Global ice volume fluctu ations, North Atlantic ice rafting events, and deep ocean circulation changes between 130 and 70 ka, Geology 27 795 798. Chapman, M. R., and N. J. Shackleton (1999b), Late Quaternary North Atlantic IRD and Isotope Data, IGBP PAGES/World Data Center A fo r Paleoclimatology Data Contribution Series #1999 053. NOAA/NGDC Paleoclimatology Program, Boulder CO, USA, in Chapman, M.R. and N.J. Shackleton (1999), Global ice volume fluctuations, North Atlantic ice rafting events, and deep ocean circulation changes b etween 130 and 70 ka, Geology 27 795 798. Clark, P. U., R. B. Alley, and D. Pollard (1999), Northern Hemisphere Ice Sheet Influences on Global Climate Change, Science 286 1104 1111.
85 Cooke, S., and E. J. Rohling (2001), Stable isotopes in foraminifera l carbonate, Southampton Oceanography Centre Internal Document, 56 pp. Cortijo, E., S. Lehman, L. Keigwin, M. Chapman, D. Paillard, and L. Labeyrie (1999), Changes in meridional temperature and salinity gradients in the North Atlantic Ocean (30 ¡ 72 ¡ N) during the last interglacial period, Paleoceanography 14 (1), 23 33. Curry, W. B., J. C. Duplessy, L. D. Labeyrie, and N. J. Shackleton (1988), Changes in the Distribution of 13C of Deep Water #CO2 Between the Last Glaciation and the Holocene, Paleocean ography 3 (3), 317 341. Darling, K. F., M. Kucera, D. Kroon, and C. M. Wade (2006), A resolution for the coiling direction paradox in Neogloboquadrina pachyderma Paleoceanography 21 PA2011, doi:10.1029/2005PA001189. de Diego Forbis, T. A., and R. Doug las (2002), Depositional paleoenvironments and stratigraphy of late Pleistocene deposits at Rancho Las Animas, Baja California Sur, Mexico, 2002 Annual Meeting, Geological Society of America, Denver, CO, Oct. 27 30, 2002. Deuser, W. G., and E. H. Ross (19 89), Seasonally abundant planktonic foraminifera of the Sargasso Sea: Succession, deep water fluxes, isotopic compositions, and paleoceanographic implications, J. Foramin. Res. 19 (4), 268 293. de Vernal, A., C. Hillaire Marcel, W. R. Peltier, and A. J. W eaver (2002), Structure of the upper water column in the northwest North Atlantic: Modern versus Last Glacial Maximum conditions, Paleoceanography 17 (4), 1050, doi:10.1029/2001PA000665. Dickens, G. R., and J. A. Barron (1997), A rapidly deposited pennate diatom ooze in upper Miocene lower Pliocene sediment beneath the North Pacific polar front, Mar. Micropaleontol. 31 (3 4), 177 182. Dickson, R. R., and J. Brown (1994), The Production of North Atlantic Deep Water: Sources, rates, and pathways, J. Geophys Res. 99 12319 12341. Duplessy, J. C., D. M. Roche, and M. Kageyama (2007), The Deep Ocean During the Last Interglacial Period, Science 316 89 91. Esat, T. M., M. T. McCulloch, J. Chappell, B. Pillans, A. Omura (1999), Rapid fluctuations in sea leve l recorded at Huon peninsula during the Penultimate Deglaciation, Science 283 197 201. Fraile, I., M. Schulz, S. Mulitza, M. Kucera (2008), Predicting the global distribution of planktonic foraminifera using a dynamic ecosystem model, Biogeosciences 5 891 911.
86 Ganssen, G. M., and D. Kroon (2000), The isotopic signature of planktonic foraminifera from NE Atlantic surface sediments: implications for the reconstruction of past oceanic conditions, J. Geol. Soc. 157( 3 ) 693 699. Gallup, C. D., H. Cheng, F. W. Taylor, and R. L. Edwards (2002), Direct determination of timing of sea level change during Termination II, Science 295 310 313. Gouzy, A., B. Malaize, C. Pujol, and K. Charlier (2004), Climatic "pause" during Termination II identified in shallow and intermediate waters off the Iberian margin, Quaternary Sci. Rev. 23 (14 15), 1523 1528. Hall, I. R., I. N. McCave, M. R. Chapman, and N. J. Shackleton (1998), Coherent deep flow variation in the Iceland and American basins during the last interglacia l, Earth Planet. Sc. Lett. 164 15 21. Hebblen, D., R. Heinrich, and K. H. Baumann (1997), Paleoceanography of the last interglacial/glacial cycle in the polar North Atlantic, Quaternary Sci. Rev. 17 125 153. Heinrich, H. (1988), Origin and consequenc es of cyclic ice rafting in the northeast Atlantic Ocean during the past 130,000 years, Quaternary Res. 29 142 152. Hemleben, C., M. Spindler, and O. R. Anderson (1989), Modern Planktonic Foraminifera 363 pp., Springer Verlag, New York, New York. Hemm ing, S. R. (2004), Heinrich Events: Massive late Pleistocene detritus layers of the North Atlantic and their global climate imprint, Review of Geophysics 42 RG1005. Hodell, D. A., and J. H. Curtis (2008), Oxygen and carbon isotopes of detrital carbonate in North Atlantic Heinrich Events, Marine Geology 256 30 35, doi:10.1016/j.margeo.2008.09.010. Hodell, D. A., J. E. T. Channell, J. Curtis, O. Romero, U. and Roehl, (2008), Onset of "Hudson Strait" Heinrich Events in the Eastern North Atlantic at the e nd of the Middle Pleistocene Transition (~640 ka)?, Paleoceanography 23 doi:10.1029/2008PA001591. Holz, C., J. B. W. Stuut, and R. Heinrich (2007), Variability in terrigenous sedimentation processes of northwest Africa and its relation to climate change s: inferences from grain size distributions of a Holocene marine sediment record, Sediment. Geol. 202 (3), 499 508, doi:10.1016/j.sedgeo.2007.03.015. Hskuldsson, ., R. Hey, E. Kjartansson, and G. B. Gumundsson (2007), The Reykjanes Ridges between 63 ¡ 10 'N and Iceland, J. Geodyn. 43 (1), 73 86, doi:10.1016.j.jog.2006.09.003. Johannesen T., and E. Jansen (1992), Distribution of oxygen isotopes in recent planktonic foraminifera from the Greenland, Iceland and Norwegian Seas and their relationship to differ ent water masses, Dr Scient thesis, University of Bergen, Norway.
87 Kawahata, H., A. Kuroyanangi, and M. Kuwae (2005), Stable isotopic composition of two morphotypes of Globigerinoides ruber (white) in the subtropical gyre in the North Pacific, Fall Meeting 2005, American Geophysical Union, San Francisco, California. Kemp, A. E. S., and J. G. Bauldauf (1993), Vast Neogene laminated diatom mat deposits from the eastern equatorial Pacific Ocean, Nature 362 141 144. Kemp, A. E. S., J. G. Bauldauf, and R. B. Pearce (1995), Origins and paleoceanographic significance of laminated diatom ooze from the eastern equatorial Pacific [ODP leg 138], Proceedings of the Ocean Drilling Program Scientific Results, 138 641 645. Kemp, A. E. S., R. B. Pearce, I. Grigorov, J Rance, C. B. Lange, P. Quilty, and I. Salter (2006), Production of giant marine diatoms and their export at oceanic frontal zones: Implications for Si and C flux from stratified oceans, Global Biogeochem. Cy. 20 GB4S04. Kidd, R. B., and P. R. Hill (19 87), Sedimentation on Feni and Gardar Sediment Drifts, Initial Report of the Deep Sea Drilling Project 94 1217 1244, doi:10.2973/dsdp/proc.94.148.1987. King, S. C., J. W. Murray, and A. E. S. Kemp (1998), Palaeoenvironments of deposition of Neogene lami nated diatom mat deposits from the eastern equatorial Pacific from studies of benthic foraminifera (sites 844, 849, 851), Mar. Micropaleontol., 35 161 177. Kukla, G., J. F. McManus, D. D. Rousseau, and I. Chuine (1997), How Long and How Stable was the La st Interglacial?, Quaternary Sci. Rev., 16 605 612. LeGrande, A. N., J. Lynch Stieglitz, and E. C. Farmer (2004), Oxygen isotopic composition of Globorotalia truncatulinoides as a proxy for intermediate depth density, Paleoceanography 19 PA4025, doi:10 .1029/2004PA001045. Lisiecki, L. E., and M. E. Raymo (2005), A Pliocene Pleistocene stack of 57 globally distributed benthic 18 O records, Paleoceanography 20 PA1003, doi:10.1029/2004PA001071. Lohmann, G. P. (1992), Increasing seasonal upwelling in the subtropical South Atlantic over the past 700,000 yrs: Evidence from deep living planktonic foraminifera, Mar. Micropaleontol., 19 1 12. Lohmann, G. P. (1995), A model for variation in the chemistry of planktonic foraminifera due to secondary calcificati on and selective dissolution, Paleoceanography 10 (3), 445 457. Lohmann, G. (1998), The influence of a Near Bottom Transport Parameterization on the Sensitivity of the Thermohaline Circulation, American Meteorological Soc. 28 2095 2103. Lohmann, G. P., and P. N. Schweitzer (1990), Globorotalia truncatulinoides' growth and chemistry as probes of the past thermocline: 1. Shell size, Paleoceanography, 5 (1) 55 75.
88 Lototskaya, A., and G. M. Ganssen (1999), The Structure of Termination II (penultimate degla ciation and Eemian) in the North Atlantic, Quaternary Sci. Rev. 18 1641 1654. McCartney, M. S. (1992), Recirculating components of the deep boundary current of the northern North Atlantic, Prog. Oceanogr. 29 283 383. McCave, I. N., B. Manighetti and S. G. Robinson (1995), Sortable silt and fine sediment size/composition slicing: parameters for palaeocurrent speed and palaeoceanography. Paleoceanography 10 593 610. McCorkle, D. C., and L. D. Keigwin (1994), Depth profiles of 13C in bottom water and core top C. wuellerstorfi on the Ontong Java Plateau and Emperor Seamounts, Paleoceanography 9 (2), 197 208. McKenna, V. S., and W. L. Prell (2004), Calibration of the Mg/Ca of Globorotalia truncatulinoides (R) for the reconstruction of marine temperatur e gradients, Paleoceanography 19 PA2006, doi:10.1029/2000PA000604. McManus, J. F., D. W. Oppo, L. D. Keigwin, J. L. Cullen, and G. C. Bond (2002), Thermohaline Circulation and Prolonged Interglacial Warmth in the North Atlantic, Quaternary Res. 58 17 21, doi:10.1006/qres.2002.2367. Moros, M., A. Kuijpers, I. Snowball, S. Lassen, D. Bckstrm, F. Gingele, and J. McManus (2002), Were glacial iceberg surges in the North Atlantic triggered by climatic warming? Mar. Geol. 192 393 417. Mortyn, P. G., a nd C. D. Charles (2003), Planktonic foraminiferal depth habitat and delta O 18 calibrations: Plankton tow results from the Atlantic sector of the Southern Ocean, Paleoceanography 18 (2), 1037, doi:10.1029/2001PA000637. Oppo, D. W., L. D. Keigwin, and J. F McManus (2001), Persistent suborbital climate variability in marine isotope stage 5 and Termination II, Paleoceanography 16 (3), 280 292, doi:2000PA000527. Oppo, D. W., J. F. McManus, and J. L. Cullen (2006), Evolution and demise of the Last Interglacia l warmth in the subpolar North Atlantic, Quaternary Sci. Rev. 25 (23 24), 3268 3277, doi:10.1016.j.quascirev.2006.07.006. Oppo, D.W., et al. 2007. Subpolar North Atlantic ODP980 MIS5 Sediment Data. IGBP PAGES/World Data Center for Paleoclimatology Data C ontribution Series # 2007 030. NOAA/NCDC Paleoclimatology Program, Boulder CO, USA. Otto Bliesner, B. L., S. J. Marshall, J. T. Overpeck, G. H. Miller, and A. Hu (2006), CAPE Last Interglacial Project members Simulating Arctic Climate Warmth and Icefield Retreat in the Last Interglaciation. Science, 311, 1751 1753.
89 Overpeck, J. T., B. L. Otto Bliesner, G. H. Miller, D. R. Muhs, R. B. Alley, and J. T. Kiehl (2006), Paleoclimatic Evidence for Future Ice Sheet Instability and Rapid Sea Level Rise, Science, 311 (5768), 1747 1750, doi:10.1126/science.1115159. Pollard, R. T., J. F. Read, N. P. Holliday, and H. Leach (2004), Water masses and circulation pathways through the Iceland Basin during Vivaldi 1996, J. Geophys. Res., 109, C04004, doi:10.1029/2003JC0020 67. Rashid, H., and E. A. Boyle (2007), Mixed Layer Deepening During Heinrich Events: A Multi Planktonic Foraminiferal 18 O Approach, Science 318 (5849), 439 441, doi:10.1126/science.1146138 Rashid, H., H. Reinhard, D. J. W. Piper (2003), Origin of unu sually thick Heinrich layers in ice proximal regions of the northwest Labrador Sea. Earth. Planet. Sc. Lett. 28 319 336. Ravelo, A. C., and R. G. Fairbanks (1992), Oxygen isotopic composition of multiple species of planktonic foraminifera: Recorders of the modern photic zone temperature gradient, Paleoceanography, 7 815 831. Revel, M., M. Cremer, F. E. Grousset, and L. Labeyrie (1996), Grain size and Sr Nd isotopes as tracer of paleo bottom current strength, Northeast Atlantic Ocean, Mar. Geol. 131 (3 4), 233 249. Reverdin, G., P. P. Niiler, and H. Valdimarsson (2003), North Atlantic Ocean surface currents, J. Geophys. Res. 108 (C1), 3002, doi:10.1029/2001JC001020. Richter, T. O., S. Lassen, T. C. E. van Weering, and H. de Haas (2001), Magnetic susce ptibility patterns and provenance of ice rafted material at Feni Drift, Rockall Trough: implications for the history of the British Irish ice sheet, Mar. Geol. 173 (1 4), 37 54, doi:10.1016/S0025 3227(00)00165 1. Risebrobakken, B., E. Balbon, T. Dokken, E. Jansen, C. Kissel, L. Labeyrie, T. Richter, and L. Senneset (2006), The penultimate deglaciation: High resolution paleoceanographic evidence from a north south transect along the eastern Nordic Seas, Earth Planet. Sc. Lett. 241 505 516. Ruddiman, W. F. (1972), Sediment redistribution on the Reykjanes Ridge: seismic evidence, Geol. Soc. Amer. Bull. 83 2039 2062. Ruddiman, W. F. (1977), Late Quaternary deposition of ice rafted sand in the subpolar North Atlantic (lat 40 ¡ to 65 ¡ N), Geol. Soc. Am. Bull 88 1813 1827. Salgueiro, E. et al. (2008), Planktonic foraminifera from modern sediments reflect upwelling patterns off Iberia: Insights from a regional transfer function, Mar. Micropaleontol. 66 (3 4), 135 164, doi:10.1016/j.marmicro.2007.09.003.
90 S nchez Goi, M. F., F. Eynaud, J. L. Turon, and N. J. Shackleton (1999), High resolution palynological record off the Iberian margin: direct land sea correlation for the Last Interglacial complex, Earth. Planet. Sc. Lett. 171 123 137. Sarthou, G., K. R. Timmermans, S. Blain, and P. Trguer (2005), Growth Physiology and fate of diatoms in the ocean: a review, J. Sea Res. 53 25 42. Saunders, P. M. (1996), The flux of dense cold overflow water southeast of Iceland, J. Phys. Oceanogr. 26 85 95. Schiebel R., B. Hiller, and C. Hemleben (1995), Impacts of storms on recent planktic foraminiferal test production and CaCO 3 flux in the North Atlantic at 47 ¡ N, 20 ¡ W (JGOFS), Mar. Micropaleontol. 26 115 129. Schiebel, R., J. Bijma, and C. Hemleben (1997), Popu lation dynamics of the planktic foraminifer Globigerina bulloides from the eastern North Atlantic, Deep Sea Res. 44 (9 10), 1701 1713. Schiebel, R., J. Waniek, M. Bork, and C. Hemleben (2001), Planktic foraminiferal production stimulated by chlorophyll re distribution and entrainment of nutrients, Deep Sea Res. I 48 721 740. Shackleton, N. J. (1974), Attainment of isotopic equilibrium between ocean water and benthonic foraminifera genus Uvigerina: isotopic changes in the ocean during the last glacial, C olloq. Int. C.N.R.S. 219 203 209. Shackleton, N. J., and N. D. Opdyke (1973), Oxygen isotope and paleomagnetic stratigraphy of equatorial Pacific core V28 238: Oxygen isotope temperatures and ice volumes on a 105 year and 106 year scale, Quaternary Res. 3 39 55. Shackleton, N. J., M. Chapman, M. F. Snchez Goi, D. Pailler, and Y. Lancelot (2002), The Classic Marine Isotope Substage 5e, Quaternary Res. 58 14 16, doi:10.1006/qres.2001.2312. Shackleton, N. J., M. F. Snchez Goi, D. Pailler, and Y. L ancelot (2003), Marine Isotope Substage 5e and the Eemian Interglacial, Global Planet. Change 36 151 155. Sharp, Z. (2006), Principles of Stable Isotope Geochemistry 360 pp., Prentice Hall. Skinner, L. C., and N. J. Shackleton (2006), Deconstructing T erminations I and II: revisiting the glacioeustatic paradigm based on deep water temperature estimates, Quaternary Sci. Rev. 25 3312 3321.
91 Spero, H. J., and D. W. Lea (1996), Experimental determination of stable isotope variability in Globigerina bulloi des : implications for paleoceanographic reconstructions, Mar. Micropaleontol. 28 (3 4), 231 246, doi:10.1016/0377 8398(96)00003 5. Steenfelt, A. (2001), Geochemical atlas of Greenland West and South Greenland: Geological Survey of Denmark and Greenland, Report 2001/46, 40 pp. Sy, A., U. Schauer, and J. Meincke (1992), The North Atlantic Current and its associated hydrographic structure above and eastwards of the Mid Atlantic Ridge Deep Sea Res., 39, 825 853. Thierry, V. E., E. de Boissson, and H. Mer cier (2008), Interannual variability of the Subpolar Mode Water properties over the Reykjanes Ridge during 1990 2006, J. Geophys. Res. 113 C04016, doi:10.1029/2007JC004443. Tolderlund, D. S., and A. W. H. B (1972), Seasonal distribution of planktonic f oraminifera in the western North Atlantic, Micropaleontology 17 297 329. Van Nieuwenhove, N., H. A. Bauch, and J. Matthiessen (2008), Last interglacial surface water conditions in the eastern Nordic Seas inferred from dinocyst and foraminiferal assembla ges, Mar. Micropaleontol. 66 (3 4), 247 263, doi:10.1016/j.marmicro.2007.10.004. Volkmann, R., and M. Mensch (2001), Stable isotope composition of living planktic foraminifers in the outer Laptev Sea and the Fram Strait, Mar. Micropaleontol. 42 (3 4), 163 188. Weinelt, M. et al. (2001), Paleoceanographic Proxies in the Northern North Atlantic, in The Northern North Atlantic: A Changing Environment Edited by P. Schfer, W. Ritzrau, M. Schlter, and J. Thiede, pp 319 352, Springer. Wilke, I., T. Bickert, and F. J. C. Peeters (2006), The influence of seawater carbonate ion concentration [CO 3 2 ] on the stable carbon isotope composition of the planktonic foraminifera species Globorotalia inflata Mar. Micropaleontol. 58 (4), 243 258, doi:10.1016/j.marmicro.20 05.11.005. Wu, G., and C. Hillaire Marcel (1994), Oxygen isotope compositions of sinistral Neogloboquadrina pachyderma tests in surface sediments: North Atlantic Ocean, Geochim. Cosmochim. Ac. 58 (4), 1303 1312. Yoder, J. A., S. G. Ackleson, R.T. Barbe r, P. Flament, and W.M. Balch (1994), A line in the sea, Nature 371 689 692.
92 BIOGRAPHICAL SKETCH Emily Kay Minth was born on Feb 10, 1984 in St. Paul, Minnesota. With one older brother she lived in Minnesota until she was 8 years old before moving to Melbourne, Florida. After a brief stay in Florida, Her family decided to move to Atlanta, Georgia for 4 years, where Emily got her first taste of the foothills of the Appalachian Mountains. After moving to Ocala, Florida, where she graduated from Bell eview High School in 2002, she moved 45 minutes away to Gainesville, Florida to attend the University of Florida. At the University of Florida, she obtained a Bachelor of Sciences degree in Geology and graduated Summa Cum Laude in 2006. Emily stayed at t he University of Florida to complete her Master of Sciences degree in Geology. In January of 2009, Emily began working for Royal Dutch Shell as a geologist in Houston, Texas.