USF Libraries
USF Digital Collections

Orbital- to millennial-scale variability in Gulf of Mexico sea surface temperature and salinity during the late Pleistocene

MISSING IMAGE

Material Information

Title:
Orbital- to millennial-scale variability in Gulf of Mexico sea surface temperature and salinity during the late Pleistocene
Physical Description:
Book
Language:
English
Creator:
Whitaker, Jessica L
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Tropics
Climate
Stable isotope
Salinity
Dissertations, Academic -- Marine Science -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Sea surface temperature (SST) reconstructions from the low latitudes indicate the tropics/subtropics warmed significantly before glacial-interglacial decreases in global ice volume, suggesting the importance of tropical and subtropical climate in driving glacial terminations. ODP Site 625, drilled at a water depth of 889 m near De Soto Canyon in the Gulf of Mexico (GOM), provides continuous records of marine isotope stages (MIS) 1-6 sampled at a mean temporal resolution of 400 years. Age control is based on 8 AMS radiocarbon dates, marine isotope stratigraphy, and Foraminifera datum levels. Results from Globigerinoides ruber (white variety) Mg/Ca-SST indicate a rise of 4.4 °C from last glacial maximum to modern conditions and a 3.2 °C rise from the penultimate glaciation to the last interglaciation. However, model results suggest reduced thermohaline circulation (THC) causes salt and heat build-up in the Atlantic Warm Pool.Paired G. ruber Mg/Ca-SST and δ¹⁸O provide evidence of sub-millennial scale variability in GOM SST and SSS that is probably influenced by the strength of NADW production, as also observed in the Western Caribbean Sea. We test the idea that widespread abrupt climate change during the last glaciation caused by millennial scale fluctuations in the intensity of THC was modulated by Laurentide ice sheet (LIS) meltwater routed to the North Atlantic. To understand LIS melting dynamics and test the Meltwater Routing Hypothesis, we investigate the phasing of GOM SST and LIS freshwater events in relationship to high latitude climate. Estimated salinities from our multi-proxy approach suggest three freshwater events with a major freshwater influx from that occurred during Heinrich Event 2. This result confirms previous studies that suggested LIS summer melting during warmings in Antarctica. We also find a climate reversal during termination II from 130.4-128.4 ka.The initial rise in GOM SST at 132.1 ka of 2.9 °C is followed by a cold reversal of 1.5 °C at 130.4 ka for 2 ky and final increase to full interglacial warmth. The reversal in GOM SST is consonant with a pause in sea level rise and reduced NADW, suggesting a reduction in THC may have caused a global two-step deglaciation.
Thesis:
Thesis (M.S.)--University of South Florida, 2008.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Jessica L. Whitaker.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 74 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002001034
oclc - 319431635
usfldc doi - E14-SFE0002550
usfldc handle - e14.2550
System ID:
SFS0026867:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Orbitalto millennial-scale variability in Gulf of Mexico sea surface temperature and salinity during the late Pleistocene by Jessica L. Whitaker A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: Benjamin P. Flower, Ph.D. David W. Hastings, Ph.D. Thomas P. Guilderson, Ph.D. Albert C. Hine, Ph.D. Date of Approval: June 26, 2008 Keywords: tropics climate stable isotope salinity Copyright 2008, Jessica L. Whitaker

PAGE 2

Table of Contents List of Tables i List of Figures ii Abstract iii 1. Introduction 1 1.1 The role of low latitudes in climate change 1 1.2 Laurentide ice sheet meltwater 2 2 Core location and methods 5 2.1 Core location 5 2.2 18O and Mg/Ca analyses 5 3. Age model 7 3.1 Foraminifera datum levels 7 3.2 Radiocarbon 9 3.3 Isotope stratigraphy 9 3.4 Depth to age (ka) conversion 10 4. Results and Discussion 11 4.1 Raw stable isotope and Mg/Ca data 4.2 Northeast GOM SSS variability 16 4.3 A comparison of AWP SSS variability: Northeast GOM versus 21 Western Caribbean 4.4 Connection between GOM SST and SSS and THC Variability 25 4.5 Meltwater Routing Hypothesis 26 4.6 Northeastern GOM paleo-productivity 28 4.7 Last local occurrence of Globorotalia tumida flexuosa 33 4.8 Two-step deglaciation during termination II 33 5. Conclusion 35 References 38 Appendices 60 Appendix A: Error analysis and supplementary tables and figures 61 Appendix B: Raw G. ruber stable isotope and Mg/Ca data with calculated 67 proxy SST, 18OSW data for ODP Site 625

PAGE 3

List of Tables Table 1 Radiocarbon ages for ODP Hole 625 C 9 Table 2 Age control for ODP Site 625 10 Table 3 Average SST at ODP Site 625 15 i

PAGE 4

List of Figures Figure 1 Location map for ODP Site 625 3 Figure 2 Raw G. ruber stable isotope and Mg/Ca data for ODP Site 625 8 versus core depth Figure 3 Age model for ODP Site 625 12 Figure 4 Paired 18 O C and Mg/Ca SST data, calculated 18 O SW versus age (ka) 18 Figure 5 18 O GOM and sea surface salinity mixing model for the Gulf of Mexico 20 Figure 6 G. ruber Mg/Ca-SST, 18 O GOM and estimated change in Gulf of Mexico 22 sea surface salinity versus age (ka) Figure 7 Comparison of the difference in 18 O composition of seawater at ODP 24 625 in the Gulf of Mexico ( 18 O GOM ) and ODP 999 in the Caribbean ( 18 O Caribbean ) relative to modern conditions versus age (ka) Figure 8 ODP Site 625 18 O GOM and Mg/Ca-SST compared to Ceara Rise benthic 27 13 C versus age (ka) Figure 9 ODP Site 625 18 O GOM and Mg/Ca-SST compared to NGRIP and 29 EDML versus age (ka) Figure 10 13 C composition of G. ruber and percent coarse fraction in ODP 625 31 versus age (ka) Figure 11 Two-step deglaciation during termination II 34 ii

PAGE 5

Orbital to millennial-scale variability in Gulf of Mexico sea surface temperature and salinity during the late Pleistocene Jessica Lorraine Whitaker ABSTRACT Sea surface temperature (SST) reconstructions from the low latitudes indicate the tropics/subtropics warmed significantly before glacial-interglacial decreases in global ice volume, suggesting the importance of tropical and subtropical climate in driving glacial terminations. ODP Site 625, drilled at a water depth of 889 m near De Soto Canyon in the Gulf of Mexico (GOM), provides continuous records of marine isotope stages (MIS) 1-6 sampled at a mean temporal resolution of 400 years. Age control is based on 8 AMS radiocarbon dates, marine isotope stratigraphy, and Foraminifera datum levels. Results from Globigerinoides ruber (white variety) Mg/Ca-SST indicate a rise of 4.4 C from last glacial maximum to modern conditions and a 3.2 C rise from the penultimate glaciation to the last interglaciation. However, model results suggest reduced thermohaline circulation (THC) causes salt and heat build-up in the Atlantic Warm Pool. Paired G. ruber Mg/Ca-SST and 18 O provide evidence of sub-millennial scale variability in GOM SST and SSS that is probably influenced by the strength of NADW production, as also observed in the Western Caribbean Sea. We test the idea that widespread abrupt climate change during the last glaciation caused by millennial scale fluctuations in the intensity of THC was modulated by Laurentide ice sheet (LIS) meltwater routed to the North Atlantic. To understand LIS melting dynamics and test the Meltwater Routing Hypothesis, we investigate the phasing of GOM SST and LIS freshwater events in relationship to high latitude climate. Estimated salinities from our multi-proxy approach suggest three freshwater events with iii

PAGE 6

a major freshwater influx from that occurred during Heinrich Event 2. This result confirms previous studies that suggested LIS summer melting during warmings in Antarctica. We also find a climate reversal during termination II from 130.4-128.4 ka. The initial rise in GOM SST at 132.1 ka of 2.9 C is followed by a cold reversal of 1.5 C at 130.4 ka for 2 ky and final increase to full interglacial warmth. The reversal in GOM SST is consonant with a pause in sea level rise and reduced NADW, suggesting a reduction in THC may have caused a global two-step deglaciation. iv

PAGE 7

1 1. Introduction 1.1 The role of low latitudes in climate change Recent evidence from the tropical Pacifi c and Atlantic Oceans suggests the low latitudes play an important role in driving glacial to interglacial climate change. The Western Pacific Warm Pool, as the largest body of warm water in the modern oceans, affects global climate thr ough variation in air-sea in teraction modulated by sea surface temperature (SST) changes within the region. Lea et al. [2000] reconstructed SST and the 18O of Western Equatorial Pacific surf ace water over the past 450 thousand years based on Ocean Drilling Program (ODP) Site 806. They discovered a 3-5 C cooling during glaciations and a 3 ky lead in SST over global ice volume during the last four glacial terminations. Thus, deep sea sedime nt records from the Western Pacific Warm Pool indicate the low latitudes, as an early responder to orbital insolation, play an important role in driving glacial to interglacial climate change. The Gulf of Mexico (GOM), Caribbean Sea, and tropical Northwest Atlantic comprise the Atlantic Warm Pool, the Atlant ic portion of the Western Hemisphere Warm Pool. The Atlantic Warm Pool constitutes a large part of the tropical heat engine, supplying moisture to the atmosphere and latent heat to North America as it evolves from early spring to early fall [ Wang and Enfield, 2001; Wang et al., 2006]. To our knowledge, few studies exist in the western low latitude Atlantic Ocean that cover multiple glacial to interglacial cycles, and th erefore the response of the Atlantic Warm Pool to orbital insolation is not well understood. In contra st, there are numerous studies of low latitude Atlantic paleoceanog raphy during the last deglaciation [e.g., Curry and Oppo, 1996; Hemming et al., 1998; Arz et al., 1999; Guilderson et al., 2001; Zahn and Stuber, 2002; Lea et al., 2003; Weldeab et al., 2006]. In particular, Atlantic Warm Pool SST during termination I [ Ruhlemann et al., 1999; Flower et al., 2004; Schmidt et al.,

PAGE 8

2 2004] and II [ Schmidt et al., 2004] exhibits an increase prior to changing ice volume, consistent with findings in th e Western Equatorial Pacific [ Lea et al., 2000]. Additionally, SST changes in the Atlantic ma y also be associated with the relative strength of North Atlantic Deep Water ( NADW) production. The rapid cooling of warm, salty surface waters from the tropics and s ubsequent formation of NADW is generally considered to drive the Atlantic thermohaline circulation (THC) [ Rahmstorf, 2006; Kuhlbrodt et al., 2007]. Export of NADW from the northern to southern hemisphere requires an equal volume of water returned through surface circulation, such that a reduction in deep water production would slow surface transport and sequester heat [ Crowley, 1992; Stocker and Johnsen, 2003] and salt in the low latitudes and southern hemisphere [ Dahl et al., 2005; Haarsma et al., 2008]. Recent evidence from the Caribbean Sea supports the notion that orbitalto millennial-scale fluctuations in Atlantic THC influence SST and sea surface salinity (SSS) in the western tropical Atlantic. Deglacial SST fr om the eastern Caribbean Sea [ Ruhlemann et al., 1999] indicates warming coeval w ith excursions to near-glacial conditions (Heinrich Event 1 and the Younger Dr yas) in the North Atlantic and reduced THC [ Boyle and Keigwin, 1987]. In addition, reconstructed Northwest Atlantic SSS variability [ Schmidt et al., 2006] is coincident with Caribbean SSS and in phase with rapid climate change of marine isotope stage (MIS) 3 in Greenland. On the orbital scale, ODP Site 999 (see Figure 1 for location) reveals increased SSS during intervals of reduced NADW production [ Schmidt et al. 2004]. These studies suggest SST and SSS in the western low latitude Atlantic is clos ely connected to the ch anging strength of THC through associated changes in the mean posit ion of the Inter-tropi cal Convergence Zone (ITCZ) and heat and salt transport variability. 1.2 Laurentide ice sheet meltwater Widespread, abrupt climate change manife sted in Greenland air temperature (i.e., Greenland stadials/interstadials or Dansgaard/Oeschger Events [ Dansgaard et al., 1993],

PAGE 9

3 tropical [ Peterson et al., 2000] and extra-tropical hydrology [ Wang et al., 2001], and tropical SST [ Lea et al., 2003] occurred in synchroneity during the last glaciation. Abrupt reorganization in ocean circulat ion, modulated by NADW formation, is the leading hypothesis to explain coeval variab ility in the northern high latitudes and low latitudes [ Broecker, 1985; Broecker et al., 1989; Clark et al., 2001]. Proxy evidence of rapid decreases and resumptions of North Atlantic deep ci rculation during termination I [ Boyle and Keigwin, 1987; McManus et al., 2004] and MIS 3 [ Keigwin and Boyle, 1999] support this hypothesis. Additionally, mode ls of NADW sensitivity to freshwater injection suggest rapidly melting Northern Hemisphere ice sheets are capable of disrupting NADW formation, consequently re ducing THC and triggering a return to glacial conditions in the North Atlantic [ Ganopolski and Ramstorf, 2001]. Figure 1. Location map for ODP Site 625 Figure 1. Map showing the location of ODP site 625 (2849.9 N, 8709.6 W) in the northeastern GOM and core locations for Orca Basin (2656.77 N, 9120.74 W), Pigmy Basin (2711.61 N, 91 24.54 W), ODP 999 (1245 N, 7844 W), and Ceara Rise EW9209-1 (5 N, 43 W).

PAGE 10

4 Melting of the Laurentide ice sheet (L IS) and subsequent production of large amounts of freshwater directed to the North Atlantic could explain rapid changes in THC [ Broecker et al., 1989; Clark et al., 2001]. The Mississippi River system, currently draining the continental U.S. from the Rockie s to the Appalachians, was one of a number of conduits for drainage of the LIS thr oughout the late Pliocene and Pleistocene. Isotopically light glacial wate rs entering the Mississippi River system along the southern margin of the LIS mixed with both Mississi ppi River water and pr ecipitation before injection to the Gulf. GOM se diment records have revealed glacial discharges during the late [ Kennett and Shackleton, 1975; Emiliani et al., 1975; Leventer et al., 1982] and early Pleistocene [ Joyce et al., 1993]. However, variability in the ice sheet geometry (e.g., isostatic rebound and melting along the southe rn margins) affect ed the continental drainage patterns, perhaps directing meltwat er through eastern outle ts, including the St. Lawrence River, to the North Atlantic. In fact, the switching between southern and eastern routed glacial meltwater is thought to trigger swings in North Atlantic climate by affecting meridional heat transport duri ng the last deglaciati on (termination I) [ Broecker et al., 1989] and over MIS 3 [ Clark et al., 2001], which we term the Meltwater Routing Hypothesis. As of yet, it is unclear whether or not the full 6/5 glacial termination (termination II or penultimate deglaciation) as well as smaller glacial-interglacial transitions, such as MIS 5d/5c, 5b/5a, and 4/ 3, were associated with similar meltwater discharge and consequential cl imate change. It is equally unknown how the timing of these events, if they existed, relates to millennial-scale warm intervals known from Greenland and Antarctic ice core records. Phasing of SST and 18O of seawater (interpreted to reflect salinity variability) in the GOM during termination I appears to be asynchronous with Greenland air temperature, and instead reflects similar pacing as Antarctic air temperature [ Flower et al., 2004]. Thus the initial fo rcing to instigate these large meltwater perturbations is still ambiguous Since the timing of southern routing of LIS meltwater has implications for potential disruption of THC during late Pleistocene millennial-scale climate changes, resolving the phasing of regional SST and LIS

PAGE 11

5 meltwater events in relationship to high la titude records will increase understanding of the role of the low and high latitudes in climate change. The northeastern GOM, with relatively hi gh sedimentation rates, provides submillennial to orbital scale records of Atla ntic Warm Pool climate change. Here we report new paired planktic foraminiferal 18O and Mg/Ca data from Northeast GOM ODP Site 625. Results indicate a 3.2-4.4 C glacial-i nterglacial SST change over the last two glacial terminations, with no substantial SST rise preceding global ice volume during the last two glacial terminations. In addi tion, significant SST and SSS changes on the millennial-scale are consistent with modulation by THC variability. 2. Core location and methods 2.1 Core location ODP Site 625 (2849.9N, 8709.6W; see Figure 1) is located in the northeastern GOM, approximately 246 km east of the current position of the Mississippi River delta. The drill site is south of the axis of De Soto Canyon, and receives fine terrigenous sediments from the northern GOM and carbonate sediments from the West Florida slope, in addition to pelagic rain [ Shipboard Scientific Party, 1985]. Two holes drilled in 1985 at Site 625 during the shake down cruise of the JOIDES Resolution recovered over 230 m of sediments in 899-900 m water depth, from which we examine the top 17.81 m. Large aragonitic pteropods and intact spines on indi vidual planktic Foraminifera indicate excellent preservation and very little to no dissolution. 2.2 18O and Mg/Ca analyses Sediment samples were processed base d on the standard procedure for the Paleoceanography, Paleoclimatology, Biogeochemi stry (PPB) Laboratory at the College of Marine Science, University of South Flor ida. Wet sediments were freeze dried to obtain dry bulk weights, washed over a 63 m si eve to remove clay a nd silts, filtered to remove water, dried in a 50 C oven, and we ighed again. When possible, each sample

PAGE 12

6 was picked for ~60 Globigerinoides ruber (white) in the 250-355 m size fraction and split into two aliquots for stable isotope a nd Mg/Ca analyses. The tests (shells) were gently sonicated for ~5 seconds in methanol dried, and weighed in order to assess the effects of post-depositional dissolution. In order to achieve homogenization of each stable isotope sample, we typically pulver ized ~30 individuals (300-400 g) between two glass plates from which we removed ~50g for isotopic analysis. For benthic stable isotope analysis, we pick ed between 5-20 individual Cibicidoides spp Three samples were lacking in Cibicidoides spp. and so Uvigerina spp. was analyzed instead. Benthic samples were prepared as previously described for G. ruber with the exception of crushing and homogenization. Since deep sea benthic Foraminifera secrete thicker tests than G. ruber we were unable to fully pulverize the sample, and instead, crushed the tests open and homogenized. Stable isot ope mass spectrometry was performed at the PPB Laboratory using a Thermo Finnigan Delta Plus XL dual-inlet mass spectrometer with an attached Kiel III carbonate preparation device. We report isotopic data on the Vienna-Pee Dee Belemnite (VPDB) scale calibra ted with the international standard NBS19. Analytical precision for stable isotope measurements is 0.06 ‰ for 18O and 0.04 ‰ for 13C based on over 500 NBS-19 standards ru n during the interval ODP 625 samples were run. Replicate precision based on 48 re plicates and triplicat es is 0.2 ‰ for 18O and 0.1 ‰ for 13C. To bring the benthic stable isotope data closer to equilibrium with ambient seawater we apply the following correction recommended by Duplessy et al. [1984]: Cibicidoides spp. = +0.64 ‰ for 18O The second aliquot of ~30 tests was gently crushed open between glass plates to remove chamber fill and subjected to an extensive cleaning process modified from Barker et al., [2003]. The procedure can be broken into three parts: samples were 1) sonicated in MilliQ water and methanol numerous times to remove fine clays, 2) bathed in 1% buffered peroxide solution within a hot water bath for approximately 10 minutes to remove organic matter, and 3) briefly s onicated in a weak acid leach (0.001 M HNO3) to

PAGE 13

7 remove any adsorbents. To minimize calcium concentra tion effects, we dissolved the calcite in sample weight-det ermined volumes of 0.075 M HNO3 to attain calcium concentrations of ~20 ppm. Samples were analyzed at the PPB Laboratory using a Perkin Elmer Optima 4300 dual-view inductively coupled plasma-optical emission spectrometer (ICP-OES). Elements Mn, Fe, a nd Al were analyzed along with Mg and Ca as a qualitative check on clay contamination. The analytic al precision for this study is <0.28 % root-mean standard deviation, ba sed on a calibrated ICP-OES solution. Replicate precision was 0.25 mmol/mol (n = 37), which translates to 0.76 C error based on the Anand et al. [2003] equation. Raw data for ODP Site 625 can be found in Figure 2. 3. Age Model Constructing a robust age model allows fo r comparison to other climate records of interest. We employed three techniques, whic h include planktic Foraminifera biozones, radiocarbon, and isotope stra tigraphy, to convert depth to calendar years. 3.1 Foraminifera datum levels The reappearance of the Globorotalia menardii complex ( G. menardii, G. tumida ), characterizes the boundary between bioz ones Z and Y, and is a well established [ Ericson and Wollin 1968; Kennett and Huddleston 1972; Kennett et al., 1985; Flower and Kennett 1990] indicator of the beginning of the Holocene in the low latitudes. We used a published radiocarbon age for the Z/Y boundary [ Kennett et al., 1985], which was found at 0.95-1 m below sea floor (mbsf) in ODP Site 625 (Table 2). Globorotalia tumida flexuosa an extinct, warm water Foraminifera species, exists in Pleistocene-age sediments. We us ed the last local occurrence (LLO) of G. flexuosa, found at 13.5-13.55 mbsf (Table 2), reported as 90 ka for th e Western GOM and 75 ka for the Caribbean Sea by Kennett and Huddleston [1972], as an indicator of stage 5 sediments (Table 2).

PAGE 14

8 Figure 2. Raw G. ruber stable isotope and Mg/Ca data for ODP Site 625 versus core depth Figure 2. Raw G. ruber (a) 18OC, (b) Mg/Ca, and (c) 13C data. Blue dots indicate replicat e measurements and pink dots indicate triplicate measurements. G. ruber 18OC(‰VPDB)G. ruber Mg/Ca (mmol/mol)Depth (mbsf) a b cG. ruber 13C (‰VPDB) G. ruber 18OC(‰VPDB)G. ruber Mg/Ca (mmol/mol)Depth (mbsf) a b cG. ruber 13C (‰VPDB)

PAGE 15

9 3.2 Radiocarbon Accelerator mass spectrometry (AMS) da ting of monospecific foraminiferal calcite from 5-9 mbsf of Hole 625 C, perf ormed at the Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratories, provide d age control from early MIS 2 to late MIS 3 (~20-36 ka) (Table 1). We used a 400 year reservoir correction and calibrate radiocarbon year s to age in thousan ds of years B.P. (ka) using a high resolution radiocarbon calibration from Cariaco Basin sediment cores [ Hughen et al., 2006]. Calibration of the Cariaco Basin gr ayscale record to 50 ka was achieved by correlation to an absolute dated (238U/232Th) Hulu Cave speleothem 18O record [ Wang et al., 2001]. Table 1. Radiocarbon Ages for ODP Hole 625 C CAMSa Number Core Depth (m) 14C AMSa Age (ka) 14C Error (y) Calibrated Age (ka) Calibrationc Error (y) Total Error (y) 130330 5.05 18.80 230 20.9 455 509 130331 5.75 23.12 220 26.05 552 594 130332 6.4 20.24 170 22.62 487 515 130333 6.55 23.43 150 26.43 556 575 130334 7.3 17.63 70 19.7b 455 460 130335 8 24.80 130 27.67 582 596 130336 8.2 28.89 220 31.41 654 690 130337 8.4 29.30 230 32.52 676 714 130338 9 32.89 350 36.21 746 824 a All AMS dates were performed at the Center for Accelerator Mass Spectrometry (CAMS), Lawrence Livermore National Laboratory, AMS is accelerator mass spectrometry b Date not included in age model due to stratigraphic inconsistencies. c Calibration error was estimated from error reported in Hughen et al, [2006] which compounds error from Cariaco 14C measurements and calibrating Cariaco grayscale to Hulu Cave 238U/232Th absolute dates.

PAGE 16

10 3.3 Isotope stratigraphy The rhythm of glacial to in terglacial climate change is generally attributed to cyclic modifications in the inte nsity of boreal summer insolation [ Emiliani et al., 1955 ; Hays et al., 1976 ; Imbrie et al., 1984 ; Raymo et al., 2006]. The SPECMAP chronology [ Imbrie et al, 1984] relies on the assumption that globa l ice volume responds to insolation variance associated with changes in the Earth’s orbital geometry (i.e., Milankovitch cycles). We identify isotopic stag e transitions and sub-stages in our G. ruber 18OC record coupled with a short benthic 18OC record from Cibicidoides spp. (Figure 3 and Table 2) and apply age based on co rrelation to the SPECMAP stack [ Imbrie et al., 1984]. a Z/Y boundary age based on radiocarbon age from Kennett et al. [1985] calibrated to calendar years using Hughen et al. [2006]. b this study; not used as age control point c Depth determined from published oxyge n isotope record on ODP Hole 625 B, [ Joyce et al., 1990] d Ages for isotopic stage transitions and sub-stages are from Imbrie et al. [1984]. Table 2. Age control for ODP Site 625 Event Core Depth (m) Age (ka)d Z/Y Boundary 0.95 1 10.65a 2.0 1.15 12 3.0 5.15 24 4.0 11.79 59 5.0 13.2 73.91 LLO G. flexuosa 13.5-13.55 78.6b 5.1 13.8 80 5.2 14.28 87 5.3 14.77 99 5.4 15.11 107 5.5 15.76 122 6.0 16 128 6.4 17.1 151 7.0 19.65c 186

PAGE 17

11 3.4 Depth to age (ka) conversion We sampled Holes 625 B and C at 2-5 cm resolution from 0-17.81 mbsf. We correlated depth between the two holes by comparing whole core measurements of magnetic susceptibility. Prior to splitti ng, each section passed through a 400 mm sensing loop where rapid measurements of magnetic susceptibility were obt ained every 10 cm. Results indicated no significant depth offsets in the 0-20 m interval between Holes 625 B and 625 C [ Shipboard Scientific Party, 1985]. We converted our composite depth for Ho les 625 B and C to age (ka) by applying a weighted curve fit with a 40% smoothing factor (Figure 3). This function fits a curve to the 21 age control points (see Ta bles 1 and 2) using the loca lly weighted least squares regression method. The 40% smooth was chos en to maximize fit and minimize large changes in sedimentation rate. Total error a ssociated with our age model ranges from 509 years on our youngest radiocarbon date to 3000 years based on correlation to SPECMAP. Error for radiocarbon dates was calculated by compounding analytical error from our 14C measurements, analytical error from 14C measurements from Cariaco Basin, and error associated with ca librating Cariaco radiocarbon to Hulu Cave as reported by Hughen et al. [2006]. Error for SPECMAP correlations is based on the most conservative estimate of error reported by Imbrie et al. [1984]. 4. Results and Discussion 4.1 Raw stable isotope and Mg/Ca data Globigerinoides ruber a symbiont-bearing species of planktic Foraminifera, inhabits the upper 50 m of the GOM [ B and Tolderlund, 1971], thrives in oligotrophic to mesotrophic environments [ Zaric et al., 2005], and is abundant in GOM deep sea sediments [ Orr 1969; Kennett and Huddleston 1972; Brunner, 1979; Kennett et al., 1985]. Well established relationships of stab le isotope and geochemical composition of G. ruber calcite to physical conditions make it an ideal proxy for sea surface conditions in the GOM.

PAGE 18

12 Figure 3. Age model for ODP Site 625 Figure 3. ODP site 625 raw G. ruber (black) and benthic (red) foraminiferal 18OC plotted with age model. (a) Age model for spliced Holes 625 B and 625C over MIS 1-6 with Foraminifera datum levels (green), 14C AMS dates (red) (calibrated to calendar years using a Cariaco Basin radiocarbon calibration [ Hughen et al., 2006]), and marine isotope stratig raphic points (blue) based on SPECMAP [ Imbrie et al., 1984]. Depth was converted to age (ka) by applying a 40% weighted smooth to 21 age control points (see Tables 1 and 2). (b) Raw G. ruber 18OC and (c) benthic 18O of Cibicidoides spp corrected +0.64 ‰ [ Duplessy et al., 1984] to compare with three Uvigerina spp. values versus core depth. Age for the Z/Y boundary was de termined from core depth in 625 and radiocarbon age from Kennett et al. [1985] calibrated to age (ka) using Hughen et al. [2006]. The last local occurrence (LLO) of G. flexuosa in the GOM [our age model suggests an age of 78.6 ka for the LLO G. flexuosa ] was not used in age model. No te that error bars are within symbols for radiocarbon. G. ruber 18OC (‰VPDB)Depth (mbsf)Benthic 18OC (‰VPDB)Age (ka) a c bG. ruber 18OC (‰VPDB)Depth (mbsf)Benthic 18OC (‰VPDB)Age (ka) a c b

PAGE 19

13 The ratio of magnesium to calcium in planktic foraminiferal calcite is thermodynamically controlled [ Chave, 1950; Savin et al., 1973; Nurnberg et al., 1996; Hastings et al., 1998; Lea et al., 1999] and therefore provides an estimate of seawater temperature at calcification depth. There have been a number of studies conducted to calibrate planktic foraminiferal Mg/Ca values to temperature that involve a variety of methods, including direct measurements of Mg -calcite growth kinetics with varying temperature [ Elderfield and Ganssen 2000], controlled culture experiments [ Lea et al., 1999], core-top Mg/Ca measurements coupled with annual SST [ Dekens et al., 2002], and Mg/Ca and 18O of the calcite ( 18OC) measurements from sediment-traps for calibration to temperature at estimated calcification depth [ Anand et al., 2003]. All experiments revealed an exponential relati onship where Mg/Ca values increase by ~8-10 % per 1 C increase in temperature. The Anand et al. [2003] calibration: Mg/Ca = 0.449(0.09 • SST (C)) (1) was used in this study because: 1) numer ous studies have confirmed the exponential constant of 0.09, 2) it represents a species -specific equation for the 250-350 m size fraction, 3) application of this equation with modern Pigmy Basin (see Figure 1) core top values of G. ruber Mg/Ca, 4.43 0.03 mmol/mol [ Richey et al., 2007], yields a SST of 25.4 C, equivalent to the modern a nnual average SST for the northern GOM [ NOAA/OAR/ESRL, 2008], and 4) to maintain intralaboratory consistency in SST estimation on G. ruber (white) from GOM sediment samples [e.g., LoDico et al., 2006; Richey et al., 2007]. A number of studies report a significant change in the magnesium content of foraminiferal calcite with large change s in salinity. Culture experiments by Nurnberg et al. [1996] showed 110 % increase in shell magne sium with a 10 psu increase in salinity. Nurnberg et al. [1996] speculated that the increased uptake in magnesium associated with higher salinities is re lated to known increased feedi ng and metabolic activity amongst foraminifers growing in more saline waters, as reported by Hemleben et al. [1987]. More recently, Ferguson et al. [2008] examined Mediterranean Sea core top Foraminifera over

PAGE 20

14 a salinity range of 36-40 psu and found Mg/C a values exceeded predicted values based on calcification temperatures. Ferguson et al. [2008] concluded salinity accounts for ~30 % change in Mg per psu change within the Mediterranean Se a. They also pointed out the effect of higher salinity during the LGM (from the 120 m drop in sea level globally) on estimating sea surface temperatures in the highly evaporative low latitudes. An examination of modern G. ruber Mg/Ca values from 78 coretops in a north-south transect of the Atlantic Ocean [ Bice et al., 2006; deMenocal et al., 2007] lends further support to these findings. Bice et al. [2006] and deMenocal et al. [2007] found significant correlation between G. ruber individuals with “excess Mg” in areas of high salinity, such as the subtropical gyres. The effects of salinity on foraminiferal Mg/Ca may be minimal in the GOM for three reasons: 1) predicted Mg/Ca values based on calcificat ion temperature, assumed to reflect the annual average GOM SST, are not exceeded on modern core top Mg/Ca measurements of G. ruber [ Richey et al., 2007]; 2) GOM Mg/Ca-SST reconstructions for the LGM [ Flower et al. 2004; Nurnberg et al., in press; this study ] are consistent with estimates based on Uk 37 [ Jasper and Gagosian, 1989 ] in that both techniques suggest greater LGM-Holocene SST than predicted by CLIMAP [ CLIMAP Project Members, 1976]. If higher salinity during the LGM acc ounted for 20-50% of the Mg incorporated in G. ruber calcite, our SST estimates would unde restimate LGM cooling and increase the inconsistency between faunal and geoche mical SST proxies. Finally, 3) De Soto Canyon SSS ranges between 32.6-35.6 psu and averages 35.5 psu annually [ Levitus and Boyer 1994], well below salinities (> 36 psu) shown to affect Mg content in foraminiferal calcite [ Nurnberg et al. 1996; Bice et al., 2006; deMenocal et al., 2007; Ferguson et al. 2008]. The G. ruber Mg/Ca values for ODP Site 625 range from 2.4 to 5.3 mmol/mol with an average of 3.65 mmol/mol (Figure 2), which converts to a te mperature range of ~19-27 C, and a mean temperature of 23.3 C, using a species-specific equation [ Anand et al., 2003] (Figure 4). To assess changes in northern GOM SST over the last 155 ky we

PAGE 21

15 compare our Mg/Ca values and SST estimates to the modern (0 yr BP core-top) Mg/Ca ratio for G. ruber (white), 4.43 0.03 mmol/mol, and associated modern SST, 25.4 C, from nearby Pigmy Basin [ Richey et al., 2007] (see Figure 1 for location and Table 3). Table 3. Average SST at ODP Site 625 MIS Mg/Ca (mmol/mol) n SST (C) Modern ( C) 1 3.79 0.62 27 23.7 2.6 -1.7 2.6 2 2.97 0.24 27 21.0 1.3 -4.4 1.3 3 3.41 0.31 105 22.5 1.4 -2.9 1.4 4 3.56 0.22 28 23.0 0.97 -2.4 0.97 5a 3.83 0.3 14 23.8 1.2 -1.6 1.2 5b 3.89 0.17 11 24.0 0.7 -1.4 0.7 5c 4.22 0.27 12 24.9 1.0 -0.5 1.0 5d 4.22 0.28 11 24.9 1.0 -0.5 1.0 5e 4.94 0.3 11 26.6 0.96 1.2 0.96 6 3.66 0.23 50 23.3 0.99 -2.1 0.99 a Marine isotope stage boundaries determined from G. ruber 18OC, supplemented by Mg/Ca-SST and benthic 18O b Values from clear transiti ons not included in average c Sea surface temperature (SST) from converting G. ruber Mg/Ca values using Mg/Ca = 0.449[0.09 • SST (C)] [ Anand et al., 2003] d Change in SST from modern value, 25.4 C, determined from Pigmy Basin core-top Mg/Ca [ Richey et al., 2007] The penultimate glaciation (MIS 6) and p eak interglacial (MIS 5e) had average Mg/Ca ratios of 3.6 0.23mmol/mol (0.77 mm ol/mol lower and 2.1 C colder than modern SST) and 4.94 0.3 mmol/mol (0.51 mmo l/mol higher and 1.2 C warmer than modern SST), respectively. The LGM and early to middle Holocene had average Mg/Ca ratios of 2.97 0.24 mmol/mol (1.46 mmo l/mol lower and 4.4 C colder than modern SST) and 3.79 0.62 mmol/mol (0.64 mmo l/mol lower and 1.7 C colder than modern SST), respectively. The last inte rglacial in the northern GOM peaked around 116-122 ka and was a substantial 1.2 0.96 C warmer than modern SST. The penultimate glaciation average Mg/Ca va lue is 0.69 0.23 mmol/mol higher than the average LGM Mg/Ca value, which indicates the penultimate glaciation was 2.3 1.63 C warmer than the LGM. The finding of wa rmer Northeast GOM conditions during MIS 6

PAGE 22

16 than MIS 2 is similar to a Mg/Ca record generated on another De Soto Canyon core, MD02-2575, extracted nearby (29N 8707.13W) [ Nurnberg et al., in press]. G. ruber 18OC values range from 1.28 to -2.6 ‰ VPDB with an average of -0.14 ‰. Glacial to interglacial 18OC amplitude is 2.7 and 3 ‰ for terminations I and II respectively. Global sea level may account for ~1‰ [ Adkins et al., 2002] and glacial to interglacial temperature change accounts for 0.77 ‰ [3.7 C (0.208 ‰ T C)] for termination I and 0.69 ‰ [3.3 C (0.208 ‰ T C )] for termination II. Therefore, the residual 18OC change, 0.93 ‰ and 1.31 ‰ respectively, reflects changes in SSS. 4.2 Northeast GOM SSS variability The use of Mg/Ca as an independent proxy for SST paired with 18OC allows us to remove the effects of temperature on the 18OC and isolate the 18O composition of the seawater in which the Foraminifera calcified (denoted 18OSW). Estimating 18OSW requires an established relati onship between temperature and 18OC for G. ruber Development of this relationship for foramini feral calcite involved culturing experiments in high-light and low-light conditions on Orbulina universa a species of subtropical, symbiont-bearing planktic Foraminifera [ Bemis et al., 1998]. Bemis et al. [1998] quantify the temperature effect on 18OC under high-light conditions: T (C) = 14.9 – 4.8 ( 18OC 18OSW) (2) Application of this equation to sediment-trap samples of G. ruber in the Santa Barbara Basin, CA shows good agreement with observed SST [ Thunell et al., 1999]. Therefore, we applied this equation to paired measurements of G. ruber Mg/Ca and 18OC from the same samples in ODP 625 to yield the 18OSW. We add 0.27 ‰ [ Hut, 1987] to convert 18OSW to Vienna-Standard Mean Ocean Wa ter (VSMOW) scale (Figure 4). On timescales greater than 10 ka, variability in 18OSW contains the effects of ice volume, which alters the 18O of seawater due to the pref erential removal and storage of

PAGE 23

17 the lighter isotope of oxygen (16O) in ice during glacials, le aving the oceans enriched in the heavier isotope (18O). For example, during the last glacial maximum (LGM), global sea level, known from coral records, was ~120 m below pres ent, effectively increasing the mean isotopic value of global seawater by ~1 ‰ [ Fairbanks, 1989; Bard et al., 1996]. LGM 18OSW was determined from measurements of pore water samples from LGM-age sediments as 0.99 0.17 ‰ VSMOW, and a relationship between 18OSW and sea level change of 0.083 ‰ 10 m-1 was established [ Adkins et al., 2002]: 0.99 0.17 ‰ 120 m = 0.0083 0.0014 ‰ per 1 m Sea level reconstruction is complicated by glacio-hydro-isostatic changes and tectonic uplift or subsidence. Cores re trieved from submerged coral reefs of Acropora palmata and other shallow water species, have the advantage of absolute dating (238U/232Th) and high resolution, yet only exis t for the last termination [e.g., Fairbanks 1989; Bard et al., 1996] and late MIS 3 [ Peltier and Fairbanks, 2006]. Coral terraces, on the other hand, provide longer re cords of sea level variability based on a few isolated high stands over the last 200 ka [ Chappell and Shackleton 1986; Chappell et al., 1996]. Sea level records determined from deep benthic foraminiferal 18OC have the advantage of supplying a continuous, high reso lution record. However, changes in deep sea temperature, salinity, and temporal vari ability in the isotopic composition of ice sheets compromise the sea level record embedded in the 18O of benthic foraminiferal calcite. To remove the effects of regi onal hydrologic variability and deep water temperature and salinity change, records from various sites are stacked and smoothed. In this study we used the highest resolu tion sea level record available for our time period (0-155 ka), which is developed from a benthic Foraminifera st ack record tuned to orbital insolation [ Waelbroeck et al., 2002]. We convert to 18OSW equivalent by applying a relationship of 0.083 ‰ pe r 10 m change in sea level [ Adkins et al., 2002], add 0.27 ‰ [ Hut, 1987] to convert VPDB to VSMOW, and denote ice volume corrected 18OSW as 18OGOM.

PAGE 24

18 Figure 4. Paired 18OC and Mg/Ca SST data, calculated 18OSW versus age (ka) Figure 4. Paired G. ruber (a) 18OC and (b) Mg/Ca-sea surface temp erature (SST) from converting G. ruber Mg/Ca values using Mg/Ca = 0.449[0.09 • SST (C)] [ Anand et al., 2003] and (c) 18OSW calculated from paired 18OC and Mg/Ca-SST using SST (C) = 14.9– 4.8 ( 18OC 18OSW [ Bemis et al., 1998] over the last 155 ky. Blue bars refer to glacial stages 2, 4, and 6. Variability in 18OGOM can arise through three different mechanisms. First, enhanced precipitation or evaporation will preferentially decrease or increase 18OGOM, 18Osw(‰VSMOW) 18Oc(‰VPDB)Mg/Ca-SST (C)Age (ka) a c b 18Osw(‰VSMOW) 18Oc(‰VPDB)Mg/Ca-SST (C)Age (ka) a c b

PAGE 25

19 respectively. Second, the amount of Mississippi Rive r discharge to our core site, which may be affected by the amount of precipita tion on the continent, will influence 18OGOM. For example, changes in the position of the Mississippi River delta may have influenced freshwater input to the northeast GOM [ Tripsanas et al., 2007]. Third, freshwater associated with melting of the LIS has been shown to influence the 18OGOM [ Kennett and Shackleton, 1975; Emiliani et al., 1975; Leventer et al., 1982; Joyce et al., 1993; Flower et al., 2004; Hill et al., 2006]. Isolating the change in the 18O of GOM surface water allows us to assess variability in SSS, depending on changing cont ributions from isotopi cally different endmembers. We estimated SSS changes by esta blishing a relationship between salinity and the 18OGOM (Figure 5). This is achieved by cr eating a simple mixing model between the modern end-member for our core site (S = 35.5 psu, [ NOAA/OAR/ESRL, 2008]; 18OSW = 1.2 ‰ VSMOW, [ Fairbanks et al., 1992]) and a low salinity e nd-member. We use three possible low salinity end-members, 3.5 ‰ for modern GOM precipitation [ Bowen and Revenaugh, 2003], -7 ‰ for the average isotopi c composition for Mississippi River [ Ortner et al., 1995], and -27.5 ‰ reflecting the averag e isotopic composition of the LIS over the last 120 ky [ Sima et al., 2006]. As evident from the box model, the more negative low salinity end-member (i.e., -27.5 ‰) yields the smallest change in salinity per unit change in 18OGOM. It should be noted that the modern salinity range for the northeast GOM is ~3 psu [ NOAA/OAR/ESRL, 2008]. Changes in SSS, defined by deviations fr om modern mean annual salinity for our core site (35.5 psu), range from as large as -6 psu and +10 psu to -1 psu and +1.5 psu, depending on the low salinity end-member (Fi gure 6). Glacial stages 4 and 6, as well as sub-stages 5a and 5c are sa ltier than present. Quantif ying the magnitude of salinity change, however, is difficult owning to the uncertainty in the relationship between salinity and 18OGOM in the past. We are unable d econvolve the different contributions of Mississippi River, precipitation, glaci al meltwater, and possible salt accumulation associated with reduced THC. Furthermore, the low-salinity end-member contributing to

PAGE 26

20 18OGOM probably changes with time, such that our calculated SSS reflects the total range of possible SSS changes. However, the unreaso nable salinity changes associated with the -3.5 ‰ end-member suggests the change 18OGOM are not likely due to precipitation variability. On the other hand, the 2-4 psu ch ange during colder periods implied by the -7 ‰ end-member is of the same order of ma gnitude of SSS change predicted for the GOM by a model of reduced NADW volume [ Haarsma et al., 2008]. Therefore, it is likely that variability in 18OGOM exceeding 0.5 ‰ (i.e., greater than total precision) is a result of some combination of glacial freshwater and increased Miss issippi River influx. Figure 5. 18OGOM and sea surface salinity mixing model for the Gulf of Mexico Figure 5. Relationship between the oxygen isotopic composition and salinity of GOM surface waters. Each line represents a conservative mixing scenario between different low salinity (i.e ., fresh) isotopic end-members and a high salinity end-member, 18OSW = 1.2 ‰ VSMOW [ Fairbanks et al., 1992] and 35.5 psu [ NOAA/OAR/ESRL, 2008]. Low salinity end-members are: (1) -3.5 ‰ for GOM precipitation [ Bowen and Revenaugh, 2003], (2) -7 ‰ for the average isotopic composition for MR [ Ortner et al., 1995], (3) 27.5 ‰ represents the average modeled isotopic composition of LIS over the last 120 ky [ Sima et al., 2006]. Note the more negative the low salinity isotopic end-member, the smalle r the change in salinity per unit change in 18OGOM. 18OGOM(‰VSMOW)Sea surface salinity (psu) 18OGOM(‰VSMOW)Sea surface salinity (psu)

PAGE 27

21 Estimating SSS during warmings is compli cated by potential meltwater influx to the GOM. The penultimate deglaciation and 4/3 transition are characterized by large changes in SSS, from saltier glacial conditions to near modern conditions in less than 2 thousand years (Figure 6). The most c onservative estimate of salinity suggests substantial glacial meltwater influence, a deduction s upported by models of reduced Atlantic THC. Haarsma et al. [2008] conclude a conseque ntial 1 psu increase in GOM SSS during reduced THC strength. Thus, change s in SSS greater than 1 psu and cannot be explained by a decrease in NADW produc tion, suggest a change in source water contribution (i.e., meltwater). Overall, this study, while unable to quantify the magnitude of salinity change, offers a possible range of salinity change. Proxy records from other archives sensitive to salinity variability ar e needed to fully characterize the changing contribution of each low salinity end-member through time. The ratio of barium to calcium in planktic Foraminifera shows a positive relationship to riverine influx [Weldeab et al., 2007] Ba/Ca analysis of G. ruber (white) from ODP 625 in conjunction with 18OGOM could provide a means to isolate salinity change. 4.3 A comparison of records of Atlantic Wa rm Pool SSS: Northeast GOM versus the Western Caribbean Sea Using the same multi-proxy approach (paired G. ruber Mg/Ca and 18OC), a similar 18OSW record was generated from ODP Site 999 in the Western Caribbean Sea. This study also used the SPECMAP chronology to constrain age beyond the range of radiocarbon. We plotted our 18O data with a 2% weighted smooth which allows for comparison on a similar temporal resolution (> 1000 years per sample). A comparison of Caribbean and GOM 18O over the past 150 ky (Figure 7) reveals some similarity, with both records exhibiting saltie r conditions (increased 18O values) during glacial stages 4 and 6. These records are consonant with a bui ld-up of salt in the Atlantic Warm Pool associated with NADW volume reduction during glacial stages 4 and 6, but not MIS 2. Furthermore, there are several in teresting differences between the 18OGOM and 18OCaribbean records. The 18OGOM record exhibits higher variab ility and amplitude

PAGE 28

22 Figure 6. G. ruber Mg/Ca-SST, 18OGOM, and estimated change in Gulf of Mexico sea surface salinity versus age (ka) Figure 6. Sea surface temperature and salinity in the northeastern Gulf of Mexico during the past 155 ky. Blue bars refer to glacial stages 2, 4, and 6. (a) ODP s ite 625 Mg/Ca-SST (C) record, (b) 18OGOM calculated by subtracting global sea level change [ Waelbroeck et al., 2002] from 18OSW (gray line) shown with 2 % weighted curve, (c) estimated change in salinity (psu) calculated by subtracting modern annual average salinity of 35.5 psu [ NOAA/OAR/ESRL, 2008] from estimated salinity based on three isotopically different end-members (see Figure 5) versus age (ka). Glacial stages 4 and 6 are characterized by higher 18OGOM, which translates to an increase SSS. Estimating the magnitude of SSS is difficult because the contribution from each low salinity end-member is likely to have changed over time. Estimated SSS is unreasonably high for precipitation (>6 psu, black line) to account for the 18OGOM over the last 155 ky. Estimated SSS indicated by MR (2-4 psu, dash ed pink) during glacials is the same order of magnitude predicted by a general circulation model of reduced THC [ Haarsma et al., 2008]. The most conservative estimated salinity changes (~1 psu, dark blue) call on significant meltwater contribution (-27. 5 ‰ end-member) to account for the decreases in 18OGOM. Estimated SSS (psu)Age (ka)18OGOM(‰VSMOW)Mg/Ca-SST (C)a b cc b a d e Estimated SSS (psu)Age (ka)18OGOM(‰VSMOW)Mg/Ca-SST (C)a b cc b a d e

PAGE 29

23 than 18OCaribbean. Additionally, 18OGOM shows both positive and negative excursions from modern conditions, whereas the 18OCaribbean record shows mostly positive excursions from modern c onditions. In particular, 18OGOM shows large negative excursions during stage 2 not seen in the 18OCaribbean record. The increase in variability and amplitude is due in part to several f actors, including sampling resolution, variability in the mean position of the ITCZ, and freshwater influence on the 18OSW. The average sedimentation rate at OD P Site 625 is 17.8 cm/ky over the last 155 ky (see Appendix A, Table 1); in contrast, the average rate at ODP Site 999 is 4 cm/ky [ Schmidt et al., 2004]. The higher sedimentation rate in ODP Site 625 allowed for a mean sampling resolution of 400 years. ODP Site 999 provides a 1.1 ky sampling resolution [ Schmidt et al., 2004]. Modern hydrology of the northeastern GOM, unlike the Western Caribbean Sea, is comparatively unaffected by the ITCZ. Recent studies [e.g. Poore et al., 2003; Poore et al., 2004; Nurnberg et al., in press] have suggested pa st fluctuations in the mean position of the ITCZ affected Loop Current va riability in the GOM. However, even at the northernmost position, during the boreal summer and early fall, the ITCZ never reaches further than 10 N [ Waliser and Gautier, 1993]. Additionally, the fact that Loop Current intrusion to the northern GOM can occur at any season [ Muller-Karger, 1991], and has periods of 6-17 months, suggests rela tively little influen ce from the seasonally migrating ITCZ [ Molinari, 1980]. Therefore we argue th at the GOM receives minimal influence from seasonal and longer timescale ch anges in the mean position of the ITCZ. The close proximity of ODP Site 625 to th e Mississippi River delta, and thus to the influence of freshwater influx, also affects the variability in 18OGOM. LIS melting produced isotopically depleted glacial meltw ater, sometimes routed to the GOM, which could have substant ially altered the 18OGOM without effecting a large salinity change (see Figures 5 and 6). Additiona lly, intervals of reduced NAD W production associated with an influx of glacial meltwater, known to ha ve occurred during the last glaciation [ Hill et al., 2006], could be elusive. An increase in salinity from reduced NADW production

PAGE 30

24 would increase the 18OGOM, whereas freshwater from glaci al runoff would deliver very depleted waters to the GOM and lower the 18OGOM. Figure 7. Comparison of the difference in 18O composition of seawater at ODP 625 in the Gulf of Mexico ( 18OGOM) and ODP 999 in the Caribbean ( 18OCaribbean) relative to modern conditi ons versus age (ka) Figure 7. Comparison of 18OGOM relative to modern conditions in the Gulf of Mexico from ODP 625 ( 18OGOM) and in the Caribbean from ODP 999 ( 18OCaribbean) over the past 155 ky. (a) ODP 625 18OGOM (gray line with symbols) shown with 2 % weighted smooth (dark black) was calculated by subtracting the modern GOM 18OSW value, 1.2 ‰ [ Fairbanks et al., 1992] from 18OGOM. (b) ODP 999 18OCaribbean was calculated by subtracting the modern Caribbean 18OSW (0.8 ‰) from 18OCaribbean [ Schmidt et al., 2004]. Negative excursions from modern indicate fresher conditions (shaded blue) an d positive excursions from modern indicate saltier conditions (shaded red). Both records show increased 18O of seawater relative to modern during glacial stages 4 and 6, but different signals during MIS 2-3. 18OGOM (‰VSMOW) 18OCaribbean (‰VSMOW)Age (ka) a b 18OGOM (‰VSMOW) 18OCaribbean (‰VSMOW)Age (ka) a b

PAGE 31

25 4.4 Connection between GOM SST and SSS and THC variability We compare our GOM SSS and SST reco rds to a benthic foraminiferal 13C record from Site EW9209-1 (5N, 43W, in 4056 m water depth) on Ceara Rise over the last 155 ky [ Curry and Oppo, 1996] (Figure 8). Site EW9209 -1 on Ceara Rise is ideally situated to record changes in the relative strength of NADW and Southern Ocean sources. Correlation of our GOM proxy records to Atlant ic deep water variability was achieved by application of the same chronology to cons train age beyond radiocarbon. To verify the correlation we compared our G. ruber 18OC to the benthic 18OC record from Ceara Rise (not shown). Results indicate synchronous variability over the orbital scale. The increase in 18OGOM during glacial stages 4 and 6, i ndicating saltier c onditions, appears concurrent with a decrease in benthic 13C, which indicates reduced NADW volume. In addition, millennial scale incr eases in SSS and SST occur after a weakening in NADW during MIS 5c, 5a, 4, and late stage 3. The lag in SST and 18OGOM response to NADW reduction ranges from 2.5-4.5 ky during MIS 5 and from 1.6-3 ky during MIS 4 and early MIS 3. These results extend the findi ngs in the Western Caribbean Sea [ Schmidt et al., 2004] from the orbital to the s ub-millennial scale, such that a sequestration of both heat and salt follows a reduction in NADW strengt h in five intervals from ca. 115-50 ka. Results from proxy records of surface ocean circulation, and therefore heat and salt transport, provide an estimate of reduced Gulf St ream rates during the LGM [ LynchStieglitz et al., 1999; LeGrande and Lynch-Stieglitz, 2007] and support the evidence of saltier conditions in the low la titude Atlantic from reduced THC. In addition, there is evidence of Western North Atla ntic SST and SSS variability concurrent with GS and GIS [ Vautravers et al., 2004; Schmidt et al., 2006], which suggests heat and salt transport is connected to changes in Greenland air temperat ure. Evidence of saltier conditions during glacial stages 4 and 6, as well as MIS substa ges 5a and 5c in the GOM (Figure 8) is consistent with increased Western North At lantic SSS during GS. Additionally, tropical to subtropical Western Atlantic SSS may act as a negative feedback during reduced THC

PAGE 32

26 by pre-conditioning surface water density fo r a rapid resumption of NADW formation after a cool event [ Latif et al., 2000; Schmidt et al., 2004]. 4.5 Meltwater Routing Hypothesis We test the Meltwater Routing Hypothesi s discussed in section 1.2 by comparing 18OGOM to ice core records of air temper ature from the Eastern Dronning Maud Land (EDML) in Antarctica [ EPICA Project Members, 2006] and North GRIP (NGRIP) in Greenland [ NGRIP Project Members, 2004] (Figure 10). The meltwater routing hypothesis predicts meltwater discharge to the GOM during Greenland interstadials [ Clark et al., 2001]. In order to determine periods of freshwater discharge to the GOM we set a threshold of 0.2 ‰, corres ponding to modern seasonal oxygen isotopic variability in northern GOM surface water, and evaluate the larger negative and positive excursions accordingly. Freshwater even ts are defined as intervals where the 18OGOM reaches 0.2 ‰ lower than modern (1 ‰ in 18OGOM and -0.2 ‰ in 18OGOM), and persists for 1 ky. We find three freshwater events (I-III) (F igure 9) from 12-49 ka that do not match the eighteen Greenland interstadi als in this interval in opposition to a simple version of the Meltwater Routing Hypothesis [ Clark et al., 2001]. The largest freshwater event, comprising IIa and IIb, occurred at the onset of glacial stage 2 at 23 ka, and appears to coincide with Heinrich Event 2 and Antarctic isotope maximum 2. This result is similar to Hill et al. [2006] where a major GOM freshwater event was coincident with Heinrich Event 4 and Antarctic isotope maximum 8 (als o known as Antarctic Warming 1, or A1). In addition, the corresponding salinity change associated with freshw ater event IIa IIb, ranges from 2-8 psu (depending on the low sali nity end-member) (Figure 6). Considering the regional climatology and also long durati on of the excursion (>3 ky), a change of 8 psu is unlikely due to the unreasonable amount of precipitation require d In contrast, if glacial freshwater contributed 100% of the is otopic change, a more conservative change in salinity of ~1 psu is estimated. Ov erall, our results are consistent with Hill et al.

PAGE 33

27 Figure 8. ODP 625 18OGOM and Mg/Ca-SST compared to a proxy record of the relative st rength of North Atlantic Deep Water (NADW) production from EW9209-1 on Ceara Rise. Mg/Ca-SST from converting G. ruber Mg/Ca values using Mg/Ca = 0.449[0.09 • SST (C)] [ Anand et al., 2003] (red) shown with 2 % weighted smooth (heavy red); 18OGOM is the isotopic composition of Gulf of Mexico seawater calculated by subtracting global sea level change [ Waelbroeck et al., 2002] from 18OSW (blue) shown with 2 % weighted curve (heavy blue); benthic 13C record from Ceara Rise (gray) shown with 1 % weighted smooth (heavy black). Yellow bars indicate decreases in 13CBenthic and increases in SST /SSS at ODP 625. Green bar indicates a climate reversal during te rmination II (see Figure 11). Triangles on the top panel indicate intervals with rad iocarbon age control. Mg/Ca-SST (C) CearaRise 13Cbenthic(‰VPDB)Age (ka)18OGOM (‰VSMOW)SST 18OGOMEW9209-1 Mg/Ca-SST (C) CearaRise 13Cbenthic(‰VPDB)Age (ka)18OGOM (‰VSMOW)SST 18OGOMEW9209-1 Fi g ure 8. ODP site 625 18O GOM and M g /Ca-SST com p ared to Ceara Rise benthic 13C versus a g e ( ka )

PAGE 34

28 [2006] which suggested summer melting of th e LIS southern margin occurred during warmings in Antarctica. The potential effects of salinity on Mg /Ca-based temperature reconstructions (e.g., Ferguson et al., 2008; Bice et al., 2007; deMenocal et al., 2007) suggests we may be overestimating the effects of temperature on 18OC and therefore the amplitude of 18OGOM change may be underestimated. Howeve r, if large salinity changes account for higher Mg content and yields an overes timated SST, the negative excursions in 18OGOM at the onset of stage 2 (fresh water event IIa IIb) would incr ease in amplitude, yielding an even more unreasonable decrease in SSS. Overall, our study may lack the tempor al resolution needed to capture all meltwater events in the GOM. Additionally, mo dern dispersal of Missi ssippi River in the GOM, while occasionally flowing eastward, predominately spreads westward [ MullerKarger et al., 1991], and meltwater events recorded to the southwest of the Mississippi River delta in Orca Basin, may be absent in De Soto Canyon [ Nurnberg et al., in press]. Nevertheless, we record a total of five freshwater events ove r the last 155 ky at 124.2119.7 ka, 110-106.7 ka, 38-36.5 ka, 26.3-23.8 ka, and 15.2-12.8 ka. 4.6 Northeastern GOM paleo-productivity The 13C isotopic composition of planktic foraminiferal calcite is controlled by the 13C composition of ambient dissolved inorgani c carbon (DIC), whic h is affected by a combination of surface water productivity, Mississippi River discharge, dissolved CO2 concentration, the amount of symbiont photosynthesis [ Jrgensen et al., 1985; Spero and Williams, 1988; Spero, 1998; Khler-Rink and Khl, 2005], and seawater pH [ Spero, 1998; McConnaughey, 2003]. In addition, on the glacial to interglacial timescale, the isotopic composition of surface water DIC is al so influenced by changes in deep water mass properties [ Spero and Lea, 2002]. Quantifying the contribution of symbiont photosynthesis is difficult because the major factors that contribute to photosynthetic rates are beyond the scope of this study (e.g., irradiance leve ls). Therefore we do not

PAGE 35

29 Figure 9. ODP Site 625 18OGOM and Mg/Ca-SST compared to NGRIP and EDML versus age (ka) Figure 9. Comparison of ODP Site 625 18OGOM and SST with ice core records over the last 60 ky. Numbers refer to Greenland interstadials (GIS) and Antarctic isotope ma ximums (AIM) (previously known as Antarctic warmings). Roman numerals refer to freshwater events (FE) in the northeast GOM. (a) 18Oice from North Greenland Ice core Project (NGRIP) [ NGRIP Project members, 2004]. (b) 18OGOM calculated by subtracting the modern oxy gen isotopic seawater value, 1.2 ‰ [ Fairbanks et al., 1992] from 18OGOM; FE (shaded green) are defined as intervals when the 18OGOM reach 0.2 ‰ lower than modern conditions and persist for 1 ky or more. (c) Sea surface te mperature (SST) from converting G. ruber Mg/Ca values using Mg/Ca = 0.449[0.09 • SST (C)] [ Anand et al., 2003] (d) 18Oice from EPICA Dronning Maud Land (EDML) [EPICA Project members 2006]. Gray bars refer to Heinrich Events (HE 1-5). There are 3 freshwater events in the northeast GOM that do not match the 18 GIS recorded in NGRIP A major freshwat er event, FE IIa IIb, occurred during HE 2 and may be associated with AIM 2. 18OGOM (‰VSMOW)Age (ka)Mg/Ca-SST (C)EDML 18Oice (‰VSMOW)a c bNGRIP 18Oice (‰VSMOW)d 18OGOM (‰VSMOW)Age (ka)Mg/Ca-SST (C)EDML 18Oice (‰VSMOW)a c bNGRIP 18Oice (‰VSMOW)d

PAGE 36

30 consider symbiont photosynthesis wh en interpreting large shifts in G. ruber 13C. To evaluate the controls on De Soto Canyon 13C, we compare the G. ruber planktic 13C record and percent coarse fraction data (or amount of material gr eater than 63 m, the majority of which is planktic foraminifera l calcite) over the last 155 ky (Figure 10). The G. ruber 13C record over the last 155 thous and years reveals orbital to millennial scale changes and has an amplitude of 1.5 ‰. The >1 ‰ difference between interglacial stages 1 and 5 a nd glacial stages 2, 4 and 6, coupl ed with coarse fraction data suggests greater productivity during warm peri ods. There is a prominent decrease in values of nearly 1 ‰ during tr ansition from MIS 5 to 4. The covariance in percent coarse fraction and decrease in 13C values during the 5/4 tran sition supports the notion of reduced productivity. In contra st, glacial stage 6 shows lower 13C values and higher percent coarse fraction compared to the reve rsal during relatively warmer stage 3. While a first order relationship suggests a positive relationship between percent coarse fraction and G. ruber 13C, there is significant sub-millennial to millennial scale variability in G. ruber 13C not reflected in percent coarse fractio n that requires further explanation. Influx of Mississippi Rive r water influences the 13CDIC by introducing freshwater with high concentrations of 12C from soil [ Tan, 1989]. Previous work has shown covariance in lower 13C values recorded in G. ruber and greater abundance of angular siliciclastic grains, indicati ng periodic Mississippi River flooding during the middle to late Holocene [ Brown et al., 1999]. Additionally, surface water enrichment in 12C could have resulted from glacial low-stands in sea le vel and a more direct source of terrestrial sediments from Mississippi Rive r input to the north east GOM. In fact, the increase in the amount of silt and clay during the LGM is apparent from core photos at Site 625 [ Shipboard Scientific Party, 1985] and lower percent coarse fraction. A comparison of 18OGOM, reflecting the difference in regional hydrologic variabili ty from modern conditions, and G. ruber 13C could potentially offer evidence of increased Mississippi River runoff.

PAGE 37

31 Figure 10. 13C composition of G. ruber and percent coarse fraction in ODP 625 versus age (ka) Figure 10. Raw G. ruber (a) 13C (gray) shown with 2 % weighted curve (heavy bl ack) and (b) percent coar se fraction in ODP 625 over the last 155 ky. The dashed line indicates samples pr ocessed in another lab and mi ssing from our dataset. Modern G. ruber 13C from Pigmy Basin is ~0.7 ‰ [ J. Richey, personal communication], and so a threshold at 0.7 ‰ pr ovides a baseline to evaluate negative excursions. Three of the five freshw ater events (see Fi gure 9 and Appendix A supplementary Figure 4) are a ssociated with more negative 13C values, suggesting a modest relationship between Mississi ppi River runoff and GOM surface water 13CDIC. However, this relationship is based on modern G. ruber 13C values from the Pigmy Basin (see Figure 1 for location) that could reflect local conditions and/or productivity changes associated with the past 50 year s. Therefore, we may be overor underestimating the change in 13CDIC from Mississippi River input. Nonetheless, the G. ruber 13C (‰VPDB)Age (ka)% Coarse fractiona b G. ruber 13C (‰VPDB)Age (ka)% Coarse fractiona b

PAGE 38

32 positive relationship between percent coarse fraction, G. ruber 13C, and 18OGOM suggests Mississippi River discharge influences the 13CDIC in the northeast GOM. Other possible influences on the G. ruber 13C record include higher surface water pH during the LGM (0.2 0.1 units) [ Sanyal et al., 1995], which would effectively decrease the 13C of G. ruber by 0.48 ‰ [ Spero, 1998], and changes in deep water mass properties [ Spero and Lea, 2002]. ODP Site 625 sits in <1 km, and is therefore bathed in Antarctic Intermediate Water (AAIW). Spero and Lea [2002] argue the carbon isotopic minimum associated with glacial maxima re sults from Southern Ocean ventilation and replenishment of 12C-enriched DIC to the surface. Once at the surface, the isotopically depleted waters sink to form AAIW. AAIW is then transported through deep advection into the southern Atlantic and finally to the northeastern GOM where it upwells and influences the 13CDIC in surface waters. More work is needed to distinguish between the contribution from AAIW and pH. For now, we consider the possibility that both factors influence the 13C towards more negative values. Interestingly, the peak in 13C during the last interglacial did not occur until substage 5a, ~80 ka, a full 40 ky after the te rmination. In contrast, MIS 2 to 1 carbon isotopic enrichment of G. ruber calcite developed in ~15 ky and 13C reaches similar values as MIS 5a by the mid-Holocene. While we have no ready explanation, the occurrence of a carbon isotopic maximum during late stage 5 in the eastern equatorial Pacific [ Spero and Lea, 2002], Western Equato rial Atlantic [ Curry and Oppo, 1996], and northeast GOM [ this study ] suggests a global response to perturbations in the carbon system. Overall, the positive relationship between percent coarse fraction and G. ruber 13C provides an explanation for the large >1 ‰ shifts on the orbital scale. Interpreting the sub-millennial to millennial scale variab ility, however, is difficult due to changing contributions from riverine influx, variability in surface water pH, changes in deep water properties, and symbiont photos ynthetic rates. This study is able to make first order

PAGE 39

33 conclusions on the relative timing of each of these factors (except photosynthetic rates) based on percent coarse fraction, 18OGOM, and previous records. 4.7 Last local occurrence of Globorotalia tumida flexuosa The last local occurrence (LLO) of G. flexuosa offers researchers working with low latitude deep sea sediments the ability to quickly distinguish se diments of Pleistocene age, adding another age control point. LLO of G. flexuosa at ODP Site 625 was originally reported at 68 ka by Joyce et al. [1990] based on low-resolution 18O data. Higher resolution 18O data correlated to the SPECMAP stack allow us to report here a new LLO date for G. flexuosa in the northern GOM at 78.6 ka (Figure 1, Table 2). The discrepancy between the dates determined from the same core site can be explained by the difference in sampling resolution and age models. Joyce et al. [1990] sampled at a mean resolution of 15 cm, which, given the se dimentation rates for the last glacial, provides an average temporal resolution of ~1500 years per sample. In contrast, our record was sampled at a mean resolution of 5 cm, which converts to a temporal resolution of <500 years per sample. We repor t a new age for the extinction of G. flexuosa at 78.6 ka within MIS 5a in the northern GOM ba sed on core depth within Hole 625 C, 13.513.55 mbsf (Table 2) and our age model. 4.8 Two-step deglacial during termination II Close examination of termination II in the GOM (Figure 11) reveals a climatic reversal. The early rise in SST of 2.9 C at 132.1 ka (n = 3) reverses and decreases by 1.5 C at 130.4 ka (n = 8; 3 replicates, 1 trip licate) for 2 ky before increasing 3.5 C (n = 8) to peak interglacial te mperatures. Similarly, the 18OGOM record exhibits a decrease in values towards fresher conditions at 137 ka before a reversal characterized by a 0.5 ‰ increase in average valu es at 131 ka. Benthic 18O also show a decrease in values starting at 131.3 ka before a reversal toward s more positive values at 129.5 ka. These findings are consonant with a number of r ecords exhibiting a two-step deglaciation during termination II including terrestrial [ Shackleton et al., 2002] and marine

PAGE 40

34 Figure 11. Two-step deglaciati on during termination II Figure 11. Two-step deglaciation during term ination II. (a) Mg/Ca-SST (C) from G. ruber Mg/Ca converted to temperature using Mg/Ca = 0.449[0.09 • SST (C)] [ Anand et al., 2003]; (b) ODP 625 18OGOM calculated by subtracting sea level [ Waelbroeck et al., 2002] from 18OSW (gray line refers to freshwater threshold as defined in text); (c) benthic 18O of Cibicidoides spp corrected +0.64 ‰ [ Duplessy et al., 1984] to compare with three Uvigerina spp. values; (d) Ceara Rise benthic 13C [ Curry and Oppo, 1996] indicate the relative strength of NADW; (e) global sea level [ Waelbroeck et al., 2002] versus age (ka). The reversal in SST rise, 18OGOM decrease, and short increase in benthic 18O coincide with a decrease in NADW, suggesti ng a link between THC and a two-step deglaciation during termination II. GOM SST initial rise (132.1 ka) precedes the rise in global s ea level (131.1 ka) by 1 ky. Based on the defined threshold for freshwater to the GOM, 18OGOM indicates three freshwater events from 124-119 and 110-105 ka. Mg/Ca-SST (C)Age (ka)Benthic 18O (‰VPDB)18OGOM(‰VSMOW)a b c e dSea Level (m)CearaRise 13Cbenthic(‰VPDB) Mg/Ca-SST (C)Age (ka)Benthic 18O (‰VPDB)18OGOM(‰VSMOW)a b c e dSea Level (m)CearaRise 13Cbenthic(‰VPDB)

PAGE 41

35 [ Siedenkrantz, 1993; Siedenkrantz et al., 1996; Lototskaya and Ganssen, 1999; Gouzy et al., 2004; Canariato and Kennett, 2005; Zhao et al., 2006; Lea et al., 2006; Skinner and Shackleton, 2006] climate records. Th e age for the climate revers al can be constrained by 238U/232Th dated corals from Barbados [ Gallup et al., 2002] (129 ka) and the Western Pacific [ Esat et al., 1999] (128 ka) that record a pause in sea level rise. Our age model suggests the climate reversal occurred in the GOM between 127 and 132.5 ka, which is well within the error ( 3 ky) and suggests the pause in sea level occurred at the same time. A pause in the deglaciation (128131 ka) reflected in benthic 18O [ Lototskaya and Ganssen, 1999; Gouzy et al., 2004; this study] coincident with a break in sea level rise [ Esat et al., 1999; Gallup et al., 2002] and Heinrich Event 11 [ Shackleton et al., 2002], suggest a cold reversal triggered by a reduction in THC similar to the Younger Dryas [ Rooth, 1982; Boyle and Keigwin, 1987; Keigwin and Lehman, 1994; McManus et al., 2004]. In addition, there is ev idence of fluctuating North A tlantic deep water sources, which resulted in millennial-scale SST fluctuations in the Western subtropical North Atlantic [ Oppo et al., 2001], that supports a reorganization of THC with global effects. Similarly, Ceara Rise benthic 13C data shows a similar pause during termination II. We argue that a return to saltier conditions prior to a decrease in SST and increase in benthic 18O in the GOM is indicative of a reduction in NADW volume. 5. Conclusion Stable isotope and Mg/Ca analyses of planktic Foraminifera from deep sea sediment cores from ODP Site 625 reveal orbi tal to millennial-scale variability in GOM SST and SSS during the late Pleistocene. Northeast GOM SST exhibits a 4.4 1.3 C and 3.2 0.99 C glacial to interglacial SST change over termination I and II respectively. Our record i ndicates the last interglacial was 1.2 0.96 C warmer than present and MIS 6 was 2.3 1.63 C warmer than the LGM.

PAGE 42

36 G. ruber 18OC values range from 1.28 to -2.6 ‰ VPDB with an average of -0.14 ‰. Glacial to interglacial 18OC amplitude is 2.7 and 3 ‰ for terminations I and II respectively. Global sea level may account for ~1‰ [ Adkins et al., 2002] and glacial to interglacial temperature change accounts for 0.77 ‰ for termination I and 0.69 ‰ for termination II. Therefore, the residual 18OC change, denoted 18OGOM, of 0.93 ‰ and 1.31 ‰ respectively, likely refl ects hydrologic variability. The hydrologic variability reflected in the difference in 18OGOM from modern ( 18OGOM) shows similarity to the difference in Caribbean Sea 18OSW ( 18OCaribbean), with the former showing greater amplitude and variability. Additionally, sub-orbital increases in GOM SST and SSS closely followed decreases in the relative strength of NADW production based on Ceara Rise benthic 13C data. This finding is consistent with the hypot hesis that the Atlantic Warm Pool responds to changes in THC by accumulation of heat a nd salt. Overall, highe r temporal resolution and comparable findings in GOM records compared with similar records from the effects. Western Caribbean suggest GOM r ecords provide more detailed records of climate variability within the Atlantic Warm Pool. A simple test of the Meltwater Routi ng Hypothesis reveals three freshwater events recorded in ODP Site 625 that do not match major Greenland interstadials, although we may lack the necessary tempor al resolution and proximity to glacial discharges in the GOM. A particularly larg e freshwater event, IIa IIb, appears to correspond to with Antarctic isotope maximu m 2 and Heinrich Event 2. These findings support previous work that suggests summer melting of the LIS occurred during Antarctic warmings [ Hill et al., 2006]. To our knowledge, we are reporting the first evidence of a cold reversal in the low latitude Atlantic Ocean during termination II. The initial ri se in GOM SST at 132.1 ka of 2.9 C is followed by a cold reversal of 1.5 C at 130.4 ka for 2 ky and final increase to full interglacial warmth. The reversal in GOM SST is consonant with a pause in sea level

PAGE 43

37 rise and reduced NADW, suggesting a reduc tion in THC may have caused a global twostep deglaciation. These findings provide importa nt clues to understanding the role of the Atlantic Warm Pool over the last glacial cycle. Howe ver, there is a need for more tropical and subtropical SST and SSS records across multiple glacial-interglacial cycles to fully understand the role of the low latitude Atla ntic Ocean in regional and global climate change. Furthermore, the discovery of a clim atic reversal during termination II raises the question of the existence of similar climate re versals during previous glacial terminations. In addition, our 18OGOM record indicates large cha nges in Northeast GOM hydrology that require further examination. Unraveling the different end-member contributions to the 18OSW is needed to better understand the role of the Atlantic Warm Pool over the last 155 ky in both regional and global climate change A more accurate quantification of the magnitude of salinity change requires a proxy sensitive to a single end-member contribution (i.e., riverine infl ux). Finally, while this study was able to test a simple version of the Meltwater Routing Hypothesi s, a similar study with higher temporal resolution would provide a more complete und erstanding of the role of LIS freshwater during abrupt climate change.

PAGE 44

38 References [1] Adkins, J. F., K. McIntyre and D. P. Schr ag (2002), The salinity, temperature, and 18O of the glacial deep ocean, Science 298, 1769-1773. [2] Anand, P., H. Elderfield, H., and M. H. Conte (2003), Calibration of Mg/Ca thermometry in planktonic foraminifera from a sediment trap time series, Paleoceanography 18, doi:10.1029/2002PA000846. [3] Andree, M., J. Beer, H. Oeschger, A.C. Mix, W.S. Broecker, N. Ragano, P. O'Hara, G. Bonani, H.J. Hofmann, E. Morenzoni, M. Nessi, M. Suter and W. Woelfli (1985), Accelerator radio-carbon ages on fo raminifera separated from deep-sea sediments. In: The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present, E.T. Sundquist and W.S. Br oecker (eds.), Geophysical Monograph 32, Amer. Geophys. Union, Washington, DC, p. 143-153. [4] Arz, H. W., J. P tzold, and G. Wefer (1999), The de glacial history of the western tropical Atlantic as inferred from hi gh resolution stable isotope records off northeastern Brazil, Earth and Planetary Science Letters 167(1-2), 105-117, doi:10.1016/S0012-821X(99)00025-4. [5] Bard, E., B. Hamelin, M. Arnold, L. M ontaggioni, G. Cabioch, and G. Faure (1996), Deglacial sea-level record from Tahiti corals and the timing of global meltwater discharge, Nature 382, 241-244, doi: 10.1038/382241a0.

PAGE 45

39 [6] Barker, S., M. Greaves, and H. Elderfield (2003), A study of cleaning procedures used for foraminiferal Mg/Ca paleothermometry, Geochem., Geophys., Geosyst 4, doi: 10.1029/2003GC000559. [7] Be, A., and D. S. Tolderlund (1971), Dist ribution and ecology of living planktonic foraminifera in the surface waters of the Atlantic and Indian oceans, in The Micropaleontology of Oceans edited by B. M. Funnell and W. R. Reidel, pp. 105–149, Cambridge Univ. Press, New York. [8] Beers, Y. (1957), Introduction to the Theory of Error 66 pp., Addison-Wesley, Boston, Mass. [9] Bemis, B. E., H. J. Spero, J. Bijma, D.W. Lea (1998), Reevaluation of the oxygen isotopic composition of planktonic foramini fera: Experimental results and revised paleotemperature equations, Paleoceanography 13, 150-160. [10] Bice, M. B., P. B. deMenocal, J. Arbu szewski (2006), Reconstructing Sea-Surface Salinity of the Mid-Atlantic Using Mg/Ca and Oxygen Isotope Ratios of G. ruber Eos Trans. AGU, 87 (52), Fall Me et. Suppl., Abstract PP21C-1703. [11] Blunier, T., and E. J. Br ook (2001), Timing of millennial-scale climate change in Antarctica and Greenland duri ng the last glacial period, Science 291, 109–112. [12] Bowen, G.J., and J. Revenaugh (2003), Interpolating the isotopic composition of modern meteoric precipitation, Water Resources Res ., 39, doi: 10.1029/2003WR002086.

PAGE 46

40 [13] Boyle, E A. (2003), Cadmium and 13C Paleochemical Ocean Distributions During the Stage 2 Glacial Maximum, review -article, Retrieved June 11, 2008, from tp://arjournals.annualreviews.or g/doi/abs/10.1146/annurev.ea.20.050192.001333. [14] Boyle, E. A., and L. D. Keigwin (1982), D eep circulation of th e North Atlantic over the last 200,000 Years: geochemical evidence, Science 218, 784-787, doi: 10.1126/science.218.4574.784. [15] Boyle, E. A. and L. D. Keigwin (198 5), Comparison of Atlantic and Pacific paleochemical records for the last 215,000 years: changes in deep ocean circulation and chemical inventories, Earth Planetary Science Letters 76, 135150. [16] Boyle, E. A. and L. D. Keigwin (1987) North Atlantic thermohaline circulation during the past 20,000 years linked to high-latitude surface temperature, Nature 330, 35-40, doi:10.1038/330035a0. [17] Broecker, W. S. (1998), Paleocean Circul ation during the last deglaciation: a bipolar seesaw?, Paleoceanography 13, 119–121. [18] Broecker, W. S., D. Peteet, and D. Rind (1985), Does the ocean-atmosphere system have more than one stable mode of operation?, Nature 315, 21-26, doi: 10.1038/315021a0. [19] Broecker, W. S., J. P. Kennett, B. P. Flow er, J. T. Teller, S. Trumbore, G. Bonnai, and W. Wolfli (1989), Routing of meltwater from the Laurentide Ice Sheet during the Younger Dryas cold episode, Nature 341, 318-321, doi:10.1038/341318a0.

PAGE 47

41 [20] Broecker, W. S G. Bond, and M. A. Klas (1990), A salt oscillator in the glacial Atlantic? 1. The concept, Paleoceanography 5, 469-477. [21] Brown, P., J. P. Kennett, and B. L. Ingram (1999), Marine Evidence for Episodic Holocene Megafloods in North America and the Northern Gulf of Mexico, Paleoceanography 14(4), 498–510. [22] Brunner, C. A. (1979), Distribution of pla nktonic foraminifera in surface sediments of the Gulf of Mexic o, Micropaleontology, 25, 325–335. [23] Cannariato, K. G., and J. P. Kennett (2005), Structure of the penultimate deglaciation along the California marg in and implicati ons for Milankovitch theory, Geology 33, 157-160, doi: 10.1130/G21065.1. [24] Chave, K. E. (1954), Aspects of the bi ogeochemistry of magnesium: calcareous marine organisms, Journal of Geology 62, 266-283. [25] Chappell, J., and N. J. Shackleton (1 986), Oxygen isotopes and sea level, Nature 324, 137-140, doi: 10.1038/324137a0. [26] Chappell, J., A. Omura, T. Esat, M. McCulloch, J. Pandolfi, Y. Ota (1996), Reconciliation of late Quaternary sea le vels derived from coral terraces at Huon Peninsula with deep s ea oxygen isotope records, Earth and Planetary Science Letters 141, 227-236, doi: 10.1016-0012-821X(96)00062-3. [27] Chiang, J. C. H., M. Biasutti, and D. S. Battisti (2003), Sensitivit y of the Atlantic Intertropical Convergence Zone to la st glacial maximum boundary conditions, Paleoceanography 18, 1094, doi: 10.1029/2003PA000916.

PAGE 48

42 [28] Clark, P. U., S. J. Marshall, G. K. C. Clarke, S. W. Hostetler, J. M. Licciardi, and J. T. Teller (2001), Freshwater forcing of abrupt climate change during the last glaciation, Science 293, 283-287, doi:10.1126/science.1062517. [29] CLIMAP Project Members (1976), The Surface of the Ice-Age Earth, Science 191, 1131-1137, doi: 10.1126/science.191.4232.1131. [30] Crowley, T. J. (1992), North Atlantic deep water cools the southern hemisphere, Paleoceanography 7(4), 489–497. [31] Curry, W. B. (1996), Late Quaternary deep circulation in the western equatorial Atlantic, in "The South Atlantic: Present and Past Circulation" edited by G. Wefer, W. H. Berger, G. Siedler, a nd D.J. Webb, Springer-Verlag, Berlin, pp. 577-598. [32] Curry, W. B., and D. W. Oppo (1996), Sync hronous, high-frequency oscillations in tropical sea surface temperatures and Nort h Atlantic deep water production during the last glacial cycle, Paleoceanography 12(1), 1–14. [33] Dahl, K. A., A. J. Broccoli, and R. J. Stouffer (2005), Assessing the role of North Atlantic freshwater forcing in millenni al scale climate variability: a tropical Atlantic perspective, Climate Dynamics 24, 325-346, doi:10.1007/s00382-0040499-5. [34] Dansgaard, W., S. J. Johnsen, H. B. Clause n, D. Dahl-Jensen, N. S. Gundestrup, C. U. Hammer, C. S. Hvidberg, J. P. Steffe nsen, A. E. Sveinbjrnsdottir, J. Jouzel, and G. Bond (1993), Evidence for general instability of past climate from a 250kyr ice-core record, Nature 364, 218-220.

PAGE 49

43 [35] Dekens, P. S., D. W. Lea, D. K. Pak, and H. J. Spero (2002), Core top calibration of Mg/Ca in tropical foraminifera: Re fining paleotemperature estimation, Geochemistry Geophysics Geosystems 3, 1022, doi:10.1029/2001GC000200. [36] deMenocal, P.B., Arbuszewski, J., Kapla n, A., and Bice, M. (2007) Fidelity of 18Oseawater estimates using foraminferal shell 18O and Mg/Ca, EOS Fall 2007 Meeting Program with abstracts. [37] Dong, B.-W. and R. T. Sutton (2002), Adjust ment of the coupled ocean–atmosphere system to a sudden change in the thermohaline circulation, Geophysical Research Letters 29, 1728, doi:10.1029/2002GL015229. [38] Duplessy, J., N. J. Shackleton, R. K. Matthews, W. Prell, W. F. Ruddiman, M. Caralp, and C. H. Hendy (1984), 13C Record of benthic fora minifera in the last interglacial ocean: implications for th e carbon cycle and the global deep water circulation, Quaternary Research 21, 225-243, doi:10.1016/00335894(84)90099-1. [39] Elderfield, H., and G. Ganssen (2000), R econstruction of temp erature and d18O of surface ocean waters using Mg/Ca of planktonic foraminiferal calcite, Nature 405, 442445. [40] Emiliani, C. (1955), Pleistocene Temperatures, Journal of Geology 63(6), 538-578. [41] Emiliani, C., S. Gartner, B. Lidz, K. Eldri dge, D. K. Elvey, T. L. Huang, J. J. Stipp, and M. F. Swanson (1975), Paleoclimatologi cal analysis of late Quaternary cores from the northeastern Gulf of Mexico, Science 189, 1083-1088.

PAGE 50

44 [42] EPICA Community Members (2006), Oneto-one coupling of glacial climate variability in Greenland and Antarctica, Nature 44, 195-198. [43] Ericson, D. B., and G. Wollin (1968), Plei stocene climates and chronology in deep sea sediments, Science 162, 1227-1234. [44] Esat, T. M., M. T. McCulloch, J. Chappe ll, B. Pillans, and A. Omura (1999), Rapid fluctuations in sea level recorded at Huon peninsula during the penultimate deglaciation, Science 283, 197–201. [45] Fairbanks, R. G. (1989), A 17,000-year glacio-e ustatic sea level re cord: influence of glacial melting rates on the Younger Drya s event and deep-ocean circulation, Nature 342(6250), 637-642, doi: 10.1038/342637a0. [46] Fairbanks, R. G., C. D. Charles, and J. D. Wright (1992), Origin of global meltwater pulses, in Taylor, R.E., et al., eds., Radiocarbon after four decades : New York, Springer-Verlag, p. 473–500. [47] Ferguson, J., G. Henderson, M. Kucera, and R. Rickaby (2008), Systematic change of foraminiferal Mg/Ca ratios acr oss a strong salinity gradient, Earth and Planetary Science Letters 265, 153-166, doi: 10.1016/j.epsl.2007.10.011. [48] Flower, Benjamin P., D. W. Hastings, H. W. Hill, and T. M. Quinn (2004), Phasing of deglacial warming and La urentide Ice Sheet meltwater in the Gulf of Mexico, Geology 32(7), 597-600, doi:10.1130/G20604.1. [49] Flower, B. P., and J. P. Kennett, J. P. (1990), The Younger Dryas cool episode in the Gulf of Mexico, Paleoceanography 5, 949-961.

PAGE 51

45 [50] Gallup, C. D., H. Cheng, F. W. Taylor and R. L. Edwards (2002), Direct determination of the timing of sea level change during termination II, Science 295, 310-313, doi: 10.1126/science.1065494. [51] Ganopolski, A., and S. Rahmstorf (2001) Rapid changes of glacial climate simulated in a coupled climate model, Nature 409, 153-158. [52] Gouzy, A., Malaiz, B., C. Pujol, and K. Charlier (2004), Climatic "pause" during Termination II identified in shallow and intermediate waters off the Iberian margin, Quaternary Science Reviews 23 (14-15), 1523-1528, doi: 10.1016/j.quascirev.2004.03.002. [53] Guilderson, T. P., R. G. Fairbanks, and J. L. Rubenstone (2001), Tropical Atlantic coral oxygen isotopes: glacial-interglacial sea surface temperatures and climate change, Marine Geology 172(1-2), 75-89. [54] Haarsma, R. J., E. Campos, W. Hazeleger, and C. Severijns (2008), Influence of the meridional overturning circulation on tr opical Atlantic climate and variability, Journal of Climate 21, 1403-1416, doi: 10.1175/2007JCLI1930.1 [55] Hastings, D. W., A. D. Russell, and S. R. Emerson (1998), Foraminiferal Magnesium in Globeriginoides Sacculifer as a Paleotemperature Proxy, Paleoceanography 13(2), 161-169. [56] Hays, J. D., J. Imbrie, and N. J. Shacklet on (1976), Variations in the earth’s orbit: Pacemaker of the ice ages, Science 194, 1121-1132.

PAGE 52

46 [57] Hemleben, C., M. Spindler, I. Breitinge r, and R. Ott (1987), Morphological and physiological responses of Globigerinoi des sacculifer (B rady) under varying laboratory conditions, Marine Micropaleontology 12, 304-324. [58] Hemming, S. R., P. E. Biscaye, W. S. Broecker, N. G. Hemming, M. Klas,and I. Hajdas (1998), Provenance change coupl ed with increased clay flux during deglacial times in the western equatorial Atlantic, Palaeogeography, Palaeoclimatology, Palaeoecology 142(3-4), 217-230, doi:10.1016/S00310182(98)00069-8. [59] Hendy, I. L. and J. P. Kennett (2000), Dansgaard-Oeschger cycles and the California Current system: planktonic fo raminiferal response to rapid climate change in Santa Barbara Basin, Ocean Drilling Program Hole 893A, Paleoceanography 15(1), 30-42. [60] Hill, H. W., B. P. Flower, T. M. Quinn, D. J. Hollander, and T. P. Guilderson (2006), Laurentide Ice Sheet meltwater and abrupt climate cha nge during the last glaciation, Paleoceanography 21, PA1006, doi:10.1029/2005PA001186. [61] Huber, C., M. Leuenberger, R. Spahni, J. Fl ckiger, J. Schwander, T.F. Stocker, S. Johnsen, A. Landais, and J. Jouzel (2006), Isotope cal ibrated Greenland temperature record over Marine Is otope Stage 3 and its relation to CH4, Earth and Planetary Science Letters 243, 504-519. [62] Hughen, K., J. Southon, S. Lehmanc, C. Bert randa, and J. Turnbull (2006), Marinederived 14C calibration and activity record for the past 50,000 years updated from the Cariaco Basin, Quaternary Science Reviews 25, 3216-3227.

PAGE 53

47 [63] Hut, G. (1987), Consultants' group meeting on stable isotope reference samples for geochemical and hydrological investigations, Report to the Director General, International Atomic Energy Agency, Vienna, April 1987. [64] Imbrie, J., J. D. Hays, D. G. Martinson, A. McIntyre, A. C. Mix, J. J. Morley, N. G. Pisias, W. L. Prell, N. J. Shackleton (1984), The orbital theory of Pleistocene climate: support from a revi sed chronology of the marine 18O record, In: Berger, A., Imbrie, J., Hays, J., Kukla, G., Saltzman, B. (Eds.), Milankovitch and climate Reidel Publishing Company, Dordrecht, pp. 269-305. [65] Jasper, J. P., and R. B. Gagosian (1989) Alkenone Molecular Stratigraphy in an Oceanic Environment Affected by Glacial Freshwater Events, Paleoceanography 4(6), 603-614. [66] Jrgensen, B.B., J. Erez, N. P. Revsbech, and Y. Cohen (1985), Symbiotic photosynthesis in a planktonic foraminiferan Globigerinoides sacculifer (Brady), studied with microelectrodes, Limnology and Oceonography 30(6), 1253-1267. [67] Joyce, J. E., L. R. C. Tjalsma, and J. M. Prutzman (1990), High-Resolution Planktic Stable Isotope Record and Spectral An alysis for the Last 5.35 M.Y.: Ocean Drilling Program Site 625 Northeast Gulf of Mexico, Paleoceanography 5(4), 507-529. [68] Joyce, J. E., L. R. C. Tjaisma and J. M. Prutzman (1993), North American glacial meltwater hisotry for the past 2.3 m.y.: Oxygen isotope evidence from the Gulf of Mexico, Geology 21, 483-486. [69] Keigwin, L. D., and E. A. Boyle (1999), Surface and Deep Ocean Variability in the Northern Sargasso Sea During Marine Isotope Stage 3, Paleoceanography 14(2), 164–170.

PAGE 54

48 [70] Keigwin, L. D., and E. A. Boyle (2008) Did North Atlantic overturning halt 17,000 years ago?, Paleoceanography 23, PA1101, doi:10.1029/2007PA001500. [71] Keigwin, L. D., and S. J. Lehman (19 94), Deep Circulation Change Linked to HEINRICH Event 1 and Younger Dryas in a Middepth North Atlantic Core, Paleoceanography 9(2), 185-194. [72] Kennett, J. P., K. Elmstrom, N. Penrose (1985), The last deglaci ation in Orca Basin, Gulf of Mexico: High resolution pl anktonic Foraminifera changes, Paleogeography, Paleoclimatology, Paleoecology 50, 189-216. [73] Kennett, J. P., and P. Huddlestun (19 72), Late Pleistocene Paleoclimatology, foraminiferal biostratigraphy and tephro chronology, western Gulf of Mexico, Quaternary Research 2, 38-69. [74] Kennett, J. P., and N. J. Shackleton (1975) Laurentide ice sheet meltwater recorded in Gulf of Mexico deep-s ea cores, Science, 188, 147-150. [75] Khler-Rink, S., AND M. Khl (2005), The chemical microenvironment of the symbiotic planktonic foraminifer Orbulina universa Marine Biology Research, 1, 68-78. [76] Kuhlbrodt, T., A. Griesel, M. Montoya, A. Levermann, M. Hofmann, and S. Rahmstorf (2007), On the driving pro cesses of the Atlantic meridional overturning circulation, Reviews of Geophysics 45, RG2001, doi: 10.1029/2004RG000166.

PAGE 55

49 [77] Latif, M., E. Roeckner, U. Mikolajewicz, and R. Voss (2000), Tropical stabilization of the thermohaline circulation in a greenhouse warming simulation, Journal of Climate 13(11), 1809-1813. [78] Lea, D. W., and E. A. Boyle (1990), Fo raminiferal Reconstruction of Barium Distributions in Water Masses of the Glacial Oceans, Paleoceanography 5(5), 719-742. [79] Lea, D. W., T. A. Mashiotta, and H. J. Spero (1999), Controls on magnesium and strontium uptake in planktonic forami nifera determined by live culturing, Geochimica et Cosmochimica Acta 63, 2369-2379. [80] Lea, D. W., D. K. Pak, and H. J. Spero, H. J. (2000), Climate impact of late Quaternary equatorial Pacific se a surface temperature variations, Science 289, 1719-1724, doi: 10.1126/science.289.5485.1719. [81] Lea, D. W., D. K. Pak, C. L. Belanger, H. J. Spero, M. A. Hall, and N. J. Shackleton (2006), Paleoclimate history of Galpagos surface waters over the last 135,000 yr, Quaternary Science Reviews 25, 1152-1167, doi:10.1016/j.quascirev.2005.11.010. [82] Lea, D.W ., C. L. Belanger, H. J. Spero, M. A. Hall and N. J. Shackleton (2003), Synchroneity of tropical and high-latitude Atlantic temperatures over the last glacial termination, Science 301, 1361-1364. [83] LeGrande, A. N. and J. Lynch-Stieglitz (2007), Surface currents in the western North Atlantic during the Last Glacial Maximum, Geochem. Geophys. Geosyst. 8 Q01N09, doi:10.1029/2006GC001371.

PAGE 56

50 [84] Leventer, A., D. F. Williams, J. P. Kenne tt (1982), Dynamics of the Laurentide ice sheet during the last deglaciation: Evidence from the Gulf of Mexico, Earth Planetary Science Letters 59, 11-17. [85] Levitus, S., and T. P. Boyer (1994), Wo rld Ocean Atlas 1994: US-Department of Commerce, NOAA Atlas NESDIS, Na tional Oceanic and Atmospheric Administration, Silver Spring, Md. [86] LoDico, J. M., B. P. Flower, and T. M. Quinn (2006), Subcentennial-scale climatic and hydrologic variability in the Gulf of Mexico during the early Holocene, Paleoceanography 21, PA3015, doi:10.1029/2005PA001243. [87] Lototskaya, A., and G. M. Ganssen ( 1999), The structure of Termination II (penultimate deglaciation and Eemian) in the North Atlantic, Quaternary Science Reviews 18, 1641-1654, doi:10.1016/S0277-3791(99)00011-6. [88] Lund, D. C., J. Lynch-Stieglitz, and W. B. Curry (2006), Gulf Stream density structure and transport during the last millennium, Nature 444, 601-604. [89] Lynch-Stieglitz, J., W. B. Curry, N. Slow ey (1999), Weaker Gulf Stream in the Florida Straits during the Last Glacial Maximum, Nature 402, 644-648. [90] Lynch-Stieglitz, J., J.F. Adkins, W.B. Cu rry, T. Dokken, I.R. Hall, J.C. Herguera, J.J.-M. Hirschi, E. Ivanova, C. Kissell, O. Marchal, T.M. Marchitto, I.N. McCave, J.F. McManus, S. Mulitza, U.S. Ninnemann, E.-F.Yu, and R. Zahn (2007), Atlantic overturning circulation dur ing the last glacial maximum, Science, 316, 66-69.

PAGE 57

51 [91] Lynch-Stieglitz, J., W. B. Curry, D. W. Oppo, U. S. Ninneman, C. D. Charles, and J. Munson (2006), Meridional overturning circ ulation in the Sout h Atlantic at the last glacial maximum, Geochemistry Geophysics Geosystems 7, Q10N03, doi:10.1029/2005GC001226 [92] Marchitto, T.M., D. W. Oppo, and W. B. Cu rry (2002), Paired benthic foraminiferal Cd/Ca and Zn/Ca evidence for a greatly in creased presence of Southern Ocean Water in the glacial North Atlantic, Paleoceanography, 17, 10.1029/2000PA000598. [93] Martin, R.E., E. D. Neff, G. W. Johnson, and D. E. Krantz (1993), Biostratigraphic expression of Pleistocene sequen ce boundaries, Gulf of Mexico, PALAIOS 8, 155-171. [94] Martinson, D.G N. G. Pisias, J. D. Hays, J. Imbrie, T. C. Moore, Jr., and N. J. Shackleton (1987), Age dating and the orbita l theory ofthe ice ages: development of a high-resolution 0 to 300 000 year chronostratigraphy, Quaternary Research 27, 1-29. [95] McConnaughey, T. A. (2003), Sub-equlibri um oxygen-18 and carbon-13 levels in biological carbonates: car bonate and kinetic models, Coral reefs 22, 316-327. [96] McManus, J. F., D. W. Oppo, J. L. Culle n (1999), A 0.5-million-year record of millennial-scale climate variabil ity in the North Atlantic, Science 283, 971-975, doi:10.1126/science.283.5404.971. [97] McManus, J. F., R. Francois, J. Gherar di, L. D. Keigwin, and S. Brown-Leger (2004), Collapse and rapid resumption of A tlantic meridional ci rculation linked to deglacial climate changes, Nature 428, 834-837, doi:10.1038/nature02494.

PAGE 58

52 [98] Molinari, R.L. (1980), Current variability and its relation to sea-surface topography in the Caribbean Sea and the Gulf of Mexico, Marine Geodesy 3, 409-436. [99] Mller-Karger, F. E., J. J. Walsh, R. H. Evans, and M. B. Meyers (1991), On the Seasonal Phytoplankton Concentration a nd Sea Surface Temperature Cycles of the Gulf of Mexico as Determined by Satellites, Journal of Geophysical Research 96(C7), 12,645-12,665. [100] NOAA/OAR/ESRL (National Oceanic & Atmospheric Administration, Earth System Research Laboratory) (1994), NODC (Levitus) World ocean Atlas Data 1994: http://www.cdc.noaa.gov/cdc/data.nodc.woa94.html (May 2008). [101] North Greenland Ice Core Project member s (2004), High-resolution record of Northern Hemisphere climate extending into the last interglacial period, Nature 431,147-151. [102] Nurnberg, D., J. Bijma, C. Hemleben (1996), Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperatures, Geochimica et Cosmochimica Acta 60, 803-814. [103] Nrnberg, D., M. Ziegler, C. Karas, R. Tiedemann, M. Schmidt (in press), Interacting Loop Current variability and Mississippi discharg e over the past 400 kyrs, Earth and Planetary Science Letters [104] Oppo, D. W., L. D. Keigwin, J. F. McMa nus, and J. L. Cullen (2001), Persistent Suborbital Climate Variability in Marine Isotope Stage 5 and Termination II, Paleoceanography 16(3), 280-292.

PAGE 59

53 [105] Orr, W.N. (1969), Variation and Distributi on of Globigerinoides ruber in the Gulf of Mexico, Micropaleontology 15, 373-379. [106] Ortner, P. B., T. N. Lee, P. J. Milne, R. G. Zika, M. E. Clarke, G. P. Podesta, P. K. Swart, P. A. Tester, L. P. Atkins on, W. R. Johnson (1995), Mississippi River flood waters that reached the Gulf Stream, Journal of Geophysical Research 100, 595601. [107] Peltier, W.R., and R.G. Fairbanks (2006) Global glacial ice volume and Last Glacial Maximum duration from an ex tended Barbados sea level record, Quaternary Science Reviews 25, 3322-3337. [108] Peterson, L., G. H. Haug, K. A. Hughen, and U. Rhl ( 2000), Rapid changes in the hydrologic cycle of the tropical Atla ntic during the last glacial, Science 290, 1947-1951. [109] Pierrehumbert, R.T. (1995), Thermostats, radiator fins, a nd the local runaway greenhouse, Journal of the Atmospheric Sciences 52(10), 1784-1806. [110] Piotrowski, A. M., S. L. Goldstein, S. R. Hemming, and R. G. Fairbanks (2005), Temporal relationships of carbon cycli ng and ocean circulation at glacial boundaries, Science 307, 1933-1938, doi: 10.1126/science.1104883. [111] Poore, R. Z., H. J. Dowsett, S. Vera rdo, and T. M. Quinn (2003), Millennial-to century-scale variability in Gulf of Mexico Holocene climate records, Paleoceanography 18, 1048, doi:10.1029/2002PA000868.

PAGE 60

54 [112] Poore, R. Z., T. M. Quinn, and S. Verar do (2004), Century-scale movement of the Atlantic Intertropical Convergence Zone linked to solar variability, Geophysical Research Letters 31, L12214, doi:10.1029/2004GL019940. [113] Rahmstorf, S. (2006) Thermohaline Ocean Circulation. In: Encyclopedia of Quaternary Sciences edited by S. A. Elias.Elsevier, Amsterdam. [114] Raymo, M.E., L. E. Lisiecki, and K.H. Nisancioglu (2006), P lio-Pleistocene Ice Volume, Antarctic Climate, and the Global 18O Record, Science 313, 492-495. [115] Richey, J.M., R. Z. Poore, B. P. Fl ower, and T. M. Quinn (2007), A 1400 year multi-proxy record of climate variability from the northern Gulf of Mexico, Geology 35, 423-426. [116] Risebrobakken, B., E. Balbon, T. Dokken, E. Jansen, C. Kissel, L. Labeyrie, T. Richter, L. Senneset (2006), The penultimate deglaciation: High-resolution paleoceanographic evidence from a north-so uth transect along the eastern Nordic Seas, Earth and Planetary Science Letters 241, 505-516, doi:10.1016/j.epsl.2005.11.032. [117] Rooth, C. (1982), Hydrology and ocean circulation, Progress in Oceanography 11, 131-149. [118] Ruhlemann, C., S. Mulitza, P. J. Muller G. Wefer, and R. Zahn (1999), Warming of the tropical Atlantic Ocean and slow down of thermohaline circulation during the last deglaciation, Nature 402, 511-514, doi: 10.1038/990069.

PAGE 61

55 [119] Sanyal, A., N. G. Hemming, G. N. Hanson, N. Gilbert and W. S. Broecker (1995), Evidence for a higher pH in the glacial ocean from boron isotopes in foraminifera, Nature 373(6511), 234-236. [120] Savin, S. M., and R. G. Douglas (1 973), Stable isotope and magnesium geochemistry of recent planktonic Foraminifera from the south Pacific, Geological Society of America Bulletins 84, 2327-2342. [121] Schmidt, M. W., H. J. Spero, and D.W. Lea (2004), Links between salinity variation in the Caribbean and North Atla ntic thermohaline circulation, Nature 428, 160163. [122] Schmidt, M. W., M. J. Vautravers, and H. J. Spero (2006), Rapid subtropical North Atlantic salinity oscillations across Dansgaard-Oeschger cycles, Nature 443, 561564, doi: 10.1038/nature05121. [123] Seidenkrantz, M.-S. (1993), Benthic forami niferal and stable isotope evidence for a 'Younger Dryas-style' cold spell at th e Saalian-Eemian transition, Denmark, Paleogeography, Paleoclimatology, Paleoecology 102, 103-120. [124] Seidenkrantz, M.-S., L. Bornmalm, S. J. Johnsen, K. L. Knudsen, A. Kuijpers, S.-E. Lauritzen, S. A. G. Leroy, I. Mergeal, C. Schweger, and B. Van Vliet-Lano (1996), Two-step deglaciation at the oxyge n isotope stage 6/5E transition: the Zeifen-Kattegat climate oscillation, Quaternary Science Reviews 15, 63-75. [125] Shackleton, N.J., M. Chapman, M. F. Sanc hez-Goni, D. Pailler, and Y. Lancelot (2002), The classic marine isotope substage 5e, Quaternary Research 58, 14-16.

PAGE 62

56 [126] Shipboard Scientific Party (1985), Ocean Drilling Program Leg 100 Report Northeastern Gulf of Mexico Ocean Drilling Program, Texas A & M University. [127] Sima, A., A. Paul, M. Schulz, J. Oerlem ans (2006), Modeling the oxygen-isotopic composition of the North American Ice Sheet and its effect on the isotopic composition of the ocean duri ng the last glacial cycle, Geophysical Research Letters 33, L15706, doi:10.1029/2006GL026923. [128] Skinner, L., and N. J. Shackleton (2006) Deconstructing Terminations I and II: revisiting the glacioeustatic paradigm ba sed on deep-water temperature estimates, Quaternary Science Reviews 25, 3312-3321, oi:10.1016/j.quascirev.2006.07.005. [129] Spero, H. J. (1998), Life Hi story and Stable Isotope Ge ochemistry of Planktonic Foraminifera. Isotope Paleobiology and Paleoe cology: The Paleontological Society Papers 4, 7-36. [130] Spero H. J., and D. F. Williams (1988), Extracting environmental information from planktonic foraminiferal 13C data, Nature 335, 717-719. [131] Spero, H. J., and D. W. Lea (2002), The cause of carbon isotope minimum events on glacial terminations, Science 296, 522-525, doi: 10.1126/science.1069401. [132] Stocker, T. F., D. G. Wright, and M. A. Mysak (1992), A zonally averaged, coupled ocean-atmosphere model for paleoclimate studies, Journal of Climate 5(8), 773797. [133] Stuiver, M. and P. M. Grootes (2000), GISP2 oxygen isotope ratios, Quaternary Research 53, 277-284.

PAGE 63

57 [134] Tan, F. C. (1989), Stable carbon isotopes in dissolved inorganic carbon in marine and estuarine environments, in Handbook of Environmental Geochemistry vol. 3A, edited by P. Fritz and J. C. Fontes, pp. 171-190, Elsevier, New York. [135] Tedesco, K.A. and R. Thunell (2003), S easonal and interannual variations in planktonic foraminiferal flux and assemb lage composition in the Cariaco Basin, Venezuela, Journal of Foraminiferal Research 33, 192-210. [136] Thunell, R., E. Tappa, C. Pride, and E. Kincaid (1999), Sea-surface temperature anomalies associated with the 1997-1998 El Nino recorded in the oxygen isotope composition of planktonic Foraminifera, Geology 27(9), 843-846. [137] Vautravers, M. J., Shackleton, N. J., C. Lopez-Martinez, and J. O. Grimalt (2004), Gulf Stream variability du ring marine isotope stage 3, Paleoceanography 19, PA2011, doi:10.1029/2003PA000966. [138] Vellinga, M. and R. A. Wood (2002), Global cl imatic impacts of a collapse of the Atlantic thermohaline circulation, Climatic Change 54, 251-267, doi: 10.1023/A:1016168827653. [139] Waelbroeck, C., L. Labeyrie, E. Michel J. C. Duplessy, J. F. McManus, K. Lambeck, E. Balbon, and M. Labracherie (2002), Sea-level and deep water temperature changes derived from bent hic Foraminifera isotopic records, Quaternary Science Reviews 21, 295-305. [140] Waliser, D. E. and C. Gautier (1993), A satellite-derived clim atology of the ITCZ, Journal of Climate 6, 2162-2174.

PAGE 64

58 [141] Wang, C. and D. B. Enfiel d (2001), The tropical wester n hemisphere warm pool, Geophysical Research Letters 28(8), 1635–1638. [142] Wang, C., D. B. Enfield, S. Lee, and C. W. Landsea (2006), Influences of the Atlantic warm pool on western hemis phere summer rainfall and Atlantic hurricanes, Journal of Climate 19(12), 3011-3028 [143] Wang, Y. J., H. Cheng, R. L. Edwards, Z. S. An, J. Y. Wu, C.-C. Shen, and J. A. Dorale (2001), A high-resolution absolute-d ated late Pleistocene monsoon record from Hulu Cave, China, Science 294, 2345-2348. [144] Weldeab, S ., R R. Schneider and M. Klling (2006), Deglacial sea surface temperature and salinity increase in th e western tropical Atlantic in synchrony with high latitude climate instabilities, Earth and Planetary Science Letters 241, 699-706, doi:10.1016/j.epsl.2005.11.012. [145] Weldeab, S., D. W. Lea, R. R. Schneid er and N. Andersen (2007), 155,000 Years of West African Monsoon and Ocean Thermal Evolution, Science, 316, doi:10.1126/science.1140461. [146] Zahn, R., and A. Stuber (2002), Suborbital intermediate water variability inferred from paired benthic foraminiferal Cd/Ca and 13C in the tropical West Atlantic and linking with North Atlantic climates, Earth and Planetary Science Letters 200(1-2), 191-205. [147] Zaric, S., G. Donner, G. Fischer, S. Mu litza, and G. Wefer (2005), Sensitivity of planktic foraminifera to sea surface te mperature and export production as derived from sediment trap data, Marine Micropaleontology 55, 75-105, doi: 10.1016/j.marmicro.2005.01.002..

PAGE 65

59 [148] Zhang, R. and T. L. Delworth (2005), Si mulated tropical respons e to a substantial weakening of the Atlantic thermohaline circulation, Journal of Climate 18(12), 1853-1860. [149] Zhao, M., C. Huang, C. Wang, and G. Wei (2006), A millennial-scale U37 K' seasurface temperature record from the S outh China Sea (8N) over the last 150kyr: Monsoon and sea-level influence, Palaeogeography, Palaeoclimatology, Palaeoecology 236, 39-55, doi: 10.1016/j.palaeo.2005.11.033.

PAGE 66

60 Appendices

PAGE 67

61 Appendix A: Error Analysis and Supplementary Tables and Figures 1. Error Analysis Calculating the error associ ated with converting foraminiferal Mg/Ca to absolute temperature requires compounding errors fr om Mg/Ca-SST calibration, analytical precision, and intra-sample precision. Anand et al. [2003] report the cal ibration error at 1.13 C, which is the largest por tion of total error. Analy tical precision for Mg/Ca was <0.28 % root-mean standard deviation (1 ). Intra-sample precision based on 11% (n = 37) replicate and triplicate measurements is 0.25 mmol/mol. To convert the precision error to error in C we first perf orm a formal propagation of error [ Beers, 1957]: [(0.001 mmol/mol)2 + (0.25 mmol/mol)2] = 0.250 mmol/mol then apply the calibration equati on to our mean Mg/Ca value: T (C) = [ln(3.655 0.250 mmol/ mol)/0.449)]/0.09 = 23.30 0.76 C To sum the entire error associated with c onverting Mg/Ca to temperature we propagate error from the calibration error associ ated with the calibration equation [ Anand et al., 2003] and our total precision converted to temperature: [(1.13 C)2 + (0.76 C)2] = 1.4 C Error from the calculation of 18OSW based on Mg/Ca-SST and 18OC is a combination of analytical error and error asso ciated with using esta blished relationships between SST, 18OC, and 18OSW. Bemis et al. [1998] calculate a precision of 0.5 C in their high-light equation. Analytical error and Mg/CaSST conversion error is as mentioned previously. Total error in our 18OSW calculation can be represented as:

PAGE 68

62 Etotal = X + Y + Z where X is the error in the Bemis et al. [1998] high-light equation, Y is the total error in converting Mg/Ca to temperature, and Z is the total anal ytical error in 18OC measurements. Since we are using the equation developed by Bemis et al. [1998] to calculate 18OSW we convert the erro r associated with 18O-temperature and Mg/Catemperature to Vienna-Pee Dee Belemnite (‰ VPDB) scale by an established relationship of 0.208 ‰ T C-1 for foraminiferal calcite: X = 0.5 C (0.208 ‰ T C-1); X = 0.1 ‰ Y = 1.4 C (0.208 ‰ T C-1); Y = 0.29 ‰ Z = 0.06 ‰ (n = >500) Etotal = [(0.1 ‰)2 + (0.29 ‰)2 + (0.06 ‰)2] = 0.32 ‰ VPDB The most conservative error reported w ith sea level reconstruction is 13 m [ Waelbroeck et al., 2002]. Thus the most conservative estimation of error related with changes in 18OSW related to sea level is 0.108 ‰ (if we assume 0.083 ‰ 10 m-1). The residual 18O value, referred to as 18OGOM, is assumed to reflect regional changes in evaporation and precipitation (i.e., isotopically different e nd-members). We report the total error in estimating 18OGOM by adding the error (all converted to ‰VPDB, see above) of each step (Calcula tion of SST) + (Calculation of 18OSW) + (Sea level correction): [(0.29 ‰)2 + (0.32 ‰)2 + (0.108)2] = 0.47 ‰ VPDB However, the above calculation ignores possi ble correlation errors between GOM records and the Waelbroeck et al. [2002] reconstructed sea level, which are difficult to quantify. Therefore, our formal calculation of tota l error represents a minimum estimate of 18OGOM error.

PAGE 69

63 2. Supplementary Tables and Figures Table 1. Sedimentation Rates MISa Depth (mbsf) Age (ka)b SRc (cm/ky) 1 0 1.65 4.75 12 26 2 1.65 5.8 12 24 36 3 5.8 11.8 24 59.3 24 4 11.8 13.2 59.3 74.8 9 5 13.2 15.7 74.8 122.5 5 6 15.7 17.81 122.5 154.5 7 a Marine isotope stage boundaries determined from G. ruber 18OC supplemented by Mg/Ca-SST and benthic 18O b Age determined from age model for ODP 625 c Sedimentation rates (SR) for ODP 625 were determined from change in core depth (m bsf) divided by change in age (ka). Figure 1. Relationship between G. ruber Mg/Ca and average G. ruber weight per interval Figure 1. Relationship between the average individual G. ruber weight per interval and measured Mg/Ca. Best-fit linear regression has an R value of 0.12 indicating variability in Mg/Ca values is not related to cha nges in the average individual weight of G. ruber G. ruber Mg/Ca (mmol/mol)G. ruber weight (g) G. ruber Mg/Ca (mmol/mol)G. ruber weight (g)

PAGE 70

64 Figure 2. Raw G. ruber Mg/Ca values and average G. ruber weight per interval versus core depth Figure 2. (a) Raw G. ruber Mg/Ca (gray with symbols) shown with 2 % weight ed smooth and (b) average individual weights (gray with symbols), shown with 2 % weighted smooth, determined fr om total sample weight divide d by the number of individual G. ruber versus core depth. Light gray lines correspond to average Mg/Ca value for ODP 625, 3.655 mmol/mol and average individual G. ruber weight, 11.6 g. G. ruber Mg/Ca (mmol/mol)G. ruber weight (g)Depth (mbsf)a b G. ruber Mg/Ca (mmol/mol)G. ruber weight (g)Depth (mbsf)a b

PAGE 71

65 Figure 3. 18O and 13C Composition of G. ruber and Cibicidoides spp. versus core depth Figure 3. (a) Raw G. ruber 18O, (b) 13C, (c) Cibicidoides spp. 13C, and (d) benthic 18O plotted against core depth (mbsf). G. ruber 13C (‰VPDB)Cibicidoidesspp. 13C (‰VPDB)Depth (mbsf)Benthic 18O (‰VPDB)G. ruber 18O (‰VPDB)a c b d G. ruber 13C (‰VPDB)Cibicidoidesspp. 13C (‰VPDB)Depth (mbsf)Benthic 18O (‰VPDB)G. ruber 18O (‰VPDB) G. ruber 13C (‰VPDB)Cibicidoidesspp. 13C (‰VPDB)Depth (mbsf)Benthic 18O (‰VPDB)G. ruber 18O (‰VPDB)a c b d

PAGE 72

66 Figure 4. Comparison of GOM 18OGOM and G. ruber 13C versus age (ka) Figure 4. Comparison of 18OGOM (a) and G. ruber 13C (b) over the past 155 ky. 13C is plotted inversely to show negative excursions in the same direction as 18OGOM. Three of the five freshwater events (g ray bars) (as defined in text) show a positive relationship with 13C. Age (ka)18OGOM(‰VSMOW13C (‰VPDB)a b Age (ka)18OGOM(‰VSMOW13C (‰VPDB)a b

PAGE 73

67 Appendix B: Raw Data Table 2. Raw G. ruber stable isotope, geochemical data with calculated proxy SST, 18OSW, and 18OGOM data for ODP 625 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) B 1H1 0-1 0 4.75 3.93 -1.23 1.29 3.960 24.19 0.98 B 1H1 5-6 0.05 4.95 13.32 -1.57 1.52 4.224 24.91 0.78 B 1H1 15-16 0.15 5.35 5.28 -1.16 0.99 4.016 24.34 1.08 B 1H1 20-21 0.2 5.55 6.27 -1.35 1.39 3.901 24.02 0.82 B 1H1 25-26 0.25 5.75 7.61 -1.25 1.29 4.658 25.99 1.33 B 1H1 30-31 0.3 5.95 5.42 -1.26 1.45 4.495 25.60 1.24 B 1H1 35-36 0.35 6.15 7.78 -1.42 1.20 4.532 25.69 1.10 B 1H1 41-42 0.41 6.39 10.33 -1.41 1.27 4.001 24.30 0.82 B 1H1 60-61 0.6 7.16 6.04 -1.66 0.89 4.370 25.28 0.77 B 1H1 65-66 0.65 7.36 13.56 -1.36 1.09 4.909 26.58 1.34 B 1H1 75-76 0.75 7.76 6.07 -1.41 1.34 4.730 26.16 1.21 B 1H1 90-91 0.9 8.36 9.30 -1.00 0.87 4.285 25.07 1.39 B 1H1 95-96 0.95 8.56 6.14 -1.12 0.88 3.835 23.83 1.01 B 1H1 100-101 1 8.75 3.05 -0.88 0.82 3.876 23.95 1.28 B 1H1 105-106 1.05 8.95 5.10 -0.56 0.70 3.710 23.46 1.49 B 1H1 110-111 1.1 9.14 3.09 -1.03 0.56 3.852 23.88 1.11 B 1H1 115-116 1.15 9.33 3.68 -0.50 0.57 3.328 22.26 1.30 B 1H1 120-121 1.2 9.50 5.20 0.00 0.10 3.938 24.13 2.19 B 1H1 140-141 1.4 10.19 3.12 0.09 0.89 3.192 21.79 1.80 B 1H2 5-6 1.55 10.71 2.94 0.22 0.70 2.995 21.09 1.78 B 1H2 10-11 1.6 10.88 3.97 0.42 0.62 2.985 21.05 1.98 B 1H2 15-16 1.65 11.05 2.32 0.80 0.73 3.330 22.26 2.60 B 1H2 20-22 1.7 11.23 4.82 0.20 0.77 3.355 22.35 2.02 B 1H2 25-27 1.75 11.40 3.46 0.85 1.00 2.841 20.50 2.28 B 1H2 30-31 1.8 11.57 2.60 0.73 0.98 3.087 21.42 2.36 B 1H2 35-37 1.85 11.74 1.83 0.42 0.88 2.742 20.10 1.78 B 1H2 40-41 1.9 11.91 1.91 0.67 0.87 3.174 21.73 2.37 B 1H2 45-46 1.95 12.09 2.44 0.38 0.78 3.497 22.81 2.30 B 1H2 50-52 2 12.26 1.33 0.05 0.72 2.789 20.29 1.44 B 1H2 56-57 2.06 12.47 2.17 -0.11 0.35 3.335 22.28 1.70 B 1H2 60-61 2.1 12.60 2.11 0.71 0.74 2.906 20.75 2.20 B 1H2 66-68 2.16 12.81 0.77 -0.01 0.77 2.582 19.44 1.21 B 1H2 75-77 2.25 13.12 0.14 0.54 1.00 B 1H2 85-87 2.35 13.46 0.11 B 1H2 115-117 2.65 14.50 1.73 0.53 0.69 2.425 18.74 1.60 B 1H2 120-122 2.7 14.67 1.66 0.17 0.64 B 1H2 126-128 2.76 14.88 0.61 0.61 0.70 B 1H2 136-138 2.86 15.22 0.25 B 1H3 15-16 3.15 16.22 0.92 -0.09 0.68

PAGE 74

68 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) B 1H3 20-22 3.2 16.39 0.61 B 1H3 25-27 3.25 16.56 0.75 0.84 0.97 2.715 19.99 2.17 B 1H3 30-32 3.3 16.73 0.40 0.34 0.81 3.233 21.94 2.08 B 1H3 35-37 3.35 16.91 0.19 0.91 0.95 B 1H3 40-42 3.4 17.08 0.32 0.66 0.36 B 1H3 46-48 3.46 17.29 1.52 0.59 0.47 3.045 21.27 2.18 B 1H3 50-52 3.5 17.42 0.89 0.50 0.48 2.723 20.03 1.84 B 1H3 55-57 3.55 17.59 0.39 0.69 0.24 B 1H3 60-62 3.6 17.77 0.36 0.92 0.36 B 1H3 65-67 3.65 17.94 0.56 1.11 -0.04 B 1H3 70-72 3.7 18.11 1.00 0.70 0.36 2.963 20.96 2.23 B 1H3 80-82 3.8 18.46 0.12 B 1H3 85-87 3.85 18.63 0.31 0.41 0.82 B 1H3 105-107 4.05 19.32 0.11 B 1H3 110-112 4.1 19.49 0.62 0.81 0.74 B 1H3 115-117 4.15 19.66 0.10 0.11 0.83 B 1H3 125-127 4.25 20.00 0.94 0.76 0.35 B 1H3 130-132 4.3 20.18 0.79 0.69 0.72 2.920 20.80 2.19 B 1H3 140-142 4.4 20.52 0.48 1.27 0.76 B 1H3 145-147 4.45 20.69 2.17 -0.04 1.29 3.067 21.35 1.57 B 1H4 10-12 4.6 21.21 0.23 0.22 0.59 B 1H4 15-17 4.65 21.38 0.39 B 1H4 20-22 4.7 21.55 0.57 0.86 0.53 2.891 20.69 2.34 B 1H4 25-27 4.75 21.73 0.90 0.70 0.65 3.278 22.09 2.47 B 1H4 30-32 4.8 21.90 0.39 0.72 0.38 B 1H4 35-37 4.85 22.07 0.22 0.58 0.49 B 1H4 40-42 4.9 22.24 0.59 0.76 0.72 2.813 20.39 2.17 B 1H4 45-47 4.95 22.41 0.28 0.74 0.65 B 1H4 51-53 5.01 22.62 0.24 0.90 0.63 B 1H4 56-58 5.06 22.78 0.35 0.60 3.053 21.30 1.95 C 1H1 15-16 5.15 22.94 -0.23 0.79 C 1H1 20-21 5.2 23.02 0.08 0.55 C 1H1 25-26 5.25 23.11 -0.06 0.80 2.908 20.76 1.43 C 1H1 30-31 5.3 23.20 -0.11 0.33 C 1H1 35-36 5.35 23.28 C 1H1 40-41 5.4 23.37 -0.41 0.77 3.103 21.48 1.24 C 1H1 45-46 5.45 23.46 0.42 0.59 2.899 20.72 1.90 C 1H1 50-51 5.5 23.55 0.23 0.72 3.030 21.22 1.82 C 1H1 55-60 5.55 23.63 -0.37 0.41 2.931 20.85 1.14 C 1H1 60-61 5.6 23.72 -0.26 0.83 2.882 20.66 1.21 C 1H1 65-66 5.65 23.81 -0.47 0.44 3.001 21.11 1.09 C 1H1 70-71 5.7 23.89 0.09 0.64 3.340 22.30 1.90 C 1H1 75-76 5.75 23.98 -0.85 0.38 3.046 21.27 0.75 C 1H1 80-81 5.8 24.07 -0.29 0.72 2.797 20.33 1.11 C 1H1 85-86 5.85 24.15 2.923 20.81 C 1H1 90-91 5.9 24.24 -0.28 0.73 C 1H1 95-96 5.95 24.32 -0.28 0.97 C 1H1 100-101 6 24.41 -0.34 0.79 3.181 21.75 1.36 C 1H1 105-106 6.05 24.49 -0.10 0.93 3.461 22.69 1.79 C 1H1 110-111 6.1 24.58 0.18 0.86 C 1H1 115-116 6.15 24.66 -0.43 0.58 3.113 21.51 1.22

PAGE 75

69 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) C 1H1 120-121 6.2 24.75 0.46 0.64 3.187 21.78 2.16 C 1H1 125-126 6.25 24.83 0.05 0.58 3.059 21.32 1.65 C 1H1 130-131 6.3 24.92 0.81 0.67 3.034 21.23 2.39 C 1H1 135-136 6.35 25.00 0.35 0.69 C 1H1 140-141 6.4 25.09 0.15 0.80 2.979 21.03 1.70 C 1H1 145-146 6.45 25.15 3.285 22.11 1.77 C 1H1 150-151 6.5 25.21 3.362 22.37 1.83 C 1H2 4-5 6.55 25.28 0.22 0.73 C 1H2 9-10 6.6 25.46 -0.10 0.93 2.845 20.51 1.34 C 1H2 14-15 6.65 25.64 0.04 0.82 3.415 22.54 1.90 C 1H2 19-20 6.7 25.82 3.151 21.65 1.68 C 1H2 24-25 6.75 26.01 C 1H2 29-30 6.8 26.19 0.34 0.63 C 1H2 34-35 6.85 26.37 0.07 0.54 3.239 21.95 1.81 C 1H2 39-40 6.9 26.55 -0.06 0.76 4.171 24.77 2.27 C 1H2 44-45 6.95 26.74 -0.13 0.72 3.904 24.03 2.04 C 1H2 49-50 7 26.92 C 1H2 54-55 7.05 27.10 3.246 21.98 C 1H2 58-59 7.09 27.25 0.53 0.90 C 1H2 65-66 7.16 27.50 -1.06 1.26 C 1H2 69-70 7.2 27.65 0.13 0.44 4.095 24.56 2.41 C 1H2 74-75 7.25 27.83 0.13 0.70 3.322 22.24 1.93 C 1H2 79-80 7.3 28.01 -0.37 0.92 3.308 22.19 1.42 C 1H2 84-85 7.35 28.20 -0.80 0.62 C 1H2 89-90 7.4 28.38 0.08 0.77 3.174 21.73 1.77 C 1H2 94-95 7.45 28.56 C 1H2 99-100 7.5 28.75 -0.28 0.75 3.513 22.86 1.65 C 1H2 104-105 7.55 28.93 3.576 23.05 1.97 C 1H2 109-110 7.6 29.11 0.02 0.60 3.596 23.12 2.00 C 1H2 114-115 7.65 29.29 0.38 0.48 3.474 22.73 2.28 C 1H2 119-120 7.7 29.48 0.10 0.35 3.212 21.86 1.82 C 1H2 124-125 7.75 29.66 0.92 0.41 3.358 22.36 2.74 C 1H2 129-130 7.8 29.84 0.49 0.77 C 1H2 134-135 7.85 30.02 -0.10 0.79 3.341 22.30 1.71 C 1H2 139-140 7.9 30.21 0.44 0.58 C 1H2 144-145 7.95 30.39 0.54 0.47 3.770 23.64 2.63 C 1H2 149-150 8 30.57 0.38 0.70 3.413 22.54 2.24 C 1H3 5-6 8.05 30.78 0.34 1.02 3.970 24.22 2.55 C 1H3 10-11 8.1 30.99 0.66 0.96 3.348 22.32 2.48 C 1H3 15-16 8.15 31.19 0.14 0.71 3.608 23.15 2.13 C 1H3 20-21 8.2 31.40 0.03 0.72 3.205 21.84 1.75 C 1H3 25-26 8.25 31.61 0.01 0.78 3.314 22.21 1.80 C 1H3 30-31 8.3 31.82 -0.40 0.91 3.672 23.35 1.63 C 1H3 35-36 8.35 32.03 0.05 0.82 3.352 22.34 1.87 C 1H3 40-41 8.4 32.24 -0.18 0.76 3.207 21.85 1.53 C 1H3 45-46 8.45 32.54 -0.25 0.97 3.550 22.97 1.70 C 1H3 50-51 8.5 32.85 -0.26 0.76 3.483 22.76 1.65 C 1H3 55-56 8.55 33.16 -0.49 1.07 3.133 21.59 1.18 C 1H3 59-60 8.59 33.40 -0.20 0.79 3.382 22.44 1.64 C 1H3 65-66 8.65 33.77 0.04 0.80 3.433 22.60 1.91 C 1H3 70-71 8.7 34.08 -0.17 0.74 3.425 22.58 1.70

PAGE 76

70 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) C 1H3 75-76 8.75 34.39 -0.71 0.93 3.565 23.02 1.25 C 1H3 80-81 8.8 34.69 1.80 0.09 0.94 3.345 22.31 1.90 C 1H3 85-86 8.85 35.00 1.79 -0.33 0.75 3.849 23.87 1.81 C 1H3 90-91 8.9 35.31 2.56 -0.18 0.78 3.643 23.26 1.84 C 1H3 95-96 8.95 35.61 2.77 -0.29 0.83 3.368 22.39 1.54 C 1H3 100-101 9 35.92 3.44 -0.33 0.73 3.246 21.98 1.41 C 1H3 105-106 9.05 36.34 2.13 -0.41 0.64 3.823 23.80 1.71 C 1H3 110-111 9.1 36.76 3.17 -0.69 0.78 3.643 23.26 1.32 C 1H3 115-116 9.15 37.17 2.00 -0.41 1.21 3.072 21.37 1.21 C 1H3 120-121 9.2 37.59 1.34 -0.17 0.78 2.809 20.37 1.24 C 1H3 125-126 9.25 38.01 2.69 -0.09 0.80 3.121 21.54 1.56 C 1H3 130-131 9.3 38.42 3.54 0.41 0.75 2.851 20.54 1.85 C 1H3 135-136 9.35 38.84 3.59 -0.16 0.83 3.288 22.12 1.61 C 1H3 140-141 9.4 39.26 4.24 -0.17 0.86 3.285 22.11 1.61 C 1H3 145-146 9.45 39.67 3.80 -0.12 0.81 3.011 21.14 1.45 C 1H3 150-151 9.5 40.09 4.61 0.15 0.73 3.160 21.68 1.83 C 1H4 4-5 9.55 40.51 2.89 0.11 0.96 3.012 21.15 1.68 C 1H4 9-10 9.6 40.93 3.18 0.23 1.14 3.285 22.11 2.00 C 1H4 14-15 9.65 41.34 3.66 -0.43 0.66 3.046 21.27 1.17 C 1H4 19-20 9.7 41.76 5.22 -0.09 0.83 3.474 22.73 1.81 C 1H4 24-25 9.75 42.18 4.91 0.12 0.78 3.216 21.88 1.85 C 1H4 29-30 9.8 42.59 5.81 0.34 0.69 3.430 22.59 2.21 C 1H4 34-35 9.85 43.01 5.40 0.00 0.89 3.081 21.40 1.63 C 1H4 39-40 9.9 43.43 4.75 0.08 0.90 3.101 21.47 1.72 C 1H4 44-45 9.95 43.85 3.02 -0.36 0.87 3.624 23.20 1.63 C 1H4 49-50 10 44.26 1.50 -0.26 0.69 3.584 23.08 1.71 C 1H4 54-55 10.05 44.68 1.24 -0.03 0.95 3.447 22.65 1.85 C 1H4 59-60 10.1 45.10 3.13 0.14 0.64 3.158 21.68 1.82 C 1H4 64-65 10.15 45.51 1.66 -0.08 0.43 3.467 22.71 1.82 C 1H4 69-70 10.2 45.93 1.55 0.07 0.34 3.163 21.69 1.75 C 1H4 74-75 10.25 46.35 1.68 0.07 1.03 C 1H4 79-80 10.3 46.77 1.59 -0.07 0.34 3.132 21.58 1.59 C 1H4 84-85 10.35 47.18 1.28 0.15 0.50 3.032 21.22 1.74 C 1H4 89-90 10.4 47.60 1.70 -0.02 0.68 3.418 22.55 1.84 C 1H4 94-95 10.45 48.02 1.83 -0.08 0.72 3.815 23.77 2.03 C 1H4 99-100 10.5 48.43 1.71 0.22 0.74 3.148 21.64 1.89 C 1H4 104-105 10.55 48.85 1.52 -0.26 0.69 3.440 22.62 1.62 C 1H4 109-110 10.6 49.27 5.37 -0.55 0.74 3.270 22.06 1.21 C 1H4 114-115 10.65 49.69 6.96 -0.36 0.65 3.157 21.67 1.32 C 1H4 119-120 10.7 50.10 7.30 -0.20 0.83 3.463 22.70 1.69 C 1H4 124-125 10.75 50.52 8.22 -0.70 0.64 3.544 22.96 1.25 C 1H4 129-130 10.8 50.94 6.68 -0.64 0.81 3.981 24.25 1.58 C 1H4 134-135 10.85 51.35 7.56 -0.37 0.79 3.610 23.16 1.62 C 1H4 139-140 10.9 51.77 4.75 -0.42 0.66 3.863 23.91 1.73 C 1H4 144-145 10.95 52.19 3.45 -0.02 0.73 4.089 24.55 2.26 C 1H4 149-150 11 52.60 4.50 -0.08 0.61 4.009 24.32 2.16 C 1H5 5-6 11.05 53.02 4.03 -0.14 0.64 3.512 22.85 1.78 C 1H5 10-11 11.1 53.44 4.91 -0.19 0.57 3.531 22.92 1.75 C 1H5 15-16 11.15 53.86 5.47 -0.29 0.66 3.363 22.37 1.54 C 1H5 20-21 11.2 54.27 6.46 -0.10 0.52 3.515 22.86 1.83 C 1H5 25-26 11.25 54.69 6.27 -0.25 0.54 3.355 22.35 1.57

PAGE 77

71 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) C 1H5 30-31 11.3 55.11 4.31 -0.48 0.74 3.476 22.74 1.43 C 1H5 35-36 11.35 55.52 2.13 -0.22 0.60 3.196 21.81 1.49 C 1H5 40-41 11.4 55.94 3.18 0.12 0.78 3.520 22.88 2.06 C 1H5 45-46 11.45 56.36 1.70 -0.35 0.68 3.617 23.18 1.64 C 1H5 50-51 11.5 56.78 2.65 -0.34 0.48 3.647 23.27 1.68 C 1H5 55-56 11.55 57.19 3.44 -0.07 0.67 3.668 23.34 1.96 C 1H5 60-61 11.6 57.61 3.23 -0.25 0.77 3.328 22.26 1.56 C 1H5 65-66 11.65 58.03 3.26 -0.79 0.86 4.566 25.77 1.75 C 1H5 70-71 11.7 58.44 4.13 -0.29 0.73 3.957 24.18 1.91 C 1H5 75-76 11.75 58.86 3.69 -0.35 0.70 3.542 22.95 1.60 C 1H5 80-81 11.8 59.31 4.99 0.00 0.56 3.875 23.95 2.16 C 1H5 85-86 11.85 59.86 3.53 0.06 0.75 3.591 23.10 2.04 C 1H5 90-91 11.9 60.41 3.07 0.46 0.71 3.708 23.46 2.51 C 1H5 95-96 11.95 60.96 2.58 0.83 0.58 3.481 22.76 2.73 C 1H5 100-101 12 61.52 2.87 0.51 0.44 3.314 22.21 2.30 C 1H5 105-106 12.05 62.07 2.02 0.67 0.44 4.008 24.32 2.90 C 1H5 110-111 12.1 62.62 3.80 0.68 0.38 3.602 23.13 2.67 C 1H5 115-116 12.15 63.18 2.78 0.28 0.57 3.639 23.25 2.29 C 1H5 120-121 12.2 63.73 2.72 0.02 0.72 3.620 23.19 2.02 C 1H5 125-126 12.25 64.28 2.57 -0.14 0.78 3.849 23.87 2.00 C 1H5 130-131 12.3 64.83 1.91 -0.04 0.84 3.214 21.87 1.68 C 1H5 135-136 12.35 65.39 1.57 -0.03 0.87 3.590 23.10 1.95 C 1H5 140-141 12.4 65.94 1.44 0.33 0.68 3.605 23.15 2.32 C 1H5 145-146 12.45 66.49 1.14 0.36 0.72 3.918 24.07 2.54 C 1H5 150-151 12.5 67.05 0.29 0.36 0.53 3.531 22.91 2.30 C 1H6 4-5 12.55 67.60 2.05 0.17 0.65 3.652 23.29 2.19 C 1H6 9-10 12.6 68.15 1.72 0.24 0.68 3.687 23.40 2.28 C 1H6 14-15 12.65 68.70 1.49 0.25 0.65 3.587 23.09 2.22 C 1H6 19-20 12.7 69.26 1.67 0.58 0.72 3.208 21.85 2.30 C 1H6 24-25 12.75 69.81 1.83 0.46 0.68 3.498 22.81 2.38 C 1H6 29-30 12.8 70.36 2.24 0.33 0.90 3.714 23.48 2.39 C 1H6 34-35 12.85 70.92 2.15 0.29 0.65 3.245 21.98 2.04 C 1H6 39-40 12.9 71.47 3.60 0.21 0.63 3.278 22.09 1.97 C 1H6 44-45 12.95 72.02 3.24 0.95 0.67 3.244 21.97 2.69 C 1H6 49-50 13 72.58 2.55 0.44 0.69 3.825 23.80 2.56 C 1H6 54-55 13.05 73.13 2.97 0.53 0.83 3.402 22.50 2.38 C 1H6 59-60 13.1 73.68 3.28 0.45 0.64 3.715 23.48 2.51 C 1H6 64-65 13.15 74.23 4.27 0.30 0.88 3.363 22.37 2.13 C 1H6 69-70 13.2 74.78 2.98 0.38 1.01 3.571 23.04 2.35 C 1H6 74-75 13.25 75.33 5.04 0.05 0.92 3.227 21.92 1.78 C 1H6 79-80 13.3 75.87 6.24 -0.55 1.02 3.476 22.74 1.35 C 1H6 84-85 13.35 76.42 3.80 -0.13 0.91 3.601 23.13 1.86 C 1H6 89-90 13.4 76.96 3.81 -0.41 0.96 3.657 23.30 1.62 C 1H6 94-95 13.45 77.51 5.87 -0.37 1.39 3.581 23.07 1.60 C 1H6 99-100 13.5 78.05 8.13 -0.43 1.30 3.989 24.27 1.79 C 1H6 104-105 13.55 78.60 10.75 -0.16 0.95 C 1H6 109-110 13.6 79.14 11.06 -0.42 1.16 4.177 24.78 1.91 C 1H6 114-115 13.65 79.68 8.40 -0.57 1.36 3.874 23.94 1.58 C 1H6 119-120 13.7 80.23 12.44 -0.47 1.31 3.898 24.01 1.70 C 1H6 124-125 13.75 80.77 14.45 -0.42 1.27 3.900 24.02 1.75 C 1H6 129-130 13.8 81.32 10.41 -0.54 1.40 4.368 25.28 1.89

PAGE 78

72 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) C 1H6 134-135 13.85 82.38 17.16 -0.69 1.46 3.944 24.14 1.51 C 1H6 139-140 13.9 83.44 13.07 -0.60 1.37 4.025 24.37 1.64 C 1H6 144-145 13.95 84.50 10.86 -0.66 1.30 3.911 24.05 1.52 C 1H6 149-150 14 85.56 10.80 -0.80 0.69 3.861 23.91 1.35 C 1H7 4-5 14.05 86.62 11.93 -0.69 1.09 3.858 23.90 1.45 C 1H7 9-10 14.1 87.68 9.62 -1.04 1.18 3.946 24.15 1.16 C 1H7 14-15 14.15 88.74 5.80 -0.48 1.37 3.901 24.02 1.69 C 1H7 19-20 14.2 89.80 4.27 -0.69 0.99 4.194 24.83 1.65 C 1H7 24-25 14.25 90.86 7.03 -0.60 1.02 3.732 23.53 1.47 C 1H7 29-30 14.3 91.93 11.52 -0.49 0.96 3.734 23.54 1.58 C 1H7 34-35 14.35 93.02 16.38 -0.60 0.84 4.036 24.40 1.65 C 1H7 39-40 14.4 94.11 15.55 -1.03 0.87 4.081 24.52 1.24 C 1H7 44-45 14.45 95.20 16.92 -0.91 0.82 3.923 24.08 1.27 C 1H7 49-50 14.5 96.29 15.03 -0.81 1.01 3.598 23.12 1.17 C 1H7 54-55 14.55 97.38 20.98 -0.65 0.91 4.008 24.32 1.58 C 1H7 59-60 14.6 98.47 21.48 -0.96 0.93 4.050 24.44 1.30 C 1H7 64-65 14.65 99.57 19.50 -1.03 1.07 4.338 25.20 1.38 C 1H7 69-70 14.7 100.66 18.34 -0.96 0.75 4.054 24.45 1.29 B 2H5 82-83 14.72 101.10 -0.48 0.92 4.117 24.62 1.82 B 2H5 84-85 14.74 101.54 -1.11 1.02 4.798 26.32 1.54 B 2H5 86-87 14.76 101.98 -1.14 0.81 4.338 25.20 1.27 B 2H5 88-89 14.78 102.42 -0.93 1.00 4.288 25.07 1.46 B 2H5 90-91 14.8 102.85 -0.70 0.72 4.182 24.79 1.63 B 2H5 92-93 14.82 103.29 -1.20 0.85 4.350 25.23 1.22 B 2H5 94-95 14.84 103.73 -0.54 0.82 3.691 23.41 1.50 B 2H5 96-97 14.86 104.17 -0.98 0.67 4.383 25.32 1.46 B 2H5 98-99 14.88 104.61 -0.51 0.81 4.160 24.74 1.81 B 2H5 100-101 14.9 105.05 -1.12 0.89 4.013 24.34 1.11 B 2H5 102-103 14.92 105.49 -1.07 0.83 4.096 24.56 1.22 B 2H5 104-105 14.94 105.93 -0.69 1.11 4.247 24.97 1.67 B 2H5 105-106 14.95 106.15 -0.65 0.62 3.914 24.06 1.53 B 2H5 106-107 14.96 106.37 -1.11 0.41 3.929 24.10 1.08 B 2H5 108-109 14.98 106.81 -0.64 0.50 4.054 24.45 1.62 B 2H5 110-111 15 107.25 -1.28 0.16 3.947 24.15 0.91 B 2H5 112-113 15.02 107.69 -0.84 0.23 3.926 24.09 1.34 B 2H5 114-115 15.04 108.13 -1.30 0.40 4.209 24.86 1.05 B 2H5 116-117 15.06 108.57 -1.53 0.69 4.068 24.49 0.74 B 2H5 118-119 15.08 109.01 -0.99 0.63 4.170 24.76 1.33 B 2H5 120-121 15.1 109.45 -1.18 0.58 3.975 24.23 1.03 B 2H5 122-123 15.12 109.89 -1.47 0.33 4.531 25.68 1.04 B 2H5 124-125 15.14 110.33 -1.38 0.98 5.074 26.94 1.40 B 2H5 126-127 15.16 110.77 -1.26 1.08 4.525 25.67 1.25 B 2H5 128-129 15.18 111.21 -1.18 0.55 4.183 24.80 1.15 B 2H5 130-131 15.2 111.66 -1.26 0.62 4.379 25.31 1.18 B 2H5 132-133 15.22 112.10 -1.51 0.69 4.498 25.61 0.99 B 2H5 134-135 15.24 112.54 -1.44 0.94 4.199 24.84 0.90 B 2H5 136-137 15.26 112.98 -1.46 0.88 4.434 25.44 1.01 B 2H5 138-139 15.28 113.42 -1.31 0.89 4.323 25.16 1.09 B 2H6 0-1 15.4 116.06 -1.72 0.75 4.781 26.28 0.92 B 2H6 3-4 15.43 116.72 -1.88 0.98 4.940 26.65 0.84 B 2H6 6-7 15.46 117.38 -1.18 1.38 5.333 27.50 1.71

PAGE 79

73 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) B 2H6 9-10 15.49 118.04 -1.88 0.79 4.964 26.70 0.85 B 2H6 12-13 15.52 118.68 -1.69 0.79 5.421 27.68 1.24 B 2H6 15-16 15.55 119.31 -1.73 0.84 4.676 26.04 0.86 B 2H6 18-19 15.58 119.94 -1.85 0.77 4.848 26.44 0.83 B 2H6 21-22 15.61 120.58 -2.64 0.30 5.069 26.93 0.13 B 2H6 24-25 15.64 121.21 -1.81 0.70 4.939 26.64 0.90 B 2H6 27-28 15.67 121.84 -1.63 0.75 5.067 26.93 1.14 B 2H6 30-31 15.7 122.47 -1.62 0.57 4.313 25.14 0.79 B 2H6 33-34 15.73 123.10 -1.99 0.87 4.591 25.83 0.56 B 2H6 36-37 15.76 123.74 -1.98 0.52 4.330 25.18 0.43 B 2H6 39-40 15.79 124.37 -0.93 0.77 4.443 25.47 1.54 B 2H6 42-43 15.82 125.00 -0.46 0.80 4.222 24.90 1.89 B 2H6 45-46 15.85 125.63 -0.58 0.53 4.038 24.40 1.67 B 2H6 48-49 15.88 126.26 -0.43 1.14 3.945 24.15 1.77 B 2H6 51-52 15.91 126.89 -0.71 0.85 3.760 23.61 1.38 B 2H6 54-55 15.94 127.53 0.50 0.82 4.158 24.73 2.82 B 2H6 57-58 15.97 128.16 4.06 -0.21 0.89 3.756 23.60 1.87 B 2H6 60-61 16 128.79 3.30 0.62 0.32 3.635 23.24 2.63 B 2H6 63-64 16.03 129.32 1.60 4.019 24.35 B 2H6 66-67 16.06 129.86 4.20 0.49 0.60 3.676 23.36 2.52 B 2H6 69-70 16.09 130.39 2.51 0.21 0.56 3.806 23.75 2.32 B 2H6 72-73 16.12 130.92 2.27 0.10 0.42 4.098 24.57 2.38 B 2H6 75-76 16.15 131.46 6.68 0.20 0.57 4.334 25.19 2.61 B 2H6 78-79 16.18 131.99 7.13 0.48 0.89 3.721 23.50 2.54 B 2H6 81-82 16.21 132.53 8.04 0.78 0.65 3.371 22.40 2.61 B 2H6 85-86 16.25 133.24 8.84 0.05 0.51 3.622 23.20 2.05 B 2H6 88-89 16.28 133.77 8.30 0.75 0.59 3.308 22.19 2.54 B 2H6 91-92 16.31 134.31 8.64 0.39 0.53 3.558 23.00 2.35 B 2H6 94-95 16.34 134.84 2.55 -0.08 0.50 3.503 22.82 1.84 B 2H6 97-98 16.37 135.37 5.93 0.27 0.44 3.388 22.46 2.11 B 2H6 100-101 16.4 135.91 6.16 0.55 0.44 3.551 22.98 2.50 B 2H6 103-104 16.43 136.44 6.77 0.09 0.44 3.937 24.12 2.28 B 2H6 105-106 16.45 136.80 6.77 0.90 0.66 3.315 22.21 2.69 B 2H6 108-109 16.48 137.33 7.36 0.56 0.43 3.273 22.07 2.32 B 2H6 111-112 16.51 137.86 7.35 0.84 0.59 3.570 23.04 2.80 B 2H6 114-115 16.54 138.40 6.62 0.48 0.51 3.561 23.01 2.44 B 2H6 117-118 16.57 138.93 5.96 0.38 0.80 3.583 23.08 2.35 B 2H6 120-121 16.6 139.46 5.98 0.89 0.67 3.702 23.44 2.94 B 2H6 123-124 16.63 140.00 7.73 0.67 0.86 4.009 24.33 2.90 B 2H6 126-127 16.66 140.53 7.33 0.95 0.63 3.905 24.03 3.12 B 2H6 129-130 16.69 141.07 5.96 0.54 0.62 3.667 23.34 2.57 B 2H6 132-133 16.72 141.60 6.42 0.54 0.60 3.597 23.12 2.52 B 2H6 135-136 16.75 142.13 5.21 0.29 0.62 3.793 23.71 2.40 B 2H6 138-139 16.78 142.67 6.75 0.49 0.41 3.575 23.05 2.45 B 2H6 141-142 16.81 143.20 4.18 0.42 0.59 3.696 23.42 2.47 B 2H6 144-145 16.84 143.73 6.32 0.15 0.59 3.595 23.11 2.13 B 2H6 147-148 16.87 144.27 4.57 0.49 0.30 3.562 23.01 2.45 B 2H6 149-150 16.89 144.62 5.56 0.71 0.44 3.855 23.89 2.86 B 2H7 5-6 16.95 145.69 6.03 0.73 0.46 3.373 22.41 2.56 B 2H7 10-11 17 146.58 4.85 0.49 0.64 3.665 23.33 2.52 B 2H7 15-16 17.05 147.47 7.26 0.54 0.78 3.666 23.33 2.57

PAGE 80

74 Sample ID Depth (mbsf) Age (ka) % Coarse 18OC (‰ VPDB) 13C (‰ VPDB) Mg/Ca (mmol/ mol) SST (C) 18OSW (‰ VSMOW) B 2H7 20-21 17.1 148.36 5.35 0.57 0.55 3.748 23.58 2.65 B 2H7 25-26 17.15 148.79 5.47 0.40 0.33 3.640 23.25 2.41 B 2H7 30-31 17.2 149.23 4.18 0.49 0.61 3.740 23.55 2.56 B 2H7 35-36 17.25 149.66 1.73 0.10 0.71 3.930 24.10 2.29 B 2H7 40-41 17.3 150.10 4.10 0.27 0.28 3.847 23.87 2.41 B 2H7 45-46 17.35 150.53 5.34 0.64 0.44 3.511 22.85 2.57 B 2H7 50-51 17.4 150.96 3.81 0.78 0.37 3.293 22.14 2.56 B 2H7 55-56 17.45 151.40 4.65 0.79 0.31 3.717 23.48 2.85 B 2H7 60-61 17.5 151.83 4.36 0.57 0.37 3.369 22.39 2.40 B CCW 2-3 17.56 152.35 2.69 0.83 0.41 3.203 21.83 2.54 B CCW 7-8 17.61 152.78 2.81 0.82 0.41 3.727 23.52 2.88 B CCW 12-13 17.66 153.22 5.38 0.80 0.63 3.882 23.97 2.96 B CCW 17-18 17.71 153.65 6.73 0.81 0.59 3.689 23.40 2.85 B CCW 22-23 17.76 154.09 6.77 0.51 0.54 3.782 23.68 2.60 B CCW 27-28 17.81 154.52 10.33 0.49 0.64 3.578 23.06 2.46


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200385Ka 4500
controlfield tag 001 002001034
005 20090427121617.0
007 cr mnu|||uuuuu
008 090427s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002550
035
(OCoLC)319431635
040
FHM
c FHM
049
FHMM
066
(S
090
GC11.2 (Online)
1 100
Whitaker, Jessica L.
0 245
Orbital- to millennial-scale variability in Gulf of Mexico sea surface temperature and salinity during the late Pleistocene
h [electronic resource] /
by Jessica L. Whitaker.
260
[Tampa, Fla] :
b University of South Florida,
2008.
500
Title from PDF of title page.
Document formatted into pages; contains 74 pages.
502
Thesis (M.S.)--University of South Florida, 2008.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Benjamin P. Flower, Ph.D.
520
ABSTRACT: Sea surface temperature (SST) reconstructions from the low latitudes indicate the tropics/subtropics warmed significantly before glacial-interglacial decreases in global ice volume, suggesting the importance of tropical and subtropical climate in driving glacial terminations. ODP Site 625, drilled at a water depth of 889 m near De Soto Canyon in the Gulf of Mexico (GOM), provides continuous records of marine isotope stages (MIS) 1-6 sampled at a mean temporal resolution of 400 years. Age control is based on 8 AMS radiocarbon dates, marine isotope stratigraphy, and Foraminifera datum levels. Results from Globigerinoides ruber (white variety) Mg/Ca-SST indicate a rise of 4.4 C from last glacial maximum to modern conditions and a 3.2 C rise from the penultimate glaciation to the last interglaciation. However, model results suggest reduced thermohaline circulation (THC) causes salt and heat build-up in the Atlantic Warm Pool.Paired G. ruber Mg/Ca-SST and O provide evidence of sub-millennial scale variability in GOM SST and SSS that is probably influenced by the strength of NADW production, as also observed in the Western Caribbean Sea. We test the idea that widespread abrupt climate change during the last glaciation caused by millennial scale fluctuations in the intensity of THC was modulated by Laurentide ice sheet (LIS) meltwater routed to the North Atlantic. To understand LIS melting dynamics and test the Meltwater Routing Hypothesis, we investigate the phasing of GOM SST and LIS freshwater events in relationship to high latitude climate. Estimated salinities from our multi-proxy approach suggest three freshwater events with a major freshwater influx from that occurred during Heinrich Event 2. This result confirms previous studies that suggested LIS summer melting during warmings in Antarctica. We also find a climate reversal during termination II from 130.4-128.4 ka.The initial rise in GOM SST at 132.1 ka of 2.9 C is followed by a cold reversal of 1.5 C at 130.4 ka for 2 ky and final increase to full interglacial warmth. The reversal in GOM SST is consonant with a pause in sea level rise and reduced NADW, suggesting a reduction in THC may have caused a global two-step deglaciation.
653
Tropics
Climate
Stable isotope
Salinity
690
Dissertations, Academic
z USF
x Marine Science
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2550