|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
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 Ka
controlfield tag 001 001795155
007 cr mnu|||uuuuu
008 070628s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001539
Hill, Heather W.
Abrupt climate change during the last glacial period :
b a Gulf of Mexico perspective
h [electronic resource] /
by Heather W. Hill.
[Tampa, Fla] :
University of South Florida,
ABSTRACT: Understanding the cause of abrupt climate change in the geologic past can help assess the potential magnitude and variability of future changes in regional and global climate. The research presented here focuses on some of the first records of hydrologic variability in the central North American continent during an interval of Marine Isotope Stage 3 (24-57 thousand years before present (ka)). Sediment core MD02-2551 from the Orca Basin, northern Gulf of Mexico, is used to document the first detailed melting history of the southern margin of the Laurentide Ice Sheet (LIS) during MIS 3, and to record terrestrial inputs from the Mississippi River related to changes in evaporation-precipitation over the mid-continent, from 28-45 ka.Paired measurements of oxygen isotopes and Mg/Ca-SST on the planktonic foraminifera Globigerinoides ruber (pink) are used to calculate the oxygen isotopic composition of seawater and test one of the key hypotheses for abrupt climate change. Five^ intervals of freshwater input from 28-45 ka do not match the abrupt Dansgaard-Oeschger temperature oscillations recorded in Greenland ice. Rather, summer melting of the LIS may have occurred during Antarctic warming and likely contributed to sea-level variability during MIS 3. A detailed assessment over one of the meltwater events, using the oxygen and carbon isotopic composition of G. ruber and the deeper dwelling Neogloboquadrina dutertrei, demonstrate that meltwater was confined to the surface layers and likely had an impact on the biological pump in the Gulf of Mexico. A similar oxygen isotopic composition of seawater record determined from the year-round white G. ruber suggests that melting was not limited to the warmest summer months. The timing of LIS meltwater input is decoupled from an interval of enhanced wet conditions over the North American continent and increased Mississippi River discharge, as shown by a suite of organic and sedimentologic proxies. Increasing summer^ insolation on the orbital scale may have led to a northward migration of the Intertropical Convergence Zone and an intensification and westward shift in the conical position of the Bermuda High, which shuttles moisture to the North American continent and contributes to flooding in the Mississippi River drainage basin.
Dissertation (Ph.D.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 116 pages.
Adviser: Benjamin P. Flower, Ph.D.
Marine isotope stage 3.
Laurentide ice sheet meltwater.
x Marine Science
t USF Electronic Theses and Dissertations.
Abrupt Climate Change During the Last Glacial Period: A Gulf of Mexico Perspective by Heather W. Hill A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Marine Science University of South Florida Major Professor: Benjamin P. Flower, Ph.D. David J. Hollander, Ph.D. Terrence M. Quinn, Ph.D. David W. Hastings, Ph.D. Albert C. Hine, Ph.D. Date of Approval: March 27, 2006 Keywords: Marine Isotope Stage 3, Laurentide Ice Sheet meltwater, Mississ ippi River, floods, ocean-continent interactions Copyright 2006, Heather W. Hill
Dedication I cannot express in words my gratitude towards my husband, Randy, who was a constant source of encouragement throughout this entire process. He was there through the good and the bad, to motivate and support, but also to keep me grounded and provide the reality check that is so often necessary. He truly was my rock. My family g ave me the strength and continual love and support that I needed to succeed.
Acknowledgments Several faculty members in the College of Marine Science served as mentors and helped guide my PhD education. Most important, I would like to thank my advisor, Dr. Benjamin Flower, for his support, his never-ending patience, and the many unique opportunities that he provided me as a graduate student. Dr. David Hollander encouraged me to think broadly and was always there to motivate and keep me positive. Dr. Terrence Quinn inspired me to be a better scientist through his high expectations of me. I also thank Dr. David Hasting particularly for his help with some of the analytical aspec ts of the project, and Dr. Albert Hine for encouraging me to think about things in the big picture. I thank all of my committee members for their various contributions along the way, for providing me with a sound knowledge base, helping me to grow academically, and instilling in me the confidence that is necessary to move forward as a scientis t. Their comments and suggestions have significantly improved the quality of the manuscripts that make up this dissertation. I am grateful to have been a part of the Paleoclimatology lab in the College of Marine Science throughout my graduate studies. The students, faculty and staff in the l ab provided a wonderful working environment that allowed for a sharing of ideas and encouraged academic growth. My research benefited tremendously from the numerous interactions I had with people in the lab. In particular, I had many scientific discussi ons and worked closely in the lab with Jenna LoDico and Jennifer Flannery. I am especially thankful to have been able to share this experience with fellow PhD students, Jennifer Smith and Hali Kilbourne. Financially, I have been supported by the VonRosenstiel Fellowship, the Paul L. Getting Fellowship, and the Gulf Oceanographic Charitable Trust Endowed Fellowship through the College of Marine Science. I also received two years of funding through the National Science Foundations GK-12 OCEANS program. Part of my research was supported by the National Science Foundation grant OCE-0318361 to B.P.F. and T.M.Q.
i Table of Contents List of Tables iii List of Figures iv Abstract vi 1. Introduction 1 1.1 Introduction 1 1.2 Climate-change paradigms 2 1.3 Cause of millennial-scale climate change 4 1.4 Climate controls on hydrologic variability 7 1.5 Orca Basin 9 1.6 Dissertation organization 9 2. Laurentide Ice Sheet meltwater and abrupt climate change during the 14 last glaciation 2.1 Abstract 14 2.2 Introduction 14 2.3 d 18 O and Mg/Ca analyses 15 2.4 Age model 16 2.5 Gulf of Mexico d 18 O of seawater 18 2.6 Conversion to sea-surface salinity 19 2.7 LIS routing hypothesis 21 2.8 Supplementary information 22 3. A multi-species approach to constraining Laurentide Ice Sheet meltwater input to the Gulf of Mexico during the last glacial period 32 3.1 Abstract 32 3.2 Introduction 32 3.3 Methods 36 3.4 Results and Discussion 37 3.4.1 d 18 O c records 37 3.4.2 d 18 O c gradients 39 3.4.3 Mg/Ca records 40 3.4.4 SST differences 41 3.4.5 d 18 O sw records 43 3.4.6 LIS meltwater input 44 3.4.7 d 13 C records 45
ii 3.4.8 Biological pump 47 3.5 Conclusions 48 4. Mississippi River flooding during the last glaciation 57 4.1 Abstract 57 4.2 Introduction 58 4.3 Scientific approach 60 4.3.1 Assessing Mississippi River discharge 60 4.3.2 Documenting Gulf of Mexico productivity 61 4.4 Sources of freshwater 62 4.4.1 Laurentide Ice Sheet meltwater 63 4.4.2 Atmospheric circulation patterns 64 4.5 Conclusions 65 5. Summary 70 References Cited 72 Appendix A: Inorganic analyses 87 Appendix B: Organic and sediment compositional analyses 110 About the Author End Page
iii List of Tables Table 1 Radiocarbon ages for MD02-2551. 24 Table 2 Volume reconstructions for d 18 O GOM record. 49 Table 2 Pink G. ruber d 18 O and Mg/Ca analyses. 89 Table 3 White G. ruber and N. dutertrei d 18 O, d 13 C, and Mg/Ca analyses. 100 Table 4 G. ruber d 18 O GOM calculations. 102 Table 5 Sediment composition and bulk organic d 13 C analyses. 112 Table 6 Low molecular weight n-alkane concentrations. 114 Table 7 High molecular weight n-alkane concentrations. 115 Table 8 d 13 C of high molecular weight n-alkanes 116
iv List of Figures Figure 1 Marine Isotope Stage 3 records. 11 Figure 2 Map of Orca Basin in the Gulf of Mexico showing location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) and the extent of the Laurentide Ice Sheet during MIS 3 12 Figure 3 Insolation at 30rN from 0-50 ka. 13 Figure 4 Map of Orca Basin in the Gulf of Mexico showing location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) and the extent of the Laurentide Ice Sheet during MIS 3. 25 Figure 5 Raw d 18 O c and Mg/Ca data and age model for MD02-2551. 26 Figure 6 Paired d 18 O c and Mg/Ca data on G. ruber from Orca Basin core MD02-2551 during MIS 3. 27 Figure 7 Comparison of Orca Basin d 18 O GOM during MIS 3 with ice core records. 28 Figure 8 Mixing model for the GOM during MIS 3. 29 Figure 9 GOM sea-surface salinity (SSS) reconstructions from 28-45 k.a. 30 Figure 10 Comparison of Orca Basin d 18 O GOM on the SFCP timescale during MIS 3 with ice core records. 31 Figure 11 Map of Orca Basin in the Gulf of Mexico showing location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) and the extent of the Laurentide Ice Sheet during MIS 3. 50 Figure 12 Comparison of Orca Basin core MD02-2551 d 18 O GOM determined from G. ruber (pink) during MIS 3 with ice core records. 51 Figure 13 Annual cycle of sea-surface temperature in the Gulf of Mexico. 52 Figure 14 Raw d 18 O c of G. ruber (pink), G. ruber (white) and N. dutertrei from Orca Basin core MD02-2551. 53
v Figure 15 Mg/Ca and SST data on pink and white G. ruber from Orca Basin core MD02-2551 during MIS 3. 54 Figure 16 d 18 O sw and d 18 O GOM of pink and white G. ruber from Orca Basin core MD02-2551. 55 Figure 17 d 13 C of G. ruber (pink), G. ruber (white) and N. dutertrei from Orca Basin core MD02-2551. 56 Figure 18 Map of Orca Basin in the Gulf of Mexico showing the location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) relative to the Mississippi River (MR) drainage basin. 67 Figure 19 Suite of organic and inorganic proxies from Orca Basin core MD02-2551 that collectively point to enhanced Mississippi River discharge and increased primary productivity in the Gulf of Mexico from 26.2-24.4 m (gray bar). 68 Figure 20 Comparison of the timing of Mississippi River flooding relative to LIS meltwater and summer insolation at 30N. 69
vi Abrupt Climate Change During the Last Glacial Period: A Gulf of Mexico Perspective Heather W. Hill ABSTRACT Understanding the cause of abrupt climate change in the geologic past can help assess the potential magnitude and variability of future changes in regional and globa l climate. The research presented here focuses on some of the first records of hydrolog ic variability in the central North American continent during an interval of Marine Is otope Stage 3 (24-57 thousand years before present (ka)). Sediment core MD02-2551 from the Orca Basin, northern Gulf of Mexico, is used to document the first detailed melting history of the southern margin of the Laurentide Ice Sheet (LIS) during MIS 3, and to record terrestrial inputs from the Mississippi River related to changes in eva porationprecipitation over the mid-continent, from 28-45 ka. Paired measurements of d 18 O and Mg/Ca-SST on the planktonic foraminifera Globigerinoides ruber (pink) are used to calculate the d 18 O of seawater ( d 18 O sw ) and test one of the key hypotheses for abrupt climate change. Five intervals of freshwater input from 28-45 ka do not match the abrupt Dansgaard-Oeschger temperature oscillations recorded in Greenland ice. Rather, summer melting of the LIS may have occurred during Antarctic warming and likely contributed to sea-level variability during MIS 3. A detailed assessment over one of the meltwater events, using the d 18 O and d 13 C of G. ruber and the deeper dwelling Neogloboquadrina dutertrei demonstrate that meltwater was confined to the surface layers and likely had an impact on the biological pump in the Gulf of Mexico. A similar d 18 O sw record determined from the year-round white G. ruber suggests that melting was not limited to the warmest summer months. The timing of LIS meltwater input is decoupled from an interval of enhanced wet conditions over the North American continent and increased Mississippi River discharge, as shown by a suite of
vii organic and sedimentologic proxies. Increasing summer insolation on the orbital scal e may have led to a northward migration of the Intertropical Convergence Zone and an intensification and westward shift in the conical position of the Bermuda High, which shuttles moisture to the North American continent and contributes to flooding in the Mississippi River drainage basin.
1 Chapter 1 Introduction 1.1. Introduction Numerous climate records now show that abrupt regional climate transitions occurred repeatedly in the past 100,000 years (e.g. Dansgaard et al., 1993), but only in the last decade has scientific research reflected the importance of this conc ept (Overpeck and Webb, 2000; Alley et al., 2003). Abrupt climate change covers time scales ranging from decades to millennia and occurs when the climate system is forced to cross s ome threshold, triggering a transition to a new state at a rate determined by the cli mate system itself and faster than the cause (National Research Council, 2002). Despite the ra pidity of the response, the forcing mechanism may be fast or slow. The largest climate c hanges have occurred during glacial time periods, where in Greenland, temperatures changed by up to 16rC and precipitation by a factor of 2 (Alley and Clark, 1999; Lang et al., 1999). These changes were concurrent with variations in the expansion of tropical wetlands (Brook et al., 1999) and the intensity of the Asian monsoon (Wang et al., 2001), suggesting a hemispheric to global response. Abrupt climate changes have also been recognized during the Holocene, where some of the greatest variability has been linke d to the hydrologic cycle (Overpeck and Webb, 2000 and references therein). Paleoclimate archives, for example, document decadeto centurylong droughts that were more severe and persistent than observed within the time frame of instrumental records. Greenhouse warming and other human alterations of the climate system may increase the probability of large, abrupt climate change (National Research C ouncil, 2002). Instrumental records show that the ocean heat content has been slowly increasing over the past 40 years (Levitus et al., 2000). The greatest change has been in the North
2 Atlantic where warming has been measurable to depths of 3000 m. Additionally, this region of the ocean has been freshening continuously for the past 40 years, with the most dramatic changes occurring in the past decade (Dickson et al., 2002; Curry et al., 2003). Freshening has been attributed to increased river discharge to the Arctic Ocean ( Peterson et al., 2002), which may be compounded by melting glaciers or Arctic sea ice, as well as increased precipitation (Hu and Meehl, 2005). Numerical climate models demonstrate that North Atlantic Deep Water (N ADW) has a threshold, called the bifurcation point, at which point the addition of more freshwater would lead to a collapse of the thermohaline circulation (Rahmstorf, 1995; 2002). Deep-water formation from the Norwegian and Greenland Seas has already diminished by 20 percent since 1950, implying a weakened global thermohaline circulation and reduced inflow of Atlantic water to the Nordic seas (Hansen et al., 2001) This is supported by recent evidence from the subtropical Atlantic, which indicates a slowing of thermohaline circulation in the past several decades (Bryden et al., 2005). A disruption of the global ocean conveyor could lead to a surface air temperature change of up to 6rC in the North Atlantic region (Rahmstorf, 1995; Vellinga and Wood, 2002), as well as widespread drought in other parts of the world (Peteet et al., 1995; Alley et al ., 1997). Clearly, more research is necessary to better understand abrupt climate cha nges in the past in order to make sound predictions for the future. 1.2. Climate-change paradigms It is now well understood that the large glaciations of the past 0.9 Myr have been dominated by a 100 thousand year (kyr) cycle (Martinson et al., 1987). This 100-kyr cycle, as first identified through the analyses of benthic oxygen isotopes in the mar ine sedimentary record (ie. Shackleton, 1987), is characterized by a saw-tooth pattern wi th a slow oscillatory buildup of ice over most of the cycle, followed by a rapid deglaciation that occurs in ~10 kyr. The benthic isotope records, which primarily reflect ice volume with a small temperature overprint, suggested that the growth and demise of the lar ge Northern Hemisphere ice sheets followed the 100-kyr cycle. The timing of the Last Glacial Maximum (LGM) in the sedimentary records was consistent with the t iming of
3 LGM ice sheet moraine deposition found on the continent. Further, the 100-kyr cycle was found in records from Antarctic ice cores of temperature and CO 2 spanning the past 400,000 years (Johnsen et al., 1972). The idea that Earths climate tends to change gradually in response to slow changes in climate forcing (ie. Milankovitch Theory) became a strong focus of paleoclimate research, where particular emphasis wa s placed on identifying the cause of the 100 kyr glacial cycle. This changed, however, in the early 1990s with the extraction of ice cores from central Greenland that allowed for the identification of numerous abrupt climate cha nges throughout the last glacial period (Dansgaard et al., 1993). The most pronounced changes are the millennial-scale Dansgaard-Oeschger (D/O) cycles, w hich occurred during Marine Isotope Stage 3 (MIS 3; 24-57 ka). D/O events have an asymmetric saw-tooth pattern that begins with an abrupt warming of 5 to 10rC within decades and is followed by a slow cooling that lasts several centuries and culminates in a rapid re turn to stadial conditions (Figure 1). Soon after, the abrupt climate oscillations first r ecorded in the Greenland records were observed in sediments from the North Atlantic region (Bond et al., 1993). In these records, several of the stadials are closely coupled with Heinri ch events (H-events) (Bond and Lotti, 1995), periods of massive episodic iceberg delivery from the Laurentide Ice Sheet, through the Hudson Strait (Bond et al., 1992). A warming to near interglacial conditions follows each H-event and successive D/O cycles get progressively cooler until the next H-event (a Bond cycle), 7-10 thousand years later (Bond et al., 1993). The signature of H-events is recorded as layers of ice-rafted debri s (IRD) in the marine sedimentary record (Heinrich, 1988; Bond et al., 1992). The discovery of the D/O events in records from the North Atlantic ocean region (Bond et al., 1993) provided the first recognition that the D/O cycles may have been a hemispheric signal. Further, the relationship between the D/O cycles and Heinric h events suggested a possible linkage between climate change in the North Atlantic regi on and the timing of Laurentide Ice Sheet fluctuations. Since the discovery of millennial-s cale climate variability in records from the North Atlantic region, many climate -change initiatives have focused on identifying D/O cycles in records across the globe (se e Voelker et al., 2002). Through time, it has become generally accepted that the Greenland
4 ice core record is a template for Northern Hemisphere climate change. Theref ore, it has also been assumed that the Northern Hemisphere ice sheets should show variability consistent with the Greenland record. One of the key findings of this dissertation is tha t fluctuations of the Laurentide Ice Sheet did not follow temperature oscillations over Greenland during the last glacial period. This challenges the assumption that Gree nland air temperature records reflect Northern Hemisphere climate change. This dissertation focuses on understanding climate variability associated wit h the D/O cycles of MIS 3. MIS 3 was a time of intermediate ice volume (Figure 2) (Dy ke et al., 2002), where sea level stood on average 80-85 m below present and fluctuated by <30m (Figure 1) (Chappell, 2002; Siddall et al., 2003). Numerous paleoclimate archives, including records of high-latitude temperature, sea level, ocean circulation and hydrologic variability, document millennial-scale climate change during MIS 3 ( Figure 1) (Voelker et al., 2002), yet the cause of these rapid climate oscillations remains a matter of debate. Mechanisms such as changes in ocean circulation and alterations in the tropica l ocean-atmosphere system have been proposed to explain MIS 3 climate variability, but many more records are needed in order to fully assess the cause of abrupt climate cha nge during this time. Two primary questions will be addressed here in an attempt to better understand abrupt climate change during MIS 3: 1. What was the role of the North American Laurentide Ice Sheet in abrupt climate change? 2. How did the North American hydrologic cycle respond to rapid climate oscillations during MIS 3? 1.3. Cause of millennial-scale climate change Several ideas have been proposed to explain the millennial-scale D/O cycles. One theory suggests that D/O cycles and H-events simply reflect an amplifica tion of a ~1,500year climate cycle that oscillates independent of the glacial/intergla cial state (Bond et al., 1997; 1999; Alley et al., 2001). The millennial-scale cycle may arise from external forcing such as solar variability (Bond et al., 2001) or internal oscillations within the deep ocean (ie. salt oscillator, Broecker et al., 1990). The theory calls upon the impact of
5 continental ice sheets on North Atlantic thermohaline circulation to explain the magnitude of abrupt temperature changes observed during the glacial period (Bond et al., 1997; Alley et al., 2001). Support for this idea is found in numerical models, which demonstrate that an increased flux of freshwater into the North Atlantic from melting ice would be ca pable of reducing or shutting down North Atlantic Deep Water (NADW) formation (Rahmstorf, 1994; Manabe and Stouffer, 1995). Although the details surrounding catastrophic ice sheet collapses are largely unresolved (MacAyeal, 1993; Andrews, 1998), the impact of melting ice on global climate would be the same. A slow down of thermohaline circulation would result in a reduction in cross-equatorial heat transport, a cooling in t he North Atlantic, and a warming in the high-latitude and tropical Atlantic (Crowley 1992; Manabe and Stouffer, 1997). This bipolar seesaw (Broecker, 1998) has been inferred from Antarctic and Greenland ice core records, where warmings in Antarctica pr ecede those in Greenland by several thousand years (Blunier and Brook, 2001). Additionally, the climate signature in Antarctica shows gradual temperature changes, whil e Greenland temperature is characterized by the higher frequency D/O events (Figure 1). Se diment archives from the western tropical Atlantic also support the modulation of climate by thermohaline circulation, as evidenced by warming that occurs during H-event 1 and the Younger Dryas cold interval (Ruhlemann et al., 1999; Huls and Zahn, 2000; Flower et al., 2004). In contrast, tropical and subtropical Atlantic records indicate SST changes that were synchronous with Greenland air temperature (Guilderson et al., 1994; 2001, Zhao et al., 1995; Lea et al., 2003; Sachs and Lehman, 1999). These records provide support for the theory that millennial-scale climate change may be initiated from the t ropics, most likely in the Pacific, which encompasses the warmest regions of the ocean and is the principal source of water vapor to the atmosphere (Cane, 1998; Cane and Clement, 1999). By analogy to modern day El Nino Southern Oscillation (ENSO) patterns, it is sugges ted that variations in the location of tropical convection influence global climate by alt ering heat and water vapor transport through atmospheric teleconnections (Cane and Clement, 1999). The rapid mixing of the atmosphere would result in a nearly synchronous
6 response of surface temperatures relative to the northern high latitudes, as simi larly observed in marine records from the Pacific (Kienast et al., 2001; 2003; Koutavas et al.,2002; Stott et al., 2002) and Indian Oceans (Bard et al., 1997). A clear discrepancy exists between the relative phasing of records from the tropical Atlantic reg ion and the northern high latitudes (e.g. Ruhlemann et al., 1999; Zhao et al., 1995), adding to the difficulties in understanding the cause of the millennial-scale D/O cycles. One of the primary goals of this dissertation is to test a key hypothesis that has been invoked to explain the millennial-scale D/O cycles, namely that changes in the routing of freshwater between the Mississippi River and eastern outlets (St. La wrence and Hudson Rivers), caused by fluctuations in the southern margin of the LIS, led to changes in the strength of NADW and regional temperature fluctuations (Clark et a l., 2001). This theory suggests that a routing of freshwater to the North Atlantic via eas tern outlets would lead to a reduction in the strength of the NADW and a regional cooling. In contrast, a southward routing of freshwater (meltwater and precipitation) to the G ulf of Mexico would allow NADW formation, thereby bringing heat to the high northern latitudes and resulting in warmer regional temperatures. Research presente d as a component of this dissertation tests the routing hypothesis of abrupt climate change during MIS 3 based on Gulf of Mexico sediments. Paired measurements of d 18 O and Mg/Ca analyses on the planktonic foraminifer Globigerinoides ruber (pink) from Orca Basin core MD02-2551 (Figure 2) are used here to calculate the d 18 O of seawater and reconstruct the timing of meltwater input to the Gulf of Mexico during D/O cycles 4-12 (Figure 1). This interval covers 2 Bond cycles, and includes one of the largest interstadials (IS 8) in the Greenland air temperature record. Additional paired measurements of d 18 O and Mg/Ca on a different planktonic foraminifer species, Globigerinoides ruber (white), over the largest LIS meltwater interval, are used to replicate the meltwater signal in the same Orca Basi n core. These measurements also provide information on the seasonality of LIS meltwater input because it is likely that the pink and white G. ruber have different seasonal preferences, based on their modern seasonal abundances in the nearby subtropical Sargasso Sea (Deuser, 1987; Deuser and Ross, 1989). The d 18 O of G. ruber (pink and white) and of
7 Neogloboquadrina dutertrei a thermocline-dwelling species, are used to address the depth of the meltwater lens. The d 13 C of these three species provide information on the productivity response to the interval of sustained freshwater input. 1.4. Climate controls on hydrologic variability Components of the hydrologic cycle have also responded to the millennial-scale variability observed in the Greenland ice core record. Major element chemistry f rom Cariaco Basin sediments indicate large-scale changes in precipitation and r iverine discharge with a similar timing to the D/O cycles and may reflect shifts in the Intertropical Convergence Zone (ITCZ) (Peterson et al., 2000). The East Asian monsoon has also been shown to fluctuate with D/O timing, where warmer Greenland temperat ures correlate with a more intense summer monsoon (Wang et al., 2001). These oscillations in the Asian monsoon are superimposed on a long-term trend that appears to follow orbital-scale northern hemisphere summer insolation. Variations in insolation on orbital time scales influence atmospheric circulation patterns and the hydrologic cycle throug h changes in the zonal and meridional gradients of atmospheric heating (Clement et al ., 2004). The results of these changes have been observed in other paleoclimate records from the tropics and subtropics, where precessional changes exert a strong control on monsoonal circulation and the position of the ITCZ, altering regional precipitation patterns (Kutzbach, 1981; Wang et al., 2001; Baker et al., 2001; Wang et al., 2004). The influence of insolation on the ITCZ has been noted in the circum-Caribbean (Hodell et al., 1991; Haug et al., 2001; Poore et al., 2003; Hillesheim et al., 2005). In particular, increased precipitation in the region (Hodell et al., 1991; Hillesheim e t al., 2005) and greater transport of Caribbean waters into the Gulf of Mexico (Poore et al., 2003) were linked to a northward shift in the ITCZ due to high solar insolation during the early Holocene. Hodell et al. (1991) and Hillesheim et al. (2005) speculated that the precipitation patterns may also have been associated with variations in the intensi ty of the North Atlantic Bermuda High atmospheric system, which is tied to the ITCZ on the annual cycle (Machel et al., 1998). The annual cycle of precipitation in the Caribbean/Gulf of Mexico region is strongly controlled by the seasonal migration of the
8 ITCZ and the Bermuda High (Hastenrath et al., 1984). During the high insolation of northern hemisphere summer, the ITCZ moves north and the Bermuda High is displaced to the west (Machel et al., 1998), shuttling moisture to the region through anticyclonic atmospheric circulation. Meteorological studies of interannual rainfall varia bility demonstrated that years of anomalously high precipitation were a result of an enhancement of this annual cycle (Hastenrath et al., 1984). The Bermuda High plays a fundamental role in advecting moisture into the North American continent in modern times, but its potential importance in precipitation patterns during times of inter mediate ice volume is yet to be determined. A second major goal of this dissertation is to document changes in hydrology in the central North American continent during MIS 3 to determine whether the continental hydrologic cycle responded to the abrupt D/O cycles, to changes in northern hemisphere summer insolation, or both. It is well recognized that the Northern Hemisphere ice she ets had a significant impact on the North American hydrologic cycle from the Last Gla cial Maximum (LGM) to the early Holocene, directly through ice sheet growth/decay ( Clark et al., 1993, Marshall and Clarke, 1999; Licciardi et al., 1999) and indirectly by altering regional precipitation patterns that led to shifts in the continental moisture bala nce (Kutzbach and Guetter, 1986; Kutzbach, 1987; Webb et al., 1993). Less is known about North American hydrologic conditions prior to the LGM, however, because few terrestrial records exist and uncertainties in the size and extent of the ice sheets have made it difficult to create model simulations. A handful of records from the west coa st of the United States suggest clear D/O cyclicity, although overall the records are ambiguous (Voelker et al., 2002). In contrast, lake records from Mexico show deep phases from 30-38 ka (Caballero et al., 1999; Bradbury et al., 2000), when summer insolation at 30rN reached a maximum (Figure 3). Bradbury et al. (2000) hypothesized that the deepening of the lakes may have resulted from a more effective Gulf of Mexico moisture sourc e. If a strengthening and westward expansion of the Bermuda High led to the deepening of these lakes, we may expect to see wet conditions on the North American continent during this time as well.
9 A suite of organic and inorganic geochemical proxies are used here to document inputs from the Mississippi River and to relate these inputs to changing moisture bala nce over the central North American continent during MIS 3. The timing of Mississippi River discharge is compared to LIS meltwater input and solar insolation to assess any linkages. Further, the records are discussed in the context of the ITCZ and Bermuda High. The response of oceanic productivity to enhanced riverine discharge is examined as well. 1.5. Orca Basin The Orca Basin, in the northern Gulf of Mexico (Figure 2), is ideally located to study changes in North American continental hydrology during MIS 3 because of its proximal location to the mouth of the Mississippi River. The Mississippi River draina ge system encompasses ~40% of the contiguous United States and therefore sediments fr om the Orca Basin should provide an integrated assessment of changes in mid-continental hydrology. In addition, the Mississippi River served as one of the main conduits for meltwater draining from the Laurentide Ice Sheet (LIS), which covered much of the northern North American continent during the last glacial period, and LIS inputs to the Gulf of Mexico should be preserved in Orca Basin sediments. A high-sedimentation rate (~50 cm/kyr) core from the Orca Basin was collected aboard the R/V Marion Dufresne in July 2002. A multi-proxy approach is used here to assess the timing and effect of LIS meltwater input to the Gulf of Mexico and to document changes in North American hydrology from 28-45 ka. The results are discussed in the context of climate change on millennial and orbital timescales during MIS 3. 1.6. Dissertation organization The dissertation is broken down into three chapters, which have been written as manuscripts for peer-reviewed journals. The text and figures in the dissertation a re almost identical to the papers. Therefore, some figures may be repetitive. The fi gures for each chapter are located at the end of the chapter.
10 In Chapter 2: Laurentide Ice Sheet meltwater and abrupt climate change during the l ast glaciation Paired measurements of oxygen isotopes and Mg/Ca of foraminiferal calcite are used to document intervals of LIS meltwater input to the Gulf of Mexico from 28-45 ka. The record of LIS melting is compared to the Greenland ice core record to test the routing hypothesis of abrupt climate change. Further, a comparison to the Antarctic i ce core temperature record and global sea level is made. Chapter 2 has been published in Paleoceanography : Hill, H.W., Flower, B.P., Quinn, T.M., Hollander, D.J., and Guilderson, T.M., 2006, Laurentide Ice Sheet meltwater and abrupt climate change during the last glac iation, Paleoceanograph y 21, PA 1006, doi: 10.1029/2005PA001186. In Chapter 3: A multi-species approach to constraining Laurentide Ice Sheet meltw ater input to the Gulf of Mexico during the last glacial period Isotopic and elemental ratios of three foraminifera with different seasonal preferences and depth habitats are used to constrain the seasonality and thickness of LI S meltwater input to the Gulf of Mexico over the largest meltwater interval. The productivity response to this interval of sustained freshwater input is also discusse d. Chapter 3 will be submitted to Earth and Planetary Science Letters, or a simila r journal. In Chapter 4: Mississippi River flooding during the last glaciation A suite of organic, inorganic, and sedimentological proxies are used to identify intervals of enhanced Mississippi River discharge and inferred wet conditions over the North American continent from 28-45 ka. The record is compared to northern hemisphere summer insolation to draw linkages between the ITCZ and the Bermuda High atmospheric systems. Chapter 4 is in review in Geology : Hill, H.W., Hollander, D.J., Flower, B.P., and Quinn, T.M., in review, Mississippi River flooding during the last glaciation, Geology.
11 -44 -43 -42 -41 -40 -39 -38 -37 -36GISP2 d18Oice ( VSMOW) -42 -41 -40 -39 -38Byrd d18Oice ( VSMOW) -100 -95 -90 -85 -80 -75 -70 -65 -60 283032343638404244Sea level (m)Calendar age (k.a.) A1 4 567 891011 12 H3 H4 Figure 1. Marine Isotope Stage 3 records. a. GISP2 d 18 O ice (Grootes et al., 1993). b. Byrd d 18 O ice (Johnsen et al., 1972) on the GISP2 timescale, based on synchronization of methane concentrations within the two ice cores (Blunier and Brook, 2001). Numbers refer to Greenland interstadials. Dark gray bars and letter H indicate Heinr ich events. A1 refers to Antarctic warming event number 1 (Blunier and Brook, 2001). c. Global sealevel record (Siddall et al., 2003).
12 Figure 2. Map of Orca Basin in the Gulf of Mexico showing location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) and the extent of the Laurentide Ice Sheet during MIS 3 (from Dyke et al ., 2002).
13 470 480 490 500 510 520Insolation (W/m2) at 30oNJune 21st 205 210 215 220 225 230 235 01020304050Insolation (W/m2) at 30oNDecember 21st Age (ka) Figure 3. Insolation at 30rN from 0-50 ka. a. June 21 st insolation and b. December 21 st insolation. Gray bar indicates time of maximum insolation from 31-38 ka referred to in text.
14 Chapter 2 Laurentide Ice Sheet meltwater and abrupt climate change during the last glac iation 2.1. Abstract A leading hypothesis to explain abrupt climate change during the last glacial cyc le calls on fluctuations in the margin of the North American Laurentide Ice Sheet (LI S), which may have routed freshwater between the Gulf of Mexico (GOM) and North Atlantic, affecting North Atlantic Deep Water (NADW) variability and re gional climate. Paired measurements of d 18 O and Mg/Ca of foraminiferal calcite from GOM sediments reveal five episodes of LIS meltwater input from 28-45 thousand years ago (ka) that do not match the millennial-scale Dansgaard-Oeschger (D/O) warmings rec orded in Greenland ice. We suggest that summer melting of the LIS may occur during Antarc tic warming and likely contributed to sea-level variability during Marine Isotope Sta ge 3 (MIS 3). 2.2. Introduction Abrupt climate changes during the last glaciation have been linked to variations in Atlantic thermohaline circulation. Numerical models demonstrate that an incre ased flux of freshwater to sites of deep-water formation decreases the strength of Nort h Atlantic Deep Water (NADW), thereby reducing meridional heat transport and causing cooling/warming in the northern/southern high latitudes (Ganopolski and Rahmstorf, 2001; Knutti et al., 2004). This bipolar seesaw (Broecker, 1998) has been invoked to explain the anti-phased relationship between climate changes in Antarctica and Greenland, where warmings in Antarctica precede those in Greenland by several thousand years (Blunier and Brook, 2001). Additionally, the climate signature in
15 Antarctica shows gradual temperature changes, while Greenland temperature i s characterized by higher frequency changes, including abrupt warmings that occur in decades, followed by slow coolings (Dansgaard-Oeschger (D/O) cycles). The North American Laurentide Ice Sheet (LIS) may have served as a source of freshwater to the North Atlantic during the last deglaciation, when ice-sheet r etreat led to the diversion of freshwater (meltwater and precipitation) from the Mississi ppi River drainage to the Hudson and St. Lawrence Rivers (Broecker et al., 1988; 1989; Shackleton, 1989; Rooth, 1990; Flower and Kennett, 1990; Clark et al., 2001; Flower et al., 2004). Meltwater routing has been suggested as a potential control of high-frequency climate variability during intervals of intermediate ice volume, such as during M arine Isotope Stage 3 (MIS 3) (Clark et al., 2001). However, evidence is needed to assess potential switches in freshwater routing during the millennial-scale D/O cy cles, which are characterized by 5-10 o C oscillations in Greenland air temperature (Dansgaard et al., 1993). Here we test whether D/O warmings correspond to freshwater routing to the Gulf of Mexico (GOM) by reconstructing the d 18 O composition of seawater ( d 18 O sw ) using paired measurements of d 18 O calcite ( d 18 O c ) and Mg/Ca of GOM foraminifera. Orca Basin (26 o 56.77N, 91 o 20.74W; Figure 4) in the northern GOM is ideally located to study freshwater input, including LIS meltwater, from the North American continent beca use of its proximal location to the mouth of the Mississippi River. 2.3. d 18 O and Mg/Ca analyses Core MD02-2551 was recovered from Orca Basin in July 2002 by the R/V Marion Dufresne as part of the IMAGES (International Marine Past Global Changes Study) program. The core was sampled at 2 cm intervals from 21-30 m. All samples were freeze-dried prior to wet sieving, and then washed over a 63m m mesh using deionized water. ~60-70 planktonic foraminifera G. ruber (pink variety) were picked from the 250-355 m m size fraction for isotopic and elemental analyses. The foraminifera were sonicated in methanol for five seconds to remove clays, and then weighed to assess downcore dissolution effects. Mean G. ruber weights are similar throughout the interval and are comparable to surface-sediment samples (LoDico, 2002). The shells were gently
16 crushed open between two glass plates and carefully homogenized using a razor blade. A ~50 m g aliquot was removed for stable isotopic analysis, which was performed at the College of Marine Science, University of South Florida using a ThermoFinnigan Delt a Plus XL dual-inlet mass spectrometer with an attached Kiel III carbonate pre paration device. The isotopic data (Figure 5) are reported on the VPDB scale calibrated wit h NBS-19. Standard deviation for the d 18 O c measurements is 0.04, based on measurements of the standard NBS-19 analyzed with MD02-2551 foraminifer samples (n=105). The remaining tests, weighing ~700 m g, were split into two aliquots that were cleaned separately for Mg/Ca analysis (Barker et al., 2003). This method involves an initial sonication to remove fine clays, oxidation of organic matter with a buffere d peroxide solution, and a dilute acid leach that eliminates any adsorbed contaminants. Samples were dissolved in weak HNO 3 to yield calcium concentrations of ~20 ppm to minimize calcium concentration effects. The Mg/Ca ratios (Figure 5) were analyzed on a Perkin Elmer Optima 4300 dual view inductively coupled plasma-optical emission spectrometer (ICP-OES). A standard instrument-drift correction technique wa s routinely used. The analytical precision for Mg/Ca determinations used in this study is <0.6% root-mean standard deviation (1 s ), based on an ICP-MS calibrated standard solution. The pooled standard deviation of 70% replicate Mg/Ca analyses is 2.5% (d.f. = 318), which is equivalent to ~0.3 o C. 2.4. Age model The age model developed for our record (Figure 5) is based on 18 AMS 14 C dates (Table 1) determined from monospecific samples (4-10 mg) of pink G. ruber which were run at the Center for Accelerator Mass Spectrometry, Lawrence Livermore N ational Laboratory. The 14 C ages were corrected for a reservoir age of 400 years and converted to the GISP2 timescale (an approximation of calendar years) using a high-resolut ion radiocarbon calibration developed on sediment cores from the Cariaco Basin (Hughen et al., 2004). Inferred minimal changes in upwelling indicate uncertainty in the reservoir correction is much better than 100 years. Age was also constrained by the Laschamp
17 geomagnetic event (Laj et al., 2000), which is recorded as a ~50 cm minimum in inclination at a depth of ~27.5 m (Kissel et al., m.s. in prep). A peak in 10 Be, which coincides with the Laschamp event in sediment cores from the North Atlantic (Robinson et al., 1995), straddles the d 18 O peak of Interstadial 10 in the Greenland ice core record (Yiou et al., 1997). The Laschamp event was therefore assigned a calendar age of 40.9 k.a. based on the age of the d 18 O peak of Interstadial 10 on the Greenland GISP2 time scale (Meese et al., 1997) (Figure 5). Depth in centimeters was converted to age by applying a weighted curve fit with a 40% smoothing factor and linearly extrapolating beyond the Laschamp event. This function fits a curve to the calibrated 14 C age control points, using the locally weighted Least Squares error method. Because of the uncertainty associated with radiocar bon dates of increasing age, including 14 C age plateaus at ~24 and ~28 14 C k.a. (Hughen et al., 2004), the weighted smooth fit provides a conservative estimate of depth vs. age. Sedimentation rates range from 25 cm/k.y. to 325 cm/k.y. Total error (1 s ) on the age model ranges from 140 calendar years at ~26 ka to a maximum of 700 calendar years at ~40 ka. Error was determined by compounding the error on the 14 C measurements from this study (Table 1), the error on the 14 C measurements from the Cariaco record and the error from the GISP2/Cariaco cal ibration reported by Hughen et al. (2004). Errors in 14 C were converted to calendar years using the Cariaco calibration. Calculating the error prior to 40 ka is difficult because of the uncertainty in the Cariaco calibration. Errors on the layer counting from the GISP2 record were not included in the total error analysis because we do not make conclusions about the absolute age of our events. Rather, we place our records on the GISP2 timescale to compare our results to Greenland air temperature history. We have also placed our data on the newly proposed age scale for the Greenland ice cores (SFCP 2004), which is based on 14 C dating of foraminifera in core MD95-2042, calibrated by paired 14 C and 230 Th measurements on corals (Shackleton et al., 2004) (see Supplementary Information). The conclusions that we report in the paper are the same regardless of which timescale we use for the Greenland ice core record.
18 2.5. Gulf Of Mexico d 18 O of seawater The G. ruber d 18 O c values range from ~ 0.5 to 2.5 (Figure 6). This 2 variability is not seen in the d 18 O c of N. dutertrei (data not shown), an inferred deepdwelling planktonic foraminifer, suggesting that surface water phenomena are cont rolling the d 18 O c The d 18 O c record exhibits four oscillations about a mean value of 1.25, from 28-45 k.a. (Figure 6). d 18 O c values are more negative than the modern core-top value of pink G. ruber (1.7) during two of these oscillations (28.7-29.2 k.a. and 37.339.8 k.a., Figure 6). Given that sea level was 63-93 m below present from 30-45 k.a. (Siddall et al., 2003), which would result in an enrichment of the foraminifera d 18 O c by ~0.5-0.75 based on the relationship 0.083 per 10m sea-level change (Adkins et al., 2001), d 18 O c values 1.7 would indicate SSTs of 30-32 o C during MIS 3, which are unreasonably high compared to the modern average summer temperature in the GOM (29 o C; June-Sep) (Levitus, 2003). A change in d 18 O sw associated with salinity variations is therefore required to explain the four negative oscillations recorded in the foram iniferal calcite. In order to isolate d 18 O sw we subtract the temperature component from the d 18 O c based on Mg/Ca data (Flower et al., 2004). The Mg/Ca ratio, a proxy for the temperature of foraminiferal calcification, is ideal for d 18 O sw calculations because it is measured on an aliquot of the calcite sample used for d 18 O c A G. ruber (pink) calibration, based on Atlantic sediment trap data (Anand et al., 2003), was applied to the Mg/Ca measurement s to calculate SST (Figure 6). We make the assumption that the effect of riverine i nput on the Mg/Ca values is minimal based on the large difference in Mg and Ca concentrati ons in the Mississippi River and the GOM (425 m M Mg vs. 53 mM Mg; 870 m M Ca vs. 10.3 mM Ca; Briggs and Ficke, 1978). Despite the lower Mg/Ca ratio of Mississippi Rive r water, oceanic Mg/Ca is not likely to be affected because the concentrations of Mg /Ca are low. A simple box model calculation shows that a 25% dilution of surface seawater (a likely maximum for G. ruber to withstand; Hemleben et al., 1989) would only decrease Mg/Ca values by <3%, which is within measurement error. The Mg-SST component was removed from the d 18 O c using a temperatured 18 O relationship (Bemis et al., 1998) appropriate for G. ruber (Thunell et al., 1999), resulting
19 in the d 18 O sw The standard deviation for d 18 O sw calculations is determined to be 0.25, based on propagating the error through the analytical errors and the combined Mg-SST and SSTd 18 O relationships (Beers, 1957). The variances used for the Mg-SST and SSTd 18 O equations are those reported in the literature. Variances for Mg/Ca and d 18 O were based on replicate analyses. The d 18 O sw variations from core MD02-2551 have similarities to the global sealevel record from MIS 3 (Siddall et al., 2003) (Figure 6). However sea-level fluctuations of <30 m during this interval (Siddall et al., 2003) can explain only 0.25 of the >1 d 18 O sw changes observed in our record, suggesting that changes in evaporation/precipitation (E-P) or freshwater input must be the dominant control on the d 18 O sw We use the sea-level record (Siddall et al., 2003) to remove the contribution of global ice volume to the d 18 O sw leaving the GOM d 18 O sw residual ( d 18 O GOM ) (Figure 7). This was accomplished by converting sea-level height to the d 18 O equivalent using the relationship 0.0083 per 1m sea-level change (Adkins et al., 2001). d 18 O GOM values reflect changes in salinity, which result from a combination of source-water variability and/or changes in the volume of water affecting the d 18 O GOM signal. The d 18 O GOM oscillates by up to 1.5, between more fresh versus more saline conditions, about a mean value of 0.45 (Figure 7). Major freshwater events, defined as intervals when the d 18 O GOM reach values <0.45 and persist for >1.5 k.a., occurred from 31.7-34 k.a. and 37.2-39.8 k.a (F2 and F4; Figure 7). The signatures of these two freshwater events are different, however: F2 is defined by a gradual change from mor e saline to more fresh conditions, while F4 is characterized by an abrupt freshening and an abrupt return to saline conditions. Three minor freshwater events, from 28.3-29.4, 35.0-35.5 and 42.9-43.8, also record values < 0.45, but persist for <1.5 k.a. (F1, F3 and F5; Figure 7). 2.6. Conversion to sea-surface salinity Conversion of d 18 O GOM estimates to sea-surface salinity (SSS) allows us to assess potential sources and magnitudes of freshwater flux to the GOM. SSS can be estimat ed
20 using a d 18 O GOM versus salinity relationship created for the GOM during MIS 3 (Figure 8). This relationship assumes conservative mixing between two end-members: high salinity GOM waters ( d 18 O sw = 1.2 and S = 36.5 psu) and a low salinity end-member. The low salinity end-member is modeled using three different compositions: 1) 1) a 3.5 value for GOM precipitation (Bowen and Revenaugh, 2003), and a Laurentide Ice Sheet (LIS) value ranging from 2) 15, reflecting the d 18 O of source waters that drained from the LIS (Yapp and Epstein, 1977), to 3) 30, the average composition of the LIS (Dansgaard et al., 1969). It should be noted that the more negative the zero salinity intercept, the smaller the changes in the estimated salinity vari ations (Figure 8). For example, a 1 change in d 18 O GOM is equivalent to ~1 psu on the -30 LIS mixing line, ~2 psu on the -15 LIS mixing line and ~8 psu on the 3.5 MR mixing line. Use of the -3.5 end-member would require changes in salinity of up to 10 psu (Figure 9) and a volume of water 3-5 times the largest historical flood (Barry, 1997), or >50X the annual precipitation in the GOM (Ropelewski et al., 1996), lasting for 3 k.y. during the largest event. It is possible that the isotopic composition of continental precipitation draining into the Mississippi River was more negative during MIS 3, due t o changes in the altitude and/or sources of precipitation. However, a minimal change in the d 18 O composition of precipitation during MIS 3 is inferred from model simulations, which show similar d 18 O precipitation values between the Last Glacial Maximum and present (Charles et al., 2001). In addition, mid-continent speleothems, which reflect the changing isotopic composition of meteoric waters, record <0.5 variations in d 18 O during this interval (Dorale et al., 1998). We cannot rule out the possibility that increas ed precipitation over the GOM may reflect an intensification of the North America n monsoon system, which is known to bring moisture to the region. However, the amount necessary to create the observed changes in the d 18 O GOM record does not support oceanic precipitation as a primary control on this signal. In contrast, meltwater derived f rom the LIS with a d 18 O composition of 15 to 30 would require only modest changes in salinity: a 15 end-member for the LIS results in a salinity change of up to 3.5 psu, while a 30 end member results in a change in salinity of up to 2 psu (Figure 9).
21 Additionally, the average SSS using a -30 endmember is 35.5 1 psu, which is within the modern salinity range in the GOM. We recognize that the source of fresh water likely changed through time and may have been a mixture of various sources (ie. meltwater and precip), and therefore the SS S calculations only reflect the endmember scenarios. Regardless, the most conser vative estimate for salinity changes indicates a substantial meltwater contri bution to d 18 O sw in the GOM, particularly when the d 18 O composition of GOM waters were most depleted This explanation is supported by recent reconstructions of the LIS during MIS 3, which place the margin of the ice sheet within the MR drainage basin (Dyke et al., 2002). 2.7. LIS routing hypothesis The uncertainty in the calibration of 14 C to calendar years precludes firm phase comparisons, but there appears to be no consistent relationship between d 18 O GOM freshwater input and Greenland interstadials. The LIS routing hypothesis would predict that the nine D/O warmings (IS 4-12) that span 28-45 k.a. (Grootes et al., 1993) should correspond to freshwater routing to the GOM (Clark et al., 2001), but only five d 18 O GOM freshwater events are recorded in the Orca Basin during this interval (Figure 7) There is no age model that we can construct with the 14 C dates that would allow the d 18 O GOM record from Orca Basin to be on the same timing as the D/O cycles in Greenland. In addition, the Laschamp event coincides with a warming in Greenland (IS 10), but a positive d 18 O excursion (more saline) in our record. If each of the D/O warmings corresponds to freshwater routing to the GOM, we would expect to see a negative d 18 O GOM excursion in our record during this interval. Although freshwater routed to eastern outlets may have led to NADW reductions and coolings in Greenland, the timing and number of d 18 O GOM freshwater events to the GOM suggest that a simple routing hypothesis cannot explain all of the MIS 3 Greenland interstadials. It appears that t he D/O warmings cannot be attributed to changes in the strength of NADW associated wi th southward routing of meltwater by the LIS, which may help explain why it has been difficult to find NADW changes during each of the D/O cycles (Curry et al., 1999; Hage n
22 and Keigwin, 2002; Vautravers et al., 2004). Additionally, SST in the GOM does not appear to be coupled to Greenland air temperature. The d 18 O GOM record has similarities to the Antarctic air temperature record (Johnsen et al., 1972), the global sea-level record from MIS 3 (Siddall et al., 2003), and to the classic MIS 3 benthic d 18 O record off Portugal (Shackleton et al., 2000). Freshwater events in the GOM have a tendency to coincide with intervals of Antarcti c warming. In particular, the largest freshwater event (F4) occurred at the sam e time as the largest warming in Antarctica (A1 centered at 39 ka; Figure 7) and a 30-m rise in s ea level also at 39 ka (Siddall et al., 2003). Our d 18 O GOM record suggests summer melting on the southern margin of the LIS during Antarctic warming, as also observed during the last deglaciation (Flower e t al., 2004). This provides evidence to support a recent modeling study that suggests that the northern hemisphere ice sheets contributed one-half of the global sea-level rises obs erved between 35-65 k.a. (Rohling et al., 2004). Our results are also consistent with a new coupled atmosphere-ocean simulation that predicts that freshwater discharge int o the Gulf of Mexico would contribute to Antarctic warming (Knutti et al., 2004). LIS melting associated with the A1 warming in Antarctica may have provided a positive feedback for Southern Hemisphere warming through changes in the strength of NADW. Similarly, our results indicate that growth/decay cycles of the LIS may have been decoupled from Greenland air temperature history during MIS 3. Our finding underscores recent work suggesting that the LIS (which is influenced by summer melting) does not follow Greenland air temperature (which is influenced by winter temperatures, partic ularly during stadials) and that seasonality is an important aspect of abrupt climate cha nge (Denton et al., 2005). 2.8. Supplementary information We have also placed our data on the newly proposed age scale for the Greenland ice cores (SFCP 2004), which is based on 14 C dating of foraminifera in core MD95-2042, calibrated by paired 14 C and 230 Th measurements on corals (Shackleton et al., 2004). This was done by first applying the SFCP timescale to the Cariaco record (Shackl eton,
23 per comm., 2004) and to the global sea-level record (Siddall et al., 2003). The sea-level record was originally correlated to the Byrd d 18 O record using a series of tie points. We used the same tie points to correlate the sea-level record to the Vostok d D record, which has been placed on the SFCP timescale. The relationship of the d 18 O GOM record to the Greenland and Antarctic air temperature records on the SFCP timescale (Fig ure 10) is consistent with the conclusions reported in the paper.
24 Table 1. Radiocarbon ages for MD02-2551. a Center for Accelerator Mass Spectrometry (CAMS), Lawrence Livermore National Laboratory b Samples not included in the age model due to stratigraphic inconsistencies. The 14 C ages at depths of 28.06, 28.46, and 29.88 m are younger than higher depths in the core. We choose not to use the 14 C age at 29.20 because it would require very large sedimentation rate changes from 30 cm/k.y. to 200 cm/k.y. Although this is possible, we instead choose to linearly extrapolate beyond the Laschamp event and are conservative with interpretations in our data prior to 41 k.a. CAMS a # Core depth (m) 14 C AMS age (k.a.) 14 C Error (+/-) Calibrated age (k.a.) 108325 19.68 23.11 160 26.40 108326 20.16 23.46 70 26.70 108327 20.62 24.22 80 27.30 90835 21.25 25.41 130 28.90 108328 22.02 25.48 90 29.00 100591 22.86 25.54 130 29.05 100592 23.20 24.21 120 27.25 100593 23.60 26.28 140 29.70 100594 24.06 26.79 150 30.00 100595 24.42 27.30 160 30.20 90836 24.75 31.17 250 34.95 100596 25.10 29.59 210 33.30 100597 25.48 31.84 270 35.50 100598 25.90 33.67 340 38.10 108329 26.48 34.20 600 39.20 90837 26.84 33.28 320 37.45 100599 27.22 35.66 420 40.55 Laschamp event 27.50 40.90 100600 27.58 36.23 460 40.75 100601 28.06 35.38 410 40.30 b 100665 28.46 34.80 500 40.00 b 108330 29.20 37.83 300 41.30 b 108331 29.88 33.17 180 37.30 b
25 Figure 4. Map of Orca Basin in the Gulf of Mexico showing location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) and the extent of the Laurentide Ice Sheet during MIS 3 (from Dyke et al., 2002).
26 -2.5 -2 -1.5 -1 -0.5 0G. ruber d18Oc ( VPDB) 3 3.5 4 4.5Mg/Ca (mmol/mol) 25 30 35 40 45 202224262830Age (k.a.)Depth (m) a b c Figure 5. Raw d 18 O c and Mg/Ca data and age model for MD02-2551. a. d 18 O c shown with 5-point smooth and b. Mg/Ca data shown with 5-point smooth on G. ruber from Orca Basin core MD02-2551 vs. depth in the core. c. Age model for our interval based on 18 14 C dates from G. ruber which were converted to the GISP2 timescale (an approximation of calendar years) using a Cariaco Basin radiocarbon calibration (Hughen et al., 2004). Age was constrained by the Laschamp geomagnetic event (Laj et al., 2000), which is recorded as a sharp peak in inclination at a depth of ~27.5 m, as indicated by light grey bar (Kissel et al., m.s. in prep).
27 -3 -2.5 -2 -1.5 -1 -0.5 0 283032343638404244d18Oc ( VPDB)Calendar age (k.a.) 22 23 24 25 26 27 28Mg-SST ( o C) -0.5 0 0.5 1 1.5 2 2.5d18Osw ( VSMOW)a b c d -100 -95 -90 -85 -80 -75 -70 -65 -60 283032343638404244Sea level (m)Calendar age (k.a.) 12 34 Figure 6. Paired d 18 O c and Mg/Ca data on G. ruber from Orca Basin core MD02-2551 during MIS 3. a. G. ruber d 18 O c shown with 5-point smooth. Mean value indicated by horizontal bar. b. G. ruber Mg/Ca converted to SST using Mg/Ca=0.38exp[0.09 X SST (C)] (Anand et al., 2003). c. Calculated d 18 O sw from d 18 O c and Mg-SST using T( o C) = 14.9-4.8*( d 18 O c d 18 O sw ) (Bemis et al., 1998). 0.27 was added to convert to VSMOW. d. Global sea-level record (Siddall et al., 2003). Numbers refer to d 18 O c oscillations referred to in text. Triangles on the bottom refer to intervals with 14 C dates.
28 -44 -43 -42 -41 -40 -39 -38 -37 -36GISP2 d 18 O ice ( VSMOW) -0.5 0 0.5 1 1.5d 18 O GOM ( VSMOW) -42 -41 -40 -39 -38 283032343638404244Byrd d 18 O ice ( VSMOW)Calendar age (k.a.)F1F2F3F4 F5 A1 4 567 891011 12 H3 H4 Fresher Saltiera b c Figure 7. Comparison of Orca Basin d 18 O GOM during MIS 3 with ice core records. a. GISP2 d 18 O ice (Grootes et al., 1993). b. Orca Basin d 18 O GOM with mean value indicated by horizontal bar. d 18 O GOM was calculated by subtracting global ice volume from the d 18 O sw record. c. Byrd d 18 O ice record (Johnsen et al., 1972) on the GISP2 timescale, based on synchronization of methane concentrations within the two ice cores (Blunier and Brook, 2001). Numbers refer to Greenland interstadials. Light grey bars and the letter F (numbered 1-5) indicate freshwater events referred to in the text. Dark grey bars and letter H indicate Heinrich events. A1 refers to Antarctic warming event number 1 (Blunier and Brook, 2001).
29 -35 -30 -25 -20 -15 -10 -5 0 5 0510152025303540d18OGOM ( VSMOW)Salinity (psu) d18OGOM = 0.13 SSS 3.5 d18OGOM = 0.44 SSS 15 d18OGOM = 0.85 SSS 30a bc a. b.c. Figure 8. Mixing model for the GOM during MIS 3. The d 18 O GOM versus salinity relationship assumes conservative mixing between two end-members: high salinit y GOM waters ( d 18 O sw = 1.2 and S = 36.5 psu) and a low salinity end-member. The low salinity end-member is modeled using three different compositions: a. 3.5 for GOM precipitation (Bowen and Revenaugh, 2003), b. 15, reflecting the d 18 O of source waters that drained from the LIS (Yapp and Epstein, 1977), and c. 30, the average composition of the LIS (Dansgaard and Tauber, 1969).
30 -0.5 0 0.5 1 1.5d 18 O GOM ( VSMOW) 22 24 26 28 30 32 34 36 38 283032343638404244 -30 end member -15 end member -3.5 end memberEstimate Salinity (psu)Calendar age (k.a.) a b Figure 9. GOM sea-surface salinity (SSS) reconstructions from 28-45 k.a. SSS is ba sed on the conversion of d 18 O GOM to salinity using a mixing model with three freshwater endmembers (see Figure 8). a. d 18 O GOM b. estimated salinity. The most conservative estimate for salinity changes indicates a substantial meltwater contri bution to d 18 O sw in the GOM.
31 -44 -43 -42 -41 -40 -39 -38 -37 -36GRIP d18Oice ( VSMOW) -0.5 0 0.5 1 1.5d18OGOM ( VSMOW) -15-10 -5 05 102832364044Vostok d D ice ( VSMOW)Calendar age (k.a.) 56 7 89101112 3 4 H4 H3 a b c A1 F1F2F3F4F5 Figure 10. Comparison of Orca Basin d 18 O GOM on the SFCP timescale during MIS 3 with ice core records. a. GRIP d 18 O ice (Johnsen et al., 2001) on SFCP timescale b. Orca Basin d 18 O GOM with mean value indicated by horizontal bar. d 18 O GOM was calculated by subtracting global ice volume from the d 18 O sw record. c. Vostok d D (Petit et al., 1999) on SFCP timescale which is normalized to remove the linear trend. Numbers refer t o Greenland interstadials. Light grey bars and the letter F (numbered 1-5) indicate freshwater events referred to in the text. Dark grey bars and letter H indicat e Heinrich events. A1 refers to Antarctic warming event number 1 (Blunier and Brook, 2001).
32 Chapter 3 A multi-species approach to constraining Laurentide Ice Sheet meltwater input t o the Gulf of Mexico during the last glacial period 3.1. Abstract Multiple species of planktonic foraminifera with different seasonal and depth preferences are a useful tool for reconstruction of temporal and vertical hydrographi c changes in the water column. In this study, isotopic and elemental ratios of Globigerinoides ruber (pink and white) and Neogloboquadrina dutertrei from Orca Basin core MD02-2551 provide information on the seasonality and thickness of Laurentide Ice Sheet (LIS) meltwater input to the Gulf of Mexico, as well as the productivity re sponse to sustained freshwater input during an interval of Marine Isotope Stage 3 (MIS 3). The d 18 O of G. ruber and N. dutertrei are used to demonstrate that meltwater was confined to the depth habitat of G. ruber a surface-dwelling species. Calculation of the d 18 O of seawater from paired measurements of d 18 O and Mg/Ca values of pink and white G. ruber suggests that LIS melting was not limited to the warmest summer months, and may have been linked to changes in seasonality in Gulf of Mexico sea-surface temperatur es. Changes in the d 13 C gradient between G. ruber and N. dutertrei shows that LIS meltwater input likely had an impact on the biological pump in the Gulf of Mexico, perhaps by providing nutrients that stimulated primary production. 3.2. Introduction Understanding the melting history of the North American Laurentide Ice Sheet (LIS) during the last glacial period is of considerable interest to paleoclima tologists because of its potential role in global climate change. It is currently unknown whether
33 this ice sheet was a passive responder to regional or global climate variabilit y, or if it served to trigger or amplify abrupt changes in climate. For example, the routing of freshwater by the LIS and the drainage of pro-glacial lakes to sites of deepwate r formation (Licciardi et al., 1999; Clark et al., 2001) has been implicated in causing changes in the strength of North Atlantic Deep Water (NADW), leading to cooling in t he North Atlantic region (Rahmstorf, 1995). Terrestrial and sedimentary archives pr ovide a detailed assessment of the timing and routing of LIS meltwater during Marine Is otope Stage 3 (e.g. Licciardi et al., 1999; Hill et al., 2006). However, many aspects of LIS input to the Gulf of Mexico, such as the seasonality of melting and the effect of meltwater on ocean productivity, remain poorly understood. The Orca Basin, in the northern Gulf of Mexico (Figure 11) is ideally located to record LIS meltwater input during the last glacial period because of its proxima l location to the mouth of the Mississippi River. Oxygen isotope data of Globigerinoides ruber from Orca Basin sediments have previously been used to document LIS meltwater input to the Gulf of Mexico during the last deglaciation (Leventer et al., 1982; 1983; Flower and Kennett, 1990). A recent study has built on this earlier work by using paired measurements of d 18 O and Mg/Ca SST of G. ruber (white) to calculate d 18 O sw and better constrain the timing and magnitude of meltwater input (Flower et al., 2004). Hill et al (2006) employed the same technique of paired d 18 O and Mg-SST on G. ruber (pink) to provide the first detailed melting history of the southern margin of the LIS during Mar ine Isotope Stage 3 (MIS 3; 24-57 ka). They document five intervals of freshwater input from 28-45 thousand years before present (ka) that do not match the abrupt Dansgaard-Oeschger temperature oscillations recorded in Greenland ice, a climate proxy commonly thought to reflect northern hemisphere temperatures (Figure 12; Hill et al., 2006). It appears, instead, that summer melting of the LIS may have occurred during Antarctic warming and likely contributed to sea-level variability during MIS 3. The geochemical composition of surface to upper thermocline dwelling (upper 100 m) foraminifera in Orca Basin sediments can be used to assess the seasonality and thickness of meltwater input, as well as the productivity response to intervals of sus tained freshwater input (Flower and Kennett, 1990). This is based on the premise that various
34 species of foraminifera have different seasonal preferences and depth habitats w ithin the water column, which is reflected in their oxygen isotopic composition. The d 18 O calcite ( d 18 O c ) recorded in foraminiferal tests is a function of the temperature and the d 18 O composition of the water ( d 18 O sw ) in which the calcite precipitated. Heavier d 18 O c values, for example, reflect a foraminifer habitat of deeper waters or colder times of the year, assuming minor changes in d 18 O sw Removal of the temperature component from the d 18 O c measurements allows the response of foraminifera species to changes in salini ty to be compared. The Mg/Ca ratio, a proxy for the temperature of foraminiferal calcification, is ideal for d 18 O sw estimation because it is measured on an aliquot of the calcite sample used for d 18 O c (Nurnberg et al., 1996; Hastings et al., 1998; Elderfield and Ganssen, 2000; Lea et al., 2000). Using multiple foraminifer species also allows reconstruction of depth and seasonally dependent changes in nutrient levels and productivity. The d 13 C of test calcite, which primarily reflects the d 13 C of the dissolved inorganic carbon pool ( d 13 C DIC ), should have a reverse trend to the d 18 O c in a depth profile where lighter d 13 C values occur with increasing depth. This is due to the utilization of 12 C by phytoplankton in the surface waters, which leaves the d 13 C DIC of the waters more enriched (Kroopnick, 1974). Reintroduction of the 12 C to the DIC pool at depth occurs during heterotrophic and bacterial respiration, completing the biological pump. An enhanced gradient betwee n surface and thermocline dwelling species would suggest a strengthening of the biolog ical pump. Globigerinoides ruber (pink), Globigerinoides ruber (white) and Neogloboquadrina dutertrei are three species found in the Gulf of Mexico (Brunner and Cooley, 1976; Be 1982) that have different seasonal preferences and depth distributions. G. ruber a symbiont-bearing spinose species, is the most shallow dwelling foraminifer (Be and Hamlin, 1967; Hemleben et al., 1989) and is therefore ideal for reconstructions of past sea surface temperature and salinity. The white and pink varieties of G. ruber are considered separate species based on their distinct occurrence, size, and isotopic compositions (Deuser, 1987; Deuser and Ross, 1989), as well as the genetic differences that have been observed through molecular phylogenetic analyses (Darling et al., 1997).
35 The pink G. ruber possesses a pigment within the test structure that gives the chambers the pink-reddish color (Be and Hamlin, 1967). MOCNESS plankton tows from the western Gulf of Mexico and the Caribbean Sea indicate that both white and pink G. ruber reside in the upper 50 m in the water column (Be 1982; Schmuker and Schiebel, 2002). Currently, there is no information on the seasonal flux of pink and white G. ruber in the Gulf of Mexico, so we use sediment trap studies from the nearby subtropical Sargasso Sea to speculate on the seasonal distribution of the two species. The Sarga sso Sea studies demonstrate that the white G. ruber are abundant throughout the year, while the pink G. ruber are more confined to the non-winter months during times of surfacewater stratification (Deuser, 1987; Deuser and Ross, 1989). The average annual flux in the number of white G. ruber tests is ~90 tests m -2 d -1 (Deuser, 1987), whereas the flux of pink G. ruber in the Sargasso Sea is an order of magnitude smaller (~5 tests m -2 d -1 ) (Deuser and Ross, 1989). Periods of reduced abundance of the white G. ruber correspond to times of greater pink G. ruber flux, suggesting that the two species occupy a different ecological niche (Deuser, 1987; Deuser and Ross, 1989). We suggest that the pink G. ruber are more representative of non-winter conditions, while white G. ruber represent a more mean annual signal in the Gulf of Mexico. The Mg/Ca-derived SST estimates on core-top samples of the two G. ruber species from the Orca Basin support this inference (Figure 13). Additional studies on the seasonal distribution of foraminifera indicate that the pink G. ruber are more abundant at higher temperatures than the white variety (Be and Hamlin, 1967; Tolderlund and Be 1971; Hemleben et al., 1989). Zaric et al (2005) quantified the temperature range of these two species by compiling flux data from t imeseries sediment trap observations throughout the Atlantic (the two species co-occur only in the Atlantic and Mediterranean; Thompson et al., 1979). They found the optimum SST range for white G. ruber to be 21.8-28.4C, while the range for pink G. ruber is 22.6-29.5C. G. ruber (white and pink) prefer high salinity waters of ~35.75-36.6 psu (Tolderlund and Be 1971; Kemle-von Mucke and Oberhansli, 1999), although they have been shown to withstand salinities from 22-49 psu in culture studies (Hemleben et al., 1989).
36 N. dutertrei is a symbiont-bearing non-spinose species that is confined to the euphotic zone, but is found deeper in the water column than G. ruber (Fairbanks and Wiebe, 1980; Be 1982; Fairbanks et al., 1982). It reaches peak abundances at depths corresponding to the thermocline and to the deep chlorophyll maximum where primary productivity is high and food supply is at a maximum (Fairbanks and Wiebe, 1980; Faribanks et al., 1982; Kemle-von Mucke and Oberhansli, 1999). In the Sargasso Sea, the greatest abundance of N. dutertrei is in the winter-spring (Deuser and Ross, 1989), with a maximum flux of ~8 tests m -2 d -1 (Deuser and Ross, 1989). N. dutertrei occurs over a temperature range of 17.2-27.0C and a salinity range of 35.75-36.63 psu in the western North Atlantic (Tolderlund and Be 1971), although it is capable of tolerating salinities between 25-46 psu and temperatures from 13-33C under laboratory conditions (Hemleben et al., 1989). The purpose of this study is to focus in detail on one of the meltwater events previously defined by Hill et al. (2006) in an attempt to better understand the seasonality and thickness of LIS meltwater, as well as the productivity response to intervals of sustained freshwater input during MIS 3. Based on an inferred non-winter preference of pink G. ruber in the Gulf of Mexico, the Hill et al. (2006) record should be a reflection of summer melting of the LIS. We choose to focus on meltwater event F4 (37.2-39.9 ka; Figure 12) because this event has the lowest d 18 O GOM ( d 18 O sw minus the contribution from ice volume) values and persists for the longest period of time. Further, F4 may be associated with the largest A1 warming in Antarctica, as recorded by oxygen is otopes in ice (Johnsen et al., 1972). There are three primary objectives: 1) to characterize t he seasonality of LIS melting and replicate the meltwater signal using a diff erent foraminifer species, 2) to determine if the freshwater lens was confined to the surface wate rs, and 3) to document any effect of meltwater on primary productivity in the Gulf of Mexico. 3.3. Methods Processing and sampling of Orca Basin core MD02-2551 and preparation of the planktonic foraminifera G. ruber (pink) samples for isotopic and elemental analyses are described in detail in Hill et al. (2006). For this study, ~40 G. ruber (white) and ~10 N.
37 dutertrei were picked from the 250-355 m size fraction over the interval 25.3-28.0 m. Depth was converted to age using the age model described in detail in Hill et al. (2006). Age control in the interval of this study is based on 7 AMS 14 C dates and the Laschamp geomagnetic event. Sedimentation rates over this interval range from ~20-70 cm/kyr The foraminifera were selected at 4 cm sample spacing within this interval, pr oviding one-half the resolution of the G. ruber (pink) analyses. The shells were gently crushed open between two glass plates and homogenized using a razor blade. A ~50 g aliquot was removed for stable oxygen and carbon isotopic analyses, which were performed at the College of Marine Science, University of South Florida using a ThermoFinnigan Delta Plus XL dual-inlet mass spectrometer with an attached Kiel III car bonate preparation device. The isotopic data are reported on the VPDB scale calibrated wit h NBS-19. Standard deviation is 0.06 for d 18 O c and 0.03 for d 13 C, based on measurements of the standard NBS-19 analyzed with MD02-2551 foraminifer samples (n=17). The remaining tests of the G. ruber (white), weighing ~350 g were cleaned separately for Mg/Ca analysis using the Cambridge method, which does not include the reductive cleaning step (Barker et al., 2003). Samples were dissolved in weak HNO 3 to yield calcium concentrations of ~20 ppm to minimize calcium concentration effects The Mg/Ca ratios were analyzed on a Perkin Elmer Optima 4300 dual view inductively coupled plasma-optical emission spectrometer (ICP-OES). A standard instrume nt-drift correction technique was routinely used. The analytical precision for Mg/Ca determinations used in this study is <0.9% root-mean standard deviation (1 s ), based on an ICP-MS calibrated standard solution. The pooled standard deviation of 23% replicate Mg/Ca analyses is 2.7% (d.f. = 14), which is <0.3 o C. Mg/Ca analyses for N. dutertrei were not performed as part of this study. 3.4. Results and Discussion 3.4.1. d 18 O c records The d 18 O c records of the three foraminiferal species (Figure 14) provide information about the northern Gulf of Mexico hydrography over the defined F4
38 meltwater interval (37.2-39.9 ka). The pink G. ruber d 18 O c values are the lowest of the three species, with a mean of 1.32 and a s.d. of 0.50. The mean d 18 O c for white G. ruber is 0.74, with a s.d. of 0.49. An abrupt shift of >1 to more depleted d 18 O c values occurs in the records of both the pink and white G. ruber at 39.9 ka, reflecting an increase in temperature or a decrease in d 18 O sw These lower d 18 O c values persist until 37.2 ka, although there are distinct differences in the trends in the two species over this interval. In particular, the white G. ruber has a two-step feature with a mean d 18 O c value from 38.3-39.9 ka (-0.9) that is ~0.5 more enriched than the value from 37.2-38.3 ka (-1.4). In comparison, the pink G. ruber shows a relatively flat trend with a mean d 18 O c value of 1.8. A positive ~1 shift in the d 18 O c values of both species occurs at 37.2 ka. Lighter values of the pink G. ruber relative to the white G. ruber species were also observed during the last deglaciation (Flower and Kennett, 1990). The mean d 18 O c value for N. dutertrei is 0.99, with a s.d. of 0.28. The d 18 O c values are consistent with those observed during the late glacial for this species but are ~1 heavier than early Holocene d 18 O c values (Flower and Kennett, 1990). There is no correlation between the d 18 O c records from the G. ruber and N. dutertrei species (R 2 =0.04 and 0.02 for white/dutertrei and pink/dutertrei, respectively). The best evidence for this is that the >1 spike observed in the G. ruber d 18 O c records during the F4 meltwater event is not seen in the N. dutertrei d 18 O c record. Despite the lack of correlation between the two records, there are intervals when the variability obse rved in the G. ruber record is also seen in the N. dutertrei record. For example, >1 negative isotopic excursions occur in both records at ~38 ka and ~40 ka, although these excursions in N. dutertrei are based on single data points and must be interpreted with caution. The fact that the large isotopic variability recorded by the G. ruber over the F4 meltwater interval is not seen in the N. dutertrei d 18 O c suggests that either the meltwater input from 37.2-39.9 ka was confined to the surface layers, or that N. dutertrei descended to greater depths to avoid the freshwater input. It is difficult to constrain the depth of t he meltwater lens without knowing the salinity tolerance of the three species. G. ruber is one of the most euryhaline species (Hemleben et al., 1989) and is shown to be one of the few foraminifera able to tolerate lower salinity waters in the natural envi ronment (van
39 Bennekom and Berger, 1984; Kemle-von Mucke and Oberhansli, 1999; Schmuker and Schiebel, 2002; Rohling et al., 2004). Plankton tow studies from the Caribbean Sea have shown that the abundance of N. dutertrei was positively related to freshwater input from the Amazon and Orinoco Rivers (Schmuker and Schiebel, 2002), suggesting that this species may be able to tolerate lower salinity waters as well. The simple st explanation, however, suggests that LIS meltwater was confined to the habitat of G. ruber likely in the surface layer. 3.4.2. d 18 O c gradients The magnitude of change observed between the N. dutertrei and G. ruber oxygen isotope records can be quantified by calculating the d 18 O gradient ( d 18 O) between the two species. An enhanced gradient indicates larger differences in d 18 O sw a greater temperature range between surface and depth, or a change in the species habitat. In order to accurately determine this gradient, it is necessary to take into account the t emperature effect on calcite precipitation (Spero et al., 2003). This results from the fact tha t G. ruber and N. dutertrei have different temperature: d 18 O c relationships based on species-specific factors that control calcification. Normalization of the N. dutertrei oxygen isotope data at 15C to G. ruber (white) requires adding +0.61 to the N. dutertrei d 18 O c record (Spero et al., 2003). Although this is an imperfect normalization method because it assumes calcite precipitation at a constant temperature of 15C, it is the best approach to com pare the two species without calcification temperature data. The d 18 O pink-dut ranges from a minimum of 2 to a maximum of 4 during peak meltwater input, while d 18 O white-dut ranges from 1.5-3.5. Normalized core top d 18 O pink-dut and d 18 O white-dut values from Orca Basin are ~2.75 and ~2.25, respectively. The normalization of the N. dutertrei d 18 O c values changes the magnitude of the difference between the N. dutertrei and G. ruber d 18 O c records, but does not change the fact that the gradient increases by ~2 during meltwater event F4. We do not attempt to normalize the G. ruber (pink) to the G. ruber (white) calibration because a separate temperature: d 18 O c calibration for pink G. ruber does not currently exist.
40 3.4.3. Mg/Ca records The mean Mg/Ca value for the pink G. ruber is 3.69 mmol/mol, with a s.d. of 0.22 mmol/mol, while the mean Mg/Ca value for the white G. ruber is 3.35 mmol/mol with a s.d. of 0.26 mmol/mol (Figure 15). We assume that post-deposition dissolution of Mg-rich parts of the foraminiferal test was minimal because the weight per foram does not change down core and the values are comparable to surface samples (15.75 ug; Appendix A). The presence of pteropods, which have an aragonitic shell and are more susceptible to dissolution, also provide support for minimal dissolution. Studies have demonstrated that Mg/Ca in shells decreases with water depth and inferred incre asing dissolution (Rosenthal and Boyle, 1993; Russell et al., 1994; Brown and Elderfield, 1996; Hastings et al., 1996). Although corrections for the effect of water depth on dissolution have been suggested (Dekens et al., 2003), the location of the Orca Basin site above the Atlantic lysocline indicates that depth had a minimal effect on Mg/Ca dissoluti on. We convert the Mg/Ca values to sea-surface temperature using calibrations tha t are based on sediment trap data from the Sargasso Sea and are appropriate to the two different G. ruber varieties (Anand et al., 2003). The Sargasso Sea study reports two separate equations for both pink and white G. ruber in the 250-355 m size fraction, which depend on whether the exponential constant (A) is fixed in the relationship Mg/Ca = B exp(AT) where A and B are constants and T is the temperature of calcificati on (see Table 3, Anand et al., 2003). The exponential constant describes the temperature sensitivity of the foraminifera. Although this may change for different species comprehensive studies are converging on an exponential value of 0.090 for all planktonic species (Elderfield and Ganssen, 2000; Dekens et al., 2002; Anand et al., 2003). This value indicates that a 9.0 0.3% change in Mg/Ca results in a 1C (+ 0.2) change in temperature. A comparison (not shown) of the temperatures derived using the equation with a fixed exponential constant, versus the equation without a fixed exponential constant for G. ruber (white) indicates a <0.3C difference between the two over the temperature range of this study, which is within measurement error. In contrast, the G. ruber (pink) temperatures show a >1.5C difference between the corresponding equations. Further,
41 the equation with the non-fixed exponential constant for pink G. ruber produces a much larger temperature range than the temperature range determined with the use of the fixed exponential equation. We choose to use the equations with the fixed exponential of 0.090 because of the convergence noted above, and because it provides a more conservative estimate of SST, particularly for the G. ruber (pink). The calibration for G. ruber (pink) in the 250-350 m size fraction with a fixed exponential constant is Mg/Ca = 0.381exp(0.090T). The calibration for G. ruber (white) in this same size fraction is Mg/Ca = 0.449exp(0.090T). These are the two equations used to derive temperature in this study. The mean temperature recorded by the pink G. ruber is ~3C warmer than the mean temperature recorded by the white G. ruber (~25C versus ~22C) (Figure 15). This is ~1C more than the temperature difference between modern pink (~27.3C) and white (~25.3C) G. ruber values as determined from coretop samples (Richey et al., unpublished data). Overall, the G. ruber (white) record is more variable than the G. ruber (pink) record, which may reflect different optimum growing seasons for the two species. In particular, there is a distinct >2r warming in the white G. ruber SST f rom 37.4-39.6 ka, during peak meltwater input, that is not observed in the pink G. ruber temperatures. 3.4.4. SST differences Differencing of the two temperature records ( T) provides information on changes in seasonality in SST and/or changes in the habitat of the two G. ruber species during LIS meltwater input, with the caveat that large changes in salinity may have affected the Mg/Ca ratios, and ultimately SST. The effect of salinity on Mg/ Ca in planktonic foraminifera is currently unknown. Initial laboratory culture studies by Nurnberg et al. (1996) and Lea et al. (1999) showed that a 1 psu salinity change would result in an average Mg/Ca change of ~6%. Given the temperature relationship of 9%/1rC, a 1 salinity unit change could account for 0.5-1rC temperature changes. However, a more recent laboratory study (Russell, 2004) showed that there was no significant change in Mg/Ca with changes in salinity. We make the assumption that
42 changes in salinity had minimal affect on Mg/Ca ratios, but also assume that any s alinity effects on Mg/Ca would have been similar between the pink and white G. ruber Simple box model calculations also demonstrate that the effect of riverine input on the Mg/Ca values is minimal based on the large difference in Mg and Ca concentrations in the Mississippi River and the Gulf of Mexico (Flower et al., 2004; Hill et al., 2006). The average T shows a distinct drop of ~1C from ~39.2-37.4 ka, during the later part of meltwater input. The rise in T at ~37.4 ka is followed by another >1C drop in the average T at ~36.2 ka, which coincides with a decrease in d 18 O c at this time (Figure 14). The decrease in T during peak meltwater input may be explained by a change in the seasonal range of SST in the Gulf of Mexico, if the pink G. ruber record non-winter conditions and the white G. ruber record mean annual conditions. A negative correlation (R 2 =0.63) between white G. ruber SST and T may indicate that warmer winter SSTs correspond to a decrease in the seasonal range of temperatures in the G ulf of Mexico. Warmer year-round temperatures in the Gulf of Mexico may have provided a source of heat to the southern LIS margin and contributed to melting. The relationship between T and LIS meltwater input needs to be assessed over several melting intervals, however, to establish any causal linkages. An alternative explanation for warmer temperatures recorded by the white G. ruber is that there was a change in the seasonal preferences of this foraminifer. The i nput of meltwater may have affected the seasonal distribution of the white G. ruber through changes in sea-surface salinity. The white G. ruber may have thrived during times of peak freshwater input in the summer months. A recent study from the Orca Basin demonstrated that a >0.6 change in d 18 O sw during the early Holocene corresponded to a reduction in the relative frequency of the pink G. ruber relative to the white G. ruber pointing to an affinity of the white variety to withstand less saline waters (LoD ico et al., in review). These results are consistent with reconstructions from the Mediter ranean, which show the resistance of white G. ruber to freshwater disturbances (Rohling et al., 2004). Changes in the seasonal concentrations of primary productivity in the Gulf of Mexico due to changes in nutrient input may have also led to a different seasonal
43 distribution of white G. ruber during times of meltwater input. In particular, meltwater flooding on the continent during summer months may have increased the nutrient supply to the northern Gulf of Mexico, stimulating primary production. In the modern system, the Mississippi River supplies nutrients that enhance primary production on the shelf (Lohrenz et al., 1990; Redalje et al., 1994). The white G. ruber may have had a preference for living during warmer months when meltwater input led to an increase i n primary productivity and hence food supply for the foraminifera. One drawback of this interpretation, however, is that neither pink or white G. ruber fluxes follow the organic carbon flux, an indicator of primary productivity, in the Sargasso Sea (Deuser, 1987; Deuser and Ross, 1989). This suggests that environmental parameters, such as temperature, may exert more control on the seasonal distribution of G. ruber than food supply alone. 3.4.5. d 18 O sw records In order to better understand the seasonal and depth influence of meltwater in the Gulf of Mexico, we removed the Mg/Ca-SST component from the G. ruber d 18 O c records. We applied the Orbulina universa high-light temperature: d 18 O relationship T=14.9-4.8( d 18 O c d 18 O sw ) (Bemis et al., 1998), which has been shown to be appropriate for G. ruber (white) (Thunell et al., 1999). There is currently not a separate temperature: d 18 O relationship for G. ruber (pink), so we apply the same equation to the pink G. ruber data and assume that the equation is appropriate to both G. ruber species. The equation allows us to solve for d 18 O sw which is a function of changes in ice volume and/or salinity variations. The white G. ruber d 18 O sw record is nearly identical to the pink G. ruber d 18 O sw record (Figure 16). A ~1 decrease in d 18 O sw occurs at 39.9 ka in the records of both species. This is followed by a prolonged interval of depleted d 18 O sw values (mean=0.7), and a return to more positive values at 37.2 ka. Sea-level fluctuations of <30 m during this interval (Siddall et al, 2003) can explain only 0.25 of the >1 changes observed in d 18 O sw based on the relationship 0.083 per 10 m sea level change (Adkins et al., 2001). Therefore, salinity variations due to changes in evaporation-
44 precipitation or freshwater input must account for most of the isotopic signal recorde d by the G. ruber We remove the contribution of global ice volume to the d 18 O sw signal using a high-resolution sea level record (Siddall et al., 2003), which leaves a residual local d 18 O sw record, defined as the d 18 O GOM (Figure 16) (Hill et al., 2006). 3.4.6. LIS meltwater input Based on the minimal contribution of sea level to the d 18 O sw record, removal of sea level changes the absolute isotopic composition, but has little effect on the overal l trends. In general, the two G. ruber d 18 O GOM records closely resemble each other, except from 41.5-40.7 ka where the pink G. ruber values are decreasing and white G. ruber values are increasing. It was previously demonstrated that the changes in the d 18 O GOM as recorded by pink G. ruber were dominantly controlled by LIS meltwater input (Hill et al., 2006). These conclusions were based on a mixing model created for the Gulf of Mexico that was used to reconstruct salinity changes using different freshwat er end members (precipitation versus LIS meltwater). The most conservative estim ates for salinity changes indicated a substantial contribution from meltwater to the d 18 O GOM signals. This is supported by recent reconstructions of the LIS, which place the southern margin of the ice sheet within the Mississippi River drainage basin during MIS 3 (D yke et al., 2002). Simple box model calculations (Table 2) also indicate that the volume of water necessary to create the observed d 18 O changes, if driven by precipitation, are unreasonable. This does not, however, rule out the possibility that the isotopic composition of the sources of moisture to the Gulf of Mexico or the Mississippi River drainage basin changed over time. More depleted source waters from high latitudes, for example, could contribute to the d 18 O GOM signal recorded by Orca Basin foraminifera. It is likely that meltwater input during MIS 3 was superimposed on a time when conditions in the Gulf of Mexico were more fresh due to enhanced precipitation. The fact that the d 18 O GOM record is consistently more negative than the modern Gulf of Mexico d 18 O sw value of 1.2 (Figure 16) would support this inference. An increase in precipitation during a time of high solar insolation may have resulted in the amount effect, which would have led to more negative d 18 O sw values It is also possible that the
45 d 18 O GOM signal represents meltwater only. Perhaps the LIS was continuously melting to the Gulf of Mexico during this time. Figure 16 demonstrates that we are able to replicate the F4 meltwater signal in a different species. Further, depending on the seasonal distribution of G. ruber during MIS 3, the d 18 O GOM signal recorded by this foraminifer may provide information on the seasonality of LIS melting, which depends on the ice sheets annual mass balance cyc le. The amount of snow and ice that is stored in the ice sheet follows a seasonal distribution of accumulation and ablation that varies with the climatic regime, especially the timing of the seasons and maximum precipitation (Benn and Evans, 1998). Typical midand high-latitude glaciers today are characterized by winter accumulation, followed by summer (June-September in the Northern Hemisphere) melting (Benn and Evans, 1998). Our limited understanding of the seasonal distribution of the white G. ruber during times of peak meltwater input preclude us from making definitive statements about the seasona lity of LIS melting. However, the Mg-SST reconstructions of the two G. ruber species indicate that the white G. ruber lived in colder times of the year, with average temperatures of ~22rC, relative to the pink G. ruber which lived in temperatures of ~25rC (Figure 15). Therefore, we can at least conclude that LIS melting was not confined to the warmest summer months. If it were, we would expect to see a reduced amplitude d 18 O GOM signal in the white G. ruber since this species appears to be recording average conditions throughout the year in the Gulf of Mexico. 3.4.7. d 13 C records The length of time that melting occurs during the year has important implications for understanding the magnitude of the effects of meltwater on the biological pump in the Gulf of Mexico. The d 13 C DIC at this site is a function of the interplay between Mississippi River waters that are depleted in 13 C (modern d 13 C = -7) and primary production, which leaves the surface waters more enriched in d 13 C (~2) due to the preferential sequestering of 12 C by photoautotrophs. Glacial meltwater contains insignificant amounts of CO 2 (Aharon, 1988), so the contribution of depleted d 13 C from Mississippi
46 River waters results from the river acquiring dissolved carbon during flooding that is sourced by 13 C-depleted soils on land. In order to use the d 13 C of the foraminiferal species to document changes in d 13 C DIC it is necessary to take into account the disequilibrium effects that occur during foraminiferal calcite precipitation (Mulitza et al., 1999; Spero et al., 1999). In gene ral, physiological processes such as symbiont photosynthesis, respiration and the carbonate ion effect (Spero et al., 1999) shift shell d 13 C away from isotopic equilibrium. Calcite d 13 C has been shown to be ~1 enriched relative to the d 13 C of DIC in seawater (Romanek et al., 1992). Recent studies have demonstrated that the magnitude of the d 13 C offset due to these physiological processes is species specific (Ortiz et a l., 1996; Mulitza et al., 1999; Spero et al., 2003). We apply a correction factor to account for the known isotopic offset between d 13 C of G. ruber (white) and N. dutertrei to the d 13 C DIC which is based on plankton tow studies from the subtropical Atlantic and Caribbean Sea (Mulitza et al., 1999; Spero et al., 2003). These d 13 C shell-DIC offsets are 0.94 for white G. ruber (Spero et al., 2003) and +0.5 for N. dutertrei (Mulitza et al., 1999). Data for G. ruber (pink) are not yet available, so we apply the same 0.94 correction that is applied to the white G. ruber record. Normalization of foraminiferal d 13 C to DIC allows us to more accurately reconstruct productivity changes in the water column (Figure 17). The more positive values of the G. ruber are consistent with removal of 12 C in the surface waters by photoautotrophs, while the more depleted values of the N. dutertrei indicate respiration of 12 C at depth. The enrichment in the pink G. ruber relative to the white G. ruber suggests that this species inhabits the more 12 Cand nutrient-depleted surface waters during surface water stratification in the summer months (Deuser and Ross, 1989). The white and pink G. ruber records in this study show similar trends to each other, with a ~0.4 decrease from ~40-41 ka, prior to peak meltwater input. The d 13 C increases from 39.9-37.2 ka, during the duration of the meltwater interval, in the records of both species, although the trend is more pronounced in the white G. ruber record. The N. dutertrei record the greatest variability in d 13 C. A >0.6 negative shift in d 13 C occurs from ~40-41 ka, prior to peak LIS melting. Values remain low from ~39.8-39.2 ka, at
47 which point they begin to increase to a maximum of ~1.0. The greater variability in the N. dutertrei relative to the G. ruber species suggests that the surface water d 13 C DIC is responding to the input of depleted Mississippi River waters in addition to enhanced surface primary productivity. Enhanced primary production would lead to more enriched d 13 C values in the G. ruber which may be balanced by freshwater with a high 12 C composition. 3.4.8. Biological pump Differencing of the G. ruber and N. dutertrei d 13 C ( d 13 C) records provides information on the strength of the biological pump (Figure 17). Both the d 13 C pink-dut and the d 13 C white-dut increase by ~0.5 from ~40-41 ka, prior to the onset of peak meltwater input. The increase in d 13 C pink-dut to a mean value of ~1.8 is higher than the modern d 13 C gradient (~1.3) between G. ruber (pink) and N. dutertrei as determined from Orca Basin core-top samples. In comparison, the d 13 C white-dut at this time is comparable to the modern d 13 C white-dut (~1.2). The increase in d 13 C is followed by maximum values of d 13 C in both species from ~39-39.9 ka. d 13 C values then decrease by ~0.5 in both species over the remainder of the meltwater interval. Meltwater input to the Gulf of Mexico may have contributed an increased supply of nutrients from the continent that stimulated primary production in the surface wate rs and led to an enhanced biological pump from ~39-39.9 ka, as indicated by the maximum d 13 C values. The decrease in d 13 C following this interval suggests that either the process of enhanced nutrient input eventually reached a threshold, at which point additional nutrients did not affect primary productivity, or there was a foraminifer al ecosystem shift in response to the increasing amount of freshwater entering the G ulf of Mexico. The increase in d 13 C relative to the defined meltwater interval may reflect the conservative criteria that were used to define meltwater events in the Gulf of M exico (Hill et al., 2006). Meltwater may have been entering the Gulf of Mexico as early a s ~40.5 ka, as indicated by both G. ruber d 18 O GOM records (Figure 16), bringing in nutrients that started to enhance the biological pump.
48 3.5. Conclusions In this paper, we used the seasonal distribution and depth preference of foraminifera from Orca Basin core MD02-2551 to better understand the seasonality and depth of LIS meltwater input and the water column response to intervals of sustained freshwater input during MIS 3. We used d 18 O c to document the response of G. ruber (white and pink) and N. dutertrei to freshwater input during a previously defined meltwater interval ca. 39 ka. We suggest that LIS melting was not limited to the warmest summer months and that it was likely confined to the surface layers. We further show that LIS meltwater likely had an impact on the biological pump, as indicated by d 13 C between the G. ruber and N. dutertrei perhaps by providing nutrients that stimulated primary production. One of the most important results from this study is that we were able to replicate the F4 meltwater event defined based on pink G. ruber calculating the d 18 O GOM from paired measurements of d 18 O c and Mg/Ca using white G. ruber It will be important to extend these studies over multiple meltwater events in order to better assess the role of the LIS in climate change.
49 Table 2. Volume reconstructions for d 18 O GOM record. Volume of water needed to satisfy the largest d 18 O GOM excursion (-0.2) using different freshwater endmembers. Precipitation volumes were compared to the annual rainfall in the Gulf of Mexico of 37.5 cm/yr (Ropelewski and Halpert, 1996). Mississippi River volumes were compared to the largest historical flood from 1927, which had a discharge of 2680 km 3 /yr (Barry, 1997). The annual Mississippi River discharge is 507 km 3 /yr (Dinnel and Wiseman, 1986). Units under the compared to columns are the number of times (X). For example, if the freshwater lens was 25 m thick and covered the entire Gulf of Mexico, 57 times the annual precipitation over the Gulf of Mexico would be necessary to satisfy the isotopic variability recorded by the foraminifera (see first line, column 3). Volume of freshwater needed (m 3 ) Depth of foram/area of GOM relative to total area Precip -3.5 Compared to annual MR -7 Compared to largest flood LIS -15 Compared to largest flood LIS -30 Compared to largest flood 25m / 100% 1.69x10 13 57 8.18x10 12 3 3.76x10 12 1.5 1.87x10 12 <1 25 m / 50% 8.43x10 12 28 4.09x10 12 2 1.88x10 12 <1 9.43x10 11 <1/2 65m / 100% 2.19x10 13 74 2.13x10 13 8 9.78x10 12 3.5 4.86x10 12 2 65m / 50% 4.38x10 13 150 1.06x10 13 4 4.89x10 12 2 2.43x10 12 <1
50 Figure 11. Map of Orca Basin in the Gulf of Mexico showing location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) and the extent of the Laurentide Ice Sheet during MIS 3 (from Dyke et al., 2002).
51 -44 -43 -42 -41 -40 -39 -38 -37 -36GISP2 d 18 O ice ( VSMOW) -0.5 0 0.5 1 1.5d 18 O GOM ( VSMOW) -42 -41 -40 -39 -38 283032343638404244Byrd d 18 O ice ( VSMOW)Calendar age (k.a.)F1F2F3F4 F5 A1 4 567 891011 12 H3 H4 Fresher Saltiera b c Figure 12. Comparison of Orca Basin core MD02-2551 d 18 O GOM determined from G. ruber (pink) during MIS 3 with ice core records. a. GISP2 d 18 O ice (Grootes et al., 1993). b. Orca Basin d 18 O GOM with mean value indicated by horizontal bar. d 18 O GOM was calculated by subtracting global ice volume from the d 18 O sw record (Hill et al., 2006). c. Byrd d 18 O ice record (Johnsen et al., 1972) on the GISP2 timescale, based on synchronization of methane concentrations within the two ice cores (Blunier and Brook, 2001). Numbers refer to Greenland interstadials. Light grey bars and the letter F (numbered 1-5) indicate freshwater events referred to in the text. Dark grey bars and letter H indicate Heinrich events. A1 refers to Antarctic warming event number 1 (Blunier and Brook, 2001).
52 20 22 24 26 28 30 32 JanFebMarAprMayJunJulAugSepOctNovDecDegrees CelsiusMonth G. ruber (pink) Anand fixed G. ruber (white) Anand fixed Figure 13. Annual cycle of sea-surface temperature in the Gulf of Mexico (Levitus 2003). Also shown are Orca Basin core-top Mg/Ca values of pink and white G. ruber converted to sea-surface temperature using Anand et al (2003). Core top values are from Richey et al. (unpublished data). Equations used to convert Mg/Ca to SST are described in detail in section 3.4.2 of the text. Note that the pink G. ruber value represents a nonwinter-weighted temperature, while the white G. ruber value represents a mean-annual temperature.
53 -3-2-1 012 343536373839404142 G. ruber (pink) G. ruber (white) N. dutertreid 18 O c ( VPDB)Age (ka) Figure 14. Raw d 18 O c of G. ruber (pink), G. ruber (white) and N. dutertrei from Orca Basin core MD02-2551. Grey bar indicates F4 Laurentide Ice Sheet meltwater inter val as defined by Hill et al. (2006).
54 2.5 3 3.5 4 4.5 G. ruber (pink) G. ruber (white)Mg/Ca (mmol/mol) 20 21 22 23 24 25 26 27Temperature ( o C) -1 0 1 2 3 4 5 6 343536373839404142D T G. ruber (pink-white) ( o C)Age (ka) a b c Figure 15. Mg/Ca and SST data on pink and white G. ruber from Orca Basin core MD022551 during MIS 3. a. Mg/Ca. b. Mg/Ca was converted to SST using Mg/Ca=0.381exp[0.09 X SST (C)] for pink G. ruber and Mg/Ca=0.449exp[0.09 X SST (C)] for white G. ruber (Anand et al., 2003). Error on temperature measurements is determined to be ~0.85C based on compounding the error from the Mg-SST calibration and replicate analyses (Beers, 1957). c. Difference between G. ruber (pink) and G. ruber (white) temperatures calculated by simple subtraction of the two records. Error is determined to be ~1.2C (Beers, 1957). Grey bar indicates F4 Laurentide Ice Sheet meltwater interval as defined by Hill et al. (2006).
55 -0.5 0 0.5 1 1.5 2 2.5 G. ruber (pink) G. ruber (white)d 18 O sw ( VSMOW) -1 -0.5 0 0.5 1 1.5 343536373839404142d 18 O GOM ( VSMOW)Age (ka) a b Figure 16. d 18 O sw and d 18 O GOM of pink and white G. ruber from Orca Basin core MD022551. a. d 18 O sw and b. d 18 O GOM d 18 O sw was calculated from d 18 O c and Mg-SST using T( o C) = 14.9-4.8*( d 18 O c d 18 O sw ) (Bemis et al., 1998). 0.27 was added to convert to VSMOW. Error on d 18 O sw is 0.25 based on propagating the error through the analytical errors and the combined Mg-SST and SSTd 18 O relationships (Beers, 1957). d 18 O GOM was calculated by subtracting ice volume from d 18 O sw using the global sea-level record (Siddall et al., 2003). Horizontal bar indicates modern d 18 O GOM Error on d 18 O GOM is 0.27 based on compounding the error from d 18 O sw and the error on the sea level record (Beers, 1957). Grey bar indicates F4 Laurentide Ice Sheet meltwate r interval as defined by Hill et al. (2006).
56 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 G. ruber (pink) G. ruber (white) N. dutertreid 13 C ( VPDB) 0 0.4 0.8 1.2 1.6 2 2.4Normalized d 13 C ( VPDB) 0.4 0.8 1.2 1.6 2 343536373839404142 pink-dut white-dutDd 13 C ( VPDB)Age (ka) a b c Figure 17. d 13 C of G. ruber (pink), G. ruber (white) and N. dutertrei from Orca Basin core MD02-2551. a. Raw d 13 C, b. Normalized d 13 C as described in text and c. d 13 C based on the difference between G. ruber (white) and G. ruber (pink) calculated by simple subtraction of the two records. Error on this measurement is 0.14 (Beers, 1957). Grey bar indicates F4 Laurentide Ice Sheet meltwater interval as defined by Hill et al. (2006).
57 Chapter 4 Mississippi River flooding during the last glaciation 4.1. Abstract Sedimentary basins proximal to major rivers can provide a coherent, highresolution assessment of the oceanic and continental responses to changing hydrologic regimes. Documenting these regime changes under climatic extremes, includi ng glacial time periods that are often considered cold and dry in temperate North America, is important to understand the potential range of variability associated with differe nt hydrologic conditions. The Orca Basin in the northern Gulf of Mexico is ideally situate d to record inputs from the Mississippi River and to relate these inputs to changing moisture balance over central North America. A suite of organic and inorganic geochemical analyses from Orca Basin core MD02-2551 is used to document a 7,000-year interval of enhanced Mississippi River discharge and inferred wet conditions ove r North America from 31-38 thousand years ago (ka), during glacial Marine Isotope Stage 3. The period of enhanced river discharge is decoupled from Laurentide Ice Sheet (LIS) meltwater input, and occurs during a time of rising summer insolation in the northern hemisphere. We speculate that increasing summer insolation on the orbital scale m ay have led to a northward migration of the Intertropical Convergence Zone and an intensification and westward shift in the canonical position of the Bermuda High, which shuttles moisture to the North American continent and contributes to air mass convergence and flooding in the Mississippi River drainage basin.
58 4.2. Introduction Over the past decade, it has become increasingly more apparent that instrumental records do not capture the full range of hydrologic variability that is manifest in our climate system (Overpeck, 1996). North American proxy records, for example, show large changes in the magnitude and duration of floods throughout the Holocene that exceed those observed over the last century (Knox, 2000, and references therein). Understanding the modes of climate variability that lead to increased flooding has important implications for impact assessment related to future changes in reg ional climate. This is especially timely given predictions that changes in the conti nental hydrologic cycle may lead to our biggest climate surprises in the next sever al decades (Overpeck, 1996). The convergence of air masses with different temperatures and moisture content is the primary cause of floods in the Mississippi River drainage basin (Hirschboec k, 1991; Knox, 2000). Warm, moist air masses from the south (Gulf of Mexico) and southeast (Atlantic Ocean) interact with cold, dry air masses from the north (A rctic region) and northwest (Pacific Ocean) to produce storm tracks over the Mississippi River watershed (Hirschboeck, 1991). Floods in the Upper Mississippi Valley, which encompasses 80% of the drainage basin, become larger when air masses originating f rom the Gulf of Mexico are present (Knox, 2000). The relative importance of the Gulf of Mexico moisture flux is largely controlled by the seasonal migration of the Atl antic Bermuda High, which brings precipitation to the central United States through anticyclonic atmospheric circulation (Mo et al., 1997). The Bermuda High is located further in the western Atlantic during the boreal summer and moves to the eastern Atlantic during the winter (Figure 18) (Machel et al., 1998). The boundary of air-mass convergence and its associated storm tracks often closely follow the jet stream axis. The extreme Mississippi River flooding in the summer of 1993 was associated with a southward displacement of the jet stream, which allowed for the cyclonic disturbances in the atmosphere to tap into the Gulf of Mexico moisture source (Trenberth and Guillemot, 1996). The southerly displaced jet stream was also accompanied by an increase in the intensity of storm tracks, which has been linked to sea-
59 surface temperatures and anomalous convection in the Pacific (Trenberth and Guillem ot, 1996, and references therein). The southward displacement of the jet stream and the convergence of air masses from the Gulf of Mexico and Canada set up a northeast/southwest trending jet axis (Figure 18) that led to the 1993 flood (Wahl et al., 1993) and has similarly been invoked to explain large floods that occurred during the Holocene (Knox, 2000). Wet conditions in the central United States and floods in the Mississippi River drainage basin have been linked to modes of climate variability on multiple timescal es from the last deglaciation to the present (Forman et al., 1995; Knox, 2000). Evidence for glacial-age floods in the Mississippi River drainage basin is limited, however because few terrestrial records exist and it is difficult to find archives that provide a continuous record of continental hydrology. Although the Northern Hemisphere ice sheets altered North American regional precipitation-evaporation (P-E) patterns, leading to shi fts in the continental moisture balance, it has been suggested that the presence of continental i ce sheets would likely suppress floods in the (Upper) Mississippi River valley due to cold, dry air masses associated with circulation from the north/northwest (Knox, 2000). In this study, we use a suite of organic and inorganic proxies from Orca Basin (2656.77N, 9120.74W) core MD02-2551 from the northern Gulf of Mexico (Figure 18), to document sediment delivery from the Mississippi River and draw linkages between freshwater input and marine production in the Gulf of Mexico. We have identified an interval of increased flooding in the Mississippi River drainage basi n during marine isotope stage 3 (MIS 3; 24-57 ka), a time of intermediate ice volume, characterized by abrupt changes in records of high-latitude temperature, ocean circulation, and hydrologic variability. Sea level stood on average 80 m below present during this time, and fluctuated by <15 m (Siddall et al., 2003). Although the ~50 year resolution of our sediment record does not allow delineation of seasonal variability such as convective processes (thunderstorms and tropical storms) during the summer/fal l, and rapid snowmelt during the spring (Hirschboeck, 1991), it provides an integrated assessment of the response of the Mississippi River drainage to changes in continent al hydrology related to air mass convergence.
60 4.3. Scientific approach 4.3.1. Assessing Mississippi River discharge The Mississippi River system drains 41% of the continental U.S. and supplies sediment to the northern Gulf of Mexico. Coarse-grained sands, silts, and dense organic matter are retained in bays, estuaries and the inner continental shelf, while fine -grained clays and soil organic matter are transported significantly offshore. On the se asonal cycle, the rivers suspended sediment load varies positively with discharge; ther efore, an increase in discharge should lead to a greater delivery of sediment to the Gulf of Me xico. Times of increased discharge over the last 100 years have been linked to well-documented flood events in the Mississippi River drainage basin (Poore et al., 2001). We use this information to infer that an increase in the concentrations of terrestria lly derived material in Orca Basin sediments results from enhanced flooding in the Mississ ippi River drainage basin. The concentrations of high-molecular weight (HMW) n-alkanes derive d from terrestrial plants (Pancost and Boot, 2004) and the weight percent insoluble residue (IR (wt%) = 100-[CaCO 3 (wt%) + total organic carbon (wt%)]) are used as proxies for continentally derived input in this study. The HMW n-alkanes are a robust proxy for the input of terrigenous organic matter because they are one of the major compound groups synthesized as epicuticular waxes of terrestrial higher plants (Pancost and Boot, 2004 and references therein). R ivers transport large amounts of vascular plant material to the ocean as suspended particul ates, and terrigenous biomarkers are an important component of sediments in coastal regions with high fluvial input (Pancost and Boot, 2004). Long chain n-alkanes are relatively resistant to degradation, and assessment of the n-alkane odd-over-even predominance (CPI index) gives us confidence that the terrestrial biomarkers used in this study have had minimal diagenetic alteration and are a reliable source for the concentration of terrigenous biomass. The HMW n-alkanes show a 5-fold increase in concentration from 26.2-24.4 m relative to intervals above and below (Figure 19). The high concentration of the HMW n-alkanes does not simply result from the dilution of marine constituents, because the simultaneous increase in the low-molecular weight (LMW) n-alkane pr oxy
61 used for marine production (see section 4.3.2) suggests that both terrestrial and marine components increased (Figure 19). The increase in the HMW n-alkanes is simultaneous with a peak in the insoluble residue over this interval (Figure 19b). The percent insoluble residue from Orca Basi n sediments is composed of silica oxides and other lithogenic constituents (Brown and Kennett, 1998), as well as authigenic sulfides. The lack of black coloring of the sediments suggests that authigenic sulfides contribute a very small percenta ge to the total insoluble residue, and that this fraction is primarily composed of terrestrially derived clays. The increase in the concentration of the percent insoluble residue from 26.2-24.4 m compared to surrounding intervals reflects an increased flux of continental mater ial to the Gulf of Mexico. 4.3.2. Documenting Gulf of Mexico productivity Northern Gulf of Mexico sediments should also record the influence of enhanced Mississippi River discharge on primary productivity. The Mississippi River car ries a high dissolved inorganic nutrient load that stimulates primary production in the Gulf of Mexico, and higher discharge has been linked to increased nutrient concentrations and higher levels of primary productivity during flood events (Rabalais et al., 1998). We use the organic d 13 C ( d 13 C org ) of the bulk sediment (Jasper and Gagosian, 1990) and the abundance of LMW n-alkanes (Ohkouchi et al., 1997) to document changes in primary production. The weight percent of the coarse fraction (>63 m) predominantly reflects variability in the percent of foraminifer tests and may be an indicator of secondary marine production. A shift from 26 to -23 occurs in the d 13 C org from 26.2-24.4 m (Figure 19c). The d 13 C org of the bulk sediment in this region is controlled by two processes: 1) changes in the relative contribution of marine ( d 13 C ~ -20) versus terrestrial ( d 13 C ~ -26) material to the bulk organic carbon pool (Jasper and Gagosian, 1990) and 2) variability in the relative contribution of C 3 ( d 13 C ~ -28) versus C 4 ( d 13 C ~ -15) plant types to the terrestrial component of the bulk organic matter (Goni et al., 1997). Compound specific carbon isotopic analyses of HMW n-alkanes (data not shown) range from 29 to 30.5,
62 confirming the dominance of C 3 plants (assuming a typical 2-4 fractionation between lipids and plant biomass). Therefore, the 3 isotopic excursion from 26.2-24.4 m likely reflects an enhancement of marine production to the bulk organic matter pool. Estimates from previous studies in the Gulf of Mexico indicate a similar positive shift in d 13 C between glacial/interglacial times, which was attributed to changes in the relative contribution of marine versus terrestrial material to the bulk organic carbon pool (Ja sper and Gagosian, 1990). The isotopic shift in the d 13 C org coincides with a delayed 5-fold increase in the concentration of the LMW n-alkanes (Figure 19d). LMW n-alkanes are synthesized as materials of buoyancy regulation, thermal insulation and energy storage in many ma rine organisms (Ohkouchi et al., 1997, and references therein). A study from the Pacific showed that the latitudinal distribution of the LMW n-alkanes varied positively with nutrient concentrations, indicating that these compounds can be used as marine biomarkers to reconstruct biological productivity (Ohkouchi et al., 1997). The delayed response in the LMW n-alkanes relative to the d 13 C org shift likely reflects an ecosystem response to changing environmental parameters where specific algae biosynthesi ze the nalkanes in relatively different proportions. Selective degradation of the signal i s not likely because there are no sedimentological characteristics in the core tha t indicate changing redox conditions, and there are no large variations in the organic carbon concentrations, which might suggest enhanced degradation. The response of secondary production to the increased primary productivity is recorded by a 15-fold increase in the percent coarse fraction (>63 m) from 26.2-24.4 m (Figure 19e). Two lines of evidence suggest that this increase is not just a preser vation signal: 1) the depths before and after the interval do not have proportionally larger amounts of broken tests and 2) the average weight per foraminifera does not change throughout the core, indicating no enhanced dissolution. 4.4. Sources of freshwater Collectively the signal in these five proxies point to intensification of Missis sippi River discharge from 26.2-24.4 m, which is ~31-38 ka when converted to age. The age
63 model is based on 18 14 C AMS dates on monospecific samples of pink Globigerinoides ruber The radiocarbon dates were corrected for an assumed constant reservoir age of 400 years and converted to calendar age using a high-resolution radiocarbon calibration developed on sediments from the Cariaco Basin. See Hill et al. (2006), and references therein, for details of the age model. The sedimentation rates from 31-38 ka do not exceed those of other intervals in the record, which suggests that our results cannot be interpreted in terms of Mississippi River delta lobe migration. There are two pr imary mechanisms to explain enhanced fluvial input from 31-38 ka: 1) LIS meltwater discharge and 2) increased P-E due to changes in atmospheric circulation patterns. We explore both of these scenarios below. 4.4.1. Laurentide Ice Sheet meltwater Meltwater megafloods during the last deglaciation produced sustained flows of low sediment concentration, which led to major downcutting in the Mississippi River system. It is believed that incision into clay-rich terrace deposits in upper Mis sissippi River tributaries produced a fine suspended sediment load that was deposited to the Gulf of Mexico and recorded in Orca Basin sediments (Brown and Kennett, 1998 and references therein). This increase in relative clay abundance correlates w ith a distinct negative anomaly in the calculated value of d 18 O of seawater ( d 18 O sw ), which has been attributed to LIS meltwater input (Flower et al., 2004). If the increase in the percent insoluble residue in our record from 31-38 ka is an indication of LIS meltwater flooding, we may expect to observe similar depleted va lues in the d 18 O sw throughout this interval. The d 18 O sw was previously calculated from paired measurements of oxygen isotopes ( d 18 O) and Mg/Ca analyses on the planktonic foraminifer G. ruber from the same samples used in this study (Hill et al., 2006). Removal of the sea-level component from the d 18 O sw isolated local d 18 O sw which was defined as d 18 O GOM and interpreted in terms of salinity changes. The d 18 O GOM record indicates five episodes of LIS meltwater input (Figure 20) (Hill et al., 2006). There appears to be no strong correlation between the insoluble residue and the d 18 O GOM (Figure 20), suggesting that meltwater flooding was not the primary control on
64 terrigenous input. In particular, the d 18 O GOM values increase during the first half of inferred terrigenous input suggesting a reduction of meltwater flow down the Missi ssippi River. In addition, an increase in terrestrial input is only observed during meltwater events 2 and 3, but not during meltwater events 1, 4, and 5 (Figure 20). The G. ruber d 13 C CaCO3 record provides further evidence that the terrestrial input from 31-38 ka was not driven by meltwater discharge (Figure 20). The d 13 C CaCO3 reflects the d 13 C of the dissolved inorganic carbon pool and is controlled by the interplay of primary production and the d 13 C of Mississippi River water (-7). Enhanced primary production leads to greater foraminiferal d 13 C values due to the preferential selection of light 12 C by photoautotrophs. Meltwater events 1, 4, and 5 correspond to negative d 13 C CaCO3 values, suggestive of riverine dominated input, but there is no d 13 C CaCO3 evidence for riverine input from 31-38 ka. Instead, we suggest that enhanced Mississippi River flooding over this interval led to increased nutrient delivery to the Gulf of Mexi co, which stimulated primary production. This resulted in more enriched d 13 C CaCO3 values from 31-38 ka relative to the d 13 C CaCO3 values corresponding to meltwater events 1, 4, and 5. 4.4.2. Atmospheric circulation patterns Alluvial records from the Upper Mississippi River Valley show an increase in the magnitude and frequency of floods in the Holocene at intervals lasting up to 2,000 years (Knox, 2000). The 7,000-year period of enhanced riverine discharge from 31-38 ka reported here is the longest documented record of Mississippi River flooding that is unrelated to LIS meltwater. A comparison of the insoluble residue to June insolation at 30N shows that the interval of enhanced Mississippi River discharge corresponds to an interval of rising summer insolation (Figure 20). The length of our record prevents us from assessing the response of Mississippi River discharge to multiple preces sional cycles, but to first order there appears to be a strong linkage between wet conditions i n the central U.S. and high summer insolation. These results support previous work that links periods of landscape stability (inferred wet times) in the central United States during the Holocene (6 ka) and the last
65 deglaciation (12 ka) to rising solar insolation (Forman et al., 1995). Forman et al. (1995) attributed wet conditions on the continent to a westward shift of the Bermuda High caused by rising insolation. The position and strength of the Bermuda High are closely tied to the position of the Intertropical Convergence Zone (ITCZ) on the annual cycle, where the maximum northward extent of the ITCZ, caused by high insolation during the Northern Hemisphere summer, corresponds to a westward migration and intensificat ion of the Bermuda High (Figure 18) (Machel et al., 1998, and references therein). We propose that rising summer insolation from 31-38 ka similarly led to a northward migration of the ITCZ and a westward expansion of the Bermuda High. The greater moisture flux into the continental U.S. could have led to an enhancement of storm fronts through the interaction of a southerly displaced jet stream in a manner simil ar to that proposed to explain the 1993 Mississippi River floods (Trenberth and Guillemot, 1996). Anomalous circulation patterns in the Pacific may have served to amplify these changes and increase flooding. Records from two lakes in central Mexico provide support for enhanced moisture over this interval. These lakes experienced relatively deep and fresh phases from 30-38 ka (Caballero et al., 1999; Bradbury, 2000), which Bradbury (2000) suggested may be attributed to a more effective moisture supply from the Gulf of Mexico. The influence of precessional forcing on precipitation patterns in the tropica l and subtropical Atlantic has been observed in other proxy records from the region (eg. Hillesheim et al., 2005 and references therein), pointing to the relative importance of orbital-scale solar insolation on the ITCZ/Bermuda High atmospheric systems 4.5. Conclusions Documenting ocean-continent linkages in the context of changing continental hydrology has important implications for understanding climate variability in the pa st and making predictions for future fluctutations in the hydrologic cycle. Here, we provide the first evidence for Mississippi River flooding during a glacial time that is unre lated to LIS meltwater. We suggest that an increase in the frequency and/or intensity of floodi ng from 31-38 ka is linked to rising summer insolation on the orbital timescale and a
66 possible westward shift of the Bermuda High. Additional records from the subtropical Atlantic region will aid in assessing the role of precession on precipitation patt erns controlled by the ITCZ and the Bermuda High during glacial times.
67 Figure 18. Map of Orca Basin in the Gulf of Mexico showing the location of core MD02-2551 (2656.77N, 9120.74W, 2248 m water depth) relative to the Mississippi River (MR) drainage basin. Core MD02-2551 was recovered in July 2002 by the R/V Marion Dufresne as part of the IMAGES (International Marine Past Global Changes Study) program. A schematic representation of the Northern Hemisphere (NH) sea sonal migration of the Intertropical Convergence Zone (ITCZ) and the Bermuda High is shown (Machel et al., 1998). Air mass trajectories (arrows) and approximate position of t he jet stream (line with arrows) represent weather patterns during the 1993 Mississi ppi River flood (Wahl et al., 1993).
68 0.5 1 1.5 2 2.5C25-31/TOC (mg/g) 75 80 85 90% Insoluble Residue -27 -26 -25 -24 -23d13Corganic ( PDB) 0 0.02 0.04 0.06 0.08 0.1C16-19/TOC (mg/g) 0 0.5 1 1.5 2 22002400260028003000% Coarse Fraction (>63 m m)Depth (cm) ab c d e Figure 19. Suite of organic and inorganic proxies from Orca Basin core MD02-2551 that collectively point to enhanced Mississippi River discharge and increased primary productivity in the Gulf of Mexico from 26.2-24.4 m (grey bar). Proxies for terrestrial input include: a. High-molecular weight n-alkanes (C 25-31 ) derived from terrestrial plants and b. Weight percent insoluble residue, primarily composed of clays. Proxies for the algal response to increased nutrient input include: c. Bulk organic d 13 C, d. Lowmolecular weight n-alkanes (C 16-19 ) derived from algal plants, and e. Percent coarse fraction.
69 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8d13CCaCO3 ( PDB) 75 80 85 90% Insoluble Residue 480 485 490 495 500 505 283032343638404244Insolation (W/m2) at 30oNAge (ka) -0.5 0 0.5 1 1.5d18OGOM ( VSMOW)a b c d 12345 Figure 20. Comparison of the timing of Mississippi River flooding relative to LIS meltwater and summer insolation at 30N. a. Weight percent insoluble residue, b. Orca Basin d 18 O GOM interpreted primarily as LIS meltwater (Hill et al., in press), c. G. ruber d 13 C CaCO3 and d. June 21 st insolation at 30N. Grey bar indicates period of inferred flooding. Numbers 1-5 indicate meltwater events defined in Hill et al., 2006.
70 Chapter 5 Summary Numerous paleoclimate archives document millennial-scale climate variabi lity during the last glacial period, yet the cause of these abrupt climate changes rem ains poorly understood. The research presented in this dissertation focuses on abrupt climate change in the North American hydrologic cycle, including Laurentide Ice Sheet (LIS ) meltwater, during an interval of Marine Isotope Stage 3. Sedimentary basins proxim al to major rivers can offer a coherent, high-resolution assessment of the oceanic and continental responses to changes in continental moisture balance. In this dissertati on, a sediment core from the Orca Basin, northern Gulf of Mexico, was used to document 1) the timing of melting of the southern margin of the LIS, 2) the seasonality and thickness of meltwater input and its effects on marine productivity, and 3) an interval of enhanced Mississippi River discharge and inferred mid-continental wet conditions, from 28-45 ka. The Orca Basins proximal location to the mouth of the Mississippi River is ideal t o record North American Laurentide Ice Sheet meltwater input because the Missi ssippi River served as one of the main conduits for glacial meltwater. Further, inputs from t he river related to changes in evaporation-precipitation over the North American contine nt offer an integrated assessment of mid-continental hydrologic change because the Mississippi River drainage basin covers ~40% of the contiguous U.S. The response of primary productivity to these two different freshwater inputs can be used to evaluate ocean-continent linkages in the context of changing hydrology. Paired measurements of d 18 O and Mg/Ca on the planktonic foraminifer Globigerinoides ruber (pink) were used to calculate the d 18 O of seawater and provide the first detailed melting history of the southern margin of the LIS during MIS 3. Five intervals of freshwater input from 28-45 ka do not match the abrupt Dansgaard-Oeschger
71 (D/O) temperature oscillations recorded in Greenland ice, a climate proxy com monly thought to reflect northern hemisphere temperature. It appears, instead, that summer melting of the LIS may have occurred during Antarctic warming and likely contribute d to sea-level variability during MIS 3. One of the most significant conclusions from thi s work is that the routing of freshwater to the Gulf of Mexico, and away from the North Atlantic, did not lead to each of the D/O warmings recorded in Greenland ice over this interval. Therefore, a simple routing hypothesis whereby changes in the strength of North Atlantic Deep Water depend on the routing of meltwater by the LIS, cannot explain the millennial-scale D/O cycles. Isotopic and elemental ratios of Globigerinoides ruber (pink and white) and Neogloboquadrina dutertrei which have different seasonal and depth preferences, provide information on the seasonality and thickness of LIS meltwater input to the Gulf of Mexico, as well as the productivity response to sustained freshwater input. The d 18 O of G. ruber and N. dutertrei were used to demonstrate that the largest meltwater interval was confined to the depth habitat of G. ruber a surface-dwelling species. Calculation of the d 18 O of seawater of pink and white G. ruber suggests that LIS melting was not limited to the warmest summer months, and may have been linked to changes in seasonality in Gulf of Mexico sea-surface temperatures. The d 13 C gradient between G. ruber and N. dutertrei shows that LIS meltwater input likely had an impact on the biological pump in the Gulf of Mexico, perhaps by providing nutrients that stimulated primary production. A suite of organic and inorganic geochemical analyses document a 7,000-year interval of enhanced Mississippi River discharge and inferred wet conditions over Nort h America from 31-38 ka that was decoupled from LIS meltwater input. The interval of increased river discharge occured during a time of rising northern hemisphere summe r insolation. Increasing summer insolation on the orbital scale may have led to a northward migration of the Intertropical Convergence Zone and an intensification and westward shift in the conical position of the Bermuda High, which shuttles moisture to the North American continent and contributes to flooding in the Mississippi River drainage basin.
72 References Cited Adkins, J.R., and D.P. Schrag, 2001, Pore fluid constraints on deep ocean temperature and salinity during the last glacial maximum, Geophysical Research Letter s, 28, 771774. Aharon, P. 1988, Oxygen, carbon and U-series isotopes of aragonites from Vestfold Hills, Antarctica: Clues to geochemical processes in subglacial environment s, Geochimica, Cosmochimica Acta, 52, 2321-2331. Alley, R.B., Sowers, T., Mayewski, P.A., Stuiver, M., Taylor, K.C., and Clark, P.U., 1997, Holocene climate instability: A prominent, widespread event 8,200 years ago, Geology, 26(6), 483-486. Alley, R.B., and Clark, P.U., 1999, The deglaciation of the Northern Hemisphere: A global perspective, Annual Reviews of Earth and Planetary Sciences, 27, 149-182. Alley, R.B., Anandakrishnan, S., and Jung, P., 2001, Stochastic resonance in the North Atlantic, Paleoceanography, 16(2), 190-198. Alley, R.B., Marotzke, J., Nordhaus, W.D., Overpeck, J.T., Peteer, D.M., Pielke Jr., R.S., Pierrenhumbert, R.T., Rhines, P.B., Stocker, T.F., Talley, L.D., and Wallace, J.M., 2003, Abrupt climate change, Science, 299, 2005-2010. Anand, P., H. Elderfield, 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. Andrews, J.T., 1998, Abrupt changes (Heinrich events) in late Quaternary North Atlantic marine environments: a history and review of data and concepts, Journal of Quaternary Science, 13(1), 3-16. Baker, P.A., Rigsby, C.A., Seltzer, G.O., Fritz, S.C., Lowenstein, T.K., Bacher, N.P., and Veliz, C., 2001, Tropical climate changes at millennial and orbital timescales on t he Bolivian Altiplano, Nature, 409, 698-701. Barber, D.C., Dyke, A., Hillaire-Marcel, C., Jennings, A.E., Andrews, J.T., Kerwin, M.W., Bilodeau, G., McNeely, R., Southon, J., Morehead, M.D., and Gagnon, J.M., 1999, Forcing of the cold event of 8,200 years ago by catastrophic drainage of Laurentide lakes, Nature, 400, 344-348.
73 Bard, E., Rostek, F., and Sonzogni, C., 1997, Interhemispheric synchrony of the last deglaciation inferred from alkenone palaeothermometry, Nature, 385, 707-710. Barker, S., M. Greaves, and H. Elderfield, 2003, A study of cleaning procedures used for foraminiferal Mg/Ca paleothermometry, Geochemistry, Geophysics, Geosyst ems, 4, doi: 10.1029/2003GC000559. Barry, J.M., 1997, Rising Tide-The Great Mississippi Flood of 1927 and How it Changed America Simon and Schuster, New York, 524 pp. Be A.W.H., 1982, Biology of Planktonic Foraminifera, in Broadhead, T. W. (ed.), Foraminifera: Notes for a short course, Studies in Geology: University of Tenness e, 6, 51-92. Be A.W.H., and Hamlin, W.H., 1967, Ecology of recent planktonic foraminifera. Part 3. Distribution in the North Atlantic during the summer of 1962, Micropaleontology, 6(4), 87-106. Beers, Y., 1957, Introduction to the Theory of Error Addison-Wesley-Longman, Rading, Mass, 66 pp. Bemis, B.E., J.H. Spero, J. Bijma, and D.W. Lea, 1998, Reevaluation of the oxygen isotopic composition of planktonic foraminifera: Experimental results and revised paleotemperature equations, Paleoceanography, 13, 150-160. Benn, D.I., and Evans, D.J.A., 1998, Glaciers and Glaciation Arnold Hodder Headline Group, UK, 734 pp. Blunier, T., and E.J. Brook, 2001, Timing of millennial-scale climate change in Antarctica and Greenland during the last glacial period, Science, 291, 109-112. Bond, G., Heinrich, H., Broecker, W., Labeyrie, L., McManus, J., Andrews, J., Huon, S., Jantschik, R., Clasen, S., Simet, C., Tedesco, K., Kias, M., Bonani, G., and Ivy, S., 1992, Evidence for massive discharges of icebergs into the North Atlantic ocean during the last glacial period, Nature, 360, 245-249. Bond, G., Broecker, W., Johnsen, S., McManus, J., Labeyrie, L., Jouzel, J., and Bonani, G., 1993, Correlations between climate records from North Atlantic sediments and Greenland ice, Nature, 365, 143-147. Bond, G., and Lotti, R., 1995, Iceberg discharges into the North Atlantic on millennial time scales during the last glaciation, Science, 267, 1005-1010.
74 Bond, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., deMenocal, P., Priore, P., Cullen, H., Hajdas, I., Bonani, G., 1997, A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates, Science 278, 1257-1266. Bond, G.C., Showers, M., Elliot, M., Evans, M., Lotti, R., Hajdas, I., Bonani, G., and Johnson, S., 1999, The North Atlantics 1-2 kyr climate rhythm: relation to Heinrich events, Dansgaard/Oeschger cycles and the Little Ice Age, in Mechanisms of Global Climate Change at Millennial Time Scales edited by P.U. Clark, R.S. Webb, and L.D. Keigwin, AGU Washington, D.C., pp. 35-58 Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffman, S., Lotti-Bond, R., Hajdas, I., and Bonani, G., 2001, Persistent solar influence on North Atlantic climate during the Holocene, Science, 294, 2130-2136. Bowen, G.J., and J. Revenaugh, 2003, Interpolating the isotopic composition of modern meteoric precipitation, Water Resources Research, 39, doi: 10.1029/2003WR002086. Bradbury, J.P., 2000, Limnologic history of Lago de Patzcuaro, Michoacan, Mexico for the past 48,000 years: impacts of climate and man: Palaeogeography, Palaeoclimatology, Palaeoecology, 163, 69-95. Briggs, J.C., and J.F. Ficke, 1978, Quality of rivers of the United States, 1975 water year, Open File Rep. 78-200, U.S. Geological Survey, 436 pp. Broecker, W.S., M. Andree, W. Wolfli, H. Oeschger, G. Bonani, J. Kennett, and D. Peteet, 1988, The chronology of the last deglaciation: Implications to the cause of the Younger Dryas event, Paleoceanography, 3, 1-19. Broecker, W.S., J.P. Kennett, B.P. Flower, J.T. Teller, S. Trumbore, G. Bonani, and W. Wolfli, 1989, Routing of meltwater from the Laurentide ice sheet during the Younger Dryas cold episode, Nature, 341, 318-321. Broecker, W.S., Bond, G., Klas, M., Bonani, G., and Wolfi, W., 1990, A salt oscillator in the glacial North Atlantic? 1. The concept, Paleoceanography, 5, 469-477. Broecker, W.S., 1998, Paleocean circulation during the last deglaciation: A bipolar seesaw?, Paleoceanography, 13(2), 119-121. Brook, E.J., Harder, S., Severinghaus, J., and Bender, M., 1999, in Mechanisms of Global Climate Change at Millennial Time Scales v. 112, edited by P.U. Clark, R.S. Webb, and L.D. Keigwin, AGU Washington, D.C., pp. 165-176 Brown, S. and Elderfield, H., 1996, Variations in Mg/Ca and Sr/Ca ratios of planktonic foraminifera caused by postdepositional dissolution-evidence of shallow Mg-dependent dissolution, Paleoceanography, 11(5), 543-551.
75 Brown, P.A., and Kennett, J.P., 1998, Megaflood erosion and meltwater plumbing changes during last North American deglaciation recorded in Gulf of Mexico sediments, Geology, 26, 599-602. Brunner, C.A., and Cooley, J.F., 1976, Circulation in the Gulf of Mexico during the Last Glacial Maximum 18,000 years ago, Geological Society of America Bulletin, 87(5), 681-686. Bryden, H.L., Longworth, H.R., and Cunningham, S.A., 2005, Slowing of the Atlantic meridional overturning circulation at 25rN, Nature, 438, 655-657. Caballero, M., Lozano, S., Ortega, B., Urrutia, J., and Macias, J.L., 1999, Environmental characteristics of Lake Tecocomulco, northern basin of Mexico, for the last 50,000 years, Journal of Paleolimnology, 22, 399-411. Cane, M.A., 1998, A role for the tropical pacific, Science, 282, 59-61. Cane, M., and Clement, A., 1999, A role of the tropical Pacific coupled oceanatmosphere system on Milankovitch and millennial time-scales. Part II: global impacts, in Mechanisms of Global Climate Change at Millennial Time Scales v. 112, edited by P.U. Clark, R.S. Webb, and L.D. Keigwin, AGU Washington, D.C., pp. 373-383 Chappell, J., 2002, Sea level changes forced ice breakouts in the Last Glacial cycle: ne w results from coral terraces, Quaternary Science Reviews, 21(10), 1229-1240. Charles, C.S., D. Rind, R. Healy, and R. Webb, 2001, Tropical cooling and the isotopic composition of precipitation in general circulation model simulations of the ice age climate, Climate Dynamics, 17(7), 489-502. Clark, P.U., S.J. Marshall, G.K.C., Clarke, S.W. Hostetler, J.M. Licciardi, and J.T. Telle r, 2001, Freshwater forcing of abrupt climate change during the last glaciation, Scienc e, 293, 283-287. Clement, A.C., Hall, A., and Broccoli, A.J., 2004, The importance of precessional signals in the tropical climate, Climate Dynamics, 22, 327-341. Crowley, T.J., 1992, North Atlantic Deep Water cools the Southern Hemisphere, Paleoceanography, 4, 489-499. Curry, R., Dickson, B., and Yashayaev, I., 2003, A change in the freshwater balance of the Atlantic Ocean over the past four decades, Nature, 426, 826-829.
76 Curry, W.B., T.M. Marchitto, J.F. McManus, D.W. Oppo, and K.L. Laarkamp, 1999, Millennial-scale changes in ventilation of the thermocline, intermediate, and dee p water of the glacial North Atlantic, in Mechanisms of Global Climate Change at Millennial Time Scales v. 112, edited by P.U. Clark, R.S. Webb, and L.D. Keigwin, AGU Washington, D.C., pp. 59-76 Dansgaard, W., and H. Tauber, 1969, Glacier Oxygen-18 Content and Pleistocene Ocean Temperatures, Science, 166, 499-502. Dansgaard, W., et al., 1993, Evidence for general instability of past climate from a 250kyr ice-core record, Nature, 364, 218-220. Darling, K.F., Wade, C.M., Kroon, D., and Brown, A.J.L, 1997, Planktic foraminiferal molecular evolution and their polyphyletic origins from benthic taxa, Marine Micropaleontology, 30, 251-266. Dekens, P.S., Lea, D.W., Pak, D.K., and Spero, H.J., 2002, Core top calibration of Mg/Ca in tropical foraminifera: Refining paleotemperature estimation, Geochemis try, Geophysics, Geosystems, 3(4), doi: 10.1029/2001GC000200. Denton, G.H., R.B. Alley, G.C. Comer, and W.S. Broecker, 2005, The role of seasonality in abrupt climate change, Quaternary Science Reviews, 24, 1159-1182. Deuser, W.G., 1987, Seasonal variations in isotopic composition and deep-water fluxes of the tests of perennially abundant planktonic foraminifera of the Sargasso Sea: Results from sediment-trap collections and their paleoceanographic significa nce, Journal of Foraminiferal Research, 17, 14-27. Deuser, W.G., and Ross, E.H., 1989, Seasonally abundant planktonic foraminifera of the Sargasso Sea: Succession, deep-water fluxes, isotopic compositions, and paleoceanographic implications, Journal of Foraminiferal Research, 19, 268-293. Dickson, B., Yashayaev, I., Meincke, J., Turrell, B., Dye, S., and Hoffort, J., 2002, Rapid freshening of the deep North Atlantic Ocean over the past four decades, Nature, 416, 832-837. Dinnel, S.P., and Wiseman, W.J., 1986, Fresh water on the Louisiana and Texas shelf, Continental Shelf Research, 6(6), 765-784. Dorale, J.A., R.L. Edwards, E. Ito, and L.A. Gonzalez, 1998, Climate and vegetation history of the mid-continent from 75 to 25 ka: A speleothem record from Crevice Cave, Missouri, USA, Science, 282, 1871-1874.
77 Dyke, A.S., J.T. Andrews, P.U. Clark, J.H. England, G.H. Miller, J. Shaw, and J.J. Veillette, 2002, The Laurentide and Innuitian ice sheets during the Last Glacial Maximum, Quaternary Science Reviews, 21, 9-31. Elderfield, H., and Ganssen, G., 2000, Past temperature and d 18 O of surface ocean waters inferred from foraminiferal Mg/Ca ratios, Nature, 405, 442-445. Fairbanks, R.G., and Weibe, P.H., 1980, Foraminifera and chlorophyll maximum: Vertical distribution, seasonal succession and paleoceanographic significance, Science, 209, 61-63. Fairbanks, R.G., Sverdlove, M., Free, R., Weibe, P.H., and Be A.W.H., 1982, Vertical distribution and isotopic fractionation of living planktonic foraminifera from Panama Basin, Nature, 298, 841-844. Flower, B.P., and J.P. Kennett, 1990, The Younger Dryas cool episode in the Gulf of Mexico, Paleoceanography, 5(6), 949-961. Flower, B.P., D.W. Hastings, H.W., Hill, and T.M. Quinn, 2004, Phasing of deglacial warming and Laurentide Ice Sheet meltwater in the Gulf of Mexico, Geology, 32, 597-600. Forman, S.L., Oglesby, R., Markgraf, V., and Stafford, T., 1995, Paleoclimatic significance of Late Quaternary eolian deposition on the Piedmont and High Plains, Central United States, Global and Planetary Change, 11, 35-55. Ganopolski, A., and S. Rahmstorf, 2001, Rapid changes of glacial climate simulated in a coupled climate model, Nature, 409, 153-158. Goni, M.A., Ruttenberg, K.C., and Eglinton, T., 1997, Sources and contribution of terrigenous organic carbon to surface sediments in the Gulf of Mexico, Nature, 389, 275-278. Grootes, P.M., M. Stuiver, J.W.C. White, S. Johnsen, and J. Jouzel, 1993, Comparison of oxygen-isotope records from the GISP2 and GRIP Greenland ice cores, Nature, 366, 552-554. Guilderson, T.P., Fairbanks, R.G., and Rubenstone, J.L., 1994, Tropical temperature variations since 20,000 years ago: Modulating interhemispheric climate change, Science, 263, 663-665. Guilderson, T.P., Fairbanks, R.G., and Rubenston, J.L., 2001, Tropical Atlantic coral oxygen isotopes: glacial-interglacial temperatures and climate change Marine Geology, 172, 75-89.
78 Hagen, S., and L.D. Keigwin, 2002, Sea-surface temperature variability and deep water reorganization in the subtropical North Atlantic during Isotope Stage 2-4, Marine Geology, 189, 145-162. Hansen, B., Turrell, W.R., and Osterhus, S., 2001, Decreasing outflow from the Nordic seas into the Atlantic Ocean through the Faroe Bank channel since 1950, Nature, 411, 927-930. Hastenrath, S., 1984, Interannual variability and annual cycle-mechanisms of circulat ion and climate in the tropical Atlantic sector, Monthly Weather Review, 112(6), 1097-1107. Hastings, D.W., Emerson, S.R., and Nelson, B.K., 1996, Vanadium in foraminifera calcite: evaluation of a method to determine paleo-seawater concentrations, Geochmica et Cosmochimica Acta, 60(19), 3701-3715. Hastings, D.W., Russell, A.D., and Emerson, S.R., 1998, Foraminiferal magnesium in Globeriginoides sacculifer as a paleotemperature proxy, Paleoceanography, 13(2), 161-160. Haug, G.H., Hughen, K.A., Sigman, D.M., Peterson, L.C., Rohl, U., 2001, Southward migration of the Intertropical Convergence Zone through the Holocene, Science, 293, 1304-1308. Heinrich, H., 1988, Origin and consequences of cyclic ice-rafting in the northeast Atlantic Ocean during the past 130,000 years, Quaternary Research, 29, 142-152. Hemleben, Ch., M. Spindler, and O.R. Anderson (1989), Modern Planktonic Foraminifera 363 pp., Springer-Verlag, New York, 363 pp. Hill, H.W., Flower, B.P., Quinn, T.M., Hollander, D.J., Guilderson, T.P., 2006, Laurentide Ice Sheet meltwater and abrupt climate change during the last glac iation: Paleoceanography, 21, doi:10.1029/2005PA001186. Hillesheim, M.B., Hodell, D.A., Leyden, B.W., Brenner, M., Curtis, J.H., Anselmetti, F.S., Ariztegui, D., Buck, D.G., Guilderson, T.P., Rosenmeier, M.F., Schnurrenberger, D.W., 2005, Climate change in lowland Central America during the late deglacial and early Holocene, Journal of Quaternary Science, 20, 363-376. Hirschboeck, K.K., 1991, Climate and floods: National Water Summary 1988-1989Hydrologic events and floods and droughts, United States Geological Survey Water Supply Paper 2375, 67-88.
79 Hodell, D.A., Curtis, J.H., Jones, G.A., Higuera-Gundy, A., Brenner, M., Binford, M.W., and Dorsey, K.T., 1991, Reconstruction of Caribbean climate change over the past 10,500 years, Nature, 352, 790-793. Hu, A.X., and Meehl, G.A., 2005, Reasons for a fresher northern North Atlantic in the late 20 th century, Geophysical Research Letters, 32(11), L11701 Hughen, K., S. Lehman, J. Southon, J. Overpeck, O. Marchal, C. Herring, and J. Turnbull, 2004, 14 C Activity and Global Carbon Cycle Changes over the Past 50,000 Years, Science, 303, 202-207. Huls, M., and Zahn, R., 2000, Millennial-scale sea surface temperature variability in t he western tropical North Atlantic from planktonic foraminiferal census counts, Paleoceanography, 15(6), 659-678. Jasper, J.P., and Gagosian, R.B., 1993, The relationship between sedimentary organic carbon isotopic composition and organic biomarker compound concentration: Geochimica et Cosmochimica Acta, 57, 167-186. Johnsen, S.J., H.B. Clausen, W. Dansgaard, and C.C. Langway, 1972, Oxygen isotope profiles through Antarctic and Greenland ice sheets, Nature, 235, 429-434. Johnsen, S.J., et al., 2001, Oxygen isotope and palaeotemperature records from six Greenland ice-core stations: Camp Century, Dye-3, GRIP, GISP2, Renland and NorthGRIP, Journal of Quaternary Science, 16(4), 299-307. Kemle-von Mcke, S., and Oberhnsli, H., 1999, The distribution of living planktic foraminifera in relation to southeast Atlantic oceanography, in Use of Proxies in Paleoceanography: Examples for the South Atlantic edited by G. Fischer, G., and G. Wefer, Springer-Verlag, Berlin, pp. 91-115. Kienast, M., Steinke, S., Stattegger, K., Calvert, S.E., 2001, Synchronous tropical South Chna Sea SST change and Greenland warming during deglaciation, Science, 291, 2132-2134. Kienast, M., Hanebuth, T.J., Pelejero, C., 2003, Synchroneity of meltwater pulse 1a and the Bolling warming: New evidence from the South China Sea, Geology, 31(1), 67-70. Knox, J.C., 2000, Sensitivity of modern and Holocene floods to climate change, Quaternary Science Reviews, 19, 439-457. Knutti, R., J. Fluckiger, T.F. Stocker, and A. Timmermann, 2004, Strong hemispheric coupling of glacial climate through freshwater discharge and ocean circulation, Nature, 430, 851-856.
80 Koutavas, A., Lynch-Stieglitz, J., Marchitto, T.M., and Sachs, J.P., 2002, El Nino-like pattern in ice age tropical Pacific sea surface temperature, Science, 297, 226-230. Kroopnick, P., 1974, The dissolved O 2 -CO 2 13 C system in the eastern equatorial Pacific, Deep Sea Research, 21, 211-227. Kutzbach, J.E., 1981, Monsoon climate of the early Holocen-climate experiment with the Earths orbital parameters for 9000 years ago, Science, 214, 59-61. Kutzbach, J.E., and Guetter, P.J., 1986, The Influence of changing orbital parameters and surface boundary conditions on climate simulations for the past 18,000 years, Journal of the Atmospheric Sciences, 43(16), 1726-1759. Kutzbach, J.E., 1987, Model simulations of the clmatic patterns during the deglaciation of North America, in The Geology of North America v. K-3, North America and Adjacent Oceans During the Last Deglaciation, edited by W.G. Ruddiman and H.E. Wright, Decade of North American Geology, pp. 425-226. Laj, C., C. Kissel, A. Mazaud, J.E.T. Channell, and J. Beer, 2000, North Atlantic Paleointensity Stack since 75 ka (NAPIS-75) and the duration of the Laschamp event, Philos. Transact. Ser. A Math Phys. Eng. Sci., 358, 1009-1025. Lang, C., Leuenberger, M., Schwander, J., and Johnsen, S., 1999, 16 degrees C rapid temperature variations in Central Greenland 70,000 years ago, Science 286, 934-937. Lea, D.W., Mashiotta, T.A., and Spero, H.J., 1999, Controls on magnesium and strontium uptake in planktonic foraminifera determined by live culturing, Geochimica et Cosmochimica Acta, 63(16), 2369-2379. Lea, D.W., Pak, D.K., and Spero, H.J., 2000, Climate impact of Late Quaternary equatorial Pacific sea surface temperature variations, Science, 289, 1719-1724. Lea, D.W., Pak, D.K., Peterson, L.C., and Hughen, K.A., 2003, Synchroneity of tropical and high-latitude Atlantic temperatures over the last glacial termination, S cience, 301, 1361-1364. Leventer, A., Williams, D.F., and Kennett, J.P., 1982, Dynamics of the Laurentide Ice Sheet during the last deglaciation: Evidence from the Gulf of Mexico, Earth and Planetary Science Letters, 59, 11-17. Leventer, A., Williams, D.F., and Kennett, J.P., 1983, Relationships between anoxia, glacial meltwater and microfossil preservation in the Orca Basin, Gulf of Me xico, Maine Geology, 53(1-2), 23-40.
81 Levitus, S., Antonov, J.I., Boyer, T.P., and Stephens, C., 2000, Warming of the world ocean, Science 287, 2225-2229. Levitus, S., 2003, National Oceanographic Data Center World Ocean Atlas 1994 at http://www.cdc.noaa.gov/ (NCDC, Boulder, Colorado, USA). Licciardi, J.M., Teller, J.T., and Clark, P.U., 1999, Freshwater routing by the Laurentide Ice Sheet during the Last Deglaciation, in Mechanisms of Global Climate Change at Millennial Time Scales v. 112, edited by P.U. Clark, R.S. Webb, and L.D. Keigwin, AGU Washington, D.C., pp. 177-201. LoDico, J.M., D.W. Hastings, B.P. Flower, and T.M. Quinn, 2002, A multi-proxy approach to distinguish between changes in SST and meltwater input in the Gulf of Mexico back to MIS 3, Eos Trans AGU 83 (47), Fall Meet. Suppl., Abstract PP62A0331. LoDico, J.M., Flower, B.P., and Quinn, T.M., in review, Sub-centennial scale climatic and hydrologic variability in the Gulf of Mexico during the early Holocene, Paleoceanography Licciardi, J.M., Teller, J.T., and Clark, P.U., 1999, Freshwater routing by the Laurentide Ice Sheet during the last deglaciation, in Mechanisms of Global Climate Change at Millennial Time Scales v. 112, edited by P.U. Clark, R.S. Webb, and L.D. Keigwin, AGU Washington, D.C., pp. 177-201. Lohrenz, S.E., Dagg, M.J, and Whitledge, T.E., 1990, Enhanced primary production at the plume/oceanic interface of the Mississippi River, Continental Shelf Resea rch, 10, p. 639-664. MacAyeal, D.R., 1993, Binge/purge oscillations of the Laurentide Ice Sheet as a cause of North Atlantics Heinrich events, Paleoceanography, 8, 775-784. Machel, H., Kapala, A., and Flohn, H., 1998, Behaviour of the centers of action above the Atlantic since 1881. Part I: Characteristics of seasonal and interannual variabi lity: International Journal of Climatology, 18, 1-22. Manabe, S., and Stouffer, R.J., 1997, Coupled ocean-atmosphere model response to freshwater input: Comparison to Younger Dryas event, Paleoceanography, 12(2), 321-336. Marshall, S.J., and Clarke, G.K.C., 1999, Modeling North American freshwater runoff through the Last Glacial Cycle, Quaternary Research, 52, 300-315.
82 Martinson, D.G., Pisias, N.G., Hays, J.D., Imbrie, J., Moore, T.C., and Shackleton, N.J., 1987, Age dating and the orbital theory of the ice ages-development of a high-resolution 0 to 300,000-year chronostratigraphy, Quaternary Research, 27(1), 1-29. Meese, D.A., A.J. Gow, R.B. Alley, G.A. Zielinski, P.M. Grootes, M. Ram, K.C. Taylor, P.A. Mayewski, and J.F. Bolzan, 1997, The Greenland Ice Sheet Project 2 depth-age scale: Methods and results, Journal of Geophysical Research, 102, 26411-26423. Mo, K.C., Paegle, J.N., and Higgins, R.W., 1997, Atmospheric processes associated with summer floods and droughts in the central United States, Journal of Climate, 10, 3028-3046. Mulitza, S., Arz, H.W., Kemle-von Mcke, Moos, C., Niebler, H.S., Ptzold, J., and Segl, M., 1999, The South Atlantic carbon isotope record of planktic foraminifera, in Use of Proxies in Paleoceanography: Examples for the South Atlantic edited by G. Fischer, G., and G. Wefer, Springer-Verlag, Berlin, pp. 427-445. National Research Council, 2002, Abrupt Climate Change: Inevitable Surprise, US National Academy of Science, Natioanl Research Council Committee on abrupt climate change, National Academy Press, 238 p. Nurnberg, D., Bijma, J., and Hemleben, C., 1996, Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperatures, Geochim ica et Cosmochimica Acta, 60(5), 803-814. Ohkouchi, N., Kawamura, K., and Taira, A., 1997, Molecular paleoclimatology: reconstruction of climate variabilities in the late Quaternary, Organic Ge ochemistry, 27(3-4), 173-183. Ortiz, J.D., Mix, A.C., Rugh, W., Watkins, J.M., and Collier, R.W., 1996, Deep-dwelling planktonic foraminifera of the northeastern Pacific Ocean reveal environmental control of oxygen and carbon isotopic disequilibria, Geochimica Cosmochimica Acta, 60, 4509-4523. Overpeck, J.T., 1996, Warm climate surprises, Science, 271, 1820-1821. Overpeck, J., and Webb, R., 2000, Nonglacial rapid climate events: Past and future, Proceedings of the National Academy of Science, U.S.A., 97(4), 1335-1338. Pancost, R.D., and Boot, C.S., 2004, The paleoclimatic utility of terrestrial biomarkers i n marine sediments, Marine Chemistry, 92, 239-261. Peteet, D.M., 1995, Global Younger Dryas?, Quaternary International, 28, 93-104.
83 Peterson, L.C., Haug, G.H., Hughen, K.A., and Rohl, U., 2000, Rapid changes in the hydrologic cycle of the tropical Atlantic during the last glacial, Science, 290, 1947-1951. Peterson, B.J., Holmes, R.M., McClelland, J.W., Vorosmarty, C.J., Lammers, R.B., Shiklomanov, A.I., Shiklomanov, I.A., and Rahmstorf, S., 2002, Increasing river discharge to the Arctic Ocean, Science, 298, 2171-2173. Petit, J.R., et al., 1999, Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica, Nature, 399, 429-436. Poore, R.Z., Darling, J., Dowsett, H.J., and Wright, L., 2001, Variations in river flow to the Gulf of Mexico: Implications for paleoenvironmental studies of Gulf of Mexico marine sediments: U.S. Geological Survey Bulletin 2187, http://pubs.usgs.gov/bulletin/b2187/. Poore, R.Z., Dowsett, H.J., Verardo, S., and Quinn, T.M., 2003, Millennialto centuryscale variability in Gulf of Mexico Holocene climate records, Paleoceanogra phy, 18(2), doi:10.1029/2002PA000868 Rabalais, N.N., Turner, R.E., Wiseman Jr., W.J., and Dortch, Q., 1998, Consequences of the 1993 Mississippi River Flood in the Gulf of Mexico, Regulated Rivers: Research and Management, 14, 161-177. Rahmstorf, S., 1994, Rapid climate transitions in a coupled ocean-atmosphere model, Nature, 372, 82-85. Rahmstorf, S., 1995, Bifurcations of the Atlantic thermohaline circulation in response to changes in the hydrologic cycle, Nature, 378, 145-149. Rahmstorf, S., 2002, Ocean circulation and climate during the past 120,000 years, Nature, 419, 207-214. Redalje, D.G., Lohrenz, S.E., and Fahnenstiel, 1994, The relationship between primary production and the vertical export of particulate organic matter in a river-impact ed coastal ecosystem, Estuaries, 17(4), 829-838. Robinson, C., G.M. Raisbeck, F. Yiou, B. Lehman, and C. Laj, 1995, The relationship between 10 Be and geomagnetic field strength records in central North Atlantic sediments during the last 80 ka, Earth Planetary Science Letters, 136, 551-557. Rohling, E.J., R. Marsh, N.C. Wells, M. Siddall, and N.R. Edwards, 2004, Similar meltwater contributions to glacial sea level changes from Antarctic and northe rn ice sheets, Nature, 430, 1016-1021.
84 Rohling, E.J., Sprovieri, M., Cane, T., Casford, J.S.L., Cooke, S., Bouloubassi, I., Emeis, K.C., Schiebel, R., Rogerson, M., Hayes, A., Jorissen, F.J., and Kroon, D., 2004, Reconstructing past planktic foraminiferal habitats using stable isotope data: a case history for Mediterranea sapropel S5, Marine Micopaleontology, 50, 89-123. Romanek, C.S., Grossman, E.L, and Morse, J.W., 1992, Carbon isotopica fractionation in synthetic aragonite and calcite-effects of temperature and precipitation r ate, 1992, Geochimica et Cosmochimica Acta, 56(1), 419-430. Rooth, C.G.H., 1990, Hydrology and ocean circulation, Progress Oceanography, 11, 131149. Ropelewski, C.F., and M.S. Halpert, 1996, Quantifying Southern OscillationPrecipitation relationships, Journal of Climate, 9, 1043-1059. Rosenthal, Y. and Boyle, E.A., 1993, Factors controlling the fluoride content of planktonic foraminifera: an evaluation of its paleoceanographic utility, Geochimic a et Cosmochimica Acta, 57(2), 335-346. Ruhlemann, C., Mulitza, S., Muller, P.J., Wefer, G., and Zahn, R., 1999, Warming of the tropical Atlantic Ocean and slowdown of thermohaline circulation during the last deglaciation, Nature, 402, 511-514. Russell, A.D., Emerson, S., Nelson, B.K., Erez, J., and Lea, D.W., 1994, Uranium in foraminiferal calcite as a recorder of seawater uranium concentrations, Ge ochimica et Cosmochimica Acta, 58(2), 671-681. Sachs, J.P., and Lehman, S.J., 1999, Subtropical North Atlantic temperatures 60,000 to 30,000 years ago, Science, 286, 756-759. Schmuker, B., and Schiebel, R., 2002, Planktic foraminifers and hydrography of the eastern and northern Caribbean Sea, Marine Micropaleontology, 46, 387-403. Shackleton, N.J., 1987, Oxygen-isotopes, ice volume and sea level, Quaternary Science Reviews, 6, 183-190. Shackleton, N.J., 1989, Deep trouble for climate change, Nature, 342, 616-617. Shackleton, N.J., M.A. Hall, and E. Vincent, 2000, Phase relationships between millennial-scale events 64,000-24,000 years ago, Paleoceanography, 15(6), 565-569. Shackleton, N.J., R.G. Fairbanks, T. Chiu, and F. Parrenin, 2004, Absolute calibration of the Greenland time scale: implications for Antarctic time scales and for D 14 C, Quaternary Science Reviews, 23, 1513-1522.
85 Siddall, M., E.J. Rohling, A. Almogi-Labin, C. Hemleben, D. Meischner, I. Schmelzer, and D.A. Smeed, 2003, Sea-level fluctuations during the last glacial cycle, Nature, 423, 853-858. Spero, H.J., Bijma, J., Lea, D.W., and Russell, A.D., 1999, Deconvolving glacial ocean carbonate chemistry from the planktonic foraminifera carbon isotope record, in Reconstructing Ocean History: A Window Into the Future edited by F. Abrantes and A. Mix, p. 329-342, Plenum, New York, pp. 329-342. Spero, H.J., Mielke, K.M., Kalve, E.M., Lea, D.W., and Pak, D.K., 2003, Multispecies approach to reconstructing eastern equatorial Pacific thermocline hydrography duri ng the past 360 kyr, Paleoceanography, 18(1), doi: 10.1029/2002PA000814 Stott, L., Poulsen, C., Lund, S., and Thunell, R., 2002, Super ENSO and global climate oscillations at millennial time scales, Science, 297, 222-226. Tedesco, K.A., and Thunell, R.C., 2003, Seasonal and interannual variations in planktonic foraminiferal flux and assemblage composition in the Cariaco Basin, Venezuela, Journal of Foraminiferal Research, 33(3), 192-210. Thompson, P.R., Be A.W.H., Duplessy, J.C., and Shackleton, N.J., 1979, Disappearance of pink-pigmented Globigerinoides ruber at 120,000 yr BP in the Indian and Pacific Oceans, Nature, 280, 554-558. 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, 843-846. Tolderlund, D.S., and Be A.W.H., 1971, Seasonal distribution of planktonic foraminifera in the western North Atlantic, Micropaleontology, 17, 297-329. Trenberth, K.E., and Guillemot, C.J., 1996, Physical processes involved in the 1988 drought and the 1993 floods in North America, Journal of Climate, 9, 1288-1298. VanBennekom, A.J., and Berger, W.H., 1984, Hydrography and silica budget of the Angola Basin, Netherland Journal of Sea Research, 17, 149-200. Vautravers, M.J., N.J. Shackleton, C. Lopez-Martinez, and J.O. Grimalt, 2004, Gulf Stream variability during marine isotope stage 3, Paleoceaonography, 19, doi: 10.1029/2003PA000966. Vellinga, M., and Wood, R.A., 2002, Global climatic impacts of a collapse of the Atlantic thermohaline circulation, Climatic Change, 54(3), 251-267.
86 Voelker, A.H.L. et al., 2002, Global distribution of centennial-scale records for marine isotope stage (MIS) 3: a database, Quaternary Science Reviews, 21, 1185-1214. Wahl, K.L., Vining, K.C., and Wiche, G.J., 1993, Precipitation in the upper Mississippi River basin, January 1 through July 31, 1993: United States Geological Survey Circular 1120-B, 13 pp. Wang, Y.J., Cheng, H., Edwards, R.L., An, Z.S., Wu, J.Y., Shen, C.C., and Dorale, J.A., 2001, A high-resolution absolute-dated Late Pleistocene monsoon record from Hulu Cave, China, Science, 294, 2345-2348. Wang, X., Auler, A.S., Edwards, R.L., Cheng, H., Cristalli, P.S., Smart, P.L., Richards, D.A., and Shen, C.C., 2004, Wet periods in northeastern Brazil over the past 210 kyr linked to distant climate anomalies, Nature, 432, 740-743. Webb, T., Bartlein, P.J., Harrison, S.P., and Anderson, K.H., 1993, Vegetation, lake levels, and climate in Eastern North America for the past 18,000 years, in Global Climate since the Last Glacial Maximum edited by H.E. Wright, J.E. Kutzbach, T. Webb, W.F. Ruddiman, F.A. Street-Perrott, P.J. Bartlein, P.J., University of Minnesota Press, pp. 415-467. Yapp, C.J., and S. Epstein, 1977, Climatic implications of D/H ratios of meteoric water over North America (9,500-22,000 B.P.) as inferred from ancient wood cellulose C-H hydrology, Earth Planetary Science Letters, 34, 333-350. Yiou, F.G., et al., 1997, Beryllium 10 in the Greenland Ice Core Project ice core at Summit, Greenland, Journal of Geophysical Research, 103, 26783-26794. aric S., Donner, B., Fischer, G., Mulitza, S., and Wefer, G., 2005, Sensitivity of planktic foraminifera to sea surface temperature and export production as derived from sediment trap data, Marine Micropaleontology, 55, 75-105. Zhao, M., Beveridge, N.A.S., Shackleton, N.J., Sarnthein, M., and Eglinton, G., 1995, Molecular stratigraphy of cores off northwest Africa: Sea surface temper ature history over the last 80 ka, Paleoceanography, 10(3), 661-675.
87 Appendix A: Inorganic analyses
88 Appendix A (Continued) Abbreviations used in Appendix A tables: depth: depth in the core (cm) calyr: calendar years before present d 13 C: d 13 C ( PDB) of foraminiferal calcite d 18 O c : d 18 O ( PDB) of foraminiferal calcite SST: Mg/Ca converted to SST (rC) using Anand et al. (2003) d 18 O sw : d 18 O of seawater ( VSMOW) calculated using Bemis et al. (1998) weight/foram: weight ( g) of forams per number of individual tests % coarse fraction: >63 m component of bulk sediment pink: pink G. ruber white: white G. ruber calyr GOM: calendar years before present for d18OGOM (resampled) d 18 O GOM : d 18 O of seawater corrected for ice volume ( VSMOW) precip endmember Salinity calculations with the use of a precipitation endmember (3.5) in the Gulf of Mexico mixing model Agassiz endmember Salinity calculations with the use of a Lake Agassiz endm ember (15) in the Gulf of Mexico mixing model LIS endmember Salinity calculations with the use of a LIS endmember (-30) in the Gulf of Mexico mixing model
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2100279530.90-1.023.6425.111.3816.070.132102279781.04-0.393.5324.751.9316.260.202104280020.95-1.123.5024.661.1816.910.152106280261.29-0.883.2423.811.2416.770.172108280511.34-0.933.7025.281.5118.290.132110280751.23-1.223.6125.031.1618.110.152112281001.05-0.473.6525.131.9316.500.162114281241.28-1.213.2923.990.9617.400.162116281481.37-0.833.6225.061.5516.710.122118281731.36-1.043.7425.401.4218.770.132120281971.40-0.953.4424.491.3119.660.222122282211.41-0.973.6125.011.4117.630.24 2123.5282401.51-1.033.8325.671.4917.510.20 2125282581.10-1.203.8925.861.3515.340.202127282731.09-1.333.2623.860.8014.170.18 2128.5282841.34-1.163.3424.141.0315.170.13 2130282951.00-1.172.8822.490.6913.690.082132283101.22-1.223.4724.561.0614.700.102134283241.29-1.103.3624.221.1215.240.072136283391.01-1.003.4924.631.3016.000.062138283540.90-1.503.7025.300.9415.560.062140283681.12-1.473.7525.451.0014.800.052142283831.01-1.203.5224.751.1315.910.032144283980.97-1.7815.710.022146284131.18-1.123.5824.941.2415.070.022148284271.22-0.763.5024.671.5415.530.022150284421.27-0.913.4324.451.3515.790.042152284571.40-1.173.4924.651.1315.950.102154284721.18-1.023.3124.051.1614.970.052156284861.15-1.3214.200.022158285011.34-1.2917.830.022160285161.23-1.153.4624.561.1316.400.122162285301.20-1.113.8825.801.4314.510.132164285450.97-1.643.6725.210.7814.500.062166285601.29-1.143.7925.551.3514.400.022168285750.97-1.163.5824.921.2016.910.162170285893.7425.4215.580.272172286041.09-1.713.5424.780.6215.490.10 2173.5286150.88-1.123.3924.301.1114.690.08 2175286261.18-1.953.6325.080.4415.910.15217728641 2178.5286521.01-1.643.4224.420.6115.350.12 2180286631.02-1.743.6525.140.6613.570.16 Table 3. Pink G. ruber d 18 O and Mg/Ca analyses. 89
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2182286780.75-1.343.5924.951.0215.140.112184286921.20-1.463.6425.110.9417.460.132186287070.99-1.863.4624.540.4216.140.112188287220.80-1.673.4724.570.6216.260.112190287371.03-1.383.4624.550.9016.920.162192287510.85-1.613.7025.290.8215.030.082194287660.95-1.423.8325.671.1015.230.052196287811.07-1.873.5124.720.4515.460.032198287960.77-1.243.4724.571.0514.740.032200288100.77-1.623.3624.210.5916.620.072202288250.94-2.123.6125.010.2515.230.132204288311.13-1.783.6325.070.6117.600.102206288370.68-1.723.1923.640.3715.630.132208288440.69-2.003.4824.610.2915.620.122210288500.87-1.623.3224.090.5615.280.062212288560.70-1.263.4724.591.0312.940.052214288620.81-1.363.3924.300.8614.090.092216288680.88-0.963.4324.441.3014.910.062218288741.08-1.623.8725.790.9216.400.062220288810.68-1.843.5024.680.4715.120.062222288870.69-1.733.5524.830.6115.230.06 2223.5288910.84-1.293.8125.611.2116.140.08 2225288961.13-1.043.8425.701.4816.400.07222728902 2228.5289070.87-1.353.7725.491.1316.050.09 2230289110.73-1.063.6024.981.3114.690.112232289180.99-1.233.6525.151.1714.970.172234289241.13-1.603.8725.790.9317.140.112236289300.93-0.673.6125.021.7114.600.072238289360.84-0.943.7725.481.5315.120.072240289420.66-1.243.8325.661.2716.110.072242289480.95-1.893.4124.390.3617.170.072244289550.84-1.143.5224.751.1814.430.062246289610.73-1.763.7925.550.7315.910.052248289671.01-1.313.3924.330.9215.120.102250289730.80-1.253.5724.881.1015.870.082252289790.95-1.243.3024.020.9315.130.062254289850.84-1.643.8225.640.8714.530.112256289920.71-1.803.5424.790.5316.330.072258289980.60-1.323.5324.781.0015.000.082260290040.93-1.6913.250.062262290100.86-1.283.3024.010.8914.880.072264290160.69-1.483.6125.010.9014.950.072266290220.82-1.633.3324.100.5620.300.072268290290.88-1.783.5824.940.5814.530.09 90
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2270290350.66-1.2214.300.072272290410.88-1.283.6625.171.1315.470.08 2273.5290450.64-0.813.1923.641.2812.420.05 2275290501.12-0.793.5924.951.5815.210.182277290560.89-0.573.7425.421.8914.520.09 2278.5290611.05-1.123.5524.841.2215.830.11 2280290660.73-1.353.5224.720.9715.110.152282290720.94-1.173.5724.881.1815.000.132284290780.93-1.204.0326.231.4314.490.072286290840.96-1.223.3024.020.9515.230.142288290950.80-1.313.7825.521.1713.140.092290291061.06-1.403.7625.451.0715.480.112292291171.00-1.8215.250.142294291280.79-1.063.7525.431.4016.100.112296291400.83-1.093.7525.421.3715.710.112298291511.11-1.673.7625.450.7914.670.112300291621.00-1.163.8625.761.3713.590.112302291730.74-1.2110.850.122304291840.84-1.173.4224.421.0815.480.172306291950.96-1.613.5124.690.7013.430.122308292060.92-1.503.6125.010.8716.480.172310292170.73-1.263.5824.921.0915.080.122312292280.85-1.464.1826.641.2514.550.092314292400.87-1.103.6725.201.3115.150.122316292510.70-1.243.5924.941.1215.900.152318292621.05-1.473.8325.671.0515.330.122320292731.15-1.813.7425.410.6516.130.152322292840.89-2.1913.330.10 2323.5292921.04-0.8710.000.09 2325293011.08-1.503.9826.101.1015.690.062327293121.06-1.283.8725.781.2514.490.16 2328.5293201.09-1.473.7925.551.0216.380.09 2330293281.00-1.313.5624.861.0315.160.102332293391.07-1.103.5424.811.2316.250.112334293511.04-1.023.7125.331.4214.900.102336293620.86-1.503.6125.020.8814.280.122338293730.87-1.203.4624.561.0911.640.102340293841.04-1.423.6025.000.9614.770.112342293950.96-1.153.7225.341.2913.930.092344294060.95-1.623.9526.020.9713.480.072346294170.93-1.743.8325.690.7814.040.042348294280.78-1.273.7425.411.1915.400.032350294390.94-1.393.8825.811.1516.640.042352294510.98-1.423.7625.451.0414.670.052354294621.07-0.963.5824.931.4017.060.02 91
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2356294730.93-0.844.1326.511.8512.630.052358294840.92-0.974.0426.281.6715.820.152360294951.08-1.403.4824.600.8916.000.152362295180.81-0.633.2923.991.5415.600.112364295410.90-0.963.4724.581.3315.720.112366295650.95-1.043.5524.821.3015.280.112368295880.98-1.143.4224.411.1115.750.092370296110.96-0.693.5224.721.6315.480.052372296341.15-1.043.5224.741.2816.330.07 2373.5296521.04-1.0816.800.08 2375296691.00-1.033.6024.981.3416.090.132377296920.87-1.263.3824.290.9715.540.15 2378.5297101.14-1.393.5624.870.9615.930.12 2380297271.00-1.223.5624.871.1316.420.142382297500.91-1.463.3324.100.7215.200.122384297740.89-1.363.5024.680.9515.500.122386297970.89-0.993.4724.561.2915.130.102388298201.29-1.373.6525.151.0316.840.152390298431.11-0.973.6225.061.4215.600.132392298661.04-1.193.6825.231.2317.220.192394298900.97-1.133.7925.541.3617.180.192396299131.10-1.033.9526.011.5515.120.122398299361.12-1.313.4824.600.9818.130.202400299591.09-0.823.7025.281.6115.250.192402299831.36-0.913.7525.441.5515.380.132404300061.21-0.623.3424.141.5714.560.132406300291.27-1.963.4024.340.2714.860.112408300851.05-1.063.4824.611.2315.030.082410301411.24-1.143.6325.081.2516.030.052412301971.19-0.793.5524.821.5414.370.072414302531.21-1.8817.500.042416303091.08-0.9911.170.022418303651.50-1.4213.500.062420304211.31-0.923.7325.371.5317.650.092422304771.38-1.193.7725.501.2917.750.20 2423.5305190.84-0.4715.690.23 2425305621.31-1.2616.770.172427306181.28-0.923.4624.561.3616.490.15 2428.5306601.08-1.463.5324.770.8616.260.08 2430307021.36-0.893.5624.851.4516.830.092432307581.22-1.003.7125.311.4318.350.102434308141.27-1.193.2623.900.9513.390.192436308701.22-0.843.5024.661.4616.960.222438309261.03-0.923.3824.301.3116.320.462440309821.25-1.083.7725.491.4017.230.25 92
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2442310381.13-0.613.5624.861.7318.300.262444311160.96-0.713.5224.721.6016.060.282446311941.23-0.963.6024.981.4216.110.232448312721.24-0.833.5624.841.5116.780.252450313511.33-1.103.5324.771.2217.070.322452314291.29-0.743.3324.131.4517.700.402454315071.43-1.013.4024.351.2317.360.412456315851.50-1.103.5524.821.2418.120.492458316631.19-0.953.4124.381.3018.800.462460317411.24-1.353.2723.930.8018.750.572462318191.42-0.993.4524.501.2813.740.482464318981.11-1.443.4224.420.8216.900.402466319761.05-1.123.4924.621.1816.500.482468320541.11-0.983.4324.461.2812.790.592470321321.35-0.973.2023.661.1316.430.432472322101.21-1.613.3224.070.5717.420.42 2473.5322691.26-1.633.2623.900.5116.510.74 2475323271.09-1.363.3324.120.8315.760.592477324061.08-1.243.3724.250.9815.170.71 2478.5324641.12-1.843.3924.320.4015.050.58 2480325231.32-1.273.0123.010.6915.120.622482326011.38-1.313.2723.920.8314.510.692484326791.55-1.303.2623.890.8518.920.842486327571.33-1.403.1323.440.6513.690.752488328351.35-1.673.1923.630.4210.130.802490329141.21-1.523.5924.960.8414.210.742492329921.20-1.203.4724.581.0811.840.922494330701.29-1.443.5324.770.8913.890.842496331481.37-1.533.2723.910.6215.821.252498332261.35-1.993.4324.440.2611.470.692500333041.34-1.593.5424.800.7414.000.572502333821.35-1.163.5724.891.1914.210.512504334611.26-1.253.3624.220.9613.830.702506335391.54-1.013.1623.531.0615.710.852508336171.37-1.093.2823.931.0715.101.002510336951.27-1.043.1723.561.0414.790.892512337871.53-1.253.2423.800.8714.020.962514338781.10-1.513.4724.560.7714.160.822516339701.50-1.273.6625.181.1416.820.912518340611.24-0.903.4024.361.3517.151.012520341531.22-1.003.6425.121.4015.830.972522342441.23-0.963.3324.121.2314.930.90 2523.5343131.49-0.923.1823.601.1614.780.80 2525343821.58-0.713.2923.971.4516.460.932527344731.49-0.983.2123.711.1316.960.97 93
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2528.5345421.29-0.713.2923.991.4615.551.00 2530346111.48-0.703.1923.631.3912.320.062532347021.30-0.493.2123.691.6113.371.562534347941.49-0.683.0323.081.3015.331.562536348861.39-0.773.1223.401.2715.071.312538349771.35-1.213.3124.040.9615.421.382540350691.09-1.153.3724.241.0613.991.262542351601.44-1.063.2523.831.0714.761.062544352521.28-0.703.2423.821.4315.961.202546353431.41-1.133.7425.401.3316.191.202548354351.27-1.913.8125.620.6014.791.382550355181.24-1.023.5724.891.3318.081.062552356021.42-0.823.5124.691.4916.380.942554356851.31-1.083.4824.621.2215.970.812556357681.39-0.953.7525.451.5217.100.922558358521.44-0.833.9826.111.7717.120.852560359351.47-0.973.7925.551.5218.790.812562360181.46-0.913.5224.731.4119.720.742564361021.50-1.083.6325.081.3116.490.582566361851.40-0.993.5324.781.3415.640.792568362681.39-0.983.8825.811.5616.550.642570363521.33-0.903.7825.531.5819.180.802572364351.63-0.984.0126.181.6418.790.86 2573.5364981.20-1.173.6225.061.2116.330.70 2575365601.08-0.713.5824.921.6414.830.692577366431.00-0.843.5724.891.5112.890.63 2578.5367061.47-0.853.6024.991.5215.620.65 2580367681.26-1.243.3824.280.9815.320.662582368521.23-1.163.4324.441.0914.630.512584369351.11-1.123.7625.461.360.522586370181.17-0.834.0026.151.7815.671.282588371021.20-1.123.8525.731.4115.831.342590371851.44-1.033.5324.771.3015.920.702592372431.15-1.713.5424.800.6214.700.672594373001.01-1.143.9826.101.4713.460.942596373581.13-1.983.5524.820.3615.870.772598374151.17-2.053.4924.630.2517.210.872600374731.11-1.503.3224.080.6816.380.792602375301.14-2.3615.731.092604375881.26-2.073.4924.640.2317.030.662606376451.24-2.013.5724.880.3415.010.412608377031.59-1.613.4824.620.6816.490.362610377601.42-1.733.3224.070.4515.280.292612378181.14-1.573.5824.920.7915.050.182614378751.25-1.653.7225.330.7917.110.32 94
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2616379331.12-2.563.6625.17-0.1616.350.212618379901.31-1.813.9325.970.7716.440.282620380481.08-2.003.5024.670.3118.300.222622381050.92-1.553.3424.130.6516.280.34 2623.5381481.14-1.803.6325.060.5816.130.24 2625381921.22-2.043.8125.610.4616.490.222627382491.23-1.933.5724.880.4216.440.13 2628.5382921.03-1.583.7325.380.8816.140.11 2630383350.97-1.893.8225.640.6214.790.102632383931.09-1.903.7325.380.5515.890.062634384501.21-1.613.9425.990.9715.130.092636385081.05-2.103.7025.280.3316.560.122638385651.24-1.903.7125.330.5516.480.192640386231.13-2.263.7425.410.2016.160.142642386801.04-1.553.7925.550.9416.180.342644387381.13-1.873.8925.840.6816.390.212646387951.19-1.983.9626.050.6116.200.112648388531.28-2.423.9626.040.1717.290.302650388911.18-2.064.2126.710.6717.900.242652389291.13-1.864.3727.1415.950.232654389671.34-1.993.6625.160.4117.260.372656390051.01-2.0916.490.332658390421.11-1.5313.310.272660390801.05-1.723.3824.290.5015.000.272662391180.94-2.123.7925.540.3615.470.452664391561.26-1.963.4224.430.3016.660.332666391940.94-2.143.4824.610.1515.900.352668392321.08-1.563.8825.800.9815.280.342670392701.22-1.553.7025.280.8916.430.342672393081.06-1.923.7225.360.5315.690.24 2673.5393360.85-1.973.6125.000.4115.070.10 2675393650.92-1.52 0.09 2677394021.09-1.644.4527.341.2214.680.32 2678.5394310.86-2.153.7825.540.3416.410.33 2680394591.03-1.913.7525.430.5516.210.282682394971.07-1.853.5824.930.5115.910.252684395351.09-2.013.7925.540.4714.820.312686395731.35-2.003.6125.010.3816.790.282688396111.33-1.513.5124.710.8115.990.352690396491.16-1.823.6525.150.5916.000.192692396870.99-1.603.8925.830.9415.090.072694397251.01-1.504.3227.011.2916.390.382696397621.46-1.603.7625.450.8717.001.882698398001.46-1.953.5024.670.3516.970.462700398381.41-1.883.8925.850.6715.660.29 95
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2702398761.29-1.493.4824.600.8016.000.282704399141.11-1.103.6125.021.2816.210.482706399521.16-1.183.6525.131.2216.000.222708399901.14-0.763.6925.241.6716.760.232710400281.22-1.043.6725.181.3816.500.212712400661.16-1.523.9526.011.0616.690.292714401031.24-1.623.6925.250.8116.030.242716401411.29-1.203.6525.131.2016.470.352718401791.20-1.383.7625.461.0916.230.282720402171.23-1.293.5724.881.0616.900.372722402551.17-0.953.5624.861.4016.530.47 2723.5402821.30-1.283.9125.911.2815.600.44 2725403091.14-1.113.8925.841.4416.930.402727403451.21-1.023.7425.401.4415.300.38 2728.5403721.17-1.273.7225.331.1817.310.77 2730403991.35-1.133.7925.571.3717.370.302732404351.14-0.703.8125.621.8016.640.142734404711.01-0.803.8525.731.7315.710.502736405071.09-0.963.8325.661.5515.960.462738405431.30-1.033.8225.631.4817.360.442740405791.23-1.163.8225.651.3517.260.312742406141.30-0.8218.110.782744406501.25-0.703.7925.561.7917.030.352746406861.37-1.453.7325.381.0118.690.332748407221.32-1.043.9225.921.5317.210.172750407581.47-0.793.8725.791.7517.440.382752407941.34-0.933.8925.841.6216.460.652754408301.33-1.0616.310.342756408661.28-1.093.7925.551.4017.210.592758409021.24-0.613.7325.371.8416.710.552760409321.11-0.543.6925.261.8816.770.352762409611.08-0.623.7825.521.8616.940.192764409911.06-1.073.7325.381.3915.480.222766410211.51-0.543.9025.862.0116.090.082768410501.28-1.163.8525.731.3716.480.112770410801.51-1.484.1326.501.2117.302.322772411101.33-0.904.0126.201.7217.535.04 2773.5411321.35-0.833.7925.551.6615.211.37 2775411541.12-0.703.9225.921.8615.480.312777411841.54-0.753.5624.851.6016.570.49 2778.5412061.45-0.913.4724.581.3816.570.48 2780412281.30-1.033.6125.021.3516.410.142782412581.25-0.663.9225.941.9115.630.262784412881.18-0.613.8725.801.9316.890.342786413171.34-0.793.5024.681.5116.910.69 96
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2788413471.46-0.653.5324.761.6716.900.602790413771.57-0.753.2924.001.4216.690.392792414061.52-0.683.4424.481.5915.900.382794414361.36-0.803.7025.301.6417.290.442796414661.37-0.853.9325.961.7316.930.432798414951.44-0.463.7825.532.0216.130.502800415251.39-0.983.1323.431.0716.900.552802415551.58-0.743.4124.381.5115.680.522804415841.51-0.573.6525.141.8416.020.292806416141.42-0.603.5024.681.7116.750.472808416441.18-0.583.8225.631.9217.000.472810416731.46-1.333.8425.691.1915.930.532812417031.19-0.763.7125.311.6816.800.592814417321.15-0.903.8425.701.6217.210.582816417621.29-0.653.7525.441.8216.260.752818417921.45-0.833.6625.181.5817.271.252820418211.33-0.693.6125.021.6815.080.662822418511.47-0.503.4524.521.7814.950.37 2823.5418731.47-0.963.4824.621.3314.410.49 2825418961.21-0.833.4024.351.4116.020.422827419251.39-1.083.7825.521.4016.330.68 2828.5419481.39-1.193.5424.781.1416.090.52 2830419701.50-0.793.3824.291.4414.840.442832419991.30-1.333.5224.730.9915.240.382834420291.25-1.103.3024.001.0615.930.792836420591.46-0.833.8725.781.7015.370.542838420881.20-1.173.3924.311.0615.740.602840421181.33-1.003.5324.771.3215.390.582842421481.22-0.813.5524.841.5316.100.602844421771.32-1.733.7025.290.7116.490.462846422071.54-1.163.6024.971.2016.590.382848422371.33-1.233.5724.901.1315.260.352850422660.82-1.594.0526.301.0513.430.342852422961.29-0.803.5924.941.5616.510.512854423261.19-0.853.6225.061.5315.660.522856423551.45-0.7416.460.342858423851.29-1.063.8325.671.4517.720.392860424151.03-1.433.8825.811.1115.360.402862424440.92-1.763.9225.920.8116.260.322864424741.12-1.073.8025.581.4317.090.412866425040.84-1.053.5824.921.3116.220.422868425331.05-1.033.7325.371.4215.740.282870425631.05-0.973.9225.941.5915.600.142872425931.04-0.9816.290.12 2873.5426150.90-1.103.9125.891.4616.830.92 97
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2875426371.20-0.983.8925.861.5817.290.532877426670.99-1.703.4224.400.5516.330.49 2878.5426891.03-0.743.0923.301.2813.960.48 2880427110.94-0.723.5124.701.5914.710.552882427410.86-1.233.9726.071.3715.340.312884427710.87-1.104.0626.311.5514.500.292886428001.28-0.843.6725.191.5816.170.612888428301.39-1.193.9225.941.3817.030.472890428601.09-0.713.4724.591.5815.630.422892428891.29-1.273.4424.481.0015.570.422894429191.17-1.443.8525.741.0817.400.502896429491.09-1.163.8825.801.3814.630.372898429781.20-1.213.8925.851.3414.140.402900430081.14-1.543.2623.900.6114.460.332902430381.12-1.023.7725.491.4615.660.432904430671.08-0.943.8825.821.6115.460.442906430970.97-1.953.4224.420.3115.550.352908431271.16-1.423.5024.660.8914.570.602910431560.95-1.403.4924.630.9013.630.232912431860.91-1.183.3324.111.0116.060.242914432161.21-1.023.1723.581.0515.140.342916432450.82-0.843.5224.741.4813.620.272918432751.03-1.013.6325.061.3715.660.342920433050.82-0.513.1723.561.5615.370.382922433340.79-0.703.3724.241.5214.030.34 2923.5433561.01-0.763.4224.401.4914.560.36 2925433790.80-0.923.4024.351.3213.750.332927434080.93-1.353.4524.500.9214.800.45 2928.5434311.03-1.493.8025.571.0114.500.38 2930434530.83-1.523.8425.711.0016.800.322932434820.92-1.593.9726.081.0115.170.342934435120.99-1.683.7625.470.7914.040.302936435421.09-1.393.8625.771.1414.400.392938435710.90-1.143.4324.451.1213.540.232940436011.09-1.273.9325.961.3015.080.322942436310.98-1.203.9025.891.3614.370.382944436600.84-2.113.7425.400.3515.400.412946436901.18-1.363.5124.710.9515.170.412948437201.28-1.333.4024.360.9115.430.542950437491.17-1.743.5024.680.5716.380.722952437791.14-1.243.6825.231.1815.570.752954438091.22-1.383.4724.580.9116.140.672956438381.16-1.623.2623.890.5216.510.832958438681.22-0.993.4024.361.2614.230.362960438980.99-1.153.5824.941.2115.000.59 98
Appendix A (Continued) depthcalyr pink d 13 C pink d 18 O c pink Mg/Ca pink SST pink d 18 O sw weight/ foram % coarse fraction 2962439270.94-0.944.5527.581.9714.260.472964439571.07-0.983.8025.601.5214.290.312966439871.13-1.533.8125.620.9715.710.242968440160.98-1.063.7225.341.3915.460.272970440461.24-1.253.6825.241.1715.950.282972440761.34-1.093.2123.711.0116.140.25 2973.5440981.26-1.653.6024.980.7217.000.21 2975441201.18-1.373.7725.491.1015.690.792977441501.42-1.143.6024.991.2315.830.56 2978.5441721.16-1.103.7425.401.3615.780.42 2980441941.07-0.993.7225.341.4615.660.322982442241.29-1.303.5124.691.0116.490.762984442541.42-0.963.4124.381.2816.311.022986442831.45-1.313.4924.650.9915.860.972988443131.54-0.903.4924.641.4015.081.572990443431.11-0.753.5824.931.6016.230.882992443721.37-1.193.5824.911.1615.491.242994444021.26-0.693.3224.081.4915.171.282996444321.16-1.543.6425.110.8616.001.592998444611.29-1.193.4524.511.0815.151.55 99
Appendix A (Continued) calyr white d 13 C white d 18 O c white Mg/Ca white SST white d 18 O sw dutertrei d 13 C dutertrei d 18 O c 346110.97-0.21 1.450.92 347940.92-0.433.3922.481.421.070.96349771.25-1.143.5522.990.821.370.99351601.41-1.463.8523.860.681.560.95353430.89-1.083.5322.900.861.441.01355180.75-0.583.2522.011.171.711.38356851.02-0.773.3622.351.051.381.21358520.79-0.763.6023.121.221.011.53360181.19-0.613.7923.701.491.751.21361851.10-0.243.2822.091.531.741.20363521.15-0.533.8523.871.611.361.47365600.89-0.673.0821.410.961.531.17367680.83-0.96 1.571.30 369350.95-0.453.0721.3126.96.36.199371020.85-0.442.9020.741.051.570.91372431.01-0.892.9520.920.631.601.03373580.77-1.342.9320.840.171.550.7637473-1.863.4222.570.011.510.80375880.98-1.703.4022.500.151.480.94377030.93-1.213.7823.680.881.101.18378180.94-1.113.6123.150.881.210.7237933-1.813.2121.84-0.091.180.09380480.97-1.713.4322.600.160.981.17381920.62-0.603.5923.091.371.221.42383350.77-1.023.8223.791.101.020.96384500.67-1.133.6323.210.870.740.87385650.72-1.043.7023.431.011.271.11386800.70-1.363.5322.910.581.301.01387950.75-0.73 0.941.31 388913.5022.820.841.15389670.74-1.453.7523.580.630.601.24390420.69-0.793.1521.640.880.931.28391180.62-0.903.5623.001.060.801.23391940.80-1.203.4722.710.700.780.88392700.76-0.833.1421.600.830.420.96393650.42-0.993.4222.570.880.951.14394590.47-0.683.1921.801.030.520.95395350.52-0.653.0321.210.941.100.74396110.45-0.342.7920.311.061.080.75396870.85-1.313.6623.320.710.470.51397620.77-0.503.1621.681.180.611.25398380.76-1.243.5422.940.711.110.68399140.360.442.8720.611.900.860.98 Table 4. White G. ruber and N. dutertrei d 18 O, d 13 C, and Mg/Ca analyses. 100
Appendix A (Continued) calyr white d 13 C white d 18 O c white Mg/Ca white SST white d 18 O sw dutertrei d 13 C dutertrei d 18 O c 399900.670.023.2421.971.770.781.08400660.71-0.403.4122.531.460.911.12401410.61-0.433.3322.271.380.841.43402170.96-0.363.1821.751.340.07403090.59-0.243.2922.141.53403990.740.093.0221.181.660.951.16404710.880.163.4622.692.050.851.11405430.78-0.013.2521.981.731.021.13406140.94-0.263.5122.861.671.270.90406860.57-0.233.3522.331.590.840.79407580.60-0.383.2121.851.331.090.90408300.69-0.652.9921.070.911.310.96409020.81-0.343.2922.121.431.310.79409610.86-0.423.3022.151.361.440.62410211.17-0.733.2522.011.031.390.82410800.95-0.783.5322.911.161.390.59411540.72-0.933.3222.230.861.390.80412280.81-0.462.9420.891.061.160.86412880.97-0.153.1921.781.551.321.20413471.28-0.443.2321.941.291.191.1741406-1.263.1121.490.391.310.94414661.07-0.603.5122.841.331.141.06 101
Appendix A (Continued) calyr GOM pink d 18 O GOM precip endmember Agassiz endmember LIS endmember white d 18 O GOM 437500.1127.7634.3435.42438000.3229.4134.8335.67438500.2428.7734.6435.58439000.8133.1235.9236.24439501.0234.7836.4136.50440000.6131.6235.4836.01440500.5731.3435.3935.97441000.3229.4034.8235.67441500.5931.4635.4335.99442000.5931.4635.4335.99442500.4530.3735.1135.82443000.5130.8735.2635.90 109
110 Appendix B: Organic and sediment compositional analyses
111 Appendix B (continued) Abbreviations used in Appendix B tables: depth: depth in the core (cm) calyr: calendar years before present d 13 C org : d 13 C ( PDB) of the bulk organic matter % total carbon Percent of total carbon in the bulk sediment % inorg carbon Percent of inorganic carbon in the bulk sediment % CaCO 3 Percent of calcium carbonate in the bulk sediment % organic Percent of organic carbon in the bulk sediment % residual Total weight percent minus the % organic and % inorganic carbon weight extracted Weight of total sediment extracted (g) weight C extracted Weight of total sediment extracted normalized to perce nt organic carbon (g) conc Concentration of n-alkanes ( g) with different carbon lengths terr/org C Total terrestrial n-alkanes (C 25 + C 27 + C 29 + C 31 ) divided by weight of organic carbon (mg/g) algal/org C Total algal n-alkanes (C 16 + C 17 + C 18 + C 19 ) divided by weight of organic carbon (mg/g) d 13 C C-25 Compound specific d 13 C on terrestrial n-alkane
Appendix B (Continued) depthcalyr d 13 C org % total carbon % inorg carbon %CaCO 3 % organic carbon % residual 297744150-25.312.601.6213.490.9885.54 113
Appendix B (Continued) calyr conc 25 (ug) conc 27 (ug) conc 29 (ug) conc 31 (ug) terr/org C (mg/g) algal/org C (mg/g) 2869231.3038.9425.5612.190.710.01288749.6117.8616.6011.540.590.012891818.4224.8715.058.950.620.042898535.6264.6759.8539.271.420.032909530.6640.1126.9517.231.010.042926223.8433.6825.3815.980.680.042943930.4441.0729.4317.831.280.042966910.1324.5925.3017.870.970.01298906.6612.1311.407.920.580.033092629.4041.8626.3213.610.910.023111658.6772.4746.8225.402.560.063361722.5741.8438.1029.971.470.093488655.3172.8747.7520.051.640.043576852.6567.5541.6322.421.460.023670653.4971.4648.4019.181.780.033724358.9880.6451.9724.791.930.033730035.1250.9536.1718.791.860.043799020.9631.8025.7918.620.670.053930837.8144.4428.3213.180.940.043980038.7745.1227.6616.530.740.013991428.6333.3221.8411.210.550.023999038.2142.0924.2714.110.720.024090229.7536.3724.7314.540.680.034118444.1447.0828.9312.500.740.024179230.6535.2423.6812.370.620.014202940.2744.3027.9012.690.850.024297838.0541.5125.1311.650.810.024321639.0046.8731.0418.040.870.034340840.3646.8324.2912.100.810.014415045.3858.0941.9526.251.420.02 Table 8. High molecular weight n-alkane concentrations. 115
Appendix B (Continued) depthage d 13 C C-25 d 13 C C-27 d 13 C C-29 d 13 C C-31 216028501-32.82-31.47-30.61-31.55218428692-29.87-29.80-30.48220428831-29.74-29.79-29.35-30.70221828874-31.40-31.04-30.35-31.26223228918-29.86-29.81-28.41-30.24225428985-30.66-30.96-30.83-31.10227329045-31.13-30.39-29.60-30.51228829095-28.45-29.01-29.92231829262-30.39-30.69-30.14-30.69235029439-29.23-29.43-28.85237529669-27.03-19.59-29.33-30.91239429890-31.52-30.59-30.30-29.46242230477-30.99-30.73-29.90-30.96243830926-30.71-30.73244431116-28.15-28.83246631976-29.72-29.34248632757-30.34-30.77-29.36-28.72250833617-31.30-31.03-30.56-29.55252234244-31.45-31.73-30.57-31.54253634886-30.48-30.19255635768-30.09-30.97257836706-29.23-30.97259437300 -30.27 261837990-30.37-30.94-30.23-31.07264638795-31.15-31.08-30.34-30.73267239308-29.47-29.16269839800-29.23-28.87270439914-30.37-30.41270839990-32.36-29.23-28.45-27.93272040255-31.99-31.63-30.67-31.56275840902-30.76-30.80-30.08-30.38277741184-29.60-29.56-29.01-29.14281841792-29.05-30.14283442029-30.00-29.33-29.90-30.20286842533-30.47-29.62-29.98-30.54289842978-29.40-29.55-29.71292743408-29.06-29.11295443809-32.06-30.94-30.83-30.80297744150-30.77-29.97-29.59-29.46 Table 9. d 13 C of high molecular weight n-alkanes. 116
About the Author Heather W. Hill was born in Huntsville, AL and raised in a military family. She received her undergraduate degree from Albion College, MI. She then attended the University of Maine, where she received a Masters degree in Geological Sci ences with Dr. Joseph Kelley, studying coastal processes. She began the Ph.D. program in the College of Marine Science, USF in August 2001 to study past climate change with Dr. Benjamin Flower. She has been afforded many unique opportunities during her tenure at USF. She participated in multiple cruises to the Gulf of Mexico, including one international cruise that served as the foundation for her research. She has also been involved in education outreach programs through USF, including two years of the NSFs GK-12 program.