|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 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004541
Late holocene climate variability from northern gulf of mexico sediments :
b merging inorganic and molecular organic geochemical proxies
h [electronic resource] /
by Julie Richey.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Dissertation (PHD)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: Accurate reconstruction of natural climate variability over the past millennium is critical for predicting responses to future climate change. In order to improve on current understanding of climate variability in the sub-tropical North Atlantic region over the past millennium, a rigorous study of Gulf of Mexico (GOM) sea surface temperature (SST) variability was conducted using both inorganic (foraminiferal Mg/Ca) and molecular organic (TEX86) geochemical proxies. In addition to generating multiple high-resolution climate records, the uncertainties of the SST proxies are rigorously assessed. There are 3 major research questions addressed: (1) What was the magnitude of GOM SST variability during the past 1,000 years, particularly during large-scale climate events such as the Little Ice Age (LIA) and the Medieval Warm Period (MWP). (2) Is the SST signal reproducible within the same sediment core, among different northern GOM basins, and using different geochemical SST proxies? (3) What are the ecological controls on the paleothermometers used to reconstruct SST variability in the GOM? Can differences in the ecology (i.e. seasonal distribution, depth habitat, etc.) of distinct paleothermometers be exploited to gain insight into changes in upper water column structure or seasonality in the GOM during the LIA and MWP? The major findings include: (1) The magnitude of temperature variability in the GOM over the past millennium is much larger than that estimated from Northern Hemisphere temperature reconstructions. The MWP (1400-900 yrs BP) was characterized by SSTs in the GOM that were similar to the modern SST, while the LIA (400-150 yrs BP) was marked by a series of multidecadal intervals that were 2-2.5C cooler than modern. (2) This LIA cooling was replicated in the Mg/Ca-SST records from three different well-dated northern GOM basins (Pigmy, Garrison and Fisk Basins), as well as in two different geochemical proxies. (3) It is determined that foraminiferal test size has a significant effect on shell geochemistry. Using core-top calibration, discrepancies in the seasonal/depth habitats between different planktonic Foraminifera, and between Foraminifera and Crenarchaeota are inferred. Downcore differences are used to make inferences about changes in GOM mixed layer depth and seasonality over the past millennium.
Advisor: Benjamin P. Flower, Ph.D
Little Ice Age
Medieval Warm Period
Gulf of Mexico
x Marine Science
t USF Electronic Theses and Dissertations.
Late Holocene Climate Variability From Northern Gulf of Mexico Sediments: Merging Inorganic and Molecular Organic Geochemical Proxies by Julie N. Richey A dissertation submitted in partial fulfillment of the requireme nts for the degree of Doctor of Philosophy College of Marine Science University of South Florida Co Major Professor: Benjamin P. Flower Ph.D. Co Major Professor: David J. Hollander Ph.D. Terrence M. Quinn, Ph.D. Richard Z. Poore, Ph.D. David W. Hastings, Ph.D. Date of Approval: July 12, 2010 Keywords: little ice age, medieval warm period, foraminifera, climate change, holocene Copyright 2010 Julie N. Richey
i Table of Contents List of Tables iii List of Figures iv Abstract vi Chaper 1 Introduction 1 1.1. Introductory Remarks 1 1.2. The Little Ice Age 5 1.3. The Medieval Warm Period 9 1.4. Reconstructing SST over the Past 2,000 Years 11 1.4.1. Molecular Organic Proxies 11 1.4.2. Foraminifer al based Proxies 15 1.5. Dissertation Organization 19 Chapter 2 Regionally Coherent Little Ice Age Cooling in the Atlantic Warm Pool 22 2.1. Abstract 22 2.2. Introduction 23 2 .3. Materials and Methods 25 2.4. Gulf of Mexico Records 28 2.5. Regional Comparisons 31 2.6. Discussion 33 2.7. Conclusions 36 Chapter 3. Merging Late Holocene Molecular Organi c and Foraminiferal Based Geochemical Records of SST in the Gulf of Mexico 38 3.1. Abstract 38 3.2. Introduction 39 3.3. Study Location 42 3.4. Methods 43 3.4.1. E xtraction and Isolation of GDGT Lipids 43 3.4.2. TEX 86 and BIT Analysis 44 3.5. TEX 86 SST Record from the Pigmy Basin 45 3.6. Influence of Terrestrial Input on Pigmy Basin TEX 86 47 3.7. Inferring Depth an d Seasonality for TEX 86 Signal in the GOM 49 3.8. Comparison of Pigmy Basin TEX 86 SST and Mg/Ca SST Record 50 3.9. Inferring Depth and Seasonality for G. ruber in the GOM 53 3.10. Mg/Ca SST to TEX 86 SST Gradients 54
ii 3.11. Potential Implications for the T Record 58 3.12. Conclusions 59 Chapter 4. Ecological Controls on the Shell Geochemistry of pink and white Globigerinoides ruber in the northern Gulf of Mexico: Implications for paleoceanographic reconstruction 61 4.1. Abstract 61 4.2. Introduction 62 4.3. Materials and Methods 63 4.4. Results 65 4.4.1. R elationship between test size and carbon isotopic composition 65 4.4.2. Relationship between test size and oxygen isotopic composition 67 4.4.3. Relationship between foraminiferal test size and Mg/Ca 69 4.5. Discussion of Size Fraction D ata 72 4.6. Comparison of downcore geochemical records for pink and white G. ruber 77 4.6.1. Comparison of white and pink Mg/Ca records 79 4.6.2. Comparison of downcore 18 O data 83 4.7. C onclusions 86 Chapter 5. Summary 88 5.1. Conclusions 88 5.2. Future Research Directions 89 References Cited 92 Appendice s 105 Appendix I. Fisk Basin (PE07 5) Data 106 AI.1. Fisk Basin Downcore Mg/Ca Data 106 AI.2. Fisk Basin Radiocarbon Data 107 Appendix II. Garrison Basin (PE07 2) Data 108 AII.1. Garrison Bas in Mg/Ca Data 108 AII.2. Garrison Basin Radiocarbon Data 110 Appendix III. Pigmy Basin GDGT Data 111 AIII.1. TEX 86 Data for Pigmy Basin (PBBC 1F) 111 AIII.2. BIT Index Data for Pigmy Basin (PBBC 1F) 113 Appendix IV. Elemental and Isotopic Size Fraction Data 116 AIV.1 Size Fraction Isotope Data for G. ruber (white) 116 AIV.2. Size Fraction Mg/Ca Data for G. ruber (white) 117 A IV.3. Size Fraction Isotope Data for G. ruber (pink) 118 AIV.4. Size Fraction Data for Mg/Ca in G. ruber (pink) 119 AIV.5. Summary of Size Fraction Data 120 About the Author End Page
iii List of Tables Table 2.1. Radiocarbon Dates 27 Table 4.1. Oxygen and Carbon isotopic data for G. ruber (pink) 68 Table 4.2. Oxygen and Carbon isotopic data for G. ruber (white) 69 Table 4.3. Mg/Ca data versus size for G. ruber (p ink) 71 Table 4.4. Mg/Ca data versus size for G. ruber (white) 71
iv List of Figures 1.1. Multiproxy Reconstruction of Northern Hemisphere Surface Temperature Variations over the Past Millennium 2 1.2. Map of the Sites for which there are Currently Published, Decadal resolution SST Proxy Records Covering at least the Past 1,000 Years 4 1.3. Solar Activity Record for the Last 1,600 Years 6 1.4. Ice Core Estimates of Global Stratospheric Sulfate Loading from Volcanoes (A.D. 500 2000) 7 1.5. Northern Hemisphere Mean Radiative Forcing 8 1.6. Long term Aridity Changes in the Western United States 10 1.7. Global Core top Calibration of TEX 86 to SST 14 1.8. Flow Chart of Paired Mg/Ca and 18 O for Estimating SST and 18 O of Seawater 16 2.1. Map of Proxy Records in the GOM Caribbean Region Exhibiting 1 3C Cooling Dur ing the LIA 24 2.2. Age Models 27 2.3. Gulf of Mexico Mg/Ca Records 29 2.4. AWP Regional SST Comparisons 32 3.1. Map of the Gulf of Mexico 42 3.2. Molecular Structures of GDGTs 45 3.3. GDGT based Proxy Records For Pigmy Basin Box Core (PBBC 1F) 46 3.4. Cross plot Between BIT index and TEX86 SST for Pigmy Basin Box Core (PBBC 1F) 48
v 3.5. Annual Cycle of Water Temperature Variabilitiy in the Up per 100 meters of the Water Column 50 3.6. TEX 86 and Mg/Ca Records Plotted as Anomalies Relative to their Respective Calibrated Core top Temperatures 52 3.7. The T (TEX 86 Mg/Ca) for the Pigmy Basin, Gulf of Mexico 55 3.8. Illustration of Mixed Layer and Seasonality Hypotheses 57 3.9. Comparison of Pigmy Basin T (TEX 86 Mg/Ca) with a Reconstruction of Tropical Cyclone Counts 59 4.1. Illustration of the relationship between d 13 C and ontoge ny in Foraminifera 65 4.2. Relationship between foraminiferal test size and d 13 C and d 18 O 67 4.3. Relationship between Mg/Ca and test size 73 4.4. Conversion of d 18 O calcite to calcificatio n temperature 76 4.5. Vertical profiles and seasonal cycles of temperature and d 18 O sw variability in the GOM 77 4.6. Downcore comparison of Mg/Ca records for white and pink G. ruber 82 4.7. Dow ncore comparison of d 18 O data for white and pink G. ruber 84 4.8. Comparison of G. ruber abundance with GOM salinity 85
vi Late Holocene Climate Variability From Northern Gulf of Mexico Sedime nts: Merging Inorganic and Molecular Organic Geochemical Proxies Julie N. Richey ABSTRACT Accurate reconstruction of natural climate variability over the past millennium is critical for predicting responses to future climate change. In order to improve on current understanding of climate variability in the sub tropical North Atlantic region over the past millennium, a rigorous study of Gulf of Mexico (GOM) sea surface temperature (SST) variability was conducted using both inorganic (foraminiferal Mg/Ca) and molecular organic (TEX 86 ) geochemical proxies. In addition to generating multiple high resolution climate records, the uncertainties of the SST proxies are rigorously assessed. There are 3 major research questions addressed: (1) What was the magnitu de of GOM SST variability during the past 1,000 years, particularly during large scale climate events such as the Little Ice Age (LIA) and the Medieval Warm Period (MWP). (2) Is the SST signal reproducible within the same sediment core, among different no rthern GOM basins, and using different geochemical SST proxies? (3) What are the ecological controls on the paleothermometers used to reconstruct SST variability in the GOM? Can differences in the ecology (i.e. seasonal distribution, depth habitat, etc.) of distinct paleothermometers be exploited to gain insight into changes in upper water column structure or seasonality in the GOM during the LIA and MWP?
vii The major findings include: (1) The magnitude of temperature variability in the GOM over the past mil lennium is much larger than that estimated from Northern Hemisphere temperature reconstructions. The MWP (1400 900 yrs BP) was characterized by SSTs in the GOM that were similar to the modern SST, while the LIA (400 150 yrs BP) was marked by a series of mu ltidecadal intervals that were 2 2.5C cooler than modern. (2) This LIA cooling was replicated in the Mg/Ca SST records from three different well dated northern GOM basins (Pigmy, Garrison and Fisk Basins), as well as in two different geochemical proxies. (3) It is determined that foraminiferal test size has a significant effect on shell geochemistry. Using core top calibration, discrepancies in the seasonal/depth habitats between different planktonic Foraminifera, and between Foraminifera and Crenarchae ota are inferred. Downcore differences are used to make inferences about changes in GOM mixed layer depth and seasonality over the past millennium.
1 Chapter 1 Introduction 1.1. Introductory Remarks Global Climate Change is one of the most important problems that we face as a society today. The warming of the planet due to the increased anthropogenic input of greenhouse gasses may have large imp lications for the global hydrologic cycle, ocean circulation, global food production, disease and biodiversity. The Intergovernmental Panel on Climate Change (IPCC) states that global temperature increased by 0.6C over the course of the 20 th century, and models of 21 st century warming predict from 0.6C to 4.0C warming over the next century, depending on greenhouse gas emissions scenarios (IPCC, 2001). In order for the scientific community to more accurately predict the future response of Earth's climat e system to anthropogenic forcing, we must improve our understanding of natural (pre industrial) climate variability. A reliable instrumental record of climate variability only extends back to 1850 A.D. (Jones et al., 1999) and thus proxy based reconstru ctions must be relied upon to investigate climate variability further into the past. Studying the past 1,000 years of climate variability allows paleoclimatologists to explore natural modes of variability without having to account for major changes in bac kground state (i.e. global ice volume, orbital scale insolation changes, or tectonic scale changes). There were also significant
2 hemispheric to global scale events such as the Little Ice Age (LIA) and the Medieval Warm Period (MWP) that occurred during th e past millennium. These events act as benchmarks for determining the magnitude of natural climatic shifts during the late Holocene. A number of global and hemispheric surface temperature reconstructions covering the past millennium have been generated b y merging proxy records from tree rings, boreholes, ice cores, corals, speleothems, and sediment cores (e.g. Mann et al., 1999; Mann and Jones, 2003; Esper et al., 2002; Moberg et al., 2005), however there is still a large degree of uncertainty in these re constructions. Figure 1.1. Multiproxy reconstruction of Northern Hemisphere surface temperature variations over the past millennium. Temperature reconstruction (blue), along with 50 year average (black), a measure of the statistical uncertainty associ ated with the reconstruction (gray), and instrumental surface temperature data for the last 150 years (red), based on the work by Mann et al. (1999). One major source of uncertainty in large scale climate reconstructions is sparse spatial and temporal c overage. The further back in time you go, the fewer records there
3 are. Many of the climate proxy records, especially ones based on archives such as corals and tree rings, involve splicing together a succession of shorter (<200 years) time series. The fu rther back in time you go, the more difficult it is to find fossil archives. A second source of uncertainty is the dearth in spatial coverage. The existing large scale climate reconstructions are primarily based on extra tropical terrestrial records, wit h very few records representing the low latitudes and/or the marine environment. The third source of uncertainty in global climate reconstructions stems from the uncertainties in the individual proxy records. Many of the climate archives are based on geochemical or biometric measurements on living and/or fossil organisms (e.g. trees, corals, Foraminifera, etc.), and the assumption is made that these organisms are passive records of their environmental conditions. The interpretation of an environmental signal in biogenic proxies can be complicated by "vital effects", which are often species specific, and affect the geochemistry in a way that is unrelated to the climate signal contained within the geochemistry. Also, biogenic proxies may be changing the ir ecologies in response to changing environmental conditions, which can lead to large uncertainties in the interpretation of their climate records. In this dissertation I addressed each of these sources of uncertainty in our collective understanding of g lobal climate change over the past 1,000 years. First, I generated decadal resolution sea surface temperature (SST) records from three different sites in the Gulf of Mexico. Globally, there are very few (<10) published continuous, decadal resolution reco rds of SST variability covering the past millennium (Figure 1.2). Improving the spatial coverage of records of ocean surface conditions during this
4 important time interval is essential to improving our understanding of both regional and global patterns of climate variability. Figure 1.2. Map of sites for which there are currently published, decadal resolution SST proxy records covering at least the past 1,000 years Pink and green markers indicate planktonic foraminiferal Mg/Ca SST records from the M akassar Straits (Newton et al., 2006; Oppo et al., 2009). Yellow marker indicates a Cariaco Basin planktonic foraminiferal Mg/Ca SST record from Black et al. (2007). The blue markers indicate planktonic foraminiferal Mg/Ca SST records from the Great Baha mas Bank and the Dry Tortugas (Lund and Curry, 2006). The red markers indicate the 3 planktonic foraminiferal Mg/Ca SST and the TEX 86 SST records from the Gulf of Mexico that were generated as part of this dissertation, and published in Richey et al. (200 7 and 2009) The issue of uncertainty in SST proxy records is addressed in this dissertation in a number of ways. First, I test the reproducibility of the foraminiferal Mg/Ca SST proxy by replicating Mg/Ca SST records in multiple species of planktonic F oraminifera within the same sediment core. Second, I compare Mg/Ca SST records generated from three different sites within the same region. Third, I use a multi proxy approach, comparing SST reconstructions based on an inorganic geochemical proxy (elemen tal ratios in planktonic Foraminifera) and molecular organic geochemical proxy (TEX 86 index). Finally, I rigorously assess the ecological controls on the different paleothermometers
5 (e.g. different species of planktonic Foraminifera and marine Crenarchaeo ta), thus improving our ability to interpret depth range and seasonality of the SST signal derived from each individual record. 1.2. T he Little Ice Age The Little Ice Age was a cool interval beginning roughly 1500 A.D. and lasting until 1850 A.D that ha s been well documented in historical records from northern Europe and eastern North America. Anecdotal evidence, including a centuries long theme of winter paintings in Europe, unprecedented advances of mountain glaciers upon alpine villages, and sharp in creases in the price of grains due to crop failure, were all indications of extraordinarily cold winters during the 17 th and 18 th centuries. Although the most dramatic impacts of the LIA seem to be confined to the Northern Hemisphere, the LIA has been ide ntified as a significant event in a number of tropical and Southern Hemisphere climate records. Evidence is beginning to support the idea that the LIA was a nearly global phenomenon, however the timing and magnitude of cooling varied significantly among d ifference regions. In many cases, solar variability has been invoked to explain the cooling experienced during the LIA. The changes in incoming solar radiation due to changes in orbital parameters (i.e. precession of the equinoxes, eccentricity of the Earth's orbit, and tilt of the Earth's axis) were minimal over the past millennium, and thus cannot be invoked to explain the LIA cooling. The amount of solar radiation emitted by the sun does vary on shorter timescales, and has to do with the occurrence of sunspots, bright faculae, and other solar phenomena. Measurements of solar variability over the last two
6 11 year sunspot cycles estimate that the total solar irradiance varied by only 0.1% (Frhlich and Lean, 2004). The production of cosmogenic nucli des (e.g. 14 C and 10 Be) varies primarily as a function of solar activity, and thus the 14 C and 10 Be records can be used as proxies for solar variability in the late Holocene (Bard et al., 1997) (Figure 1.3). From the 14 C and 10 Be records (as well and hist orical sunspot counts), quasi periodic cycles have been identified in the record of solar irradiance. For example, there is an 11 year sunspot cycle, and a 200 year cycle called the Suess Cycle. Recent minima in the Suess Cycle of solar variability: the Oort (ca. 1100 A.D. ), Wolf (ca. 1300 A.D. ), Sprer (ca. 1500 A.D. ), and Maunder (ca. 1700 A.D. ) minima, have been correlated with episodes of glacial advance and cooling associated with the LIA. Figure 1.3. Solar Activity Record for the last 1,600 years A decadally averaged plot of the 14 C record is plotted for the past 1,600 years (Stuiver et al., 1998). Note that the y axis is inverted, and increasing 14 C production indicated decreased solar activity. The correlation of these strong solar minima with cooling events during the LIA is highly suggestive of a solar forcing for the LIA. However, the total reduction in solar irradiance during the Maunder Minimum was <0.6%, which is equivalent to a forcing of
7 ~0.7 watts per square meter (W/m 2 ) (Bard et al., 2000). This forcing is two orders of magnitude smaller than the solar forcing associated with changes in orbital parameters and glacial interglacial climate variability. Thus it is not likely that solar forcing alone caused the dramatic climate response during the LIA. It has been proposed that volcanic forcing may have been an important factor in LIA cooling (Crowley, 2008). Volcanic eruptions add large amounts of ash and sulfur gasses to the atmosphere, diminishing the amount of solar radiation reachi ng the surface, thus causing the Earth to cool. The degree of radiative forcing from a volcanic eruption depends on the magnitude and location, as well as the composition of the ejecta (Robock, 2000). Figure 1.4. Ice core estimates of global stratosp heric sulfate loading from volcanoes (A.D. 500 2000). (from Gao et al., 2008) For example, sulfate aerosols from tropical volcanic eruptions are transported globally by high altitude winds, thus causing widespread cooling. High latitude eruptions tend t o be more spatially restricted, and have less effect on global temperature. Ejecta composed of large volcanic ash particles settles quickly, and only causes regional cooling that lasts from a few days to a few weeks. Explosive eruptions release sulfur ga sses, which
8 combine with water vapor to form sulfate aerosols. In large eruptions, these sulfate aerosols are injected high into the atmosphere, where they can remain for years, having a longer term cooling effect. Figure 1.4 illustrates a proxy record o f sulfate loading from volcanic eruptions, reconstructed from ice cores. It appears that there were 3 pulses of unusually high volcanic activity between the 14 th and 19 th centuries, which may have contributed to observed LIA cooling. Figure 1.5. Northe rn Hemisphere Mean Radiative Forcing Hegerl et al. (2006) estimate radiative forcing due to greenhouse gasses (Ghg), tropospheric aerosols, solar variability and volcanism over the past 1,000 years. Although solar variability is often cited as the pr imary forcing mechanism for LIA cooling, volcanism must also be considered as a potential agent for the LIA. Hegerl et al. (2003) did a comprehensive test of this issue and found volcanism to be substantially more important than solar variability, explain ing 40% of the cooling during the LIA. There has been some suggestion that the LIA and MWP were not forced by external
9 factors like solar variability or volcanism, but by a millennial scale internal oscillation. Broecker (2001) hypothesized that the MWP t o LIA transition was the penultimate cycle in a series of quasi periodic oscillations called "Bond Cycles", with a periodicity of roughly 1500 years (Bond, 1997). Another potential mechanism for LIA cooling may have been a slow down in the thermohaline ci rculation. Lund et al. (2006) found proxy evidence for a substantial decrease in the flow through the Florida Straits from 1200 1850 AD suggesting a reduction in Gulf Stream transport during that time. A reduction in reduction in heat transport to northe rn Europe via the Gulf Stream may have contributed to observed LIA cooling. 1.3. The Medieval Warm Period The so called Medieval Warm Period (MWP) was much more heterogeneous than the LIA, but is generally described as a warm interval preceding the LIA ( ca. 1000 1300 AD ) in which global temperatures were similar to the 20 th century (Crowley and Lowery, 2000). A network of borehole temperature estimates suggests that global temperatures from 500 1000 AD were warmer than 20 th century temperatures (Huang et al ., 1997), while large scale climate reconstructions vary significantly in their portrayal of the MWP (e.g. Mann et al., 1999 versus Esper et al., 2002). The MWP is often used as a benchmark for pre industrial warming, and is compared to the 20 th century t o make arguments for whether today's warming is caused by natural climate variability, or anthropogenic input of greenhouse gasses. For this reason it is important to improve upon our understanding of the spatial and temporal patterns of climate variabili ty during this time interval.
10 Evidence for the MWP is found throughout the globe, but it doesn't manifest exclusively as a warming event (for this reason, many studies now refer to this time interval as the "Medieval Climate Anomaly"). In the tropics a nd sub tropics there seems to be a large hydrologic response during the MWP. Droughts in the Yucatan associated with the MWP have been connected to the termination of the Classical Mayan Civilization (Hodell et al., 2005), while unprecedented warmth allowe d the Vikings to sustain successful agricultural based settlements in Greenland. The North American Monsoon was weakened during the MWP (900 1300 AD ) causing widespread drought conditions in western North America (Cook et al., 2004) (Figure 1.6). A recons truction of Atlantic tropical cyclone activity suggests that the MWP was a time of more frequent tropical cyclones, with tropical cyclone counts similar to the 20 th century (Mann et al., 2009). Figure 1.6. Long term aridity changes in the Western United States The figure shows the Drought Area Index (DAI) for the Western U.S. as reconstructed by tree rings, both annual in pale brown and 60 year low pass ltered in black. The red and blue lines are mean DAI for the MWP (ca. 900 1300 AD ) and the 20th centu ry out to 2003, respectively. This record shows that the MWP was much more arid on average than the 20th century. Figure from Cook et al. (2010).
11 1.4. Reconstructing SST over the past 2,000 years Understanding how ocean surface temperatures have vari ed over the past 2,000 years is critical for data model comparisons, and accurate prediction of future climate change. Nevertheless, paleoceanographers have traditionally neglected this important time interval. There are a number of factors that make obt aining high resolution SST records from the late Holocene difficult. Sedimentation rates are much too low in most marine depositional environments to resolve the past 2,000 years of climate variability. Marine environments with sedimentation rates high e nough to resolve decadal to centennial scale variability (i.e. >40cm/kyr) are limited to coastal margins near major river systems, drift sites, etc. Secondly, temperature variability during the late Holocene is subtle relative to the glacial interglacial shifts that paleoceanographers generally reconstruct. Attempting to resolve the small (<2C) temperature shifts of the most recent millennia is pushing the limits of existing SST proxies. In this section I will outline the advantages and limitations of e xisting geochemical proxies for reconstructing SST in the late Holocene. 1.4.1. Molecular Organic Proxies With the development of new HPLC MS (High Performance Liquid Chromatography Mass Spectrometry) techniques (Hopmans et al., 2000) as well as techniqu es for compound specific radiocarbon dating (see Eglinton and Eglinton, 2008 for a review), the field of molecular organic geochemistry is rapidly expanding in the realm of paleo SST reconstruction. The HPLC MS allows geochemists to easily identify, quant ify and isolate large polar compounds, which cannot be effectively analyzed via
12 traditional gas chromatographic (GC) techniques. This has led to the recent development of a new paleothermometer, the TEX 86 index, which is based on relative abundances of la rge 86 carbon polar lipids derived from marine Crenarchaeota. Compound specific radiocarbon dating has allowed for the independent dating of molecular components of sediments, thus enabling paleoceanographers to identify discrepancies in source region as well as age offsets between organic geochemical and foraminiferal based paleoceanographic records. Presently there are two widely accepted molecular organic SST proxies: the TEX 86 index and the U k' 37 index. The TEX 86 index is a paleothermometer based on the composition of membrane lipids called glycerol dialkyl glycerol tetraethers (GDGTs) found in marine Crenarchaeota. Schouten et al. (2002) discovered that the number of cyclopentane moieties on these GDGTs in sedimentary membrane lipids varies as a fu nction of local mean annual SST. It is thought that Crenarchaeota adjust the number of cyclopentane moieties in these GDGT lipids with temperature to regulate membrane fluidity. The latest calibration of Kim et al. (2008), which uses a much more extensiv e network of core top samples, confirms that TEX 86 correlates with mean annual SST, globally (Figure 1.7). The U k' 37 index is based on the ratio of di to tri unsaturated alkenones (Brassell et al., 1986; Prahl and Wakeham, 1987), which are 37 carbon compo unds produced as membrane lipids by haptophyte algae (coccolithophorids). The index varies between 0 and 1, which corresponds to a temperature range of 0 26C, and is generally considered to represent mean annual SST. Both the GDGT and alkenone lipids are contained within the fine fraction of sediments, and thus are subject to lateral transport, sometimes over large distances. The
13 development of compound specific radiocarbon dating has allowed for the investigation of the origin of specific molecular c omponents of sediment. The techniques for this involve isolating individual compounds using bench chemistry, preparative capillary gas chromatography (PCGC), or preparative HPLC. Alkenones can be isolated for radiocarbon dating using a series of Si gel c olumns, according to the techniques outlined in Ohkouchi et al. (2005). GDGTs have been isolated for radiocarbon dating using preparative HPLC (Shah et al., 2008). Results of radiocarbon dating of these two compound classes suggest that they have differ ent labilities. In other words, alkenones are relatively refractory, and can survive lateral transport over long distances (Englebrecht and Sachs, 2005). GDGTs, on the other hand, are relatively labile, and do not effectively survive transport ( Mollenhau er et al., 2007; Shah et al., 2008 ). The implications are that in some locations, the U k' 37 signal may be neither local, nor contemporaneous with the foraminiferal record. The TEX 86 signal, on the other hand, is much more likely to contain the signal of overlying waters, and have minimal age offset from the foraminiferal record. The TEX 86 proxy has both advantages and disadvantages over the U k' 37 and foraminiferal Mg/Ca proxies. Culture studies suggest that unlike alkenones, the composition of GDGTs does not seem to be effected by crenarchaeal growth rate (Wuchter et al., 2004). The same study tests the effect of salinity on TEX 86 and concludes that salinity has no influence on GDGT composition, while recent studies suggest a considerable salinity e ffect on Mg/Ca (e.g. Ferguson et al., 2008). TEX 86 is calibrated over a larger range of SSTs, and has been found to be an effective SST proxy in both the arctic (Sluijs et al., 2009) and the tropics (Tierney et al., 2008). TEX 86 has
14 even been used to mak e estimates of SST as far back as the Cretaceous (Jenkyns et al., 2004). Alkenones, on the other hand are not an effective SST proxy at SSTs greater than 27C (Pelejero and Calvo, 2003), as concentrations of tri unsaturated alkenone drop below detection l evels. Figure 1.7. Global Core top calibration of TEX 86 to SST The TEX 86 index from 223 core top samples, distributed globally, is calibrated to corresponding mean annual SST. Open circles represent Pacific and Indian Ocean sites, while crosses repres ent Atlantic Ocean sites. This results in the following equation: SST(C)= 10.78 + 56.2*TEX 86 from (Kim et al., 2008). Investigating the limitations of the TEX 86 proxy is currently an area of active research in the organic geochemistry community. One i ssue is the ubiquitous nature of Crenarchaeota in the marine environment. Although the TEX 86 signal consistently reflects mean annual SSTs, live crenarchaeal communities are living and producing GDGTs throughout the water column (Karner et al., 2001), and within sub surface
15 sediments ( Lipp et al., 2008 ). There is also the issue of terrestrial GDGTs. Branched GDGTs are produced by soil bacteria, and are abundant in terrestrial organic matter (TOM). In marine depositional basins proximal to major sources of terrestrial input, the TEX 86 could be biased by terrestrially derived GDGTs. The BIT index (branched to isoprenoid tetraether index) was developed by Hopmans et al. (2004), and is used as a proxy for TOM input. Samples with BIT values greater than 0.3 are considered suspect for bias in the TEX 86 record (Weijers et al., 2006). 1.4.2. Foraminiferal based proxies T he use of the stable oxygen isotope composition of planktonic Foraminifera to reconstruct paleotemperatures was pioneered in the mid 20 th cen tury by Urey (1947) and Emiliani (1 954, 1955). Now the 18 O of planktonic foraminifera is one of the most commonly used paleoceanographic tools (see Lea, 2003 for a review). However, the 18 O of foraminiferal calcite is controlled not only by calcification temperature, but also by the oxy gen isotopic composition of ambient seawater. Beginning in the mid 1990's, paleoceanographers began to use the Mg/Ca ratio of foraminiferal calcite as an SST proxy (e.g. Nrnberg et al., 1996; Rosenthal et al., 1997; Hastings et al., 1998; Lea et al., 19 99; Elderfield and Ganssen, 2000). The use of Mg/Ca as an SST proxy is based on the fact that Mg 2+ substitutes for Ca 2+ in the calcite lattice, with an exponential temperature dependence. The exact Mg/Ca SST calibration varies slightly among different sp ecies of Foraminifera, but general equates to a ~9% increase in Mg/Ca per 1C in water temperature (Anand et al., 2003).
16 Paired Mg/Ca and 18 O measurements can be used to derive records of both temperature and the oxygen isotopic composition of seawat er (Figure 1.8) if the assumption is made that the Mg/Ca of foraminiferal calcite is controlled exclusively by temperature (see discussion of caveats below). Then, changes in 18 O of seawater can then be interpreted in terms of salinity variability. Thus a major advantage of paired Mg/Ca and 18 O measurements on Foraminifera is the ability to generate a record of SST and salinity variability from a single archive. A number of studies over the past decade have suggested that there are additional physical parameters that can affect foraminiferal Mg/Ca, thus complicating interpretation of Mg/Ca records strictly in terms of SST. Additionally, it has been shown that post depositional diagenetic processes can have a large impact on the elemental composition of foraminiferal calcite. Figure 1.8. Flow Chart of Paired Mg/Ca and 18 O for estimating SST and 18 O of seawater. The equation for converting Mg/Ca to SST is the equation for white G. ruber (Anand et al., 2003). The 18 O paleotemperature equation is f rom Bemis et al. (1998).
17 The Mg/Ca ratio in seawater is spatially constant and unlikely to change on timescales of less than 1 million years due to the very long residence times of both Mg and Ca in the oceans (Broecker and Peng, 1982). There are a few fa ctors that complicate the use of Mg/Ca as a paleotemperature proxy, however. The post mortem addition of diagenetic phases to foraminiferal calcite was recognized by Boyle (1983). Specifically, the addition of Mn rich carbonate overgrowths (as well and M n and Fe rich oxyhydroxides) has been a problem for trace metal analysis (Cd/Ca, Ba/Ca, etc.), because these phases also contain other elements that alter the original shell chemistry. One of these phases that has been demonstrated to dramatically effect shell Mg/Ca is the carbonate mineral kutnahorite. Pena et al. (2005) showed that the presence of this phase biased foraminiferal Mg/Ca to 7 36% higher values (equivalent to a 0.9 6.2C temperature bias). In order to remove these potentially problematic diagenetic carbonate phases, a reductive cleaning step is used (Barker et al., 2003). The reductive cleaning step can also preferentially remove Mg from the primary calcite, thus reducing shell Mg/Ca by up to 15% (Rosenthal et al., 2004). Precautions nee d to be taken to monitor elements associated with diagenetic alteration (i.e. Mn and Fe) when performing Mg/Ca analyses. Also, care must be taken when comparing Mg/Ca records in which different cleaning methods were used, as there may be systematic offset s in the temperature estimates due to cleaning. Selective dissolution of high Mg carbonate, as the foraminifera are affected by calcite undersaturated deep waters and porewaters during sedimentation can lead to alteration of the original Mg/Ca signal ( e.g. Dekens et al., 2002). It has been noted in a number of studies that dissolution can lower foraminiferal Mg/Ca in sediments
18 ( Rosenthal et al., 1993; Russell et al., 1994; Brown and Elderfield, 1996; Hastings et al., 1998). Waters are caustic to carb onates when the in situ concentration of carbonate ion is less than the saturation concentration of carbonate ion. i.e. CO 3 2 = [ CO 3 2 ] in situ [ CO 3 2 ] saturation (Berger et al., 1982) Calcite that is more enriched in Mg is more susceptible to di ssolution, and high Mg calcite can be susceptible to dissolution well above the lysocline (Brown and Elderfield, 1996). Factors that can affect the degree of dissolution include the water depth, the age of the deep waters, and the concentration of organi c matter in sediments. Dekens et al. (2002) have provided a depth correction for the Mg/Ca paleotemperature calibration in the foraminifer, Globigerinoides ruber in order to correct for the dissolution effect on Mg/Ca. Laboratory culture studies have sho wn additional physical parameters (e.g. pH, carbonate ion concentration and salinity of seawater) act as controls on Mg/Ca ratios in Foraminifera, but suggest that their influence is small in comparison with temperature (Lea et al., 1999; Nrnberg et al., 1996; Russell et al., 2004; Elderfield et al., 2006). In a laboratory culture study, Lea et al. (1999) looked at the effect of salinity on entire test Mg/Ca ratios, and found a small increase with salinity of 43% per psu in Orbulina universa Another st udy of the final chambers of Globigerinoides sacculifer grown over a range of salinities (26 44 psu) showed increases of Mg/Ca of over 100% at higher salinity, or approximately 11% per psu (Nrnberg et al.,1996). A field study in the Mediterannean found a significant relationship between Mg/Ca and calcification salinity, in which Mg/Ca increased by 15 59% per psu (Ferguson et al., 2008). This study was conducted over a salinity range of 36 40 psu, which is significantly higher than most
19 open ocean setting s, and may not be applicable at lower salinities. Each of the caveats discussed in this section, for both molecular organic and foraminiferal geochemistry, are addressed in the dissertation as they are specifically related to my study sites in the Gulf of Mexico. 1.5. Dissertation Organization This main body of this dissertation is organized into three chapters, which are written in the format of separate manuscripts for peer reviewed journals. Figures are, for the most part, identical to those used i n the publication manuscripts. Therefore there may be some redundancy with respect to certain data sets being plotted more than once. References from all 3 manuscripts are combined at the end of this dissertation. In Chapter 2: Regionally Coherent Litt le Ice Age Cooling in the Atlantic Warm Pool In this chapter I present two new Mg/Ca SST records from the Fisk and Garrison Basins in the northern Gulf of Mexico. The aim of this study was to test the regional reproducibility of a Mg/Ca SST record from the Pigmy Basin (GOM) that was published as part of my masters thesis (Richey et al., 2007). A large (~2C) cooling during the LIA was present in all three SST records, and comparison with other records within the Atlantic Warm Pool (AWP) suggest that thi s timing and magnitude of LIA cooling was consistent throughout the sub tropical and tropical western Atlantic Ocean. Potential mechanisms for this regional cooling are proposed. This study was published in the journal, Geophysical Research Letters in 20 09:
20 Richey, J. N., R. Z. Poore, B. P. Flower, and T. M. Quinn, and D. J. Hollander (2009), Regionally coherent Little Ice Age cooling in the Atlantic Warm Pool, Geophys. Res.Lett., 36, L21703, doi:10.1029/2009GL040445. In Chapter 3: Merging late Hol ocene molecular Organic and Foraminiferal Based Geochemical 2 Records of SST in the Gulf of Mexico In this chapter I used a molecular organic geochemical approach (TEX 86 ) to generate an additional SST record from the Pigmy Basin, Gulf of Mexico. The TEX 86 SST record is remarkably similar to the Mg/Ca SST from the same core, despite the fact that the two proxies are subject to separate diagenetic processes and are contained within different sediment fractions. The relative seasonal and depth distribution s of the Globigerinoides ruber and marine Crenarchaeota are rigorously assessed for the northern Gulf of Mexico, and those relative differences are exploited to make inferences about changing mixed layer depth and seasonality over the past 1,000 years. Th is study is currently under review at the journal, Paleoceanography : Richey, J. N., D. J. Hollander, B. P. Flower and T. I. Eglinton (2010), Merging late Holocene molecular Organic and Foraminiferal Based Geochemical 2 Records of SST in the Gulf of Mex ico, Paleoceanography in review. In Chapter 4: Ecological controls on the shell geochemistry of pink and white Globigerinoides ruber in the northern Gulf of Mexico: Implications for paleoceanographic reconstruction In this chapter I examine the relat ionship between foraminiferal test size and shell
21 geochemistry ( 13 C, 18 O and Mg/Ca) for the pink and white sub species of Globigerinoides ruber These data provide insights into ecological and metabolic controls on shell geochemistry, allowing for bette r constraints on paleoceanographic data. We conclude that there is a significant positive relationship between size and 13 C, which is most likely related to growth rate. A significant decrease in 18 O and increase in Mg/Ca with size, suggests that large r individuals have higher calcification temperatures than smaller individuals. High resolution down core comparisons of Mg/Ca and 18 O data for pink and white G. ruber are made in order to assess whether the two planktonic species have distinct seasonal d istributions in the Gulf of Mexico. I conclude that the pink G. ruber signal is summer weighted, while the white G. ruber signal represents mean annual surface conditions. The results of this study are being prepared for submission to the journal, Marine Micropaleontology Richey, J. N., R. Z. Poore, D. J. Hollander, and B. P. Flower, (2010), Ecological controls on the shell geochemistry of pink and white Globigerinoides ruber in the northern Gulf of Mexico: Implications for paleoceanographic reconstr uction, Marine Micropaleontology in preparation.
22 Chapter 2 Regionally Coherent Little Ice Age Cooling in the Atlantic Warm pool 2.1. Abstract We present 2 new decadal resolution foraminiferal Mg/Ca SST records covering the past 6 8 ce nturies from the northern Gulf of Mexico (GOM). These records provide evidence for a Little Ice Age (LIA) cooling of 2 3C, consistent with a published Mg/Ca record from Pigmy Basin. Each of the GOM basins exhibits SST minima within the Dalton, Maunder a nd Sprer sunspot minima, with a general warming trend over the past 150 years. Comparison of these 3 records with existing SST proxy records from the GOM Caribbean region show that the magnitude of LIA cooling in the Atlantic Warm Pool (AWP) was significa ntly larger than the mean hemispheric cooling of <1C. We propose that a reduction in the intensity and spatial extent of the AWP during the LIA, combined with associated changes in atmospheric circulation may account for the regional SST patterns observe d in the GOM Caribbean region during the LIA.
23 2.2. Introduction Accurate reconstruction of high resolution sea surface temperature (SST) records during time intervals of societal importance, such as the Little Ice Age (LIA), and through the 20 th c entury, is important in determining the magnitude of pre industrial climate variability. Given the uncertainties inherent to most SST proxies, as well as the influence of local climatology, replication of SST records is critical to understanding regional responses to climate forcings. The LIA generally spans the time interval from 1400 1850 AD, although the timing and magnitude of cooling varies widely throughout the Northern Hemisphere Temperature reconstructions (primarily based on extratropical terres trial proxy records) suggest that the Northern Hemisphere experienced modest cooling of 0.6 0.8C during the 15 th 19 th centuries (Mann et al., 1998, 1999; Esper et al., 2002, Moberg et al., 2005. Here we show that SST proxy records from the low latitude N orth Atlantic Ocean experienced significantly larger cooling than the hemispheric average, and may be have been particularly sensitive to climate perturbations on multi decadal to centennial timescales during the LIA. The Atlantic Warm Pool (AWP), defined by the >28.5C SST isotherm, develops annually in the northern Caribbean during early summer (June) and expands into the GOM and western tropical North Atlantic through the late summer (July October) (cf. Wang et al., 2008a). Multidecadal variability in the size of the AWP is correlated with rainfall anomalies in the Caribbean region, formation and intensification of North Atlantic hurricanes, and variability in moisture transport to the North American continent via interactions with atmospheric circulati on (Wang et al., 2008a). The geographic area covered by an anomalously large AWP can be 3 times larger than an anomalously small
24 AWP (Wang et al., 2006), thereby altering the SST and atmospheric circulation patterns in the GOM, Caribbean and western tropi cal North Atlantic. A number of geochemical proxy records from corals, sclerosponges and foraminifera in the region encompassed by the AWP show a 1 3C cooling during the LIA ( Winter et al., 2000; Watanabe et al., 2001; Nyberg et al., 2002; Haase Schramm et al., 2003, 2005; Lund and Curry, 2006; Richey et al., 2007; Black et al., 2008; Kilbourne et al., 2008) (Figure 2.1), implying that the AWP may have been particularly sensitive to climate forcing during the LIA. Figure 2.1. Map of proxy records in the GOM Caribbean region exhibiting 1 3C cooling during the LIA The Fisk (open square) and Garrison (closed square) basins are the 2 new Mg/Ca SST records presented in this study. The September (maximum seasonal geographic extent) AWP (28.5C isotherm) is p lotted using the Reynolds and Smith OISST V2.0 dataset (1x1 grid, averaged from 1981 2009). Mean LIA cooling is indicated in parentheses for each region.
25 Assessing the fidelity of these SST proxy records is especially critical for the interpretation of decadal to sub decadal resolution low latitude records covering critical time intervals such as the LIA 20th century. The uncertainties inherent to foraminiferal Mg/Ca based SST estimates can exceed the environmental signal in some cases. Factors that le ad to these uncertainties include, but are not limited to, diagenetic overgrowths (e.g. Boyle 1983; Barker et al., 2003; Pena et al., 2005), salinity (e.g. Nrnberg et al., 1996; Ferguson et al., 2008), and dissolution (e.g. Dekens et al., 2003). Each of these factors has the potential to overprint the Mg/Ca signal of the downcore record, and the influence of these factors is variable, and often basin specific. Thus, replication among a grouping of regional cores is essential to developing a coherent regi onal record of climate variability. In this paper we present 2 new foraminiferal Mg/Ca SST records spanning the past 600 800 years from the northern Gulf of Mexico. These new records replicate the magnitude and pattern of SST variability recorded in a pu blished Mg/Ca record from the Pigmy Basin (Richey et al., 2007), and further corroborate a large magnitude (1 3C) cooling in the GOM Caribbean region during the LIA. We highlight the regional coherence among all published Caribbean Gulf of Mexico SST pro xy records during this time interval, and discuss potential mechanisms for this large low latitude Atlantic cooling overprinted on the modest hemispheric cooling during the LIA. 2.3. Materials and Methods The Fisk Basin (PE07 5I; 817 m depth; 2733 .0' N, 9210.1' W) and Garrison Basin (PE07 2; 1570 m depth; 2640.5' N, 9355.5' W) box cores were collected onboard the R/V Pelican in 2007. Pigmy Basin box core (PBBC 1; 2259 m depth; 2711.61'N,
26 9124.54'W) was collected in 2002 onboard the R/V Longho rn. All of these basins are located on the continental slope in the northern Gulf of Mexico, and have a relatively high sediment accumulation rate (20 40 cm/kyr) due to large inputs of terrigenous material via the Mississippi River. For each of the box c ores the sediment water interface was recovered, and AMS 14 C dates with bomb radiocarbon confirm that the core top samples include the most recently deposited sediments. We set the core top age to 2000 AD for each of these GOM records for ease of comparis on with other regional, absolutely dated records. Radiocarbon ages (Table 2.1) were calibrated using the Calib 5.0 program with a 400 year reservoir correction. Radiocarbon AMS dates were determined using 6 8mg of mixed planktic foramnifers. The Fisk B asin age model (Figure 2.1a) was constructed by fitting a least squares regression line through the 4 radiocarbon AMS dates and setting the intercept to 0. Radiocarbon dates below 20 cm core depth indicate much lower sediment accumulation rates prior to 80 0 yrs BP. Due to the uncertainties in the age model below 20 cm, we do not plot the Mg/Ca data in this paper. The linear sedimentation rate was determined to be [Cal Age (years BP)=3.7602*core depth (mm)]. The Garrison Basin age model (Figure 2.1b) was constructed by fitting a third order polynomial through the 8 radiocarbon AMS dates [Cal Age (Yrs BP)= 2e 05 x 3 + 0.022x 2 + 2.26x; x= core depth (mm)]. An 18% decrease in the foraminiferal weights below 13 cm in the Garrison Basin core indicates a potenti al problem with calcite dissolution, and thus we exclude Mg/Ca data below this core depth.
27 Core Depth (mm) 14 C age (yrs) error (yrs) Cal Yrs BP Fisk Basin 0 5 215 30 0 150 155 925 35 523 160 165 1090 30 652 200 205 1210 30 743 Garrison Basin 0 5 140 35 0 10 15 455 25 41 70 75 935 35 186 120 125 1435 45 604 200 205 2080 20 1248 270 275 2495 35 1666 340 345 3300 35 2695 390 395 3640 30 3064 Table 2.1. Radiocarbon dates. 6 AMS radiocarbon dates are shown for Fisk Basin and 4 dates for Garrison Basin. The error column indicates the analytical error on the 14 C age. The radiocarbon ages were converted to calendar ears using the CALIB 5.0 program, with a 400 year reservoir correction. Figure 2.1. Age models a) Fisk Basin age model and b) Garrison Basin age model. For the Fisk Basin, a least squares regression is fit through the 4 AMS radiocarbon dates. For the Garrison Basin a second order polynomial is fit through the 7 AMS radiocarbon dates.
28 For all down core elemental analyse s "60 foraminifera were picked from the 250 300mm size fraction of Globigerinoides ruber (white). Foraminifera were lightly crushed and underwent a cleaning process that includes multiple clay removal steps, an oxidative step to remove organic material, a nd an acid leaching step to remove adsorbed metals (Barker et al., 2003). A reductive cleaning step was not performed on these samples. Elemental analyses were performed on a Perkin Elmer Optima 4300 dual view inductively coupled plasma optical emission spectrometer (ICP OES) at the College of Marine Science, U niversity of South Florida. In all cases where there are replicate Mg/Ca analyses, replicates are based on aliquots of "60 foraminifera that have been crushed, cleaned and analyzed separately. 2.4. Gulf of Mexico Mg/Ca records In order to test the rep roducibility of the Pigmy Basin Mg/Ca record (Richey et al., 2007), we generated Mg/Ca records in 2 additional Gulf of Mexico basins: Garrison Basin (box core PE07 2) and Fisk Basin (box core PE07 5I). The upper 13 cm of the Garrison Basin box core cover s the past ~600 yrs (age control provided by 4 AMS radiocarbon date, see supplemental materials). An 18% decrease in the foraminiferal weights below 13 cm indicates potential problems with calcite dissolution, and thus we exclude Mg/Ca data below this cor e depth. The core top Mg/Ca value is 4.43 mmol/mol ( 0.16 mmol/mol), based on 2 replicate measurements, and corresponds to an SST of 25.4C (using [Mg/Ca=0.449*exp(0.09*SST)], from Anand et al., 2003), the modern annual average for the Gulf of Mexico (Lev itus, 2003). This is equivalent to the core top Mg/Ca value of 4.43 ( 0.03 mmol/mol) that was generated from replicate measurements
29 of 3 different sub cores from the Pigmy Basin box core (Richey et al., 2007). The mean precision for replicate analyses o f the Garrison Basin downcore record is 0.14 mmol/mol ( 0.3C), with 60% of the samples run in duplicate. The major features of this record include 3 distinct SST minima (ca. 1450 1550, 1700 1750, and 1900 AD), that are ~2C cooler than the core top SST. These minima appear to correspond with the Sprer, Maunder, and Dalton sunspot minima, respectively (Figure 2.3). Figure 2.3. Gulf of Mexico Mg/Ca Records (a.) Garrison Basin (b.) Fisk Basin and (c.) Pigmy Basin (from Richey et al., 2007) are plotted on the same Mg/Ca scale, with age control points indicated by arrows. Corresponding SST scale is given on a secondary y axis, using the relationship [Mg/Ca=0.449*exp(0.09*SST], from Anand et al. (2003). Lines are plotted on each curve representing the lin ear warming trend over the past 250 years.
30 The upper 20 cm of the Fisk Basin box core span the past ~800 years (age control provided by 4 AMS radiocarbon dates). Radiocarbon dates below 20 cm core depth indicate much lower sediment accumulation rates pri or to 800 yrs BP. Due to the uncertainties in the age model below 20 cm, we focus on the uppermost 20 cm of the box core, which has a sedimentation rate of ~26.5cm/kyr (sampling resolution of ~18 yrs per 0.5 cm sample). The core top Mg/Ca value is 4.75 m mol/mol ( 0.17 mmol/mol), based on 3 measurements, which corresponds to an SST of 26.2C ( 0.4C), and is slightly higher (by 0.8 0.4C) than the core top Mg/Ca SST for Pigmy and Garrison Basins. The Fisk Basin Mg/Ca record shows a similar pattern of vari ability to the other 2 Gulf of Mexico records over the past 6 centuries, with SST minima ca. 1550 and 1750 1850 AD that are ~3C cooler than the core top SST (Figure 2.3). LIA cooling in all 3 GOM Mg/Ca records is preceded by an interval of warmth in wh ich Mg/Ca is as high or higher than the mean core top value of 4.4 mmol/mol. The timing of the warm interval in the Pigmy and Fisk basins is similar (ca. 1500 and 1450 AD, respectively), while it is slightly later in Garrison Basin (~1600 AD). All 3 bas ins reach maximum cooling ca. 1750 AD. The linear warming trend from maximum LIA cooling (1750 AD) to the core top is similar in the Pigmy and Garrison basins (~0.007C/yr), while the slope of the warming trend is slightly steeper in Fisk Basin (Figure 2. 3). In actuality however, given the uncertainty of the age models and Mg/Ca SST estimates, the timing of the onset as well as the magnitude of LIA cooling is consistent among these 3 GOM sites.
31 2.5. Regional Comparisons We have generated a 600 year s tacked SST record for the northern GOM based on the Fisk, Garrison and Pigmy Basin Mg/Ca SST records (Fig. 2.4d). In order to generate the stacked SST record, the mean was removed from each record for the period in which there is overlap between the 3 r ecords (from 2000 1420 AD), (i.e. SST is the SST relative to the 580 year mean). The SST records were re sampled at a constant T of 20 years, and then the mean of the 3 re sampled SST records was calculated to generate the GOM SST stack. The uncerta inty in the stack is 0.4C (indicated by the error bar in Figure 2.4d.). This stack represents the multi centennial trend that is common to the 3 independent GOM SST records, and is used here to draw regional comparisons. We compare the GOM SST stack to a 250 year continuous coral Sr/Ca SST record from the species Montastraea faveolata from La Parguera, Puerto Rico (Kilbourne et al., 2008). Both records (plotted their own independent timescale, and calibration to SST) show that it was ~2C cooler ca. 1 750 AD than modern, and they also can both be described by a linear warming trend of 0.007C/yr from the LIA toward the present (Figure 2.4, a and d). There are 2 additional coral based SST records from Puerto Rico that compare brief time intervals during the LIA to late 20 th century SSTs. Winter et al. (2000) infer that LIA SSTs were 2 3C cooler than modern while Watanabe et al. (2001) suggest that LIA SSTs were 2C cooler than modern (from 18 O and Mg/Ca data in M. faveolata respectively). In summary, these 3 different Puerto Rico coral based geochemical proxies agree that early 18 th century SSTs were 2 3C cooler than modern.
32 Figure 2.4. AWP Regional SST Comparison. (a) La Parguera, Puerto Rico coral Sr/Ca SST (Kilbourne et al., 2008). The coral Sr /Ca is calibrated to SST using the equation [Sr/Ca= 0.047*T+10.3726], and the plot is a 5 year running mean. (b) Montego Bay, Jamaica sclerosponge Sr/Ca SST (Haase Schramm et al., 2003). The sclerosponge Sr/Ca data was calibrated to SST using the equation [Sr/Ca= 0.102*T+12.5] from Rosenheim et al. (2004). The record was re sampled at a constant T of 4 years, and then smoothed using a 5 point running mean (c) Cariaco Basin G. bulloides Mg/Ca SST calibrated to SST using the equation Mg/Ca=0.0.368exp(0.092* T), (d ) Gulf of Mexico #SST stack (solid line) and (e) Pigmy Basin, GOM foraminiferal Mg/Ca SST (dashed line)(Richey et al., 2007). Each record is plotted on its own independent timescale, and SST is scaled identically in each panel. The error bar in panel (d) i ndicates the uncertainty in the GOM SST stack of 0.4C.
33 A Sr/Ca SST record spanning the past 650 years from a Jamaican sclerosponge shows mean LIA conditions that were ~3C cooler than modern (Haase Schramm et al., 2003). Although the magnitude of coo ling inferred from this sclerosponge record is slightly larger than that observed in other circum Caribbean SST records, the general pattern of variability is very similar to other regional records (Figure 2.4b). A Cariaco Basin foraminiferal Mg/Ca SST record shows a similar pattern of centennial scale variability to the GOM Mg/Ca SST records over the past 8 centuries (Black et al., 2007). Both show a period of warmth ca. 1500 AD., during which SSTs are similar to core top SSTs, followed by a period of s ignificant cooling during the LIA, and a rapid warming over the past 100 years (Figure 2.4c). The amplitude of SST variability reported in Black et al. (2007) is muted relative to GOM SST variability; however, the use of an alternate Mg/Ca SST calibration equation yields an SST record that exhibits a ~2C LIA cooling, and an amplitude of variability that is similar to that of the GOM SST records. The similarities between the SST proxy records that span the GOM Caribbean region, in spite of the uncertainti es in both the Mg/Ca and Sr/Ca temperature proxies, suggests that the trends recorded in the northern GOM over that past 6 centuries are representative of a regional climate signal. 2.6. Discussion There are important implications of a regionally coher ent >2C cooling in the Gulf of Mexico Caribbean region during the LIA. Northern Hemisphere temperature reconstructions, which are based predominantly on mid to high latitude records, show a
34 <1C hemispheric cooling (Mann et al., 1998, 1999; Esper et al ., 2002; Moberg et al., 2005). Model results using both medium and high solar radiative forcing estimate Northern Hemisphere LIA cooling of <1C (Ammann et al., 2007). Simulated Northern Hemisphere temperature response to volcanic forcing shows cooling o f 0.2 0.5C, and is in good agreement with Northern Hemisphere temperature reconstructions (Gao et al., 2008). While model simulations and temperature reconstructions suggest a <1C LIA cooling for the Northern Hemisphere, proxy records estimate that cool ing in the high northern latitudes is on the order of 1 3C (Overpeck et al., 1997). Based on the concept of polar amplification, one would predict much more subtle temperature changes in the subtropical Atlantic Ocean than at the high northern latitudes. We suggest, based on the weight of evidence, that there was a large cooling (1 3C) in the GOM Caribbean region during the LIA (ca. 150 400 yrs BP), indicating that this particular region of the sub tropical Atlantic Ocean was especially sensitive to cli mate perturbations during this time interval. Timing of local SST minima in existing high resolution continuous records from the GOM Caribbean correspond roughly to minima in solar insolation associated with sunspot minima (The Dalton, Maunder and Spre r Minima). The reduction in solar irradiance (0.25 0.65%) attributed to these sunspot minima indicates a very small change in radiative forcing (Bard et al., 2000). A model simulation by Ammann et al. (2007) shows that the low latitude North Atlantic has a relatively low sensitivity to solar forcing (<0.05C/watt m 2 ) compared with the mid to high latitudes. Modeled surface air temperature response to irradiance changes, similar to the decrease in irradiance associated with the Maunder Minimum, show a $T of ~0.3C in the GOM Caribbean
35 region (Shindell et al., 2006), which is much smaller than the observed SST variability from proxy records in this region during these solar minima. Thus direct solar forcing alone is not likely responsible for the large LIA cooling in the low latitude North Atlantic, but perhaps positi ve feedbacks (e.g. intensification of the subtropical high) amplified cooling during periods of reduced solar irradiance. The observed LIA cooling in the Caribbean and northern GOM may have been driven by changes in the size of the AWP, which can vary b y 50% (Wang et al., 2008b) and the SST anomaly within the AWP can vary by 0.6C (Wang et al., 2006) on multidecadal timescales. During the LIA, it is possible that there was a dramatic reduction in the geographic extent and intensity of the AWP, thus re ducing summer SSTs in regions on the periphery of the AWP (e.g. Puerto Rico and the northern GOM) for prolonged time intervals. Model results of Wang et al. (2008a) suggest that an anomalously small AWP, coupled with associated changes in atmospheric circ ulation can lead to an increase in the mid summer drought in the Central America/Yucatan region, and an increase in moisture transport from the GOM to the North American continent via a strengthening of the Caribbean low level jet (CLLJ) and the Great Plai ns low level jet (GPLLJ). Proxy records of hydrologic variability suggest that this was the likely regional climate scenario during the LIA. A GOM record of terrigenous input (via the Mississippi River) suggests wetter conditions in North American during the LIA (Flannery et al., 2008). Records from the Yucatan Peninsula suggest drier conditions during the LIA (Hodell et al., 2005), while bulk d 18 O from the Blue Hole in Belize suggest drier and/or cooler conditions in Central America (Gischler et al., 20 08). An increase in salinity (inferred from an increase in the d 18 O of seawater) in the GOM (Richey et al., 2007) and
36 Florida Current (Lund and Curry, 2006) during the LIA further supports increased evaporation minus precipitation in the GOM Caribbean. T he LIA cooling in the low latitude North Atlantic is consistent with an increase in the north south sea level pressure (SLP) gradient associated with the positive phase of the North Atlantic Oscillation (NAO). Increased levels of Na + and K + in a glacioche mical series from the GISP2 ice core suggest multiple intervals of increased north Atlantic storminess during the LIA (Meeker and Mayewski, 2002), which suggests an increase in the pressure gradient between the Icelandic Low and the North Atlantic subtropi cal high (Maasch et al., 2005). This positive NOA like pattern is characterized by an increase in North Atlantic trade wind strength, and a cooling in northern hemisphere tropical and subtropical SSTs (Marshall et al., 2001). A centennial scale strengthe ning of the trade winds is consistent with evidence for cooler and drier conditions observed throughout the Gulf of Mexico Caribbean region. Although a recent reconstruction of the NAO suggests a shift to weaker NAO conditions during the LIA (Trouet et al ., 2009), evidence from the subtropical North Atlantic is consistent with a persistent, enhanced positive NAO pattern of SLP and SST during the LIA. 2.7. Conclusions Despite uncertainties in foraminiferal Mg/Ca SST proxy data and radiocarbon dating, th e 3 late Holocene Mg/Ca SST records generated from the northern Gulf of Mexico show very similar variability over the past 6 centuries, corroborating observations from throughout the Gulf of Mexico Caribbean region that there was a prominent Little Ice Age cooling of 1 3C. This suggests that the tropical subtropical
37 North Atlantic may be more dynamic than previously thought during the late Holocene. A reduction of the AWP, coupled with reorganization of atmospheric circulation patterns during the Little Ice Age, may explain the observed cooling in this region. Additional high quality SST proxy records from the subtropical North Atlantic Ocean are needed to establish the spatial extent and timing of this Little Ice Age cooling. Additional terrestrial pro xy records of regional hydrologic variability will aid in understanding the ocean atmosphere dynamics during this climatically important interval. Models including solar and volcanic forcings during the LIA have not been able produce a >1C cooling in the GOM Caribbean region, thus more work needs to be done to better understand the regional climate dynamics that could lead the observed cooling.
38 Chapter 3 Merging late Holocene molecular Organic and Foraminiferal based Geochemical R ecords of SST in the Gulf of Mexico 3.1. Abstract A molecular organic geochemical proxy (TEX 86 ) for sea surface temperature (SST) is compared with a Foraminifera based SST proxy (Mg/Ca) in a decadal resolution marine sedimentary record spanning the last 1,000 years from the Gulf of Mexico (GOM). We assess the relative strengths of the organic and inorganic paleoceanographic techniques for reconstructing high resolution SST variability during recent climate events, including the Little Ice Age (LIA) and t he Medieval Warm Period (MWP). SST estimates based on the molecular organic proxy TEX 86 show a similar magnitude and pattern of SST variability to foraminiferal Mg/Ca SST estimates, but with some important differences. For instance, both proxies show a s ignificant cooling (1.5 2.5C) of GOM SSTs during the LIA. During the MWP, however, Mg/Ca SSTs are similar to near modern SSTs, while TEX 86 indicates SSTs that were significantly cooler than core top. Using the respective SST calibrations for each proxy results in TEX 86 SST estimates that are 2 to 4¡C warmer than Mg/Ca SST throughout the 1,000 year record. We interpret the TEX 86 SST as a summer weighted SST signal from the upper mixed layer, whereas the Mg/Ca SST better reflects the mean annual SST Dif ferences in the SST estimates between the two proxies are interpreted in the context of varying seasonality and/or changing water column temperature gradients.
39 3.2. Introduction Accurate reconstruction of the spatial and temporal patterns of sea surface temperature (SST) variability in the low latitude oceans is critical to understanding the range of natural climate variability over the past millennium. While the low latitude oceans are a major source of heat and moisture to the mid and high latitudes, global and hemispheric climate reconstructions are predominantly based on extra tropical terrestrial proxy records (c.f. National Research Council 2006). There are few locations in the low latitude oceans from which continuous decadal resolution SST prox y records covering the past 1,000 years have been published ( Lund and Curry, 2006; Newton et al., 2006; Richey et al., 2007; Black et al., 2008; and Oppo et al., 2009 ). These SST reconstructions, although widely distributed geographically, all show signif icant SST fluctuations (1 2C) over the past 1,000 years. Paleoclimate records from the Atlantic Warm Pool (AWP), which includes much of the western tropical/sub tropical Atlantic Ocean, provide further evidence for significant climate fluctuations over t he past 6 centuries (c.f. Richey et al., 2009). The AWP is defined by the >28.5C SST isotherm, and encompasses the northern Caribbean, GOM and western tropical North Atlantic during the summer (cf. Wang et al., 2008). Multi decadal variability in the siz e/intensity of the AWP is correlated with rainfall anomalies in the Caribbean region, formation and intensification of North Atlantic hurricanes, and variability in moisture transport to the North American continent via interactions with atmospheric circul ation ( Wang et al., 2008). A number of geochemical proxy records from the region encompassed by the AWP provide evidence for a large (2 3C) cooling during the LIA (ca. 400 150 yrs BP)
40 ( Winter et al., 2000; Watanabe et al., 2001; Nyberg et al., 2002; Ha ase Schramm et al., 2003; Richey et al., 2007; Black et al., 2007; Kilbourne et al., 2008; Richey et al., 2009 ). However, two additional SST proxy records from within the AWP show a subtler LIA cooling (~1C): A Mg/Ca SST record from a sediment core from the Great Bahamas Bank ( Lund and Curry 2006), and a coral growth rate based SST record from the Bahamas ( Saenger et al., 2009). Additional multi proxy studies are needed to determine whether the discrepancies observed are real regional differences in th e climate response during the LIA, or whether there are site and proxy specific factors that are influencing the SST records. Presently, there are three widely used SST proxies derived from marine sedimentary archives: U K' 37 TEX 86 and foraminiferal Mg/ Ca. The U K' 37 index, an organic geochemical SST proxy based on the ratio of long chain diunsaturated to triunsaturated alkenones ( Brassell et al ., 1986), is not ideal in low latitude marine settings where SSTs exceed 28C. The U K' 37 index approaches 1.0 at temperatures > 28C ( Prahl and Wakeham, 1987) due to insufficient production of the triunsaturated alkenones. Although Jasper and Gagosian (1989) generated a low resolution 100 kyr U K' 37 SST record from Pigmy Basin sediments, there are insufficient al kenone concentrations in late Holocene sediments to provide an SST record of the past 1,000 years. A novel molecular organic SST proxy, TEX 86 (the TetraEther IndeX of tetraethers with 86 carbon atoms), is based on the relative abundance of isoprenoid glyc erol dialkyl glycerol tetraethers (GDGTs) with varying numbers of cyclopentane moieties ( Schouten et al., 200 2). GDGTs are membrane lipids biosynthesized by marine Crenarchaeota, and the number of cyclopentane moieties in these crenarchaeotal membrane lipi ds has been
41 observed to increase with increasing growth temperature ( Wuchter et al., 2004). Although Crenarchaeota are ubiquitous throughout the water column (Karner et al., 2001) and have been found living deep within sub surface sediments ( Lipp et al ., 2 008), TEX 86 has been shown to correlate linearly with mean annual SST ( Schouten et al., 2002, Kim et al., 2008) throughout the global oceans. Mesocosm studies suggest that salinity and nutrients are not significant factors in TEX 86 measurements ( Wuchter et al 2004). It has been shown that terrestrial organic matter often contains isoprenoid GDGTs ( Weijers et al., 2006), and thus TEX 86 SST estimates may be biased in marine settings with large terrestrial input. However, the contribution of terrestrial r elative to marine GDGTs can be monitored downcore via the BIT (Branched to Isoprenoid Tetraether) index ( Hopmans et al. 2004). Foraminiferal Mg/Ca has been widely accepted as an SST proxy, however factors such as diagenetic overgrowths (e.g. Boyle, 1983; Barker et al., 2003; Pena et al., 2005), salinity ( e.g. Nrnberg et al., 1996; Ferguson et al., 2008) and dissolution (e.g. Dekens et al., 2003) have been shown to influence Mg/Ca of foraminiferal calcite. In this study we present the first direct compar ison of foraminiferal Mg/Ca with a molecular organic (TEX 86 ) SST proxy, from co occurring sediments in a decadally resolved 1,000 year long sedimentary record from the Gulf of Mexico. Using a multi proxy approach we can better constrain the effects of the local ecology and other oceanographic factors on each paleothermometer. Coupling of the two SST proxies expands our ability to assess SST conditions, and thus will provide a more complete picture of regional ocean climate variability (e.g. seasonality, v ertical temperature gradients) during the late Holocene.
42 3.3. Study Location The location of our study is the Pigmy Basin an intraslope basin located in the northern Gulf of Mexico (2711.61 N, 9124.54 W, water depth 2259 m), ~ 200 km south of the Mis sissippi River mouth (Figure 3.1). The Mg/Ca and TEX 86 records discussed in this study are derived from two different sub cores (PBBC 1E and PBBC 1F, respectively) isolated from a single box core recovered from the Pigmy Basin in 2003 aboard the RV Longhor n. Sedimentation rates during the late Holocene are relatively high (43 cm/kyr) as a result of the large volume of terrigenous material delivered via the Mississippi River. The high sedimentation rate combined with a 0.5 cm sampling interval allow for de tailed study of multi decadal to centennial scale climate variability of the past millennium. Figure 3.1. Map of the Gulf of Mexico. Location of the Pigmy Basin (2711.61'N, 9124.54'W, 2259 meters water depth) is indicated by marker.
43 An age model was constructed for sub core PBBC 1E based on 7 accelerator mass spectrometer radiocarbon dates on planktonic Foraminifera ( Richey et al ., 2007). Calibrated calendar ages (calibrated using the CALIB 5.0 program with a 400 year reservoir correction [Stuiver e t al., 1998]) were plotted against core depth, and a least squares linear regression (r 2 =0.995) indicates a sedimentation rate of 43 cm/kyr. The core top radiocarbon date indicates post 1950 deposition, and therefore we infer a near modern core top. The age model for PBBC 1E was projected onto sub core PBBC 1F (i.e. the core top age was set to 0 yrs BP, and a 12.3 yr interval was assigned to each 0.5 cm sample). The discrepancy between the lengths of the two sub cores (PBBC 1E is 59 cm long and PBBC 1F is 44 cm long) is due to their relative positions within the box core. The shovel of the box core is curved such that cores taken in the center of the box core are longer than core taken on the sides of the box core. 3.4. Methods 3.4.1. Extraction and Isolation of GDGT Lipids Core PBBC 1F was sampled at 0.5 cm intervals and freeze dried. Samples were solvent extracted with a DIONEX Accelerated Solvent Extractor (ASE 200) using a solvent mixture of 9:1 dichloromethane (DCM) to methanol (MeOH) at the C ollege of Marine Science, University of South Florida. The resulting total lipid extract (TLE) then underwent a base hydrolysis (in 0.5M KOH in MeOH), and was separated into an acid and neutral fraction via liquid liquid extraction under neutral and acidi c conditions, respectively. The neutral fraction was then separated into an apolar, ketone and polar
44 fraction via silica pipette column chromatography using hexane, 3:2 (vol:vol) hexane/DCM, and 1:1 (vol:vol) DCM/MeOH, respectively. 3.4.1. TEX 86 and BIT Analysis The polar fraction, containing the GDGTs, was dissolved in a 99:1 (vol:vol) mixture of hexane:propanol, then filtered through 0.45 m PFTE filters. Analyses of GDGTs for TEX 86 and BIT index determination were performed by high pressure liquid ch romatography mass spectrometry (HPLC MS) at the Woods Hole Oceanographic Institution. Samples were analyzed on an Agilent 1200 series LC/MSD SL operating in positive APCI mode, with an autoinjector and Chemstation software. A Prevail Cyano column (150 x 2 .1 mm, 3 m from Grace Davison Discovery Sciences) was used with 99:1 hexane:isopropanol (vol:vol) as an eluent. After the first 5 min, the eluent increased by a linear gradient up to 1.8% isopropanol (vol) over the next 45 min at a flow rate of 0.2 mL/min Scanning was performed in single ion monitoring (SIM). The TEX 86 indices were calculated according to the following equations: TEX 86 = ([II]+[III]+[IV'])/[I]+[II]+[III]+[IV']) (from Schouten et al., 2002) BIT = ([V]+[VI]+[VII])/([V]+[VI]+[VII]+[IV]) (f rom Hopmans et al ., 2004) where the roman numerals refer to the GDGT structures shown in Figure 3.2. The TEX 86 index was then converted to SST according to the following equation from Kim et al ., (2008): T = 10.78 + 56.2*TEX 86 Each of the 88 GDGT sampl es in this study was analyzed in triplicate. The average standard deviation for TEX 86 among triplicate analyses is 0.007, which corresponds to
45 0.4C using the calibration equation of Kim et al. (2008). For the BIT index, the average standard deviation among triplicates is 0.006. Figure 3.2. Molecular structures of GDGTs. A) isoprenoid and B) branched glycerol dialkyl glycerol tetraethers (GDGTs) used in the TEX 86 ( Schouten et al ., 2002) and BIT ( Hopmans et al., 2004) indices. 3.5. TEX 86 SST record from Pigmy Basin The TEX 86 record from the Pigmy Basin indicates significant SST fluctuations over the past 1,000 years (Figure 3.3a). The TEX 86 varies between 0.65 and 0.70 (25.8 28.5C) and there is a general warming trend over the length of the recor d, with the core top recording the warmest TEX 86 SST (28.5C) of the past millennium. The time interval 1,100 600 yrs BP is relatively stable with a mean TEX 86 SST of 27C, with the exception of a century long SST excursion (1,000 900 yrs BP) in which tem peratures were 1C cooler. This period is followed by a rapid transition to ~1C warmer SSTs ca. 600 yrs BP. The LIA (400 150 yrs BP) is marked by a 0.5C drop in mean SSTs from the
46 preceding two centuries (600 400 yrs BP), and followed by a rapid 1.5C warming into the 20 th century. Figure 3.3. GDGT based proxy records for a Pigmy Basin box core (PBBC 1F). A) TEX 86 record with corresponding temperature scale, calibrated using the equation: T= 10.78 + 56.2*TEX 86 from Kim et al. (2008). The pooled stan dard deviation of triplicate TEX 86 measurements is indicated (0.007) which corresponds to 0.4C. B) BIT index for PBBC 1F. The pooled standard deviation among triplicate analyses is 0.006. The dashed lined indicates the threshold BIT value of 0.3, above which input of terrestrial organic matter may influence TEX 86 ( Weijers et al., 2006).
47 3.6. Influence of terrestrial input on Pigmy Basin TEX 86 In marine marginal settings proximal to major fluvial systems such as the Mississippi River, there is concern that the TEX 86 SST proxy may be influenced by delivery of isoprenoid GDGTs derived from terrestrial sources. The Branched and Isoprenoid Tetraether (BIT) index, based on the relative abundance of terrestrially derived branched tetraether lipids (GDGT V, VI, and VII) versus the marine derived crenarchaeol (GDGT IV), can be used to monitor the relative contribution of terrestrial organic matter (TOM) to sediments ( Hopmans et al., 2004). Weijers et al. (2006) found that in marine sediments with a large cont ribution of terrestrially derived organic matter, TEX 86 values tended to be biased toward warmer temperatures. Using a two end member mixing model (GDGT distribution in African soils versus GDGT distribution in marine sediments of the Niger deep sea fan), Weijers et al (2006) predicted a +1C temperature bias in the TEX 86 SST of sediments with BIT values of 0.2 0.3, and the influence of TOM on TEX 86 temperature estimates was found to increase non linearly at BIT values > 0.3. However, this specific tempe rature bias depends heavily upon the composition and source of TOM, and therefore is not necessarily applicable to other marine basins. In the Pigmy Basin (core PBBC 1F) the BIT index varies between 0.14 and 0.40 (mean=0.25) over the past 1000 years (Figu re 3.3b), which lies between the values observed for coastal and open marine environments ( Hopmans et al ., 2004). In order to quantify the potential bias to the TEX 86 SST record introduced by the moderately elevated BIT index in Pigmy Basin sediments, the GDGT composition of TOM delivered to the GOM via the Mississippi River would have to be characterized. Although we cannot directly assess to what degree, if any, terrestrially derived GDGTs are influencing
48 the TEX 86 measurements, we adopt the working hyp othesis that downcore variability in TOM input does not significantly affect downcore TEX 86 variability. Figure 3.3 shows a comparison of the downcore records of both the BIT and TEX 86 indices from Pigmy Basin, and illustrates that the two indices do not covary (the correlation between the TEX 86 and BIT indices is r = 0.2, which is small, but significantly different from 0, at the 90% confidence level). The lack of correlation is further illustrated in a cross plot of the BIT index versus TEX 86 SSTfrom th e Pigmy Basin (r 2 = 0.04 ) (Figure 3.4). We conclude that, while there is a possibility that the absolute SST may be biased, the pattern of TEX 86 variability over the past 1,000 years in the Pigmy Basin is not systematically influenced by changes in terres trially derived GDGT input. Figure 3.4. Cross plot between BIT index and TEX 86 SST for Pigmy Basin box core (PBBC 1F) .There is no significant correlation between TEX86 and BIT indices (r 2 =0.04).
49 3.7. Inferring depth and seasonality for TEX 86 signal i n the GOM The global core top calibration of Kim et al (2008) indicates that TEX 86 best reflects mean annual SST of the upper mixed layer (0 30 m ). Although Crenarchaeota are living and biosynthesizing GDGTs throughout the water column, Wuchter et al (2006) showed that the TEX 86 signal recorded in deep sediment trap samples (> 500 m) still reflects surface temperatures, rather than water temperatures resulting from deep water column GDGT production. This is most likely due to the effective packaging a nd export of GDGT containing crenarchaeal cells from the upper water column in fecal pellets via zooplankton grazing. Due to lack of a packaging process and mechanism of transport to sediments, deep water crenarchaeotal production likely has an insignifi cant effect on TEX 86 SST estimates ( Wuchter et al., 2005). The core top TEX 86 value of 0.70 (0.008) in the Pigmy Basin corresponds to an SST of 28.5C (0.5C), which is ~3C warmer than the mean annual SST for the GOM (25.4C), and equivalent to the mea n summer (June Sept.) SST for the GOM mixed layer ( Levitus 2004) (Figure 3.5a). This suggests that the TEX 86 record in the Pigmy Basin is a heavily summer weighted signal, and that crenarchaeotal production and/or export via zooplankton grazing must be s ignificantly higher during the summer in the Gulf of Mexico. This is contrary to the observation of Wuchter et al. (2005) that GDGTs occurred in higher abundances during the winter and spring in a number of marine settings, including the subtropical and t ropical Atlantic sites, BATS (Bermuda Atlantic Time series Study) and in the Cariaco Basin. In contrast, Shah et al. (2008) found in Bermuda Rise core top sediments that TEX 86 indicated SSTs ~2C warmer than those estimated from co occurring foraminiferal 18 O, a fact consistent with our findings in the
50 Pigmy Basin. At present there are no data from the GOM on seasonal changes in crenarchaeotal production. Figure 3.5 Annual cycle of water temperature variability in the upper 100 meters of the water co lumn Data are monthly averages from the climatic means for the Gulf of Mexico ( Levitus 2004). A) The circled region indicates the dominant depth and seasonal range of the TEX 86 signal, based on the core top TEX 86 SST of 28.5C. B) The circled regions ind icate the two possible distributions of G. ruber in the Gulf of Mexico. The modern SST (25.4C) recorded by G. ruber indicates that they are living in the upper mixed layer (0 30 meters) throughout the year, or they are limited to the summer months (Jun Se pt), but living over a greater depth range (0 75 meters). 3.8. Comparison of Pigmy Basin TEX 86 SST to Mg/Ca SST Record Comparison of two or more different paleotemperature proxy records from co occurring sediments often reveals strikingly different clim ate histories. This can be attributed to a number of physical, biological and chemical factors, including separate transport mechanisms to the sediment, different ecologies of the signal carrying organisms, and susceptibility to separate diagenetic proces ses of the different sediment components. It has been well documented that organic compounds, which are attached to fine grained particles, can be laterally transported long distances (e.g. Ohkouchi et al., 2002; Mollenhauer et al., 2005) causing age and temperature offsets between molecular
51 organic proxies (e.g. U K' 37 and TEX 86 ) and planktonic Foraminifera based proxies. Compound specific radiocarbon dating suggests that GDGTs may be more labile than alkenones, and thus don't survive transport over long distances under oxic conditions as well as alkenones ( Mollenhauer et al., 2007; Shah et al., 2008 ). Thus GDGTs are more likely to contain a local signal that is contemporaneous with the foraminiferal signal. Each paleo SST proxy is based on the geochemi stry of a planktonic organism, each of which has a different depth and seasonal distribution. These depth and seasonal distributions for a signal carrier can vary depending on geography and local controls on productivity For example, Huguet et al (2006 ) find a 2.5C difference between the estimates of LGM to present warming between U K' 37 SST and TEX 86 SST in the Arabian Sea. They attribute these discrepancies to differences in seasonality between the crenarchaeota and haptophyte algae, which are the pl anktic source of the TEX 86 and U K' 37 signals, respectively. Castaeda et al. (2010) found that U K' 37 and TEX 86 based SST estimates in the eastern Mediterranean were similar during the LGM, but TEX 86 SSTs were generally 1 2C warmer than U K' 37 SSTs during the Holocene. They also attribute this discrepancy to changing seasonality of crenarcaeota and haptophyte blooms during the last deglaciation. The TEX 86 based SST record generated from the Pigmy Basin (core PBBC 1F) shows similar patterns of variabili ty to the Mg/Ca based SST record (core PBBC 1E) previously published by Richey et al (2007) from the same box core, but with important differences. Due to the potential implications of offsets in absolute SST calibration between the two different paleote mperature proxies (TEX 86 and Mg/Ca), similarities in the decadal to centennial scale patterns of variability are discussed first. Figure 3.6
52 shows the down core SST variability in both records plotted as a temperature anomaly relative to their respective core top SSTs. Both the Mg/Ca and TEX 86 based SST records indicate that SSTs ca. 500 yrs BP were ~1C cooler than core top SSTs. The Mg/Ca record shows that the mean temperature during the LIA (ca. 400 150 yrs BP) was 2.0C (0.6C) cooler than today, an d the TEX 86 record indicates a LIA mean that was 1.2C (0.6C) cooler than today. From 850 600 yrs BP both records indicate SSTs that were ~1.5C cooler than their modern core top SSTs. The major discrepancies between the 2 records are 1) The LIA is not the coolest interval during the past millennium as recorded by TEX 86 and 2) The period prior to 900 yrs BP was similar to the core top SST as recorded by foraminiferal Mg/Ca, but cooler in the TEX 86 record. Figure 3.6. TEX 86 and Mg/Ca records plotted a s anomalies relative to their respective calibrated core top temperatures. Individual data points are shown with open circles (TEX 86 ) and open triangles (Mg/Ca). Both records are derived from separate subcores with in the same Pigmy Basin box core. Sub cor e PBBC 1F (TEX 86 ) is 15 cm shorter than PBBC 1E (Mg/Ca), due to position within the box core.
53 3.9. Inferring Depth and Seasonality for G. ruber in the GOM The white variety of the planktonic foraminifer, Globigerinoides ruber is abundant throughout the tropical and subtropical oceans, and is constrained to the euphotic zone by its photoautotrophic dinoflagellate symbionts. The seasonal and depth distribution of G. ruber (white) has been reported from sediment trap (e.g. Deuser, 1987; Tedesco et al., 20 03; Tedesco et al., 2009) and plankton tow studies (e.g. Tolderlund and B, 1971; Fairbanks et al., 1980; B, 1982) in a number of different locations proximal to the Gulf of Mexico. The modern depth preference of G. ruber (white) has been documented in a number of Sargasso Sea plankton tows studies, but there are limited data specific to the Gulf of Mexico. Fairbanks et al. (1980) found that G. ruber (white) was common throughout the upper 100 meters of the water column in a November 1975 plankton tow st udy, while a series of monthly plankton tows in the Sargasso Sea indicated that G. ruber (white) was most abundant in the uppermost 10 meters of the water column ( Tolderlund and B, 1971). B ( 1982) confirmed the presence of G. ruber (white) throughout th e upper 50 meters of the water column in the western Gulf of Mexico. The Mg/Ca of G. ruber (white) is typically interpreted as a mean annual mixed layer signal in the Gulf of Mexico (e.g. Flower et al., 2004; LoDico et al., 2006; Richey et al. 2007). Flux data from a sediment trap study in the Sargasso Sea indicate that G. ruber are present throughout the annual cycle, with peak fluxes in early spring and late summer ( Deuser 1987). A one year sediment trap study in the northern Gulf of Mexico indicat es that G. ruber (white) is present throughout the year, with small peaks in flux in early spring and late summer ( Tedesco et al 2009).
54 The Mg/Ca record in the Pigmy Basin was generated from a well constrained size fraction (250 300 m) of the white variety of G. ruber ( Richey et al., 2007) A mean Mg/Ca core top value of 4.43 mmol/mol (0.03) for the Pigmy Basin was determined from replicate analyses of core top (0 0.5 cm) samples from 7 GOM core tops. This robust core top value corresponds to an S ST of 25.4C (Mg/Ca = 0.449 exp(0.09 SST) (from Anand et al., 2003), which is the modern mean annual SST for the GOM ( Levitus et al ., 2004 ). Unlike the TEX 86 signal, which is well constrained as a surface summer signal, the core top Mg/Ca SST of 25.4 C can be produced by a number of different scenarios (Figure 3.5b). 1) The G. ruber (white) depth habitat is limited to the uppermost 30 meters of the water column and flux to the sediments is equally weighted throughout the year. 2) The flux of G. rube r (white) is summer weighted, but G. ruber (white) has a greater range (0 75 meters) in the water column. In either case, increases (decreases) in SST, as recorded by the Mg/Ca of G. ruber may be influenced by warmer (colder) winters or deeper (shallower ) mixed layers. 3.10. Mg/Ca SST to TEX 86 SST gradients Previous studies have interpreted the difference in paleo temperature records between species of Foraminifera with known differences in seasonal and depth distribution ( Williams et al ., 2009). Usi ng the assumptions we have made about the depth and seasonal distribution of GDGTs versus G. ruber (white) in the GOM we can exploit the differences between the two SST records to make inferences about changing upper water column structure and/or seasonal ity over the past millennium. To do this we
55 take the difference ( T) between the TEX 86 and Mg/Ca based SST records using their respective paleotemperature calibrations (Figure 3.7). The results of this exercise are plotted in Figure 3.7 and suggest that the greatest difference between the TEX 86 and Mg/Ca SST occurs between 200 and 300 yrs BP, during the maximum LIA cooling. Figure 3.7. The !T (TEX 86 Mg/Ca) for the Pigmy Basin, Gulf of Mexico. A) The SST records derived from TEX 86 (open circles) and M g/Ca (closed circles), calibrated to SST using their independent paleotemperature equations. B) The #T (TEX 86 Mg/Ca) for the Pigmy Basin, Gulf of Mexico. The largest #T (>4C) occurs during the LIA, while minimum #T (<1C) occurs during the MWP. Note that when uncertainties associated with respective calibrations to SST and analytical errors associated with both proxies are compounded, the error of the #T record is 2C.
56 The maximum is controlled by greater cooling of Mg/Ca SST. This implies that the LIA cooling observed in the GOM may have been dominated by enhanced winter cooling relative to summer cooling, or shoaling of the thermocline (shallow mixed layer). The minimum difference between the TEX 86 and Mg/Ca SST records occurs 1100 900 yrs BP, during the MWP. This implies that during this time there was a significant decrease in seasonality and/or increase in mixed layer depth. Without seasonally resolved flux data for Crenarchaeota or G. ruber specific to the Gulf of Mexico, we cannot evaluate wh ether the differences between TEX 86 and Mg/Ca SST estimates result from differences in seasonal or depth distribution. Figure 3.8(a d) depicts the different scenarios by which the T between the two proxies could change over time. Figures 3.8a and 3.8c d epict the theoretical LIA scenarios, in which there was an especially shallow mixed layer (causing a steep thermal gradient in the upper 100 meters of the water column) or increased seasonality. In either of these scenarios we would expect the T between T EX 86 and Mg/Ca temperatures to be greater. The converse is shown in Figures 3.8b and 3.8d, where the mixed layer is deeper (causing a decreased thermal gradient in the upper water column), or decreased seasonality. In these scenarios, a smaller T betwe en TEX 86 and Mg/Ca temperatures would be expected. It must also be recognized that the seasonal or depth preferences of either signal carrier may be influenced by some other environmental factor over the past 1,000 years, such as changes in nutrient input or salinity.
57 Figure 3.8. Illustrations of Mixed Layer and Seasonality Hypotheses. An illustration of the working hypotheses that A & B) the changes in T (TEX 86 Mg/Ca) over time may be indicative of changes in the thermal gradient in the upper water c o lumn in the Gulf of Mexico and/or C & D) changes in #T over time may be indicative of changes in the seasonality in the Gulf of Mexico. The "LIA scenario", in which we observe the greatest T, is illustrated by a steeper thermal gradient (or shallow mixed layer) in panel A, or by a greater seasonal range of SST in the Gulf of Mexico in panel C. The "MWP scenario", in which we observe the smallest T, is illustrated by a reduced thermal gradient (or deeper mixed layer) in panel B, or by a reduced seasonal ra nge in the Gulf of Mexico in panel D. Both scenarios require the assumption that the Mg/Ca and TEX 86 signals maintain their relative depth and/or seasonal distributions throughout the past millennium.
58 3.11. Potential Implications for the T record The c hanges in T between the TEX 86 and Mg/Ca SST records in the Pigmy Basin can be interpreted in terms of changes in the amplitude of the seasonal SST cycle and/or changes in mixed layer depth. Either scenario, 1) a deepening of the mixed layer, or 2) a decr eased seasonal SST variability with reduced winter cooling, would result in increased heat storage in the upper ocean on an annual basis. The Gulf of Mexico is in the path of, or the birthplace for a large number of Atlantic tropical cyclones, and thus it follows that a persistent annual build up of heat in the upper water column may have been a factor in enhancing tropical cyclone activity in the past. The MWP (ca. 1100 900 yrs BP) is marked by the minimum T for the past 1,000 years in the Pigmy Basin, suggesting a centennial scale period in which more heat was being stored in the upper ocean. Greater heat storage in the GOM via a warm mixed layer that is thicker and/or more seasonally persistent is consistent with reconstructions of greater tropical cy clone frequency in the Atlantic basin during the MWP (Figure 3.9) ( Mann et al., 2009). Maximum T in the Pigmy Basin record is observed during the LIA, suggesting a period of enhanced seasonality and/or a decreased mixed layer depth. This observation is a lso consistent with a minimum in reconstructed tropical cyclone frequency during the LIA by Mann et al. (2009).
59 Figure 3. 9. Comparison of Pigmy Basin !T (TEX 86 Mg/Ca) with a reconstruction of tropical cyclone counts. A) Grey curve is a 3 point running mean of the #T record from the Pigmy Basin. We interpret decreasing #T in terms of increasing mixed layer depth and/or decrea sed seasonality. B) The dashed line is a multidecadal smoothed record of Atlantic tropical cyclone counts over the past 150 years. The solid black line is the proxy reconstructed Atlantic tropical cyclone counts based on statistical model from Mann et al. . The LIA is indicated by the shaded bar and highlights a minimum in reconstructed tropical cyclone counts, and a maximum #T. The MWP, indicated with a shaded bar, can be characterized by a maximum in reconstructed tropical cyclone counts and a minim um in #T. 3.12. Conclusions We present the first comparison of a decadal resolution TEX 86 based SST record with a foraminiferal Mg/Ca based SST record for the past 1,000 years from marine sediments. There are similarities in the magnitude and pattern of SST variability recorded
60 by TEX 86 and Mg/Ca in co occurring sediments of the Pigmy basin box core, especially over the past four centuries, including a substantially cooler LIA (by 1.5 2.5C). There are however, significant differences in the two SST rec ords that can most likely be attributed to changes in the seasonal and/or depth distribution of crenarchaeota versus G. ruber (white). The core top TEX 86 SST indicates that TEX 86 in the Gulf of Mexico is a summer weighted, upper mixed layer signal, while the Mg/Ca SST indicates that G. ruber are most likely living throughout the year, and/or deeper in the water column. The difference ($T) between the two proxy records indicates changes in seasonality and/or mixed layer depth over the past millennium with the LIA characterized by enhanced seasonality and/or a shallow mixed layer whereas the MWP was recognized by a decrease in the seasonal temperature gradient and/or a deeper mixed layer. The increased ability to store heat in the GOM surface waters during the MWP may be a link to understanding historical changes in past Atlantic tropical cyclone activity.
61 Chapter 4 Ecological controls on the shell geochemistry of pink and white Globigerinoides ruber in the northern Gulf of Mexico: Implications fo r paleoceanographic reconstruction 4.1. Abstract We assess ecological controls on shell geochemistry for two of the most commonly used planktonic foraminifers for paleoceanographic reconstruction in the subtropical Atlantic Ocean: the pink and white var ieties of Globigerinoides ruber We evaluate the relationship between foraminiferal test size and shell geochemistry ( 13 C, 18 O and Mg/Ca) using temporally well constrained core top samples from the Pigmy and Garrison Basins, in the northern Gulf of Mexi co. The core top samples are from the uppermost 0.5cm of the respective box cores, and represent ~10 30 years of sedimentary deposition. For each size fraction, multiple separate analyses (2 10 aliquots) of 60 foraminifera were analyzed. Data show a sig nificant positive relationship between Mg/Ca and test size, with a range of 1.1 mmol/mol (~2.5C) from the smallest (150 212 m) to largest (>500 m) size fractions of G. ruber (pink), but no significant relationship in G. ruber (white). There is a depletion in 18 O of 0.26 per 100 m increase in test size for both pink and white G. ruber The increase in Mg/Ca and decrease in 18 O is consistent with an increase in calcification temperature of 0.66C per 100mm increase in test size. Overall, these results stress the necessity for using a consistent size fraction. In addition, we compare downcore records of 18 O and Mg/Ca from b oth pink and white G. ruber and make inferences about the relative seasonal distribution and depth habitat of the 2 varieties.
62 4.2. Introduction Improving understanding of the biological, chemical and physical factors that contribute to uncertainty i n foraminiferal geochemical proxies is critical to the interpretation of paleoceanographic records. The uncertainties inherent to Mg/Ca based SST estimates can exceed the environmental signal in some cases, especially in low latitude records covering the past few millennia, when SST variability was likely small (<2C) relative to glacial interglacial changes. Some sources of uncertainty in Mg/Ca SST estimates include, but are not limited to diagenesis (e.g. Boyle 1983; Barker et al., 2003; Pena et al., 20 05), salinity (e.g. Nrnberg et al., 1996; Ferguson et al., 2008; Kisakrek et al., 2008), shell heterogeneity (e.g. Eggins et al., 2004; Sadekov et al., 2008) and dissolution (e.g. Dekens et al., 2003). Each of these factors has the potential to overprin t the Mg/Ca signal of the downcore record, and the influence of these factors is variable, and often basin specific. One issue that has been explored to some degree is the effect of foraminiferal test size on Mg/Ca (Elderfield et al., 2002; Ni et al., 2 007). Elderfield et al. (2002) illustrated that there is a positive correlation between Mg/Ca and test size in a number of species of planktonic foraminifera; however it is difficult to quantify the relationship in that study due to the small number of ind ividuals analyzed for each size fraction (~20), and insufficient temporal constraint in their sediment sample (their sample represented ~800 yrs (Elderfield et al., 2002)). Ni et al. (2007) found no relationship between test size and Mg/Ca in the white va riety of Globigerinoides ruber but their sample also represented multiple centuries of deposition, and they had few replicate analyses.
63 In this study we improve upon previous studies by presenting a temporally well constrained data set to investigate the relationship between test size and Mg/Ca in the white and pink varieties of G. ruber The 0.5cm zero age dated core top samples from the two high accumulation rate (40cm/kyr) basins (Pigmy and Garrison basins) in the Gulf of Mexico (GOM) represent the mo st recent few decades, and between 2 and 10 aliquots of 60 individual Foraminifera were analyzed for each size fraction of pink and white G. ruber We propose an internally consistent explanation of the changes in three geochemical parameters with size ( 13 C, 18 O and Mg/Ca) as well as for observed offsets in the geochemistry of the pink and white varieties of G. ruber The pink and white varieties of G. ruber are the two most abundant foraminifers in modern GOM sediments, and when combined, make up >45 % of the total assemblage, respectively (Brunner, 1979; Kennett et al., 1985; Dowsett et al., 2003). The two are morphologically very similar, are both hosts to dinoflagellate photosymbionts, and accordingly reside within the euphotic zone. Despite their similarities, we observe differences in their shell geochemistry. We compare downcore geochemical records of pink and white G. ruber in a high resolution, recent sedimentary series in order to assess their relative depth and seasonal distributions. 4 .3. Materials and Methods The samples used for the size fraction study are from the core top (top 0.5 cm) of the Pigmy Basin box core (PBBC 1; 2711.61'N, 9124.54'W; 2259 meters water depth), and the Garrison Basin box core (PE07 2; 2640.5' N, 9355.5' W; 1570 meters water depth). Each of these basins is an intraslope basin in the northern Gulf of Mexico, and
64 has a relatively high sediment accumulation rate (20 40 cm/kyr) due to the input of terrigenous material from the Mississippi River. For each of the box cores the sediment water interface was recovered, and therefore the core top samples incorporate the most recently deposited sediments. Radiocarbon dates from a mixed assemblage of planktonic foraminifera, for each of the core top samples, reflect a modern age (< 0 after reservoir correction; Richey et al., 2007, 2009). Size fractions where separated by sieving, and between 2 and 10 different aliquots of "60 foraminifera were analyzed for each size fraction of both pink and white G. ruber Downcore analyses were performed on the 250 300 m size fraction of pink and white G. ruber from the same subcore of the Pigmy Basin box core, PBBC 1. The age model for PBBC 1, as well as the oxygen isotope and Mg/Ca records from the white variety of G. ruber was previously published by Richey et al. (2007). Age control is based on 7 accelerator mass spectrometer (AMS) radiocar bon dates. The sampling interval of 0.5 cm combined with the linear sedimentation rate of 43 cm/kyr yields a sampling resolution of ~12 years. For all elemental a nalyses "60 Foraminifera were picked from each size fraction of Globigerinoides ruber (white and pink) Foraminifera were lightly crushed and underwent a cleaning process that includes multiple clay removal steps, an oxidative step to remove organic mater ial, and an acid leaching step to remove adsorbed metals (Barker et al., 2003). A reductive cleaning step was not performed on th ese samples. Elemental analyses were performed on a Perkin Elmer Optima 4300 dual view inductively coupled plasma optical emi ssion spectrometer (ICP OES) at the College of Marine Science, University of South Florida (CMS, USF). In all cases where there are replicate Mg/Ca
65 analyses, replicates are based on se parate aliquots of "60 foraminifera that have been crushed, cleaned and analyzed separately. Oxygen and carbon isotope ratios were measured on a ThermoFinnigan Delta Plus XL light stable isotope ratio mass spectrometer (SIRMS) at the CMS, USF. The 18 O ca lcite and 13 C calcite are reported on the VPDB scale. 4.4. Results 4.4.1. Relationship between test size and carbon isotopic composition The positive relationship between the 13 C and foraminiferal test size has been well established in number of stu dies using a variety of different planktonic foraminifera species across a range of oceanographic settings (e.g. Berger et al., 1978; Curry and Matthews, 1981; Oppo and Fairbanks, 1989; Ravelo and Fairbanks, 1995; Elderfield et al., 2002). Berger et al. ( 1978) proposed that foraminiferal 13 C increases with test size as a function of ontogeny, due to the fact that metabolic rates are highest in the early ontogenetic stages (i.e. smallest individuals), and therefore are the most depleted relative to equilib rium. Figure 4.1. Illustration of the relationship between 13 C and ontogeny in Foraminifera. This illustration was redrawn from Berger et al. (1978).
66 As the foraminiferan matures, the metabolic rate slows and the carbon isotopic composition of the te st becomes less depleted relative to the 13 C of seawater %CO 2 The formation of a deep calcite crust by mature planktonic Foraminifera accounts for the observed depletion in the largest individuals (Figure 4.1). It has also been proposed that the incr ease in 13 C with increasing size can be attributed to symbiont photosynthesis (Oppo and Fairbanks, 1989; Spero and Lea, 1993). Symbionts preferentially utilize 12 C during photosynthesis, leaving the carbon pool in the calcification microenvironment enric hed with respect to 13 C (Spero and Williams, 1988). Photosynthetic activity increases as a function of increasing light level (Spero and Lea, 1993), and the increase in symbiotic density with advancing ontogeny (Spero and Parker, 1985). A culture study i n which O. universa and G. sacculifer were grown under variable light levels revealed that the 13 C of the Foraminifera was controlled by symbiont photosynthesis, rather than ontogeny. However, the observation of increasing 13 C with increasing test size in non symbiont bearing foraminifera (e.g. Ravelo and Fairbanks, 1995; Elderfield et al., 2002) suggests that symbiont photosynthesis is not the sole contributing factor to the 13 C size relationship. Our observation of the relationship between 13 C and t est size in white and pink G. ruber is consistent with previous studies. For G. ruber (pink) the 13 C ranges from 0.5 in the smallest size class (150 212 m), to 2.0 in the largest size class (425 500 m ). G. ruber (white) ranges from 0.3 in the smalles t size class (150 212 m) to 1.3 in the largest size class (355 425 m) (Figure 4.2.a). The rate of change per increase in size slows as the forams get larger, which is consistent with a smaller
67 fractionation with decreasing metabolic rate. The pink varie ty of G. ruber is consistently ~0.4 more enriched than the white variety. The overall range in 13 C from the smallest to largest individuals exceeds the downcore 13 C variability over the past 1500 years in both white and pink G. ruber thus emphasizing the importance of using a narrow size range for all downcore paleoceanographic studies. Figure 4.2. Relationship between foraminiferal test size and 13 C and 18 O A) The carbon isotopic composition is plotted against test size for pink (open circle) a nd white (square) G. ruber A 3 order polynomial is fit through both data sets. B) The oxygen isotopic composition is plotted against test size for pink (open circle) and white (square) G. ruber A linear least squares regression line is fit through bot h data sets, and they both have the same slope. The mean offset between pink and white G. ruber in 18 O is 0.27. Error bars represent the standard deviation among all replicate measurements for each size fraction. 4.4.2. Relationship between test size and oxygen isotopic composition A number of studies have investigated the relationship between 18 O and size in planktonic foraminifera, however, unlike with 13 C, the relationship does not seem to be
68 systematic. Elderfield et al. (2002) found that ther e was a general decrease in 18 O in three species of planktonic foraminifera ( Orbulina universa Neogloboquadrina dutertrei and white G. ruber ), but 14 other planktonic species analyzed showed either the opposite relationship, or no significant relationsh ip at all. Ravelo and Fairbanks (1992, 1995) observed a significant decrease in 18 O of G. ruber (white) with increasing test size, but no relationship in G. ruber (pink). Curry and Matthews (1981), however, looked at the relationship between 18 O and tes t size in G. ruber (white), and found no significant relationship. In a culture study, Spero and Lea (1993 ) concluded that ontogeny did not have an effect on the shell 18 O of G. sacculifer but they observed a significant depletion in 18 O with increasin g light levels for the symbiont bearing O. universa and G. sacculifer Size Fraction ( m) n 18 O (permille) STDEV 13 C (permille) STDEV 150 212 4 1.38 0.06 0.52 0.14 212 250 7 1.58 0.12 1.01 0.14 250 300 9 1.52 0.14 1.26 0.11 300 355 7 1.66 0.09 1.70 0.16 355 425 8 1.96 0.08 1.87 0.20 425 500 6 2.09 0.11 1.99 0.22 Table 4.1. Oxygen and Carbon isotopic data for G. ruber (pink) Data in this table represent the mean values for all measurements in each size fraction. "n" is the total number of measurements for each size fraction, and each measurement is based on "60 individual Foraminifera. STDEV represents the standard deviation among replicate measurements for each size fraction.
69 Size Fraction ( m) n 18 O (permille) STDEV 13 C (pe rmille) STDEV 150 212 5 1.17 0.20 0.26 0.10 212 250 7 1.22 0.18 0.56 0.05 250 300 8 1.41 0.18 0.86 0.27 300 355 7 1.39 0.15 1.24 0.15 355 425 3 1.70 0.16 1.34 0.17 Table 4.2. Oxygen and Carbon isotopic data for G. ruber (white). Data in this t able represent the mean values for all measurements in each size fraction. "n" is the total number of measurements for each size fraction, and each measurement is based on "60 individual Foraminifera. STDEV represents the standard deviation among replicate measurements for each size fraction. Our results indicate that 18 O decreases with increasing test size in both white and pink G. ruber (Figure 4.2.b). In the white variety the 18 O of foraminiferal calcite decreases from 1.1 from the smallest individuals (150 212 m) to 1.6 in the largest individuals (355 425 m), while pinks range from 1.4 in the smallest size class (150 212 m) to 2.0 in the largest siz e class (425 500 m). A least squares linear regression through the mean 18 O values for each size fraction indicates a significant negative correlation between 18 O and size for both pink and white G. ruber (r 2 = 0.94 and 0.84, respectively). The slopes of the relationships are nearly identical for pink and white G. ruber and indicate a 0.26 decrease in 18 O per 100 m increase in size. There is a consistent offset between pink and white G. ruber in which the pink variety is depleted by ~0.27 relative to the white. 4.4.3. Relationship between foraminiferal test size and Mg/Ca In G. ruber (pink) there is a significant increase in Mg/Ca with increasing test size from 150 m to >500 m (Figure 4.3.a). The overall range in Mg/Ca from the smallest size (150 212 m) to the largest size (>500 m) is 1.1 mmol/mol. This corresponds to an
70 SST difference of 2.5C from the smallest to largest size fraction based on the Mg/Ca paleotemperature equation for G. ruber (pink): (1) Mg/Ca=0.38exp(0.09*T) from Anand et al. (2003) A temperature range of 2.5C exceeds the range of SST variability expected for most late Holocene records. The increase in Mg/Ca with test size appears to be linear, with a positive slope of 0.27 mmol/mol (0.66C) per 100 m increase in test size. The Mg/Ca values for the 250 300 m and 300 355 m size fractions are not significantly different from each other, and have Mg/Ca values (mean=4.30.02mmol/mol) that correspond to the modern summer weighted (April October) SST for the Gulf of Mexico (27.0C ), when calibrated using equation (1). This exercise was repeated for a second depth interval (320 325mm, 650 yrs B.P.) in the Pigmy Basin box core (PBBC 1) in order to test whether the relationship between test size and Mg/Ca is stationary through time The interval ca. 650 yrs B.P. was 2C colder than the modern core top in the downcore record, as indicated by the Mg/Ca SST from the 250 300 m size fraction of G. ruber (white) (Richey et al., 2007). Figure 4.3.b illustrates a nearly identical slope o f the size Mg/Ca relationship for the both the modern core top sample and for the sample at 650 yrs B.P., indicating that the increase in Mg/Ca with test size is a robust relationship, even under different climatic conditions. The fact that each size frac tion from the interval 650 yrs B.P. is consistently offset ~2C cooler than the corresponding size from the modern core top SST, further corroborates the 2C cooling observed in the downcore record. It is also worth noting that there were no G. ruber (pin k) in the >500 m size fraction for the 650 yrs B.P. sample, while there were
71 more than 60 individuals in the >500 m size fraction in the core top sample. This supports the hypothesis that Foraminifera grow larger under warmer conditions. Size Fracti on ( m) n Mg/Ca (mmol/mol) STDEV SST (C) 150 212 5 4.00 0.32 26.1 212 250 5 4.19 0.19 26.7 250 300 10 4.35 0.23 27.1 300 355 7 4.37 0.19 27.1 355 425 7 4.57 0.30 27.6 425 500 6 4.83 0.27 28.2 >500 1 4.92 n/a 28.5 Table 4.3. Mg/Ca data versus si ze for G. ruber (pink) The data listed in this table represent the midpoint of n measurements made for each size fraction, where n is the number of replicate measurements for each size fraction. STDEV is the standard deviation among the n replicate measu rements. The SSTs listed are calibrated using the equation for G. ruber (pink), Mg/Ca=0.38exp(0.09*SST) (Anand et al., 2003). Size Fraction ( m) n Mg/Ca (mmol/mol) STDEV SST (C) 150 212 5 4.36 0.10 25.2 212 250 6 4.29 0.27 25.1 250 300 8 4.32 0.16 25.2 300 355 5 4.30 0.22 25.1 355 425 2 4.73 0.19 26.2 Table 4.4. Mg/Ca data versus size for G. ruber (white) The data listed in this table represent the midpoint of n measurements made for each size fraction, where n is the number of replicate measu rements for each size fraction. STDEV is the standard deviation among the n replicate measurements. The SSTs listed are calibrated using the equation for G. ruber (white), Mg/Ca=0.449exp(0.09*SST) (Anand et al., 2003). The relationship between test siz e and Mg/Ca is not as clear in the white variety of G. ruber In our samples, G. ruber (white) is not abundant enough in the >425 m size fractions to make a Mg/Ca measurement. The four size fractions between 150 m and 355 m (150 212 m, 212 250 m, 250 3 00 m and 300 355 m) have the same Mg/Ca value of 4.32mmol/mol (within the analytical error) (Figure 4.3.a). The Mg/Ca value of
72 the largest size fraction, 355 425 m, does increase significantly, by 0.41 mmol/mol, consistent with the increase in Mg/Ca for this size fraction in pink G. ruber Each of the size fractions between 212 m and 425 m in G. ruber (white) falls, within 1 # error, on the trend line relating Mg/Ca to test size for G. ruber (pink). However, looking at the test size vs. Mg/Ca data for ju st the G. ruber (white) data, there is no significant relationship between Mg/Ca and size. 4.5. Discussion of Size Fraction Data Unlike the positive test size 13 C relationship that has been observed for nearly all species of planktonic Foraminifera inv estigated (e.g. Curry and Williams, 1981; Ravelo and Fairbanks, 1995; Elderfield et al., 2002), the relationship between test size and the Mg/Ca and 18 O parameters varies between positive, negative and no relationship for a variety of planktonic Foraminif era (e.g. Curry and Matthews, 1981, Elderfield et al., 2002; Ni et al., 2007). Spero and Lea (1993) concluded from analysis of individual chambers that ontogeny has no significant effect on shell 18 O in two species of symbiont bearing planktonic Foramini fera ( O. universa and G. sacculifer ), therefore changes in metabolic rate are not likely to be the cause of 18 O size relationships. Likewise, in a study of intra test Mg/Ca variability in Globigerinoides ruber Sadekov et al. (2008) found no significant differences in the mean Mg/Ca values of the different chambers within a single test. This finding suggests that changing metabolic rates as a foraminiferan grows does not have a significant effect on shell Mg/Ca.
73 Figure 4.3. Relationship between Mg/ Ca and test size A) Mg/Ca is plotted against test size for pink (open circles) and white (black squares) G. ruber Error bars represent standard deviation among replicate measurements. A linear least squares regression is plotted through the data for G. ruber (pink). There is no significant relationship in G. ruber (white), but the values for G. ruber (white) are not significantly different from the values for G. ruber (pink). B) Size versus Mg/Ca for G. ruber (pink) is plotted from two different depths in the Pigmy Basin box core. The slope of the relationship is identical in the two different samples, and there is a 2C offset, which is consistent with the downcore Mg/Ca that suggests that SST was 2C cooler 650 yrs BP.
74 We hypothesize that changes in Mg/Ca and 18 O with test size observed for G. ruber in this study reflect differences in calcification temperature among the different size fractions. The increase in Mg/Ca and decrease in 18 O with increasing test size is consistent with larger indiv iduals calcifying in warmer waters than smaller individuals. In a culture experiment Spero and Lea (1993) demonstrated that the two symbiont bearing foraminifers, O. universa and G. sacculifer grow significantly larger under "high light" culture conditio ns than "low light" culture conditions. They also found that foraminiferal 18 O decreased with increasing light levels, most likely as a result of increased symbiont photosynthesis. In the marine environment, the highest light conditions are found at sha llow depths in the water column or during the summer season, so it follows that size fractions in G. ruber may vary in their depth/seasonal distribution in the GOM. We established that the increase in Mg/Ca with size was equivalent to a 0.66C increase in calcification temperature per 100 m increase in test size, with a total range of 2.5C among the size fractions of G. ruber (section 4.4.3.). The vertical temperature gradient in the upper 50 meters of the water column in the GOM is ~4C, and the seasona l range in mixed layer (0 30 meters) temperature is ~7C (Figure 4.5). Therefore, if we adopt the working hypothesis that the Mg/Ca change we observe with test size is a function of calcification temperature, it is reasonable to assume that the range in c alcification temperatures observed could result from differences in depth and/or seasonal distribution of the different sized Foraminifera.
75 In order to assess whether the change in Mg/Ca based calcification temperature with size is consistent with 18 O based calcification temperatures,, we use the following paleotemperature equation to convert 18 O to temperature, T=16.5 4.80* ( 18 O c 18 O sw ) (2) Where T is temperature (C), 18 O c is the measured 18 O of foraminiferal calcite, and 18 O w is the o xygen isotopic composition of seawater (converted from the VSMOW scale to VPDB by subtracting 0.27). Equation (2) is the paleotemperature equation developed for Orbulina universa under "high light" culture conditions by Bemis et al. (1998), and has been determined to be appropriate for use in G. ruber (Thunell et al., 1999). If we assume a constant 18 O w of 0.7 for the Gulf of Mexico mixed layer, the increase in 18 O calcification temperature with size is equal to a 1.3C increase in temperature for eac h 100 m increase in test size for both pink and white G. ruber This is twice as large as the temperature change with test size predicted from Mg/Ca in this study. However, if we assume a decrease in 18 O w of 0.06 with each larger test size (i.e. we ass ume a 18 O w of 0.7 for the 150 212 m size fraction, decreasing to 0.37 for the 425 500 m size fraction), then the estimated increase in calcification temperature with increasing size is equivalent using both Mg/Ca and 18 O paleotemperature equations. T his change in 18 O over the range of sizes is equivalent to the change in 18 O expected over the upper 50 meters of the water column, or over the annual cycle in the Gulf of Mexico.
76 Figure 4.4. Conversion of 18 O calcite to calcification temperature A) 18 O c data for G. ruber (white) is converted to calcification temperature using the paleotemperature equation for Orbulina universa "high light" from Bemis et al. (1998): T=16.5 4.80* (d 18 O c d 18 O sw ). The 18 O c data are converted to SST using a fixed 18 O sw value of 0.7 for each size fraction (open squares), and using a variable 18 O sw (solid squares). For the variable 18 O sw a value of 0.7 is assigned to the smallest size fraction (150 212mm) and the 18 O sw is decreased by 0.07 for each subsequen t size fraction. B) The same calculation is performed for G. ruber (pink). When the 18 O sw is fixed at 0.7, the resulting change in calcification temperature with size is 1.3C increase per 100 m increase in size. When the variable 18 O sw is used, the r esult is a change in calcification temperature of 0.66C per 100 m increase in size. Since 18 O c is a function of both calcification temperature and the oxygen isotopic composition of ambient seawater, we convert 18 O c to calcification temperature ag ain, but this time assuming a decrease of 0.07 in 18 O w for each larger size fraction (i.e. the 150 212 m size fraction calcifies in seawater with a 18 O of 0.7 , while the 212 250 m size fraction calcifies in 0.63 seawater, and so on). With this assume d change in 18 O w the increase in calcification temperature is equal to 0.66C per 100mm increase in test size. This is equivalent to the increase in calcification temperature with size estimated by Mg/Ca in this study.
77 Figure 4.5. Vertical pro files and seasonal cycles of temperature and 18 O sw variability in the GOM A) Summer weighted (April September) vertical profile of 18 O sw inferred from salinity data in the GOM. B) Seasonal cycle of mixed layer (upper 30 meters) 18 O sw inferred from sali nity in the GOM. The 18 O sw was calculated from salinity data from the Levitus (2004) data set, using the equation: 18 O sw = 0.557*salinity 19.98. C) Summer weighted (Apr. Sep.) vertical temperature profile for the GOM. D) Seasonal cycle of mixed layer (up per 30 meters) temperature for the GOM. Temperature data are also from the Levitus (2004) data set. 4.6. Comparison of downcore geochemical records for pink and white G. ruber Relative offsets in the Mg/Ca and 18 O of pink and white G. ruber have been exploited in downcore records to make inferences about changing seasonality in the past (e.g. Williams et al., 2009). This is based on the premise that G. ruber (white) reflects mean annual sea surface conditions, while the geochemical signal from G. rub er (pink) is summer weighted. We compare near modern core top data with Gulf of Mexico climatologic data to make inferences about depth/seasonal habitat preferences for pink
78 and white G. ruber Comparison of the downcore oxygen isotopic and Mg/Ca records of pink and white G. ruber from a high resolution sedimentary record spanning the past 1500 years provides additional insights into the differences between the two sub species of G. ruber White and pink G. ruber are the two most abundant species of pl anktonic Foraminifera in the Gulf of Mexico assemblages, comprising >45% of the total assemblage in core top sediments (Kennett, 1985). Both live in tropical to sub tropical surface waters, and are confined to the euphotic zone by their photoautotropic di noflagellate symbionts. Sediment trap data from twenty global sites suggest the optimum SST range for G. ruber (white) is 22 31¡C, while G. ruber (pink) has a more limited ideal range of 23 30¡C (&ari' et al., 2005). Plankton tow data from the Sargasso S ea reveal that white and pink G. ruber are most commonly found in SSTs ranging from 18 26¡C, with highest concentrations seen at 23 27¡C. The pink variety is found at even warmer temperatures up to 28¡C (B and Hamlin, 1967) Modern depth preferences of planktic Foraminifera such as G. ruber have been studied extensively in the Sargasso Sea yet little work has focused on GOM. Plankton tow results taken in November 1975 in the Sargasso Sea suggest that white G. ruber is commonly present in the top 100 m while the pink variety is found at low concentrations up to 200 meters (Fairbanks et al., 1980). Monthly plankton tows also from the Sargasso Sea showed that white and pink G. ruber were most abundant in the top 10 meters (Tolderlund and B, 1971). A si ngle tow in April 1980 in the western GOM revealed that the G. ruber (white) is present from 0 50 meters water depth, while the less abundant pink G. ruber is found slightly deeper from 25 50 meters (B, 1982).
79 From the limited sediment trap data availabl e for the subtropical Atlantic Ocean, white and pink G. ruber appear to have distinct seasonal distributions. Sediment trap data from the Sargasso Sea indicate that G. ruber (white) is present in high abundance throughout the year (with fluxes >50 tests m 2 day 1 during all months) (Deuser et al., 1981; Deuser, 1987; Deuser and Ross, 1989), suggesting that G. ruber (white) is representative of mean annual sea surface conditions in the sediment record. Data indicate that G. ruber (pink) exhibits peak abundan ces from April October (with fluxes of >5 tests m 2 day 1 ), and fluxes drop to nearly 0 during the winter (December April). A recent sediment trap study in the GOM (January December 2008) (Tedesco et al., 2009) shows that white G. ruber (white) is present throughout the year (although in much lower abundances than in corresponding core top sediments). The flux of G. ruber (pink) in the GOM sediment trap study is broadly consistent with the Sargasso Sea data, such that fluxes are low (near zero) during the coldest months (December April), while maximum fluxes occur during the summer months (July September). 4.6.1. Comparison of white and pink Mg/Ca records A Mg/Ca record was generated from the 250 300 m size fraction of G. ruber (pink) for comparison wi th the previously published Mg/Ca record from white G. ruber (Richey et al., 2007) from the Pigmy Basin. The raw downcore Mg/Ca values, as well as the overall pattern of variability are nearly identical ( 0.23 mmol/mol) for both pink and white G. ruber re cords. Centennial scale features, such as a 1.1 mmol/mol increase in Mg/Ca from ca. 300 yrs B.P. to the 20 th century, are present in both records. Additionally, the abrupt transition from elevated Mg/Ca values that occurs ca. 950 yrs
80 B.P., as well as the 3 multi decadal intervals of elevated Mg/Ca that occur between 1000 and 1400 yrs B.P., are present in both the pink and white G. ruber records. This indicates that the original Mg/Ca record from white G. ruber is highly reproducible from a second surface dwelling species from the same box core. Anand et al. (2003) determined that the temperature dependence (i.e. the exponential constant of the Mg/Ca paleotemperature equation) was approaching 9% increase in Mg/Ca per 1C for nearly all planktonic forami nifers in their study. However the pre exponential constants were determined to be different among species. Therefore, using the respective Mg/Ca paleotemperature equations for pink and white G. ruber results in a downcore SST record in which the variabi lity is the same, but the G. ruber (pink) is consistently 1.8C warmer than the G. ruber (white) record (Figure 4.6.a). If the assumption is made that there is a real difference in the distribution coefficient for Mg/Ca between white and pink G. ruber th e results suggest that G. ruber (pink) is consistently calcifying at significantly warmer SSTs than G. ruber (white) over the past 1400 years. This may be attributed to a seasonal distribution for G. ruber (pink) that is weighted toward warmer months of t he year, or G. ruber (pink) may be living at a shallower depth in the water column than G. ruber (white). In order to test whether a consistent 1.8C offset between the white and pink G. ruber downcore records can be attributed to differences in season al distribution, we compared the core top Mg/Ca SST of pink and white G. ruber with monthly instrumental data for the Gulf of Mexico. The mean annual SST for the Gulf of Mexico is 25.4C, while the mean summer SST (April October) is 27.0C ( Levitus, 2004 ) The downcore Mg/Ca records for pink and white G. ruber vary in step, with G. ruber (pink) consistently
81 recording 1.8C warmer SSTs when their respective paleotemperature equations are used. Therefore, this interpretation of differing seasonal distribu tion leads to the inference that the climatic variations observed in this 1400 year record indicate mean state changes, and not changes in seasonality. Alternatively, the 1.8C offset between white and pink G. ruber could be explained by a difference in m ean depth habitat. The mean temperature gradient across the upper 50 meters of the water column from April October (when the upper 50 meters of the water column in the northern Gulf of Mexico are not fully mixed) is ~4C (Levitus, 2003). Therefore it is plausible that G. ruber (pink) could be consistently calcifying at shallower depths than G. ruber (white), thus producing a downcore Mg/Ca SST record that is weighted 1.8C warmer. Our conclusion from this exercise is that it is possible to explain a cons istent 1.8C offset by either differences in preferred season or depth habitat.
82 Figure 4.6. Downcore comparison of Mg/Ca records for white and pink G. ruber A) Pink and white G. ruber Mg/Ca records are calibrated using their respective paleotemperatu re equations [Mg/Ca= 0.38exp(0.09*SST), and Mg/Ca=0.449exp(0.09*SST) from Anand et al., (2003)]. The mean summer (Apr Nov) SST and mean annual SST are indicated. B) Raw Mg/Ca records for pink and white G. ruber are plotted.
83 4.6.2. Comparison of downc ore 18 O data The 18 O of foraminiferal calcite is a function of both calcification temperature, and the oxygen isotopic composition of ambient seawater. Unlike the raw Mg/Ca records of pink and white G. ruber the 18 O calcite records do not covary. The major difference between the two 18 O calcite records occurs when there are excursions to more enriched 18 O calcite in white G. ruber record. During these excursions of increased 18 O calcite in white G. ruber (ca. 1000 yrs B.P., 600 yrs BP, etc.) the p ink G. ruber 18 O calcite record actually records a depletion in 18 O calcite (Figure 4.7). When the 18 O calcite record of G. ruber (white) is converted to a 18 O seawater record by removing the temperature effect on the 18 O calcite record (using equation 2) those excursions to more enriched 18 O calcite are dominated by increases in the 18 O seawater. The first 18 O seawater excursion occurs during a warm period (as recorded by the Mg/Ca SST of both pink and white G. ruber ), while the second excursion occurs d uring an SST minimum in the Mg/Ca records. Interestingly the % abundance of both pink and white G. ruber drop to their lowest over the 1500 year record (to <5% and <10%, respectively) during these inferred high salinity excursions. This is consistent with other studies in the Gulf of Mexico that show large drops in G. ruber abundance during high salinity events (LoDico et al., 2006).
84 Figure 4.7. Downcore comparison of 18 O data for white and pink G. ruber Downcore raw Mg/Ca is indicated by dashed l ines, and multi decadal smoothed record is illustrated with solid lines. Shaded bars indicate intervals of high salinity.
85 Figure 4.8. Comparison of G. ruber abundance with GOM salinity A) % abundance for G. ruber (pink). B) % abundance for G. rub er (white). C) 18 O seawater record for Pigmy Basin (Richey et al., 2007). 18 O seawater record in panel C. was generated using paired Mg/Ca 18 O measurements on the 250 300 m size fraction of the G. ruber (white), and using equations (1) and (2).
86 4.7. Conclusions This study presents a detailed assessment of the relationship between foraminiferal test size and geochemical parameters commonly used for paleoceanographic reconstruction (i.e. 13 C, 18 O and Mg/Ca) in the pink and white varieties of G. ruber A systematic relationship between foraminiferal test size and either 18 O or Mg/Ca has not been demonstrated in previous studies. However, this study finds a significant decrease in 18 O, and a significant increase in Mg/Ca with increasing test size for pink and white G. ruber We hypothesize that these relationships are the result of larger individuals calcifying at higher temperatures then smaller individuals. An increase in calcification temperature of 0.66C per 100 m increase in size is found when both Mg/Ca and 18 O data are converted to temperature using their respective paleotemperature equations, providing an internally consistent explanation for these observations. The proposed increase in calcification temperature with increasing test size i s likely the result of differences in the seasonal and/or depth distribution of different size fractions (i.e. white and pink G. ruber grow larger during the summer and/or at depths closest to the surface). The overall range of calcification temperature ( ~2.5C) over the entire range of size fractions is reasonable, given the seasonal range of temperature and/or the thermal gradients in the mixed layer for the northern Gulf of Mexico. Although many studies suggest that ontogeny plays a role in the 13 C size relationship, the enrichment in 13 C that we observe with increasing size is also consistent with a symbiont photosynthesis influence on the increase in 13 C with size. Larger individuals living at shallower depths and/or with a warmer seasonal bias, would lead to
87 enhanced symbiont photosynthesis with increasing size, resulting in increasing 13 C enrichment with increasing size. This study also presents a downcore comparison of a decadal resolution 18 O and Mg/Ca records for white and pink G. ruber from co occurring sediments in the Pigmy Basin. Results indicate an offset in the raw 18 O of pink and white G. ruber while the raw Mg/Ca of the two records is identical within analytical error. Using a 18 O paleotemperature equation indicates that the average 18 O depletion of 0.27 in G. ruber (pink) relative to the white variety corresponds to G. ruber (pink) having a calcification temperature that is ~1.20.4C warmer than G. ruber (white). Although the raw Mg/Ca values are the same for both pink a nd white G. ruber using their respective Mg/Ca paleotemperature equations results in the pink variety being offset by 1.80.8C warmer than the white variety. This supports the hypothesis that pink G. ruber is consistently calcifying in warmer waters tha n white G. ruber either due to a more summer weighted seasonal distribution, or a shallower depth habitat.
88 Chapter 5 Summary 5.1. Conclusions The Little Ice Age and Medieval Warm Period are two of the most prominent climatic events of the pa st millennium, recorded in climate proxy records throughout the mid to high latitude Northern Hemisphere. However, there are still gaps in our knowledge of the magnitude and spatial extent of these events, the low latitude oceans being an area from which there are very few proxy records. In this thesis I contribute to the understanding of sub tropical Atlantic Ocean variability by presenting multiproxy records of sea surface temperature variability from three different sites in the Gulf of Mexico (GOM). Data from the GOM indicate a significant Little Ice Age cooling of 2C. This cooling is observed in the foraminiferal Mg/Ca records from three different GOM basins, in the Mg/Ca records of two different planktonic foraminifers in the same basin (pink and white Globigerinoides ruber ), and among the SST signals recorded in an inorganic and molecular organic proxy from the same GOM basin (Mg/Ca and TEX 86 ). The similarities in the timing and magnitude of the Little Ice Age cooling among the three different GO M Basins suggests that the Mg/Ca proxy is recording a robust, reproducible regional climate signal in these basins. Comparison of the Mg/Ca SST record with the TEX 86 SST record from Pigmy Basin reveals significant similarities between the two proxy reco rds. For instance, the timing and magnitude of Little Ice Age cooling are similar for the two proxy records.
89 There are also important differences between the two records, which I hypothesize result from differences in the seasonal/depth distribution of Foraminifera and marine Crenarchaeota. These offsets are exploited to make inferences about changing seasonality/mixed layer depth over the past 1,000 years. For example, the offset between TEX 86 and Mg/Ca is minimal during the medieval Warm period, sugg esting that there was reduced seasonality and/or increased mixed layer depth during that time. Maximum offset between the two records during the Little Ice Age suggests enhanced seasonality and/or a shallow mixed layer during that time. In addition to the paleoclimate records presented in this thesis, I present a detailed study on the effect of foraminiferal test size on the different geochemical parameters commonly used for paleoenvironmental reconstruction. The change in 13 C, 18 O and Mg/Ca over the size range from 150 m to 500 m of the pink and white variety of G. ruber is larger than the observed variability in each of these parameters over the past 1,000 years. There is an increase in Mg/Ca and decrease in 18 O with increasing size that equates t o a 0.66C increase in calcification temperature with 100 m increase in shell size. These data suggest that larger individuals are calcifying under warmer conditions, and may be weighted toward a warmer season or shallower depth habitat than the smaller i ndividuals. This study emphasizes the importance of using a well constrained size range of Foraminifera for paleoenvironmental reconstruction. 5.2. Future Research Directions The work presented in this dissertation has provided a number of importan t answers regarding climate variability in the subtropical Atlantic Ocean over the past
90 millennium, but has also led to further questions. These additional questions generate new potential research directions that could be pursued in the future. Below is a brief outline of those potential research projects that would further develop or complement the research presented in this dissertation. In this dissertation, I make large assumptions about the ecologies of planktonic Foraminifera and Crenarchaeota i n the Gulf of Mexico. These assumptions have significant implications for the paleoclimatic interpretations, and thus it is important to determine the modern depth habitat and seasonal distribution for these proxy recorders in the Gulf of Mexico. Sedimen t trap and water column filtration studies in the GOM are essential to better understanding the ecology of the biogenic proxies used in paleoenvironmental reconstruction. Globally, the TEX 86 signal corresponds to mean annual sea surface temperatures. How ever, the TEX 86 data presented in this dissertation from the Pigmy Basin are summer weighted. Due to the delivery of large amounts of terrestrial organic matter to the Pigmy Basin via the Mississippi River, it is possible that the TEX 86 signal is being in fluenced by terrestrially derived isoprenoid GDGTs. In order to determine whether or not the Pigmy Basin TEX 86 is being biased toward warmer temperatures by terrestrially derived GDGTs, I propose to analyze the GDGT composition of surface sediments across a transect of surface sediments from the Mississippi River Delta to the Pigmy Basin on the continental slope. The molecular compounds commonly used for paleo environmental reconstruction (e.g. alkenones, GDGTs, fatty acids, n alkanes, etc.) are contained within the fine sediment fraction (<63 m), and can have different transport mechanisms to
91 depositional basins than Foraminifera. In order to directly compare downcore records of sea surface variability derived from both Foraminifera and molecula r fossils, it is best to have an independent chronology based on compound specific radiocarbon dating of the molecular fraction. Molecular radiocarbon dating of the isoprenoid GDGTs used for the TEX 86 index would verify that the TEX 86 and Mg/Ca records in the Pigmy Basin are contemporaneous. Isolation and radiocarbon dating of terrestrially derived compounds (e.g. long chain n alkanes and fatty acids) from Gulf of Mexico sediments can provide information about residence times of terrestrial carbon on the North American continent, as well as sediment transport to marine depositional basins. There are very few records of terrestrial climate variability from the subtropical Atlantic region. Lake Tulane, located in central Florida ( 27.5853N, 81.5034W ), is a high sedimentation rate lake with a small catchment basin. It is an ideal setting for comparing terrestrial temperature and hydrologic variability in an environment that is closely linked to the GOM. I plan to use the TEX 86 proxy to reconstruct lake s urface temperature. Additionally, the hydrogen isotopic composition ( D) of terrestrially derived fatty acids can be used to reconstruct regional hydrologic variability.
92 R eferences Ammann, C.N., F. Joos, D.S. Schimel, B.L. Otto Bliesner, and R. A. Tomas (2007), Solar influence on climate during the past millennium: Results from transient simulations with the NCAR Climate System Model, PNAS 104(10), 3713 3718, doi:10.1073/pnas0605064103. Anand, P., H. Elderfield, and M.H. Conte (2003), Cali bration of Mg/Ca thermometry in planktonic foraminifera from a sediment trap time series, Paleoceanography 18, 1050, doi: 10.1029/2002PA000846. Bard, E., G. Raisbeck, F. Yiou, and J. Jouzel (1997), Solar modulation of cosmogenic nucide production ove r the last millennium: comparison between 14 C and 10 Be records, Earth and Planetary Science Letters 150, 453 462. Bard, E., G. Raisbeck, F. Yiou, and J. Jouzel (2000), Solar irradiance during the last 1200 years based on cosmogenic nuclides. Tellus 5 2B: 985 992. Barker, S., M. Greaves, and H. Elderfield (2003), A study of cleaning procedures used for foraminiferal Mg/Ca paleothermometry, Geo chem. Geophys. Geosyst ., 4, doi:10.1029/2003GC000559. B, A.W.H. (1982), Biology of planktonic foraminifer a, in Foraminifera: notes for a short course edited by T. W. Broadhead, pp. 51 92, Department of Geological Sciences, Knoxville, Tennessee. B, A.W.H., and W. H.Hamlin (1967), Ecology of Recent Planktonic Foraminifera: Part 3: Distribution in the Nort h Atlantic during the Summer of 1962, Micropaleontology 13 (1), 87 106. Berger, W.H., J.S. Killingley, and E. Vincent (1978), Stable isotopes in deep sea carbonates: Box core ERDC 92, west equatorial Pacific, Oceanol. Acta 1, 203 216. Berger, W.H., M .C. Bonneau, and F.L. Parker (1982), Foraminifera on the deep sea floor: Lysocline and dissolution rate, Oceanol. Acta 5, 249 258. Black, D.E., M.A. Abahazi, R.C. Thunell, A. Kaplan, E.J. Tappa, and L.C. Peterson (2007), An 8 century tropical Atlantic SST record from the Cariaco Basin: Baseline variability, twentieth century warming, and Atlantic hurricane frequency, Paleoceanography 22, PA4204, doi:10.1029/2007PA001427.
93 Bond, G., W. Showers, M. Cheseby, R. Lotti, P. Almasi, P. deMenocal, P. Priore, H. Cullen, I. Hajdas, and G. Bonani (1997), A Pervasive Millenial Scale Cycle in North Atlantic Holocene and Glacial Climates, Science 278, doi:10.1126/science.278.5341.1257. Boyle, E. A. (1983), Manganese carbonate overgrowths on foraminifera tests Earth Planet. Sci. Lett ., 53, 11 35. Brassell, S.C., G. Eglinton, I.T. Marlowe, U. Pflaumann, and M. Sarnthein (1986), Molecular stratigraphy: A new tool for climatic assessment, Nature 320, 129 133, doi:10.1038/320129a0. Broecker, W.S., and T.H. P eng (1982), Tracers in the Sea Lamont Doherty Geological Observatory, Palisades, N. Y. Broecker, W.S. (2001), Was the Medieval Warm Period Global?, Science 291, 1497 1499. Brown, S.J., and H. Elderfield (1996), Variations in Mg/Ca and Sr/Ca ratios of planktonic foraminifera caused by postdepositional dissolution: Evidence of shallow Mg dependent dissolution, Paleoceanography 11, 543 551, 1996. Brunner, C.A. (1979), Distribution of planktonic Foraminifera in surface sediments in the Gulf of Mexico Micropaleontology 25, p.325 335. Castaeda, I.S., E. Schefu§, J. Ptzold, J.S. Sinninghe Damst, S. Weldeab, and S. Schouten (2010), Millennial scale sea surface temperature changes in the eastern Mediterranean (Nile River Delta region) over the last 27,000 years, Paleoceanography 25, PA1208, doi:10.1029/2009PA001740. Cook E.R., C. Woodhouse, C.M. Eakin, D. M. Meko, and D.W. Stahle (2004), Lo ngterm aridity changes in the western United States, Science 306, 1015 1018. Cook, E.R., R. Seager, R.R. Heim Jr., R.S. Vose, C. Herweijer, and C. Woodhouse (2010), Megadroughts in North America: placing IPCC projections of hydroclimatic change in a l ong term palaeoclimate context, J. Quaternary Sci ., 25, 48 61, ISSN 0267 8179. Crowley, T.J, and T. Lowery (2000), How Warm was the Medieval Warm Period?, Ambio v. 29, p. 51 54. Crowley, T.J. G. Zielinski, B. Vinther, R. Udisti, K. Kreutz, J. Cole Da i, and E. Castellano (2008), Volcanism and the Little Ice Age, PAGES News 16, 22 23.
94 Curry, W.B., and R.K. Matthews (1981), Equilibrium 18 O fractionation in small size fraction planktic foraminifera: Evidence from recent Indian Ocean sediments, Marin e Micropaleontology 6, 327 337. Dekens, P.S., D.W. Lea, D.K. Pak, and H.J. Spero (2002), Core top calibration of Mg/Ca in tropical foraminifera: Refining paleotemperature estimation, Geo chem. Geophys. Geosyst ., v. 3, doi: 10.129/2001GC000200. deMeno cal, P., J. Ortiz, T. Guilderson, and M. Sarnthein (2000), Coherent High and Low Latitude Climate Variability During the Holocene Warm Period, Science 288(5474), 2198 2202. Deuser, W.G., E.H. Ross, C. Hemleben, and M. Spindler (1981), Seasonal change s in species composition, numbers, mass, size, and isotopic composition of planktonic foraminifera settling into the deep Sargasso Sea, Palaeogeography Palaeoclimatology Palaeoecology 33, 103 127. 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 significance, Journal of Foraminiferal Research 17, 14 27. Deuser W.G., and E.H. Ross (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. Dowsett, H.J., S. Verardo, and R.Z. Poore (2003), Gulf of Mexico planktonic foraminifer transfer function GOM2: Preliminary Report, USGS Open File Report OF 03 61. Eggins, S.M., A.Y. Sadekov, and P. De Deckker (2004), Modulation and daily banding of Mg/Ca in Orbulina u niversa tests by symbiont photosynthesis and respiration: A complication for seawater thermometry?, Earth Planet. Sci. Lett ., 225, 411 419. Eglinton, T.I., and G. Eglinton (2008), Molecular proxies for paleoclimatology, Earth and Planetary Science Lette rs 275, 1 16, doi:10.1016/j.epsl.2008.07.012. Elderfield, H., and G. Ganssen (2000), Past temperature and 18 O of surface ocean waters inferred from foraminiferal Mg/Ca ratios, Nature 405, 442 445. Elderfield, H., M. Vautravers, and M. Cooper (2002), The relationship between shell size and Mg/Ca, Sr/Ca, d 18 O and d 13 C of species of planktonic foraminifera, Geochemistry Geophysics Geosystems 3, doi: 10.1029/2001GC000194.
95 Elderfield, H., J. Yu, P. Anand, T. Kiefer, B. Nyland, (2006), Calibrations for benthic foraminiferal Mg/Ca paleothermometry and the carbonate ion hypothesis. Earth Planet. Sci. Lett ., 250 (3 4), 633 649. Emiliani, C. (1954), Depth habitats of some species of pelagic foraminifera as indicated by oxygen isotope ratios, American Jo urnal of Science 252, 149 158. Emiliani, C. (1955), Pleistocene temperatures, Journal of Geology 63, 538 578. Englebrecht, A.C., and J.P. Sachs (2005), Determination of sediment provenance at drift sites using hydrogen isotopes and unsaturation ratio s in alkenones, Geochimica et Cosmochimica Acta 69, 4253 4265. Esper, J., E.R. Cook, and F.H. Schweingruber (2002), Low frequency signals in long tree ring chronologies for reconstructing past temperature variability, Science 295, 2250 2254, doi: 10 .1126/science.1066208. Fairbanks, R.G., P.H. Wiebe, and A.W.H. B (1980), Vertical distribution and isotopic composition of living planktonic Foraminifera in the western North Atlantic, Science 207, 61 63. Ferguson, J.E., G.M. Henderson, M. Kucera, an d R.E.M. Rickaby (2008), Systematic change of foraminiferal Mg/Ca rations across a strong salinity gradient, Earth Planet. Sci. Lett ., 265, doi:10.1016/j.epsl.2007.10.011. Flannery, J.A., D.J. Hollander, J.N Richey, and B.P. Flower (2008), A 1400 Year Sedimentary Record from the Pigmy Basin, Gulf of Mexico Reveals Strong Decadal Scale Linkages Between Solar Variability, Gulf of Mexico Moisture Balance, and Hydrologic Conditions over the North American Continent, Eos Trans. AGU, 89 (53), Fall Meet. Sup pl., Abstract PP13E 06. 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, doi: 10.1130/G20604.1. Frhlich, C. and J. Lean (2004), Solar Radiative Output and its Variability: Evidence and Mechanisms, Astron. and Astrophys. Rev. 12 273 -320, doi: 10.1007/s00159 004 0024 1. Gao, C., A. Robock, and C. Ammann (2008), Volcanic forcing of climate over the past 1500 years: An improved ice corebased index for climate models, Journal of Geophysical Research 113, D23111, doi:10.1029/2008JD010239.
96 Gischler, E., E A. Shinn, W. Oschmann, J. Fiebig, and N.A. Buster (2008), A 1500 year Holocene Caribbean climate archive from the Blue Hole, Lighthouse Reef, Belize, Journal of Coastal Research 24, 1495 1505, doi: 10.2112/07 0891.1. Goodkin, N.F., K.A. Hughen, A.L. Cohen, and S.R. Smith (2005), Record of Little Ice Age sea surface temperatures at Bermuda using a growth dependent calibration of coral Sr/Ca, Paleoceanography 20, PA4016, doi:10.1029/2005PA001140. Haase Schramm, A., F. Bhm, A. Eisenhauer, W. C. Dullo, M.M. Joachimski, B. Hansen, and J. Reitner (2003), Sr/Ca ratios and oxygen isotopes from sclerosponges: Temperature history of the Caribbean mixed later and thermocline during the Little Ice Age, Paleoceanography ,18, doi:10.1029/2002PA000830. Haase Schramm, A., F. Bhm, A. Eisenhauer, D. Garbe Schnberg, W. C. Dullo, and J. Reitner (2005), Annual to interannual temperature variability in the Caribbean during the Maunder sunspot minimum, Paleoceanography, 20, PA4015, doi: 10.1029/2005PA001137. Hastings, D.W., A.D. Russell, and S.R. Emerson (1998), Foraminiferal magnesium in Globeriginoides sacculifer as a paleotemperatu re proxy, Paleoceanography 13, 161 169. Hegerl, G.C., T.J. Crowley, S.K. Baum, K. Y Kim, and W.T. Hyde (2003), Detection of volcanic, solar, and greenhouse gas signals in paleo reconstructions of Northern Hemispheric temperature. Geophysical Reearch L etters 30, 1242, doi:10.1029/2002GL016635. Hegerl, G.C., T.J. Crowley, W.T. Hyde, and D.J. Frame (2006), Climate sensitivity constrained by temperature reconstructions over the past seven centuries, Nature 440, doi:10.1038/nature04679. Hodell, D.A., M. Brenner, J.H. Curtis, R. Medina Gonzales, E. Ildefonso Chan Can, A. Albornaz Pat, and T. P. Guilderson (2005), Climate change on the Yucatan Peninsula during the Little Ice Age, Quaternary Research 63, 109 121, doi:10.1016/j.yqres.2004.11.004. E.C. Hopmans, S. Schouten, R.D. Pancost, M.J.T. van der Meer, J.S. Sinninghe Damst (2000), Analysis of intact tetraether lipids in archaeal cell material and sediments using high performance liquid chromatography/atmospheric pressure ionization mass spectr ometry, Rapid Commun. Mass Spectrom ., 14, 585 589. Hopmans, E.C., J.W.H. Weijers, E. Schefu§, L. Herfort, J.S. Sinninghe Damst, and S. Schouten (2004), A novel proxy for terrestrial organic matter in sediments based on branched and isoprenoid tetraeth er lipids, Earth and Planetary Science Letters 224, 107 116, doi:10.1016/j.epsl.2004.05.012.
97 Huang, S., H.N. Pollack, and P.Y. Shen (1997), Late Quaternary temperature change seen in world wide continental heat flow measurements, Geophysical Research Letters 24, 1947 1950. Huguet, C., J. H. Kim, J.S.S. Damst, and S. Schouten (2006), Reconstruction of sea surface temperature variations in the Arabian Sea over the last 23 kyr using organic proxies (TEX 86 and U K' 37 ), Paleoceanography 21, PA3003, do i:10.1029/2005PA001215. IPCC (Intergovernmental Panel on Climate Change) (2001), Climate Change 2001: The Scientific Basis. Cambridge University Press, Cambridge, UK. Jasper, J.P. and R.B. Gagosian (1989), Alkenone molecular stratigraphy in an oceanic environment affected by glacial freshwater events, Paleoceanography 4, 603 614. Jenkyns, H., A. Forster, S. Schouten, and J.S. Sinninghe Damste (2004), High temperatures in the Late Cretaceous Arctic Ocean, Nature 432, 888 891. Jones, P.D., M. New, D .E. Parker, S. Martin, and I.G. Rigor (1999), Surface Air Temperature and its Changes Over the Past 150 Years, Rev. Geophys. 37, 173 199. Karner M., E.F. DeLong, and D.M. Karl (2001), Archaeal dominance in the mesopelagic zone of the Pacific Ocean. N ature 409, 507 510. Keigwin, L. D. (1996), The Little Ice Age and Medieval warm period in the Sargasso Sea, Science 274, 1504 1508. Kennett, J.P., K. Elmstrom, and N. Penrose (1985), The last deglaciation in Orca Basin, Gulf of Mexico: High resolut ion planktonic foraminiferal changes, Paleogeography Paleoclimatology Paleoecology 50, 189 216. Kilbourne, K.H., T.M. Quinn, R. Webb, T. Guilderson, J. Nyberg, and A. Winter (2008), Paleoclimate proxy perspective on Caribbean climate since the year 17 51: Evidence of cooler temperatures and multidecadal variability, Paleoceanography 23, PA3220, doi: 10.1029/2008PA001598. Kim, J H., S. Schouten, E.C. Hopmans, B. Donner, and J.S. Sinninghe Damst (2008), Global sediment core top calibration of the T EX 86 paleothermometer in the ocean, Geochimica et Cosmochimica Acta 72, 1154 1173, doi:10.1016/j.gca.2007.12.010.
98 Kisakrek, B., A. Eisenhauer, F. Bhm, D. Garbe Schnberg, and J. Erez (2008), Controls on shell Mg/Ca and Sr/Ca in cultured planktonic f oraminiferan, Globigerinoides ruber (white), Earth Planet. Sci. Lett. 273 260 269. Lea, D.W., T.A. Mashiotta, and H.J. Spero (1999), Controls on magnesium and strontium uptake in planktonic foraminifera determined by live culturing, Geochem. Cosmochim Acta 63, 2369 2379. Lea, D.W. (2003), Elemental and isotopic proxies of past ocean temperatures. In: Elderfield, H. (Ed.), The Oceans and Marine Geochemistry Elsevier, New York, NY, pp. 365 390. Levitus, S. (2003), National Oceanographic Data Cente r World Ocean Atlas 1994, at http://www.cdc.noaa.gov/, Clim. Data Cent., Boulder, Colo. Lipp, J.S., Y. Morono, F. Inagaki, and K U. Hinrichs (2008), Significant contribution of Archaea to extant biomass in marine subsurface sediments, Nature 454, 991 9 94, doi:10.1038/nature07174. LoDico, J.M., B.P. Flower, and T.M. Quinn (2006), Subcentennial scale climatic and hydrologic variability in the Gulf of Mexico during the early Holocene: Paleoceanography 21, PA3015, doi: 10.1029/2005PA001243. Lund, D.C. and W.B. Curry (2004), Late Holocene variability in Florida Current surface density: Patterns and possible causes, Paleoceanography 19, PA4001, doi:10.1029/2004PA001008. Lund, D.C., and W. Curry (2006), Florida current surface temperature and salini ty variability during the last millennium, Paleoceanography 21, PA2009, doi: 10.1029/2005PA001218. Lund, D.C., J. Lynch Stieglitz, and W. Curry (2006), Gulf Stream density structure and transport during the past millennium, Nature 444, doi:10.1038/na ture05277. Maash, K.A., P.A. Mayewski, E.J. Rohling, J.C. Stager, W. Kaln, L.D. Meeker, and E.A. Meyerson (2005), A 2000 year context for modern climate change, Geogr. Ann., Ser. A, Phys. Geogr ., 87, 7 15, doi:10.1111/j.0435 3676.2005.00241.x. Mann, M.E., R.S. Bradley, and M.K. Hughes (1998), Global scale temperature patterns and climate forcing over the past six centuries, Nature 392, 779 787. Mann, M.E., R.S. Bradley, and M.K. Hughes (1999), Northern Hemisphere temperatures during the past mill ennium: Interferences, uncertainties, and limitations, Geophysical Research Letters 26, 759 762.
99 Mann, M.E. and P.D. Jones (2003), Global surface temperatures over the past two millennia. Geophysical Research Letters 30, 1820. Marshall, J., Y. Kushn ir, D.S. Battisti, P. Chang, A. Czaja, R. Dickson, J. Hurrell, M. McCartney, R. Saravanan, and M. Visbeck (2001), North Atlantic climate variability: Phenomena impacts and mechanisms, Int. J. Climatol ., 21, 1863 1898. doi:10.1002/ioc.693. Meeker, L.D., and P. A. Mayewski (2002), A 1400 year high resolution record of atmospheric circulation over the North Atlantic and Asia, Holocene 12, 257 266, doi:10.1191/0959683602hl542ft. Moberg, A., D.M. Sonechkin, K. Holmgren, N.M. Datsenko, and W. Karln (2005 ), Highly variable Northern Hemisphere temperatures reconstructed from low and high resolution proxy data, Nature 433, 613 617, doi:10.1038/nature03265. Mollenhauer G., M. Kienast, F. Lamy, H. Meggers, R.R. Schneider, J.M. Hayes, and T.I. Eglinton (20 05), An evaluation of C 14 age relationships between co occurring foraminifera, alkenones, and total organic carbon in continental margin sediments. Paleoceanography 20, doi:10.1029/2004PA001103. Mollenhauer G., M. Inthorn, T. Vogt, M. Zabel, J.S. Sinn inghe Damst, and T.I. Eglinton (2007), Aging of marine organic matter during cross shelf lateral transport in the Benguela upwelling system revealed by compound specific radiocarbon dating, Geo chem. Geophys. Geosyst., 8, doi:10.1029/2007GC001603. Na tional Research Council. Surface Temperature Reconstructions for the Last 2,000 Years (National Academy Press, 2006); available at
100 Nyberg, J., B.A. Malmgren, A. Kuijpers, and A. Winter (2002), A centennial scale variability of tropical North Atlantic surface hydrography during the late Holocene Palaeogeography Palaeoclimatology Palaeoecol ogy 183, 25 41. Ohkouchi N., T.I. Eglinton, L.D. Keigwin, and J.M. Hayes (2002), Spatial and temporal offsets between proxy records in a sediment drift, Science 298, 1224 1227. Ohkouchi, N., L. Xu, C.M. Reddy, D. Montluon, and T.I. Eglinton (2005), Radiocarbon dating of alkenones from marine sediments: I. Isolation Protocol, Radiocarbon 47, 401 412. Oppo, D.W., and R.G. Fairbanks (1989), Carbon isotope composition of tropical surface water during t he past 22,000 years, Paleoceanography 4, 333 351. Oppo, D.W., Y. Rosenthal, and B.K. Linsley (2009), 2,000 year long temperature and hydrology reconstructions from the Indo Pacific warm pool, Nature 460, 1113 1116, doi:10.1038/nature08233. Overpeck, J.T., K. Hughen, D. Hardy, R. Bradley, R. Case, M. Douglas, B. Finney, K. Gajewski, G. Jacoby, A. Jennings, S. Lamoureux, A. Lasca, G. MacDonald, J. Moore, M. Retelle, S. Smith, A. Wolfe, G. Zielinski (1997), Artic environmental change of the last four centuries, Science 278, 1251 1256, doi:10.1126/science.278.5341.1251. Pelejero, C., and E. Calvo (2003), The upper end of the U K' 37 temperature calibration revisited, Geochem. Geophys. Geosyst ., 4(2)1014, doi:10.1029/2002GC000431. Pena, L.D., E. Calv o, I. Cacho, S. Eggins, and C. Pelejero (2005), Identification and removal of Mn Mg rich contaminent phases on foraminiferal tests: Implications for Mg/Ca past temperature reconstructions, Geo chem. Geophys. Geosyst ., 6, doi:10.1029/2005GC000930. Prahl F. G. and S.G. Wakeham (1987), Calibration of unsaturation patterns in long chain ketone compositions for paleotemperature assessment, Nature, 330, 367 369. Ravelo, C.A., and R.G. Fairbanks (1992), Reconstructing the photic zone temperature range usin g 18 O measured on multiple species of planktonic foraminifera. Paleoceanography 7, 815 832. Ravelo, A.C., and R.G. Fairbanks (1995), Carbon isotope fractionation in multiple species of planktonic foraminifera from coretops in the tropical Atlantic, Jo urnal of Foraminiferal Research 25, 53 74.
101 Richey, J.N., R.Z. Poore, B.P. Flower, and T.M. Quinn (2007), 1400 yr multiproxy record of climate variability from the northern Gulf of Mexico, Geology 35, 423 426, doi:10.1130/G23507A.1. Richey, J.N., R.Z Poore, B.P. Flower, and T.M. Quinn, and D.J. Hollander (2009), Regionally coherent Little Ice Age cooling in the Atlantic Warm Pool, Geophys. Res.Lett., 36, L21703, doi:10.1029/2009GL040445. Robock, A., (2000), Volcanic eruptions and their impact on c limate, Earth in Space 12, 9 10. Rosenheim, B.E., P.K. Swart, S.R. Thurrold, P. Willenz, L. Berry, and C. Latkoczy (2004), High resolution Sr/Ca records in sclerosponges calibrated to temperature in situ, Geology 32, 145 148. Rosenthal, Y., and E. A. Boyle (1993), Factors controlling the fluoride content of planktonic foraminifera: An evaluation of its paleoceanographic applicability, Geochim. Cosmochim. Acta 57, 335 346. Rosenthal, Y., E.A. Boyle, and N. Slowey (1997), Temperature control on th e incorporation of magnesium, strontium, fluorine, and cadmium into benthic foraminiferal shells from Little Bahama Bank: Prospects for thermocline paleoceanography, Geochem. Cosmochim. Acta 61, 3633 3643. Rosenthal, Y., et al. (2004), Interlaboratory comparison study of Mg/Ca and Sr/Ca measurements in planktonic foraminifera for paleoceanographic research, Geochem. Geophys. Geosyst ., 5, Q04D09, doi:10.1029/2003GC000650. Russell, A.D., S. Emerson, B.K. Nelson, J. Erez, and D.W. Lea (1994), Uranium i n foraminiferal calcite as a recorder of seawater uranium concentrations, Geochim. Cosmochim. Acta 58, 671 681. Russell, A., B. Hnisch, H. Spero, D. Lea (2004), Effects of seawater carbonate ion concentration and temperature on shell U, Mg, and Sr in cultured planktonic foraminifera. Geochim. Cosmochim. Acta 68, 4347 4361. Sadekov, A., S.M. Eggins, P. De Deckker, and D. Kroon (2008), Uncertainties in seawater thermometry deriving from intratest and intertest Mg/Ca variability in Globigerinoides r uber Paleoceanography 23, PA1215, doi:10.1029/2007PA001452. Saenger, C., A.L. Cohen, D.W. Oppo, R.B. Halley, and J.E. Carilli (2009), Surface temperature trends and variability in the low latitude North Atlantic since 1552, Nature Geoscience 2, 492 495, doi: 10.1038/NGEO552.
102 Schouten, S., E.C. Hopmans, E. Schefu§, and J.S. Sinninghe Damst (2002), Distributional variations in marine crenarchaeotal membrane lipids: A new tool for reconstructing ancient sea water temperatures?, Earth Planet. Sci. Le tt ., 204, 265 274. Shah, S.R., G. Mollenhauer, N. Ohkouchi, T.I. Eglinton, and A. Pearson (2008), Origins of archaeal tetraether lipids in sediments: Insights from radiocarbon analysis, Geochimica et cosmochimica Acta 72, 4577 4594, doi:10.1016/j.gc a.2008.06.021. Shindell, D.T., G. Faluvegi, R.L. Miller, G.A. Schmidt, J.E. Hansen, and S. Sun (2006), Solar and anthropogenic forcing of tropical hydrology, Geophys. Res. Lett. 33, L24706, doi:10.1029/2006GL027468. Sluijs, A., S. Schouten, T.H. Donde rs, P.L. Schoon, U. Rhl, G.J. Reichart, F. Sangiorgi, J. H. Kim, J.S. Sinninghe Damst and H. Brinkhuis (2009), Warm and wet conditions in the Arctic region during Eocene Thermal Maximum 2, Nature Geosciences 2, 777 780, doi:10.1038/NGEO668. Spero, H. J., and D.W. Lea (1993), Intraspecific stable isotope variability in the planktic foraminifera Globigerinoides sacculifer: Results from laboratory experiments, Marine Micropaleontology 22, 221 234. Spero, H.J., and S.L. Parker (1985), Photosynthesis in the symbiotic planktonic foraminifer Orbulina universa and its potential contribution to oceanic primary productivity, Journal of Foraminiferal Research 15, p. 273 281, doi: 10.2113/gsjfr.15.4.273. Spero, H.J., and D.F. Williams (1998), Extracting e nvironmental information from planktonic foraminiferal 13 C data, Nature 375, 717 719. Stuiver, M., P.J. Reimer, E. Bard, J.W. Beck, G.S. Burr, K.A. Hughen, B. Kromer, F. G. McCormac, J. Van Der Plicht, and M. Spurk (1998), INTCAL98 radiocarbon age c alibration 24,000 0 cal BP, Radiocarbon v. 40, p. 1041 1083. Tedesco, K. A., J.W. Spear, E. Tappa, and R.Z. Poore (2009), Seasonal flux and assemblage composition of planktic foraminifera from the northern Gulf of Mexico: U.S. Geological Survey Open F ile Report 2009 1293 19 p. Tedesco, K.A, and R.C. Thunell (2003), Seasonal and interannual variations in planktonic foraminiferal flux and assemblage composition in the Cariaco Basin, Venezuela, Journal of Foraminiferal Research 33, 192 210.
103 Thunel l, 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. Tierney, J.E., J.M. Russell, Y. Huang, J.S. Sinninghe Damst, E.C. Hopmans, and A.S. Cohen (2008), Northern Hemisphere controls of tropical Southeast African climate during the past 60,000 years, Science 322, 252 255. Tolderlund, D.S., and A.W.H. B (1971), Seasonal distribution of Foramin ifera in the western North Atlantic, Micropaleontology 17, 297 329. Trouet, V., J. Esper, N.E. Graham, A. Baker, J.D. Scourse, and D.C. Frank (2009), Persistent Positive North Atlantic Oscillation Mode Dominated the Medieval Climate Anomaly, Science 324, doi: 10.1126/science.1166349. Urey, H.C., (1947), The thermodynamic properties of isotopic substances, Journal of the Chemical Society, 562 581. Wang, C., D.B. Enfield, S. K. Lee, and C.W. Landsea (2006), Influences of the Atlantic warm pool on We stern Hemisphere summer rainfall and Atlantic hurricanes, J. Clim ., 19, 3011 3028, doi:10.1175/JCLI3770.1. Wang, C., S. K. Lee, and D.B. Enfield (2008a), Climate response to anomalously large and small Atlantic warm pools during the summer, J. Clim. 21 2437 2450, doi: 10.1175/2007JCLI2029.1. Wang, C., S. K. Lee, and D.B. Enfield (2008b), Atlantic Warm Pool acting as a link between Atlantic Multidecadal Oscillation and Atlantic tropical cyclone activity, Geochem. Geophys. Geosyst. 9, Q05V03, doi:10 .1029/2007GC001809. Watanabe, T., A. Winter, and T. Oba (2001), Seasonal changes in sea surface temperature and salinity during the Little Ice Age in the Caribbean Sea deduced from Mg/Ca and 18 O/ 16 O rations in corals, Marine Geology 173, 21 35, doi: 10 .1016/S0025 3227(00)00166 3. Weijers, J.W.H., S. Schouten, O.C. Spaargaren, and J.S. Sinninghe Damst (2006), Occurrence and distribution of tetraether membrane lipids in soils: Implications for the use of the TEX 86 proxy and the BIT index, Org. Geoch em ., 37, 1680 1693, doi:10.1016/j.orggeochem.2006.07.018. Williams, C., B.P. Flower, D.W. Hastings, T.P. Guilderson, and A.C. Hine (2009), Meltwater and Abrupt Climate Change During the Last Deglaciation: A Gulf of Mexico Perspective, Master's thesis, University of South Florida.
104 Winter, A., H. Ishioroshi, T. Watanabe, T Oba, and J. Christy (2000), Caribbean Sea surface temperatures: Two to three degrees cooler than present during the Little Ice Age, Geophys. Res. Lett. 27, 3365 3368, doi: 10.1029/2 000GL011426. Wuchter, C., S. Schouten, M.J.L. Coolen, and J.S. Sinninghe Damst (2004), Temperature dependent variation in the distribution of tetraether membrane lipids of marine Crenarchaeota: Implications for TEX 86 paleothermometry, Paleoceanograph y 19, PA4028, doi:10.1029/2004PA001041. Wuchter, C., S. Schouten, S.G. Wakeham, and J.S. Sinninghe Damst (2005), Temporal and spatial variation in tetraether membrane lipids of marine Crenarchaeota in particulate organic matter: Implications for TEX 8 6 paleothermometry, Paleoceanography 20, PA3013, doi:10.1029/2004PA001110. Wuchter, C., S. Schouten, S.G. Wakeham, and J.S. Sinninghe Damst (2006), Archaeal tetraether membrane lipid fluxes in the northeastern Pacific and the Arabian Sea: Implicatio ns for TEX 86 paleothermometry, Paleoceanography 21, PA4208, doi:10.1029/2006PA001279. &ari', S., B. Donner, G. Fischer, S. Mulitza, and G. Wefer (2005), Sensitivity of planktic foraminifera to sea surface temperature and export production as derived fr om sediment trap data, Marine Micropaleontology 55 75 105.
105 A ppendices
106 Appendix I: Fisk Basin (PE07 5) Data AI.1. Fisk Basin Downcore Mg/Ca Data Core Depth (mm) Calendar Age (Yrs BP) Mg/Ca (mmol /mol) SST (C) 0 0.0 4.75 26.2 5 18.8 4.37 25.3 10 37.6 4.13 24.7 15 56.4 4.16 24.7 20 75.2 4.29 25.1 25 94.0 4.28 25.0 30 112.8 3.72 23.5 35 131.6 3.77 23.6 40 150.4 3.88 24.0 45 169.2 3.61 23.1 50 188.0 3.75 23.6 55 206.8 3.69 23.4 60 225.6 3.60 23.1 65 244.4 3.69 23.4 70 263.2 3.82 23.8 75 282.0 3.90 24.0 80 300.8 3.89 24.0 85 319.6 3.95 24.1 90 338.4 3.88 24.0 95 357.2 3.94 24.1 100 376.0 3.72 23.5 105 394.8 4.08 24.5 110 413.6 3.95 24.2 115 432.4 4.26 25.0 120 451.2 3.69 23.4 125 470.0 4.18 24.8 130 488.8 4.28 25.0 135 507.6 3.92 24.1 140 526.4 4.05 24.4 145 545.2 4.53 25.7 150 564.0 4.20 24.9 155 582.8 4.03 24.4 160 601.6 4.25 25.0 165 620.4 4.04 24.4 170 639.2 4.14 24.7 175 658.0 4.10 24.6 180 676.8 4.24 25.0 185 695.6 4.09 24.5 190 714.4 4.22 24.9
107 195 733.2 4.48 25.6 200 752.0 4.63 25.9 205 not calibrated 4.45 25.5 210 not calibrated 4.24 25.0 215 not calibrated 4.37 25.3 220 not calibrated 4.46 25.5 225 not calibrated 4.50 25.6 230 not calibrated 4.34 25 .2 235 not calibrated 4.42 25.4 240 not calibrated 4.73 26.2 245 not calibrated 4.52 25.7 250 not calibrated 4.28 25.1 255 not calibrated 4.24 24.9 260 not calibrated 4.37 25.3 AI.1. Fisk Basin Downcore Mg/Ca Data. Each Mg/Ca measurement is based on 60 G. ruber (white) from the 250 300 m size fraction. Shaded rows represent the interval of the core for which there are large uncertainties in the age model, and therefore a calibration to calendar age was not made. AI.2. Fisk Basin Radiocarbon Data Sample Depth (mm) AMS 14 C Date error 1 sigma (low) 1 sigma (high) Cal yrs BP STDEV 0 5 215 30 0 0 0 0 150 155 925 35 494 552 523 41 160 165 1090 30 628 676 652 34 200 205 1210 30 703 782 743 56 220 225 1675 35 1201 1276 1239 53 230 235 1610 35 1135 1230 1183 67 AI.2. Fisk Basin Radiocarbon Data. Raw radiocarbon dates were made on mixed assemblages of planktonic Foraminifera, and measured via Accelerator Mass Spectrometer at Lawrence Livermore National Laboratory AMS facility. Raw AMS 14 C dates ar e reported in the second column with instrumental 14 C error in the third column. Radiocarbon ages were converted to calendar years using the CALIB 5.0 program, with an assumed constant 400 year reservoir correction. 1 sigma (high and low) range is reported
108 Appendix II. Garrison Basin (PE07 2) Data AII.1. Garrison Basin Mg/Ca Data Core Depth (mm) Calendar Age (Yrs BP) Mg/Ca (mmol/mol) SST (C) 0 0 4.43 25.4 5 11.8 4.29 25.1 10 24.7 4.42 25.4 15 38.6 4.26 25.0 20 53.6 4.13 24.7 25 69.6 4.06 24.5 30 86.6 4.21 24.9 35 104.5 3.81 23.8 40 123.5 3.96 24.2 45 143.3 4.33 25.2 50 164.2 3.95 24.2 55 185.9 4.00 24.3 60 208.6 3.80 23.7 65 232.1 4.01 24.3 70 256.6 3.73 23.5 75 281.9 3.78 23.7 80 308.0 4.02 24.4 85 335.0 3.91 24.1 90 362 .8 3.90 24.0 95 391.4 4.51 25.6 100 420.8 4.19 24.8 105 451.0 3.79 23.7 110 481.9 4.07 24.5 115 513.6 3.78 23.7 120 546.0 3.76 23.6 125 579.2 3.88 23.9 130 data not used 3.52 22.9 135 data not used 3.27 22.1 140 data not used 3.49 22.8 145 data not used 3.84 23.9 150 data not used 3.36 22.3 155 data not used 3.15 21.6 160 data not used 165 data not used 3.66 23.33 170 data not used 3.40 22.48 175 data not used 4.13 24.64 180 data not used 185 data not used 4.32 25.2 190 data not used 4.11 24.6 195 data not used 3.49 22.8 200 data not used 3.80 23.7 205 data not used 3.51 22.9
109 210 data not used 3.19 21.8 215 data not used 4.08 24.5 220 data not used 3.80 23.7 225 data not used 3.86 23.9 230 data not used 3.84 23.8 235 data not used 3.81 23.8 240 data not used 3.76 23.6 245 data not used 3.62 23.2 250 data not used 3.97 24.2 255 data not used 3.93 24.1 260 data not used 3.90 24.0 265 data not used 4.13 24.7 270 data not used 4.27 25.0 275 data not used 4.02 24.4 280 data not used 3.83 23.8 285 data not used 4.26 25.0 290 data not used 4.05 24.4 295 data not used 4.46 25.5 300 data not used 4.32 25.2 305 data not used 4.51 25.6 310 data not used 4.35 25.2 315 data not used 4.24 25.0 320 data not used 4.50 25.6 325 data not used 4.35 25.2 330 data not used 4.05 24.4 335 data not used 4.09 24.5 340 data not used 4.30 25.1 345 data not used 4.15 24.7 350 data not used 4.04 24.4 355 data not used 4.00 24.3 360 data not used 3.95 24.2 365 data not used 3.98 24.2 370 data not used 4.35 25.2 375 data not used 3.92 24.1 380 data not used 4.06 24.5 385 data not used 3.99 24.3 390 data not used 3.88 24.0 395 data not used 3.74 23.6 400 data not used 4.17 24.8 405 data not used 3.85 23.9 410 data not used 4.11 24.6 415 data not used 3.90 24.0 420 data not used 3.83 23.8 425 data not used 4.11 24.6 430 data not used 4.05 24.4 AII.1. Garrison Basin Mg/Ca Data. Each Mg/Ca measurement is based on 60 G. ruber (white) from the 250 300 m size fraction. Sha ded rows represent data that were not used for climatic interpretation due to indications of diagenetic alteration of the foraminiferal calcite (i.e. low weight per foram and visual evidence of manganese coatings).
110 AII.2. Garrison Basin Radiocarbon Data Core Depth (mm) AMS 14 C age error 1 sigma (high) 1 sigma (low) Cal Yr BP 0 5 140 35 n/a n/a 0 10 15 455 25 0 81 41 70 75 935 35 127 244 186 120 125 1435 45 563 645 604 200 205 2080 20 1221 1274 1248 270 275 2495 35 1610 1722 1666 340 345 3300 35 2 655 2735 2695 390 395 3640 30 3002 3125 3064 AII.2. Garrison Basin Radiocarbon Data. Raw radiocarbon dates were made on mixed assemblages of planktonic Foraminifera, and measured via Accelerator Mass Spectrometer at Lawrence Livermore National Laborator y AMS facility. Raw AMS 14 C dates are reported in the second column with instrumental 14 C error in the third column. Radiocarbon ages were converted to calendar years using the CALIB 5.0 program, with an assumed constant 400 year reservoir correction. 1 si gma (high and low) range is reported
111 Appendix III. Pigmy Basin GDGT Data AIII.1. TEX 86 Data for Pigmy Basin (PBBC 1F) Core Depth (mm) Calendar Yrs BP TEX 86 Index SST (C) STDEV (C) 0 0 0.70 28.6 0.59 5 12 0.69 28.1 0.2 6 10 25 0.70 28.5 0.88 15 37 20 49 0.69 28.1 0.28 25 62 0.68 27.4 0.39 30 74 0.69 28.0 0.36 35 86 0.70 28.3 0.75 40 98 0.69 27.8 0.33 45 111 0.68 27.7 0.43 50 123 0.69 27.7 0.36 55 135 0.67 26.9 0.79 60 148 0.67 26.9 0.28 65 160 0.68 26.8 0.50 70 172 0.69 27.4 0.42 75 185 0.67 26.7 0.47 80 197 0.68 27.6 0.18 85 209 0.68 27.5 0.60 90 221 0.68 27.3 0.32 95 234 0.68 27.4 0.19 100 246 0.68 27.2 0.50 105 258 0.67 26.8 1.11 110 271 0.69 27.6 0.43 115 283 0.68 27.3 0.29 120 295 0.68 27 .2 0.33 125 308 0.68 27.5 0.66 130 320 0.67 27.2 0.18 135 332 0.68 27.4 0.34 140 344 0.68 27.3 0.38 145 357 0.67 26.9 0.41 150 369 0.69 28.0 0.14 155 381 0.68 27.5 0.69 160 394 0.69 27.8 0.44 165 406 0.68 27.5 0.22 170 418 0.69 27.7 0.56 175 431 0.67 27.1 1.09 180 443 0.68 27.5 0.71 185 455 0.69 28.0 1.43 190 467 0.68 27.6 0.58 195 480 0.68 27.4 0.54
112 200 492 0.68 27.7 0.56 205 504 0.69 27.7 0.47 210 517 0.68 27.5 0.45 215 529 0.68 27.7 0.46 220 541 0.68 27.6 0.26 225 554 0.68 27.3 0.32 230 566 0.68 27.2 0.49 235 578 0.68 27.5 0.34 240 590 245 603 0.68 27.4 0.92 250 615 0.67 27.1 0.14 255 627 0.66 26.1 0.21 260 640 0.67 26.8 0.25 265 652 0.67 27.0 0.37 270 664 0.66 26.6 0.79 275 677 0.68 27.6 0.86 280 689 0.67 26.6 0.48 285 701 0.68 27.1 0.34 290 713 0.68 27.3 0.20 295 726 0.67 27.1 0.53 300 738 0.68 27.3 0.87 305 750 0.67 27.0 0.40 310 763 0.66 26.5 0.12 315 775 0.68 27.3 0.30 320 787 0.67 26.7 0.73 325 800 0.68 27.3 1.19 330 812 0.68 27.4 0.25 335 824 0.66 26. 5 0.59 340 836 0.68 27.6 1.17 345 849 0.67 26.7 0.24 350 861 0.68 27.2 0.44 355 873 0.67 27.0 0.22 360 886 0.66 26.1 0.74 365 898 0.66 26.5 0.27 370 910 0.67 26.6 0.46 375 923 0.65 26.0 0.53 380 935 0.66 26.2 0.91 385 947 0.65 25.9 0.15 390 959 0.65 25.7 0.11 395 972 0.67 27.0 0.55 400 984 0.66 26.5 0.38 405 996 0.67 27.0 0.40 410 1009 0.67 26.6 0.40 415 1021 0.68 27.5 0.35 420 1033 0.67 27.1 0.51 425 1046 0.67 27.0 0.68
113 430 1058 0.67 26.6 0.26 435 1070 0.67 26.8 0.15 440 1082 0.66 26.5 0.02 AII.1. TEX 86 Index Data for Pigmy Basin (PBBC 1F). The TetraEther indeX of 86 carbon GDGTs (TEX 86 ) was calculated using the equation ([II]+[III]+[IV'])/ ([I]+[II]+[III]+[IV']) (see Figure 3.2. for compound structures). SST was calculated using T= 1078 + 56.2*TEX 86 (from Kim et al., 2008). STDEV listed in the fifth column represents the standard deviation of the SST (C) of six replicate injections for each sample. AIII.2. BIT Index Data for Pigmy Basin (PBBC 1F) Core Depth (mm) Calendar Yrs BP BIT Index STDEV 0 0 0.144 0.002 5 12 0.149 0.003 10 25 0.181 0.003 15 37 20 49 0.206 0.011 25 62 0.264 0.003 30 74 0.237 0.006 35 86 0.147 0.006 40 98 0.272 0.004 45 111 0.293 0.005 50 123 0.276 0.008 55 135 0.347 0.007 60 148 0.305 0.003 65 160 0.268 0.005 70 172 0.270 0.015 75 185 0.269 0.004 80 197 0.257 0.002 85 209 0.269 0.012 90 221 0.278 0.006 95 234 0.295 0.007 100 246 0.308 0.002 105 258 0.313 0.005 110 271 0.313 0.004 115 283 0.317 0.007 120 295 0.321 0.005 125 308 0 .329 0.009 130 320 0.338 0.003 135 332 0.308 0.002 140 344 0.265 0.006 145 357 0.263 0.004 150 369 0.238 0.013 155 381 0.194 0.003
114 160 394 0.199 0.003 165 406 0.202 0.002 170 418 0.213 0.002 175 431 0.218 0.001 180 443 0.220 0.004 185 455 0.228 0.003 190 467 0.199 0.008 195 480 0.208 0.012 200 492 0.219 0.004 205 504 0.216 0.001 210 517 0.226 0.008 215 529 0.236 0.002 220 541 0.176 0.005 225 554 0.230 0.003 230 566 0.214 0.007 235 578 0.227 0.015 240 590 245 603 0.202 0.004 250 615 0.217 0.008 255 627 0.245 0.009 260 640 0.243 0.002 265 652 0.227 0.007 270 664 0.287 0.001 275 677 0.345 0.008 280 689 0.298 0.010 285 701 0.211 0.007 290 713 0.178 0.003 295 726 0.155 0.002 300 738 0.167 0.003 305 750 0.215 0.004 310 763 0.251 0.005 315 775 0.237 0.000 320 787 0.229 0.020 325 800 0.212 0.003 330 812 0.230 0.008 335 824 0.255 0.011 340 836 0.255 0.006 345 849 0.338 0.016 350 861 0.368 0.029 355 873 0.266 0.007 360 886 0.256 0.008 365 898 0.269 0.011 370 910 0.26 2 0.008 375 923 0.306 0.008 380 935 0.324 0.010 385 947 0.288 0.011
115 390 959 0.271 0.006 395 972 0.239 0.001 400 984 0.160 0.007 405 996 0.146 0.004 410 1009 0.171 0.006 415 1021 0.196 0.007 420 1033 0.212 0.008 425 1046 0.212 0.006 430 1058 0.2 15 0.007 435 1070 0.192 0.006 440 1082 0.201 0.005 AII.2. BIT Index Data for Pigmy Basin (PBBC 1F). The branched to isoprenoid tetraether (BIT) index was calculated using the equation ([V]+[VI]+[VII])/ ([V]+[VI]+[VII]+[IV]) (see Figure 3.2. for compoun d structures). STDEV listed in the forth column represents the standard deviation of six replicate injections for each sample.
116 Appendix IV. Elemental and Isotopic Size Fraction Data AIV.1 Size Fraction Isotope Data for G. ruber (white) Size Fraction ( m) Sediment Core top 13 C (permille) 18 O (permille) 150 212 PBBC 1 0.32 1.33 150 212 PBBC 1 0.39 1.35 150 212 PBBC 1 0.26 1.24 150 212 PE07 2 0.21 0.93 150 212 PE07 2 0.12 0.99 212 250 PBBC 1 0.60 0.92 212 250 P BBC 1 0.60 1.09 212 250 PBBC 1 0.59 1.25 212 250 PBBC 1 0.51 1.39 212 250 PE07 2 0.47 1.26 212 250 PE07 2 0.55 1.46 212 250 PE07 2 0.61 1.15 250 300 PBBC 1 1.03 1.30 250 300 PBBC 1 0.84 1.53 250 300 PBBC 1 0.27 1.42 250 300 PBBC 1 0.90 1 .62 250 300 PBBC 1 1.02 1.05 250 300 PE07 2 1.11 1.41 250 300 PE07 2 1.02 1.55 250 300 PE07 2 0.72 1.41 300 355 PBBC 1 1.12 1.29 300 355 PBBC 1 1.30 1.38 300 355 PBBC 1 1.19 1.31 300 355 PBBC 1 0.98 1.54 300 355 PBBC 1 1.31 1.64 300 355 PE07 2 1.39 1.27 300 355 PE07 2 1.39 1.29 355 425 PBBC 1 1.52 1.54 355 425 PBBC 1 1.33 1.70 355 425 PE07 2 1.17 1.87 Appendix III.1. Size Fraction Isotope Data for G. ruber (white). Raw carbon and oxygen isotopic data for G. ruber (white) rep resent measurements on separate aliquots of 60 individuals from the respective size fractions. Foraminifera were picked from two different box core core top samples: Pigmy Basin (PBBC 1) and Garrison Basin (PE07 2).
117 AIV.2. Size Fraction Mg/Ca Data for G. ruber (white) Size Fraction ( m) Sediment Core top Mg/Ca (mmol/mol) 150 212 PBBC 1 4.30 150 212 PBBC 1 4.39 150 212 PE07 2 4.35 150 212 PE07 2 4.50 150 212 PE07 2 4.23 212 250 PBBC 1 4.64 212 250 PBBC 1 4.36 212 250 PE07 2 4.41 212 250 PE0 7 2 4.38 212 250 PE07 2 3.86 212 250 PE07 2 4.10 250 300 PBBC 1 4.40 250 300 PBBC 1 4.32 250 300 PBBC 1 4.56 250 300 PBBC 1 4.31 250 300 PE07 2 4.39 250 300 PE07 2 4.17 250 300 PE07 2 4.02 250 300 PE07 2 4.40 300 355 PBBC 1 4.22 300 355 PBBC 1 4.66 300 355 PBBC 1 4.28 300 355 PE07 2 4.05 300 355 PE07 2 4.30 355 425 PBBC 1 4.59 355 425 PE07 2 4.86 Appendix III.2. Size Fraction Mg/Ca Data for G. ruber (white). Raw Mg/Ca data for G. ruber (white) represent measurements on separate aliquot s of 60 individuals from the respective size fractions. Foraminifera were picked from two different box core core top samples: Pigmy Basin (PBBC 1) and Garrison Basin (PE07 2).
118 AIV.3. Size Fraction Isotope Data for G. ruber (pink) Size Fract ion ( m) Sediment Core top 13 C (permille) 18 O (permille) 150 212 PBBC 1 0.68 1.35 150 212 PBBC 1 0.52 1.46 150 212 PE07 2 0.56 1.39 150 212 PE07 2 0.33 1.32 212 250 PBBC 1 1.29 1.46 212 250 PBBC 1 1.10 1.62 212 250 PBBC 1 0.92 1.76 212 250 PE07 2 0.97 1.65 212 250 PE07 2 0.89 1.51 212 250 PE07 2 1.00 1.44 212 250 PE07 2 0.93 1.64 250 300 PBBC 1 1.24 1.58 250 300 PBBC 1 1.38 1.57 250 300 PBBC 1 1.26 1.57 250 300 PBBC 1 1.26 1.64 250 300 PE07 2 1.21 1.62 250 300 PE07 2 1.22 1.40 250 300 PE07 2 1.05 1.23 250 300 PE07 2 1.43 1.46 250 300 PE07 2 1.29 1.65 300 355 PBBC 1 1.79 1.67 300 355 PBBC 1 1.43 1.60 300 355 PBBC 1 1.65 1.69 300 355 PE07 2 1.92 1.59 300 355 PE07 2 1.81 1.67 300 355 PE07 2 1.67 1.82 300 35 5 PE07 2 1.61 1.57 355 425 PBBC 1 1.79 1.94 355 425 PBBC 1 1.60 1.96 355 425 PBBC 1 2.23 1.90 355 425 PBBC 1 1.92 2.02 355 425 PE07 2 1.98 1.98 355 425 PE07 2 1.66 1.81 355 425 PE07 2 1.92 1.97 355 425 PE07 2 1.85 2.10 425 500 PBBC 1 1.87 1.98 425 500 PBBC 1 1.79 1.98 425 500 PBBC 1 1.81 2.22 425 500 PE07 2 1.95 2.00 425 500 PE07 2 2.36 2.19 425 500 PE07 2 2.16 2.18 Appendix III.3. Size Fraction Mg/Ca Data for G. ruber (pink). Raw carbon and oxygen isotopic data for G. rube r (pink) represent measurements on separate aliquots of 60 individuals from the respective size fractions. Foraminifera were picked from two different box core core top samples: Pigmy Basin (PBBC 1) and Garrison Basin (PE07 2).
119 AIV.4. Size Fraction Data for Mg/Ca in G. ruber (pink) Size Fraction ( m) Sediment Core top Mg/Ca (mmol/mol) 150 212 PBBC 1 3.82 150 212 PBBC 1 3.77 150 212 PE07 2 4.01 150 212 PE07 2 4.54 150 212 PE07 2 3.85 212 250 PBBC 1 4.01 212 250 PBBC 1 4.46 212 250 PE07 2 4.22 21 2 250 PE07 2 4.00 212 250 PE07 2 4.24 250 300 PBBC 1 4.47 250 300 PBBC 1 4.55 250 300 PBBC 1 4.48 250 300 PBBC 1 3.96 250 300 PBBC 1 4.05 250 300 PE07 2 4.05 250 300 PE07 2 4.55 250 300 PE07 2 4.46 250 300 PE07 2 4.45 250 300 PE07 2 4.43 300 35 5 PBBC 1 4.11 300 355 PBBC 1 4.33 300 355 PBBC 1 4.55 300 355 PE07 2 4.55 300 355 PE07 2 4.41 300 355 PE07 2 4.49 300 355 PE07 2 4.11 355 425 PBBC 1 4.56 355 425 PBBC 1 4.50 355 425 PBBC 1 4.34 355 425 PE07 2 4.44 355 425 PE07 2 4.65 355 425 PE 07 2 5.19 355 425 PE07 2 4.31 425 500 PBBC 1 5.00 425 500 PBBC 1 4.48 425 500 PBBC 1 4.88 425 500 PE07 2 4.77 425 500 PE07 2 4.61 425 500 PE07 2 5.23 >500 PE07 2 4.92 Appendix III.2. Size Fraction Mg/Ca Data for G. ruber (pink). Raw Mg/Ca data for G. ruber (pink) represent measurements on separate aliquots of 60 individuals from the respective size fractions. Foraminifera were picked from two different box core core top samples: Pigmy Basin (PBBC 1) and Garrison Basin (PE07 2).
120 AIV.5. Summary of Size Fraction Data AIII.5.a. Summary of Size Fraction Isotopic Data for G. ruber (pink) Size Fraction ( m) n 18 O (permille) STDEV 13 C (permille) STDEV 150 212 4 1.38 0.06 0.52 0.14 212 250 7 1.58 0.12 1.01 0.14 250 300 9 1.52 0.14 1.26 0.11 300 355 7 1.66 0.09 1.70 0.16 355 425 8 1.96 0.08 1.87 0.20 425 500 6 2.09 0.11 1.99 0.22 AIII.5.b. Summary of Size Fraction Isotopic Data for G. ruber (white) Size Fraction ( m) n 18 O (permille) STDEV 13 C (permille) STDEV 150 212 5 1.17 0.20 0. 26 0.10 212 250 7 1.22 0.18 0.56 0.05 250 300 8 1.41 0.18 0.86 0.27 300 355 7 1.39 0.15 1.24 0.15 355 425 3 1.70 0.16 1.34 0.17 AIII.5.c. Summary of Size Fraction Mg/Ca Data for G. ruber (pink) Size Fraction ( m) n Mg/Ca (mmol/mol) STDEV SST (C) 150 212 5 4.00 0.32 26.1 212 250 5 4.19 0.19 26.7 250 300 10 4.35 0.23 27.1 300 355 7 4.37 0.19 27.1 355 425 7 4.57 0.30 27.6 425 500 6 4.83 0.27 28.2 >500 1 4.92 n/a 28.5 AIII.5.d. Summary of Size Fraction Isotopic Data for G. ruber (white) Size Fraction ( m) n Mg/Ca (mmol/mol) STDEV SST (C) 150 212 5 4.36 0.10 25.2 212 250 6 4.29 0.27 25.1 250 300 8 4.32 0.16 25.2 300 355 5 4.30 0.22 25.1 355 425 2 4.73 0.19 26.2 AIII.5a d. Summary of Size Fraction Data. These tables summarize the data in AIII.1 AIII.4, by presenting the mean values for all replicate measurements within each size fraction for G. ruber (pink) and G. ruber (white), respectively. n represents the number of measurements made in each size fraction, and the STDEV is the stand ard deviation of those n measurements.
About the Author Julie N. Richey was born in Cincinnati, Ohio in 1981. She earned a B.S. Degree in Biological Sciences and a B.A. Degree in Geology from The Ohio State University in Columbus, Ohio in 2004. She was awarded a Presidential Doctoral Fellowship at the University of South Florida, where she began as a doctoral student in the College of Marine Science in 2004. She received her M.S. Degree from USF in 2007, and will complete her Ph.D in Marine Science in August, 2010. During her studie s at USF, she has published 4 peer reviewed papers, and presented her research findings at more than 10 international research conferences. She has participated in 3 research cruises to the Gulf of Mexico, Caribbean Sea and Tropical North Atlantic Ocean. She has been awarded a NOAA UCAR Global Climate Change Postdoctoral Research Fellowship, which she will carry out at the University of Washington's School of Oceanography in Seattle, WA, beginning in August 2010.