PP11A-0539 0800h
The `Antarctic signal' in Northeast Atlantic benthic $\delta^{18}$O during MIS 3 attributed primarily to deep-water temperature variations
A key, and unresolved though much theorised aspect of millennial climate variability during the last glacial period is the consistent asynchrony of planktonic and benthic foraminiferal $\delta^{18}$O recorded in the Northeast Atlantic (Shackleton et al., Paleoceanography 15, 565, 2000), and in particular the apparent coupling of such variations with an identical asynchrony between Greenland and Antarctic temperatures (Blunier et al., Nature 394, 739, 1998). These relationships result in an apparent coupling of Antarctic temperature and benthic $\delta^{18}$O variations throughout Marine Isotope Stage (MIS) 3. It was originally proposed that a climatic parameter with a long time-constant, such as global ice-volume, must account for the millennial benthic $\delta^{18}$O signal. Based on reconstructions of Red Sea salinity and coral terrace facies variations (Chappell, QSR 21, 1229, 2002; Siddall et al., Nature 423, 853, 2003) it has since been suggested that benthic $\delta^{18}$O has indeed responded to rapid glacioeustatic sea level fluctuations. Importantly, the sea level variations thus reconstructed could only account for approximately half of the observed amplitude of benthic $\delta^{18}$O; or if invoked as a complete explanation for the benthic $\delta^{18}$O signal would require a sensitivity of Atlantic deep-water $\delta^{18}$O to ice-volume change that was two times larger than on glacial - interglacial time-scales. Here, we present preliminary measurements of Mg/Ca ratios in benthic foraminifera from core MD01-2444 (recovered from 2,656m on the Iberian Margin) that indicate that most of the `Antarctic' benthic $\delta^{18}$O signal may be attributed to local deep-water temperature variations, and hence that sea-level fluctuations co-ordinated with Antarctic climate cannot account entirely for the benthic signal. Furthermore, it appears that sea-level change may not be in phase with the benthic $\delta^{18}$O, and that the timing of sea-level high-stands should not be inferred from the benthic $\delta^{18}$O record. The deep-water temperature variations reconstructed during MIS 3 are suggested to arise from local changes in deep-water circulation, including variations in deep-water sourcing and/or mode of formation in a manner similar to that proposed for the late glacial and across Termination I (Skinner et al., G$^{3}$ v.4 (12), 2003). The proven validity of these results would place important and much needed constraints on the mechanisms that may be invoked to explain the so-called `bi-polar see-saw' of Greenland and Antarctic temperatures, and their apparent links to other global phenomena, including sea level and atmospheric CO$_{2}$ fluctuations.
PP11A-0540 0800h
Refining Of The MIS 5 Tephrostratigraphic Record In Marine Cores From The S-Norwegian Margin; A Tool For High-Resolution Dating And Correlation Of Climatic Archives
Tephrochronology has proved to be an excellent tool for correlation and dating of sediment sequences of various ages and in different archives. It is, however, well known that the limitation of applying tephra from the whole range of basaltic eruptions, taking place in Iceland during is partly due to the distribution and thickness of some of these tephra layers but importantly also due to the geochemical identification of these tephra layers. Knowing that during at least the last 1100 years has the frequency of volcanic eruption in Iceland been of an order of one eruption every 5th year, the potential for utilising the entire tephrostratigraphic record for dating and correlation is great. Still only few tephra layers have been applied for dating and correlation from this period as ca. 90 % of these tephra layers are originated from basaltic volcanoes where the geochemical identification has been a limited factor due to internal homogeneity within the basaltic volcanic systems. A new analytical method, which uses a broader spectrum of geochemical elements as the tracer and the rare earth elements REE, will offer new possibilities to identify each of these basaltic layers on a single grain basis. This method will offer a great possibility to extend the application of the tephra records significantly as a dating and correlation tool especially where these tephra events can be combined with spectra obtained through e.g. stable isotope analyses or through non-destructive core detection. Here, Marine Isotope Stage 5, MIS 5, tephra layers in two giant piston cores, i.e. MD99-2283 and MD99-2289 from the S-Norwegian Margin, have been identified and analysed geochemically for major, trace and REE. The EMP (electron microprobe) technique has been applied on the major elements and Laser Ablation ICP-MS (213 nm) on the trace and the REE. At least 14 tephra layers, both previously described and identified, have been found and will be presented. These findings are used to correlate the two cores and other sediment/ice core sequences from the same time span, and to resolve the timing of important climatic changes in MIS 5. The application of Laser Ablation ICP-MS on geological fingerprinting problems will also be discussed.
PP11A-0541 0800h
Middle Eocene to early Oligocene paleoceanography of the Southern Ocean from foraminiferal stable isotope and Mg/Ca records
The middle Eocene to early Oligocene is a critical phase of Cenozoic climatic evolution, characterized by long-term cooling of deep waters and the first appearance of large Antarctic ice sheets. The Southern Ocean is a key area for monitoring Eocene-Oligocene climatic variability, and development of high-resolution paleoceanographic records from this region is crucial to evaluating possible mechanisms of long-term climate change. These records will ultimately help constrain the timing and nature of middle-to-late Eocene cooling and ice buildup on Antarctica in relation to the timing of major paleogeographic changes, especially the opening of Southern Ocean gateways, and independent records of {\it p}CO2 variability. We have generated high-resolution stable isotope records from ODP Sites 689, 702, 738, and 748 in the Atlantic and Indian sectors of the Southern Ocean. These sites are separated by considerable distances, but the records display coherent trends with several unique features. These features enable detailed site-to-site correlation and indicate that a common paleoceanographic history is shared by all study sites. The late middle Eocene records are punctuated by a strong, transient warming event at $\sim$41 Ma. This event is followed by an increase in $\delta$$^{18}$O values that culminates at $\sim$37 Ma, indicative of long-term cooling and/or growth of small ice sheets. This is followed by an inferred brief warming in the early late Eocene ($\sim$36.5-36.0 Ma), and a slight cooling step at ($\sim$35.5 Ma). A period of relative stability in the latest Eocene precedes the well-documented $\sim$1.4$\permil$ increase in $\delta$$^{18}$O (the Oi-1 event) in the earliest Oligocene. In addition to the stable isotope records, we are developing high-resolution planktonic foraminiferal Mg/Ca records. The primary goal of this work is to separate the ice-volume and temperature contributions to the Eocene-Oligocene $\delta$$^{18}$O signal. Our initial focus is on the early Oligocene Oi-1 event, which is thought to represent the first major Cenozoic buildup of large ice sheets in East Antarctica. At Site 748, Mg/Ca data from the thermocline-dweller Subbotina angiporoides indicate a significant decrease in temperature in the earliest Oligocene. These data are interpreted to represent a 2-3$\deg$C cooling at the Oi-1 transition. Removal of this temperature component from the oxygen isotope signal retains a significant ice volume signature. A minimum of 60% modern ice volume is estimated during peak Oi-1 glaciation.
PP11A-0542 0800h
Orbitally paced climate change across the middle Miocene climate transition
Spectral analyses of an orbitally-tuned 3-my-long geochemical record of Miocene (16.5-13.5 Ma) climate and carbon cycling from the South Tasman Rise, Southern Ocean (paleolatitude: $\sim$$55\deg$S) provide insight into the processes and feedbacks involved in the middle Miocene climate transition (MMCT; 14.2-13.8 Ma), one of the three major steps in Earth's Cenozoic climate evolution. Spectral power in benthic foraminifer ({\it Cibicidoides mundulus}) $\delta$$^{18}$O and $\delta$$^{13}$C is concentrated in all primary Milankovitch frequency bands, with significant power in the 406- and $\sim$100-ky eccentricity bands. Cross-spectral analyses indicate that {\it C. mundulus} $\delta$$^{18}$O and $\delta$$^{13}$C co-vary and are coherent with orbital cyclicity. Variations in Earth's orbit preceded both Antarctic ice growth and global carbon cycling by $\sim$60-ky during the MMCT. At 14.5 Ma, a shift in power from the $\sim$100- to the 406-ky band occurred in both the $\delta$$^{18}$O and $\delta$$^{13}$C records. Band-pass filtering of these records reveals an increase in the amplitude of the 406-ky eccentricity cycle at 14.2 Ma and a decrease in the amplitude of the 100-ky eccentricity cycle at 14.5 Ma, even though the calculated eccentricity signal was relatively stable throughout the interval. Changes in eccentricity power and amplitude occur before the global $\sim$1$\permil$ $\delta$$^{18}$O increase at 13.9 Ma, suggesting that strong climate feedbacks were involved in the MMCT. Sensitivity to long-period eccentricity forcing increased at 14.2 Ma, immediately following peak warmth of the Miocene climatic optimum ($\sim$17-14 Ma). The shift in sensitivity may have resulted from changes in global carbon cycling associated with reorganization of the climate system, specifically a tectonically mediated reduction in meridional heat/vapor transport related to the constriction of the eastern Tethys Seaway. Inferred low atmospheric {\it p}CO$_{2}$ levels may have further enhanced the global response to this reorganization.
PP11A-0543 0800h
Isotopic Constraints on the Neogene Flow of Northern Component Water
Benthic foraminiferal isotope data have been assembled to examine Neogene $\delta^{13}$C gradients between the major ocean basins. These data are reported on the astronomical time scale of Lourens et al. (2004) and correlated to high resolution benthic $\delta^{18}$O records, resulting in a higher precision of correlation between cores. A local linear kernel regression procedure (i.e. a type of nonparametric regression) is used to produce a regional estimate of $\delta^{13}$C in each of the North Atlantic, Southern and Pacific Oceans. Confidence bands for the regression estimates are found through both analytic and bootstrap techniques and a simulation extrapolation procedure is applied to understand the effect of errors in age determinations. Estimated uncertainties are used to establish the value of interbasinal $\delta^{13}$C gradients as a tool for Neogene paleoceanographic reconstructions. Uncertainties in carbon isotope measurements are considerably larger than instrument error: it is mostly a result of sample variance, comprising variable vital effect corrections, diagenesis, etc. During intervals of reduced $\delta^{13}$C gradients, the uncertainties are such that the use of carbon isotope techniques in understanding paleoceanographic circulation patterns is severely limited. Confidence bands overlap for much of the Neogene period, and $\delta^{13}$C values diverge only in the last 8 Ma. Interpretation of $\delta^{13}$C gradients before 8 Ma is particularly troublesome. Possible causes for changes in the influence of northern-derived Northern Component Water (NCW) in the Southern Ocean are examined. We assess the importance of mantle-plume control of the northern gateway around Iceland. Temperature fluctuations of mantle-plume material beneath the region cause changes in the bathymetry of the Greenland-Scotland Ridge, and are believed to be a contributing factor to fluctuations in %NCW throughout the Neogene period. Lourens et al. 2004. The Neogene Period. In F.M. Gradstein, J.G.Ogg, A.G. Smith et al. - A Geologic Time Scale 2004. Cambridge University Press, UK.
PP11A-0544 0800h
Oceanographic Influences in the Variability Of $\delta$$^{13}$C and $\delta$$^{18}$O in Benthic and Planktonic Foraminiferal From the Gulf of California
The analysis of multicore samples collected during CALMEX-NH01 cruise (2001) in two locations of the Gulf of California provide information regarding changes in climate and oceanography of the region. Alfonso Basin on the western side and Pescadero Basin on the center slope of the east side of the Gulf, are marginal basins with sills or shoreward slopes in the Oxygen Minimum Zone. Each basin represents different oceanographic conditions in the Gulf, which can be recognized in the sedimentary history. In Pescadero Basin, accumulation rates of 0.5 to 1.1mm/yr were calculated by $^{210}$Pb dating. For Alfonso basin accumulation rates between 0.24 and 0.32mm/yr were obtained by AMS radiocarbon analysis in foraminifera. The multicores were sampled at 2.2mm intervals (~3-5yr/sample in Pescadero, 6-8yr/sample in Alfonso) to produce a high-resolution record of $\delta$$^{18}$O and $\delta$$^{13}$C based on benthic ({\it Bolivina subadvena}) and planktonic ({\it Globigerina bullloides}) foraminifera. For the past 100 years, a visual comparison between {\it G. bulloides} $\delta$$^{18}$O calculated sea surface temperature (SST) time series from Pescadero Basin and a 5-years moving average of the annual Pacific Decadal Oscillation (PDO), revealed a good agreement between both variables. The same comparison but using {\it G. Bulloides} $\delta$$^{18}$O calculated SST from Alfonso Basin, shown no similarities between SST and PDO time series. Such an important contrast between both records evidences a major influence of the Pacific Ocean on the east side of the Gulf due to the entrance of water from the California Current (CC), and a more direct influence of meteorological processes in the west side mainly due to the presence of Gulf of California Water (GCW) from the Northern Gulf. Isotopic ratios of benthic foraminifera for both basins also evidence oceanographic differences. For the last 40 years, the calculated average {\it B. subadvena} $\delta$$^{13}$C in Alfonso Basin is -0.832%o when Pescadero exhibits an average $\delta$$^{13}$C {\it B. subadvena} for the same period of -1.608%o, indicating differences in surface productivity rates for both sides and also a different quality of the organic matter reaching the bottom. However, downcore $\delta$$^{18}$O variations show that changes in SST associated with ENSO events leave a measurable signal in the composition of planktonic foraminifera in both basins, indicating that both basins are sensitive to major climatic changes. For the last 50yr, {\it G. bulloides} $\delta$$^{18}$O records from Alfonso and Pescadero Basins show negative excursions that can be associated with the 1972-73, 1982-83, 1986-88, and 1997-98 ENSO events.
PP11A-0545 0800h
Changes in Deep Sea Temperature and Ice Volume Based on Paired Measurements of Benthic Foraminiferal Mg/Ca and $\delta$$^{18}$O: Evidence from ODP Site 849, Equatorial Pacific for Marine Isotope Stages 1-3
Most continuous records of sea level changes are based on benthic foraminiferal $\delta$$^{18}$O, which is a measure of both the $\delta$$^{18}$O of seawater and temperature. New calibrations [Martin et al., 2002; Lear et al., 2002] between temperature and Mg/Ca ratios in benthic foraminiferal calcite provide a means to differentiate the separate effects of temperature and global ice volume on the benthic foraminiferal $\delta$$^{18}$O signal. We have measured the Mg/Ca ratio of the benthic foraminifera species, {\it Cibicidoides wuellerstorfi}, throughout Marine Isotope Stages (MIS) 1-3 (0-60 Ka) in sediments from Eastern Equatorial Pacific, ODP Site 849 (0$\deg$11'N, 110$\deg$ 31'W; 3851 m). Using published Mg/Ca -temperature relationships for this species, we estimate a 2.5$\deg$C increase in bottom water temperature at this location over Termination 1. The change in benthic foraminiferal $\delta$$^{18}$O associated with this deglaciation is ~1.6$\permil$ [Mix et al., 1995]. Using a conversion factor of 0.23 $\permil$ per 1$\deg$C [Kim and O'Neil, 1997], the $\delta$$^{18}$O change accounted for by a 2.5$\deg$C change in temperature is approximately 0.58$\permil$, suggesting a residual 1.02$\permil$ change to be accounted for by the change in global ice volume over Termination 1. This is consistent with previous estimates of the global $\delta$$^{18}$O change of seawater due to melting of continental ice sheets since the Last Glacial Maximum.
PP11A-0546 0800h
Milankovitch Forcing at the Eocene/Oligocene Climate Transition from Benthic Foraminiferal Isotope Analysis
A critical component missing from our knowledge of Antarctic cryosphere evolution is a comprehensive understanding of the role of orbital forcing in ice sheet development during the Eocene/Oligocene climate transition. Existing marine records lack the resolution and preservation to reconstruct variability in the high-frequency precessional band, thought to be potentially significant in early Oligocene ice sheet development and stability. A new E/O boundary section from Ocean Drilling Program Site 1263, on the northwest flank of Walvis Ridge in the eastern South Atlantic Ocean, yields a benthic foraminiferal isotopic record with an average time step of 3.5 kyrs, sufficient to resolve the 19- and 23-ky precessional Milankovitch bands. The E/O boundary at this site is associated with a gradual increase in sediment color lightness (L$\ast$),attributed to an increase in carbonate content and accumulation rate and a general improvement in microfossil preservation. The 1263 L$\ast$ increase is accompanied by an enrichment of $\sim$1.4$\permil$ in $\delta$$^{18}$O and $\sim$0.8$\permil$ in $\delta$$^{13}$C of bulk carbonate across the 5-meter interval that represents the period between 33.7 and 33.4 Ma. Multiple single-species isotope analyses of {\it Oridorsalis umbonatus}, {\it Nuttallides trumpeyi}, and {\it Cibicidoides spp.} produce consistent isotope values across a series of horizons, however deviation from these patterns observed in some larger specimens may be indicative of reworking. Using smaller specimens, we will present a high-resolution oxygen isotope record across the E/O transition interval in order to evaluate the significance of precessional forcing in early Oligocene glaciation.
PP11A-0547 0800h
New Insights into Inferring Climate Variability from Records of Planktonic Foraminiferal Mg/Ca, Oxygen Isotope and Shell Weight in the Southern Ocean
Temperature is the dominant control on foraminiferal Mg/Ca, although other factors are important and can bias paleoceanographic reconstructions. In order to investigate the controls on Mg/Ca, we compare records from two sites at different water depths located north and south of the Subtropical Convergence in the Southern Ocean (ODP Site 1123 and MD97-2120). Mg/Ca, d18O, and shell weight were obtained from two narrow size fractions (250-300 and 300-355 um) of the planktonic foraminifer Globigerina bulloides. Combining these data allows us to investigate the effects of hydrography, dissolution and foraminiferal habitat as controls on Mg/Ca. Differences are observed between the records from these locations. Comparison of Mg/Ca records from both size fractions at MD97-2120 implies similar calcification temperatures. In contrast, Mg/Ca-derived paleotemperatures from the 250-300 um fraction at ODP Site 1123 are higher than in the larger size fraction. Mg/Ca records from both size fractions at MD97-2120 are similar across Termination 1 and imply a glacial-interglacial temperature change of 8 degrees C. The temperature record from the 250-300 um fraction from ODP Site 1123 shows a glacial-interglacial temperature change of 4 degrees across Termination 1, whereas in the 300-355 um fraction no temperature change is observed. At ODP Site 1123, the contrast between size fractions in Mg/Ca-derived temperatures is even more striking across Termination 2. The differences in Mg/Ca records are associated with offsets in d18O and shell weight records from both size fractions. The larger fraction exhibits heavier d18O and larger weight variability. Possible explanations for these observations include hydrographic differences between the two sites, foraminiferal habitat, dissolution susceptibility, and preservation history.
PP11A-0548 0800h
Oligocene Climate Forcing and Palaeoceanography of the Equatorial Pacific
A planktonic and benthic foraminiferal stable isotope stratigraphy of the Oligocene equatorial Pacific (Ocean Drilling Program, Site 1218) was generated at 6 kyr resolution between magnetochrons C9n and C11n.2n ($\sim$26.4-30 Ma on a newly developed astronomically calibrated time scale). Our data allow a detailed examination of Oligocene paleoceanography, the evolution of the early cryosphere and the influence of orbital forcing on glacioeustatic sea level variations. The Oligocene climate and ice sheet dynamics were strongly influenced by orbital forcing. Spectral analysis reveals power and coherency for obliquity (40 kyr period) and eccentricity ($\sim$110 kyr, 405 kyr) orbital bands, with an additional strong imprint of the eccentricity and 1.2 Myr obliquity amplitude cycle, which drove ice sheet oscillations in the southern hemisphere. Heavy oxygen isotope intervals correspond to maxima in carbon isotopes, indicating substantial changes in the carbon cycle that accompanied the glacial events. Planktonic and benthic foraminifera $\delta^{18}$O are used to constrain the magnitude and timing of major fluctuations in ice volume and global sea level change. Glacial episodes, related to obliquity and eccentricity variations, occurred at 29.16, 27.91 and 26.76 Ma. These correspond to glacioeustatic sea level fluctuations of 50 to 65 meters. High amplitude cyclic variations are recorded in the carbon isotope signal of planktonic and benthic foraminifera, the water column carbon isotope gradient and estimated percent carbonate of bulk sediment. Maxima in $\delta^{13}$C and the increased $\Delta\delta^{13}$C values are associated with each of the glacial events. Alteration of high latitude temperatures and Antarctic ice volume thus had a significant impact on the global carbon burial and equatorial productivity. We investigate the implications of a close correspondence between oxygen and carbon isotope events and long term amplitude envelope extrema in astronomical calculations during the Oligocene, and develop a new naming scheme for stable isotope events, based on the 405 kyr eccentricity cycle count.
PP11A-0549 0800h
Ocean Circulation and Gateway Closures During the Late Miocene (~13-5 Ma)
Long-term climate change is driven by tectonic influences, including changes in ocean circulation that are the result of ocean gateway closure. During the middle to late Miocene (~13-5 Ma), both tropical ocean circulation and deep water production were reorganized due to the increasing constriction of the Indonesian and Central American seaways. For example, the waters of the modern Pacific equatorial current system do not move freely into the Indian Ocean (i.e., via the Indonesian Throughflow, ITF) but instead pile up to form the Western Pacific Warm Pool (a thermal anomaly that greatly influences tropical Pacific climate and ocean circulation). Here we use a continuous record of multispecies stable isotope stratigraphy and foraminiferal assemblage counts from Ontong Java Plateau to demonstrate that during middle to late Miocene time, progressive restriction of the ITF, modulated by sea level fluctuations, resulted in the waxing and waning of a proto-warm pool in the western equatorial Pacific (WEP). The proto-warm pool profoundly affected the early late Miocene "carbonate crash" (an anomalous decrease of carbonate in deep sea sediments) and the late Miocene "biogenic bloom" (sharp increase in carbonate accumulation rates across the tropical Indo-Pacific). We hypothesize that El Niño/La Niña-like alternations of tropical carbonate preservation and productivity between the western and eastern equatorial Pacific during the late Miocene were the consequence of early warm pool development and decay. A proto-warm pool was formed ~12.1-10.6 Ma, which initiated a nutrient-rich Equatorial Undercurrent and/or increased Trade Wind strength. These La Niña-like conditions sustained carbonate productivity in the eastern equatorial Pacific (EEP) at a time when carbonate preservation sharply declined in the Caribbean. Proto-warm pool weakening and El Niño-like conditions ~10.6-8.8 Ma intensified a "carbonate crash" in the EEP, while resurgence of the warm pool and La Niña-like conditions after ~6.5 Ma spurred a "biogenic bloom". The production of deep water in the northern North Atlantic (i.e., Northern Component Water, NCW) may also have been modulated by sea level fluctuations as the Central American Seaway became increasingly constricted by the uplift of the Panama sill during the late Miocene. We suggest that the sea level fluctuations that facilitated the early development of a proto-warm pool in the WEP, particularly the Mi5 event at 11.4 Ma, also constricted flow through the Central American Seaway and controlled NCW production at this time in the North Atlantic.
PP11A-0550 0800h
Orbitally Paced Climate Variability During the Middle Miocene: High Resolution Stable Isotopes and Fe Records From the Western and Southeastern Pacific
We present continuous high resolution middle Miocene benthic foraminiferal isotope ($\sim$ 4-ky) and XRF scanner Fe ($ < $ 1-ky) records from two deep-sea cores in the western and southeastern Pacific (ODP Sites 1146 and 1237). These records provide new insights into the pacing of climatic variability across the middle Miocene climate transition. Spectral analyses of the $\delta$$^{13}$C records from Sites 1146 and 1237 and the Fe record from Site 1237 reveal spectral power concentrated in the eccentricity band (400-ky, 100-ky) over the time interval from 13 to 17 Ma. In the $\delta$$^{18}$O records, a switch occurs at $\sim$ 14.9 Ma from predominant 100-ky eccentricity to 41-ky obliquity periodicity, suggesting a shift in the ocean-climate response to orbital forcing from the low to the high latitudes. The $\delta$$^{18}$O obliquity signal, mainly attributed to the waxing and waning of an expanded southern polar ice sheet, is pervasive in the $\delta$$^{18}$O records until $\sim$ 13.9 Ma, when a sharp increase in $\delta$$^{18}$O values indicates a further climatic transition (major expansion of East Antarctic Ice Sheet). This critical step into the "icehouse-world" coincided with a further switch in the $\delta$$^{18}$O records from the obliquity (41-ky) to the eccentricity (100-ky) band.
PP11A-0551 0800h
The Carbon and Sulfur Cycles through the Cenozoic: Insight from Oxygen Isotopes in Marine Sulfate
Marine sulfate plays an important role in the cycling of biochemicals in organic rich sediments, serving as the terminal electron acceptor in the remineralization of organic matter and responsible for nearly all anaerobic methane oxidation. Because sulfur isotopes are largely conserved during sulfur cycling in organic rich sediments, they reflect mostly changes in net sulfur burial, and have been used to study fluctuations in sulfur mineral burial over Earth history. Recently, we have shown that temporal variability in oxygen isotopes measured in marine sulfate (d18O-SO4) highlight changes the pathways of sulfur cycling on continental margins because the d18O-SO4 is reset during sulfate reduction and sulfide reoxidation. The fluxes associated with sulfur cycling, predominantly in shallow sediments, are nearly three times larger than riverine input. We present a continuous record of d18O-SO4 in marine barite over the Cenozoic. There is considerable variability in the d18OSO4, with major peaks 55, 15, and 3 million years ago. There is little correlation between sulfur isotopes in marine sulfate and d18O-SO4, illustrating the fact that different processes control the sulfur and oxygen isotopic composition of sulfate. The peaks in the d18O-SO4 at 55 and 15 Ma coincide with peaks in the d13C of benthic foraminifera, highlighting the connection between the carbon and sulfur cycles in organic rich sediments. In addition, the increase in the d18O of the ocean (measured in benthic foraminifera) between 34 and 28 Ma coincides with a slight increase in the d18O-SO4. We have modeled the sulfur cycle for both sulfur and oxygen isotopes and will show model results and interpretation over several key intervals over the Cenozoic, including the Mid-Miocene Climate Optimum, the Eocene-Oligocene boundary, and the Paleocene productivity high.
PP11A-0552 0800h
Theoretical estimates of equilibrium $^1^3C-^1^8O$ clumping in carbonates and organic acids
The development of techniques for measuring small gas-phase molecules containing more than one rare stable isotope (e.g., $^1^3C^1^8O^1^6O$) at natural, ppm-level abundances$^{1,2}$ has made it possible to track sources and sinks of atmospheric gases from a new perspective. Similar measurements of $^1^3C-^1^8O$ clumping in ancient samples could improve our understanding of ancient climates, if the abundances of `clumped' rare stable isotopes in materials that retain isotopic signatures over geologic time can be measured with sufficient precision. This theoretical study estimates the abundances of such $^1^3C-^1^8O$ `clumps' in carbonates and organic acids and discusses their potential applications. Accompanying abstracts by Eiler et al. and Ghosh et al. will present the analytical methods and some initial data for carbonate minerals to examine the applicability of our theoretical models. Equilibrium isotopic speciations in carbonate minerals and organic acids are calculated from the reduced partition function ratios of isotopically substituted crystals and molecules. Vibrational frequencies used as input for these calculations come from {\it ab initio} force fields, determined using density functional theory. Our calculations indicate that carbonate minerals, including calcite, dolomite, and aragonite, when equilibrated at earth-surface temperatures, will have a slight overabundance of $CO_3^{2-}$ groups containing both $^1^3C$ and $^1^8O$ (i.e., $^1^3C^1^8O^1^6O_2^{2-}$) relative to what would be expected if carbon and oxygen isotopes were distributed randomly in the crystal lattice. Calcite and dolomite crystals are predicted to have 0.4$\permil$ excesses of $^1^3C^1^8O^1^6O_2^{2-}$ at 298 K; in aragonite the excess will be about 0.05$\permil$ larger. The excesses are smaller for crystals formed or equilibrated at higher temperatures, decreasing by 0.003\permil/$^o$C at room temperature and essentially disappearing at temperatures of 1000 K or higher. Similarly, there is an excess of both $^1^3C^1^8O^1^6OH$ and $^1^3C^1^6O^1^8OH$ groups in organic acids like formic acid ($HCOOH$) and pyruvic acid ($CH_3COCOOH$) that equilibrate at low temperatures. For gas-phase carboxylic acids, $^1^3C-^1^8O$ clumping in the $COOH$ group is strongest at the $C=O$ double bond, with an 1.0-1.1$\permil$ excess at room temperature. The $C-O-H$ subgroup has an ~0.4$\permil$ $^1^3C-^1^8O$ excess, and thus the average anomaly for the whole $COOH$ group is 0.7-0.8\permil. As with carbonate minerals, these excesses decrease at higher temperatures. The magnitude of $^1^3C-^1^8O$ clumping in carboxylic acid is similar to gas-phase $CO_2 ^{1,2,3}$, while in carbonate minerals the effect is about one-half as large. The temperature sensitivity of these isotopic clumping effects suggests that measurements of abundances of $^1^3C-^1^8O$ bonds in carbonates and organic acids could be useful for paleothermometry. The clumping equilibrium is an internal property of each phase, so temperature information can be obtained even when the isotopic composition of the fluid phase from which a sample precipitated is unknown. Clumping effects may also be able to distinguish pristine, unaltered sedimentary and biogenic carbonates and organic deposits from those that have undergone post-depositional diagenesis or metamorphism, even in samples that have not suffered extensive open-system exchange. Refs: $^1$Eiler et al. 2004, GCA {\it in press}; $^2$Schauble et al. {\it in prep.}; $^3$Wang et al. 2004, GCA {\it in press}.
PP11A-0553 0800h
The Marine Os Isotopic Record in the Eocene-Oligocene Sections From the Equatorial Pacific
The marine Osmium isotope record has great potential as a tool for chemostratigraphic correlation, and as a proxy for tracking changes in inputs to seawater. In order to evaluate the utility of this record for correlation, 45 analyses have been performed on sediment samples from the Eocene-Oligocene (E-O) section of the equatorial Pacific (Leg 199, Sites 1219 and 1218) spanning ca. 2.6 Myr. The results capture the pronounced Late Eocene $^{187}$Os/$^{188}$Os minimum and the subsequent rapid rise of $^{187}$Os/$^{188}$Os across the E-O transition, and are in general agreement with those reported earlier in E-O sections from the Pacific, the Atlantic and the Tethys (Ravizza and Peucker-Ehrenbrink, EPSL v. 210, 151-165). In detail there are important new features in the Leg 199 data. The amplitude of the Late Eocene $^{187}$Os/$^{188}$Os minimum is similar to that measured in the pelagic clay sequence of the Pacific core LL44-GPC3 but is smaller than recorded in the other pelagic carbonate records. In addition, $^{187}$Os/$^{188}$Os of samples from the E-O transition at the site 1219A are systematically higher by ca. 5-10 % compared to those of the age equivalent samples from 1218A. Such comparisons suggest that marine Os isotopic records allow general stratigraphic correlation among various sections; however, precise age correlation using $^{187}$Os/$^{188}$Os may be compromised by such small shifts in Os isotopic composition. These shifts in $^{187}$Os/$^{188}$Os among sections may be a result of hiatus in sedimentation and/or lithological variation among cores. Further work is underway that would allow comparison of variations in $^{187}$Os/$^{188}$Os in Late Eocene sections of 1219A and 1218A that are devoid of carbonates, where detailed logging data allow precise correlation between the two holes.
PP11A-0554 0800h
Ratio of multi substituted isotopologues in natural carbonates, a novel tool to study palaeo temperature.
It has recently been shown that gaseous molecules containing more than one rare stable isotope (e.g., 18O18O, 13C18O16O, etc.) can be precisely analyzed at their natural, ppm-level abundances. Application of such measurements to the geologic record will require development of analogous measurements of `clumped' rare stable isotopes in condensed phases. This study demonstrates that 13C-18O `clumps' can be measured in calcite and that their concentration reflects the temperature of carbonate growth (perhaps among other factors not yet recognized). Accompanying abstracts by Eiler et al. and Schauble et al. discuss relevant mass spectrometric methods and theoretical models of isotopic `clumping' in carbonate minerals. The concentration of 13C-18O bonds in carbonates has potential as a thermometer because they form as a result of an order/disorder reaction, such as: 40Ca13C16O3 + 40Ca12C18O16O2 = 40Ca13C18O16O2 + 40Ca12C16O3 and this reaction has a temperature-dependent equilibrium constant. The most important feature of this reaction is that it involves a homogeneous equilibrium (that is, a reaction among components of one phase, rather than between two or more phases), and therefore rigorously constrains temperature without knowing the 18O of waters from which carbonates precipitated. We analyzed the concentration of 13C18O16O in CO2 extracted from various calcite samples by phosphoric acid extraction at 25 to 50 ˚C (McCrea, 1950), with the aim of determining whether it is related to the concentration of 13C18O16O2= ionic groups in the reactant carbonate, which theoretical models presented by Schauble et al. (this volume) show should have distinctive and temperature-sensitive concentrations in calcite grown at thermodynamic equilibrium. In reporting these data, we define the variable ∆47 as the difference in permil between the measured value of R47sample (=[mass 47 isotopologues]/[mass 44 isotopologue] in the sample) and the value of R47 sample expected in that sample if its stable C and O isotopes are randomly distributed among all isotopologues- a case we refer to as the stochastic distribution. First, we analyzed a sample of MZ carbonate that had been re-crystallized at 1000 ˚C, 800˚C, 600˚C and 450˚C, yielding CO2 with ∆47 of 0.022 %, 0.1%, 0.32% and 0.45% respectively. Second, analyses of a variously purified Red Sea coral imply an uncontaminated ∆47 value of 0.598ñ0.002 (n=4); this coral grew at ~27.5 ˚C. We also analyzed a deep-sea coral, 85080 (which grew at ~4 ˚C), yielding a ∆47 value of 0.764. Finally, we analyzed a modern surface coral collected just west of Sumatra, yielding a ∆47 value of 0.637 (needing no correction after purification). The growth temperature of this coral is unknown but was likely 28-30 ˚C based on average sea surface temperatures from this area. Taken together, these data suggest that the abundance of 13C-18O bonds in carbonate has a temperature sensitivity broadly resembling, but stronger than, predicted from lattice vibration models. Potential usage of such isotopic discrimination to constrain the past global temperature history require calibration of the temperature scale with laboratory precipitated carbonates and natural carbonates cultured at a controlled temperature settings, and careful study of the effects of re-setting over geological timescales.
PP11A-0555 0800h
An orbital origin for large oxygen isotopic shifts and sea-level changes during the Oligocene
Recently, it was shown that a relatively rare orbital congruence involving low-amplitude variance in obliquity and a minimum in eccentricity coincides at the Miocene/Oligocene boundary with a large increase in oxygen isotopes (Mi1 event) and ice volume, suggesting a possible causal mechanism for such events at the million-year timescale. We tested this idea against the Oligocene oxygen isotopic and sea-level records and the astronomical timescale of Shackleton et al. (1999). As in the Miocene, large isotopic shifts (Oi events; $\delta$$^{18}$O $ > $0.5\permil, with maximum values $ > $2.7\permil) and sea-level changes (30-70 m) occurred in the Oligocene at the million-year timescale. Smaller variations in sea level (and corresponding isotopic values) characterize obliquity (10$^{4}$ years) and eccentricity (10$^{5}$ years) timescales: 15-20 m and 20-30 m, respectively. Our analysis shows that, as with the Mi1 event, Oi events relate to the anticipated orbital congruence - resolving one of the outstanding conundrums of the late Paleogene paleoclimate. It also provides support for the robustness of the astronomical time scale of Shackleton et al. (1999) for the Oligocene Epoch.
PP11A-0556 0800h
The Nature of the $\delta$$^{13}$C of Periplatform Sediments: Implications for stratigraphy
The carbon isotopic composition ($\delta$$^{13}$C) of bulk carbonates, obtained from a transect of sites drilled through platform and periplatform sediments of Holocene to Early Miocene age, have been compared to ascertain whether changes in the $\delta$$^{13}$C can be correlated between sediments of equivalent ages and whether such changes can be related to global changes in the $\delta$$^{13}$C of the dissolved inorganic carbon in the oceans over this time period. Five of the sites were drilled during Leg 166 of the Ocean Drilling Project (1003-1007) in a transect ranging from five km to 25 km away from the platform margin and penetrating sediments of Holocene to Oligocene age that are contained in 17 depositional sequences (A-Q). Two shallow water sites, Clino and Unda were situated on a extension of the same transect on Great Bahama Bank in a water depth of 10-15 m. With the exception of Unda and Clino, the $\delta$$^{13}$C of the carbonates range in isotopic composition from +5% in the younger sequences to +1% in the Early Miocene. In each of the sites the $\delta$$^{13}$C is strongly positively correlated with the percentage of aragonite. As a consequence the $\delta$$^{13}$C of sequences A through F is strongly correlated, reflecting the decreasing amount of aragonite with increasing depth. In the two platform sites, the $\delta$$^{13}$C is significantly more negative in the younger portions of the cores as a result of the influences of meteoric diagenesis during repeated exposure during the Pleistocene. Although the $\delta$$^{13}$C of the individual sequences can be correlated in most instances between the ODP holes, the changes are not related to global changes in the 13C of the oceans which in contrast to the $\delta$$^{13}$C of the platform sediments become isotopically more negative towards the present day. Instead the $\delta$$^{13}$C appears to be related to varying contributions from $\delta$$^{13}$C rich bank-top sediments. While the results from this study may not be universally applicable to all carbonates associated with shallow water platforms formed throughout geological time, they certainly cast a note of caution that apparently related changes in the carbon isotopic composition can be caused by mechanisms other than global changes in the carbon cycle.
PP11A-0557 0800h
Cenozoic variations in the South Atlantic carbonate saturation profile: Insights from the Walvis depth-transect (ODP Leg 208)
Ocean Drilling Program Leg 208 Science Party (D. Kroon, J. C. Zachos, P. Blum, J. Bowles, P. Gaillot, T. Hasegawa, E. C. Hawthorne, D. A. Hodell, D. C. Kelly, J. Jung, S. M. Keller, Y. Lee, D. C. Leuschner, Z. Liu, K. C. Lohmann, L. Lourens, S. Monechi, M. Nicolo, I. Raffi, C. Riesselman, U. Röhl, D. Schmidt, A. Sluijs, D. Thomas, E. Thomas, H. Vallius) Carbonate saturation profiles are complex and dynamic products of processes operating on temporospatial scales from the "short-term local" (e.g. carbonate export production) to the "long-term global" (e.g. carbonate-silicate weathering, shelf:basin carbonate partitioning). Established, if admittedly crude, proxies for reconstructing carbonate saturation from sediments include wt% carbonate, where values of 0-20% are typically attributed to deposition below the carbonate compensation depth (CCD), and planktonic foraminifer fragmentation, where enhanced fragmentation is typically attributed to deposition below the lysocline. Ocean Drilling Program Leg 208 successfully drilled a six-site Walvis Ridge depth-transect spanning modern water depths from 2,717 to 4,755 m. Exceptional core recovery, well-constrained biomagnetostratigraphy, and standard crustal subsidence corrections provide a working age-depth framework for contouring ship-board wt% carbonate determinations and identifying the following first-order features of the regional CCD: (1) $ > $3.5 km position from 60-48 Ma punctuated by a major transient shoaling to $ < $2 km during the Paleocene-Eocene Thermal Maximum at $\sim$55 Ma; (2) shoaling to $\sim$2.75 km from 48 to 44 Ma; (3) subsequent deepening to $ > $4.25 km from 37 to 28 Ma; (4) marked high amplitude fluctuations from 28 to 20 Ma followed by deepening to $ > $4.75 km; (5) transient shoaling to $\sim$4 km around 15 Ma followed by deepening to $ > $4.75 km by $\sim$12 Ma. These first-order features are broadly congruent with classic Atlantic CCD reconstructions by van Andel (1975) and Berger and Roth (1975). A wealth of higher frequency variation in carbonate saturation is clearly preserved within the Leg 208 depth-transect. Ongoing shore-based analyses aim to transform cm-scale variations in core physical properties (i.e. magnetic susceptibility, color reflectance) into synthetic records of wt% carbonate. These data, combined with other proxies (e.g., planktonic foraminifer fragmentation, stable isotopes) and placed within the evolving post-cruise biomagnetostratigraphic and cyclostratigraphic age-model, will provide valuable constraints on cyclic and secular fluctuations in the South Atlantic carbonate saturation profile and their relation to other major components of the earth system (e.g. pCO$_{2}$, eustacy).