U31A-0001 0800h
The EPICA challenge - predicting greenhouse gas concentrations over 800,000 years: (2)
The European Project for Ice Coring in Antarctica (EPICA) ice core from Dome C (Antarctica) has provided a new record of climate extending to 800,000 years. The imminent appearance of a dataset of greenhouse gas concentrations covering the same period has presented a unique opportunity for a test of our understanding, particularly of the climate/carbon system. The EPICA challenge (issued in Eos earlier this year) was that modeling groups and others should predict the dataset (of CO$_2$ and/or CH$_4$) before it appears, with the aim of promoting an open discussion of the underlying ideas and assumptions. People were invited to send their ideas to the Past Global Changes (PAGES) Project Office for collation. This poster, and the one that accompanies it, will summarise the responses received, and will try to highlight the underlying assumptions and uncertainties that have led to the suggested profiles. The main focus of this poster (CO$_2$ or CH$_4$) depends on the balance of responses received.
U31A-0002 0800h
The EPICA challenge - predicting greenhouse gas concentrations over 800,000 years: (2) CH4
The European Project for Ice Coring in Antarctica (EPICA) ice core from Dome C (Antarctica) has provided a new record of climate extending to 800,000 years. The imminent appearance of a dataset of greenhouse gas concentrations covering the same period has presented a unique opportunity for a test of our understanding, particularly of the climate/carbon system. The EPICA challenge (issued in Eos earlier this year) was that modeling groups and others should predict the dataset (of CO2 and/or CH4) before it appears, with the aim of promoting an open discussion of the underlying ideas and assumptions. People were invited to send their ideas to the Past Global Changes (PAGES) Project Office for collation (contact: Christoph Kull, kull@pages.unibe.ch). This poster, and the one that accompanies it, will summarise the responses received, and will try to highlight the underlying assumptions and uncertainties that have led to the suggested profiles. It is likely that this poster will highlight results for CH4, while part 1 will concentrate on CO2, although this may alter depending on the balance of responses received.
U31A-0003 0800h
Glacial-Interglacial Atmospheric CO2 Changes--The Glacial Burial Hypothesis
Organic carbon buried under the great ice sheets of the Northern Hemisphere is suggested to be the missing link in the atmospheric CO$_2$ change over the glacial-interglacial cycles. At glaciation, the advancement of continental ice sheets buries vegetation and soil carbon accumulated during warmer periods. At deglaciation, this burial carbon is released back into the atmosphere. In a simulation over two glacial-interglacial cycles using a synchronously coupled atmosphere-land-ocean carbon model forced by reconstructed climate change, I found a 547 Gt terrestrial carbon release from glacial maximum to interglacial, resulting in a 60 Gt (about 30 ppmv) increase in the atmospheric CO$_2$, with the remainder absorbed by the ocean in a scenario in which ocean acts as a passive buffer. This is in contrast to previous estimates of a land uptake at deglaciation. This carbon source originates from glacial burial, continental shelf and other land areas in response to changes in ice cover, sea level, and climate. The input of light isotope enriched terrestrial carbon causes atmospheric $\delta^{13}$C to drop by about 0.3\permil at deglaciation, followed by rapid rise towards a high interglacial value in response to oceanic warming and regrowth on land. Together with other ocean based mechanisms such as change in ocean temperature, the glacial burial hypothesis may offer a full explanation of the observed 80-100 ppmv atmospheric CO$_2$ change.
U31A-0004 0800h
Role of the Ocean in Changes of Atmospheric CO$_2$ on Glacial-Interglacial and Millennial Timescales
Measurements of air trapped in ice cores show changes in atmospheric CO$_2$ contentrations of ~80 ppmv on glacial-interglacial timescales and intra-glacial variability of ~20 ppmv on millennial timescales. Using highly idealized (box or zonally averaged) models, it has been suggested that the ocean accounts both for glacial to interglacial changes of atmospheric CO$_2$ as well as for the intra-glacial variability. Here I reassess both problems with a global coupled climate model (UVic ESCM) containing state-of-the-art 3D ocean circulation, ecosystem and carbon cycle modules. Simulated variations of atmospheric CO$_2$ in response to changes in North Atlantic Deep Water (NADW) formation are much smaller (~2 ppmv) than previously found. This suggests a much larger role of the terrestrial biosphere in millennial CO$_2$ variations than previously assumed. Much larger variations (~20 ppmv) are simulated for glacial-interglacial changes, although still largely underestimated compared to the observations. The processes leading to these modelled changes, particularly deep ocean ventilation, air-sea gas exchange and productivity, are analysed. The potential role of missing processes such as increased productivity through iron fertilisation and calcium carbonate compensation within the sediments are discussed. I conclude that we are still a far way from understanding the natural variability of the carbon cycle and that the simulation of both glacial-interglacial and millennial variability remains a key challenge for Earth System Models.
U31A-0005 0800h
Northern-Southern Hemisphere Comparison of Aeolian Dust Records Over the Last 800 Kyr.
The 3130m-deep ice core from Dome C obtained in the frame of the European Project of Ice Coring in Antarctica (EPICA) depicts 8 glacial periods. Here we present the results of dust content and we extend the record down to 3200 m, corresponding to circa 800 kyr B.P. The dust has been measured by using two independent techniques: a laser system deployed on the field and used for continuous measurements and counting in laboratory by using Coulter Counter (Multisizer II) on discrete samples from about 1200 levels. The laser and Coulter EPICA dust profiles are well consistent. Amongst Antarctic records, over the last 4 climate cycles, the EPICA-Dome C dust record mimics the Vostok one and several dust events could be confidently used as stratigraphic markers since the records share the same dust sources. The high dust input (up to a factor 50 with respect to present time) characterising the high latitude records during glacial climate is linked to the synergetic effect of several factors, among which the extension of ice caps and periglacial areas, the bigger continental aridity and reduced hydrologic cycle, as well as the change in the atmospheric circulation. At the first order, the dust concentration could be linked to glacial conditions and therefore to the global ice volume. The EPICA dust record is extended to MIS 20. The record shares several common pattern with the global ice volume, and also with the Northern Hemisphere dust records from the Chinese loess and with marine records from western Pacific ocean. Among patterns, the glacial stages MIS 14 and MIS 16 appear less and well defined respectively while the interglacial period MIS 13 and MIS 15 are not well marked on all records. For prior periods, EPICA record displays better resolution and marked glacial MIS 18 and 20 stages. Interestingly, the long term progressive increasing glaciation depicted from western Pacific record likely associated to the Himalayan uplift, is also suggested from the EPICA dust record.
U31A-0006 0800h
In Situ Production Of CO$_2$ In The Antarctic Siple Dome Ice Core Record
Understanding how the atmospheric concentration of CO$_2$ changed in the past in response to other changes in the climate system provides us with a better understanding of how current and future changes in the carbon cycle will influence our future climate. CO$_2$ records from Antarctic ice cores are considered to be representative of paleoatmospheric concentrations because its low dust content might not allow significant in situ production of CO$_2$ by carbonate-acid reaction as shown in Greenland ice cores. Despite similarities in general time series of CO$_2$ records with other Antarctic ice cores, at some depth intervals, the Siple Dome ice shows higher CO$_2$ concentrations by up to 20 ppm (ݍmol CO$_2$/mol air) than those in other Antarctic ice cores. In order to examine the possible mechanisms of CO$_2$ production in the Siple Dome ice core, we check the CO$_2$ difference between Siple Dome and Taylor Dome or Dome C. With the latest results of chemical and physical studies of the Siple Dome ice we discuss three possible mechanisms for the in situ production of CO$_2$ in the Siple Dome ice core: (1) carbonate-acid reaction, (2) oxidation of organic compounds, and (3) snowmelting/refreezing.
U31A-0007 0800h
Mineral Dust Elemental Composition Over the Last 220 Kyr from the EPICA-Dome C ice core (East Antarctica)
Mineral windblown aerosol (dust) represents a small fraction of the total mass of atmospheric aerosols. However, it may play an important role in climate and, in turn, it is itself sensitive to climatic changes. The investigation of the temporal variability of dust flux as well as the changes of its mineralogical composition within different climatic periods offers a unique way of assessing the source-related environmental changes, the variation in atmospheric circulation, and the potential influence of dust on biogeochemical cycles, for several key elements (such as Si and Fe). The mineral composition of dust found in ice cores is still poorly known, because few techniques are able to provide information on the chemical composition of the insoluble aerosol fraction, because of the very low mass of dust usually present in polar ice samples. Total volume of mineral content is provided by Coulter Counter measurements, and the dust record shows higher dust input during glacial periods than during interglacials (Delmonte et al., 2004). Sr and Nd isotopic tracers for identification of dust provenance to East Antarctica suggest southern South America as the dominant common source for dust in glacial periods of the late Pleistocene (Delmonte et al., 2004), whereas dust provenance during interglacials periods is still highly uncertain. Elemental composition of mineral dust from the Dome C ice core (75° 06' S, 123° 21' E) drilled in the framework of the European Project for Ice Coring in Antarctica (EPICA) is presented, showing measurements obtained by the Proton Induced X-ray Emission (PIXE) technique applied on insoluble atmospheric dust in ice cores. Within this work, substantial improvements to the experimental set-up and data processing have been made, compared to previously presented works on the first 2200 m of the EDC ice core, corresponding to about 220 kyr of climatic history (Marino et al., 2004; Ghermandi et al., 2003). The PIXE technique, based on X-ray spectrometry, allows direct measurements on filters of the insoluble dust fraction without sample pre-treatment, with analytical detection limits less than 1 ppb. The minimum required mass density of sample material on the filter is about 0.1 microg cm-2. Improvements made here allowed us to obtain accurate records of 8 major and minor crustal elements (Si, Al, Fe, Ti, K, Ca, Mg, Na). The data for these 8 elements, as well as their proportions as oxides, are shown for different climatic periods. Because O, Si, Al, Fe, Ti, K, Ca, Mg, Na are the principal constituents (99.90 %) of the Earth's Continental Crust, a comparison with literature data can reveal changes in relative elemental abundances, and hence variable composition of dust under different climatic conditions.
U31A-0008 0800h
900 kyr of Paleo-Volcanism From the EPICA Ice Core (Dome C - Antarctica).
In the framework of the EPICA program (European Project for Ice Coring in Antarctica), a deep ice core was drilled at Dome C (East Antarctic Plateau) reaching the depth of about 3200 (corresponding to about 900,000 years) at the end of the '02-'03 drilling season. High-resolution measurements of sulfate (every 1.5-2 cm, corresponding to 1-50 years depending on depth) were carried out by Fast Ion Chromatography (FIC) directly in the field, allowing the reconstruction of the paleo-volcanism from the Dome C record covering the last 9 glacial cycles. The volcanic signatures were distinguished from the biogenic sulfate background by evaluating a threshold by a statistical approach. For the last millennium the Dome C volcanic record was compared with those coming from several Antarctic ice cores, and differences in the depositional fluxes of the events memorized at different sites were interpreted in terms of changes in snow accumulation rate and atmospheric circulation. For older ages, the correlation between changes in the frequency of volcanic signatures and climatic proxies (such as dD) was carried out aiming to enlighten possible links between climatic changes and volcanism. The sharp decrease of the event frequencies recorded in the deeper part of the core is interpreted in terms of both a decrease of temporal resolution and diffusion of sulfate in the ice. Anomalously high sulfate spikes recorded in the deepest part of the ice core could be interpreted as the effect of depositional processes of local-volcanism products coupled with dust spikes able to prevent the sulfate diffusion or caused by aggregation processes of non-volcanic sulfate-salt particles.
U31A-0009 0800h
EPICA-Dome C Ice Core FIC Analysis: the Longest Chloride, Nitrate and Sulfate High Resolution Record From an Ice Core (900 kyr).
In the framework of the EPICA project, Fast Ion Chromatographic (FIC) analysis of the 3200 m of the ice core drilled at Dome C yielded a continuous high resolution record of Cl-, NO3- and SO42-, spanning the last 900 kyr and covering the last nine glacial/interglacial cycles. About 140 kdata for each component, with resolutions ranging from 2.0 to 4.0 cm (covering from 1.5 to 50 years), were performed in the field and in the cold laboratory of AWI on continuously melted firn and ice core sections. Cl-, NO3- and SO42- profiles were compared together with the dust and isotopic (dD) smoothed profiles in order to enlighten leads and lags between environmental and climatic changes. Particular attention was devoted to the chemical trend interpretation of the fast climatic variations occurring during the glacial onsets and terminations and the interstadial periods. Since Cl- and NO3- are not irreversibly fixed in the snow layers and can be re-emitted into the atmosphere as gaseous acidic species, their preservation in the snow is affected by accumulation rate and acidity. In particular, the NO3-profile shows a close positive relationship with dust content demonstrating that the acidity neutralization drives the preservation of NO3- in low accumulation periods. Cl-, which are stable for relatively low accumulation rate (higher than 80 kg m-2 yr-1) shows a higher sensitivity to accumulation rate changes. Indeed, high Cl- levels are recorded both in glacial period characterized by high dust content and in high accumulation rate periods such as interglacial climatic optima and interstadials. The SO42- profile is marked by volcanic signatures superimposed to the biogenic background contribution (mainly biogenic). Preliminary evidences showed biogenic SO42- depositional fluxes in the snow was not significantly different in the different climatic stages. In this, way, the SO42- background profile is only tuned by changes in snow accumulation rate constituting a potential proxy-marker for accumulation rate.
U31A-0010 0800h
A Record of Multiple Sea-Level High-Stands for the Last 650 ka Preserved at Henderson Island, a Coral Atoll in the South Pacific
The U-series chronometer has a wide range of applications in paleoclimatology. Uranium-series dating of fossil coral reefs provides an excellent record of past sea-level high-stands and offers an independent test of the validity of the Milankovitch theory of climate change. However, U-series ages have a number of limitations, as many corals tend to display poorly understood open-system behaviour due to diagenesis, often leading to inaccurate ages. Also, the poor knowledge of the marine $^{234}$U/$^{238}$U in the past limits the accuracy of the method. Fossil corals from the last interglacial warm period, or MIS 5.5 ($\sim$125 ka), have been studied extensively, however data from older interglacial and interstadial periods are limited. This is mainly due to fewer accessible sample sites, and the tendency for older corals to exhibit a larger degree of diagenetic effects compared with younger MIS 5.5 corals. Another important issue for older samples is the necessity for superior measurement precision, in order to obtain precise ages from samples in which the radioactive isotopes have evolved to a state close to equilibrium. We present new U-series ages of corals from Henderson Island, a fossil coral atoll in the southern Pacific. Previous U-series ages for this locality show that reef growth occurred at $\sim$320 and $\sim$630 ka, correlating respectively with the MIS 9.3 and MIS 15 sea-level high-stands (Stirling {\ et al.}, Science 291, 290-93, 2001). Improved analytical techniques have been developed for the purpose of acquiring highly precise ages for old corals, such as those on Henderson Island. Measurements are conducted using a Nu Instruments Nu Plasma MC-ICPMS (Andersen {\ et al.}, Int. J. Mass Spec 237, 107-118, 2004). Our technique obtains superior precision compared to earlier U-series protocols, allowing 300 ka samples to be dated with age uncertainties of better than \pm 1 ky (2\sigma). Our newly acquired U-series ages for Henderson Island correlate with MIS 15, 9.3, 7.3 and 7.1. These data not only offer the possibility to test the validity of the Milankovitch theory for climate episodes occurring before MIS 5.5, but also allow processes of reef diagenesis to be constrained, and provide critical information on the marine $^{234}$U/$^{238}$U over the past 650 kyrs.
U31A-0011 0800h
Warmer tropics during stage 11 - evidence from the Cariaco Basin
Marine Isotope Stage (MIS) 11 stands out as the most extreme interglacial episode of the last million years. But evidence for unusual warmth in the tropics at this time remains equivocal. A proxy SST record from the equatorial western Pacific (ODP 806B - Lea et al., 2000; Medina et al., this meeting) suggests that MIS 11 was the warmest time interval of the last million years. A key question that directly addresses the hypotheses proposed to explain MIS 11 is whether anomalously warm SSTs were a common feature of the tropics at this time. The Cariaco Basin, an anoxic marine basin on the northern shelf of Venezuela, preserves a unique high resolution climate record of the southeastern Caribbean/tropical Atlantic. This record is continuous through the last 580,000 years and includes multiple varved intervals. Records from this basin have been instrumental in establishing ties between tropical climate records and high latitude ice core and oceanic records. Paleothermometry based on planktic foraminiferal Mg/Ca works well in the Cariaco Basin because of the presence of a wide range of planktic species and superb preservation of carbonates, including pteropods. We have analyzed Mg/Ca in samples from ODP Hole 1002C (10°42.73'N, 65°10.18'W, 893m water depth) that span MIS 11. Our preliminary results, covering the interval 402-409 ky BP, from surface dwelling Globigerinoides ruber yield Mg/Ca values of 4-5 mmol/mol, equivalent to SSTs of 26.5-28 degrees C. For comparison, Mg/Ca values in the Holocene range from 4-4.5 mmol/mol, equivalent to SSTs of 26-27.5. We are in the process of adding further results to evaluate how systematic this difference is. Poore and Dowsett (2001) reported oxygen isotope results from the same sequence and species and found that MIS 11 values were 0.3 permil more depleted than the Holocene. Our preliminary Mg/Ca-SST results suggest that most of this difference is not due to higher SST, suggesting either lower Cariaco salinity at this time and/or a negative shift in mean ocean O18 related to higher sea level during MIS 11. These findings will be considered in light of the unusual climate state that prevailed during MIS 11.
<a href='http://www.geol.ucsb.edu/faculty/lea/' >http://www.geol.ucsb.edu/faculty/lea/
U31A-0012 0800h
The Thermal Evolution of the Western Equatorial Pacific During the Midde and Early Pleistocene
Magnesium/calcium data from planktonic foraminifera in the equatorial Pacific sediment cores suggested that tropical Pacific sea surface temperatures (SST's) were about 3 deg.C colder than modern conditions during glacial episodes of the last 500 thousand years (ky). We have extended the Western Equatorial Pacific (WEP) Globigerinoides ruber Mg/Ca and d18O records, from the Ontong Java Plateau, Ocean Drilling Program Hole 806B (0 19.11'N, 159 21.69'E, 2520m), back to 1.3 ma. For temperature conversion we used a G. ruber calibration based on core-tops from the tropical Pacific (Lea and Martin, 1996), which yields the following relationship: Mg/Ca(mmol/mol)= 0.30exp[0.089 X SST deg. C)]. Our SST record shows that the previously observed 3 deg C-colder SST's than modern conditions during glacial episodes are generally the rule for the entire 1.3 ma period. Glacial-interglacial temperature differences as great as 4 deg. C are observed, even in the Early Pleistocene (1-1.8 ma). Preliminary observations suggest the presence of ~40 Ky SST cycles during the Early Pleistocene, of similar amplitude to the dominant SST cycles seen in Late Pleistocene Tropical records. Early Pleistocene WEP SST's, as suggested by our record, oscillated between 26 and 30 deg. C. MIS 11 stands out as the most prominent feature of the WEP SST record. The potential bias on temperature estimates due to the influence of changes in lysocline depth (Farrell and Prell, 1989) and decrease preservation with depth (Lea et al., 2000) is ± 0.8 deg.C. Preliminary point to point comparison between the SST and the d18O records shows that Mg/Ca-based temperatures lead over d18O by about 3 ky as previously determined by Lea et al. (2000).
U31A-0013 0800h
Relatively high sea levels of the last million years
Estimates of the elevations of relatively high sea levels (above present) of the last million years continue to be problematical because of critical unknown factors. Such sea levels have been estimated: by inferences drawn from oxygen isotope measurements on benthic foraminifera - using Mg/Ca ratios to eliminate the temperature term; and from marine and littoral stratigraphic units and shoreline geomorphology, the actualities of which are essential for confirming estimates by other means. Both approaches encounter major difficulties: the former because local or regional hydrologic effects are difficult to estimate; the latter, because of the ambiguity of water levels inferred from corals, marine or littoral lithofacies, repeated marine occupation of shorelines on relatively low uplift coastlines (do `stable` coastlines exist?) and, not least, considerable uncertainty for shoreline age estimates, especially for those older than event 5.5. On rapidly uplifting coastlines, at or near plate margins, some success has been obtained in separating the uplift and sea level terms. But a long-standing problem at intraplate locations is the need to quantify medium-term uplift rates, between shorter-term (Holocene) and longer-term (earlier Pleistocene) ones. Approaches to these problems are reviewed and, using boundary parameters defined by late Pleistocene and late Pliocene shorelines, sea-level estimates are presented from Australasian and Atlantic intraplate locations and Pacific plate margin locations. These are correlated with benthic oxygen isotope stratigraphy from Site 849 and with Antarctic (EPICA) palaeotemperatures. Such correlations allow inferences to be drawn about trends in the evolution of relatively high sea levels, ice-sheet evolution and melt-water sources. A major reorganization of the `ocean-ice-sheet-sea-level system` occurred after oxygen isotope stages 12/11; thereafter, the role of the Laurentide Ice Sheet became progressively more important, and relatively higher sea levels obtained.
U31A-0014 0800h
A Million-Year Record of Glaciation in the Tropical Andes
We present a longterm record of glaciation in the tropical Andes based on cosmogenic dating ($^{10}$Be) of boulders on moraines. Well-preserved moraines in deglaciated valleys bordering the Junin Plain in central Peru ($\sim$11$\deg$S, 76$\deg$W, 4000 m) were deposited during several glacial cycles extending back more than one million years before present (1 Myr BP). The presence of boulders with zero-erosion $^{10}$Be exposure ages $>$1 Myr constrains boulder erosion rates to relatively low values. For boulders at high altitudes, however, even low boulder erosion rates (0.3 to 0.5 m/Myr) make calculated old exposure ages markedly older [e.g., $\sim$20% older for a zero-erosion age of 400,000 $^{10}$Be years (400 $^{10}$Be kyr)]. Exposure ages recalculated with boulder erosion rates of 0.3 m/Myr straddle interglacial marine isotope stage (MIS) 11 ($\sim$430-390 kyr BP), fall within glacial MIS 12 ($\sim$480-430 kyr BP), but skip over glacial MIS 16 ($\sim$670-630 kyr BP), perhaps the largest ice volume of the past 2 Myr. Increasing the erosion rate used in the calculations to 0.5 m/Myr moves ages into both MIS 11 and MIS 16. If we assume that the older Andean glaciations were indeed synchronous with global ice volume, our data suggest that boulder preservation cannot be treated as a simple linear process. Conversely, the data may be suggesting correctly that glaciation of the tropical Andes was not synchronous with the global glaciations as inferred from the marine isotope record. Our chronology for the last glacial maximum (LGM) in the region supports the idea of asynchrony between the global ice volume record and the terrestrial record of glaciation in the tropical Andes. The LGM in the Junin region of Peru and in the Cordillera Real of Bolivia (16$\deg$S 68$\deg$W) occurred $\sim$34 to 22 $^{10}$Be kyr BP and was less extensive than older glaciations. Asynchrony between the LGM in the Northern Hemisphere ($\sim$21 kyr BP) and the tropical Andes suggests that previous glaciations in the tropical Andes may have been similarly out of step.
U31A-0015 0800h
Radioisotopically Dated Climate Record Spanning the Last Interglacial in Ice from Mount Moulton, West Antarctica
Age models for paleoclimate records extending into and beyond the last interglacial largely rely on either orbital tuning (Milankovitch theory of climate change) or ice flow modeling. Few late Pleistocene paleoclimate records have been radioisotopically dated for testing the Milankovitch theory of global climate change. Within the ice cap near the summit of Mount Moulton, West Antarctica a series of stratigraphic englacial volcanic ash layers (tephra) have been precisely dated with $^{40}$Ar/$^{39}$Ar to between 10 and 500 kyr bp. Using these dated tephra layers as control points we developed radioisotopically dated paleoclimate records. Coherent $\delta$$^{18}$O$_{ice}$, CH$_{4}$, and $\delta$$^{18}$O$_{atm}$ records have been established throughout the last interglacial and the start of the last glacial periods. The Moulton $\delta$$^{18}$O$_{ice}$ record is the first such record of the last interglacial in West Antarctica. The integrity of the Moulton $\delta$$^{18}$O$_{ice}$ record is reinforced by its many shared features with ice stable isotope records at Vostok and EPICA-Dome C in East Antarctica. Similarly, the records derived from the trapped gases (CH$_{4}$ and $\delta$$^{18}$O$_{atm}$) at Moulton correlate well with those of Vostok. A time scale independent of flow modeling or orbital tuning was developed for the Moulton sequence based on four datable tephra layers. One well-dated tephra layer falls in ice that was deposited just before the penultimate glacial termination and suggests that this transition began shortly after 135.8 $\pm$ 1.1 kyr bp. For the penultimate glacial termination, the Moulton age of 135.8 $\pm$ 1.1 agrees extremely well the same feature in the Vostok ice core based on the EGT4 timescale. We discuss the potential for the radioisotopically derived Moulton age model to be applied to any ice core climate record for this interval via gas correlation.
U31A-0016 0800h
Late Pleistocene Sea-level and Deep-sea Temperature Changes Constrained by U.S. Mid-Atlantic Margin Sequences
We assembled and dated a late Pleistocene (last 130 kyr) sea-level record based on sequence stratigraphy from the U.S. middle Atlantic margin. The timing and magnitude of these sea-level changes are similar to those reported from uplifted coral terraces in New Guinea and Barbados, suggesting that we have established a global record of late Pleistocene sea-level change. Comparison of this eustatic record with benthic foraminiferal oxygen isotope records shows that the deep sea cooled ~2.5\deg C between Marine Isotope Chrons (MIC) 5e and 5d (~120-110 ka) and that near freezing conditions persisted until Termination 1a (14-15 ka). The pattern of deep-sea cooling follows a hysteresis loop between two stable modes of operation. Cold, near freezing deep-water conditions characterize most of the past 130 kyr. In contrast, two warm intervals (the Holocene/MIC 1 and MIC 5e) resulted from rapid warming during the terminations; rapid cooling followed the peak warmth of 5e and presumably the same would be beginning today if not for anthropogenic warming.
U31A-0017 0800h
Sulfur Mass Independent Fractionation in the Oxidizing Atmosphere?
The discovery of mass-independent fractionation (MIF) in sulfur isotopes in Archean sediments and a lack of MIF in sulfur in the younger rocks (Farquhar, 2000) places a strict upper constraint on the amount of oxidants and oxygen in the ancient atmosphere prior to 2.3 Gyr ago (Pavlov and Kasting, 2002). However recent Antarctic ice core measurements show a small, but non-zero, MIF in sulfate correlated with the large volcanic eruptions (Savarino et al., 2003). Moreover, recent measurements of isotopic compositions in present-day sulfate aerosols also exhibit non-zero MIF (Romero and Thiemens, 2003). MIF must be produced by UV photolysis in the stratosphere or mesosphere, but sulfur re-entering the troposphere takes only one chemical form (sulfate), making preservation of any sulfur MIF signature difficult to explain. We added sulfur isotopes to the Garcia-Solomon 2D dynamical/chemical model with aerosol microphysics and looked at how strong the initial atmospheric fractionation has to be in order to be preserved upon descent to the troposphere. Our calculations can quantitatively constrain the oxidative capacity of the atmosphere over the period of the ice core record. We will discuss two possibilities to preserve small sulfur MIF in the present atmosphere: 1) Preservation through stratospheric and mesospheric sulfate aerosols 2) Preservation through the alternative pathway of SO2 oxidation (Savarino et al., 2003).
U31A-0018 0800h
Glacial Sea Level Variations and Their Effects on Resonance of the Ocean Tide in the Labrador Sea: Possible Implications for Ice Sheet Dynamics and Millennial Scale Climate Variation
During the last 800,000 years, the ice age Earth was subject to a series of glacial cycles of period $\sim$100,000 years. Over much of this time, large ice complexes covered most of Canada and Northern Europe. At the various glacial maxima, sea level was 100 m or more lower than during interglacial periods. As a result, the geometry of ocean basins was altered from that of the present day. High-resolution sea-level, sediment, and ice core data for the last glacial cycle have provided compelling evidence of a rich temporal evolution in ice age climate. Throughout the globe, a profound millennial scale variability has been found that correlates with Heinrich events, which are believed to represent massive discharges of ice from the Labrador Sea. The cause of this instability is a matter of ongoing debate. In this talk we report on work that suggests that Labrador Sea tides were anomalously large in glacial times. We therefore suggest that tides played a catalytic role in destabilizing the Hudson Strait ice stream and Labrador Sea ice shelf, and that tides thus provide an important, missing link in understanding the origin of millennial scale climate events. Present-day semi-diurnal and diurnal tides have been interpreted as resonant normal modes of the ocean basins, and thus relatively small changes in the shape of ocean basins may substantially alter the tidal dynamics. As a first step towards understanding this sensitivity, we have investigated tides over the last 65,000 years using a new global numerical model that captures 92 percent of the present-day open-ocean tidal height variance from ``first-principles'' (i.e. without assimilating observations). The tide model utilizes a new formalism for predicting gravitationally self-consistent sea level change on realistic, viscoelastic Earth models. The formulation yields maps of sea level change that exhibit complex spatial patterns reflecting the deformation and self-gravitation associated with the (ice plus ocean) load. From approximately 65,000 to 7,000 years before present, Hudson Bay was ice covered and the tides at the debouchement point of the Hudson Strait ice stream in the Labrador Sea are predicted to be exceptionally large. M$_{2}$ tides are 6 meters peak-to-peak and spring tides are 10 meters peak-to-peak. Since observations in present-day Antarctica demonstrate that tides significantly impact the dynamics of both continental ice streams and floating ice shelves, our results suggest that large paleotides in the Labrador Sea weakened the ice shelves and streams and served as a precondition for the Heinrich event instability.
U31A-0019 0800h
Simulating and Investigating the 100ka ice age cycle with a three dimensional ice sheet model and GCM.
One of the challenges of earth system modelling is to confirm the theories of ice age cycle by simulating the realistic response of the climate system to the change of orbital parameters, known as Milankovitch forcing, by phisically based models instead of conceptual models. Here we simulate the ice age cycle and investigate the origin of 100ka cycle using a three dimensional ice sheet model which includes the thermo-mechanical coupling (verified by the simulation of Antarctica and Greenland ice sheet) with the process of isostatic rebound. We also use an atmospheric GCM (CCSR/NIES) coupled to a slab ocean to estimate the climate sensitivity to Milankovitch forcing and atmospheric CO2 indicated by ice core data. Within the range of possibilities of the physically based model, ice age cycles with a sawtooth shape 100 ka cycle and the major NH ice sheet (Laurentide ice sheet and FennoScandian ice sheet) volume and geographical distribution in the past 400,000 years are successfully simulated. It is shown that both contribution of CO2 and Milankovitch forcing are necessary for the ice age cycle, although the CO2 change affects the global climate change. The effect of orbital forcings, long-term CO2 change, response time of crustal rebound and continental distribution upon the origin of 100 ka cycle will be investigated and discussed.