PP13B-01 INVITED 13:40h
The role of changing seasonality in North Atlantic abrupt climate changes
Cooling around the North Atlantic associated with ice-age millennial oscillations was primarily a wintertime phenomenon, enforced and amplified by sea-ice formation. As summarized by Denton et al. (submitted, QSR), Greenland ice-core indicators of mean-annual temperature change show the Younger Dryas much colder than the Little Ice Age and almost as cold as the glacial maximum, whereas indicators primarily of summer temperature in Greenland and NW Europe, such as moraine positions, indicate a Younger Dryas much warmer than glacial maximum and not much colder than the Little Ice Age. A large, Antarctic-like seasonal cycle in Younger Dryas sea ice explains these observations, and is consistent with a wide range of data and models. The "conveyor" circulation is important not just for heat transport to the North Atlantic but also for salt transport that maintains high surface density and allows wintertime deep mixing to suppress sea-ice formation. The abrupt-climate-change "switch" for the ice-age events may be as simple as whether North Atlantic water sinks before it freezes, or freezes before it sinks.
PP13B-02 13:55h
Pleistocene Glaciation and the Stability of North Atlantic Thermohaline Circulation
Geochemical profiles from the North Atlantic Ocean suggest that the vertical $\delta13C$ structure of the water column at intermediate depths did not change significantly between glacial and interglacial time over much of the Pleistocene, despite large changes in ice volume and iceberg delivery from nearby landmasses. After the Mid-Pleistocene Transition (MPT), warmer interglaciations may have resulted in periodically (and present-day) open sea ice conditions in the Norwegian-Greenland Sea. The high $\delta13C$ deep water that forms in the NGS region today may be geologically unusual and restricted to the few extreme interglaciations of the late Pleistocene. Over most of the Pleistocene, Norwegian Sea Overflow Water appears to have had low $\delta13C$ values, similar to Southern Ocean Water, and is seen as a mid-depth low-$\delta13C$ layer between 1 and 2 km depth south of the Greenland-Scotland Ridge. If northern source deep waters can have variable $\delta13C$, then this likelihood must be considered when inferring past circulation changes from benthic $\delta13C$ records. The larger ice volumes attained after the MPT may also have resulted in the periodically large and rapid discharges of freshwater associated with Heinrich events. Only such extreme freshwater perturbations appear to be able to significantly disrupt the THC from its normal mode that characterizes most of the Pleistocene. With the modern configuration of land and ice, it seems unlikely that future alterations to high latitude run-off would constitute a density perturbation as large as a Heinrich event. The question remains whether proposed hydrographic changes at high latitudes would fall outside the range of the glacial-interglacial boundary conditions under which North Atlantic THC appears to have been relatively stable.
PP13B-03 INVITED 14:10h
Glacial Water Mass Geometry and the Distribution of $\delta^{13}$C of $\Sigma$CO$_{2}$ in the Western Atlantic Ocean
Oxygen and carbon isotopic data were produced on the benthic foraminiferal taxa {\it Cibicidoides} and {\it Planulina} from twenty five piston cores, gravity cores and multi-cores from the Brazil margin. The cores span water depths from about 400 m to 3000 m and intersect the major water masses in this region. The Holocene and glacial bathymetric profiles of benthic foraminifera $\delta^{13}$C show significant differences. The Holocene bathymetric profile along the Brazil margin shows the presence of North Atlantic Deep Water as a local maximum in $\delta^{13}$C centered at about 2500 m, intersecting the northward flowing water masses Antarctic Intermediate Water/Circum-polar Deep Water and Antarctic Bottom Water. The glacial bathymetric profile of $\delta^{13}$C of $\Sigma$CO$_{2}$ requires the presence of three distinct water masses in the glacial Atlantic Ocean: a shallow ($\sim$1000 m), southern-source water mass with an end-member $\delta^{13}$C value of about 0.3-0.5 $\permil$ VPDB, a mid-depth ($\sim$1500 m), northern-source water mass with an end-member value of about 1.5 $\permil$, and a deep ($ > $2000 m), southern-source water with an end-member value of $ < $-0.2 $\permil$ and perhaps as low as the -0.9 $\permil$ values observed in the South Atlantic sector of the Southern Ocean [Ninnemann and Charles, 2002]. The origins of the water masses are supported by the meridional gradients in benthic foraminiferal $\delta^{18}$O: at similar depths in the water column, the southern water masses are colder, less salty and have a higher benthic foraminiferal $\delta^{18}$O value than the water mass originating in the north. Although there is evidence for significant aging and mixing of the water mass along its trajectory, the mid-depth northern-source water mass crossed the equator and could still be identified as a unique water mass as far south as 30$\deg$ S. In contrast, the shallow, southern-source water mass did not exert a significant influence in the northern subtropical Atlantic. These new data fill a critical gap in the South Atlantic Ocean and provide the motivation for updating the classic glacial Atlantic $\delta^{13}$C transect of Duplessy {\it et al.} [1988]. A revised glacial section of western Atlantic $\delta^{13}$C of $\Sigma$CO$_{2}$ documents the positions and gradients among these intermediate and deep water masses. The strong hydrographic gradients observed below 2000 m in the North and South Atlantic most likely result from mixing along the mid-Atlantic Ridge system between water masses of very different end-member $\delta^{13}$C composition.
PP13B-04 INVITED 14:25h
Calcareous Nannofossils and Orbitally Tuned Cyclostratigraphy in the Cenozoic
The well known reputation of calcareous nannofossils as powerful biostratigraphic and dating tools has been improved in the last two decades, with a decisive advance towards consistently successful biostratigraphy. This has been obtained by improving the nannofossil database, following the pioneering paper of Backman and Shackleton (1983), with the application of a strategy that combines quantitative data-gathering techniques with high resolution sampling. The effort to optimize the analytical methodologies, in the study of an increasing numbers of suitable sedimentary successions, has generated excellent results for different intervals of geological time. The improvement of database is linked to improvements in the quality of sedimentary successions, as the high-quality sediment cores recovered by the Ocean Drilling Program (ODP). It has been proved that high-quality biostratigraphic data provide the potential for high resolution calibration between nannofossil biohorizons and the other records used for dating as, specifically, orbitally tuned cyclostratigraphy. In this way the accurate time resolution, which distinguishes this data sequence, is transferred to the biohorizons, and a precise biochronology is obtained. On the other hand, detailed nannofossil biostratigraphy can be used to provide preliminary control points for the astronomical tuning of cyclostratigraphic records. This rigorous approach in the study and use of calcareous nannofossils has been greatly stimulated by Nick Shackleton. The introduction of orbitally tuned time-scales has been crucial for the chronological precision of nannofossil biohorizons, and has provided an accurate chronologic framework which is continuously improved, in various intervals of the Cenozoic. Furthermore, an accurate database on ancient nannofossil communities provides informations useful for clarifying their possible paleoceanographic meanings and their paleoecological affinities, and for describing the evolutionary "behaviour" of nannofossils during prominent climatic episodes and climatic transitions.
PP13B-05 14:40h
Oxygen Isotopic Stratigraphy and Geomagnetic Field Intensity
Nick Shackleton intellectual leadership in isotope stratigraphy had profound implications for paleoclimatology and paleoceanography. What is not so well enough known is that Nick contributed also to significant advances in studies of variations of the Earth's magnetic field. The first link between the two disciplines was certainly the paper that he produced with Neil Opdyke in 1973. About twenty years later Nick Shackleton obtained a very detailed isotope stratigraphy after analyzing the oxygen isotopes of bulk sediment of two cores from the Somali basin characterized by high resolution records of relative paleointensity for the past 140 kyr. This stratigraphy allowed us to correlate these two records with other independent data from the Mediterranean sea and to propose that the signal which was recorded at the two locations was global and thus of geomagnetic origin. The fast growing database made possible a stacking of the results for the past 200 ka and then for the past 800 kyr. The resulting curve was constructed from 33 paleointensity records and Shackleton's isotopic records were essential in many cases. Indeed without high resolution stratigraphy much information could not be retrieved due to uncertainties in correlating different records. The results revealed the very variable character of the field with large 20 to 60 kyr oscillations and changes in amplitude that can exceed a factor of five, but no apparent periodicity. Short periods of very low intensity occur at more or less regular intervals (roughly every 100 kyr) and correspond with geomagnetic excursions. The next step was to obtain a much longer record that would document the field changes across reversals and during entire polarity intervals. The opportunity was met during ODP Leg 138 with the recovery of beautifully magnetized sequences that covered at least 4 Myr of geomagnetic history. Nick was responsible for correlating sedimentary columns taken at each site. He performed a tremendeous task by orbitally tuning the density variations for thousands of meters of sediment. This allowed us to correlate the paleointensity signals from several holes and to produce the first long dataset with a very accurate time-depth control. Very recently the accumulation of data made it possible to stack records from different oceans for the past 2 Myr and to extract features of field intensity which add significant constraints to the modelling of the geodynamo. Alternatively this curve can be used as a stratigraphic reference, similarly to isotopes records.
PP13B-06 14:55h
Sea-surface and deep-sea temperatures and seawater d18O in the Southern Ocean over the last 440,000 years
The circulation of Antarctic Bottom Water is one of the controlling factors in the Earth's heat budget, and therefore Southern Ocean paleoceanography is critical to understanding glacial-interglacial climate change. Ocean Drilling Program (ODP) Site 1123 (3290 m water depth) on Chatham Rise (41° 47.15' S, 171° 28.94' W) is located just north of the present-day Subtropical Convergence, below the Pacific Deep Western Boundary Current, within the depth range of Antarctic Bottom Water. The recovered sedimentary sequence from ODP Site 1123 contains a record of Southern Ocea, hydrographic variations over several glacial-interglacial cycles: We present a 440,000 year record of surface and deepwater hydrographic variability based measurements of Mg/Ca and stable isotope ratios in planktic and benthic foraminifera. Planktic Mg/Ca-based temperatures indicate a maximum glacial-interglacial temp. change of 7.5°C. The planktic record shares many features in common with a record from a shallower site located just south of the Subtropical Convergence excepth that temp. changes across Termin. 1 and 2 aren't seen at Site 1123. Possible reasons for this include hydrographic differences between sites, foraminiferal habitat, and dissolution artifacts. Glacial-interglacial benthic Mg/Ca differences are variable through the record, with coldest temperatures inferred for Stages 8 and 10. Deep-water temps. based on Mg/Ca indicate LGM values of -0.8°C (calibration of Martin et al. 2002) and a seawater d18O of 1.0 per mil (VSMOW), similar to the porewater-based reconstruction at this site (Adkins et al. 2002).
PP13B-07 INVITED 15:10h
Reanalysis and Extension of the Barbados Sea Level Record
The development and application of advanced multi-collector inductively coupled magnetic sector mass spectrometry and improvements in accelerator mass spectrometers warrant a reanalysis of the Barbados sea level curve (Fairbanks, 1989) at higher analytical precision and in more detail. Several hundred new 230Th/234U and radiocarbon dates and the addition of more Barbados offshore cores results in an extended and more detailed sea level record and oxygen isotope record for this important location. Among our new observations, the more detailed Barbados sea level record now resolves a Younger Dryas still stand and curiously indicates that the LGM may have occurred as early as 26,000 years ago. The Isotope Stage 3 interstadial ended with sea level near 87.5 meters below present at 29,500 years ago before dropping to full glacial levels. We incorporate these new sea level data into a refined model of continental deglaciation and an accurate methodology for prediction of the changes of sea-level that accompany changes of land ice volume. Our results not only provide unambiguous evidence that the post LGM rise of eustatic sea-level was very close to the widely supported estimate of 120 m, thus ruling out the occurrence of any significant meltwater pulse at 19 ka, but also that the system remained near the fully glaciated state for almost 8,000 years.
PP13B-08 15:25h
Dansgaard-Oeschger Cycles in Marine Isotope Stages 3 and 6
The oxygen isotope record of Marine Isotope Stage (MIS) 3 planktonic foraminifera in deep-sea sediment cores from the Iberian Margin is readily correlated to the oxygen isotope record in Greenland ice (Johnsen et al., J.Quat. Sci. 16, 299, 2001). Shackleton et al. (Paleoceanography 15, 565, 2000) analysed core MD95-2042 and placed the data on a Greenland ice age scale; they then showed that the benthic record closely resembles the D/H record of theAntarctic Vostok ice core (Petit et al., Nature 399, 429, 2000) when this is put on a consistent age scale using methane in the air bubbles to synchronise the records (Blunier et al., Nature 394, 739, 1998). We have analysed core MD01-2444 from 2656m water depth and have replicated this relationship; indeed with a closer sampling interval (3-cm) the resemblance between the benthic oxygen isotope (Globobulimina affinis) and the Antarctic D/H records is even more striking. The estimated standard error for G. affinis analysed in groups of about 12 is ñ0.06%. Analyses for Hoeglundina elegans, Uvigerina peregrina and Cibicidoides wuellerstorfi (groups of 1 - 3) are slightly more scattered (about ñ0.13%). Globobulimina affinis lives in the sediment where dissolved oxygen approaches zero, so that its carbon isotopic composition provides information on the water mass at its source, rather than providing information on the deepwater mass at the core site. The G. affinis carbon isotope record in MD01-2444 parallels the G. affinis oxygen isotope record, suggesting that this oxygen isotope record may be dominantly a water mass record and not primarily an ice-volume record. Core MD01-2444 extends to late MIS7 at about 190 ka. The benthic oxygen isotope record of MIS6 (sampled at 3-cm intervals, equivalent to about 300 years) contains a pattern of stadial-interstadial variability that can be convincingly correlated to the D/H record in the Vostok ice core. The triplet of interstadials older than 165 ka appear to have the same planktonic-benthic phase relationship as is observed in MIS3, suggesting that the "bipolar seesaw" was operating during that interval as during MIS3.