PP31A-0887 0800h
Planktonic Foraminifera Study at Site ODP 999A (Caribbean Sea): Insights into Oceanic Exchange and Paleocirculation During the last 450 Kyrs.
A recent micro-palaeontological study of a core from the South Atlantic (Peeters et al., Nature. 430, 661, 2004) stresses the importance of the Agulhas leakage intensity as a key control on North Atlantic Thermohaline operation. Planktonic foraminifera counts were conducted on the Colombian Basin core ODP 999A ($12\deg$N, $74\deg$W, 2878 m); results are presented on a SPECMAP-based age model (Schmidt et al., in prep 2004). At present both climate and surface circulation are driven by the motion of the Intertropical Convergence Zone (ITCZ), which governs dry/wet seasons and upwelling intensity. Caribbean surface water originates from both the south Atlantic, which re-circulates Indian Ocean water, and the north Atlantic subtropical gyre. Similarly the deep-water is a mixture of AAIW and UNADW, originating in the southern and northern hemispheres respectively. A CaCO$_{3}$ preservation index, which is deduced from planktonic fragments and pteropod remains, is used to trace changes in the intermediate water mixture. This study confirms that the sediments are better preserved during glacial times. Furthermore, it shows that the sediments of MIS11 are intensely dissolved, suggesting a maximal contribution of AAIW. Over the last 450 kyr, we observe a decreasing contribution of the AAIW during both interglacial and glacial times. In addition, the Holocene appears unusual, with a low AAIW to UNADW ratio in comparison to previous warm intervals. An examination of the percentages of the surface Indo-Pacific species {\it G. menardii} seems to confirm part this result, identifying MIS11 as a time of maximal Indian-Atlantic exchange via the Agulhas valve. It also confirms that during MIS2, 3 and 4, Caribbean surface waters were unusually unfavourable for this species and/or that the Agulhas connection was absent. The % of {\it G. Hexagona}, another Indo-Pacific species, which lives in deeper thermocline waters, peaks during MIS11, MIS9e, MIS7c and MIS5e but is absent during the Holocene and MIS7e. This observation suggests a slightly different scenario for the intermediate water flowing through the valve. This confirms the peculiarity of the Holocene as was found on the basis of benthic \delta$^{13}$C measurements in the North Atlantic (Raymo et al., Paleoceanography. 19, PA2008, 2004). Finally, on the basis of the species mentioned above in the Caribbean Sea we cannot find evidence of enhanced exchange between the Indian and the Atlantic at the end of the past five glacial as found in the Agulhas area. This disagreement and the factors controlling the late reappearance of these species in the Caribbean Sea need further investigation.
PP31A-0888 0800h
A Model Perspective on Factors Influencing the Stability Characteristics of the North Atlantic Meridional Overturning Circulation During the Last Glacial Cycle
Much of the variance in Holocene and Late Pleistocene climate is attributed to changes in the stability characteristics of the North Atlantic meridional overturning circulation (MOC). Proxy records indicate that global climate was relatively stable throughout the Holocene, moderately stable during the coldest periods of the last glaciation (Stages 2 and 4), and least stable throughout Stage 3. An examination is made of how changes during the last 120 kyr in glacial-cycle forcings (waxing and waning of continental ice sheets, variations in atmospheric carbon dioxide concentration, and changes in Earth's orbital geometry) affected the stability thresholds of the MOC to freshwater forcing. These inferences will be drawn from the results of runs of the seasonally resolving Bern climate model (Schmittner and Stocker, 1999, J Clim), which is composed of a zonally averaged ocean model (Wright and Stocker, 1992, J Clim) coupled to an energy-balance model of the atmosphere that parameterizes the meridional transport of heat and moisture.
PP31A-0889 0800h
Entrainment of Meltwaters in Hyperpycnal Flows During Deglaciation Superfloods in the Gulf of Mexico: Implications for Freshwater Outflows in the Climate System
Climate models have indicated that exceptionally high freshwater discharge outbursts into the ocean may offset the ocean thermohaline circulation (THC) and the North Atlantic Deep Water (NADW) production causing sudden climate instabilities. Verification of the numerical model predictions requires the establishment of phase and amplitude relation between the freshwater outflow events into the ocean and the warming/cooling of the atmosphere during the deglaciation. A case in point is provided by the ongoing controversy regarding the question of coupling between the onset of the global meltwater pulse mwp-1A with either the rapid B\/{o}lling warming event at 14.7 cal ka or with the Older Dryas cold event at 14 cal ka. Were the B\/{o}lling warm event and the global meltwater pulse mwp-1A synchronous, then the implications are that (i) the rapid melting of the ice sheets was in response to the dramatic warming event; (ii) mwp-1A corresponds with an intensification, rather than a weakening, of the THC and a regular NADW production, and (iv) the absence of a coupling between the mwp-1A and a slowdown of the THC offers additional evidence of an Antarctic source for the mwp-1A. In contrast, a causative coupling between mwp-1A and the Older Dryas cold event implies disturbances in the THC and NADW formation in response to the meltwater input from a Laurentide Ice Sheet (LIS) source and massive climate reversals. The validity of the proposed arguments and correlation can be readily tested by comparing the Greenland ice record with the contemporaneous meltwater fluxes in the Gulf of Mexico derived from the receding LIS. It has been argued, however, that the isotopic evidence of meltwater flows into the GOM are inconsistent with the volumes of freshwater required to account for the mwp-1A event. Moreover, the size of the mwp-1A would have required melting of the entire southern section of LIS, a scenario that is unsupported by geological evidence. Here I show that previous reliance on oxygen isotopes of planktonic foraminifera as sole archives of the meltwater events in GOM led to a gross underestimation of the volume of meltwater discharges, and to displaced timing. Oxygen isotopes of depth-stratified foraminifera exhibit unusual inversions at the onset of a succession of meltwater discharges into the Gulf of Mexico from the receding LIS. These isotope inversions document entrainment of $^{18}$O-depleted meltwaters in dense flows reaching the seafloor and rising to the surface as buoyant plumes. The exceptional meltwater fluxes of 0.28 to 0.33 Sv, coeval with a prominent erosion intensity peak in continental cryohydrology, are likely the imprints of meltwaters stored in proglacial lake basins along the ice margins and catastrophically released at 14.23-13.87 cal ka in a rapid succession of dam failures. Correspondence between the torrential floods and the Older Dryas abrupt cold reversal in the Greenland ice record supports climate models advocating a coupling between abrupt freshwater flooding of the ocean and climate instabilities.
PP31A-0890 0800h
Holocene Sea-Level Rise in New Jersey and W. North Atlantic Reefs: Untangling Natural and Anthropogenic Effects
The history of pre-anthropogenic Holocene global sea level has not been well constrained. We provide Holocene sea-level estimates for five new boreholes on the New Jersey (NJ) coast (Rainbow Island, Great Bay I, Great Bay II, Cape May, and Island Beach). We analyzed facies, radiocarbon dated marsh deposits, and derived a sea-level record by compiling new and previously published NJ data. Our sea-level record shows a constant rise of $\sim$2 mm/yr from $\sim$7000 yrBP to present. This contrasts sharply with previous NJ estimates that suggested a slowing in rise since 2000 yrBP. Comparison with other NJ locations suggests surprising uniformity in the rate of rise amongst sites as far flung as Cape May and Cheesequake (200 km apart), suggesting a far-field response to the Laurentide ice sheet. The "Barbados/western North Atlantic reefs sea-level" record shows a major decrease in the rate of rise from 12 mm/yr to $\sim$2 mm/yr between 7000 and 8000 yrBP. Data from NJ and western North Atlantic reefs indicate a constant rate of rise of $\sim$2 mm/yr since $\sim$7000 yrBP. This suggests a background, pre-anthropogenic sea-level rise of 2 mm/yr for the entire east coast of the U.S. This background includes both the global (water volume) rise and far-field geoidal subsidence due to removal of the Laurentide ice sheet and water loading (estimated as 1 mm/yr in the modern). Applying the modern subsidence rate to the Holocene suggests a eustatic rise of $\sim$1 mm/yr since 7000 yrBP. Based on tide gauge data, regional sea-level rise averaged 3 mm/yr from 1900-1995, with higher rates (4 mm/yr) locally due to compaction and groundwater withdrawal. We conclude that the anthropogenically induced rise in sea level was $\sim$1 mm/yr from 1900-1995.
PP31A-0891 0800h
Oceanic Response to Idealized Net Atmospheric Freshwater in the Pacific at the Decadal Timescale
In the study of decadal variations of the Pacific Ocean circulations and temperature, the role of anomalous net atmospheric freshwater (NAFW; evaporation minus precipitation) has received scant attention even though ocean salinity anomalies are long lived and can be expected to have more variance at low frequencies than at high frequencies. To explore the magnitude of salinity and temperature anomalies and their generation processes, we studied the response of the Pacific Ocean to idealized NAFW anomalies in the tropics and subtropics, using an ocean general circulation model developed in the Massachusetts Institute of Technology. Simulations showed that salinity anomalies generated by the anomalous NAFW were spread throughout the Pacific basin by mean flow advection. This redistribution of salinity anomalies caused adjustments of basin-scale ocean currents, which further resulted in basin-scale temperature anomalies due to changes in heat advection caused by anomalous currents. In our study, the response of the Pacific Ocean to magnitudes and locations of anomalous NAFW was linear. When forced with a positive NAFW anomaly (anomalous evaporation) in the subtropical North (South) Pacific, a cooling appeared in the western North (South) Pacific, which extended to the tropical and South (North) Pacific; and a warming emerged in the eastern North (South) Pacific. When forced with a negative NAFW anomaly (anomalous precipitation) in the tropical Pacific, a warming occurred in the tropical Pacific and western North and South Pacific; and a cooling occurred in the eastern North Pacific near 30\deg N and the South Pacific near 30\deg S. The warming (cooling) in the tropical Pacific was associated with the weakening (strengthening) of the South Equatorial Current. The warming (cooling) in the east and cooling (warming) in the west in the subtropical North and South Pacific were associated with a spin-down (spin-up) of the subtropical gyres. The temperature anomalies propagated from the tropical Pacific to the subtropical North and South Pacific via equatorial divergent Ekman flows and poleward western boundary currents; and they propagated from the subtropical North and South Pacific to the western tropical Pacific via equatorward coastal Kelvin waves and to the eastern tropical Pacific via eastward equatorial Kelvin waves. The timescale of temperature response was typically much longer than that of salinity response due to slow adjustment times of ocean circulations. These results imply that the slow response of ocean temperature due to anomalous NAFW in the tropics and subtropics may play an important role in the Pacific decadal variability.
PP31A-0892 0800h
Coral Evidence for Abrupt Changes in Ocean-Atmosphere Dynamics in the SW Pacific since 1565 AD
A coral-based multi-tracer approach can give an overview of the whole tropical ocean-atmosphere system. Key indicators are sea surface temperature (SST), which sets climate boundary conditions, sea surface salinity (SSS), which provides a measure of energy transfer through the evaporation-precipitation balance, and river runoff, which can establish the strength and variability of precipitation. We present palaeoenvironmental records from eight massive { \it Porites} coral colonies, spanning 120 to 420 years of continuous growth, collected from the central Great Barrier Reef, Australia. Stable isotopes (\delta$^{18}$O and \delta$^{13}$C), Sr/Ca, U/Ca, and Ba/Ca ratios were measured in 5-year increments and a record of annual UV luminescence was developed. By replicating the measurements between colonies we demonstrate how faithfully corals record changes in their environment over decadal-to-centennial timescales, constructing composite records in a manner analogous to dendroclimatology and confidence intervals for each proxy. The competing environmental influences affecting a number of tracers can be distinguished by comparison between the SST-tracers (Sr/Ca, U/Ca, \delta$^{18}$O), the freshwater flux tracers (\Delta\delta$^{18}$O, Ba/Ca and luminescence) and tracers of water mass characteristics (\delta$^{18}$O, \delta$^{13}$C, and \Delta$^{14}$C). The coral palaeothermometers Sr/Ca and U/Ca ratios, measured in tandem with \delta$^{18}$O, allow the separation of SST changes from changes in seawater \delta$^{18}$O, thereby resolving SSS. The composite Sr/Ca and U/Ca are in excellent agreement back to 1565, and capture the 20$^{th}$ century warming trend, up to the 1980s when the cores were collected. The most remarkable feature of the 420-year record is that SSTs were consistently as warm as the second half of the 20$^{th}$ century from the early 18$^{th}$ and through most of the 19$^{th}$ centuries. Changes in the evaporation-precipitation balance dominate the \delta$^{18}$O record. A striking 0.2\permil\ shift from the 1850s to modern values in the 1870s, indicates an abrupt freshening, which is common to coral \delta$^{18}$O records throughout the SW Pacific. Interdecadal variability of \Delta\delta$^{18}$O, Ba/Ca and luminescence correspond strongly in response to the variable strength of the Australian summer monsoon circulation. We investigate the scenario that a strengthened latitudinal temperature gradient may have prevailed prior to the 1870s, with evidence for an intensified Hadley circulation impacting evaporation rates and the surface-ocean circulation in the SW Pacific, in addition to altered precipitation patterns. The late-19$^{th}$ century demise of the extratropical 'Little Ice Age' in the Northern Hemisphere coincides with cooling and abrupt freshening of the SW Pacific.
PP31A-0893 0800h
Climate change at 8200 yrs BP: A synthesis of paleoclimate proxy records
Many paleoclimate proxy records show an abrupt climate change at 8200 yrs BP and there is strong evidence linking this event to the rapid drainage of Lake Agassiz through the Hudson Strait into the North Atlantic. This event is particularly important because it occurred during a time when temperatures were similar to or slightly higher than today, providing an example of how a warm, interglacial climate could be altered by an anomalous freshwater flux to the North Atlantic. Despite the importance of this event, no comprehensive global map of climate anomalies exists for this time. In this research, we have compiled nearly 100 high-resolution, well-dated proxy records that span the early Holocene. We have used statistical tests to objectively identify anomalies associated with the abrupt event at 8200 yrs BP. Regional patterns and magnitudes of climate change will be described. Because so many high-resolution records are available from this time period and the cause of the event is known relatively well, it provides an excellent opportunity for a comparison to coupled climate model simulations. We will discuss how our results can be compared with model output in order to understand the response of the climate system to freshwater anomalies and to test model sensitivity.
PP31A-0894 0800h
Cosmogenic Dating Evidence for the Routing of Lake Agassiz Overflow During the Younger Dryas
Two sources of fresh water releases in the North Atlantic are possible causes for abrupt climate changes that punctuated the last glaciation. One possible cause is the release of iceberg armadas from a surging Laurentide Ice Sheet (LIS) and the other is the release of meltwater from proglacial Lake Agassiz. This additional fresh water in the North Atlantic would have reduced the surface water density, which would have shut down the thermohaline circulation, resulting in large scale climate changes. The release of iceberg armadas has been well studied in deep sea sediments and has been identified by ice-rafted detritus and called Heinrich events. The sudden release of meltwater could have been the trigger for three cold periods recorded in the $\delta$$^{18}$O records of Greenland ice cores: the Younger Dryas (YD), the Preboreal Oscillation (PBO) and the 8.2 cal ka cold event. The best hypothesis so far to explain these three cold snaps is that outbursts from Lake Agassiz happened via different pathways that abruptly opened during the retreat of the LIS. The final two Greenland isotopic events (PBO and the 8.2 cal ka event) have been correlated to Lake Agassiz bursts, and these occurred through northern outlets: the Athabasca-Mackenzie Valley and Hudson Strait respectively. The isotopic signal associated with the YD event may be related to one of two possible routings: (1) the NW outlet via the Athabasca-Mackenzie Valley and (2) the E outlets into the Great Lakes. Although there is evidence for one flood to the NW around 10 ka, there may have been an earlier event, but that record has not yet been identified. We investigated the second possible routing through Thunder Bay, Ontario, to Lake Superior. Thunder Bay is a strategic area because paleotopographic reconstructions suggest that if ice was not present, meltwater would have overflowed directly into the Lake Superior Basin and from there into the St Lawrence Valley to the North Atlantic Ocean. However, geomorphic evidence for a flood through the Thunder Bay area is not convincing, and an alternative routing may be required. One possibility is that the NW outlet carried overflow during the YD. Another possibility is that flow was eastward, subglacially via the Lake Nipigon Basin to the N, which was buried by the LIS at this time; excellent geomorphic and sediment evidences exist in that area for strong meltwater discharge. To verify this hypothesis we used the surface exposure dating technique and specifically the cosmonuclide $^{10}$Be. We sampled granitic boulders on the Dog Lake moraine (associated with the YD cold event) and granitic boulders from the drainage channels west of Nipigon Lake (presumed to be post-YD in age). Here we present the first results of our investigation.
PP31A-0895 0800h
Fluctuations of Sulfur and Strontium Isotope Ratios of Fish Specimens in Lake Biwa, Japan, During Recent 40 Years
Sulfur and strontium isotopic fluctuations of lake water were reconstructed using fish specimens collected and preserved over 40 years. In this study, the first year and/or the second year fishes of Chaenogobius isaza from the northern part of Lake Biwa, Japan, were used. Each C. isaza was dissected, and the muscle tissue and the bone were subjected to sulfur and strontium isotope analyses, respectively. Delta 34S values of C. isaza showed monotonous decrease from +2.0 per mil of 1963 and -2.5 per mil of 2001. Considering sulfur isotopic fractionation between lake water sulfate to muscle tissue of C. isaza, it is suggested that delta 34S of lake water sulfate decreased from +5.0 per mil to +0.5 per mil during recent 40 years. A relative increase of sulfur contribution originated by human activities (~ - 4.0 per mil; Nakai et al., 1991) might be one of the main factors of the sulfur isotopic decrease, supposing from a rise in population and industrializing of the catchment area. Strontium isotopic ratios (87Sr/86Sr) of C. isaza slightly decreased from 0.7126 to 0.7124 during recent 40 years with distinct negative spikes (0.71195 to 0.71175) at 1980, 1983 and 1986. No specific change was shown in the trends of sulfur (this study) and nitrogen (Ogawa et al., 2001) isotope values at each of three years. Therefore, the drastic changes related to strontium circulation in Lake Biwa may have no influence on sulfur and nitrogen circulations.
PP31A-0896 0800h
LGM Summer Climate on the Southern Margin of the Laurentide Ice Sheet: Wet or Dry?
High-resolution regional climate simulations are conducted using Polar MM5 with a domain centered over North America to explore the summer climate of the Last Glacial Maximum (LGM) roughly 21,000 calendar years ago. The simulations use appropriate LGM boundary conditions: orbital forcing, continental ice sheets, reduced trace gas concentrations, lowered sea level, and paleovegetation. A tailored NCAR Community Climate Model version 3 (CCM3) simulation of the LGM climate provides the initial and lateral boundary conditions for Polar MM5 as well as prescribed sea surface temperatures and sea ice extent. The LGM summer climate features a pronounced low level thermal gradient along the southern margin of the Laurentide Ice Sheet, which anchors the mid-tropospheric jet stream and facilitates the development of synoptic cyclones that track over the ice sheet; some of these produce heavy {\it rain} along the southern margin. Precipitation on the southern margin is orographically enhanced as moist southerly low-level flow from the Gulf of Mexico (resembling a contemporary Great Plains low level jet configuration) is drawn up the ice sheet slope in advance of the cyclones. Composites of wet and dry periods on the Laurentide Ice Sheet southern margin illustrate two distinctly different atmospheric flow regimes. Given the episodic nature of the rain events and the interannual variability simulated in Polar MM5, the modeled wet conditions in the Great Plains during the LGM summer are reconcilable with general aridity across the region inferred from geological proxy evidence.
PP31A-0897 0800h
Asian monsoon change in the Mid-Holocene: Influence of ocean coupling for GCM simulations
In order to clarify the role of ocean for changes on Asian monsoon in response to the insolation forcing in 6000 years before present (mid-Holocene, 6ka), simulations using (A) an CCSR/NIES/FRCGC ocean-atmosphere coupled GCM, (B) an AGCM extracted from the coupled GCM coupled with a mixed ocean model, and (C) the AGCM with prescribed SST are compared. Paleoclimate evidences of lake status records show conditions wetter than present in central Asia, from India to northern China and Mongolia and more arid condition along the coast of China (Yu and Harrison 1996). However precipitation in Asia in all three simulations becomes more vigorous with 6ka insolation forcing, ocean feedbacks in (A) and (B) simulations suppress the enhancement of precipitation. These suppressions are not consistent to cooler SST along the coast of India in (A) and (B) simulations suggesting more vigorous monsoon. This suggests that more than simple understanding of monsoon mechanisms plays a role in the change in Asian monsoon. Effect of developed anticyclone by cooler SST in Indian Ocean on relationship between African and Asian monsoons may have to be considered. The wetter conditions from India to northern China and the more arid condition along the coast of China are represented by the three model configurations. Precipitation reduction in the coastal China is expressed better by (A) and (B) than by (C). This may be resulting from weaker development of the Pacific summer anticyclone with ocean feedbacks, attributed to the warmer SST in the Pacific. It is found that the introduction of ocean dynamics strengthens the tendency of changes of Asian monsoon in 6ka but it is less important than the introduction of heat exchange on the ocean surface.
PP31A-0898 0800h
Interhemispheric precipitation seesaw: mirror images of oxygen isotopic records from caves in S. Brazil and E. China
We have obtained a high resolution oxygen isotopic record of cave calcite from Caverna Botuver\'{a} ($27\deg$13\'S, $49\deg$09\'W), South Brazil. The chronology was determined by 16 high-resolution U/Th ages. We interpreted our record, spanning from 40 ka through 15 ka, in terms of meteoric precipitation changes in this low-latitude location. The oxygen isotopic profile broadly follows local insolation changes and shows millennial-scale variation with amplitudes as large as 3 per mil, similar to (but not in phase with) some Northern Hemispheric observations (e.g., Wang et al., 2001; Bar-Matthews et al., 2003; Yuan et al., 2004). Using independent age scales, we compare our record to contemporary records from Hulu Cave, East China (Wang et al., 2001), and the GISP2 Ice Core, Greenland (Grootes and Stuiver, 1997). Minima in $\delta$$^{18}$O (wet periods) at our site are synchronous with maxima in $\delta$$^{18}$O (dry periods) in East China and vice versa. Our wet phases also correspond to minima in $\delta$$^{18}$O (cold periods) in Greenland and vice versa. This anti-phased precipitation relationship between Botuvera and Hulu, two low-latitude locations, may be related to asymmetry in Hadley cell circulation in two hemispheres, associated not with seasonal changes as observed today, but with millennial-scale climate shifts. If the Botuvera and Hulu records are broadly representative of low-latitude rainfall in their respective hemispheres, and if rainfall is directly related to absolute humidity, then the atmospheric see-saw pattern that we observe could have important consequences because of water vapor's greenhouse properties. Reference: Bar-Matthews M. et al., 2003. Geochimica et Cosmochimica Acta 67, 3181-3199. Wang Y.J. et al., 2001. Science 294, 2345-2348. Yuan D.X. et al., 2004. Science 304, 575-578. Grootes P.M. and Stuiver M, 1997. Journal of Geophysical Research 102, 26455-26470.
PP31A-0899 0800h
Meltwater Outflow and Laminated Plume Deposition in the Southern Norwegian Sea Preceding Heinrich Event 1.
Rapidly deposited laminated sediments dated between 18 and 22 ka BP have been observed at several locations in the Northern North Sea and the Southern Norwegian Sea. Sedimentation rates range from 20 cm/ka up to 2000 cm/ka, compared to a normal background sedimentation rate in the area of 5 to 10 cm/ka. Detailed sedimentological work revealed that the laminations are possibly marine annual varves, indicated by cyclic changes in the number of planktonic foraminifera and lithology. The deposits have been identified as the product of a large suspension plume and are associated with lighter d18O values compared to the region, indicating meltwater inflow. The laminated deposits identified along the upper continental margin off southern Norway occurred mainly during the early deglaciation at 18.6 ka BP and are believed to originate from the presence and fast retreat of the Norwegian Channel Ice Stream. Simultaneously, increased fluxes were observed on the Barra Fan, the North Sea Fan, the Lofoten contourite and the Svalbard margin, preceding increased fluxes on the Greenland margin and the Labrador Sea, associated with Heinrich event 1. Between 21 and 18 ka BP planktonic foraminifera indicate warm water inflow to the area. This, possibly together with an increase in sea level, probably triggered the melting of the low latitude parts of the Fennoscandian ice sheet, leading to a massive output of material to the continental slope. A similar response of the Laurentide ice sheet occurred 500 to 1000 yrs later during Heinrich event 1.
PP31A-0900 0800h
Catastrophic Drainage of Glacial Lake Iroquois Down the Hudson River Valley: A Potential Trigger for the Intra-Allerod Cold Period
Meltwater discharge to the North Atlantic during deglaciation likely influenced global oceanic circulation and climate. However, directly linking meltwater discharge events with individual climate oscillations has been difficult because of challenges in determining the location, timing and amount of meltwater discharge. New evidence from the Hudson River Valley (HRV) and the Northeastern US continental margin constrain the timing of the catastrophic draining of Glacial Lake Iroquois, that breached the moraine dam at the Narrows in New York City, eroded the HRV, and deposited large sediment lobes on the exposed New York and New Jersey continental shelf, to about 13,350 years before present (BP). Including the contribution of Glacial Lake Vermont and Glacial Lake Albany, which also drained in this event, the meltwater release was likely about 3.5x1012 m3. We regard this as a minimum estimate, as additional meltwater from the west most likely contributed to Glacial Lake Iroquois at this time and further meltwater may have been provided by subglacial reservoirs. The timing of this meltwater pulse is coincident with the onset of the Intra-Allerod Cold Period (IACP) recorded in the GISP II oxygen isotopes and the grey-scale record from the Cariaco Basin. The IACP probably correlates to short-lived cold reversals in Central North America, Atlantic Canada (Killarney Oscillation), and Central Europe (Gerzensee Oscillation). Excess 14C in Cariaco Basin sediments likely indicates a slowdown in thermohaline circulation and heat transport to the North Atlantic that initiated about 13,350 years ago. Thus the meltwater discharge out the HRV may have played an important role in slowing thermohaline circulation triggering the IACP.
PP31A-0901 0800h
Isotopic modelling of Heinrich Events : a method to constrain duration & freshwater release
During the last glacial period, abrupt climate events due to ice-sheet instabilities, called Heinrich events, are recorded in sediment cores retrieved in the north Atlantic. Model based studies have described likely physical mechanisms explaining these events, involving large-scale instabilities of the Laurentide (and fennoscandian) ice-sheets. But no general agreement exists on the characteristics of an Heinrich Event, in terms of duration and freshwater release. To further investigate this question, we have included oxygen-18 in a climate model of intermediate complexity, CLIMBER-2, to simulate Heinrich Event 4 (40kyr BP), enabling us a direct comparison with data. As the model used is a model of intermediate complexity, and therefore is fast to run, we compare a large ensemble of scenarii to oxygen-18 data measured in foraminiferal calcite of a comprehensive set of sediment cores to assess the duration, the rate of discharge and the contribution to sea-level rise of Heinrich Event 4. A statistical analysis is then used to minimize differences between model and data. Our approach enables us to diagnose these values with greatly reduced uncertainties. We show that Heinrich Event 4 lasted 250ñ150 years and released an amount of ice corresponding to about 2ñ1 meters sea level equivalent. Our results are in good agreement with the sum of arising sedimentological evidences that Heinrich events were shorter than a millennium. We demonstrated that during HE4 the only consistency between model results and data involves a complete, or nearly complete, shutdown of the THC. These results emphasizes the need for isotopic tracer simulation to analyze thoroughly paleo-records, allowing us to gain confidence in the model behavior during rapid climate changes. To pursue even further the comparison with data, we used the carbon cycle module of CLIMBER analyze the signal in terms of carbon 13. An interpretation of this comparison will be suggested. The implications of these result on the comprehension of the mechanisms of glacial climate variability will be investigated in the frame of succession of events.
PP31A-0902 0800h
ENSO Response to a Change in Thermohaline Circulation Caused by Surface Fresh Water Forcing of the North Atlantic Basin
Previous studies of the coupled atmosphere-ocean response of the climate system to collapse of the North Atlantic thermohaline circulation (THC), and subsequent cooling of the northern hemisphere, have alluded to, though not established, the existence of a possible connection to Tropical Pacific climate variability. A series of fresh water forcing (FWF) anomalies of 0.1 and 1Sv, each of 100 years duration, have been applied to the surface of the North Atlantic between 50 and 70 degrees north latitude in an equilibrated NCAR CCSM1.4 modern pre-industrial simulation which fully resolves the ENSO process. The climate is allowed to readjust after this forcing is removed. The changes in both the atmosphere and oceans that result in a strengthened ENSO during the transient period of THC collapse caused by the freshwater forcing will be described. Of particular interest is the existence of an atmospheric teleconnection between the North Atlantic and the Tropical Pacific, whereby a PNA type pattern induced by a quasi-stationary train of planetary Rossby waves modifies the surface wind stresses in the tropical Pacific. This sets up in concert with a significant increase in the strength of the Hadley circulation of the tropical atmosphere. The increase in the strength of both the ENSO process and the Hadley circulation are essentially identical to those previously found to be characteristic of Last Glacial Maximum conditions (Peltier and Solheim, QSR 23, 335-357, 2004). A number of results will be presented that attempt to resolve the issue as to whether the tropical system is solely responding to a Northern Hemisphere cooling or whether a feedback mechanism might exist that modifies both the tropical Pacific and Northern Hemisphere climate regimes.
PP31A-0903 0800h
The Impact of the Opening of the Bering Strait on the Atlantic Fresh Water Budget and Climate
It is often suggested that reorganizations of the overturning circulation is the cause of abrupt climate changes. Such a re-organization probably happened in the early Holocene when mean Greenland temperatures suddenly increased. We hypothesize that the abruptness of the temperature change in the early Holocene, and the increase in mean temperature is due to the opening of the Bering Strait (BS), which plays a pivotal role in the re-organization. With a present day depth of merely 50 meters, the strait opens and closes with deglaciation/glaciation (which cause changes of as much as 140 meters in the sea level). We use laboratory experiments and climate models to examine this idea. Assuming that the oceans are in a state of the Last Glacial Maximum (LGM) and that the BS Strait is initially closed, deglaciation starts melting continental ice sheets. This melt water finds its way to the North Atlantic creating a low salinity surface layer, which inhibits convection. Since the BS is closed, this anomaly cannot be flushed out of the system quickly, and convection would recover only after a few hundred or thousand years (via processes such as diffusion). However, as deglaciation continues, global sea level rises and eventually opens the BS. The pressure head created on the Atlantic side by the Southern Ocean winds attempts to flush this low salinity layer out into the Pacific. As it does this, icebergs in the vicinity of the BS move towards the strait creating a temporary dam, because the water level is too shallow to allow them to pass through. Global temperatures continue to rise and eventually the pressure head on the Arctic side is able to break this temporary dam, flushing the low salinity layer into the Pacific and kick starting convection in the North Atlantic. A simple laboratory box model is used to investigate this temporary dam and it's subsequent release. This experiment was done for five different strait widths, using polyethylene balls as icebergs. For each width we see the temporary damming effect, as mentioned earlier, releasing abruptly once the pressure head on the Arctic side is large enough to force both balls and water through. Increasing the width of the strait decreases the number of temporary dam occurrences as well as the length of each occurrence. The transition from LGM to Holocene conditions is examined using the University of Victoria Earth System Climate Model (Uvic ECSM). Specific focus is placed on recreating (1) a reversal of flow through the Bering Strait, (2) an increase in mean atmospheric and oceanic temperature in the North Atlantic and (3) a resumption of convection in the North Atlantic.
PP31A-0904 0800h
Simulated Global Response to a Substantial Weakening of the Atlantic Thermohaline Circulation
In this study we use a newly developed fully coupled global ocean-atmosphere general circulation model to investigate the global scale response of the climate system to a sustained addition of fresh water to the model's North Atlantic, such as may have occurred during glacial periods. In response to this forcing, the model's thermohaline circulation weakens substantially, thereby cooling the North Atlantic and warming the South Atlantic. The associated global response involves a southward shift of the intertropical convergence zone (ITCZ) over both the Atlantic and Pacific sectors, an El Nino like condition and weakened Walker circulation in the southern tropical Pacific, a La Nina like condition and strengthened Walker circulation in the northern tropical Pacific, a dipole subsurface temperature response in the western tropical Pacific, and weakened Indian and east Asian summer monsoons. The substantial weakening of the thermohaline circulation leads to a more symmetric annual mean zonally averaged ITCZ and zonally integrated Hadley circulation about the Equator. These responses are consistent with the global-scale synchronization of millennial-scale abrupt climate change as indicated by paleoclimate records.
PP31A-0905 0800h
Speed, extent, and sequence of abrupt climate change 8200 years ago
The atmospheric methane concentration can be viewed as a qualitative indicator of globally integrated terrestrial hydrological conditions owing to the dominant methane source from wetland areas. Coupled with nitrogen isotope analyses in the trapped air in an ice core (which gives a local temperature signal in Greenland), the methane record provides critical information on the speed, extent, and sequence of past climate change. We conducted high resolution (5-year interval for the cooling period in gas age) methane analyses on the 8.2k event in the GISP2 ice core. The record shows that atmospheric methane decreased from about 650 ppb to about 570 ppb synchronous (within a decade or so) with a nitrogen isotope decrease. This suggests that the widespread drying and cooling events inferred from many paleoclimatic proxies around this time were simultaneous events. The widespread distribution of methane sources at that time further supports the notion that this climate event had a footprint of at least hemispheric scale. The magnitude of the methane drop (about 80ppb) was about one third that of the Younger Dryas event. The oxygen isotope record of ice suggests that the event occurred very rapidly, taking only about 5 years. To quantitatively reconstruct the rate of the change in the atmospheric methane concentration and emissions from wetlands, we used a simple box model. Assuming a constant atmospheric OH concentration, a preliminary result shows that the rate of atmospheric methane change is -10$\pm$ 5 ppb/year, and the emission change occurred within 10$\pm$ 10 years. In addition, we will present a new temperature record from 7,600 B.P. to 3,700 B.P. derived from nitrogen and argon isotope analyses from GISP2 ice core.
PP31A-0906 0800h
Can a freshwater forcing in high latitudes induce a permanent El Ni\~{n}o in the tropics - an overlooked mechanism for a rapid climate change.
A rapid climate change in response to the current rise in the atmospheric concentration of greenhouse gases is a matter of considerable concern. Thus far attention has focused almost entirely on changes associated with the deep thermohaline circulation. Little attention has been paid to possible changes in the wind-driven circulation of the ventilated thermocline. This shallow wind-driven circulation controls a significant transport of heat from low to high latitudes and is therefore subject to the constraint of a balanced heat budget: on a certain time-scale, on the order of decades, the heat gain in the upwelling zones of low latitudes must equal the heat loss in high latitudes. Can a freshening of the surface waters in high latitudes significantly reduce this heat transport? How will that affect the tropics? To explore this matter we use a number of idealized general circulation models of varying complexity that capture the main aspects of the thermocline ventilation. Our calculations show that a freshening of the surface waters in high latitudes does reduce the poleward heat transport and deepens the tropical thermocline. For a sufficiently strong freshwater forcing, the poleward heat transport all but vanishes, and permanent El Ni\~{n}o-like conditions prevail in the tropics. Paleo data indicate that this have happened before - there was no equatorial SST gradient in the Pacific Ocean during the early Pliocene.
http://www.geology.yale.edu/~avf5