C34B-01 INVITED
A Synthesis of Late Miocene Through Pliocene Evolution of Glaciation as Inferred From Deep Sea Geochemical Records
During the Miocene through Pliocene climate evolved from an early Miocene climatic optimum (~16 Ma) followed by major expansion of the Antarctic ice sheet during the middle Miocene (~15 Ma), to an early Pliocene interval of relative global warmth (5 to 3.5 Ma) followed by the onset of wide spread Northern Hemisphere Glaciation during the late Pliocene (~3 Ma). Here I review the evolution of Miocene/Pliocene glaciation as recorded in the geochemistry of deep sea sediments. Much of what we know about past climate change comes from the oxygen isotopic composition of benthic foraminifera. Although this proxy outlines large scale changes in the degree of polar glaciation, the absolute magnitude and the relationship between ice extent and ocean temperature cannot be uniquely determined. The recent development of foraminiferal Mg/Ca ratios as a proxy for paleotemperatures provides an opportunity to improve our understanding of Miocene/Pliocene climate change on both tectonic and orbital time scales. For example, paired δ18O and Mg/Ca deep water records show that middle Miocene expansion of ice predates cooling of Southern Ocean surface waters providing evidence for the importance of heat and moisture transport in Antarctic ice growth (Shevenell and Kennett, 2007). Relatively few deep sea studies have focused on late Miocene climate, and foraminiferal δ18O records do not support major oceanographic and climatic changes. Although, the late Miocene may have been a time of global cooling, especially in the circum- Antarctic region, with the establishment of a grounded West Antarctic ice sheet. The early Pliocene, in contrast, has been the focus of much research because of the relevance to understanding intervals of sustained global climatic warmth with near modern-day tectonic configuration, warm upwelling regions, and elevated CO2 levels with respect to the pre-industrial atmosphere. The deep sea δ18O record, however, suggests that Antarctic ice sheet size remained relatively stable during this particular warm period. Early Pliocene climatic warmth ends with the beginning of wide spread Northern Hemisphere Glaciation at ~3 Ma, which is followed by a time of enhanced sensitivity of ice sheet variations to obliquity forcing. Although a detailed picture of Miocene/Pliocene evolution of the cryosphere has emerged, there are a number of open questions such as the causes of middle Miocene climatic warmth, the stability of the East Antarctic ice sheet during the early Pliocene, and the triggers for significant Northern Hemisphere Glaciation during the late Pliocene.
C34B-02
Modeling West Antarctic Ice Sheet Growth and Retreat Through the Last 5 Million Years
The West Antarctic Ice Sheet, grounded mostly below sea level and fringed by floating ice shelves, is considered to be vulnerable to future anthropogenic warming. However, projections of its future behavior are hampered by limited understanding of past variations and the main forcing mechanisms. Here a combined ice sheet-shelf model with imposed grounding-line fluxes following C. Schoof (J. Geophys. Res., 2007) is used to simulate Antarctic variations over the last 5 million years. We argue that oceanic melting below ice shelves is an important long-term forcing, controlled mainly by far-field influences that can be correlated with deep-sea- core d18O records. Model West Antarctic configurations range between full glacial extents with grounding lines near the continental shelf break, intermediate states similar to modern, and brief collapses to isolated ice caps on small West Antarctic islands. Transitions between these states can be relatively rapid, taking one to several thousand years. Several aspects of our simulation agree with a sediment record recently recovered beneath the Ross Ice Shelf by ANDRILL (MIS AND-1B core), including a long-term trend from more frequently collapsed to more glaciated states, and brief but dramatic collapses at Marine Isotope Stage 31 (~1 Ma) and other super-interglacials. Higher-resolution nested simulations over the Ross Embayment resolve Siple Coast ice streams, Transantarctic outlet glaciers, and details of shelf flow. Correlations between modeled local conditions near the AND-1B core site and the overall West Antarctic state are examined, along with implications for the AND-1B lithologic record.
C34B-03
Climatic forcing of West Antarctic Ice Sheet variability on orbital timescales
The numerical modeling effort associated with the ANDRILL program is using a hierarchy of models to explore the climatic and glacial evolution of Antarctica on orbital and longer (Cenozoic) timescales. Modeling work over the past several years has concentrated on the development of a combined ice sheet-shelf model, the adaptation of a Regional Climate Model to the Antarctic region, GCM-ice sheet coupling, and the development of asynchronous coupling techniques allowing long time-continuous climate-ice sheet simulations on orbital and longer timescales. To complement recent findings by the ANDRILL field program, simulations to date have concentrated on the Plio-Pleistocene, and in particular on Marine Isotope Stage 31 (~1.07 Ma); a super-interglacial interval when the Ross Embayment appears to have been ice-free and the West Antarctic Ice Sheet was partially or completely collapsed. Here, we show results from model simulations exploring the potential range of surface mass balance forcing and oceanic, sub-ice-shelf melt in response to orbital forcing and interglacial greenhouse gas concentrations. The climate model output is used to drive an ice sheet-shelf model and the results are compared with 5-myr time-continuous sheet-shelf simulations forced by parameterized climate and sub-ice melt derived from orbital solutions and deep-sea-core δ18O. Our climate model results reinforce the conclusions of parameterized 5-myr ice sheet-shelf simulations, and point to the overriding importance of sub-ice melt to Ross Ice Shelf and WAIS variability on these timescales. More explicit, high-resolution ocean modeling of West Antarctic embayments will be needed to directly address the potential impact of increasing ocean heat content in past and future scenarios of WAIS retreat.
C34B-04
Sequence stratigraphy of the ANDRILL Southern McMurdo Sound (SMS) project drillcore, Antarctica: an expanded, near-field record of Antarctic Early to Middle Miocene climate and relative sea-level change
Present understanding of Antarctic climate change during the Early to Middle Miocene, including definition of major cycles of glacial expansion and contraction, relies in large part on stable isotope proxy records from Ocean Drilling Program cores. Here, we present a sequence stratigraphic analysis of the Southern McMurdo Sound drillcore (AND-2A), which was acquired during the Austral Spring of 2007. This core offers a hitherto unavailable ice-proximal stratigraphic archive of the Early to Middle Miocene from a high-accommodation Antarctic continental margin setting, and provides clear evidence of repeated fluctuations in climate, ice expansion/contraction and attendant sea-level change over the period 20-14 Ma, with a more fragmentary record of the post-14 Ma period. A succession of seventy sequences is recognized, each bounded by a significant facies dislocation (sequence boundary), composed internally of deposits of glacimarine to open shallow marine environments, and each typically dominated by the transgressive systems tract. From changes in facies abundances and sequence character, a series of long-term (m.y.) changes in climate and relative sea-level is identified. The lithostratigraphy can be correlated confidently to glacial events Mi1b and Mi2, to the Miocene Climatic Optimum, and to the global eustatic sea-level curve. SMS provides a detailed, direct, ice-proximal reference point from which to evaluate stable isotope proxy records for Neogene Antarctic paleoclimate.
C34B-05
Colluvium Erosion Rates in the McMurdo Dry Valleys, Antarctica
We analyzed cosmic-ray-produced Be-10 and Al-26 in quartz sands from colluvium deposits directly under two ash deposits in the McMurdo Dry Valleys (MDV), Antarctica. In Arena Valley we sampled the Monastery Colluvium that is overlain by the Arena Valley Ash, which has an Ar-40/Ar-39 age of 4.34 Ma. In Wright Valley we sampled undifferentiated colluvium that lies directly beneath the Hart Ash, which has an Ar-40/Ar-39 age of 3.9 Ma. We collected a series of sediment samples (3 cm thick) from below these ash layers, as well as a sample of soil directly below the modern desert pavement that armors each ash deposit. Concentrations of Be-10 and Al-26 are significantly less than expected from the minimum limiting age of these in situ, air fall ashes. The ratio of Be-10/Al-26 allows us to rule out burial or ice cover to explain these low concentrations. The concentrations of Be-10and Al-26 in the soil profile do not show signs of mixing, creep, or sublimation of ground ice, and the profiles are best interpreted as indicating steady degradation of the surface. In Arena Valley, the erosion rate of the colluvium above the ash that best explains the observed isotope profiles is 0.3 m/Ma for the last two million years. In Wright Valley the erosion rate that best explains the observed isotope profiles is 4.2 m/Ma for the last two hundred thousand years. The range of erosion rates determined here can be explained in part by the difference in elevation of the two sites (~1,000 m), which in the MDV has been shown to have a strong effect on geomorphic activity. The preservation of ash deposits, at the present-day surface has been interpreted as indicating that the climate in the MDV has remained cold and dry for millions of years and that little to no landscape degradation has occurred over this time. These results are consistent with the Ar-40/Ar-39 ages of these ash deposits only if slow, but geomorphically significant degradation has occurred for the last few hundred thousand to million years.
C34B-06
Pliocene Glacial-Deglacial Cycles Deciphered From Southern Kerguelen Plateau Benthic Foraminiferal Assemblages
Kerguelen Plateau offers a unique perspective on late Neogene paleoclimate because it is located downstream from the Antarctic cryosphere at a critical location within thermohaline circulation. The Pliocene on the Kerguelen Plateau was a time of dynamic paleoclimatic change that has been documented from planktonic microfossils (Bohaty and Harwood, 1998; Whitehead and McMinn, 2002). Using benthic foraminifera as a proxy for paleoenvironmental and paleoclimatic conditions, intervals of change (glacial- deglacial cycles) were deciphered. Benthic foraminiferal data from Ocean Drilling Program (ODP) Sites 747, 748, and 751 (Mackensen, 1992) were compared with those data from ODP Site 744 in order to develop a regional Pliocene history of the Southern Kerguelen Plateau. Carbonate values fluctuate dramatically throughout the late Neogene in the region. Pliocene Dissolution Events (PDEs) are recorded at Sites 751, 748, and 744 (Ehrmann, 1991; Mackensen, 1992). During these events calcareous benthic foraminifera are dissolved, thus affecting the record of benthic assemblages. Carbonate availability may in part be controlling test abundances and species richness in biocoenoses and thanatocoenoses at the Southern Kerguelen Plateau, especially during PDEs. While benthic foraminifera have limited utility biostratigraphically during the Pliocene on the Kerguelen Plateau, they do offer great value in deciphering paleoenvironmental seafloor conditions and trophic relationships. Benthic assemblages on the Southern Kerguelen Plateau during the Pliocene are dominated by Epistominella exigua (Schroeder-Adams, 1991; Mackensen, 1992). Thus, the major factors driving benthic assemblages during the Pliocene in the region are organic carbon flux to the seafloor and carbonate availability.
C34B-07 INVITED
Late Cretaceous to Cenozoic Antarctic Ice-Sheet Evolution From Sea-Level and Deep- Sea Isotope Changes
The imperfect direct record of Antarctic glaciation has led to the delayed recognition of "initiation" of a continent-sized ice sheet: early studies interpreted initiation in the middle Miocene (ca. 15 Ma), most current studies place the first ice sheet in the earliest Oligocene (33.55 Ma), followed by dynamic ice sheets until the middle Miocene (14 Ma) and a persistent ice sheet from that time on. Though there are inherent limitations in sea-level and deep-sea isotope records, both place constraints on the size and extent of Late Cretaceous to Cenozoic Antarctic ice sheets. We present revised sea-level, benthic foraminiferal oxygen isotope, and Mg/Ca syntheses including the first backstripped estimates of the past 8 myr and use them to extract a temperature history of the deep-sea that reflects surface waters near Antarctica. Sea-level records argue that small- to medium-sized (typically 10-12 x 106 km3), ephemeral ice sheets occurred during the greenhouse world of the Late Cretaceous to middle Eocene. Deep-sea δ18O records show increases associated with many of these greenhouse sea-level falls. This is consistent with their attribution to ice-sheet growth back to 100 myr. The earliest Oligocene Oi1 event was the first continent-scale glaciation of the past 100 myr. Concordance of temperature estimates from sea-level and Mg/Ca is remarkable over the past 40 myr and support some Eocene ice. Our temperature estimates are similar to those of Hansen et al. (2008) for the period prior to 40 Ma and after 14 Ma, but differ for the time in between, reconciling better with the record from Antarctic margin drilling. New sea-level estimates from backstripping in Virginia for the past 8 myr suggest that sea level peaked in the Pliocene <20 m above present, suggesting the loss of Greenland ice sheet, West Antarctic ice sheet and some ice from the East Antarctic margin, but no collapse of the main ice sheet as originally proposed.
C34B-08 INVITED
A Turning Point in the Cenozoic Greenhouse to Icehouse Transition during the Late Middle Eocene
A fundamental shift from greenhouse to icehouse climates occurred during Eocene-Oligocene time interval, characterized by extensive global cooling and the development of large polar ice sheets. Although the long- term climate trends through this interval have been well established, very few paleoceanographic records provide the necessary resolution to document short-term climatic variability. Compilation of benthic foraminiferal and bulk δ18O records from Southern Ocean sites provides high-resolution data coverage spanning most of the Eocene. Based on these highly-coherent records, there is now sufficient resolution to make general inferences concerning both the long-term and short-term climate variability through the middle and late Eocene. The Middle Eocene Climatic Optimum (MECO) at ~40 Ma stands out as the most prominent reversal in the long-term cooling trend through the middle and late Eocene, representing the lowest δ18O values (warmest temperatures) in all intervals younger than ~48 Ma. Thus, the compiled δ18O records indicate that the warm temperatures sustained during the MECO represent a brief return to temperatures experienced during the early Eocene, and warming of a similar degree did not occur during the remainder of the Eocene, indicating that the MECO was the final Eocene 'hyperthermal' event. Post-MECO cooling was rapid and initiated a long-term cooling trend that culminated near the middle-late Eocene boundary at ~37 Ma. The highest benthic foraminiferal δ18O values of the entire Eocene characterize this interval, representing the most likely period of ice-sheet development prior to the Eocene-Oligocene boundary. This interval of cooling and/or glaciation at 37 Ma was subsequently followed by several late Eocene warming and cooling cycles preceding the major increase in δ18O across the Eocene-Oligocene transition. In the broad perspective, the MECO event appears to represent a major turning point in the Eocene climate evolution, separating greenhouse and icehouse climate states. Support for this interpretation is provided by clay mineral assemblage from Southern Ocean deep-sea sites, which indicate that widespread physical weathering by ice sheets did not begin until at least the late Eocene in East Antarctica. This interpretation contrasts with recent estimates of eustatic variation and deep-sea proxy studies that have been interpreted to indicate large-scale glaciation in several intervals throughout the middle Eocene.