Timing of the most recent Neoglacial advance and retreat in Potter Cove, Antarctic Peninsula

The timing of the most recent Neoglacial advance in the Antarctic Peninsula is important for establishing global climate teleconnections and providing important post-glacial rebound corrections to gravity-based satellite measurements of ice loss. However, obtaining accurate ages from terrestrial geomorphic and sedimentary indicators of the most recent Neoglacial advance in Antarctica has been hampered by the lack of historical records and the difficulty of dating materials in Antarctica. Here we use a new approach to dating flights of raised beaches in the South Shetland Islands of the northern Antarctic Peninsula to bracket the age of a Neoglacial advance that occurred between 1500 and 1700 AD, broadly synchronous with compilations for the timing of the Little Ice Age in the northern hemisphere. Our approach is based on optically stimulated luminescence of the underside of buried cobbles to obtain the age of beaches previously shown to have been deposited immediately inside and outside the moraines of the most recent Neoglacial advance. In addition, these beaches mark the timing of an apparent change in the rate of isostatic rebound thought to be in response to the same glacial advance within the South Shetland Islands. We use a Maxwell viscoelastic model of glacial-isostatic adjustment (GIA) to determine whether the rates of uplift calculated from the raised beaches are realistic given the limited constraints on the ice advance during this most recent Neoglacial advance. Our rebound model suggests that the subsequent melting of an additional 16-22% increase in the volume of ice within the South Shetland Islands would result in a subsequent uplift rate of 12.5 mm/yr that lasted until 1840 AD resulting in a cumulative uplift of 2.5 m. This uplift rate and magnitude are in close agreement with observed rates and magnitudes calculated from the raised beaches since the most recent Neoglacial advance along the South Shetland Islands and falls within the range of uplift rates from similar settings such as Alaska.

CaCO3, density, sedimentation and accumulation rates at DSDP Hole 68-503B

Pelagic sediments from DSDP Hole 5O3B contain, in their carbonate abundance data, a clear record of glacial-interglacial cycles. The eolian component of those sediments was analyzed over the past four carbonate cycles, and the mass accumulation rate (MAR) and grain size of the eolian component was determined. Eolian MARs range from 24 to 169 mg/cm**2/10**3y. and commonly are higher by a factor of three to five during times of glacial retreat. Reduced contribution during periods of glaciation most likely reflects glacial-age humidity in the American source. Grain-size values (phi50) range from 8.25 to a minimum of 8.79phi-a variation in grain mass by a factor of 3.1. Larger grains reflect more vigorous atmospheric circulation, but sizes do not covary with the carbonate or eolian accumulation curves. These data suggest that the intensity of atmospheric circulation in the tropics may reflect the 42,000 y.-tilt cycle rather than the 100,000 y.-cycle of glacial advance.

Average dip directions from intervals of dipmeter log in drill hole CRP-2A

Bedding dips in the CRP-2A drillhole were determined in two ways (1) analysis of a dipmeter log, and (2) identification of bed boundaries on digital images of the outer core surface. The two methods document the downhole increase in structural dip, to a maximum of 15° in the lowest 150 m of the hole. Dipmeter data, which are azimuthally oriented, indicate a 75° azimuth for structural tilting, in agreement with seismic reflection profiles. Core and log dips indicate that structural dip increases by 5-7° between 325 and 480 mbsf. Both, however, also exhibit high dip inhomogeneity because of depositional (e.g., cross bedding) and post-depositional (e.g., softsediment deformation) processes. This variability adds ambiguity to the search for angular unconformities within the CRP-2A drillhole. Dip directions of different lithologies are generally similar, as are dip directions for the four kinds of systems tracts. Downdip azimuths of sands and muds are slightly different from those of diamicts, possibly reflecting the divergence between ENE offshore dip and ESE glacial advance.

Description of lithofacies associations and their interpreted settings from sequences of sediment core CRP-1 (Table 2)

Cape Roberts Project drillcore 1 was obtained from Roberts Ridge, a sea-floor high located at 77°S, 16 km offshore from Cape Roberts in western McMurdo Sound, Antarctica. The recovered core is about 147 m long with the upper 43.15 metres below the sea floor (revised figure) being dated as Quarternary and the older part of the sequence being Miocene. The core includes nine facies: sandy diamict, muddy diamict, gravel/conglomerate, mud(stone), clay(stone) and carbonate. These facies occure in associations that are repeated in particulare sequences throughout the core, and are interpreted as representing different depositional environments through time. Seven lithofacies associations are interpreted as representing offshore shelf, ice protected/below wave-base; prodeltaic/offshore shelf; delta front/sandy shelf; ice system; subglacial till/rainout diamict/debris flow diamicts singly or in combination; and a carbonate-rich shelf bank. The facies associations are used to infer that the Quaternary section represents deposition on a polar shelf with perhaps two or three glacial fluctuations. The Quaternary carbonate unit indicates a period of ice sheet retreat, but local glacial activity may have increased with an increase in costal precipitation. The Miocene section represents polythermal glacial systems. The older Miocene section is glacially dominated whereas the younger section is much less so. The glacially dominated section may provide evidence for a major glacial advance thar resulted un a low stand of global eustatic sea level at that time. After the low stand, eustatic sea level was gradually rising during deposition of the younger section dominated more by non-glacial processes.

Bulk components and ice rafted debris of five sediment cores of the Sophia Basin, Arctic Ocean

Periods of enhanced terrigenous input to the ocean's basins of the North Atlantic have been reported for the last glacial period. We present a set of new sediment cores recovered from the Sophia Basin north of Svalbard which exhibit wide spread IRD layers reflecting enhanced terrigenous input throughout the last ~200 kyr. BP. Their consistent stratigraphic position, sedimentological character, high sedimentation rate and geochemical characteristic point to synchronously deposited layers which we name terrigenous input events (TIEs). Due to their higher densities, they generate excellent reflectors for sediment penetrating acoustic devices and prominent acoustic layers in the imagery of sedimentary structures. Therefore TIEs can be used for regional acoustic stratigraphy. Each of the events can be linked to major glacial activity on Svalbard. However, the Early Weichselian glaciation is not recorded as a TIE and, in agreement with other work, might not have occurred on Svalbard as a major glacial advance to the shelf break. Non-synchronous timing of western and northern sources on Svalbard points against sea-level induced iceberg discharge events.

Firn line elevation of the Dickson Land peninsula

The Dickson Land peninsula is located in central West-Spitsbergen between the NNE branches of Isfjorden. The climatic firn line lying at 500 m causes plateau glaciers with outlet tongues which are characteristic of S-Dickson Land. The distribution of valley glaciers and the variations of the orographic firn line depend on wind direction. In comparing the firn lines established by the methods of LICHTENECKER (1938) and VISSER (1938), to the values calculated by the method of v. HÖFER (1879), differences of up to l07 m are found. These differences may depend on the inclination and distance relationships of the glaciers above and below the real firn lines. During the latest glacial advance, Dickson Land was located on the peripheries of two local glaciation centers. At that time an inland glaciation of West-Spitsbergen did not exist . The formation of a subglacial channel system dates back to the maximum extent of the late glacial phase before 17500 B.P, (+2000/-1375 years). A correlation of postglacial stadia and 14C dated marine terraces (FEYLING-HANSSEN & OLSSON, 1960; FEYLING-HANSSEN, 1965) is possible. Considering isostatic movement and the difference between calculated and real firn lines, a postglacial stadium at about 10400 B. P. can be reconstructed with a firn line lying 265 m above former sea level. On average, the absolute depression below the recent firn line amounted to 246 m. Stagnation at 9650 B.P. coincided with a firn line at 315 m above former sea level and a depression of 173 m. Around 1890 A.D., glacial fluctuations corresponded to a firn line at 415 m (depression: 64 m). To some extent the morphology of the main valleys appears to depend on structure and petrography. Therefore their value as indicators of former glaciations is questionable. The periglacial forms are shown on a large-scale map. At the time of the "Holocene warm interval", between 7000 and 2000 B.P. (FEYLING-HANSSEN, 1955a, 1965), an increase of periglacial activity seems likely. This can be explained by a simultaneous increase in the depth of the active layer in both soil and bedrock.

Late Glacial and Holocene data from sediment cores MD99-2236 and MD99-2256

The Laurentide Ice Sheet (LIS) was a large, dynamic ice sheet in the early Holocene. The glacial events through Hudson Strait leading to its eventual demise are recorded in the well-dated Labrador shelf core, MD99-2236 from the Cartwright Saddle. We develop a detailed history of the timing of ice-sheet discharge events from the Hudson Strait outlet of the LIS during the Holocene using high-resolution detrital carbonate, ice rafted detritus (IRD), d18O, and sediment color data. Eight detrital carbonate peaks (DCPs) associated with IRD peaks and light oxygen isotope events punctuate the MD99-2236 record between 11.5 and 8.0 ka. We use the stratigraphy of the DCPs developed from MD99-2236 to select the appropriate DeltaR to calibrate the ages of recorded glacial events in Hudson Bay and Hudson Strait such that they match the DCPs in MD99-2236. We associate the eight DCPs with H0, Gold Cove advance, Noble Inlet advance, initial retreat of the Hudson Strait ice stream (HSIS) from Hudson Strait, opening of the Tyrrell Sea, and drainage of glacial lakes Agassiz and Ojibway. The opening of Foxe Channel and retreat of glacial ice from Foxe Basin are represented by a shoulder in the carbonate data. DeltaR of 350 years applied to the radiocarbon ages constraining glacial events H0 through the opening of the Tyrell Sea provided the best match with the MD99-2236 DCPs; DeltaR values and ages from the literature are used for the younger events. A very close age match was achieved between the 8.2 ka cold event in the Greenland ice cores, DCP7 (8.15 ka BP), and the drainage of glacial lakes Agassiz and Ojibway. Our stratigraphic comparison between the DCPs in MD99-2236 and the calibrated ages of Hudson Strait/Bay deglacial events shows that the retreat of the HSIS, the opening of the Tyrell Sea, and the catastrophic drainage of glacial lakes Agassiz and Ojibway at 8.2 ka are separate events that have been combined in previous estimates of the timing of the 8.2 ka event from marine records. SW Iceland shelf core MD99-2256 documents freshwater entrainment into the subpolar gyre from the Hudson Strait outlet via the Labrador, North Atlantic, and Irminger currents. The timing of freshwater release from the LIS Hudson Strait outlet in MD99-2236 matches evidence for freshwater forcing and LIS icebergs carrying foreign minerals to the SW Iceland shelf between 11.5 and 8.2 ka. The congruency of these records supports the conclusion of the entrainment of freshwater from the retreat of the LIS through Hudson Strait into the subpolar gyre and provides specific time periods when pulses of LIS freshwater were present to influence climate.

Last glacial maximum sediment record of the southern Weddell Sea shelf

Abstract: The history of grounded ice-sheet extent on the southern Weddell Sea shelf during the Last Glacial Maximum (LGM) and the timing of post-LGM ice-sheet retreat are poorly constrained. Several glaciological models reconstructed widespread grounding and major thickening of the Antarctic Ice Sheet in the Weddell Sea sector at the LGM. In contrast, recently published onshore data and modelling results concluded only very limited LGM-thickening of glaciers and ice streams feeding into the modern Filchner and Ronne ice shelves. These studies concluded that during the LGM ice shelves rather than grounded ice covered the Filchner and Ronne troughs, two deep palaeo-ice stream troughs eroded into the southern Weddell Sea shelf. Here we review previously published and unpublished marine geophysical and geological data from the southern Weddell Sea shelf. The stratigraphy and geometry of reflectors in acoustic sub-bottom profiles are similar to those from other West Antarctic palaeo-ice stream troughs, where grounded ice had advanced to the shelf break at the LGM. Numerous cores from the southern Weddell Sea shelf recovered sequences with properties typical for subglacially deposited tills or subglacially compacted sediments. These data sets give evidence that grounded ice had advanced across the shelf during the past, thereby grounding in even the deepest parts of the Filchner and Ronne troughs. Radiocarbon dates from glaciomarine sediments overlying the subglacial deposits are limited, but indicate that the ice grounding occurred at the LGM and that ice retreat started before ~15.1 corrected 14C kyrs before present (BP) on the outer shelf and before ~7.7 corrected 14C kyrs BP on the inner shelf, which is broadly synchronous with ice retreat in other Antarctic sectors. The apparent mismatch between the ice-sheet reconstructions from marine and terrestrial data can be attributed to ice streams with very low surface profiles (similar to those of "ice plains") that had advanced through Filchner Trough and Ronne Trough at the LGM. Considering the global sea-level lowstand of ~130 metres below present, a low surface slope of the expanded LGM-ice sheet in the southern Weddell Sea can reconcile grounding-line advance to the shelf break with limited thickening of glaciers and ice streams in the hinterland. This scenario implies that ice-sheet growth in the Weddell Sea sector during the LGM and ice-sheet drawdown throughout the last deglaciation could only have made minor contributions to the major global sea-level fluctuations during these times.

Facies distribution in CRP sediment cores from the Ross Sea, Antarctica

Clasts from the Cape Roberts Project cores CRP-2/2A and CRP-3 provide indications of glacially influenced depositional environments in Oligocene and Miocene strata in the western Victoria Land Basin, Antarctica. CRP-2/2A is interpreted to represent strongly glacially influenced, unconformity bound depositional sequences produced by repeated advance and retreat of floating and grounded ice across the shelf. A similar interpretation is extended to the upper 330 meters of the CRP-3 core, but the lower part of the core records shallow marine deposition with significantly less glacial influence. Clast shape analysis from selected coarse-grained facies throughout the cored interval indicates that most clasts are glacially sourced, with little distinction between diamictite and conglomeratic facies. Three dimensional clast fabric analysis from units immediately above sequence boundaries generally display weak or random fabrics and do not suggest that grounded ice actually reached the drillsite at these intervals. Striated and outsized clasts present in fine-grained lithofacies throughout the cores provide further evidence of sub-glacially transported sediment and iceberg rafting. The distribution of these striated and out-sized clasts indicate that a significant glacial influence persisted through most of the time represented by the cores with glaciers actively calving at sea-level introducing ice-berg rafted glacial debris even in the earliest Oligocene.

Facies investigations on sediment core CRP-3 from the Ross Sea, Antarctica

The Cenozoic Victoria Land Basin (VLB) stratigraphic section penetrated by CRP-3 is mostly of Early Oligocene age. It contains an array of lithofacies comprising fine-grained mudrocks, interlaminated and interbedded mudrocks/sandstones, mud-rich and mud-poor sandstones, conglomerates and diamctites that are together interpreted as the products of shallow marine to possibly non-marine environments of deposition, affected by the periodic advance and retreat of tidewater glaciers. This lithofacies assemblage can be readily rationalised using the facies scheme designed originally for CRP-2/2A, and published previously. The uppermost 330 metres below sea floor (mbsf) shows a cyclical arrangement of lithofacies also similar to that recognised throughout CRP-2/2A, and interpreted to reflect cyclical variations in relative sea-level driven by ice volume fluctuations ('Motif A'). Between 330 and 480 mbsf, a series of less clearly cyclical units, generally fining-upward but nonetheless incorporating a significant subset of the facies assemblage, has been identified and noted in the Initial Report as 'Motif B' Below 480 mbsf, the section is arranged into a repetitive succession of fining-upward units, each of which comprises dolerite clast conglomerate at the base passing upward into relatively thick intervals of sandstones. The cycles present down 480 mbsf are defined as sequences, each interpreted to record cyclical variation of relative sea-level. The thickness distribution of sequences in CRP-3 provides some insights into the geological variables controlling sediment accumulation in the Early Oligocene section. The uppermost part of the section in CRP-3 comprises two or three thick, complete sequences that show a broadly symmetrical arrangement of lithofacies (similar to Sequences 9-11 in CRP-2/2A). This suggests a period of relatively rapid tectonic subsidence, which allowed preservation of the complete facies cycle. Below Sequence 3, however, is a considerable interval of thin, incomplete and erosionally truncated sequences (4-23), which incorporates both the remainder of Motif A sequences and all Motif B sequences recognised. The thinner and more truncated sequences suggest sediment accumulation under conditions of reduced accommodation, and given the lack of evidence for glacial conditions (see Powell et al., this volume) tends to argue for a period of reduced tectonic subsidence. The section below 480 mbsf consists of a series of fining-upward, conglomerate to sandstone intervals which cannot be readily interpreted in terms of relative sea-level change. A relatively mudrock-rich interval above the basal conglomerate/breccia (782-762 mbsf) may record initial flooding of the basin during early rift subsidence. The lithostratigraphy summarised above has been linked to seismic reflection data using depth conversion techniques (Henrys et al., this volume). The three uppermost reflectors ('o', 'p' and 'q') correlate to the package of thick sequences 1-3, and several deeper reflectors can also be correlated to sequence boundaries. The package of thick Sequences 1-3 shows a sheet-like cross-sectional geometry on seismic reflection lines, unlike the similar package recognised in CRP-2/2A.

(Table 1) Maceral composition and organic carbon chemistry of ODP Leg 104 samples

The Cenozoic sediments sampled in ODP Leg 104 on the Vøring Plateau show a distinct variability of the total organic carbon content (TOC) and the accumulation rates of TOC. Based on the geochemical and organic-petrographic characterization of the sedimentary organic matter (OM), the allochthonous and autochthonous proportion of the OM could be quantified. The results clearly demonstrate that high TOC percentages and TOC accumulation rates in Cenozoic sediment sections display a generally high input of allochthonous organic matter. Oxidized and partly well-rounded organic particles built up the main portion of OM within the Miocene, TOC-rich sediments. The most probable source of this oxidized OM are reworked sediments from the Scandinavian shelf. Changes in the input of these organic particles are to some degree correlative with sea-level changes. The Cenozoic accumulation of autochthonous OM is low and does not reveal a clear variation during the Miocene and early Pliocene. In spite of a high accumulation rate of biogenic opal during the Early Miocene, the accumulation rate of autochthonous TOC is low. The autochthonous particle assemblage is dominated by relatively inert OM, like dinoflagellate cysts. This points to an intensive biological and/or early diagenetic degradation of the marine OM under well oxidized bottom water conditions during the last 23 Myr. Nevertheless, a continuation of marine OM degradation during later stages of diagenesis cannot be excluded. A prominent dominance of allochthonous OM over autochthonous is documented with the beginning of the Pliocene. At 2.45 Ma the episodic occurrence of ice-rafted, thermally mature OM reflects the onset of the glacial erosion of Mesozoic, coal and black shale bearing sediments on the Scandinavian and Barents Sea shelves. The first occurrence of these, in view of the actual burial depth, thermally overmature OM particles is, therefore, a marker for the beginning of the strong Scandinavian glaciation and the advance of the glacial front toward the shelves.

Elevation lines of modelled ice sheet extension during last glacial maximum

A numerical ice-sheet model was used to reconstruct the Late Weichselian glaciation of the Eurasian High Arctic, between Franz Josef Land and Severnaya Zemlya. An ice sheet was developed over the entire Eurasian High Arctic so that ice flow from the central Barents and Kara seas toward the northern Russian Arctic could be accounted for. An inverse approach to modeling was utilized, where ice-sheet results were forced to be compatible with geological information indicating ice-free conditions over the Taymyr Peninsula during the Late Weichselian. The model indicates complete glaciation of the Barents and Kara seas and predicts a "maximum-sized" ice sheet for the Late Weichselian Russian High Arctic. In this scenario, full-glacial conditions are characterized by a 1500-m-thick ice mass over the Barents Sea, from which ice flowed to the north and west within several bathymetric troughs as large ice streams. In contrast to this reconstruction, a "minimum" model of glaciation involves restricted glaciation in the Kara Sea, where the ice thickness is only 300 m in the south and which is free of ice in the north across Severnaya Zemlya. Our maximum reconstruction is compatible with geological information that indicates complete glaciation of the Barents Sea. However, geological data from Severnaya Zemlya suggest our minimum model is more relevant further east. This, in turn, implies a strong paleoclimatic gradient to colder and drier conditions eastward across the Eurasian Arctic during the Late Weichselian.

(Table 1) Age determination of Cape Shpindler, Yugorski Peninsula, northwest Russia

Depositional environments, stratigraphic relations, and 35 new AMS 14C dates at Cape Shpindler, Yugorski Peninsula, help constrain the late Pleistocene glacial and environmental history of the southern Kara Sea region. Fifteen- to fifty-meter-high coastal exposures reveal a complex package of shallow marine, fluvial, glacial, and postglacial deposits, and are documented here in a 19-km-long cross-section and eight vertical sections. The shallow marine (Unit A), estuarine or prodeltaic (Unit B), and fluvio-deltaic (Unit C) deposits contain an interglacial molluscan fauna, yield radiocarbon dates greater than 40 ka, and may correspond with a regional sea-level highstand during the Eemian. These units are overlain by a diamicton (Unit D), and are pervasively deformed by folds and low- to high-angle faults into a stacked glaciotectonic accretionary complex. The diamicton (Unit D) is a subglacial till, and associated massive ground ice with deformed debris bands (Unit E) appears to be relict glacier ice. Glaciotectonic structures document both southward- and northward-directed glacier movement. Above the till and associated glaciotectonic horizons lies 0- to 11-m-thick postglacial deposits of peatland, eolian, fluvial, and primarily lacustrine origin (Unit F). The postglacial deposits yield radiocarbon ages of 12.8 to 0.8 ka. Thus, at least one regional glaciation is prominently represented in the stratigraphy, and occurred probably after the Eemian but before 12.8 ka. We infer that the bulk of the glacial record corresponds with southward advance by an early Weichselian Kara Sea Ice Sheet, in agreement with other recently documented, regional records from Yamal Peninsula and the Pechora Basin. The timing and source of northward-directed glacier ice are less well constrained. Across the broad expanse of the Eurasian Arctic, Quaternary stratigraphy is still sparsely documented. The new data from Cape Shpindler fill a spatial gap in paleoenvironmental research.