Search

Tunnel valleys are widespread in formerly glaciated regions such as the North Sea and record sediment transport beneath ice sheets undergoing deglaciation. However, their complex infill architecture often makes their implications for ice-sheet processes difficult to unravel. Here, we use high resolution 3D (HR3D) seismic data, improved-resolution conventional 3D seismic-reflection data, and geotechnical information from industry-acquired boreholes to image the infill architecture of buried Quaternary tunnel valleys in the North Sea in unprecedented detail. Ten cross-cutting generations of tunnel valleys are mapped beneath the seafloor of the North Sea where only seven were visible previously. Each generation of tunnel valleys potentially reflects a different glaciation, although our evidence may imply that it is possible to rapidly erode and infill multiple generations of tunnel valleys within a single glacial cycle. The infill of the oldest tunnel valley generations reflects sedimentation during relatively gradual ice-sheet retreat, with occasional episodes of overriding by re-advancing grounded ice. Tunnel valleys formed in more recent glaciations are characterised by more variable sedimentation patterns that reflect dynamic fluctuations of the ice margin, including readvances and stagnation, during valley filling and ice retreat. Numerous subglacial landforms are also imaged within the tunnel valleys; these sometimes contain shallow gas accumulations that represent geohazards for seafloor installations. In addition, we document examples where salt diapirism has caused fluids to migrate upwards from depth through faults and into the near-surface tunnel valleys. In instances where this occurs, the relatively porous and often highly continuous subglacial landforms present within their infill may allow these fluids to spread laterally for kilometres or even escape from the seafloor; it is therefore important to consider tunnel valleys when monitoring possible CO2 lea

The geological record of landforms and sediments produced beneath deglaciating ice sheets offers insights into inaccessible glacial processes. Large subglacial valleys formed by meltwater erosion of sediments (tunnel valleys) are widespread in formerly glaciated regions such as the North Sea. Obtaining a better understanding of these features may help with the parameterisation of basal melt rates and the interplay between basal hydrology and ice dynamics in numerical models of past, present, and future ice-sheet configurations. However, the mechanisms and timescales over which tunnel valleys form remain poorly constrained. Here, we present a series of numerical modelling experiments, informed by new observations from high-resolution 3D seismic data (6.25 m bin size, ∼4 m vertical resolution), which test different hypotheses of tunnel valley formation and calculate subglacial water routing, seasonal water discharges, and the rates at which tunnel valleys are eroded beneath deglaciating ice sheets. Networks of smaller or abandoned channels, pervasive slump deposits, and subglacial landforms are imaged inside and at the base of larger tunnel valleys, indicating that these tunnel valleys were carved through the action of migrating smaller channels within tens of kilometres of the ice margin and were later widened by ice-contact erosion. Our model results imply that the drainage of extensive surface meltwater to the ice-sheet bed is the dominant mechanism responsible for tunnel valley formation; this process can drive rapid incision of networks of regularly spaced subglacial tunnel valleys beneath the fringes of retreating ice sheets within hundreds to thousands of years during deglaciation. Combined, our observations and modelling results identify how tunnel valleys form beneath deglaciating mid-latitude ice sheets and have implications for how the subglacial hydrological systems of contemporary ice sheets may respond to sustained climate warming.

Reconstructing the advance and retreat of past ice sheets provides important long-term context for recent change(s) and enables us to better understand ice sheet responses to forcing mechanisms and external boundary conditions that regulate grounding line retreat. This study applies various radiocarbon dating techniques, guided by a detailed sedimentological analyses, to reconstruct the glacial history of Anvers-Hugo Trough (AHT), one of the largest bathymetric troughs on the western Antarctic Peninsula (WAP) shelf. Existing records from AHT indicate that the expanded Antarctic Peninsula Ice Sheet (APIS) advanced to, or close to, the continental shelf edge during the Last Glacial Maximum (LGM; 23-19 cal kyr BP [ = calibrated kiloyears before present]), with deglaciation of the outer shelf after ∼16.3 cal kyr BP. Our new chronological data show that the APIS had retreated to the middle shelf by ∼15.7 cal kyr BP. Over this 600-year interval, two large grounding-zone wedges (GZW) were deposited across the middle (GZW2) and inner shelf (GZW3), suggesting that their formation occurred on centennial rather than millennial timescales. Expanded sequences of sub-ice shelf sediments occur seaward of the inner GZW3, which suggests that the grounding line remained stationary for a prolonged period over the middle shelf. Grounding-line retreat rates indicate faster retreat across the outer to middle shelf compared to retreat across the middle to inner shelf. We suggest that variable retreat rates relate to the broad-scale morphology of the trough, which is characterised by a relatively smooth, retrograde seabed on the outer to middle shelf and rugged morphology with a locally landward shallowing bed and deep basin on the inner shelf. A slowdown in retreat rate could also have been promoted by convergent ice flow over the inner shelf and the availability of pinning points associated with bathymetric highs around Anvers Island and Hugo Island.

Following the Last Glacial Maximum (LGM; ca. 23-19 calibrated [cal.] kyr before present [BP]), atmospheric and oceanic warming, together with global sea-level rise, drove widespread deglaciation of the Antarctic Ice Sheet, increasing the flux of freshwater to the ocean and leading to substantial changes in marine biological productivity. On the Antarctic continental shelf, periods of elevated biological productivity, often preserved in the sediment record as laminated (and sometimes varved) diatomaceous oozes (LDO), have been reported from several locations and are typically associated with the formation of calving bay re-entrants during ice sheet retreat. Understanding what drives the formation and deposition of LDOs, and the impact of deglacial processes on biogenic productivity more generally, can help inform how Antarctic coastal environments will respond to current and future ice sheet melting. In this study we utilise a suite of sediment cores recovered from Anvers-Hugo Trough (AHT), western Antarctic Peninsula shelf, which documents the transition from subglacial to glacimarine conditions following retreat of an expanded ice stream after the LGM. We present quantitative absolute diatom abundance (ADA) and species assemblage data, to investigate changes in biological productivity during the Last Glacial Transition (19-11 cal kyr BP). In combination with radiocarbon dating, we show that seasonally open marine conditions were established on the mid-shelf by 13.6 cal kyr BP, but LDOs did not start to accumulate until ∼11.5 cal kyr BP. The ∼1.4 kyr delay between the onset of seasonally open marine conditions and LDO deposition indicates that physiographic changes, and specifically the establishment of a calving bay in AHT, is insufficient to explain LDO deposition alone. LDO deposition in AHT coincides with the early Holocene climatic optimum (∼11.5 – 9.0 kyr) and is therefore explained in terms of increased atmospheric/ocean temperatures, high rates of sea and gl

The present-day morpho-stratigraphy of the Ross Sea is the result of Cenozoic tectonic and cryospheric events, and constitutes a key record of Antarctica's cryospheric evolution. An enduring problem in interpreting this record in a broader regional context is that the correlation between eastern and western Ross Sea stratigraphy has remained uncertain due to the limited number of drill sites. We correlate the glacial-related features observed on a dense network of seismic reflection profiles in McMurdo Sound with those identified in the Nordenskjöld and Drygalski Basins, as well as the basins farther east in the central Ross Sea. We present an improved correlation of the regional patterns of early to middle Miocene ice-sheet variance across the Ross Sea constrained by new evaluation of seismic facies and age models from one site recovered by the Antarctic Drilling Project (ANDRILL) in the southwestern most part of McMurdo Sound. We also integrate this correlation with the recently published seismic framework in the central Ross Sea. The formation of U-shaped valleys during the early Miocene in McMurdo Sound, together with prograding sedimentary wedges in the western-most basins, and the central Ross Sea, suggest two major phases of overall advance of a marine-terminating ice sheet between ~18 Ma and ~ 17.4 Ma. Widespread formation of turbiditic channel-levee systems in McMurdo Sound and rapid sediment deposition in Nordernskjöld Basin point to subsequent ice-sheet retreat between ~17.4 Ma and ~ 15.8 Ma, coinciding with the onset of the Miocene Climate Optimum (MCO; ~17–14.5 Ma). However, the carving of troughs and formation of irregular morphologic features suggest that an extensive ice sheet still remained along the western Ross margin at ~17.4 Ma and a brief episode of ice-sheet advance occurred at ~16.8 Ma in the earliest interval of the MCO. Subsequent marine-based ice sheet advance during the Middle Miocene Climate Transition (MMCT, ~14.0–13.8 Ma) is indicated

The end of the Eocene greenhouse world was the most dramatic phase in the long-term Cenozoic cooling trend. Here we use 75,000 km of multi/single channel seismic reflection data from offshore Prydz Bay, Antarctica, to provide new insight on the Paleogene stratigraphic transition from greenhouse to icehouse conditions and reorganizing the ocean circulation changes that were invigorated by the cooling and glaciation. We identify a new prominent Paleogene transitional phase (Greenhouse to Icehouse) preserved in the deep-water sedimentary record by correlating from shelf to the continental slope. The occurrence of mega-Mass Transport Deposits (MTDs) on the slope during an early stage in the transition suggests significant instability and collapse of the upper part of the continental margin. A second stage of the transition is represented by the growth of a well-defined set of continental slope clinoforms. We estimate the formation age of the MTDs and clinoforms to be around Eocene-Oligocene Transition. The formation of the clinoforms in the transitional phase indicates sea level has risen, and large volumes sediment delivered to the margin by marine-terminating glaciers on the shelf. Finally, a subsequent marked migration of the margin depocenter toward the west and northwest, attests the onset of drift sedimentation and full glacial conditions, suggesting a more vigorous ocean circulation as the Earth entered the icehouse conditions after the Eocene-Oligocene boundary.

The legacy of the Scientific Committee on Antarctic Research’s (SCAR) PAIS strategic research programme includes not only breakthrough scientific discoveries, but it is also the story of a long-standing deep collaboration amongst different multi-disciplinary researchers from many nations, to share scientific infrastructure and data, facilities, and numerical models, in order to address high priority questions regarding the evolution and behaviour of the Antarctic ice sheets (AIS). The PAIS research philosophy is based on data-data and data-model integration and intercomparison, and the development of ‘ice-to-abyss’ data transects and paleo-environmental, extending from the ice sheet interior to the deep sea. PAIS strives to improve understanding of AIS dynamics and to reduce uncertainty in model simulations of future ice loss and global sea level change, by studying warm periods of the geological past that are relevant to future climate scenarios. The multi-disciplinary approach fostered by PAIS represents its greatest strength. Eight years after the start of this programme, PAIS achievements have been high-profile and impactful, both in terms of field campaigns that collected unique data sets and samples, and in terms of scientific advances concerning past AIS dynamics, that have measurably improved understanding of ice sheet sensitivity in response to global warming. Here we provide an overview and synthesis of the new knowledge generated by the PAIS Programme and its implications for anticipating and managing the impacts of global sea-level rise.

The past three decades have seen a sustained and coordinated effort to refine the seismic stratigraphic framework of the Antarctic margin that has underpinned the development of numerous geological drilling expeditions from the continental shelf and beyond. Integration of these offshore drilling datasets covering the Cenozoic era with Antarctic inland datasets, provides important constraints that allow us to understand the role of Antarctic tectonics, the Southern Ocean biosphere, and Cenozoic ice sheet dynamics and ice sheet–ocean interactions on global climate as a whole. These constraints are critical for improving the accuracy and precision of future projections of Antarctic ice sheet behaviour and changes in Southern Ocean circulation. Many of the recent advances in this field can be attributed to the community-driven approach of the Scientific Committee on Antarctic Research (SCAR) Past Antarctic Ice Sheet Dynamics (PAIS) research programme and its two key subcommittees: Paleoclimate Records from the Antarctic Margin and Southern Ocean (PRAMSO) and Palaeotopographic-Palaeobathymetric Reconstructions. Since 2012, these two PAIS subcommittees provided the forum to initiate, promote, coordinate and study scientific research drilling around the Antarctic margin and the Southern Ocean. Here we review the seismic stratigraphic margin architecture, climatic and glacial history of the Antarctic continent following the break-up of Gondwanaland in the Cretaceous, with a focus on records obtained since the implementation of PRAMSO. We also provide a forward-looking approach for future drilling proposals in frontier locations critically relevant for assessing future Antarctic ice sheet, climatic and oceanic change.

A detailed morphologic and seismic stratigraphic analysis of the continental margin offshore of Adelaide Island on the Pacific Margin of the Antarctic Peninsula (PMAP) is described based on the study of a regular network of reflection multichannel seismic profiles and swath bathymetry. We present an integrated study of the margin spanning the shelf to the continental rise and establish novel chronologic constraints and offer new interpretations on tectonic evolution and environmental changes affecting the PMAP. The stratigraphic stacking patterns record major shifts in the depositional style of the margin that outline three intervals in its evolution. The first non-glacial interval (Early Cretaceous to middle Miocene) encompasses a transition from an active to a passive margin (early Miocene). The second glacially-influenced interval (middle to late Miocene) is marked by pronounced aggradational sedimentary stacking and subsidence. Ice sheets advanced over the middle shelf of the margin at the end of this second interval, while the outer shelf experienced rare progradational events. The third, fully glaciated interval shows clear evidence of glacially dominated conditions on the margin. This interval divides into three minor stages. During the first stage (late Miocene to the beginning of the early Pliocene), frequent grounded ice advances to the shelf break began, depositing an initial progradational unit. A major truncation surface marked the end of this stage, which coincided with extensive mass transport deposits at the base of the slope. During the second progradational glacial margin stage (early Pliocene to middle Pleistocene), stacking patterns record clearly prograding glacial sequences. The beginning of the third aggradational glacial margin stage (middle Pleistocene to present) corresponded to an important shift in global climate during the Mid-Pleistocene Transition. Morphosedimentary characteristics observed along the margin today began to develop durin

This paper compiles and reviews marine and terrestrial data constraining the dimensions and configuration of the Antarctic Peninsula Ice Sheet (APIS) from the Last Glacial Maximum (LGM) through deglaciation to the present day. These data are used to reconstruct grounding-line retreat in 5 ka time-steps from 25 ka BP to present. Glacial landforms and subglacial tills on the eastern and western Antarctic Peninsula (AP) shelf indicate that the APIS was grounded to the outer shelf/shelf edge at the LGM and contained a series of fast-flowing ice streams that drained along cross-shelf bathymetric troughs. The ice sheet was grounded at the shelf edge until ∼20 cal ka BP. Chronological control on retreat is provided by radiocarbon dates on glacimarine sediments from the shelf troughs and on lacustrine and terrestrial organic remains, as well as cosmogenic nuclide dates on erratics and ice moulded bedrock. Retreat in the east was underway by about 18 cal ka BP. The earliest dates on recession in the west are from Bransfield Basin where recession was underway by 17.5 cal ka BP. Ice streams were active during deglaciation at least until the ice sheet had pulled back to the mid-shelf. The timing of initial retreat decreased progressively southwards along the western AP shelf; the large ice stream in Marguerite Trough may have remained grounded at the shelf edge until about 14 cal ka BP, although terrestrial cosmogenic nuclide ages indicate that thinning had commenced by 18 ka BP. Between 15 and 10 cal ka BP the APIS underwent significant recession along the western AP margin, although retreat between individual troughs was asynchronous. Ice in Marguerite Trough may have still been grounded on the mid-shelf at 10 cal ka BP. In the Larsen-A region the transition from grounded to floating ice was established by 10.7–10.6 cal ka BP. The APIS had retreated towards its present configuration in the western AP by the mid-Holocene but on the eastern peninsula may have approached its pre

Marine and terrestrial geological and marine geophysical data that constrain deglaciation since the Last Glacial Maximum (LGM) of the sector of the West Antarctic Ice Sheet (WAIS) draining into the Amundsen Sea and Bellingshausen Sea have been collated and used as the basis for a set of time-slice reconstructions. The drainage basins in these sectors constitute a little more than one-quarter of the area of the WAIS, but account for about one-third of its surface accumulation. Their mass balance is becoming increasingly negative, and therefore they account for an even larger fraction of current WAIS discharge. If all of the ice in these sectors of the WAIS were discharged to the ocean, global sea level would rise by ca 2 m. There is compelling evidence that grounding lines of palaeo-ice streams were at, or close to, the continental shelf edge along the Amundsen Sea and Bellingshausen Sea margins during the last glacial period. However, the few cosmogenic surface exposure ages and ice core data available from the interior of West Antarctica indicate that ice surface elevations there have changed little since the LGM. In the few areas from which cosmogenic surface exposure ages have been determined near the margin of the ice sheet, they generally suggest that there has been a gradual decrease in ice surface elevation since pre-Holocene times. Radiocarbon dates from glacimarine and the earliest seasonally open marine sediments in continental shelf cores that have been interpreted as providing approximate ages for post-LGM grounding-line retreat indicate different trajectories of palaeo-ice stream recession in the Amundsen Sea and Bellingshausen Sea embayments. The areas were probably subject to similar oceanic, atmospheric and eustatic forcing, in which case the differences are probably largely a consequence of how topographic and geological factors have affected ice flow, and of topographic influences on snow accumulation and warm water inflow across the continental sh

Glaciers flowing into the Amundsen Sea Embayment (ASE) account for >35% of the total discharge of the West Antarctic Ice Sheet (WAIS) and have thinned and retreated dramatically over the past two decades. Here we present detailed marine geological data and an extensive new radiocarbon dataset from the eastern ASE in order to constrain the retreat of the WAIS since the Last Glacial Maximum (LGM) and assess the significance of these recent changes. Our dating approach, relying mainly on the acid insoluble organic (AIO) fraction, utilises multi-proxy analyses of the sediments to characterise their lithofacies and determine the horizon in each core that would yield the most reliable age for deglaciation. In total, we dated 69 samples and show that deglaciation of the outer shelf was underway before 20,600 calibrated years before present (cal. yr BP), reaching the mid-shelf by 13,575 cal. yr BP and the inner shelf to within c.150 km of the present grounding line by 10,615 cal. yr BP. The timing of retreat is broadly consistent with previously published radiocarbon dates on biogenic carbonate from the eastern ASE as well as AIO 14C ages from the western ASE and provides new constraints for ice sheet models. The overall retreat trajectory – slow on the outer shelf, more rapid from the middle to inner shelf – clearly highlights the importance of reverse bedslopes in controlling phases of accelerated groundling line retreat. Despite revealing these broad scale trends, the current dataset does not capture detailed changes in ice flow, such as stillstands during grounding line retreat (i.e., deposition of grounding zone wedges) and possible readvances as depicted in the geomorphological record.

The Weddell Sea sector is one of the main formation sites for Antarctic Bottom Water and an outlet for about one fifth of Antarctica's continental ice volume. Over the last few decades, studies on glacial–geological records in this sector have provided conflicting reconstructions of changes in ice-sheet extent and ice-sheet thickness since the Last Glacial Maximum (LGM at ca 23–19 calibrated kiloyears before present, cal ka BP). Terrestrial geomorphological records and exposure ages obtained from rocks in the hinterland of the Weddell Sea, ice-sheet thickness constraints from ice cores and some radiocarbon dates on offshore sediments were interpreted to indicate no significant ice thickening and locally restricted grounding-line advance at the LGM. Other marine geological and geophysical studies concluded that subglacial bedforms mapped on the Weddell Sea continental shelf, subglacial deposits and sediments over-compacted by overriding ice recovered in cores, and the few available radiocarbon ages from marine sediments are consistent with major ice-sheet advance at the LGM. Reflecting the geological interpretations, different ice-sheet models have reconstructed conflicting {LGM} ice-sheet configurations for the Weddell Sea sector. Consequently, the estimated contributions of ice-sheet build-up in the Weddell Sea sector to the {LGM} sea-level low-stand of ~130 m vary considerably. In this paper, we summarise and review the geological records of past ice-sheet margins and past ice-sheet elevations in the Weddell Sea sector. We compile marine and terrestrial chronological data constraining former ice-sheet size, thereby highlighting different levels of certainty, and present two alternative scenarios of the {LGM} ice-sheet configuration, including time-slice reconstructions for post-LGM grounding-line retreat. Moreover, we discuss consistencies and possible reasons for inconsistencies between the various reconstructions and propose objectives for future resear

Five quantitatively distinct gully types are identified on the Antarctic continental margin from swath bathymetric data of over 1100 individual features. The gullies differ in terms of length, width, depth, width/depth ratio, cross-sectional shape, branching order, sinuosity and spatial density. Quantitative analysis suggests that Antarctic gully morphology varies with local slope character (i.e. slope geometry, gradient), regional factors (i.e. location of cross-shelf troughs, trough mouth fans, subglacial meltwater production rates, drainage basin size), sediment yield and ice-sheet history. In keeping with interpretations of previous researchers, most gullies are probably formed by hyperpycnal flows of sediment-laden subglacial meltwater released from beneath ice-sheets grounded at the continental shelf edge during glacial maxima. The limited down-slope extent of gullies on the western Antarctic Peninsula is explained by the steep gradient and slope geometry at the mouth of Marguerite Trough, which cause flows to accelerate and entrain seawater more quickly, resulting in a reduction of the negative buoyancy effect of the sediment load. Due to pressure gradients at the ice-sheet bed caused by variations in ice thickness inside and outside palaeo-ice stream troughs, subglacial meltwater flow was generally focussed towards trough margins. This has resulted in gullies with larger cross-sectional areas and higher sinuosities at the trough margins. A unique style of gullying is observed off one part of the western Antarctic Peninsula, corresponding to an area in which the ice-sheet grounding line is not thought to have reached the shelf edge during the Last Glacial Maximum. We interpret the features in this area as the cumulative result of slope processes that operated over a long period of time in the absence of hyperpycnal meltwater flows.

Multichannel seismic reflection lines collected in the western Amundsen Sea Embayment (ASE) provide an insight into the sedimentary cover on the shelf, which documents glacial processes. Numerous columnar, reflection-poor structures penetrating the sedimentary sequences on the middle shelf form the focus of this study. The features range between 50 and 500 m in width, and from a few metres up to 500 m in height. The columns originate and end at different depths, but do not seem to penetrate to the seafloor. They show well-defined vertical boundaries, and reflection signals can be identified below them. Hence, we exclude gas-bearing chimneys. Based on the general seismic reflection characteristics we suggest that the columns originate from dewatering processes which occur close to glaciated areas where fluids are pressed out of rapidly loaded sediments. Likely several mud-diapirs rise from water-rich mud layers within a mixed sedimentary succession and penetrate overlying denser and coarse-grained sediment strata. The presence of fluid-escape veins indicates a glacial origin and overprinting of the older sedimentary sequences on the ASE. The locations of the structures indicate that grounded ice sheets reached at least onto the middle shelf during former glacial periods.

We present a reconstruction of the Antarctic topography at the Eocene–Oligocene (ca. 34 Ma) climate transition. This provides a realistic key boundary condition for modeling the first big Antarctic ice sheets at this time instead of using the present day bedrock topography, which has changed significantly from millions of years of tectonism and erosion. We reconstruct topography using a set of tools including ice sheet-erosion models, models of thermal subsidence and plate movement. Erosion estimates are constrained with offshore sediment volumes estimated from seismic stratigraphy. Maximum and minimum topographic reconstructions are presented as indicators of the range of uncertainty. Our results point to a significant upland area in the Ross Sea/Marie Byrd Land and Weddell Sea sectors. In addition, East Antarctic coastal troughs are much shallower than today due to the restoration of material that has been selectively eroded by the evolving ice sheets. Parts of East Antarctica have not changed since the E–O boundary because they were protected under non-erosive cold-based ice. The reconstructions provide a better-defined boundary condition for modeling that seeks to understand interaction between the Antarctic ice sheet and climate, along with more robust estimates of past ice volumes under a range of orbital settings and greenhouse gas concentrations.

The Last Glacial Maximum (LGM; ca 23–19 ka BP) extent of grounded ice in the Filchner Trough, a major cross shelf trough extending seaward from the Filchner Ice Shelf in the southern Weddell Sea, has been much debated. Here we present data from the first extensive multibeam swath bathymetry and sub-bottom acoustic profiling surveys in the Filchner Trough that include several parallel survey lines with overlapping swaths. We interpret these new data, combined with published observations and radiocarbon dates from sediment cores, as indicating that the grounding line in the Filchner Trough during the LGM advanced beyond the middle shelf, probably to within 40 km of the shelf break, and possibly reached the shelf break. Three different hypotheses are discussed that could reconcile this interpretation with interpretations, based on ice coring and surface exposure age data, that LGM ice surface elevations in areas draining into the Filchner and Ronne ice shelves were no more than a few hundred metres higher than today: (1) ice plain conditions extended along most of the Filchner Trough; (2) the ice shelf advanced and thickened so that it “touched down” on the continental shelf for a short period; (3) LGM ice drainage pathways in the interior of the Weddell Sea embayment were different from those observed today.

Multichannel seismic reflection lines collected in the western Amundsen Sea Embayment (ASE) provide an insight into the sedimentary cover on the shelf, which documents glacial processes. Numerous columnar, reflection-poor structures penetrating the sedimentary sequences on the middle shelf form the focus of this study. The features range between 50 to 500 m in width, and from a few metres up to 500 m in height. The columns originate and end at different depths, but do not seem to penetrate to the seafloor. They show well-defined vertical boundaries, and reflection signals can be identified below them. Hence, we exclude gas-bearing chimneys. Based on the general seismic reflection characteristics we suggest that the columns originate from dewatering processes which occur close to glaciated areas where fluids are pressed out of rapidly loaded sediments. Likely several mud-diapirs rise from water-rich mud layers within a mixed sedimentary succession and penetrate overlying denser and coarse-grained sediment strata. The presence of fluid-escape veins indicates a glacial origin and overprinting of the older sedimentary sequences on the ASE. The locations of the structures indicate that grounded ice sheets reached at least onto the middle shelf during former glacial periods.

The Amundsen Sea Embayment (ASE) drains approximately 35% of the West Antarctic Ice Sheet (WAIS) and is one of the most rapidly changing parts of the cryosphere. In order to predict future ice sheet behaviour, modellers require long-term records of ice-sheet melting to constrain and build confidence in their simulations. Here, we present detailed marine geological and radiocarbon data along three palaeo-ice stream tributary troughs in the western ASE to establish vital information on the timing of deglaciation of the WAIS since the Last Glacial Maximum (LGM). We have undertaken multi-proxy analyses of the cores (core description, shear strength, x-radiographs, magnetic susceptibility, wet bulk density, total organic carbon/nitrogen, carbonate content and clay mineral analyses) in order to: (1) characterise the sedimentological facies and depositional environments; and (2) identify the horizon(s) in each core that would yield the most reliable age for deglaciation. In accordance with previous studies we identify three key facies, which offer the most reliable stratigraphies for dating deglaciation by recording the transition from a grounded ice sheet to open marine environments. These facies are: i) subglacial, ii) proximal grounding line, and iii) seasonal open marine. In addition, we incorporate ages from other facies (e.g., glaciomarine diamictons deposited at some distance from the grounding line, such as glaciogenic debris flows and iceberg-rafted diamictons and turbates) into our deglacial model. In total, we have dated 78 samples (mainly the acid insoluble organic (AIO) fraction, but also calcareous foraminifers), which include 63 downcore and 15 surface samples. Through careful sample selection prior to dating, we have established a robust deglacial chronology for this sector of the WAIS. Our data show that deglaciation of the western ASE was probably underway as early as 22,351 calibrated years before present (cal yr BP), reaching the mid-shelf by 13,837 cal

A major trough ("Belgica Trough") eroded by a palaeo-ice stream crosses the continental shelf of the southern Bellingshausen Sea (West Antarctica) and is associated with a trough mouth fan ("Belgica TMF") on the adjacent continental slope. Previous marine geophysical and geological studies investigated the bathymetry and geomorphology of Belgica Trough and Belgica TMF, erosional and depositional processes associated with bedform formation, and the temporal and spatial changes in clay mineral provenance of subglacial and glaciomarine sediments. Here, we present multi-proxy data from sediment cores recovered from the shelf and uppermost slope in the southern Bellingshausen Sea and reconstruct the ice-sheet history since the last glacial maximum (LGM) in this poorly studied area of West Antarctica. We combined new data (physical properties, sedimentary structures, geochemical and grain-size data) with published data (shear strength, clay mineral assemblages) to refine a previous facies classification for the sediments. The multi-proxy approach allowed us to distinguish four main facies types and to assign them to the following depositional settings: 1) subglacial, 2) proximal grounding-line, 3) distal sub-ice shelf/sub-sea ice, and 4) seasonal open-marine. In the seasonal open-marine fades we found evidence for episodic current-induced winnowing of near-seabed sediments on the middle to outer shelf and at the uppermost slope during the late Holocene. In addition, we obtained data on excess Pb-210 activity at three core sites and 44 AMS C-14 dates from the acid-insoluble fraction of organic matter (AIO) and calcareous (micro-) fossils, respectively, at 12 sites. These chronological data enabled us to reconstruct, for the first time, the timing of the last advance and retreat of the West Antarctic Ice Sheet (WAIS) and the Antarctic Peninsula Ice Sheet (APIS) in the southern Bellingshausen Sea. We used the down-core variability in sediment provenance inferred from clay mi

The submarine glacial geomorphology and sedimentology of the cross-shelf troughs and the adjacent continental slope around the Antarctic Peninsula and West Antarctica have been the focus of a series of marine geophysical and geological investigations over the last decade (e.g., [Shipp et al., 1999], [Canals et al., 2000], [Wellner et al., 2001], [Lowe and Anderson, 2002], [Ó Cofaigh et al., 2002], [Dowdeswell et al., 2004], [Heroy and Anderson, 2005], [Evans et al., 2005], [Domack et al., 2006], [Mosola and Anderson, 2006], [Wellner et al., 2006], [Dowdeswell et al., 2006], [Ó Cofaigh et al., 2007], Ó Cofaigh et al., 2008 C. Ó Cofaigh, J.A. Dowdeswell, J. Evans and R.D. Larter, Geological constraints on Antarctic palaeo-ice stream retreat, Earth Surface Processes and Landforms 33 (2008), pp. 513–525.[Ó Cofaigh et al., 2008] and [Dowdeswell et al., 2008]). These studies have interpreted sets of characteristic streamlined glacial bedforms and sediments on the shelf, which in numerous cases occur in front of modern ice streams, as largely the product of fast-flowing palaeo-ice streams that drained across the shelf during or following the last glacial maximum. A key feature of these studies is the observation of highly attenuated bedforms known as mega-scale glacial lineations (MSGL), formed in soft sediment on the outer continental shelf. These lineations are regarded as key evidence for streaming flow. In their recent paper, Shaw et al. present a radically different interpretation of the glacial geomorphology of Antarctic cross-shelf troughs in terms of catastrophic discharge of subglacial meltwater floods across the shelf. In their interpretation, MSGL are regarded as the product of erosion by turbulent meltwater flow. The following comment discusses a number of the key assertions made in their paper.

Seismic stratigraphic studies and scientific drilling of the Antarctic continental margin have yielded clues to the evolution of Cenozoic climates, depositional paleoenvironments and paleoceanographic conditions. This paper draws on studies of the former Antarctic Offshore Stratigraphy Project and others to review the geomorphic and lithostratigraphic offshore features that give insights into the long-duration (m.y.) and short-term (k.y.) changes that document the great variability of Cenozoic Antarctic paleoenvironments. The lithologic drilling record documents non-glacial (pre-early Eocene) to full-glacial (late Pliocene to Holocene) times, and documents times of cyclic ice-sheet fluctuations at k.y. scales (early Miocene to Pliocene and Holocene). Times of significant change in types and/or amounts of glaciation are also seen in the offshore lithologic record (early Oligocene, mid-Miocene, early Pliocene). Seismic data illustrate large-scale geomorphic features that point to massive sediment erosion and dispersal by ice sheets and paleoceanographic processes (e.g. cross-shelf troughs, slope-fans, rise-drifts). The commonality of these features to East and West Antarctica since late Eocene time points to a continent that has been intermittently covered, partially to completely, by glaciers and ice sheets. The greatest advances in our understanding of paleoenvironments and the processes that control them have been achieved from scientific drilling, and future progress depends on a continuation of such drilling.

Swath bathymetric and sub-bottom profiler data reveal a variety of submarine landforms such as gullies, slide scars, subtle shelf edge-parallel ridges and elongated depressions, and small debris flows along the continental shelf break and upper slope of West Antarctica. Gullies cutting through debris flow deposits on the Belgica Trough Mouth Fan (TMF) suggest formation after full-glacial deposition on the continental slope. The gullies were most likely eroded by sediment-laden subglacial meltwater flows released from underneath the ice margin grounded at the shelf edge at the onset of deglaciation. Scarcity of subglacial meltwater flow features on the outer shelf suggests that the meltwater reached the shelf edge mainly either through the topmost layer of soft diamict or in the form of dispersed flow beneath the ice, although locally preserved erosional channels indicate that more focused and higher-energy flows also existed. Concentration of gullies on the upper continental slope in front of the marginal areas of the major cross-shelf troughs, as contrasted to their axial parts, is indicative of higher-energy gully-eroding processes there, possibly due to additional subglacial meltwater flow from beneath the slow moving ice lying over the higher banks of the troughs. The shallow and sinuous gully heads observed on the outermost shelf within the Pine Island West Trough may indicate postglacial modification by near-bed currents resulting either from the subglacial meltwater flow from underneath the ice margin positioned at some distance landward from the shelf edge, or from the currents formed by brine rejection during sea ice formation. On the continental slope outside major troughs, slide scars as well as shelf-edge parallel ridges and elongated depressions indicate an unstable and failure-prone uppermost slope, although failures were probably mainly associated with rapid sediment loading during glacial periods. Complex, cauliflower- and amphitheatre-shaped gully h

The presence of a complex bedform arrangement on the sea floor of the continental shelf in the western Amundsen Sea Embayment, West Antarctica, indicates a multi-temporal record of flow related to the activity of one or more ice streams in the past. Mapping and division of the bedforms into distinct landform assemblages reveals their time-transgressive history, which implies that bedforms can neither be considered part of a single downflow continuum nor a direct proxy for palaeo-ice velocity, as suggested previously. A main control on the bedform imprint is the geology of the shelf, which is divided broadly between rough bedrock on the inner shelf, and smooth, dipping sedimentary strata on the middle to outer shelf. Inner shelf bedform variability is well preserved, revealing information about local, complex basal ice conditions, meltwater flow, and ice dynamics over time. These details, which are not apparent at the scale of regional morphological studies, indicate that past ice streams flowed across the entire shelf at times, and often had onset zones that lay within the interior of the Antarctic Ice Sheet today. In contrast, highly elongated subglacial bedforms on sedimentary strata of the middle to outer shelf represent a timeslice snapshot of the last activity of ice stream flow, and may be a truer representation of fast palaeo-ice flow in these locations. A revised model for ice streams on the shelf captures complicated multi-temporal bedform patterns associated with an Antarctic palaeo-ice stream for the first time, and confirms a strong substrate control on a major ice stream system that drained the West Antarctic Ice Sheet during the Late Quaternary. (C) 2009 Elsevier Ltd. All rights reserved.

An animated reconstruction shows South Pacific plate kinematics, in the reference frame of West Antarctica, between 55 Ma and the present-day. The ocean floor in the region formed due to seafloor spreading between the Antarctic, Pacific, Phoenix and Nazca plates (a plate formed by fragmentation of the Farallon plate early in Oligocene times). The Pacific-Antarctic Ridge remained fairly stable throughout this time, migrating relatively northwestwards, by various mechanisms, behind the rapidly-moving Pacific plate. The Nazca and Phoenix plates also moved quickly, but relatively towards the cast or southeast, and were subducted in these directions beneath the South American and Antarctic plates. Segments of spreading centres forming at the trailing edges of the Nazca and Phoenix plates periodically collided with these subduction zones, resulting in the total destruction of the Nazca Phoenix spreading centre and the partial destruction of the Nazca-Antarctica spreading centre (the Chile Ridge) and Antarctic-Phoenix Ridge, which ceased to operate shortly before its northeasternmost three segments could collide with the Antarctic margin. Following collision of segments of the Chile Ridge, parts of the Antarctic plate underwent subduction at the Chile Trench. After these collisions, slab windows Should have formed beneath both the South American and Antarctic convergent margins, and the animation shows Occurrences of alkaline volcanism that have been, or can newly be, related to them. Further occurrences of alkali basalts, at the margins of the Powell Basin and, more speculatively, James Ross Island, can be related to the formation of a slab window beneath them following the collision of segments of the South America Antarctica spreading centre in the northwest Weddell Sea. (C) 2007 Elsevier B.V. All rights reserved.

This chapter explores the Middle Miocene to Pliocene terrestrial and marine records of Antarctica and the Southern Ocean. The structure of the chapter makes a clear distinction between terrestrial and marine records as well as proximal (on or around Antarctica) and more distal records (Southern Ocean). Particular geographical regions are identified that reflect the areas for which the majority of palaeoenvironmental and palaeoclimatic information exist. Specifically, the chapter addresses the terrestrial sedimentary and fjordal environments of the Transantarctic Mountains and Lambert Glacier region, the terrestrial fossil record of Antarctic climate, terrestrial environments of West Antarctica, and the marine records of the East Antarctic Ice Sheet (EAIS), the West Antarctic Ice Sheet (WAIS) and the Antarctic Peninsula Ice Sheet (APIS), as well as the marine record of the Southern Ocean. Previous and current studies focusing on modelling Middle Miocene to Pliocene climate, environments and ice sheets are discussed.

Multichannel seismic reflection profiles provide a record of the glacial development in the western Amundsen Sea Embayment during the Neogene. We identified a northwest-dipping reflector series of more than 1 s TWT thickness (>800 m) on the middle continental shelf indicating well-layered sedimentary units. The dipping strata reveal a striking alternation of reflection-poor, almost transparent units and sequences of closely spaced, continuous reflectors. We suggest that the distinct changes in reflection character represent episodes of ice sheet advance and retreat forced by climate changes. Boundaries between acoustic units are sharp, but without chronological data we cannot constrain the rapidity of glacial advances and retreats. Due to the similarity between the seismic stratigraphy and the lithology in bore-hole records from the Amundsen Sea and Ross Sea, we infer that dipping strata have accumulated since an intensification of glaciation in the Miocene. On the inner and middle shelf we can identify at least four episodes of ice sheet expansion. We conclude that the West Antarctic Ice Sheet has responded sensitively to climate variations since the Miocene.

The Belgica Trough and the adjacent Belgica Trough Mouth Fan in the southern Bellingshausen Sea (Pacific sector of the Southern Ocean) mark the location of a major outlet for the West Antarctic Ice Sheet during the Late Quaternary. The drainage basin of an ice stream that advanced through Belgica Trough across the shelf during the last glacial period comprised an area exceeding 200,000 km(2) in the West Antarctic hinterland. Previous studies, mainly based on marine-geophysical data from the continental shelf and slope, focused on the bathymetry and seafloor bedforms, and the reconstruction of associated depositional processes and ice-drainage patterns. In contrast, there was only sparse information from seabed sediments recovered by coring. In this paper, we present lithological and clay mineralogical data of 21 sediment cores collected from the shelf and slope of the southern Bellingshausen Sea. Most cores recovered three lithological units, which can be attributed to facies types deposited under glacial, transitional and seasonally open-marine conditions. The clay mineral assemblages document coinciding changes in provenance. The relationship between the clay mineral assemblages in the subglacial and proglacial sediments on the shelf and the glacial diamictons on the slope confirms that a grounded ice stream advanced through Belgica Trough to the shelf break during the past, thereby depositing detritus eroded in the West Antarctic hinterland as soft till on the shelf and as glaciogenic debris flows on the slope. The thinness of the overlying transitional and seasonally open-marine sediments in the cores suggests that this ice advance occurred during the last glacial period. Clay mineralogical, acoustic sub-bottom and seismic data furthermore demonstrate that the palaeo-ice stream probably reworked old sedimentary strata, including older tills, on the shelf and incorporated this debris into its till bed. The geographical heterogeneity of the clay mineral assemblage

A 330-km length of the little known continental shelf edge and slope of the Bellingshausen Sea, West Antarctica, is investigated using multibeam swath-bathymetric and sub-bottom profiler evidence. The shelf break is at 650-700 m across the 150-km wide Belgica Trough, and to either side is about 500 m. When full-glacial ice advanced across the shelf to reach the shelf break, it was partitioned into fast- and slow-flowing elements, with an ice stream filling the trough. This had important consequences for the nature and rate of sediment delivery to the adjacent continental slope. Off Belgica Trough, the upper continental slope has convex-outward contours indicating a major sedimentary depocentre of gradient 1-2 degrees. Acoustic profiles and cores from the depocentre show a series of diamictic glacigenic debris flows. The depocentre is interpreted as a trough-mouth fan, built largely by debris delivered from the ice stream. The slope is steeper beyond the trough margins at up to 6 degrees. The main morphological features on the Bellingshausen Sea slope are gully systems and channels. Major canyons and Late Quaternary slides are absent. Most gullies and channels are found on the fan. Gullies are about 15-25 m deep, a few hundred metres wide and some are >25 km long. The largest channel is over 60 km long, about a kilometre wide and 10 to 15 m deep. The channels provide pathways for sediment by-passing of the upper slope and transfer to the continental rise and beyond by turbidity currents. Gullies on the Bellingshausen Sea margin cut through debris flows on the slope. Assuming the debris flows are linked mainly to downslope transport of diamictic debris when ice was at the shelf edge under full-glacial conditions, then those gullies cut into them formed during deglaciation. Belgica Fan is >22,000 km(2) in area and about 60,000 km(3) in volume. It is the largest depocentre identified to date on the continental margin of the West Antarctic Ice Sheet, fed by an interior i

Seismic reflection profiles, bathymetric and magnetic data collected along and across the continental margin of the Bellingshausen Sea provide new constraints and interpretations of the oceanic basement structure and Cenozoic glacial history of West Antarctica. Evidence for tectonic boundaries that lie perpendicular to the margin has been identified on the basis of one previously unpublished along-slope multichannel seismic reflection profile. By combining several magnetic data sets, we determined basement ages and verified the positions of possible fracture zones, enabling us to improve previous tectonic and stratigraphic models. We establish three main sediment units on the basis of one seismic along-slope profile and by correlation to the continental shelf via one cross-slope profile. We interpret a lowermost unit, Be3 (older then 9.6 Ma), as representing a long period of slow accumulation of mainly turbiditic sediments. Unit Be2 (from about 9.6 to 5.3 Ma) may represent a period of short-lived ice advances on the continental shelf. The uppermost unit, Be1 (from about 5.3 Ma to present), apparently consists of rapidly deposited terrigenous sediment that we interpret as having been transported to the shelf edge by frequent advances of grounded ice. Listric faults are observed in Be1 and indicate sediment instability due to interactions between different depositional processes. Correlation of the sediment classification scheme with the continental rise of the western Antarctic Peninsula shows obvious differences in sediment depositional patterns. We estimate a very high sedimentation rate for Unit Be1 (up to 295 m/my) which points to an increase in glacial sediment supply due to major glacial outlets that flowed to nearby parts of the shelf edge in Pliocene and Quaternary times. This is in contrast to the situation at the adjacent Antarctic Peninsular margin and many other parts of the continental rise around Antarctica.

The Pacific margin of the Antarctic Peninsula to the southwest of the Hero Fracture Zone (HFZ) is a former subducting margin which became inactive following the arrival of ridge crest segments of the Antarctic - Phoenix ridge at the margin during the Tertiary. In contrast, the part of the margin to the northeast of the HFZ remains active. Tertiary convergence was approximately perpendicular to the margin and ongoing motion is thought to have the same orientation. A new seismic reflection profile running along Boyd Strait, just northeast of the landward projection of the HFZ, shows major structural components similar to those typically observed along the margin to the southwest of the HFZ. In order of increasing proximity to the margin, these components are: the inner shelf, the shelf basin, the mid-shelf basement high (MSBH), and the outer shelf. The continuation of the post-subduction margin structures to the active margin suggests that the boundary between crust with passive and active margins characteristics is not sharply defined. Our postulated scenario for tectonic evolution along Boyd Strait is that: (1) before the arrival of the last ridge crest segment to the southwest of the HFZ, the inner shelf and the shelf basin were part of a Cretaceous-Tertiary arc and forearc area, (2) after the arrival, thermal effects resulting from interaction of the ridge crest with the margin just southwest of the HFZ lead to the formation of the MSBH to the northeast, but MSBH uplift in Boyd Strait did not prevent concurrent cross-shelf sediment transport contributing to development of an extensive outer shelf on the seaward flank of the MSBH, (3) Recent extension in Bransfield Strait, a marginal basin to the northeast of the landward projection of the HFZ, has caused about 10 kin of seaward deflection in the strike of the part of the MSBH to the northeast of the projection of the HFZ.

Despite a spreading rate of 65–70 km Ma−1, the East Scotia Ridge has, along most of its length, a form typically associated with slower rates of sea floor spreading. This may be a consequence of cooler than normal mantle upwelling, which could be a feature of back-arc spreading. At the northern end of the ridge, recently acquired sonar data show a complex, rapidly evolving pattern of extension within 100 km of the South Sandwich Trench. New ridge segments appear to be nucleating at or near the boundary between the South American and Scotia Sea plates and propagating southwards, supplanting older segments. The most prominent of these, north of 56°30′S, has been propagating at a rate of approximately 60 km Ma−1 for at least 1 Ma, and displays a morphology unique on this plate boundary. A 40 km long axial high exists at the centre of this segment, forming one of the shallowest sections of the East Scotia Ridge. Beneath it, seismic reflection profiles reveal an axial magma chamber, or AMC, reflector, similar to those observed beneath the East Pacific Rise and Valu Fa Ridge. Simple calculations indicate the existence here of a narrow (<1 km wide) body of melt at a depth of approximately 3 km beneath the sea floor. From the topographic and seismic data, we deduce that a localised mantle melting anomaly lies beneath this segment. Rates of spreading in the east Scotia Sea show little variation along axis. Hence, the changes in melt supply are related to the unique tectonic setting, in which the South American plate is tearing to the east, perhaps allowing mantle flow around the end of the subducting slab. Volatiles released from the torn plate edge and entrained in the flow are a potential cause of the anomalous melting observed. A southward mantle flow may have existed beneath the axis of the East Scotia Ridge throughout its history.

The outer shelf on the Antarctic Peninsula Pacific margin south of 63°30′S is underlain by Pliocene-Pleistocene prograding sequences which have been produced mainly by the action of ice sheets grounded out to the shelf edge at times of glacial maximum. Most sediment in these sequences has probably been transported to the margin in a deforming basal till, which implies deposition on a broad front: a “line source”. A representative prograding sequence mapped across an extensive network of multichannel seismic reflection lines has an elongate depocentre on the upper palaeoslope, which is consistent with the grounded ice sheet model. However, it is likely that each sequence recognized on existing multichannel seismic data represents several ice advances. Depth-to-surface maps reveal a broad variation along the margin in the amount of progradation, reflecting differences in sediment supply. The pattern of progradation and the bathymetry of the outer shelf suggest that the main depocentre in the area studied was fed by an ice stream at times of glacial maximum. Seismic lines across the margin farther to the southwest indicate the existence of other depocentres. Several broad depositional lobes have probably coalesced to form the extensive outer shelf. The present continental slope is smooth and steep, and is not cut by major canyons. A downslope change in seismic facies and scouring on the uppermost rise probably reflect downslope transitions from slumps to debris flows to turbidity currents. These processes are likely to be most active at times of glacial maximum. Deep drilling data indicate that the rise sediments consist mainly of terrigenous turbidite and ice-rafted detritus. A marked upward change in seismic facies on the continental rise indicates a change to a higher energy sedimentary regime and appears to correlate with the start of glacial progradation on the shelf.