Search

Quantitative satellite observations provide a comprehensive assessment of ice sheet mass loss over the last four decades, but limited insights into long-term drivers of ice sheet change. Geological records can extend the observational record and aid our understanding of ice sheet–climate interactions. Here we present the first millennial-scale reconstruction of David Glacier, the largest East Antarctic outlet glacier in Victoria Land. We use surface exposure dating of glacial erratics deposited on nunataks to reconstruct changes in ice surface elevation through time. We then use numerical modelling experiments to determine the drivers of glacial thinning. Thinning profiles derived from 45 10Be and 3He surface exposure ages show that David Glacier experienced rapid thinning up to 2 m/yr during the mid-Holocene (~ 6,500 years ago). Thinning stabilised at 6 kyr, suggesting initial formation of the Drygalski Ice Tongue at this time. Our work, along with terrestrial cosmogenic nuclide records from adjacent glaciers, shows simultaneous glacier thinning in this sector of the Transantarctic Mountains occurred ~ 3 kyr after the retreat of marine-based grounded ice in the western Ross Embayment. The timing and rapidity of the reconstructed thinning at David Glacier is similar to reconstructions in the Amundsen and Weddell embayments. In order to identify the potential causes of these rapid changes along the David Glacier, we use a glacier flow line model designed for calving glaciers and compare modelled results against our geological data. We show that glacier thinning and marine-based grounding line retreat is initiated by interactions between enhanced sub-ice shelf melting and reduced lateral buttressing, leading to Marine Ice Sheet Instability. Such rapid glacier thinning events are not captured in continental or sector-scale numerical modelling reconstructions for this period. Together, our chronology and modelling suggest a ~ 2,000-year period of dynamic thinning in the recent geological past.

Beryllium-10 (10Be) in quartz represents the most common in situ cosmogenic nuclide used for quantifying Earth-surface processes, primarily due to the prevalence of quartz in the Earth's crust. However many landscapes lack quartz-bearing rocks, thus other nuclide-mineral pairs are required for geochronometric and geomorphic applications. Here we describe the successful isolation and measurement of in situ 10Be concentrations in pyroxene from two mafic sample sets: (i) andesite boulders of the Murimotu Formation debris avalanche on Mt. Ruapehu, New Zealand, and (ii) dolerite cobbles deposited in a ∼100 m vertical transect at Mt. Gran by Mackay Glacier, Antarctica. Precise radiocarbon age constraint of the New Zealand site provides further geological constraint of the reference (at sea level and high latitude) 10Be production rate in pyroxene, which we find to be indistinguishable from a previous estimate. Combining our results with previous data yields a reference production rate of 3.2 ± 0.8 at. g−1 yr−1 (n=5; ‘Lm’ scaling). Application of this rate to the glacial cobbles at Mackay Glacier yields a relatively coherent chronology of ice surface lowering between ∼14 and 6 ka, which is broadly consistent with a well-constrained quartz-based 10Be chronology from nearby nunataks. Improving the viability of in situ 10Be for geological applications in mafic domains requires increased analytical precision beyond current levels. This improvement may be best achieved by further modification of the quartz-based methodologies for 10Be purification, in order to better handle the high cationic contaminant loads of ferromagnesian minerals. In addition, further 10Be measurements from suitable mafic sedimentary deposits with independent age control (e.g. existing cosmogenic 3He calibration sites) will help to refine estimates of the reference production rate.

Oceanic and atmospheric warming were dominant controls of past ice sheet retreat in Antarctica’s largest catchment., Modern observations appear to link warming oceanic conditions with Antarctic ice sheet grounding-line retreat. Yet, interpretations of past ice sheet retreat over the last deglaciation in the Ross Embayment, Antarctica’s largest catchment, differ considerably and imply either extremely high or very low sensitivity to environmental forcing. To investigate this, we perform regional ice sheet simulations using a wide range of atmosphere and ocean forcings. Constrained by marine and terrestrial geological data, these models predict earliest retreat in the central embayment and rapid terrestrial ice sheet thinning during the Early Holocene. We find that atmospheric conditions early in the deglacial period can enhance or diminish ice sheet sensitivity to rising ocean temperatures, thereby controlling the initial timing and spatial pattern of grounding-line retreat. Through the Holocene, however, grounding-line position is much more sensitive to subshelf melt rates, implicating ocean thermal forcing as the key driver of past ice sheet retreat.

The extent, basal conditions and retreat history of a Loch Lomond Stadial glacier are reconstructed based on detailed geomorphological and sedimentological assessment. We present new evidence from the vicinity of Coire Ardair that supports the former existence of a warm-based, locally-fed valley glacier, with probable cold-based ice on the surrounding plateau. This is broadly consistent with modelled creep-dominated flow in the upper catchment and sliding-dominated flow throughout much of the valley. A dense suite of moraines, primarily formed in ice-marginal environments, records a multi-phase recessional history: (1) active and oscillatory retreat; (2) a prolonged ice stillstand; (3) partial ice stagnation with occasional minor readvances; (4) increased oscillatory retreat with a substantial readvance event; and (5) rapid and uninterrupted retreat. We propose that a Coire Ardair glacier responded to sub-centennial scale climate fluctuations, possibly associated with the periodic delivery of warmer air masses to the region, rather than to a single, prominent shift in climate.

Quantifying the contribution of the East Antarctic Ice Sheet (EAIS) to sea-level rise during the last deglaciation is complicated by the limited opportunities to constrain ice-sheet models. The nunatak, Escalade Peak, provides a gauge for past ice surface elevation changes and behaviour throughout the last glacial cycle. Geomorphological mapping, geological evidence and 10Be cosmogenic-nuclide exposure dating at Escalade Peak, provide new constraints on the ice surface history of the Skelton Neve since the Last Glacial Maximum (LGM). An elevation transect from the eastern margin of Escalade Peak indicates that the ice surface of the Skelton Neve was at least 50 m and perhaps >120 m higher than present during the LGM. In contrast, surface-exposure ages from a suite of inner moraines (blue-ice moraines) adjacent to Escalade Peak do not provide independent ice surface elevation constraints, but may provide an indirect constraint on the timing of thinning due to exhumation-ablation processes. Maximum simple exposure ages from the inner moraines suggest ice surface ablation was initiated by 19.2 ka, but the majority of ice surface lowering at Escalade Peak likely occurred after ∼15 ka and reached the present-day ice level at ∼6 ka. These findings suggest that slow flowing inland sites of EAIS outlet glaciers, such as southern Skelton Neve, experienced minimal ice surface elevation change since the LGM and record an EAIS outlet glacier and western Ross Sea retreat signature rather than widespread Ross Sea retreat. The ice surface lowering is likely to have been in response to retreat of the grounded ice in the western Ross Embayment causing a reduction in buttressing of the Skelton Glacier and draw down into the Ross Sea.

Highlights • Exposure ages that constrain ice sheet thickness collated from an online database. • Thinning rates are reconstructed from 23 sites across Antarctica. • Palaeo-thinning rates are comparable to modern observations. • Wide-spread thinning during the Holocene, but after Meltwater Pulse 1A. Abstract Constraining Antarctic ice sheet evolution provides a way to validate numerical ice sheet models that aid predictions of sea-level rise. In this paper we collate cosmogenic exposure ages from exposed nunataks in Antarctica that have been used, or have the potential to be used, to constrain rates of thinning of the Antarctic Ice Sheets since the Last Glacial Maximum. We undertake quality control of the data and adopt a Bayesian approach to outlier detection. Past thinning rates are modelled by Monte Carlo linear regression analysis. We present thinning rates from 23 sites across Antarctica. The resulting data set is the first Antarctic-wide collation of past ice sheet thinning rates and provides an empirical starting point for future model-data comparisons. Palaeo-thinning rates are spatially variable with high rates appearing to correlate to areas of contemporary rapid changes. On centennial timescales past thinning rates are comparable to modern day observations implying that modern day thinning has the potential to persist for centuries in numerous parts of Antarctica. The onset of abrupt thinning from all sites post-dates Meltwater Pulse 1A suggesting that its source region(s) are distal to areas where exposure age constraints on ice surface geometry exist.

Geological records of past glacier extent can yield important constraints on the timing and magnitude of pre-historic climate change. Here we present a cosmogenic 3He moraine chronology from Mt. Ruapehu in central North Island, New Zealand that records fluctuations of New Zealand's northernmost glaciers over the last 14,000 years. In the Mangaehuehu Glacier catchment, exposure ages and geomorphic observations at our most distal moraine site, relative to the current glacier margin, show that glacial retreat during the last glacial termination was interrupted by one or more readvances between 14 and 11 ka. Further up-valley, moraine boulders deposited at 4.5 ka indicate that glaciers were more extensive during the mid-Holocene than at any later time. The youngest exposure ages in our study, 150–450 years, constrain the formation of moraines situated within 500 m from the present-day Mangaehuehu Glacier terminus. These ages are in agreement with historic sources that show the glacier close to these moraine limits approximately 100 years ago. Overall our data show that glaciers on Mt. Ruapehu retreated c. 3 km since 14 ka, in response to a net rise in the local equlibrium line altitude of c. 400 m. The magnitude of post-industrial snowline rise and glacier retreat on Mt. Ruapehu is also consistent with instrumental temperature data that record approximately 0.9 ∘C of warming during the 20th century.

We present a diagnostic glacier flowline model parameterized and constrained by new velocity data from ice-surface GPS installations and speckle tracking of TerraSAR-X satellite images, newly acquired airborne-radar data, and continental gridded datasets of topography and geothermal heat flux, in order to better understand two outlet glaciers of the East Antarctic ice sheet. Our observational data are employed as primary inputs to a modelling procedure that first calculates the basal thermal regime of each glacier, then iterates the basal sliding coefficient and deformation rate parameter until the fit of simulated to observed surface velocities is optimized. We find that the two glaciers have both frozen and thawed areas at their beds, facilitating partial sliding. Glacier flow arises from a balance between sliding and deformation that fluctuates along the length of each glacier, with the amount of sliding typically varying by up to two orders of magnitude but with deformation rates far more constant. Beardmore Glacier is warmer and faster-flowing than Skelton Glacier, but an up-glacier deepening bed at the grounding line, coupled with ice thicknesses close to flotation, lead us to infer a greater vulnerability of Skelton Glacier to grounding-line recession if affected by ocean-forced thinning and concomitant acceleration.

The East Antarctic Ice Sheet (EAIS) is the largest continental ice mass on Earth, and documenting its evolution since the Last Glacial Maximum (LGM) is important for understanding its present-day and future behaviour. As part of a community effort, we review geological evidence from East Antarctica that constrains the ice sheet history throughout this period (similar to 30,000 years ago to present). This includes terrestrial cosmogenic nuclide dates from previously glaciated regions, C-14 chronologies from glacial and post-glacial deposits onshore and on the continental shelf, and ice sheet thickness changes inferred from ice cores and continental-scale ice sheet models. We also include new C-14 dates from the George V Land Terre Adelie Coast shelf. We show that the EAIS advanced to the continental shelf margin in some parts of East Antarctica, and that the ice sheet characteristically thickened by 300-400 m near the present-day coastline at these sites. This advance was associated with the formation of low-gradient ice streams that grounded at depths of >1 km below sea level on the inner continental shelf. The Lambert/Amery system thickened by a greater amount (800 m) near its present-day grounding zone, but did not advance beyond the inner continental shelf. At other sites in coastal East Antarctica (e.g. Bunger Hills, Larsemann Hills), very little change in the ice sheet margin occurred at the LGM, perhaps because ice streams accommodated any excess ice build up, leaving adjacent, ice-free areas relatively unaffected. Evidence from nunataks indicates that the amount of ice sheet thickening diminished inland at the LGM, an observation supported by ice cores, which suggest that interior ice sheet domes were similar to 100 m lower than present at this time. Ice sheet recession may have started similar to 18,000 years ago in the Lambert/Amery glacial system, and by similar to 14,000 years ago in Mac.Robertson Land. These early pulses of deglaciation may have been responses to abrupt sea-level rise events such as Meltwater Pulse la, destabilising the margins of the ice sheet. It is unlikely, however, that East Antarctica contributed more than similar to 1 m of eustatic sea-level equivalent to post-glacial meltwater pulses. The majority of ice sheet recession occurred after Meltwater Pulse la, between similar to 12,000 and similar to 6000 years ago, during a period when the adjacent ocean warmed significantly. Large tracts of East Antarctica remain poorly studied, and further work is required to develop a robust understanding of the LGM ice sheet expansion, and its subsequent contraction. Further work will also allow the contribution of the EAIS to post-glacial sea-level rise, and present-day estimates of glacio-isostatic adjustment to be refined.