Your search

Understanding long-term climate variability in the high latitudes of the Southern Hemisphere is critical due to the key role of the Southern Ocean in the global climate system. However, sparse observations (in space and time) coupled with strong internal variability limit our ability to interpret the origin of recent changes, and their longer-term context. Here we present a dynamically consistent reconstruction of the Antarctic atmosphere and Southern Ocean from 1700 to 2023. We first use data assimilation (DA)-based Antarctic atmospheric reanalyses that combine instrumental observations (1958–2023) and paleoclimate proxies (1700–2000) with Earth System Models to reconstruct key surface climate fields. We then drive a global ocean–sea-ice model with this atmospheric reanalysis to simulate historical ocean conditions, including temperature, salinity, currents, and sea-ice-related variables at 1° resolution. This reconstruction provides the first long-term physically consistent dataset of Antarctic atmosphere–ocean variability, suitable for studying low-frequency climate variability, evaluating climate models, and potentially driving regional atmospheric and ocean models as well as ice sheet models.

The global overturning circulation (GOC) is the largest scale component of the ocean circulation, associated with a global redistribution of key tracers such as heat and carbon. The GOC generates decadal to millennial climate variability, and will determine much of the long-term response to anthropogenic climate perturbations. This review aims at providing an overview of the main controls of the GOC. By controls, we mean processes affecting the overturning structure and variability. We distinguish three main controls: mechanical mixing, convection, and wind pumping. Geography provides an additional control on geological timescales. An important emphasis of this review is to present how the different controls interact with each other to produce an overturning flow, making this review relevant to the study of past, present and future climates as well as to exoplanets’ oceans.

Enhanced Antarctic ice sheet mass loss yields ocean surface freshening, cooling and sea ice expansion, which result in changes in the atmospheric conditions. Using the Southern Ocean Freshwater Input from Antarctica (SOFIA) multi-model ensemble, we study the atmospheric response to a 100-year idealized freshwater release of 0.1 Sv. All models simulate a surface-intensified tropospheric cooling and lower-stratospheric warming south of 35°S. Tropospheric cooling is attributed to sea ice expansion and the associated albedo enhancement in winter and a colder sea surface in summer. This cooling yields a downward displacement of the tropopause, reduced stratospheric water vapor content and ultimately warming around 200 hPa. An enhanced southward eddy heat flux explains warming at 10?100 hPa during austral winter. Despite a temporally (and spatially) uniform prescribed freshwater flux, a prominent sea ice seasonal cycle and atmosphere dynamics result in a distinct seasonal pattern in the occurrence and magnitude of the temperature responses.

In recent years, the Antarctic sea ice has experienced major changes, which are neither well understood nor adequately reproduced by Earth system models. To support model development with an aim to improve Antarctic sea ice and upper-ocean predictions, the impacts of updating the sea ice model and the atmospheric forcing are investigated. In the new MetROMS-UHel-v1.0 (henceforth MetROMS-UHel) ocean–sea ice model, the sea ice component has been updated from CICE5 to CICE6, and the forcing has been updated from ERA-Interim (ERAI) to ERA5 reanalyses. The two versions of MetROMS evaluated in this study use a version of the regional ROMS ocean model including ice shelf cavities. We find that the update of CICE (Community Ice CodE) and ERA reduced the negative bias of the sea ice area in summer. However, the sea ice volume decreases after the CICE update but increases when the atmospheric forcing is updated. As a net result after both updates, the modelled sea ice becomes thinner and more deformed, particularly near the coast. The ROMS ocean model usually yielded a deeper ocean mixed layer compared to observations. Using ERA5, the situation was slightly improved. The update from CICE5 to CICE6 resulted in a fresher coastal ocean due to a smaller salt flux from sea ice to the ocean. In the ice shelf cavities, the modelled melt rates are generally underestimated compared with observations, with the largest underestimation coming from the ice shelves in the too cold Amundsen and Bellingshausen seas as well as from the Australian sector in East Antarctica. These identified sea ice and oceanic changes vary seasonally and regionally. By determining sea ice and oceanic changes after the model and forcing updates and evaluating them against observations, this study informs modellers on improvements and aspects requiring attention with potential model adjustments.

Observations of water stable isotopes in Antarctic surface snow, precipitation and water vapor are key for improving our understanding of the atmospheric water cycle and past climate reconstructions from ice cores. In this study, we use isotopic observations in Antarctica to assess the skill of the isotope-enabled atmospheric general circulation model LMDZ6, nudged to ERA5 above the boundary layer (1980?2023 period). The model has no significant bias for time-mean temperature and snow accumulation over the ice sheet. Sensitivity test on parameterized supersaturation strength highlights its opposite effect on precipitation ${\delta }^{18}$O and d-excess. Selecting an intermediate supersaturation strength resulted in a minimal bias for surface snow ${\delta }^{18}$O across the continent, with a reduced but systematic positive bias in surface snow d-excess ( ${\sim} $5?). We then assessed seasonal and diurnal isotope variability with daily precipitation and continuous vapor isotopes at Dumont d?Urville (DDU, coastal station) and Concordia (inland station). On a seasonal scale, LMDZ6iso accurately reproduces the seasonal cycle of precipitation ${\delta }^{18}$O and d-excess at both stations. Moving from statistical evaluation to physical analysis, we use the individual process contributions to boundary-layer water vapor isotopes to identify the main drivers controlling the clear-sky isotopic daily cycles. At Concordia, daily isotope variations are mainly driven by surface sublimation, whereas at DDU they are driven by surface sublimation and advection by the katabatic flow. Our results suggest that to further improve water isotopes in LMDZ6iso, fractionation during surface sublimation should be included and fractionation at condensation for low temperature should be better constrained.

Over the last decade, the Southern Ocean has experienced episodes of severe sea ice area decline. Abrupt events of sea ice loss are challenging to predict, in part due to incomplete understanding of processes occurring at the scale of individual ice floes. Here, we use high-resolution altimetry (ICESat-2) to quantify the seasonal life cycle of floes in the perennial sea ice pack of the Weddell Sea. The evolution of the floe chord distribution (FCD) shows an increase in the proportion of smaller floes between November and February, which coincides with the asymmetric melt–freeze cycle of the pack. The freeboard ice thickness distribution (fITD) suggests mirrored seasonality between the western and southern sections of the Weddell Sea ice cover, with an increasing proportion of thicker floes between October and March in the south and the opposite in the west. Throughout the seasonal cycle, there is a positive correlation between the mean chord length of floes and their average freeboard thickness. Composited floe profiles reveal that smaller floes are more vertically round than larger floes and that the mean roundness of floes increases during the melt season. These results show that regional differences in ice concentration and type at larger scales occur in conjunction with different behaviors at the small scale. We therefore suggest that floe-derived metrics obtained from altimetry could provide useful diagnostics for floe-aware models and improve our understanding of sea ice processes across scales.

Antarctic sea ice has changed significantly over the past four decades; yet limited understanding of fundamental processes, including its seasonal cycle, hinders our ability to interpret these changes. Here, we examine the processes determining the moment when sea ice locally disappears each spring, defined as the retreat date, using satellite observations over 1994?2020. We find that climatological retreat date is driven by sea ice melt in most of the seasonal ice zone and strongly constrained by the seasonal maximum ice thickness. Ice removal due to drifting ice export predominantly drives retreat only in coastal polynyas. At interannual timescales, retreat date anomalies are also preconditioned by prior maximum ice thickness, which affects melt-driven spring ice loss through the ice-albedo feedback, though this effect appears limited to specific regions. Winds emerge as a primary driver of interannual variability in the retreat date, influencing both drift- and melt-related spring ice removal processes.

During the Quaternary, ice sheets experienced several retreat–advance cycles, strongly influencing climate patterns. In order to properly simulate these phenomena, it is preferable to use physics-based models instead of parameterizations to estimate the surface mass balance (SMB), which strongly influences the evolution of the ice sheet. To further investigate the potential of these SMB models, this work evaluates the BErgen Snow SImulator (BESSI), a multi-layer snow model with high computational efficiency, as an alternative to providing the SMB for the Earth system model iLOVECLIM for multi-millennial simulations as in paleostudies. We compare the behaviors of BESSI and insolation temperature melt (ITM), an existing SMB scheme of iLOVECLIM during the Last Interglacial (LIG). Firstly, we validate the two SMB models using the regional climate model Mod- èle Atmosphérique Régional (MAR) as forcing and reference for the present-day climate over the Greenland and Antarctic ice sheets. The evolution of the SMB over the LIG (130–116 ka) is computed by forcing BESSI and ITM with transient climate forcing obtained from iLOVECLIM for both ice sheets. For present-day climate conditions, both BESSI and ITM exhibit good performance compared to MAR despite a much simpler model setup. While BESSI performs well for both Antarctica and Greenland for the same set of parame- ters, the ITM parameters need to be adapted specifically for each ice sheet. This suggests that the physics embedded in BESSI allows better capture of SMB changes across varying climate conditions, while ITM displays a much stronger sen- sitivity to its tunable parameters. The findings suggest that BESSI can provide more reliable SMB estimations for the iLOVECLIM framework to improve the model simulations of the ice sheet evolution and interactions with climate for multi-millennial simulations.

Abstract Basal melting of ice shelves is fundamental to Antarctic ice sheet mass loss, yet direct observations remain sparse. We present the first year-round melt record (2017-2021) from a phase-sensitive radar on Fimbulisen, one of the fastest flowing ice shelves in Dronning Maud Land, East Antarctica. The observed long-term mean ablation rate at 350 m depth below the central ice shelf was 1.0 ± 0.5 m yr?1, marked by substantial sub-weekly variability ranging from 0.4 to 3.5 m yr?1. 36-h filtered basal melt rate fluctuations closely align with ocean velocity. On seasonal time scales, melt rates peak during austral spring to autumn (September-March), driven by both elevated ocean velocities and thermal driving near the base. The combined effect of thermal driving and current speed explains the majority of the melt rate variability (r = 0.84), highlighting the dominant role of shear-driven turbulence. This relationship enables parameterization of melt rates for the decade-long ocean record (2010?2021), although deviations appear under low and high forcing conditions. Both observed and parameterized melt rates show similar yearly mean magnitudes compared to satellite-derived melt rates but with a tenfold lower seasonal amplitude and a 3-month delay in seasonality. These detailed concurrent ice?ocean observations provide essential validation data for remote sensing and numerical models that aim to quantify and project ice-shelf response to a change in ocean forcing. In situ measurements and continued monitoring are crucial for accurately assessing and modeling future basal melt rates, and for understanding the complex dynamics driving ice-shelf stability and sea-level change.

Meltwater ponding along the margins of Antarctica poses a threat to ice shelf stability, increasing the risk of accelerated inland ice mass loss. Understanding the key drivers of supraglacial lake formation is therefore essential for assessing the vulnerability and future stability of Antarctic ice shelves. In this study, we combine high-resolution simulation from the regional climate model Modèle Atmosphérique Régional (MAR) with satellite-derived records of supraglacial lakes in coastal Dronning Maud Land to investigate the role of topographic downslope winds on spatial lake distribution. We find that persistent katabatic winds and episodic foehn winds are key controls on the observed regional patterns of lakes. Katabatic winds, most persistent in eastern Dronning Maud Land, exert a sustained impact near grounding zones through snow erosion, scouring and sublimation. Foehn winds predominantly affect ice shelves on the lee (western) side of large ice rises and promontories, causing considerable surface warming. While these downslope winds directly contribute to surface melt and ponding during summer, they also precondition the surface year-round through wind-driven warming and sublimation. Statistical analysis of downslope wind exposure further allows us to identify other Antarctic ice shelves that may become vulnerable to future ponding as firn retention capacity is diminished.

The Quaternary climate is characterized by glacial–interglacial cycles, with the most recent transition from the last glacial maximum to the present interglacial (the last deglaciation) occurring between ∼ 21 and 9 ka. While the deglacial warming at high southern latitudes is mostly in phase with atmospheric CO2 concentrations, some proxy records indicate that the onset of the warming occurred before the CO2 increase. In addition, high southern latitudes exhibit a cooling event in the middle of the deglaciation (15–13 ka) known as the “Antarctic Cold Reversal”. In this study, we analyse transient simulations of the last deglaciation performed with six different climate models as part of the 4th phase of the Paleoclimate Modelling Intercomparison Project (PMIP4) to understand the processes driving high-southern-latitude surface temperature changes. As the protocol of the last deglaciation sets the choice of freshwater forcing as flexible, the freshwater forcing is different in each model, thus complicating the multi-model comparison. While proxy records from West Antarctica and the Pacific sector of the Southern Ocean suggest the presence of an early warming before 18 ka, only half the models show a significant warming at this time (∼ 1 °C or ∼ 10 % of the total deglacial warming). All models simulate a major warming during Heinrich Stadial 1 (18–15 ka), concurrent with the CO2 increase and with a weakening of the Atlantic Meridional Overturning Circulation (AMOC) in some models. However, the simulated Heinrich Stadial 1 warming over Antarctica is smaller than the one suggested from ice core data. During the Antarctic Cold Reversal, simulations with an abrupt AMOC strengthening exhibit a high-southern-latitude cooling of 1 to 2 °C, in relative agreement with proxy records, while simulations with rapid North Atlantic meltwater input exhibit a warming. Using simple models to extract the relative AMOC contribution, we find that all climate models simulate a high-southern-latitude cooling in response to an AMOC increase with a response timescale of several hundred years, suggesting the choice of the North Atlantic meltwater forcing substantially affects high-southern-latitude temperature changes. Thus, further work needs to be carried out to reconcile the deglacial AMOC evolution with the Northern Hemisphere ice sheet disintegration and associated meltwater input. Finally, all simulations exhibit only minimal changes in Southern Hemisphere westerlies and Southern Ocean meridional circulation during the last deglaciation. Improved understanding of the processes impacting Southern Hemisphere atmospheric and oceanic circulation changes accounting for deglacial atmospheric CO2 increase is needed.

Water stable isotope records in polar ice cores have been largely used to reconstruct past local temperatures and other climatic information such as evaporative source region conditions of the precipitation reaching the ice core sites. However, recent studies have identified post-depositional processes taking place at the ice sheet's surface, modifying the original precipitation signal and challenging the traditional interpretation of ice core isotopic records. In this study, we use a combination of existing and new datasets of precipitation, snow surface, and subsurface isotopic compositions (δ18O and deuterium excess (d-excess)); meteorological parameters; ERA5 reanalyses; outputs from the isotope-enabled climate model ECHAM6-wiso; and a simple modelling approach to investigate the transfer function of water stable isotopes from precipitation to the snow surface and subsurface at Dome C in East Antarctica. We first show that water vapour fluxes at the surface of the ice sheet result in a net annual sublimation of snow, from 3.1 to 3.7 mm w.e. yr−1 (water equivalent) between 2018 and 2020, corresponding to 12 % to 15 % of the annual surface mass balance. We find that the precipitation isotopic signal cannot fully explain the mean, nor the variability in the isotopic composition observed in the snow, from annual to intra-monthly timescales. We observe that the mean effect of post-depositional processes over the study period enriches the snow surface in δ18O by 3.0 ‰ to 3.3 ‰ and lowers the snow surface d-excess by 3.4 ‰ to 3.5 ‰ compared to the incoming precipitation isotopic signal. We also show that the mean isotopic composition of the snow subsurface is not statistically different from that of the snow surface, indicating the preservation of the mean isotopic composition of the snow surface in the top centimetres of the snowpack. This study confirms previous findings about the complex interpretation of the water stable isotopic signal in the snow and provides the first quantitative estimation of the impact of post-depositional processes on the snow isotopic composition at Dome C, a crucial step for the accurate interpretation of isotopic records from ice cores.

The global overturning circulation (GOC) is the largest scale component of the ocean circulation, associated with a global redistribution of key tracers such as heat and carbon. The GOC generates decadal to millennial climate variability, and will determine much of the long-term response to anthropogenic climate perturbations. This review aims at providing an overview of the main controls of the GOC. By controls, we mean processes affecting the overturning structure and variability. We distinguish three main controls: mechanical mixing, convection, and wind pumping. Geography provides an additional control on geological timescales. An important emphasis of this review is to present how the different controls interact with each other to produce an overturning flow, making this review relevant to the study of past, present and future climates as well as to exoplanets’ oceans.

Sea ice is a composite solid material that sustains large fracture features at scales from meters to kilometres. These fractures can play an important role in coupled atmosphere-ocean processes. To model these features, brittle sea ice physics, via the Brittle-Bingham-Maxwell (BBM) rheology, has been implemented in the Lagrangian neXt generation Sea Ice Model (neXtSIM). In Arctic-only simulations, the BBM rheology has shown a capacity to represent observationally consistent sea ice fracture patterns and breakup across a wide range of time and length scales. Still, it has not been tested whether this approach is suitable for the modeling of Antarctic sea ice, which is thinner and more seasonal compared to Arctic sea ice, and whether the ability to reproduce sea ice fractures has an impact on simulating Antarctic sea ice properties. Here, we introduce a new 50-km grid-spacing Antarctic configuration of neXtSIM, neXtSIM-Ant, using the BBM rheology. We evaluate this simulation against observations of sea ice extent, drift, and thickness and compare it with identically-forced neXtSIM simulations that use the standard modified Elastic-Visco-Plastic (mEVP) rheology. In general, using BBM results in thicker sea ice and an improved correlation of sea ice drift with observations than mEVP. We suggest that this is related to short-duration breakup events caused by Antarctic storms that are not well-simulated in the viscous-plastic model.

The Antarctic Circumpolar Current (ACC) is the world’s strongest ocean current and plays a disproportionate role in the climate system due to its role as a conduit for major ocean basins. This current system is linked to the ocean’s vertical overturning circulation, and is thus pivotal to the uptake of heat and CO2 in the ocean. The strength of the ACC has varied substantially across warm and cold climates in Earth’s past, but the exact dynamical drivers of this change remain elusive. This is in part because ocean models have historically been unable to adequately resolve the small-scale processes that control current strength. Here, we assess a global ocean model simulation which resolves such processes to diagnose the impact of changing thermal, haline and wind conditions on the strength of the ACC. Our results show that, by 2050, the strength of the ACC declines by ∼20% for a high-emissions scenario. This decline is driven by meltwater from ice shelves around Antarctica, which is exported to lower latitudes via the Antarctic Intermediate Water. This process weakens the zonal density stratification historically supported by surface temperature gradients, resulting in a slowdown of sub-surface zonal currents. Such a decline in transport, if realised, would have major implications on the global ocean circulation.

Dynamical modeling is widely utilized for Antarctic sea ice prediction. However, the relative impact of initializing different model components remains unclear. We compare three sets of hindcasts of the Norwegian Climate Prediction Model (NorCPM), which are initialized by ocean, ocean/sea-ice, or atmosphere data and referred to as the OCN, OCNICE, and ATM hindcasts hereafter. The seasonal cycle of sea ice extent (SIE) in the ATM reanalysis shows a slightly better agreement with observations than the OCN and OCNICE reanalyzes. The trends of sea ice concentration (SIC) in the OCN and OCNICE reanalyzes compare well to observations, but the ATM reanalysis is poor over the western Antarctic. The OCNICE reanalysis yields the most accurate estimation of sea ice variability, while the OCN and ATM reanalyzes are comparable. Evaluation of the hindcasts reveals the predictive skill varies with region and season. Austral winter SIE of the western Antarctic can be skillfully predicted 12 months ahead, while the predictive skill in the eastern Antarctic is low. Austral winter SIE predictability can be largely attributed to high sea surface temperature predictability, thanks to skillful initialization of ocean heat content. The ATM hindcast from July or October performs best due to the effective initialization of sea-ice thickness, which enhances prediction skills until early austral summer via its long memory. Meanwhile, the stratosphere-troposphere coupling contributes to the prediction of springtime. The comparable skill between the OCN and OCNICE hindcasts implies limited benefits from SIC data on prediction when using ocean data.

Antarctic sea ice has exhibited significant variability over the satellite record, including a period of prolonged and gradual expansion, as well as a period of sudden decline. A number of mechanisms have been proposed to explain this variability, but how each mechanism manifests spatially and temporally remains poorly understood. Here, we use a statistical method called low-frequency component analysis to analyze the spatiotemporal structure of observed Antarctic sea ice concentration variability. The identified patterns reveal distinct modes of low-frequency sea ice variability. The leading mode, which accounts for the large-scale, gradual expansion of sea ice, is associated with the Interdecadal Pacific Oscillation and resembles the observed sea surface temperature trend pattern that climate models have trouble reproducing. The second mode is associated with the central Pacific El Niño–Southern Oscillation (ENSO) and the Southern Annular Mode and accounts for most of the sea ice variability in the Ross Sea. The third mode is associated with the eastern Pacific ENSO and Amundsen Sea Low and accounts for most of the pan-Antarctic sea ice variability and almost all of the sea ice variability in the Weddell Sea. The third mode is also related to periods of abrupt Antarctic sea ice decline that are associated with a weakening of the circumpolar westerlies, which favors surface warming through a shoaling of the ocean mixed layer and decreased northward Ekman heat transport. Broadly, these results suggest that climate model biases in long-term Antarctic sea ice and large-scale sea surface temperature trends are related to each other and that eastern Pacific ENSO variability is a key ingredient for abrupt Antarctic sea ice changes.

Future climate and sea level projections depend sensitively on the response of the Antarctic Ice Sheet to ocean-driven melting and the resulting freshwater fluxes into the Southern Ocean. Circumpolar Deep Water (CDW) transport across the Antarctic continental shelf and into cavities beneath ice shelves is increasingly recognised as a crucial heat source for ice shelf melt. Quantifying past changes in the temperature of CDW is therefore of great benefit for modelling ice sheet response to past warm climates, for validating paleoclimate models, and for putting recent and projected changes in CDW temperature into context. Here we compile the available bottom water temperature reconstructions representative of CDW over the past 800 kyr. Estimated interglacial warming reached anomalies of +0.6 +/- 0.4 degrees C (MIS 11) and +0.5 +/- 0.5 degrees C (MIS 5) relative to present. Glacial cooling typically reached anomalies of ca. -1.5 to -2 degrees C, therefore maintaining positive thermal forcing for ice shelf melt even during glacials in the Amundsen Sea region of West Antarctica. Despite high variance amongst a small number of records and poor (4 kyr) temporal resolution, we find persistent and close relationships between our estimated CDW temperature and Southern Ocean sea surface temperature, Antarctic surface air temperature, and global deep-water temperature reconstructions at glacial-cycle timescales. Given the important role that CDW plays in connecting the world's three main ocean basins and in driving Antarctic Ice Sheet mass loss, additional temperature reconstructions targeting CDW are urgently needed to increase temporal and spatial resolution and to decrease uncertainty in past CDW temperatures - whether for use as a boundary condition, for model validation, or for understanding past oceanographic changes.

Supraglacial lakes on Antarctic ice shelves can have far-reaching implications for ice-sheet stability, highlighting the need to understand their dynamics, controls and role in the ice-sheet mass budget. We combine a detailed satellite-based record of seasonal lake evolution in Dronning Maud Land with a high-resolution simulation from the regional climate model Modèle Atmosphérique Régional to identify drivers of lake variability between 2014 and 2021. Correlations between summer lake extents and climate parameters reveal complex relationships that vary both in space and time. Shortwave radiation contributes positively to the energy budget during summer melt seasons, but summers with enhanced longwave radiation are more prone to surface melting and ponding, which is further enhanced by advected heat from summer precipitation. In contrast, previous winter precipitation has a negative effect on summer lake extents, presumably by increasing albedo and pore space, delaying the accumulation of meltwater. Downslope katabatic or föhn winds promote ponding around the grounding zones of some ice shelves. At a larger scale, we find that summers during periods of negative southern annular mode are associated with increased ponding in Dronning Maud Land. The high variability in seasonal lake extents indicates that these ice shelves are highly sensitive to future warming or intensified extreme events.

We are in a period of rapidly accelerating change across the Antarctic continent and Southern Ocean, with land ice loss leading to sea level rise and multiple other climate impacts. The ice-ocean interactions that dominate the current ice loss signal are a key underdeveloped area of knowledge. The paucity of direct and continuous observations leads to high uncertainty in the glaciological, oceanographic and atmospheric fields required to constrain ice-ocean interactions, and there is a lack of standardised protocols for reconciling observations across different platforms and technologies and modelled outputs. Funding to support observational campaigns is under increasing pressure, including for long-term, internationally coordinated monitoring plans for the Antarctic continent and Southern Ocean. In this Practice Bridge article, we outline research priorities highlighted by the international ice-ocean community and propose the development of a Framework for UnderStanding Ice-Ocean iNteractions (FUSION), using a combined observational-modelling approach, to address these issues. Finally, we propose an implementation plan for putting FUSION into practice by focusing first on an essential variable in ice-ocean interactions: ocean-driven ice shelf melt.