Snow atop Antarctic sea ice plays a critical role in modulating sea ice growth, surface energy balance, and ocean–atmosphere interactions. However, it also introduces substantial uncertainty into satellite altimeter-based sea ice thickness (SIT) estimates. Ku-band radar altimeters, such as CryoSat-2 (CS-2), are often processed using threshold-based retrackers that implicitly assume the maximum radar intensity return originates near the snow–ice interface. In practice, layered snowpacks featuring wet snow, brine infiltration, and ice lenses can shift the primary scattering contribution upward, leading to overestimated freeboard and higher SIT estimates. In Part I of this study, we used physically based waveform decomposition to quantify the vertical distribution of radar backscatter under Weddell Sea conditions. Building on these insights, Part II introduces an optimized threshold first-maximum retracker algorithm (TFMRA) for CS-2, tuned using airborne observations from NASA’s Operation IceBridge (OIB) over the Weddell Sea. We identify a 70% retracking threshold that minimizes freeboard bias and improves consistency with independent observations. Applying this snow-aware retracker to 46 CRYO2ICE collocated tracks (2020–2022), we retrieve snow depth from the ICESat-2 and CS-2 freeboard difference and reduce mean SIT by ∼ 0.1 m relative to the ESA Baseline-E product in the southern Weddell Sea. Monte-Carlo (MC) perturbations of OIB snow retrievals, combined with CS-2 threshold-sensitivity tests, indicate an intrinsic ∼ 0.2 m uncertainty in OIB snow depth and a similar lower-bound CRYO2ICE snow-depth uncertainty of ∼ 0.21-0.24 m at 10 km scales. Our results offer practical guidance for altimeter algorithm development and are directly relevant to upcoming dual-frequency radar missions such as ESA’s CRISTAL. • A 70% TFMRA threshold reduces radar freeboard bias, and improves Antarctic CryoSat-2 SIT. • Snow-aware CS-2 retracking tuned with OIB and CRYO2ICE sharpens snow depth and SIT. • OIB and CRYO2ICE snow depth errors ≥ 0.2 m constrain altimetry and guide CRISTAL.

Abstract. The Arctic during the Last Interglacial period (LIG) was considered warmer than it is today. The previous study points to a large difference in the degree of simulated annual-mean Arctic warming among models. While recent reconstructions suggest the disappearance of summer sea ice in the Arctic at the LIG, many climate models fail to capture this feature. It is thus essential to investigate sources of uncertainty in climate models. The current study examines the impact of the temperature-cloud phase relationship. Sensitivity studies are conducted for the first time to explore the potential importance of this relationship in simulating the LIG climate. Two different cloud parameter sets are used for an atmosphere-ocean general circulation model with and without the dynamic vegetation feedback. The model with cloud parametrization that permits liquid water at lower temperatures and a larger fraction of supercooled liquid water at the same temperature simulates a warmer preindustrial (PI) climate, greater annual-mean Arctic warming at the LIG, and substantially reduced summer sea ice cover at the LIG. It is demonstrated that the low-level clouds play a crucial role in controlling the Arctic response via the greenhouse effect. The result indicates the importance of the temperature-cloud phase relationship in simulating the Arctic climate at the LIG. It also highlights the importance of accurately simulating modern sea ice thickness and representing the processes that affect the fraction of supercooled liquid water in clouds.

Abstract. It has been hypothesized that the Earth may have experienced snowball events in the past, during which its surface became completely covered with ice. Previous studies used general circulation models to investigate the onset and climate of such snowball events. Using the MIROC4m coupled atmosphere–ocean climate model, this study examined the changes in the oceanic circulation during the onset of a modern snowball Earth and elucidated their evolution to steady states under the snowball climate. Abruptly changing the solar constant to 94 % of its present-day value caused the modern Earth climate to turn into a snowball state after ∼ 1300 years and initiated rapid increase in sea ice thickness. During onset of the snowball, extensive sea ice formation and melting of sea ice in the mid-latitudes caused substantial freshening of surface waters and salinity stratification. By contrast, such salinity stratification was absent if the duration between the change in the solar flux and the snowball onset was short. After snowball onset, the global sea ice cover and the buildup of salinity stratification caused drastic weakening in the deep ocean circulation. However, the meridional overturning circulation resumed within several hundred years after the snowball onset because the density flux by sea ice production weakens the salinity stratification. While the evolution of the oceanic circulation would depend on the continental distribution and the evolution of continental ice sheets, our results highlight the gradual growth of sea ice and associated brine rejection are essential factors for the transient evolution of the oceanic circulation in the snowball events.

The Arctic sea ice cover and thickness have rapidly declined in recent years, with snow cover on sea ice playing a key role in driving this variability and trend. Arctic sea ice and snow properties also strongly regulate heat and momentum exchange between the ocean and atmosphere, making their accurate representation in climate models essential. Yet in situ observations of Arctic sea ice and snow thickness are scarce. The year-long Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition provided valuable, high-resolution measurements of these properties, but this dataset is short-term and localized compared to the observational products typically used for model evaluation. We examine whether free-running climate model simulations can be meaningfully compared to point observations, such as those from MOSAiC, to assess model performance. To address this, we employ multiple methodological approaches to generate representative seasonal cycles of simulated snow and sea ice thickness: a standard 30-year climatology, two proxy-year methods (based on sea ice area (SIA) and the Arctic Oscillation (AO) index), atmospherically nudged simulations, and a Monte Carlo random-year benchmark. We find that the SIA-based proxy method performs comparably to the 30-year climatology. In contrast, the AO-based proxy method reduces bias relative to the SIA method for snow thickness comparisons. However, both methods nevertheless fail to accurately reproduce the amplitude of the observed snow thickness cycle, suggesting unresolved processes in models. Overall, these findings show potential for snow and sea ice evaluation against localized measurements and demonstrate that proxy methods can provide viable alternatives when nudging or direct temporal overlaps are unavailable. Finally, this study highlights the need for an improved representation of modeled sea ice and snow processes to enhance the next generation of global climate models.

Abstract Solar radiation is the key energy input to the ocean. In the Arctic Ocean and its peripheral seas, the distribution of solar radiation is strongly modulated by the presence of sea ice. In this study, we combined satellite and model products to investigate solar radiation partitioning between reflection to the atmosphere, absorption in the ice, and transmission to the ocean over 1984–2024. We present total annual solar heat partitioning, relative contributions to energy deposition from ice and open water, and trends in large‐scale partitioning. The Arctic exhibited a decreasing trend in albedo (0.019 decade −1 ) due to decreasing sea ice areal coverage and thickness. Consequently, solar transmittance into the ocean increased by 0.031 decade −1 , resulting in an additional ∼300 MJ m −2 of heat input over 1984–2024. A brighter, warmer ocean contributes to Arctic Amplification and may alter the functioning of the Arctic marine ecosystem.

The Arctic is warming more than four times as fast as any other region of our planet. This warming has led a dramatic decline in seasonal sea-ice coverage and thickness and to a poleward extent of the Atlantic water. The Fram Strait is the only deep gateway of the Arctic Ocean where the West Spitsbergen Current transports the Atlantic water northward and the East Greenland Current the polar water southward. This is a highly dynamic oceanographic regime in the zone (central Fram Strait) between the two currents with its mixed water in combination with the semi-permanent sea-ice edge and large horizontal density gradients observed in the marginal ice zone. Long-term high-spatial resolution data on phytoplankton distribution and its community composition help to assess the impact of these physical changes on the biological processes in the surface waters of Fram Strait. In this study, we resolve the implications of changes in sea-ice conditions and physical hydrographic properties in Fram Strait to the chlorophyll-a concentration (Chl-a) of the whole phytoplankton community and of the major phytoplankton groups (PGs) contributing to its biomass. We extend the formerly collected total Chl-a (TChl-a) and PG Chl-a data sets of High-Performance Liquid Chromatography and of hyperspectral particulate absorption data sets from high-frequency underway spectrophotometry measurements to include the five recent East Greenland Sea expeditions from 2019 and 2021 to 2024. We adapt well-established methods to retrieve from the hyperspectral data as baseline for more high-spatial resolution TChl-a and PG Chl-a retrievals. Cross validation and independent validation confirm the robustness of our 2015 to 2024 data sets. We analyse the ship-based high-spatial resolution (around 300 m) continuous Fram Strait time series data along the ship transect of the eight expeditions. We observe diatoms and haptophytes, the key functional groups of the marine ecosystem, together with chlorophytes to represent most of the phytoplankton community. We identify a shift in spring blooms’ phytoplankton community composition responding to the distinct sea-ice thickness conditions in different years. We observe in the West Spitsbergen Current that the mid-summer TChl-a and the contribution of haptophytes to the phytoplankton community are much lower during the latest years (2022 and 2024) identified as cold as opposed to the warm (2015 to 2017) temperature anomaly years. We recommend continuing the underway spectrophotometry measurements in future and combine these long-term high-spatial resolution time series on phytoplankton community composition with satellite measurements to track the effects of the changes of the physical environment to global warming on the main primary producer in the Arctic Ocean. • High-spatial resolution Chl-a data, including phytoplankton groups’ contribution. • Spring bloom phytoplankton composition responds to sea-ice thickness conditions. • Chl-a is lower in mid-summer during cold opposed to warm temperature anomaly years. • Haptophytes contribute less during cold as opposed to warm anomaly years.

Sea ice thickness surveys with a drone-borne multi-frequency EM sensor

Ground-based GNSS-R dual path adaptive retrieval of sea ice thickness with spectral estimability criterion and dual-frequency fusion

Abstract Subseasonal prediction systems based on coupled atmosphere‐ice‐ocean models are vital tools for polar environmental forecasting. However, their predictive skill for Antarctic sea ice thickness (SIT), which is a critical variable governing atmosphere‐ocean energy and mass exchange, remains inadequately assessed. In this study, we evaluated Antarctic SIT prediction skill in four dynamical systems (China Meteorological Administration, Environment and Climate Change Canada, European Centre for Medium‐Range Weather Forecasts, and Météo‐France) participating in the Subseasonal to Seasonal Prediction Project using the Soil Moisture and Ocean Salinity satellite‐derived thin SIT retrievals. Results indicate that dynamical systems show limited overall skill against observation‐based benchmarks, primarily due to excessive initial errors. Removing climatological biases reduces errors, yielding skill in the Ross Sea comparable to the damped anomaly persistence prediction benchmark. Furthermore, the dynamical system surpassed the benchmark significantly from March to April, with three systems skillfully predicting region‐wide SIT anomaly evolution. Predictive skill for thin ice in the Weddell Sea and the coastal regions around Antarctica is a key driver of inter‐model skill differences. Overall, these results suggest that dynamical systems have the potential to surpass benchmark skill via improved SIT initialization, reduced climatological bias, and improved process representation of newly formed sea ice that enhances coastal thin‐ice simulation. Daily pan‐Antarctic SIT observations are urgently needed for future comprehensive evaluation.

Abstract. We conducted a series of perfect model experiments using Icepack, a one-dimensional single-column sea ice model, to assess the potential of data assimilation (DA) to improve predictions of the mean sea ice state through the incorporation of sea ice albedo (SIAL) observations in addition to sea ice concentration (SIC) and sea ice thickness (SIT) observations. One ensemble member is designated as the TRUTH, and synthetic observations drawn from it are assimilated into the remaining ensemble members. DA is carried out using the Data Assimilation Research Testbed (DART) Quantile Conserving Ensemble Filtering Framework (QCEFF), which accounts for the bounded nature of sea ice variables. Icepack ensembles were spun-up for five Arctic locations based on small perturbations to atmospheric forcing. Results show that assimilating SIAL has the potential to improve reanalysis products when concurrently assimilated with the more commonly assimilated observables SIC and SIT at three of the five discrete points examined in the Arctic Ocean, when observational uncertainty in SIAL is reduced below current literature estimates. These findings underscore the value of leveraging existing SIAL observations and expanding their temporal and spatial coverage in the Arctic. Furthermore, the study highlights the critical need to better constrain the observational uncertainty of SIAL. Enhanced observational networks would provide the necessary validation data, enabling more accurate uncertainty characterization and improving sea ice forecasts in a rapidly evolving polar climate.

Abstract In this study, we propose a regional deep learning model (SISNet) to evaluate the feasibility of data-driven model in seasonal prediction of Arctic ice-ocean coupling. SISNet integrates key multi-sphere information of the ocean and sea ice, which enables the prediction of the joint evolution of the ocean-sea ice coupled system over the next 12 months. During the 10-year testing period, SISNet outperforms persistence model and other deep learning baseline models by a significant margin. Specifically, the root mean square errors (RMSE) of sea ice concentration (SIC) and sea ice thickness (SIT) predicted by SISNet are approximately 10% and 0.2 m, respectively. By comparing SISNet with reanalysis and satellite observations, this study finds that SISNet can accurately capture the seasonal evolution characteristics of the Arctic ice-ocean coupling system. Meanwhile, SISNet can significantly enhance the prediction performance of the sea ice melting season. The anomaly correlation coefficients (ACC) for SIC and SIT both exceed 0.6. Experiments conducted with different starting months reveal that the spring predictability barrier for Arctic sea ice has been effectively mitigated. The results show that SISNet exhibits more reliable and stable performance in multivariate predictions for the Arctic over the next 12 months. The study highlights that our modeling approach provides valuable guidance for ice-ocean coupling prediction.

Abstract The rapid decline of Arctic sea ice requires accurate prediction of its concentration (SIC) and thickness (SIT). We introduce IceCT, a deep learning model using a Mixture‐of‐Experts (MoE) framework to generate monthly SIC and SIT forecasts at a 12.5 km resolution up to 6 months ahead. Its architecture features two specialized experts—one for SIC, one for SIT—enhanced with Multi‐Head Self‐Attention (MHSA) to capture spatial patterns. A novel Spatial Gate adaptively weights the experts' contributions across the grid. Evaluations show IceCT outperforms other deep learning baselines, achieving an overall Mean Absolute Error (MAE) of 4.56% for SIC and 0.140 m for SIT, and achieves enhanced predictive skill in forecasting sea ice during the crucial summer melting period. Additionally, the model mitigates the Arctic spring predictability barrier by learning to rely more on SIT information for forecasts initialized in spring.

Daily Arctic first-year sea ice thickness observed by China's GNSS-R satellite constellation.

ABSTRACT Accurate estimation of snow depth on Antarctic sea ice is critical for understanding ice mass balance, surface thermodynamics, and satellite-altimetry-based sea ice thickness retrievals. This study introduces a dual-mode retrieval framework for deriving a snow depth product on Antarctic sea ice using the Microwave Radiation Imager (MWRI). The MWRI snow depth product outperforms existing passive-microwave products in accuracy, seasonal adaptability, and temporal consistency. Validation against multiple in situ datasets shows that MWRI snow depth achieves superior performance across most Antarctic regions, notably with an RMSE below 9.0 cm and a correlation coefficient above 0.70 in West Antarctica. During the melt season, validation with AWI snow buoys yielded an RMSE of 8.5 cm, demonstrating robustness under complex surface conditions. Time-series comparisons with ICESat-2 snow depth demonstrate that the MWRI snow depth effectively captures seasonal variability (r = 0.79), accurately reproducing the winter-to-spring snow-accumulation trend. Seasonally, snow depth rises in winter and early spring and diminishes during summer melt and compaction. Interannually, long-term snow-covered zones—particularly the Weddell and Amundsen Seas—remain relatively stable and thick, while marginal ice areas exhibit a clear thinning trend. The dataset is available at https://doi.org/10.57760/sciencedb.25039.

Abstract. A major shift in Arctic sea ice occurred in 2007, transitioning from thicker, deformed ice to thinner, more uniform ice with reduced surface roughness. This abrupt change likely altered the dynamic and thermodynamic interactions between sea ice and ocean, with potential implications for nutrient and biogeochemical cycles in both sea ice and the upper ocean. In this study, we present a suite of regional coupled ocean-sea ice simulations designed to assess the potential impact of the regime shift on sea ice–ocean interactions, with a regional focus on the Atlantic sector of the Arctic Ocean. The different sea ice regimes are represented by changes in ice thickness distribution described by ice thickness classes in the sea ice model, and the effects of the different regimes are simulated through variations in the drag coefficient diagnosed from the ice thickness distribution. We emulate the two sea ice regimes by prescribing sea ice properties at the model's lateral boundaries. The results suggest a weaker dynamical coupling between sea ice and ocean in the new sea-ice regimes, leading to enhanced surface stratification, suppression of vertical mixing and momentum transfer to deeper layers. The simulation framework and the physical analyses presented here serve as a basis for ocean biogeochemical modelling studies that aim at understanding ocean ecosystem responses to changing Arctic sea ice.

Abstract. Jones Sound is one of three critical waterways in the Canadian Arctic Archipelago that regulate liquid exchange between the Arctic Ocean and northern Atlantic Ocean. However, to date, no high-resolution ocean circulation model exists to study the recent evolution of Jones Sound, meaning that our understanding of circulation within the sound is based either on temporally and spatially sparse oceanographic observations or on extrapolating conditions within Baffin Bay, which has a more dense observational record. To address this, we develop a high-resolution (1/120°, 0.9 km) Jones Sound configuration of the Massachusetts Institute of Technology general circulation model and perform coupled ocean–sea ice–biological productivity simulations between 2003–2016. We find that circulation through Lady Ann Strait, Fram Sound, and Glacier Strait comprises 71 %, 14 %, and 15 % of the volumetric transport into and out of Jones Sound, with tidal flushing enhancing the magnitude of volumetric transport through Fram Sound. Warming Atlantic Water within western Baffin Bay flows into Jones Sound through Lady Ann Strait, becomes well-mixed, and circulates counterclockwise, encroaching on the terminus of most tidewater glaciers that line the eastern periphery of the sound. Furthermore, we find that sustained atmospheric and oceanic warming drives an 11 % reduction in the 2003–2016 mean summertime sea ice area, decreased wintertime sea ice thickness, and delayed onset of sea ice refreeze in the fall (thus lengthening the amount of time during which Jones Sound is ice-free). Tidal flushing through Cardigan Strait is critical in triggering melt-back of sea ice across northern Jones Sound. Lastly, this decline in sea ice increases light availability and, when coupled with warming of the subsurface waters in Jones Sound, facilitates enhanced primary productivity down to ∼ 21 m depth. While we note that the modeled warming signal in Baffin Bay is overestimated relative to observations, the results presented here improve our general understanding of how this critical waterway might change under continued polar-amplified global warming and underscores the need for sustained oceanographic observations in this region.

Abstract Previous work documented both strongly positive and strongly negative impacts of cyclones on sea ice, with differing conclusions on their overall effect. To better quantify this effect, we attribute sea ice tendencies to cyclones and other weather features in a long-term pan-Arctic simulation with the next-generation sea-ice model (neXtSIM). We find the net impact of cyclones on the Arctic sea ice to remain small, although cyclones can locally and instantaneously strongly affect sea ice. This result is due to a combination of cyclones occurring relatively infrequently and many other processes affecting sea ice also in the absence of cyclones. In contrast to previous studies, we find the weakest impact of cyclones on sea ice during summer. We further show that focusing on different types of midlatitude weather systems in the Arctic captures a more nuanced impact on sea ice, with fronts or pronounced moisture transport having a stronger impact on sea ice than cyclones, in particular during winter. Even though fronts and moisture transport do not necessarily occur independently of cyclones, they highlight different aspects and sometimes extend significantly beyond the area associated with cyclones. Climatologically, however, the impact of fronts and moisture transport axes remains small because these features occur even less frequently than cyclones. Finally, we show that cyclones have a stronger impact on sea ice concentration than sea ice thickness. Significance Statement Cyclones can have strongly positive and negative effects on sea ice. On the one hand, they are often associated with strong wind and warm air, causing melt both at the surface and by mixing up warmer waters from below. On the other hand, the colder air in the cold sector can accelerate sea ice formation. In addition, the wind associated with a cyclone can spatially redistribute the ice thickness. Previous work has come to different conclusions on which processes dominate the overall impact. We show that the vast majority of the sea ice changes in the Arctic over four recent decades occurred in the absence of cyclones. Cyclones had a weak net-negative impact on sea ice, which was least pronounced during summer.

Abstract Polynyas, thin-ice or open water regions within the sea ice, have regularly been observed in the Arctic since satellite observations began in the 1970s. Their opening, in response to complex interactions between several drivers, significantly influences the regional weather and climate, ecosystem, and ocean circulation. Yet their monitoring at the pan-Arctic scale is rare since their detection is not trivial. Here, we use three sea ice satellite data products to detect and investigate major Arctic polynya events since 1978, focusing on their winter locations and total area. We compute the polynyas’ recurrence percentage, total number and area, varying the sea ice concentration (30 – 60%) and thickness (10 – 30 cm) thresholds to enhance our analysis robustness. We find that the most active polynya regions are along the coasts of the Laptev Sea, Kara Sea, Franz-Josef Land, northwestern Greenland, and Chukchi Sea. Both total and cumulative polynya areas have significant increasing trends in these regions and at the pan-Arctic scale between 1978 and 2024. In these regions, we find that wind speed and direction have a prominent one-day lag effect on polynya openings, suggesting that they are latent heat polynyas. The air temperature plays a preconditioning role in many regions, but seems to impact most the daily area extent, after the polynyas formed. Under rising temperatures and stronger extreme winds, our results suggest an increase in Arctic polynya activity, although polynyas might then extend into the open ocean, where different processes would drive their opening.

Abstract. The EU and ESA plan to launch a dual-frequency Ku- and Ka-band polar-orbiting synthetic aperture radar (SAR) altimeter, the Copernicus Polar Ice and Snow Topography Altimeter (CRISTAL), by 2027 to monitor polar sea ice thickness (SIT) and its overlying snow depth, among other applications. However, the interactions of Ku- and Ka-band radar waves with snow and sea ice are not fully understood, demanding further research effort before we can take full advantage of the CRISTAL observations. Here, we use three ongoing altimetry missions to mimic the sensing configuration of CRISTAL over Arctic sea ice and investigate the derived snow depth estimates obtained from dual-frequency altimetry. We apply a physical model for the backscattered radar altimeter echo over sea ice to CryoSat-2's (CS2's) Ku-band altimeter in SAR mode and to the SARAL mission's AltiKa (AK) Ka-band altimeter in low-resolution mode (LRM), and then we compare it to reference laser altimetry observations from ICESat-2 (IS2). ICESat-2 snow freeboards (snow + sea ice) are representative of the air–snow interface, whereas the radar freeboards of AltiKa are expected to represent a height at or close to the air–snow interface, and CryoSat-2 radar freeboards are expected to represent a height at or close to the snow–ice interface. The freeboards from AltiKa and ICESat-2 show similar patterns and distributions; however, the AltiKa freeboards do not thicken at the same rate over winter, implying that Ka-band height estimates can be biased low by 10 cm relative to the snow surface due to uncertain penetration over first-year ice in spring. Previously observed mismatches between radar freeboards and independent airborne reference data have frequently been attributed to radar penetration biases, but they can be significantly reduced by accounting for surface topography when retracking the radar waveforms. Waveform simulations of CRISTAL in its expected sea ice mode reveal that the heights of the detected snow and ice interfaces are more sensitive to multi-scale surface roughness than to snow properties. For late-winter conditions, the simulations suggest that the CRISTAL Ku-band radar retrievals will track a median elevation 3 % of the snow depth above the snow–ice interface because the radar return is dominated by surface scattering from the snow–ice interface which has a consistently smoother footprint-scale slope distribution than the air–snow interface. Significantly more backscatter is simulated to return from the air–snow interface and snow volume at Ka band, with the radar retrievals tracking a median elevation 10 % of the snow depth below the air–snow interface. These model results generally agree with the derived satellite radar freeboards, which are consistently thicker for AltiKa than CryoSat-2, across all measured snow and sea ice conditions.

Abstract This article estimates the future transit capacity of the Northern Sea Route (NSR) in consideration of the ice navigation capabilities of the world fleet and the escort capacity of the current and planned Russian icebreakers. The work employs two different storyline simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6) to account for the future development of sea ice extent and thickness between 2024 and 2050. In both simulations, the transit traffic is expected to remain seasonal and highly dependent on limited icebreaking capacity, affecting the potential of liner shipping in particular. In the analyzed simulations, the current and estimated maximum transit capacity of the NSR significantly exceeds currently realized transport volumes, confirming prior assumptions that volumes on the route are not a capacity issue but are instead mostly caused by a lack of time savings, poor economic viability, and navigational safety concerns.