The study revisits and dramatically expands earlier work published in 2015 and 2017, which had relied on a smaller set of climate projections drawn from the Coupled Model Intercomparison Project Phase 3, or CMIP3. By folding in newer simulations driven by CMIP5 global climate models, Groll and Grabemann assembled an ensemble of up to 14 regional wave projections for the North Sea and 8 for the Baltic Sea. Each projection combines a different global climate model, a different regional climate model used to downscale atmospheric conditions to local scales, different initial conditions, and different greenhouse gas emission scenarios. This deliberate diversity is what allows the researchers to separate genuine climate signals from the noise of natural variability.

At the heart of the analysis is the significant wave height — a standard oceanographic measure representing roughly the average height of the highest one-third of waves at a given location. It is the metric that coastal engineers, shipping companies, and offshore wind farm operators care about most, because it governs the forces that waves exert on breakwaters, ships, and turbine foundations. The researchers ran the third-generation spectral wave model WAM, forced by downscaled wind fields from the regional climate models, at a fine horizontal resolution of approximately 5 by 5 kilometres — fine enough to resolve the complex coastlines, shallow banks, and island chains that shape wave behaviour in these semi-enclosed seas.

The results paint a picture of a North Sea divided. By the final three decades of the 21st century, the ensemble projects a clear west-to-east gradient in wave height change. Along the British coast, median significant wave heights decrease by up to 5 percent, while in the eastern reaches — including the German Bight and the Skagerrak — annual maxima increase by more than 5 percent under high-emission scenarios. The 99th percentile of wave height, a proxy for extreme storm-driven seas, shows increases of 2.5 to 5 percent across the southern and eastern North Sea. These patterns are not uniform across all model runs, but they emerge as statistically robust when at least 11 of the 14 ensemble members agree on the direction and magnitude of change.

The Baltic Sea tells a different story altogether. Rather than a gradient, the ensemble projects a more widespread increase in wave heights across most of this brackish inland sea. Maximum significant wave heights rise by more than 5 percent in the central Baltic Proper, the Gulf of Riga, the Gulf of Finland, and areas west of Bornholm. Under high-emission scenarios, some central areas see increases exceeding 7.5 percent. The researchers attribute part of this Baltic-specific signal to a factor absent from the North Sea analysis: shrinking seasonal sea ice. As the climate warms, ice cover in the Bothnian Sea, the Bay of Bothnia, and the Gulf of Finland retreats, exposing previously protected waters to wind-driven wave generation for longer periods each year — a mechanism that amplifies the wave height increases in the northern Baltic.

What makes the study particularly valuable for climate policy is its explicit comparison of high- and low-emission futures. The researchers split their ensemble into two sub-groups based on projected atmospheric carbon dioxide concentrations in the year 2100, with a threshold of 700 parts per million separating the two. High-emission scenarios — including the SRES A1B and A2 pathways and the RCP8.5 pathway — consistently produce stronger and more spatially extensive wave height changes than their low-emission counterparts such as RCP2.6, B1, and B2. The spatial pattern of change remains the same in both cases, but the amplitude scales with the emissions trajectory, offering a direct illustration of how mitigation choices translate into physical ocean conditions.

Temporal analysis at four sentinel locations adds a crucial dimension of when these changes will become detectable. In the German Bight, the ensemble median of the 99th percentile wave height shows a statistically significant increase of more than 5 percent by the end of the century under high emissions, while low-emission scenarios produce changes too small to distinguish from natural variability. At Tyne, on the British coast, a steady and statistically significant decrease emerges from around 2030 onward under high emissions, reaching roughly 5 percent by 2100. In the Baltic, the Central Baltic Proper shows the most consistent signal, with high-emission scenarios producing statistically significant increases from mid-century onward — up to 10 percent in maximum wave height — while the Arkona Basin in the western Baltic shows more muted and delayed changes.

The physical mechanism behind these patterns lies in shifting atmospheric circulation. Previous analyses of the underlying climate models have identified an increased frequency of westerly wind patterns and a tendency toward more positive phases of the North Atlantic Oscillation, which steer stronger winds from the British Isles toward the Baltic. Zappa and colleagues showed in 2013 that the North Atlantic storm track is projected to extend toward Europe in winter under CMIP5 scenarios, and Ruosteenoja and colleagues found in 2019 that strong westerly winds become more frequent while easterly winds decline. The wave height changes documented by Groll and Grabemann are the ocean surface’s integrated response to these wind shifts — waves grow where westerly fetch lengthens and shrink where it shortens.

The authors are careful to frame their findings within the substantial internal variability that characterises regional climate projections. Individual ensemble members can deviate by as much as 20 percent in either direction from the ensemble mean, reflecting the chaotic nature of atmospheric dynamics on decadal timescales. This is precisely why the ensemble approach matters: no single simulation can be trusted to represent the future, but the consensus signal across many independent runs — each representing an equally plausible realisation of the climate system — provides a far more reliable picture. The researchers deliberately treated all ensemble members as equally likely, following the

Subject of Research: Projected changes in significant wave height in the North Sea and Baltic Sea under CMIP3 and CMIP5 climate scenarios

Article Title: An ensemble of regional CMIP3/CMIP5 wind wave scenarios for the North Sea and the Baltic Sea: a revisit

Article References: Groll, N., & Grabemann, I. (2026). An ensemble of regional CMIP3/CMIP5 wind wave scenarios for the North Sea and the Baltic Sea: a revisit. Ocean Science, 22(5), 3185-3206. https://doi.org/10.5194/os-22-3185-2026

Image Credits: AI Generated

DOI: 10.5194/os-22-3185-2026

Keywords: North Sea, Baltic Sea, significant wave height, wave climate, CMIP3, CMIP5, climate change, sea ice, regional climate models, WAM wave model, emission scenarios, coastal protection

The study, led by Theresa Mathes of the Technische Universität Berlin and the Leibniz Institute for Tropospheric Research, together with colleagues from the University of Leeds, Stockholm University, Uppsala University and TROPOS in Leipzig, was carried out as part of the ARTofMELT expedition, which sailed from Svalbard between 8 May and 15 June 2023. The campaign’s broader goal was to understand how atmospheric rivers, narrow corridors of warm, moist air, trigger the onset of sea ice melt. But the flux measurements collected along the way, published in the journal Atmospheric Chemistry and Physics, provide one of the longest continuous eddy covariance records of particle exchange ever obtained over the Arctic Ocean, and they carry implications that reach well beyond the expedition’s original purpose.

Aerosol particles matter enormously in the Arctic. They scatter and absorb sunlight directly, and they act as cloud condensation nuclei and ice nucleating particles, shaping the formation, brightness and lifetime of clouds. Because particle concentrations in the polar boundary layer are often very low, particularly in autumn, even small changes in the particle budget can ripple through cloud microphysics and the regional radiation balance. The Arctic is warming two to seven times faster than the global average since 1979, a phenomenon known as Arctic amplification, and shifts in the sources and sinks of aerosols are one of the feedbacks that models struggle to capture. Yet direct observations of turbulent particle fluxes over sea ice have been strikingly scarce, with most previous ship-based campaigns lasting only days to a couple of weeks.

The technical challenge of making these measurements from a moving ship is considerable. Eddy covariance, the gold-standard technique, requires capturing rapid fluctuations in vertical wind speed and particle concentration simultaneously, at sampling rates of around 10 hertz. The team mounted a sonic anemometer 20.3 meters above the surface on Oden’s foremast, alongside a heated inlet feeding a mixing-type condensation particle counter capable of detecting particles down to 7 nanometers. Because the instrument sampled air through a long tube, the researchers had to correct carefully for the time lag between wind and concentration fluctuations, for particle losses to the tubing walls, and for the motion and flow distortion of the ship itself, using motion sensors and computational fluid dynamics corrections developed for Oden. Ship exhaust contamination, an ever-present hazard on a vessel, was rigorously screened out, removing 17 percent of the data.

To interpret the fluxes, the team classified the surface upwind of the ship into three categories: closed ice, leads, and open water. This was done automatically using a machine-learning image segmentation algorithm, a random forest classifier, applied to hourly photographs taken by cameras on the ship’s decks, with the relevant camera chosen according to the prevailing wind direction. The approach allowed the researchers to attribute each 20-minute flux measurement to the type of surface it had sampled, something few previous studies have done systematically.

The headline result is a study in contrasts. Under low to moderate turbulence, closed ice behaved as expected: it was a net sink for particles, with a median deposition flux of −0.02 × 10⁶ particles per square meter per second, and deposition dominated roughly 70 percent of the intervals measured over ice. But when the wind picked up, the picture reversed. During eight high wind speed events identified during the campaign, five showed net particle emission, and during the longest of these, a storm on 24 to 26 May with winds reaching 19 meters per second, the team recorded net emission fluxes of up to 0.98 × 10⁶ particles per square meter per second, a statistically robust figure with a 95 percent confidence interval excluding zero. The inferred emission strength, calculated by subtracting a modeled deposition flux from the measured net flux, peaked at 1.57 × 10⁶ particles per square meter per second during a second storm at the end of May.

The most likely culprit is blowing snow. Strong winds lift snow grains from the surface, and because Arctic snow contains sea salt, sulfate, nitrate, organic matter and even black carbon, the sublimation of these wind-blown particles releases aerosols into the boundary layer. The particle size distributions measured during the storm events support this interpretation: elevated concentrations appeared across a broad range of diameters, roughly 30 to 300 nanometers, rather than showing the narrow nucleation-and-growth signature expected from new particle formation driven by precursor gases. The researchers also found that the strength of the particle flux correlated strongly with friction velocity, a measure of near-surface turbulence, with a correlation coefficient of 0.88, reinforcing the link between turbulent wind stress and particle exchange. The threshold wind speed for blowing snow is not fixed; it depends on temperature and snow properties, and at low temperatures mobilization could begin at winds as gentle as 6 meters per second.

Over leads and open water, the story was more nuanced. Both surfaces showed a mixture of emission and deposition events, with median fluxes close to zero, suggesting that while open ocean and cracks in the ice can act as particle sources, through mechanisms such as bubble bursting and sea spray, their net contribution during the campaign was modest. The researchers note that whitecaps in pack ice regions are smaller and less frequent than in the open ocean, which may explain the weak net emission observed. They also caution that open water periods were underrepresented in the filtered dataset because the ship was typically underway when open water was present, and surface classification was restricted to periods of slow drift.

What makes these findings consequential is the trajectory of the Arctic itself. Climate models project continued sea ice loss and a corresponding expansion of open water, which is expected to change the frequency and intensity of the particle exchange processes documented here. As the ice retreats, local particle sources, whether from leads, open water or wind-blown snow, are likely to play a growing role in the regional aerosol budget, with knock-on effects for cloud formation and the surface energy balance. Whether the net climatic effect is cooling, through brighter clouds reflecting more sunlight, or warming, through changes in cloud lifetime and phase, remains poorly constrained, and competing source and sink processes make the outcome genuinely uncertain.

By providing hard observational constraints on particle fluxes over three distinct surface types, the ARTofMELT dataset gives modelers something they have long lacked: direct numbers against which to test deposition schemes and emission parameterizations in the High Arctic. The team’s data are publicly available through the PANGAEA repository. As the researchers conclude, quantifying particle deposition on closed ice and emissions from leads and storm-whipped snowfields is a step toward reducing the stubborn uncertainties that still surround Arctic aerosol-cloud-climate interactions, in a region where the stakes of getting the feedbacks right could not be higher.

Subject of Research: Turbulent aerosol particle exchange between the atmosphere and sea ice, leads and open water in the High Arctic

Article Title: Turbulent atmospheric particle fluxes in the High Arctic: surface type dependence and variability

Article References: Mathes, T., Brooks, I., Murto, S., Tjernström, M., Kojoj, J., Zieger, P., Prytherch, J., Wehner, B., & Held, A. (2026). Turbulent atmospheric particle fluxes in the High Arctic: surface type dependence and variability. Atmospheric Chemistry and Physics, 26(19), 14507-14530. https://doi.org/10.5194/acp-26-14507-2026

Image Credits: AI Generated

DOI: 10.5194/acp-26-14507-2026

Keywords: Arctic, aerosol particles, eddy covariance, sea ice, blowing snow, particle fluxes, cloud condensation nuclei, Arctic amplification, leads, sea spray aerosol, ARTofMELT, dry deposition

At the center of the analysis is a quantity that fire scientists watch closely: vapor pressure deficit, or VPD. VPD measures the gap between how much water vapor the air actually holds and how much it could hold when saturated. When this gap widens, the atmosphere effectively becomes a sponge, pulling moisture out of vegetation, soils, and dead organic matter. High VPD desiccates fuels, lowers the moisture content of needles, leaves, and peat, and makes it far easier for any ignition source, whether lightning or a human spark, to set off a blaze that spreads rapidly and burns hot. Decades of observations across boreal regions have tied rising VPD to worsening fire seasons, which is why understanding what drives VPD upward is central to anticipating future fire risk.

Previous work had established one major pathway: large-scale atmospheric circulation. When air sinks over a region in descending branches of circulation patterns, it warms by compression and dries out, pushing VPD higher through purely atmospheric means. What remained unclear was how much of the circulation-driven VPD increase is realized indirectly, through feedbacks that the circulation triggers at the land surface. The new study set out to separate these direct and indirect contributions, and the answer it finds is striking.

According to the researchers, atmospheric circulation explains roughly 60 percent of the summer VPD trend over eastern Siberia between 2004 and 2024. That figure alone confirms the dominant role of circulation in setting the region’s fire weather trajectory. But the team’s decomposition went further, quantifying how much of that circulation contribution arrives through land-atmosphere feedback rather than through direct warming and drying of the air column. Their result: circulation-triggered land-atmosphere feedback accounts for about 42 percent of the circulation contribution, a substantial amplification pathway that earlier attribution studies had largely left unquantified.

The mechanism begins far to the north, over the Laptev Sea, a sector of the Arctic Ocean along Siberia’s northern coast. The study finds that reduced sea ice concentration in the Laptev Sea favors the development of an anomalous anticyclonic circulation over eastern Siberia. Anticyclones are systems of high pressure in which air spirals slowly downward and outward. The sinking motion suppresses cloud formation and precipitation, so less rain falls over the region during the summer months. With rainfall diminished, soil moisture steadily depletes through the season, and the land surface begins to dry out.

This is where the physics of the land surface takes over. A moist soil can cool itself efficiently: incoming solar energy is spent evaporating water, a process known as latent heat flux, which returns moisture to the atmosphere. But when the soil dries, that evaporative escape valve closes. The surface energy budget shifts, with a growing fraction of the sun’s energy converted instead into sensible heat, directly warming the air near the ground. Warmer air can hold more water vapor, so even if the absolute moisture in the air stays the same, the vapor pressure deficit climbs. The drying land thus feeds back into the very atmospheric dryness that caused it, tightening a self-reinforcing loop in which circulation suppresses rain, rain shortage dries the soil, and the parched soil amplifies the atmospheric thirst that defines extreme fire weather.

The quantitative structure of the finding matters for how scientists attribute fire risk. If atmospheric circulation directly explains about 60 percent of the VPD trend, and feedbacks triggered by that circulation supply roughly 42 percent of the circulation contribution, then a meaningful slice of the trend that appears circulation-driven is in fact mediated through the land surface. This distinction is more than bookkeeping. It means that the state of the soil, and upstream of that the state of Arctic sea ice, shapes how severe a given circulation anomaly becomes in practice. Two summers with identical large-scale wind and pressure patterns could produce very different fire weather depending on how wet or dry the ground is when the pattern sets in.

The link to sea ice adds an Arctic dimension to Siberia’s fire problem. The Laptev Sea has experienced pronounced declines in summer ice concentration as the Arctic warms, and the study identifies this loss as a factor favoring the anticyclonic circulation over eastern Siberia. In other words, changes at the top of the planet can cascade southward: less ice alters the atmospheric circulation, the altered circulation suppresses precipitation, the suppressed rain dries the soils, and the dry soils intensify the fire weather. This chain connects the fate of sea ice, a hallmark of Arctic climate change, to the frequency of smoke plumes over boreal forests thousands of kilometers away.

The implications extend well beyond Siberia. The authors describe their results as revealing a key land-surface pathway through which circulation intensifies fire weather risk, and as providing a broader framework for understanding similar changes in other fire-prone regions. Many of the world’s fire belts, from the Mediterranean to Australia to the western United States, feature the same essential ingredients: circulation anomalies that modulate rainfall, soils whose moisture state feeds back on near-surface temperature and humidity, and vegetation that becomes flammable as VPD rises. The Siberian case demonstrates that attribution studies which consider only direct atmospheric drivers may systematically understate the role of land-atmosphere coupling, and that models used for seasonal fire outlooks need to capture soil moisture evolution realistically if they are to anticipate the worst conditions.

For a region like eastern Siberia, the stakes are unusually high. The boreal forests and permafrost landscapes there store enormous quantities of carbon, and fires that burn deeper into soils and organic layers release that carbon to the atmosphere while degrading the frozen ground beneath. An intensifying fire weather regime, amplified by the feedback loop the study documents, therefore threatens not only ecosystems and communities in the region but also a globally significant carbon reservoir. By quantifying how much of the rising fire weather risk flows through the land surface, the research provides a clearer target for monitoring: sea ice conditions in the Laptev Sea, soil moisture across Siberia, and the anticyclonic patterns that tie them together may together offer early warning of the seasons when the region’s forests are most likely to burn.

Subject of Research: Land-atmosphere feedbacks amplifying circulation-driven fire weather risk over eastern Siberia

Article Title: Land-atmosphere feedback amplifies circulation-dominant fire weather risk over eastern Siberia

Article References: Bai, D., Yu, H., Ma, K., Guo, R., Hu, Y., Jiang, Z., Wu, H., & Fu, M. (2026). Land-atmosphere feedback amplifies circulation-dominant fire weather risk over eastern Siberia. Communications Earth & Environment. https://doi.org/10.1038/s43247-026-04150-2

Image Credits: AI Generated

DOI: 10.1038/s43247-026-04150-2

Keywords: eastern Siberia, wildfire, vapor pressure deficit, land-atmosphere feedback, Laptev Sea, sea ice, soil moisture, anticyclonic circulation, fire weather, atmospheric dynamics, boreal forests, climate change

The research team ran six-member ensembles of an Earth system model in which CO2 emissions were switched off instantaneously at different points along a business-as-usual trajectory. The switch points were defined by cumulative emission totals ranging from 800 to 2000 petagrams of carbon, a span that covers everything from aggressive early mitigation to decades of continued delay. This experimental design isolates a question that conventional scenario simulations rarely address cleanly: what happens to the climate system after emissions end, and does the answer depend on when we end them? For the globe as a whole, the answer is reassuring in one respect. Consistent with a large body of prior work, global mean temperature stabilizes once CO2 emissions cease, because the cooling effect of declining heat uptake by the ocean roughly balances the warming effect of declining radiative forcing.

The Arctic, however, refuses to follow the global script. In the simulations, the region exhibits what the authors describe as nonlinear and bifurcating behavior. Below a cumulative emission threshold of about 1700 petagrams of carbon, the Arctic continues to warm and lose sea ice even after emissions stop, drifting further from its historical state. Above that threshold, something entirely different happens: the Arctic abruptly cools and its sea ice expands. The same quantity of additional carbon, in other words, can push the region in opposite directions depending on which side of the divide it lands on. This is the signature of a tipping element, a component of the climate system whose response to forcing is not proportional to the forcing itself but instead involves a sudden transition between alternative states.

The mechanism behind the abrupt cooling is the Atlantic Meridional Overturning Circulation, or AMOC, the vast system of ocean currents that carries warm, salty water northward through the Atlantic and returns cold, deep water southward. In the high-emission simulations, the AMOC undergoes a near-collapse, sharply reducing the amount of heat that the ocean transports poleward into the North Atlantic and Arctic. Deprived of this oceanic heat supply, the surface of the far North Atlantic and the Arctic cools rapidly. The authors identify two amplifying feedbacks that magnify this cooling. The first is the lapse-rate feedback, which is unusually strong in polar regions: because the Arctic atmosphere is stably stratified, surface cooling there is not efficiently mixed upward and is therefore retained and intensified near the surface. The second is the surface-albedo feedback: as the cooled surface allows sea ice to expand, the bright ice reflects more sunlight, cooling the surface further and allowing still more ice to grow.

Perhaps the most sobering detail in the study is the timing of the bifurcation. The AMOC collapse in the simulations occurs when CO2 emission mitigation has been delayed long enough that global mean temperature exceeds the 2 degrees Celsius limit enshrined in the Paris Agreement. In other words, the dangerous branch of the Arctic’s future is not triggered by the amount of warming alone but by the path taken to get there. A world that burns roughly 1700 petagrams of carbon or more before cutting emissions to zero crosses into territory where the Arctic’s post-emissions trajectory becomes qualitatively different from that of a world that acts earlier. The finding reframes the Paris target not merely as a guardrail against gradual harm but as a boundary beyond which abrupt, hard-to-reverse regional transitions become plausible.

There is also an element of unpredictability woven through the results that the authors characterize as stochastic tipping behavior. Within each ensemble, the individual members share identical forcing but differ in their internal climate variability, the natural chaos of weather and ocean eddies that no model or observation can predict in detail. The tipping transition is modulated by this internal variability, meaning that even two otherwise identical worlds with the same cumulative emissions could diverge, with one tipping into the cold state and the other continuing to warm. This stochastic character makes the transition impossible to forecast with precision and underscores why the authors describe it as an abrupt shift to an alternative Arctic climate state rather than a gradual, predictable drift.

The implications for sea ice deserve particular attention. In the warming branch, the simulations show continued sea-ice loss after emissions cease, extending the ice-free conditions that have become increasingly common in recent decades. In the cooling branch, sea ice expands as the AMOC-driven cooling takes hold. Counterintuitively, then, a delayed-mitigation world could end up with more Arctic sea ice than an early-mitigation world, but for deeply troubling reasons: the ice returns because the ocean circulation that warms the region has broken down. Such a state would carry its own consequences, including altered patterns of ocean heat and carbon uptake, disrupted marine ecosystems adapted to a warming Arctic, and potential knock-on effects on weather patterns over the North Atlantic and Europe that are known to be sensitive to AMOC strength.

The study’s methodology is worth appreciating on its own terms. By using six-member ensembles at each emission cessation point rather than a single simulation, the researchers could distinguish the forced response of the climate system from the noise of internal variability, which is essential when the signal being sought is a bifurcation. By defining the experiment in terms of cumulative emissions, the design aligns with the scientific understanding that peak warming scales approximately linearly with total carbon emitted, giving the results a direct policy interpretation. And by focusing on the post-emissions period, the study addresses a blind spot: most climate projections end at the year 2100 under a given scenario, leaving the long-term behavior of sensitive regions like the Arctic underexplored precisely at the moment when mitigation decisions would take effect.

As with any modeling study, there are caveats that the authors and the broader community will want to weigh. The results come from a single Earth system model, and the precise threshold of 1700 petagrams of carbon should be understood as a model-specific estimate rather than a firm planetary boundary; other models may place the bifurcation at different cumulative emissions or may represent the AMOC’s sensitivity to freshwater and heat fluxes somewhat differently. The stochastic nature of the tipping behavior also means that the probability of crossing the threshold in the real world depends on details of internal variability that remain uncertain. Nonetheless, the qualitative finding, that the Arctic’s post-emissions future can branch in opposite directions depending on the timing of mitigation, is a robust and consequential result that other modeling groups are likely to test with independent tools.

For policymakers and the public, the message is a sharpened version of a familiar warning. Delaying CO2 emissions mitigation does not simply add a few tenths of a degree to the eventual peak temperature; it can change the character of the climate response itself, converting a region that would have stabilized into one that undergoes an abrupt, internally triggered transition. The Arctic has long served as the planet’s early-warning system, warming faster than anywhere else and broadcasting its distress through shrinking ice and thawing permafrost. This new work suggests that the region may hold yet another surprise in store, one in which the price of procrastination is not a warmer Arctic but a fundamentally rearranged one, flipped by a faltering ocean conveyor into a state that no one planned for and no one can easily undo.

Subject of Research: Arctic climate tipping points and AMOC collapse under delayed CO2 emissions mitigation

Article Title: Arctic climate tipping points under delayed CO2 emissions mitigation

Article References: Wu, Y.-T., Previdi, M., Romanou, A., Lerner, P., & Polvani, L. M. (2026). Arctic climate tipping points under delayed CO2 emissions mitigation. Communications Earth & Environment. https://doi.org/10.1038/s43247-026-03872-7

Image Credits: AI Generated

DOI: 10.1038/s43247-026-03872-7

Keywords: Arctic, climate tipping points, AMOC, CO2 emissions mitigation, sea ice, Earth system modeling, Atlantic Meridional Overturning Circulation, surface albedo feedback, lapse-rate feedback, Paris Agreement, cumulative carbon emissions, abrupt climate change

The idea rests on a curious property of seawater. When ocean water freezes, most of the salt is expelled from the ice crystals, concentrating in pockets of brine. But as temperatures drop far below the normal freezing point of seawater, the brine itself becomes saturated and the salt begins to precipitate out as solid crystals. For the entire seawater system, this eutectic point sits at about minus 36 degrees Celsius. Crucially, laboratory experiments and field measurements by other research groups have shown that these salt crystals, particularly the mineral hydrohalite, are extraordinarily reflective, with an albedo of about 0.93, meaning they bounce back nearly all of the sunlight that strikes them. Fresh snow is one of the few natural surfaces that can compete.

For such crystals to matter climatically, they must sit at the surface where sunlight can reach them. This is where the peculiar atmospheric circulation of Snowball Earth comes in. Previous modeling work by Raymond Pierrehumbert, Dorian Abbot, and colleagues indicates that on a fully glaciated planet, a net ablation zone would have existed at lower latitudes, a band where sublimation of ice exceeds precipitation. There, the frozen ocean surface would have been slowly eaten away by the dry air, exposing bare sea ice for thousands of years. As that ice sublimated, any salt crystals precipitating within it would have been left behind, accumulating into what scientists call a lag deposit, a reflective crust of salt sitting directly on the tropical sea ice.

Samuelsberg and his colleagues asked a simple question: what happens if you add this salt-albedo feedback to a climate model? To find out, they used a diffusive one-dimensional energy balance model, a classic and well-studied framework that describes the zonal mean surface temperature of the planet. The model balances incoming solar radiation against outgoing longwave radiation, redistributes heat from the equator toward the poles through diffusion, and includes the Planck response along with the familiar nonlinear ice-albedo feedback. The team modified the model’s albedo function so that latitudes colder than a critical isotherm carry snow or ice albedo, bare sea ice in the net ablation zone carries a lower albedo, and any region where temperatures fall below the eutectic temperature takes on the extremely high albedo of the salt deposit.

Solving the model analytically using boundary integral methods, the researchers traced out bifurcation diagrams showing all the stable and unstable climate states the system can occupy as radiative forcing changes. The results were striking. In addition to the familiar hothouse, high-latitude ice line, and Snowball Earth states, and the Waterbelt states with a tropical water belt produced by the so-called Jormungand mechanism, the salt feedback gave rise to two new, stable, coexisting Snowball Earth states: one with a salt deposit covering the bare sea ice of the deep tropics, and one without. The salt-deposit state is significantly colder than its salt-free counterpart, because its surface reflects so much more sunlight. Several unstable states with smaller or multiple salt deposits also appeared in the analysis.

Which of these states would the real Snowball Earth have occupied? To answer this, the team applied a test previously used to evaluate Waterbelt hypotheses: a candidate climate state must be accessible from a warm climate, and it must persist over a large range of atmospheric carbon dioxide concentrations, consistent with the well-established hysteresis loop in which the planet plunges into glaciation and later escapes as carbon dioxide builds up. The geological record adds a constraint here, because the roughly 1.5 billion years before the Neoproterozoic glaciations, known as the Proterozoic glacial gap, show no evidence of ice sheets, meaning the climate system must have rested in a warm, largely ice-free state before the plunge.

The verdict was unambiguous. For plausible model parameters, the Snowball Earth state without a salt deposit is essentially inaccessible from a warm climate. Instead, once global glaciation begins and temperatures in the tropical ablation zone fall below the eutectic point, the salt-albedo feedback activates and drives the system into the colder, salt-covered state. The analysis showed that this outcome is robust: higher rates of meridional heat diffusion and larger ice-albedo contrasts favor the direct transition, and no matter at what temperature the salt precipitation begins, the system ultimately collapses into the fully salt-covered state. Only very large bare sea ice regions combined with extremely low eutectic temperatures would allow the planet to avoid it. In other words, it may have been very difficult for Snowball Earth to cool down without salt crystals amplifying the chill.

The findings carry implications for one of the thorniest puzzles surrounding Snowball Earth: how the planet ever deglaciated. In the model, escaping the salt-deposit state requires a substantially higher atmospheric carbon dioxide concentration than escaping the classical salt-free state, because the reflective salt crust suppresses the warming that volcanic carbon dioxide would otherwise deliver. Several earlier modeling efforts have tried to lower the carbon dioxide deglaciation threshold to match geochemical proxy data, and the new results suggest that if a salt deposit persisted for a significant time, it would have raised that threshold even further, making the escape from global glaciation harder to explain. The authors also note that the full seawater system contains numerous salt species that begin precipitating well above the eutectic temperature, with mirabilite crystallizing around minus 8 degrees and hydrohalite around minus 23 degrees, meaning some salt precipitation could be expected even before the deepest cold set in, potentially helping to trigger the glaciation in the first place.

Important caveats remain, and the authors are careful to frame the work as an initial investigation. The simplified model includes no ice dynamics, yet it is widely recognized that thick sea glaciers must have formed and flowed equatorward on Snowball Earth, gradually freshening the ice in the ablation zone as meteoric ice arrived from higher latitudes. Previous work by Jacob Goodman found that only a narrow band of marine ice remained near the equator over million-year timescales, so the salt deposit probably became less climatically important in the later stages of the glaciations, and the study cannot determine how long such a deposit lasted. The strength of the feedback is also uncertain: clouds can mask the albedo of both bare sea ice and salt crusts, surface dust may gradually dim the crystals, and winds could blow salt into regions of net precipitation, keeping larger areas salt-free. Yet even with a lower assumed crystal albedo, the qualitative structure of the model’s solutions holds. What the study establishes is that salt precipitation, long absent from Snowball Earth simulations, is a physical process powerful enough to reshape the planet’s climate trajectory, and one that future models of Earth’s deepest freeze can no longer afford to ignore.

Subject of Research: A salt-albedo feedback from precipitated salt crystals on tropical sea ice amplifying cooling during Snowball Earth

Article Title: Amplified cooling of Snowball Earth from a salt-albedo feedback

Article References: Samuelsberg, A., Jakobsen, P. K., & Rypdal, M. (2026). Amplified cooling of Snowball Earth from a salt-albedo feedback. Climate of the Past, 22(8), 1499-1506. https://doi.org/10.5194/cp-22-1499-2026

Image Credits: AI Generated

DOI: 10.5194/cp-22-1499-2026

Keywords: Snowball Earth, salt-albedo feedback, sea ice, hydrohalite, eutectic temperature, energy balance model, Neoproterozoic, albedo, climate modeling, degaciation, hysteresis, Climate of the Past

The problem the team set out to solve is known in the forecasting community as underdispersion. Ensemble forecasts work by running a model many times with slightly different starting conditions and perturbed physics, producing a cloud of possible futures whose spread is meant to mirror the true uncertainty of the forecast. When the ensemble spread is much smaller than the average forecast error, the ensemble is underdispersive: it is overconfident, presenting a range of outcomes that is too narrow to encompass what actually happens. Sea ice forecasts, the authors note, are systematically and often severely underdispersive — even at the moment they are initialised, before any forecast error has had time to grow. This is likely tied to the large biases that models exhibit in sea ice and to the poor representation of unresolved processes such as ice fracturing, ridging, and the turbulent exchange of heat between ice, ocean, and atmosphere.

The remedy tested in the study is a stochastically perturbed parameterisation, or SPP, scheme applied to SI³, the state-of-the-art sea ice model now coupled to ECMWF’s Integrated Forecast System. Rather than perturbing the ice directly, the scheme perturbs nine physical parameters inside the model — including the ice strength parameter P*, the ice–ocean drag coefficient, and snow-related properties — multiplying each by a log-normally distributed random field that evolves in space and time. The random field is generated by repeatedly applying a Shapiro filter to white noise, producing smooth spatial correlations on scales of roughly 750 kilometres, and it decorrelates over about ten days. Because the perturbations act on parameters rather than on tendencies or energies, the scheme conservatively respects tracers, momentum, and sea ice volume, keeping it physically consistent with the stochastic schemes already used in the atmospheric component of the model since the late 1990s.

To evaluate the scheme, the team ran 50-member ensemble forecasts initialised on 1 November and 1 May for every year from 1993 to 2023, each running six months to cover full winter and summer seasons. The experiments used a prototype of cycle 49R2 of the IFS, the configuration that will underpin the next-generation seasonal forecast system SEAS6 and the ERA6 reanalysis. Two ensembles were compared: a control run and an otherwise identical run with sea ice SPP switched on. The only difference between them was the stochastic noise in the ice, which means any divergence between the two ensembles must ultimately trace back to the sea ice itself.

The headline result is a robust increase of roughly 10 percent in ensemble spread for both sea ice concentration and sea ice thickness, in both summer and winter, bringing the spread–error ratio closer to the ideal value of one that defines a well-calibrated probabilistic forecast. But the noise did something unexpected as well. Despite being mean-zero by construction, the perturbations systematically shifted the mean state: sea ice concentration decreased on average, thickness increased, and in winter the ice was redistributed outward from the central Arctic pack toward the ice edge, with drift changes on the order of 20 percent of baseline velocities. The authors explain this through a simple but powerful conceptual model built on two asymmetries. First, sea ice concentration is bounded between zero and 100 percent, and its distribution is strongly bimodal: perturbations can only push near-total ice cover downward, while gridpoints knocked to zero effectively vanish from the scheme’s view, so the mean tends to fall. Second, in winter the dense ice pack resists convergence — piling ice up costs energy — whereas there is little resistance to moving ice toward the open edge, so random perturbations to advection preferentially push ice outward along the path of least resistance. A toy model based on repeatedly perturbing a truncated bimodal distribution reproduces exactly this signature of a declining mean.

These ice changes did not stay confined to the cryosphere. Because fluctuations in ice concentration and thickness modulate surface heat fluxes and albedo, the perturbations altered air temperature up to at least the 850 hectopascal level, with changes of order half a degree Celsius. In winter, the central Arctic warmed where ice cover thinned, allowing more heat to escape the ocean, while regions along the Greenland, Barents, and Kara sea ice edges cooled where the perturbations increased ice concentration. Circulation anomalies, including a cyclonic feature near the Ural Mountains that advected cold Siberian air over the Barents and Kara seas, reinforced these thermodynamic responses. Ensemble spread in air temperature also increased by around 10 percent, both over the Arctic and to some extent at midlatitudes. In summer, by contrast, the atmospheric response was much smaller, as Arctic surface temperatures are pinned near the freezing point and the ice has little leverage on the air above it.

The most striking finding concerns forecast skill beyond the ice itself. In winter, the anomaly correlation coefficient of 500 hectopascal geopotential height — the standard yardstick for seasonal circulation forecasts — increased substantially over the subpolar North Atlantic and extended into Scandinavia, with the difference peaking near 0.5. Averaged over a broad subpolar North Atlantic box, the correlation rose from about 0.3 in the control to about 0.6 with the perturbations switched on, and the winter North Atlantic Oscillation, the dominant mode of European winter variability, was forecast somewhat better as well. The authors are careful to stress the statistical caveats: with only about 30 winters in the sample, field significance was not achieved, and Monte Carlo resampling suggests a roughly 40 percent chance that the pattern of significant gridpoints could arise by chance. Yet the same improvement appeared in several additional hindcasts with different configurations, and the region of enhanced skill coincides with where the ice was most strongly affected, lending credibility to the signal.

The proposed mechanism is as intriguing as the result. The control model carries a uniform warm bias across the Arctic, which weakens the meridional temperature gradient in the subpolar North Atlantic. By increasing ice concentration along the Greenland, Barents, and Kara sea edges, the stochastic perturbations cool the overlying atmosphere and partially correct that warm bias, potentially strengthening the temperature gradient that steers the North Atlantic jet stream. Bootstrap resampling of the perturbed ensemble revealed a statistically significant correlation of −0.28 between the degree of cooling along the ice edge and the skill of the North Atlantic circulation forecast, consistent with this pathway. The authors caution that this accounts for only about 8 percent of the variability, that causality is difficult to establish in a system this noisy, and that some of the improvement may reflect bias compensation rather than a genuinely better simulation of the coupled system. Whether the benefit transfers to other models will likely depend on their own baseline biases.

The study also uncovered a subtlety with practical consequences for how such schemes should be deployed. In uncoupled ocean-and-ice-only simulations, the perturbations were dramatically more effective at generating spread than in fully coupled forecasts, where the atmosphere dominates the error growth and may already saturate the achievable spread; switching to independent random fields for each parameter boosted spread by up to 60 percent in the uncoupled case but made little difference when coupled. This suggests the scheme could be especially valuable in data assimilation, where enhancing the spread of ocean and ice initial conditions is a recognised priority, and work is already under way to apply a variant of the scheme in that context. The team also found no evidence that the perturbations strengthened the much-debated teleconnection between Barents–Kara sea ice and the North Atlantic Oscillation reported in earlier climate-model work, hinting that in these forecasts the relevant pathway may run instead through Greenland sea ice.

For operational forecasting, the message is that a modest dose of well-designed randomness in the ice can pay dividends far beyond the polar cap. The scheme achieved its primary goal of better-calibrated sea ice probabilities, and it did so while introducing mean state changes small enough — around 4 percent in concentration and 10 centimetres in thickness — that modest retuning could offset them if desired. As ECMWF prepares its next generation of seasonal and sub-seasonal systems, and as climate change continues to make Arctic sea ice both more variable and more consequential for midlatitude weather, the idea that noise, properly structured, is a form of signal may prove one of the more quietly important lessons in modern numerical weather prediction.

Subject of Research: Stochastic sea ice perturbation schemes for improving seasonal ensemble forecasts of sea ice and the midlatitude atmosphere

Article Title: The impact of stochastic sea ice perturbations on seasonal forecasts

Article References: Strommen, K., Mayer, M., Storto, A., Spaeth, J., & Tietsche, S. (2026). The impact of stochastic sea ice perturbations on seasonal forecasts. Weather and Climate Dynamics, 7(3), 1593-1618. https://doi.org/10.5194/wcd-7-1593-2026

Image Credits: AI Generated

DOI: 10.5194/wcd-7-1593-2026

Keywords: sea ice, seasonal forecasting, stochastic parameterisation, ensemble forecasts, ECMWF, Arctic, North Atlantic Oscillation, underdispersion, sea ice concentration, jet stream, weather and climate dynamics, coupled modelling

The research, led by Xiner Wu of the Université du Québec à Montréal together with Anne de Vernal and Paul G. Myers, forms part of the Paleoclimate Modelling Intercomparison Project, or PMIP4, a coordinated international effort that uses past climates as out-of-sample tests for the same numerical models that project future warming. The mid-Holocene is an ideal target for such a test. At 6,000 years before present, continental geometry was essentially identical to today’s and greenhouse gas concentrations were close to preindustrial levels, but subtle differences in Earth’s orbit redistributed sunlight across latitudes and seasons, producing warmer Northern Hemisphere summers and cooler winters. If a model cannot faithfully reproduce how the ocean responded to that modest, well-documented forcing, confidence in its response to the far larger forcing of the coming century is necessarily shaken.

To establish what the ocean actually did during that interval, the team turned to an ingenious biological archive: dinoflagellate cysts. These microscopic, extraordinarily resistant structures are produced by single-celled plankton in the surface ocean, and the species composition of the assemblages preserved in layered marine sediments reflects the environmental conditions under which the organisms lived. Crucially, recent work by the same group showed that dinocyst assemblages respond to wintertime deepening of the mixed layer associated with deep convection, allowing quantitative reconstruction of past mixed layer depths. Drawing on twenty-four Holocene sediment records from the modern deep convection regions of the subpolar North Atlantic, the researchers applied a modern analogue technique with a demonstrated error of about 41 metres for winter mixed layer depth — a precision that makes meaningful model comparison possible for the first time.

The comparison was conducted using anomalies rather than absolute values, subtracting the preindustrial baseline from the mid-Holocene simulation for each model and from each proxy record. This sidesteps systematic biases that plague both models and reconstructions. The team recomputed mixed layer depth from modelled temperature and salinity fields using a standardized density threshold, since different modelling groups define the mixed layer in incompatible ways, and they evaluated agreement with two complementary statistics: root mean square error, which measures the magnitude of disagreement, and Cohen’s kappa, which tests whether a model captures the direction of change — deeper or shallower — better than random chance would allow.

The results revealed a strikingly wide spread across the fifteen PMIP4 models. Some, such as ACCESS-ESM1.5, produced very large winter mixed layer anomalies, while others, like INM-CM4-8, barely responded at all. A recurring pattern emerged in several models: deepening in the Labrador Sea alongside a northwestward shift of the deep-mixing zone in the Nordic Seas, a response the authors link to reduced winter sea ice. Higher Northern Hemisphere summer insolation melts more Arctic sea ice, reducing its winter export into the subpolar North Atlantic; where ice retreats, the ocean loses heat to the atmosphere more readily, buoyancy fluxes intensify, and convection deepens — a self-reinforcing feedback in which rising warm, salty subsurface water further suppresses local ice formation.

Yet when the models were confronted with the proxy data, the picture fractured along a geographic line that carries real weight for future projections. In the Nordic Seas, the ensemble mean aligned broadly with the reconstructions, which indicate increased winter mixing at western and southeastern sites and shallower mixing in the east. In the Labrador Sea, however, the disagreement was stark and systematic. All four proxy sites there recorded shallower winter mixed layers during the mid-Holocene, while nearly every model — from both the current PMIP4 generation and the older PMIP3 ensemble — simulated the opposite, a deeper mixed layer. Only one model, MRI-ESM2.0, achieved statistically significant agreement with the direction of reconstructed change, and even the ensemble mean failed the test, demonstrating that averaging models together does not guarantee better performance.

The explanation the authors propose is meltwater — a forcing that the mid-Holocene experiment deliberately omits. The protocol prescribes modern ice sheets in equilibrium, but the real Labrador Sea of 6,000 years ago was anything but equilibrium. Residual ice caps lingered in eastern North America until roughly 6,000 to 5,500 years ago, and the Greenland Ice Sheet did not reach its minimum extent until about 4,000 years ago. Marine records show persistently low sea surface salinity in the Labrador Sea through the Early and Middle Holocene, and earlier proxy work suggests a buoyant, stratified surface layer that suppressed deep-water formation until well into the Holocene. Freshwater discharged from melting ice sheets would have capped the ocean with a light, low-density lid, preventing the dense overturning that models, deprived of this input, freely simulate. Notably, the Nordic Seas appear largely insulated from this effect, consistent with modern studies showing that most meltwater from the East Greenland Current ends up in the Labrador Sea and Baffin Bay rather than crossing eastward.

Transient simulations of the last deglaciation, which do include time-varying meltwater fluxes, lend qualified support to this hypothesis: they overestimate Labrador Sea mixed layer depth less severely than the equilibrium experiments. But the deglaciation models disagree wildly with each other and with the records, and none reproduces the reconstructed late-Holocene trend toward modern-like convection in the Labrador Sea, suggesting that the ocean’s memory in these long simulations may be unrealistically long. The team also found that model details matter enormously — different tuning of the same model changed the mixed layer response more than different ice sheet reconstructions did, and swapping the atmospheric component of a coupled model could reverse the sign of simulated change, underscoring that air-sea-ice interactions, not ocean physics alone, govern open-ocean convection.

Why should this matter beyond the paleoclimate community? Because the Labrador Sea is one of the few places on Earth where the atmosphere can reach down and set the deep ocean in motion, and because the same mechanism — ice sheet melt freshening the surface — is now operating in reverse time. As Greenland’s ice sheet accelerates its melt under global warming, the buoyant lid that suppressed convection 6,000 years ago could reassemble, with consequences for the overturning circulation, European climate and ocean carbon uptake. The study’s central message is that deep convection in the Labrador Sea may be uniquely and disproportionately sensitive to meltwater, and that current climate models, which handle this forcing imperfectly even in purpose-built paleoclimate experiments, may be underestimating that vulnerability. The mid-Holocene, it turns out, was not just an ancient curiosity but a rehearsal — and the models, on this crucial measure, are still learning their lines.

Subject of Research: Evaluation of mixed layer depth in PMIP4 mid-Holocene climate simulations against proxy reconstructions in North Atlantic deep convection regions

Article Title: Mixed layer depth in the PMIP4 midHolocene simulations: comparison to proxy data in North Atlantic deep convection regions

Article References: Wu, X., de Vernal, A., & Myers, P. G. (2026). Mixed layer depth in the PMIP4 midHolocene simulations: comparison to proxy data in North Atlantic deep convection regions. Climate of the Past, 22(9), 1609-1629. https://doi.org/10.5194/cp-22-1609-2026

Image Credits: AI Generated

DOI: 10.5194/cp-22-1609-2026

Keywords: mixed layer depth, PMIP4, mid-Holocene, Labrador Sea, Nordic Seas, deep convection, AMOC, dinocyst proxies, meltwater forcing, sea ice, climate models, paleoclimate

The stakes are considerable. The Antarctic Ice Sheet as a whole holds the equivalent of 57.9 metres of global sea level rise, and the East Antarctic Ice Sheet alone accounts for 52.2 metres of that total, an order of magnitude more than its western counterpart. While West Antarctica has dominated headlines with losses concentrated around Pine Island and Thwaites glaciers, Wilkes Land has quietly emerged as the region of East Antarctica losing the most mass. Previous work identified Frost and Holmes glaciers, both of which drain into Porpoise Bay, as among the top contributors to cumulative ice loss in East Antarctica between 1979 and 2017, behind only Totten and Denman glaciers. Yet despite this signal, the glaciers of Porpoise Bay had never been examined in the detail that their importance demands.

Weatherley and her colleagues set out to close that gap by measuring four key parameters across four outlet glaciers: Frost, Holmes East, Holmes West, and a smaller flow feature the team calls Glacier 1. They mapped the position of the ice-shelf calving fronts using imagery from Landsat, ASTER and Sentinel satellites, extending the record back to 1963 with declassified ARGON photography. They tracked ice surface elevation using multiple satellite altimetry datasets, measured ice velocity from the ITS_LIVE product, and located the grounding lines, the critical boundary where grounded ice begins to float, using differential synthetic aperture radar interferometry, or DInSAR, supplemented by a neural network trained to delineate grounding lines automatically from interferograms.

The results reveal a pattern of change that is hard to dismiss as noise. Frost Glacier’s grounding line retreated 17.17 kilometres between 1996 and 2014, an average rate of roughly 0.95 kilometres per year, with the fastest retreat, more than two kilometres per year, occurring between 1996 and 2004. Holmes East Glacier retreated up to 15.24 kilometres in just five years at the start of that period. If the estimates hold, Frost would rank among the fastest-retreating glaciers in East Antarctica on decadal timescales, with rates comparable to those recorded at Thwaites and Pine Island glaciers in West Antarctica. The authors are careful to flag substantial uncertainties, since the comparison combines datasets that delineate different parts of the grounding zone using different methods, but the retreat is broadly consistent with independent estimates from other studies.

The thinning signal is equally striking. Between 2003 and 2017, the grounded ice at Holmes West Glacier lowered at an average rate of up to 1.22 metres per year at the grounding line, a figure that places it among the most rapidly thinning glaciers in East Antarctica, rivalled in the published record only by Totten Glacier. The thinning propagated inland, reaching 0.34 metres per year ten kilometres up-ice, a spatial pattern characteristic of dynamic thinning driven by changes at the coast rather than by shifts in local snowfall. Because no long-term trends in precipitation or firn densification have been reported for the region, the researchers attribute the elevation loss to the ice itself flowing faster and drawing down its surface, a hallmark of dynamically forced mass loss.

What is driving these changes? The evidence points firmly toward the ocean. Warm, salty modified Circumpolar Deep Water, or mCDW, was detected at the continental shelf break near Porpoise Bay through measurements collected by instrumented seals from the MEOP consortium and by Argo profiling floats between 2004 and 2021. The water masses observed there, with temperatures above freezing at depth and salinities exceeding 34.5 grams per kilogram, match the signature of mCDW in East Antarctica. Reanalysis data from the EN4 dataset further suggest that water warmer than zero degrees Celsius has been present below roughly 450 metres depth on the continental shelf since at least the 1990s. Crucially, gravity-derived bathymetry reveals deep sub-ice-shelf cavities, between 650 and 750 metres deep and deepening inland, beneath Frost, Holmes East and Holmes West glaciers, providing plausible pathways for this warm water to reach the ice from below.

The melt rates implied by this oceanic heat delivery are extraordinary. Previous work calculated an average basal melt rate of 13.3 metres per year beneath the Holmes Ice Shelf between 1994 and 2018, the highest of any ice shelf in East Antarctica and fourth highest in all of Antarctica, exceeding even the melt rate beneath the Totten Ice Shelf. The ice shelf has also been thinning at its surface, with satellite altimetry recording losses of up to 3.07 metres per year on the floating ice, and pinning-point analysis suggests the thinning began at least fifty years ago. The picture that emerges is one of warm water circulating beneath the ice, melting it from below, thinning the floating tongues, reducing their buttressing effect, and allowing the grounded glaciers upstream to accelerate, thin and retreat.

The study also documents a striking rhythm of calving. The ice shelves of Porpoise Bay underwent near-synchronous calving events in 2006 to 2008, again in 2016 to 2017, and in a previously unidentified episode in 2021 to 2022, each time advancing back to within about a kilometre of their previous maximum extent before calving again. The team found that these events coincided with unusually low sea-ice concentrations in the preceding winter and spring, and with the break-up of the iceberg mélange, the jumble of sea ice and icebergs that presses against the ice front. Sea ice and mélange exert a resistive backstress on calving fronts, and their removal has been linked to ice-shelf breakup elsewhere in Antarctica. Because calving events at Porpoise Bay removed more than the roughly 18 percent of ice-shelf area that modelling identifies as dynamically passive, the losses translated into measurable speed-ups of the glaciers behind them, with velocity increases of several percent following the largest events.

Perhaps the most consequential finding concerns Holmes West Glacier. Its grounding line currently rests on a local bedrock high that appears to have shielded it from the full force of the warm water, which may explain its comparatively modest retreat rate of around 70 metres per year. But that pinning point sits at the very edge of a steep retrograde bed. Within two kilometres of inland retreat, the bed drops by 650 to 700 metres, and different bathymetric datasets suggest the reverse slope may continue for 25 to 35 kilometres inland. If the glacier unpins from its bedrock high, which the authors suggest could happen within a decade at current retreat rates, warm water would gain access to deep grounded ice, potentially initiating marine ice sheet instability, the same self-reinforcing retreat mechanism feared at Thwaites Glacier. The Holmes catchment holds the equivalent of roughly 11 centimetres of global sea level rise, and an adjacent ice rise that currently buttresses the flow could also lose its grip.

The authors emphasise that urgent improvements to bed elevation data are needed before the region’s future can be modelled with confidence, since the bathymetric uncertainties in this poorly surveyed corner of Antarctica are large enough to change the prognosis entirely. Nevertheless, the convergence of evidence, thinning that propagates inland, grounding lines retreating across retrograde slopes, modest but persistent flow acceleration, warm water at the shelf break, and deep cavities connecting ocean to ice, paints a coherent picture of a region responding to oceanic forcing. As Antarctic sea ice enters an era of record lows and increasing anomaly persistence, the stabilising seasonal blanket that has repeatedly broken and reformed over Porpoise Bay may grow thinner and less reliable. What happens next at these unassuming glaciers could help determine how much of East Antarctica’s enormous ice reservoir stays put in a warming world.

Subject of Research: Ocean-driven dynamic thinning and grounding line retreat at outlet glaciers in Porpoise Bay, Wilkes Land, East Antarctica

Article Title: Dynamic thinning and grounding line retreat in Porpoise Bay, Wilkes Land, East Antarctica

Article References: Weatherley, M., Stokes, C. R., Jamieson, S. S. R., Ramanath, S., & Silvano, A. (2026). Dynamic thinning and grounding line retreat in Porpoise Bay, Wilkes Land, East Antarctica. The Cryosphere, 20(9), 5509-5532. https://doi.org/10.5194/tc-20-5509-2026

Image Credits: AI Generated

DOI: 10.5194/tc-20-5509-2026

Keywords: East Antarctic Ice Sheet, Porpoise Bay, Wilkes Land, grounding line retreat, dynamic thinning, Circumpolar Deep Water, ice shelf calving, sea ice, marine ice sheet instability, Aurora Subglacial Basin, satellite remote sensing, sea level rise

Using rare in situ measurements from wave buoys deployed in the Vincennes Bay Polynya, a research team led by Joey Voermans of the University of Melbourne, working with colleagues from China, Germany, the United Kingdom, the United States, Norway and Australia, has documented something unexpected. The strong offshore winds that keep the polynya open do more than simply blow newly formed ice away from the coast. They also generate energetic wind waves within the open water itself, and those locally generated waves are powerful enough to fracture the surrounding ice cover. Crucially, the breakup propagates in the opposite direction to the pattern scientists usually associate with wave damage in polar seas.

The conventional picture of wave-induced sea-ice breakup comes from the Southern Ocean, where long-period swells generated by distant storms travel hundreds or even thousands of kilometres before reaching the ice edge. When these swells arrive, they flex and fatigue the ice at its outer boundary, breaking floes apart progressively from the outside in. The fractures radiate inward, and the ice edge erodes over time. This outside-inward mechanism has been studied extensively, both from satellites and in wave tanks, and it underpins most current models of how waves and sea ice interact across the vast marginal ice zones that ring Antarctica.

The Vincennes Bay observations reveal a mirror image of that process. Because the polynya is a region of open water embedded within the ice pack, waves generated inside it by the katabatic and synoptic winds grow across the fetch of the polynya and then strike the surrounding ice from within. The result is an inside-outward breakup pattern: fracturing begins at the polynya’s edge and works its way outward into the surrounding ice field. The team’s buoy records captured energetic wave conditions during strong wind events, providing direct observational evidence that the polynya is not merely a passive hole in the ice but an active wave-generation zone capable of reshaping its own boundaries.

This distinction matters because the two breakup regimes operate on different wave physics. Southern Ocean swells are long, low-frequency waves that can propagate through sea ice over considerable distances, attenuating gradually as their energy is scattered by floes and dissipated by ice deformation. Locally generated wind waves inside a polynya are typically shorter and steeper, having grown only across the limited fetch of open water available to them. Their interaction with the ice edge is correspondingly more abrupt, concentrating flexural stress near the boundary where the ice is thinnest and most vulnerable. The study demonstrates that these shorter, locally born waves are nonetheless energetic enough to fracture the ice cover and expand the region of broken floes around the polynya.

To assess how significant these events are beyond the duration of a single buoy deployment, the researchers turned to two decades of satellite records spanning 2003 to 2022. By examining sea-ice concentration data alongside atmospheric conditions, they identified extreme wave-induced breakup events and traced their consequences across the surrounding ice pack. The satellite analysis showed that these events coincide with substantial reductions in sea-ice concentration, and, strikingly, that the reductions persist for at least a week after the breakup itself. In other words, a single episode of wave-driven fracturing leaves a lasting imprint on the ice landscape, long after the winds that generated the waves have subsided.

The satellite records also revealed a shift in the type of ice present after breakup events. Following wave-induced fracturing, the affected areas showed increased transitions to frazil ice, the loose, slushy crystals that form in supercooled, turbulent water before consolidating into new ice sheets. Frazil ice is the raw material of the polynya’s ice factory, and an increase in frazil production signals that the open, wave-agitated water is actively churning out new ice. The waves, in effect, feed back into the very process that defines the polynya: by breaking the surrounding ice, they enlarge and sustain the open water area, which in turn allows more wind-driven heat loss, more supercooling, and more frazil generation.

This feedback loop has implications that extend well beyond the immediate vicinity of Vincennes Bay. Antarctic coastal polynyas are disproportionately important sites of dense water formation, a key link in the global overturning circulation that ventilates the deep ocean. They are also biological hotspots, where the combination of open water, intense mixing and rapid ice production supports productive marine ecosystems. If waves play a larger role in polynya dynamics than previously recognized, then the models used to simulate sea-ice production, dense water formation and polar ocean circulation may need to account for wave-ice interactions inside polynyas, not just at the open-ocean ice edge where most wave-ice research has traditionally focused.

The findings arrive at a moment when Antarctic sea ice is under scrutiny. After decades of relative stability, Antarctic sea-ice extent has shown pronounced variability and record lows in recent years, prompting intense debate about the role of ocean heat, atmospheric forcing and internal feedbacks. As the climate changes, wind patterns, storm tracks and ocean conditions around Antarctica are all shifting. The authors note that as Antarctic coastal polynyas evolve under a changing climate, wave activity may become increasingly important in shaping their sea-ice conditions and their evolution. A polynya that experiences more frequent or more energetic wave events could break its surrounding ice more often, stay open longer, and produce different ice types than one governed by wind alone.

What makes the study particularly valuable is the rarity of its observations. Deploying and maintaining wave buoys inside an Antarctic polynya is logistically demanding, requiring support from teams at Australia’s Casey Station and instrumentation capable of surviving one of the harshest marine environments on Earth. Direct measurements of waves within polynyas are scarce, which is precisely why polynya dynamics have been framed almost exclusively in terms of wind and thermodynamics. By pairing these hard-won buoy records with a twenty-year satellite perspective, the researchers have transformed a theoretical possibility into a documented mechanism. The Vincennes Bay Polynya, it turns out, does not simply exist because the wind says so. It exists because the wind and the waves work together, with the waves carving the ice from the inside out and leaving fingerprints on the Antarctic ice pack that satellites can still see a week later.

Subject of Research: Wave-driven inside-outward breakup of sea ice in an Antarctic coastal polynya

Article Title: Inside-outward breakup of sea ice by locally generated waves in the Vincennes Bay Polynya

Article References: Voermans, J. J., Liu, Q., Cao, L., Kousal, J., Heil, P., Fraser, A. D., Collins, C. O., Rabault, J., & Babanin, A. V. (2026). Inside-outward breakup of sea ice by locally generated waves in the Vincennes Bay Polynya. Nature Communications. https://doi.org/10.1038/s41467-026-77979-y

Image Credits: AI Generated

DOI: 10.1038/s41467-026-77979-y

Keywords: sea ice, polynya, ocean waves, Antarctica, Vincennes Bay, wave-ice interaction, frazil ice, Southern Ocean, katabatic winds, sea-ice concentration, polar oceanography, climate change

The iron hypothesis, proposed by oceanographer John Martin in the late 1980s, rests on a simple biological fact. In vast stretches of ocean known as High-Nutrient Low-Chlorophyll regions, which cover up to twenty percent of the ocean surface, phytoplankton starve for lack of iron even though nitrate and phosphate abound. These microscopic plants use iron as a co-factor in the enzymes of photosynthesis and nitrogen fixation, so when iron arrives, blooms follow. Martin argued that during cold, dusty glacial periods, windblown dust delivered extra iron to these waters, stimulating marine productivity and drawing down atmospheric carbon dioxide, which was eighty to one hundred parts per million lower during ice ages than today. Subsequent research confirmed the mechanism but steadily shrank its estimated contribution, with models attributing perhaps twenty parts per million of the glacial decline to iron-driven fertilization.

To test the hypothesis where it should have mattered most, the team turned to the EGRIP ice core, drilled within the Northeast Greenland Ice Stream about 360 kilometers from the Greenland Summit. The ice they analyzed spans the Pleistocene-Holocene transition, from 13,000 to 10,300 years before present, capturing the Younger Dryas cold event, the warmer Bølling-Allerød interval before it, and the Early Holocene after. Crucially, much of the dust that settled on Greenland during the Younger Dryas originated in the Gobi and Taklamakan deserts of Asia and the Sahara, and a large share of it traveled across the North Pacific before reaching the ice sheet. That geography means Greenland ice records changes in atmospheric iron deposition over the iron-starved Subarctic North Pacific, the very region where marine sediment cores show that productivity stubbornly refused to rise during dusty cold periods.

The technical challenge lies in measuring not just iron, but the right kind of iron. The researchers used the Bern Continuous Flow Analysis system, which melts ice at 2.8 centimeters per minute and channels the meltwater from the pristine inner core of each sample to an array of detectors. Two iron fractions emerged from this pipeline. The first, called Fe_ICP, was quantified by single-particle inductively coupled plasma time-of-flight mass spectrometry and captures most of the atmospheric iron pool, excluding only iron locked inside refractory silicate minerals. The second, dubbed labile iron or LFe, was measured by online acidification of the meltwater at a pH of roughly 1.6 followed by spectrophotometric detection. This mild acid treatment releases iron bound in colloids, hydroxides and organic complexes, the forms most easily grabbed by the siderophore molecules that phytoplankton deploy to scavenge iron from seawater. Labile iron is therefore an upper bound on the iron potentially available to marine life.

The contrast between the two fractions is the heart of the discovery. Median Fe_ICP concentrations surged seventeen-fold between the Early Holocene and the Younger Dryas, rising from 1.76 micrograms per liter to 30.16 micrograms per liter, a signal consistent with the known dustiness of the cold period. Yet labile iron climbed only twenty-nine percent, from a median of 0.51 to 0.66 micrograms per liter after excluding volcanic horizons, a statistically robust difference confirmed by a Wilcoxon rank sum test on more than 21,000 data points. As a result, the labile share of total iron collapsed from roughly forty percent during the Holocene to a mere two percent during the Younger Dryas. In other words, the ocean received vastly more dust, but almost none of the extra iron came in a form plankton could use.

The culprit, the researchers argue, was chemistry. Ice-core acidity measurements reveal that Greenland snow was mildly acidic during the Early Holocene, at a median of 1.06 micro-equivalents per liter, but turned distinctly alkaline during the Younger Dryas, at minus 4.93 micro-equivalents per liter, because carbonate-rich dust from the deserts neutralized the aerosol. Iron solubility is exquisitely sensitive to pH: theoretical calculations show that a single unit increase in pH can reduce dissolved iron(III), the thermodynamically stable form, by nearly two orders of magnitude. That swing matches the gap between the seventeen-fold rise in total iron and the modest rise in labile iron almost exactly. The Bølling-Allerød, still slightly alkaline, showed even lower labile iron concentrations than the Holocene, reinforcing the link between alkalinity and iron lock-up.

The hemispheric comparison makes the case even more striking. In Antarctic ice from EPICA Dome C, labile iron rose as much as twelve-fold between the Holocene and the Antarctic Cold Reversal, contributing up to sixty-four percent of the total iron content during that cold interval. Antarctica stayed acidic even during glacial times, likely because abundant biogenic sulfur dioxide emissions prevented complete neutralization of the aerosol, keeping iron dissolved and potentially bioavailable. This chemical divergence helps explain why sediment records show that iron fertilization boosted productivity in the Subantarctic zone of the Southern Ocean during ice ages, while the North Pacific, bathed in alkaline dust, saw no comparable response over the past 800,000 years.

If not iron, what did control North Pacific productivity during the last glacial transition? The study points to sea ice and ocean stratification. Sediment records from the eastern and western Subarctic Pacific and the Bering Sea indicate extensive spring sea-ice cover during the Last Glacial Maximum, precisely when iron fluxes peaked, and sea ice acts as a physical barrier that blocks both light and direct iron deposition. Productivity rose as perennial ice retreated, peaked during the largely ice-free Bølling-Allerød, and dipped again amid variable ice during the Younger Dryas. Meanwhile, nitrogen isotope measurements from foraminifera show that nitrate consumption was more complete during the Younger Dryas despite low productivity, a paradox explained by stronger stratification: cold, fresh surface waters suppressed vertical mixing, cutting off the upwelling of nutrient-rich deep water even as iron arrived from above.

There is one dramatic exception to the iron drought, and it comes from volcanoes. By identifying ten volcanic horizons through spikes in acidity and electrical conductivity, the team found that labile iron soared to as high as 17 micrograms per liter during eruptions, more than twenty-five times background levels. Acidic volcanic gases react with ash surfaces to create soluble iron salts such as iron chlorides, fluorides and sulfate hydrates, and ash also carries phosphorus, silica, zinc, manganese and copper. At least three interhemispheric eruptions, dated to 10,481, 12,917 and 13,028 years before present, could have triggered transient phytoplankton blooms in the North Pacific, though such blooms would last only a few years and leave no trace in sediment cores. The authors also note a modern echo: rising aerosol pH from changing industrial emissions may have locally altered iron availability in recent decades.

The study does carry caveats. Microbial activity in ice can reduce insoluble iron(III) to more soluble iron(II), and post-depositional mineral weathering can form jarosite and other poorly soluble secondary minerals, but the authors judge these effects limited and broadly uniform across the shallow, 150-meter depth interval examined, leaving the relative changes intact. Even so, the implications are significant for climate science and for any future proposals to fertilize the ocean deliberately. The iron hypothesis, the work suggests, is not a simple matter of dumping dust on the sea; whether iron nourishes plankton depends on the acidity of the atmosphere that carries it, the chemistry of the aerosol, and the physical state of the ocean below. During the Younger Dryas, the Northern Hemisphere failed that test on every count, and the North Pacific stayed hungry despite a seventeen-fold feast of dust.

Subject of Research: Atmospheric iron bioavailability and its role in North Pacific marine productivity during the Younger Dryas

Article Title: Limited atmospheric iron availability increase during the Younger Dryas in the Northern Hemisphere

Article References: Burgay, F., Derrod, H., Erhardt, T., Scoto, F., Segato, D., Maffezzoli, N., Dallo, F., Zannoni, D., Spagnesi, A., Kjær, H.-A., Fischer, H., Varin, C., Barbante, C., & Spolaor, A. (2026). Limited atmospheric iron availability increase during the Younger Dryas in the Northern Hemisphere. Climate of the Past, 22(9), 1675-1689. https://doi.org/10.5194/cp-22-1675-2026

Image Credits: AI Generated

DOI: 10.5194/cp-22-1675-2026

Keywords: iron hypothesis, EGRIP ice core, Younger Dryas, North Pacific, labile iron, aerosol acidity, phytoplankton, HNLC regions, marine productivity, volcanic ash, sea ice, Climate of the Past