From Fimbulisen In Norway To The Filchner-Ronne In Antarctica: A Body Of Evidence Building Across Nature Communications, Nature Climate Change And IPCC Reports That Should Reshape How Humanity Thinks About Sea Level Rise
This extensive analysis-driven article on the current state of Antarctica was compiled with the help of the following sources: ScienceDaily, PreventionWeb, EcoMagazine, EurekAlert!, Nature, Antarctic and Southern Ocean Coalition and PubMed Central.
There is a version of Antarctica that climate models have been describing for decades. It involves ice shelves slowly thinning, glaciers gradually accelerating toward the ocean, and sea levels rising by somewhere between half a meter and one meter by the end of the century. It is an alarming picture. But a growing body of peer-reviewed research now suggests it may also be a dangerously incomplete one.
A study published on May 10, 2026 in Nature Communications, led by Tore Hattermann of the iC3 Polar Research Hub in Tromsø and Qin Zhou of Akvaplan-niva, identifies a mechanism that current global climate models do not capture at all. Beneath floating ice shelves, long channels carved into the underside of the ice can trap relatively warm ocean water, creating small overturning circulation cells that hold heat against the ice rather than allowing it to pass through and disperse. The researchers found that melting within these channels increases by roughly an order of magnitude in some locations compared to smoother sections of the same shelf.
The study focused on the Fimbulisen Ice Shelf in East Antarctica, a region historically considered among the most stable portions of the continent. The researchers combined high-resolution mapping of the ice shelf base with advanced ocean cavity modelling, testing both smooth and channelled ice surfaces under cooler and warmer ocean conditions. The results showed that even modest intrusions of warm deep water produce dramatically amplified melting where the topography concentrates that heat.
The significance of this lies not just in the specific finding about Fimbulisen but in what it implies about the entire East Antarctic ice sheet, which contains far more ice than West Antarctica and has generally sat outside the most urgent projections.
Hattermann’s warning is direct and worth quoting in full. “Current climate models do not capture this effect. This means that they risk underestimating the sensitivity of cold ice shelves along East Antarctica’s coastline to small changes in coastal water temperatures. Such changes have already been observed and are projected to increase.”
The IPCC has long acknowledged the problem these researchers are now quantifying. In its 2021 Assessment Report, the IPCC projected sea levels would rise between 28 and 55 centimetres by 2100, but explicitly noted that a rise of over six feet could not be ruled out under a worst-case scenario, and identified polar ice shelf instability as the primary source of that upper-end uncertainty.
The Norwegian study aligns directly with that acknowledgement while sharpening its edge. Separately, research published in Nature Climate Change identified another dimension of the same structural problem. Using a regional ocean model to project ocean-driven ice shelf melt in the Amundsen Sea, researchers concluded that already-committed rapid ocean warming drives increased basal melt regardless of emissions scenario, suggesting that extensive ice loss from West Antarctica may now be unavoidable even with aggressive greenhouse gas mitigation. The implication carries serious weight: even optimistic emissions pathways may not prevent the destabilisation of ice sheets that contain enough ice to raise global sea levels by more than five meters.
An independent line of research reinforces the channel-trapping mechanism from a different angle. A study from the ARC Australian Centre for Excellence in Antarctic Science found that ice shelves in East Antarctica experience summer melting spikes when sea ice retreats and warm ocean water flows beneath them. The study’s lead author Dr Fabio Boeira Dias described basal melting as a major driver of Antarctic Ice Sheet instability and identified it as largely overlooked by current climate models. Co-author Dr Adele Morrison from the Australian National University stated explicitly that if basal melting in East Antarctica is not properly accounted for, future sea level rise may be considerably underestimated.
Research published in Science Advances on the Ross Ice Shelf adds another observational dimension. Scientists at the University of East Anglia identified a 50-metre-thick intrusion of warm surface water immediately beneath the Ross Ice Shelf and found that the heat transport driving near-front melting has increased over the past four decades, driven by rising heat content in coastal waters.
A critical counterpoint comes from modelling work published in Nature, which examined tipping points in ice shelf cavities using an Earth system model with interactive ice shelf dynamics. Those researchers found that previous ice sheet modelling may have overestimated future ice shelf melt in some regions, particularly the Filchner-Ronne system, which they showed responds abruptly once a warm-water intrusion threshold is crossed rather than degrading gradually. This nuance matters because it means the risks are not evenly distributed across Antarctica and that some regions may remain more stable for longer before crossing irreversible thresholds, not because they are safe but because they behave differently.
What emerges from synthesising these perspectives is a picture that is more complex than any single study can convey. The Norwegian research identifies a previously unmodelled mechanism operating in East Antarctica. The Australian work shows that seasonal dynamics in the same region are also missing from standard climate models. The Nature Climate Change study suggests that West Antarctic warming has already crossed a commitment threshold. The UEA research demonstrates that observational heat transport data confirms increasing energy entering ice cavities over decades. And the Nature tipping point study warns that some systems may cross thresholds abruptly rather than gradually, with the Filchner-Ronne ice shelf showing potential for abrupt warm-water intrusion and a massive basal melt increase within this century.
The convergence of these findings across independent institutions, methodologies and regional focuses points toward a shared conclusion that current climate models systematically underestimate both the rate and the variety of mechanisms driving Antarctic ice loss. This is not a marginal discrepancy. Ice shelves function as structural buttresses for the enormous land ice behind them. When they thin and weaken, they stop slowing glacial flow. When glaciers accelerate, land ice enters the ocean. When land ice enters the ocean, sea levels rise.
Hattermann and Zhou argue that integrating channelled ice shelf topography into global climate models is now essential, not just for improving scientific projections but for informing coastal adaptation planning and infrastructure decisions worldwide that currently rely on projections that may significantly understate the threat.
The scientific community does not yet agree on precisely how much additional sea level rise these missing mechanisms represent. That uncertainty is itself the problem. When the upper bound of a risk involves the potential inundation of coastal cities housing hundreds of millions of people, the argument for caution and for urgent model improvement does not require consensus on the exact number. It requires only the recognition that the number is larger than we thought, and that the processes making it larger are already underway.
Antarctica is melting from below. The models that inform global climate policy do not fully see it yet. Like this analysis-heavy extensive article? Read our previous article here.
