Hidden warm-water traps beneath Antarctica may be melting the continent’s ice far faster than scientists realized.
Global sea levels could rise faster than scientists once predicted, according to new research pointing to a hidden source of Antarctic ice loss. The study suggests that warmer ocean water is melting Antarctic ice shelves from underneath much more efficiently than expected.
Ice shelves are giant floating extensions of glaciers that help slow the movement of massive amounts of ice into the ocean. Researchers in Norway have now identified a process that may weaken these natural barriers. Long channel-like formations beneath the ice shelves can trap relatively warm seawater, dramatically increasing melting in specific areas.
If these ice shelves become thinner and less stable, the glaciers behind them may flow into the sea more quickly. That could speed up global sea level rise well beyond many current estimates.
Scientists have already seen similar patterns in other parts of Antarctica. The Intergovernmental Panel on Climate Change (IPCC) has identified unstable polar ice shelves as a major climate concern, although the process remains difficult to fully understand and model.
Hidden Channels Beneath Antarctic Ice Shelves
The team focused on the Fimbulisen Ice Shelf in East Antarctica to better understand how underwater melting occurs. Their results showed that the shape of the underside of the ice shelf has a major influence on how ocean water circulates below it.
In areas where the underside contains channels, water movement can form small circulation systems that keep warmer water trapped against the ice rather than allowing it to move away. This lingering heat intensifies melting in those regions.
Researchers found that melting rates inside these channels can increase by roughly an order of magnitude locally. In simple terms, the shape of the ice shelf determines where ocean heat gathers and how much melting it can trigger.
“We found that the shape of the ice shelf underside is not just a passive feature. It can actively trap ocean heat in exactly the places where extra melting matters most,” lead author Tore Hattermann from the iC3 Polar Reseach Hub in Tromsø, Norway explains.
Fimbulisen Ice Shelf lies in East Antarctica, an area generally considered colder and less vulnerable than other parts of the continent.
“We observed beneath the Fimbulisen Ice Shelf that even small amounts of warmer water can substantially increase melting within the channels,” Tore Hatterman says. “As a result, the channels can grow and, in the worst case, weaken the stability of the entire ice shelf.”
Qin Zhou, who co-led the study, adds that “What is striking is that even modest inflows of warmer deep water can have a large effect when the ice shelf base is channeled. That means some ice shelves that scientists usually think of as cold may be more fragile than expected.”
Modeling Antarctic Ice Melt
To study the phenomenon, researchers combined a detailed map of the underside of the Fimbulisen Ice Shelf with a high-resolution computer model of the ocean cavity beneath it.
The team compared simulations using smoother ice shelf bases with versions that included realistic channels, under both cooler and slightly warmer ocean conditions. This approach allowed them to isolate how the channels affect water circulation, mixing, and melting.
The work also incorporated earlier field observations from the region. Researchers say combining long-term measurements with advanced modeling is critical for understanding the small-scale features hidden beneath Antarctic ice shelves. Hattermann himself has spent hundreds of days conducting fieldwork on Antarctic ice shelves.
Why Faster Antarctic Ice Melt Matters
Scientists warn that stronger melting within the channels could create a dangerous cycle. As channels deepen and widen, parts of the ice shelf may thin unevenly, weakening the shelf’s overall structure.
A weaker ice shelf is less able to slow the glaciers behind it, potentially allowing more land ice to flow into the ocean.
“Current climate models do not capture this effect,” Tore Hattermann warns. “This means that they risk underestimating the sensitivity the ‘cold’ ice shelves along East Antarctica’s coastline to small changes or warming in coastal waters. Such changes have already been observed, and are projected to increase in the future.”
The findings could have major implications for climate science and coastal planning. Researchers say ice sheet and climate models need to better account for these small-scale melting processes to improve future sea level projections. The changing flow of meltwater could also affect ocean circulation and marine ecosystems around Antarctica.
The study, “Channelized topography amplifies melt-sensitivity of cold Antarctic ice shelves,” was published in the journal Nature Communications.
Reference: “Channelized topography amplifies melt-sensitivity of cold Antarctic ice shelves” by Qin Zhou, Tore Hattermann, Chen Zhao, Rupert Gladstone, Julius Lauber, Petteri Uotila and Ashley Morris, 7 May 2026, Nature Communications.
DOI: 10.1038/s41467-026-71828-8
The research was led by Tore Hattermann from the iC3 Polar Research Hub and Qin Zhou from Akvaplan-niva (joint first authors). Both scientists are based in Tromsø, the capital of Arctic Norway. Tore is an assistant lead of the iC3 research group that develops and applies novel technologies for cryospheric science.
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
5 Comments
A simple notion. As for ice-shelves slowing glaciers’ flow rates. What price inertia?
I would suggest that as the ice shelves grow in the Winter, the new ice added onto the leading edge of the floating glacier will inherit the inertia of the glacier, thus not impeding its forward motion at all.
Saltwater typically freezes/melts at about -2°C, varying with the salinity of the water, as does the density; however, the brines that are extruded from ice crystals that are growing can potentially stay liquid above -21°C, therefore, they sink rapidly once released into saltwater, and immediately start mixing with the seawater. Albeit, Wikipedia states, “The coldest seawater still in the liquid state ever recorded was found in 2010, in a stream under an Antarctic glacier: the measured temperature was −2.6°C.” Saltwater at a typical salinity of 35 parts per thousand reaches a maximum density at −1.3°C. These are not unimportant facts! Small changes can influence what happens under glaciers and therefore their interpretations of measurements.
Unfortunately, no temperatures can be found in the above press release summary. The best that one can do is guess that “warm deep water” is warmer than the glacier ice. But by how much? Is it very much warmer, or just slightly warmer? I’m not just criticizing this research, I’m pointing out that it is common for at least summary papers in climatology to skimp on numbers. Even for this peer review paper one has to burrow about 2,400 words into the text to discover that their working definition for “warmer water” is >−1.7°C. It is not good form to make their peers guess just what they did and what measurements they relied on because it makes it more difficult to evaluate their claims.
In closing, I’d like to draw the reader’s attention to the following quote from the actual peer-reviewed article: “However, processes governing the ocean-driven thinning remain poorly constrained, leading to uncertainty in sea-level rise projections.” I would call that an understatement.
Uncertainty is not your friend.
The ice shelves are becoming an upside down Canyonlands.