Antarctica’s ice cliffs are collapsing faster than models predicted, but the retreat stops short of the runaway chain reaction scientists had feared, according to a study published in the journal Science in October 2026.
Researchers from MIT and the Woods Hole Oceanographic Institution (WHOI) analyzed 15,330 satellite observations of ice cliffs at the margins of Earth’s ice sheets, collected between 2019 and 2023 by NASA’s Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2), a spacecraft that measures ice and ocean elevation with a laser altimeter.
The finding cuts both ways. Tall cliffs did fracture and retreat faster than existing models anticipated, which points to more short-term ice loss.
Yet every glacier that retreated eventually stabilized rather than collapsing inland without limit.
“We found that when the cliffs were retreating, they were doing so faster than what has been modeled, which would mean more ice loss and higher contributions of sea level rise,” said the study’s lead author, a graduate student in the MIT-WHOI Joint Program.

“But because we found that the retreat was short-lived, and all the glaciers eventually stabilized, that would suggest that things won’t be so bad. We don’t see this runaway effect.”
Scientists Put the Collapse Theory to the Test
The study set out to test marine ice cliff instability, or MICI, a theory that has shaped sea-level discussions for years. It holds that once an ice shelf breaks away and exposes a cliff taller than roughly 100 meters, about the height of the Statue of Liberty, the ice should collapse under its own weight, expose an even taller wall inland, and set off a self-sustaining cycle of retreat.
Models built on that idea have been simple. “Predictive ice-sheet models that have been used so far to represent ice cliff instability just take into account cliff height,” one researcher said. “That’s the only variable that matters. And once the cliff height reaches the critical threshold, usually 100 meters, it should initiate this runaway effect of ice sheet retreat.”
The satellite record suggested otherwise. The average cliff in the survey stood about 30 meters tall, well below the danger line, and only a small number crossed 100 meters.
| Measured cliff height | What the survey found |
|---|---|
| ~30 meters | Average cliff height across more than 15,000 observations |
| 100 meters | The threshold where instability models predict runaway retreat |
| 127 meters | Tallest of four cliffs the team tracked in detail |
Tall Cliffs Collapse, Then Stop
After the global survey, the team zoomed in on four locations around Antarctica with the tallest cliffs on record — each roughly 100 meters high, the tallest reaching 127 meters — and reconstructed how they changed at three-month intervals across the four-year study.
At first the results seemed to confirm the theory. The towering cliffs did collapse, and the glaciers behind them retreated at speeds substantially greater than MICI models predicted.
“But that retreat was short-lived,” the lead author said.
Instead of collapsing their way inland, the glaciers stabilized and in places advanced back toward the ocean. “It’s this rapid retreat, punctuated by periods of stabilization,” the researcher said. “That was puzzling to us. Then, we thought: We need to look at all the things that affect cliff height.”
Two overlooked factors emerged. One was mélange, the slushy mix of sea ice and iceberg fragments piling up in front of a cliff. “We measured mélange height and found taller cliffs have thicker mélange, which could be a built-in stabilizing effect,” the researcher said. Deeper water, which adds buoyant support, also appeared to matter, according to the study coverage.
The other was the bedrock beneath the ice. Raised sections of the underlying terrain thin the ice above them, so a tall cliff that retreats onto higher ground quickly shrinks into a shorter one that resists further collapse. That turns a supposedly endless chain reaction into bursts of fast retreat followed by calm.
What This Means for Sea-Level Projections
The implications for coastal planning run in both directions. If models underestimate how fast tall cliffs break apart, they may understate the pace of ice loss over short periods. But treating every collapse as the start of an unstoppable cascade could exaggerate long-term retreat, and the flooding risks built on those projections.
The study was supported by the National Science Foundation and the DOMINOS project, part of the International Thwaites Glacier Collaboration, according to ScienceDaily. Its co-authors include an assistant scientist at WHOI and an associate professor at the University of Michigan.
The researchers are careful about the limits. The findings do not mean Antarctica’s ice sheet is safe from further losses, and they do not rule out future episodes of sustained instability. Outcomes will depend on changing climate conditions and the physical makeup of individual glaciers.
The next task, the team says, is building bedrock topography, ice thinning and water conditions into ice-sheet models. “Now we know what to look for,” the lead author said. What remains unknown is whether the stabilization seen between 2019 and 2023 will hold under a warming climate.