Melting From Below: Rising Ocean Heat Could Trigger Sudden Collapse of Greenland’s Ice Mélange

Along the frozen frontiers of Greenland, where colossal tidewater glaciers meet the sea, a fragile barrier of floating ice plays an outsized role in deciding how fast the island’s ice sheet slips into the ocean. This barrier, known as ice mélange, is a dense, jumbled raft of icebergs and sea ice fragments packed tightly against glacier fronts. It may look like chaotic debris, but it behaves in ways that matter enormously for sea level. New research published in Nature Communications by Paul T. Summers of Tufts University and the Georgia Institute of Technology and colleagues shows that these ice fields are far more precarious than previously appreciated: as warming ocean waters eat away at them from below, ice mélange may not weaken gradually but instead collapse suddenly, with potentially dramatic consequences for glacier retreat.

The stakes are high. The Greenland Ice Sheet has been responsible for roughly 20 percent of global sea level rise since 2002, and nearly half of that ice loss is moderated by the presence of ice mélange at the fronts of its tidewater glaciers. These floating aggregations of ice act as a physical brake on calving, the dramatic process by which chunks of glacier ice break off and become icebergs. When mélange is thick and rigid, it presses back against the glacier terminus, suppressing the release of large icebergs and slowing the glacier’s seaward flow. When mélange weakens or disappears, that restraint vanishes, and glaciers can accelerate their discharge of ice into the ocean.

Yet scientists have long struggled to explain why ice mélange behaves so differently from one Greenland fjord to another. Observations reveal a remarkable spread of behavior: some fjords see their mélange collapse seasonally each spring, disintegrating over just a few days, while others host mélange that persists year-round, seemingly indefinitely. Understanding what tips a mélange from durable to doomed has been a stubborn puzzle, and one with direct implications for forecasting how quickly Greenland’s glaciers will shed ice in a warming climate.

To crack the puzzle, the research team built a coupled model that brings together two worlds of physics that are usually simulated separately. The first component captures high-resolution ocean circulation within the fjord, including the turbulent plumes of meltwater that rise along glacier fronts and the transfer of heat from the ocean to the ice. The second component models the mélange itself as a dynamic granular material, a flowing collection of discrete ice blocks that jam, shear, and fracture much like sand or gravel. This granular treatment is essential, because ice mélange does not behave like a continuous sheet of ice; its strength emerges from the crowding and friction among thousands of individual bergs pressing against one another.

Running multi-year simulations with this coupled framework, the researchers were able to reproduce the observed seasonal variability of ice mélange in real fjords. In their simulations, the mélange follows an annual rhythm governed by the competition between ice supplied by calving at the glacier front and ice removed by melting and by export out of the fjord. During cold seasons, newly calved bergs consolidate into a rigid mass, and the mélange grows in extent and strength. As conditions warm, the balance shifts, and the fragile structure becomes vulnerable to rapid disintegration.

The crucial insight from the new work is the mechanism of collapse itself. Rather than eroding steadily in step with melt rates, the model reveals that mélange possesses a threshold behavior. Below a critical level of submarine melting, the mélange can persist indefinitely, repairing itself seasonally with fresh calved ice. But once melting from below pushes the system past that threshold, the internal structure of the mélange loses its rigidity. The granular fabric that gives the mélange its buttressing power unravels quickly, and the entire ice field can collapse in a matter of days, mirroring the abrupt springtime disintegrations that observers have documented in Greenlandic fjords.

To test whether this modeled threshold holds in the real world, the team parameterized the collapse condition and compared it against observations from five fjords across Greenland. The comparison showed that the model’s threshold captures the observed spread of mélange behavior, distinguishing fjords where mélange persists from those where it breaks down each year. This kind of validation matters, because it suggests the threshold could serve as a practical diagnostic, a way to assess which of Greenland’s glaciers are currently protected by stable mélange and which are hovering near the tipping point.

The findings carry an unsettling implication. Perennial ice mélange is currently present at the majority of the largest glaciers in Greenland, which means some of the ice sheet’s biggest contributors to sea level rise are being held in check by these floating braces. The new results indicate that many of these barriers may be capable of sudden collapse if subjected to increased submarine melt, the process by which relatively warm Atlantic-sourced waters, circulating deep in the fjords, erode ice from beneath the waterline. Because ocean heat content around Greenland has been rising, and because warm waters are increasingly able to reach the deep channels that feed glacial fjords, the reservoirs of melt potential pressing against these mélange fields are growing.

If a perennial mélange does collapse, the consequences would cascade. Freed from the back-pressure of the mélange, the glacier terminus becomes free to calve more vigorously, and the removal of resistive floating ice can allow the glacier to speed up its flow toward the sea. Faster flow means more ice delivered across the grounding line and into the ocean, accelerating glacier mass loss and, ultimately, sea level rise. The nonlinear nature of the collapse means this acceleration could arrive abruptly rather than gradually, complicating efforts to project Greenland’s contribution to future sea levels on decadal timescales.

The study, which was supported by the National Science Foundation through the GLACIOME project, represents a step forward in treating the ocean, the ice, and the fractured mélange between them as a single interacting system. It also highlights how much remains to be learned: fjord bathymetry, seasonal oceanography, and calving dynamics all shape where each glacier sits relative to its collapse threshold. But the central message is clear. Greenland’s glacial fjords are not destined to lose their ice mélange slowly and predictably. Some of these frozen barriers could hold for years and then vanish in days, and the warming ocean pressing against their undersides is steadily pushing more of them toward that edge.

Subject of Research: Threshold collapse of ice mélange at tidewater glaciers driven by increasing submarine melt

Article Title: Sudden collapse of ice mélange under increasing submarine melt at tidewater glaciers

Article References: Summers, P. T., Robel, A. A., Jackson, R. H., King, M. D., Amundson, J. M., & Burton, J. C. (2026). Sudden collapse of ice mélange under increasing submarine melt at tidewater glaciers. Nature Communications. https://doi.org/10.1038/s41467-026-78423-x

Image Credits: AI Generated

DOI: 10.1038/s41467-026-78423-x

Keywords: Greenland Ice Sheet, ice mélange, tidewater glaciers, submarine melting, calving, sea level rise, granular flow, fjord oceanography, glacier retreat, Nature Communications, climate change, cryosphere