Between January 2022 and March 2023, Hektoria Glacier on the Antarctic Peninsula lost roughly 25 kilometers of grounded ice, about 15 miles, retreating faster than any grounded glacier in the modern observational record. The most violent phase lasted just two months. In November and December 2022, more than 8 kilometers of the glacier’s front broke apart in rapid succession, a rate nearly ten times faster than any previously documented episode of grounded ice loss.
Two peer-reviewed studies, published in Nature Geoscience and The Cryosphere, have now reconstructed the collapse in detail. Together with seismic records and publicly available satellite data, they reveal a glacier that crossed a physical threshold scientists had theorized about but never witnessed at this scale. NASA’s Earth Observatory has described the episode as one of the fastest grounded ice losses ever recorded.
The trigger: a shield of sea ice shatters
For years, a thick layer of multi-year landfast sea ice filled the Larsen B embayment, the bay where Hektoria meets the ocean. That frozen barrier acted as a buttress, pressing against the glacier’s front and helping hold it in place. In January 2022, the landfast ice broke apart and drifted away.
The Cryosphere study documented the breakout and the glacier’s immediate response. With its protective cover gone, Hektoria’s terminus was suddenly exposed to open water and the mechanical forces of waves and currents. The retreat began almost at once.
A collapse mechanism scientists had not seen at this scale
What followed was not ordinary iceberg calving, where cracks propagate through a glacier’s face and chunks fall into the sea. The Nature Geoscience study identified a different process: buoyancy-driven calving on an ice plain.
An ice plain is a section of glacier that barely touches the seafloor. Ocean water circulates beneath it, thinning the ice from below. As the thinning progressed through 2022, sections of Hektoria’s ice plain reached a point where they could no longer stay grounded. They lifted off the bed, became buoyant, and broke apart in large blocks.
The study measured 8.2 ± 0.2 kilometers of retreat during November and December 2022 alone. “That is nearly an order of magnitude faster than previously published values for grounded ice loss,” the authors reported, making it the most extreme short-duration retreat on record for ice that had been anchored to bedrock.
How satellites and seismometers tracked the collapse
Multiple independent observation systems captured the event in real time. Terminus positions were mapped using imagery from Landsat, WorldView, and Planet satellites. The resulting shapefiles and retreat-rate calculations are publicly available through the U.S. Antarctic data portal, allowing any researcher to verify the measurements independently.
ICESat-2 laser altimetry revealed something that satellite photos alone could not: the glacier’s surface was sinking during the austral winter of 2022, even when the terminus appeared stable in optical imagery. That thinning was the precursor. The ice plain was being eaten from below, setting the stage for the explosive calving that erupted months later.
Three seismic networks, POLENET-Antarctica, the Antarctic Seismographic Argentinean Italian Network, and the Global Seismograph Network, detected glacial earthquakes during the retreat. These quakes have distinct signatures that differ from tectonic events, and their timing confirmed the transition from grounded to floating conditions at the calving front. Each seismic signal marked a moment when a section of ice detached from the bed or from adjacent blocks.
Where Hektoria fits in the bigger picture
Hektoria is not a household name like Thwaites, the so-called “Doomsday Glacier” in West Antarctica. It is smaller, and its direct contribution to sea-level rise from this single event is modest, likely on the order of millimeters. But glaciologists are paying close attention because of what the collapse reveals about process.
A continent-wide grounding-line migration record covering 1992 through 2025, built from multi-mission radar interferometry and published in the Proceedings of the National Academy of Sciences, shows that the northeast Antarctic Peninsula is experiencing accelerated grounding-line retreat. Hektoria’s collapse fits squarely within that pattern: warm ocean water is reaching glacier bases across the region, destabilizing ice that had been anchored to bedrock for decades or longer.
The concern is that the mechanism documented at Hektoria, buoyancy-driven calving on a thinning ice plain, could operate at larger glaciers with far greater ice volumes. If warm water continues to intrude beneath marine-terminating glaciers around Antarctica, the same sequence of thinning, flotation, and rapid breakup could repeat at scales with significant consequences for global sea levels.
What scientists still do not know
Despite the strength of the satellite and seismic evidence, several critical gaps remain.
No direct, in-situ measurements of ocean temperature or current velocity beneath Hektoria during the 2022 retreat have been published. Scientists infer that warm water drove the thinning based on surface-elevation changes captured by ICESat-2 and the physical behavior of the ice plain, but the precise thermal conditions under the glacier during the critical months are not directly observed. That gap limits the ability to model how fast similar glaciers elsewhere might respond to comparable ocean forcing.
The grounding-line positions during the retreat were tracked primarily through optical satellite imagery of the terminus, not through contemporaneous radar interferometry that can directly detect where ice lifts off the bed. The PNAS dataset provides three decades of context, but its temporal resolution may not capture the month-by-month dynamics of a collapse this rapid. Pinning down exactly when individual sections transitioned from grounded to floating would require higher-frequency radar observations targeted specifically at Hektoria during the critical window.
There is also no published attribution of the January 2022 sea-ice breakout to specific climate drivers. Whether the breakout resulted from anomalous ocean heat, atmospheric warming, unusual wind patterns, or some combination has not been isolated in the peer-reviewed literature. The broader warming trend in the region suggests such events are likely to become more frequent, but the timing and triggers of future breakouts remain uncertain.
What happens to Hektoria now
The most vulnerable, near-floating sections of the glacier’s front have already broken away. What comes next is an open question.
One possibility is a period of relative stability as the glacier adjusts to a new grounding-line position farther inland. Another is continued retreat if warm ocean water keeps accessing deeper parts of the bed. Current ice-sheet models cannot resolve the trajectory with confidence because they must approximate poorly known variables: subglacial topography, basal friction, and the geometry of whatever ice plain remains.
As of mid-2026, researchers are watching Hektoria closely with satellite imagery and planning future field campaigns to deploy oceanographic instruments beneath the glacier. The data from those efforts will determine whether the 2022-2023 collapse was a one-time adjustment or the opening phase of a longer disintegration.
Either way, the event has already reshaped how glaciologists think about the speed at which grounded ice can fail. A glacier that had been stable behind its sea-ice shield for years lost 15 miles of ice in 15 months once that shield disappeared. The physical process responsible, buoyancy-driven calving on a thinning ice plain, is now documented in detail for the first time at this magnitude. For other glaciers around Antarctica sitting on similar ice plains, with warm water creeping beneath them, Hektoria is both a case study and a warning.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
