Ice shelves, circling the edges of the Antarctic Ice Sheet as large floating aprons of ice, play an important role in the stability of the continential ice sheet. Like a viscous fluid, the ice sheet flows under its own weight and out towards the ocean, with most of the ice concentrated into glaciers and ice streams. These glaciers, once they hit the ocean, go afloat and form ice shelves; often hundreds of meters thick and 10s of kilometers across. Essentially, ice shelves then act as a “buttress” to hold back the ice sheet upstream, modulating its flow into the ocean. One might imagine this scenario as a cork in a bottle, laid on its side. Holding liquid inside, corks work best at restraining flow when they are intact and properly sized for the bottle. Cutting out some of the cork, or having a crumbling cork, would not properly contain the liquid inside. Subsequently, when the ice sheet is in balance, accumulation and compaction of snowfall replenishes any ice mass lost from iceberg calving and melt from the ice shelves, and results in no change to sea level. However, when ice shelves thin, retreat, fracture, and crevasse, their buttressing capabilities are diminished, and like a crumbly cork, let more and more of the ice sheet behind them flow out into the ocean, adding to sea level rise.

 Ice shelves, particularly those around the West Antarctic Ice Sheet, have been losing mass since the 1990s. Ice shelves in the Amundsen Sea, for example Thwaites Glacier, are known to be holding back enough ice to raise global sea levels by several meters; their rapid weakening is a significant cause for concern when it comes to predicting sea level rise in the coming decades. Timely prediction of global sea level rise depends on our ability to understand and predict how ice shelves will evolve. Crucially, one outstanding question looms large: what happens when an ice shelf completely collapses?

In the satellite record, we have observed roughly eight to ten complete ice-shelf collapses in the dramatic “disintegration” style, where most of the ice shelf is lost over a period of days to weeks. This style of breakup is striking in its suddenness compared to the slow processes that drive most glacier dynamics over years to decades. Most of these events, so far, have occurred on the Antarctic Peninsula, where warm air temperatures enables significant surface melting, resulting in the formation of surface melt ponds. Strikingly, in 2022, during an observation campaign over a neighboring ice shelf in East Antarctica, we noticed that the Conger-Glenzer ice shelf, about 1,200 square kilometers, suddenly disappeared over just a few days’ time. In those final days, there was no sign of surface melt; just a dry, weathered surface filled with cracks and depressions.

Using the ever-increasing satellite record back to the 1990s (and in some cases, the 1960s!), our new study follows the Conger-Glenzer Ice Shelf’s evolution through time and finds that long-term thinning over decades, induced by a warming ocean, weakened the ice shelf slowly. Over that time, its pinning points, small rocky islands propping up the ice shelf over the ocean and typically stabilizing the flow of ice, became like slow-moving rocks in a windshield – maybe thick glass wouldn’t shatter, but thin glass would. These features transitioned into destructive features in its final years, driving its fractured appearance that we observed in its final days. Finally, in its weakened state, an unprecedented atmospheric river made landfall nearby in March 2022, bringing with it strong winds and large ocean swells. While the storm didn’t cause the collapse, its approach did hasten Conger-Glenzer’s demise, with the 1,200 square kilometer ice shelf disintegrating over a few days before the height of the storm.

While the Conger-Glenzer Ice Shelf was a small ice shelf by Antarctic standards, fed by relatively small and slow-moving glaciers–thus its collapse has little impact on sea level rise–the event marks a first for East Antarctica. Long considered more stable than the West Antarctic Ice Sheet, this event added an exclamation point to the accelerating narrative of unexpected changes being observed in East Antarctica. What’s more, the collapse was driven by ocean-induced thinning and structural weakening rather than by surface melt. Though most other ice shelves around East Antarctica are much thicker than Conger-Glenzer was, this event raises the spectre of ice shelf collapse in locations that are not susceptible to surface melt but do have the potential to significantly impact sea level, i.e., much of East Antarctica, which holds about 90% of Antarctica’s ice. In this way, Conger-Glenzer’s collapse serves as a small example of how to observe future changes and highlights our evolving understanding of early warning indicators for potential ice shelf collapse in the future.