Tall ice cliffs retreated faster than models predict without runaway collapse, Needell's team reports

"Up to three times" is a ceiling. New Scientist reports that the thickest glaciers may calve icebergs up to three times faster than previously predicted, and that such retreat may not be sustained over long periods ^[1]. New Scientist did not report how many glaciers were measured, over what period, or whether Thwaites sits at the top of that range.

Until now the runaway scenario rested on idealised mathematical models, because there was not enough empirical data to judge how likely it was ^[6]. Caroline Needell of the MIT-WHOI Joint Program and her colleagues used satellite data to make the first direct, large-scale observations of how the tallest cliffs behave ^[7]. The physical idea under test is that if ice flows fast enough, it thins at the glacier's edge and lowers the cliff, even where the ice is thicker inland ^[9].

"None of the glaciers we observed underwent unstable retreat, even though they did retreat much faster than our current models predict," Needell said ^[11]. That rules out reading fast retreat as proof of instability.

The "bursts" framing is New Scientist's, and the report hedges it with a "may" ^[2]. Douglas Benn at the University of St Andrews offered a picture of how it could look at Thwaites. "Over the next decades, it's going to jump from one pinning ridge to another," he said ^[15]. We read that as the stop-start pattern the report describes: retreat speeds up between ridges and pauses at each one.

The pauses end where the ridges do. "The bed then begins to get real deep," Benn said, and at that point "the whole structural integrity of the ice becomes unviable" ^[16]. New Scientist reports the glacier may reach that point within the current century ^[16].

The study does not show what a glacier does after its last ridge. Needell noted that most glaciers in the study are buttressed by sea-floor features ^[14], so the observations come mostly from glaciers still held in place from below. Her study suggests that thinning might restabilise a glacier once the bed deepens, but the question remains open ^[18].

In our view the open question is now whether runaway collapse happens: retreat faster than models allow has been observed ^[11], and runaway collapse has not. Robert Larter at the British Antarctic Survey points out that it has not yet been seen, though one study suggests it may have happened in the vicinity of Thwaites around 12,000 years ago ^[12]. "The problem is, nobody was around to take observations at the time," he said ^[13].