Unlike other climate sciences—oceanography, for instance—there’s not a lot of great historical data on what glaciers in the world’s remote parts have been doing over the course of the last few hundred years. But Willis’s new paper might point to something historically notable. In general, glaciers like the Vavilov Ice Cap are predictable They’re machines trying to balance the input of snow and the output of climate influences on the ice, and generally, in the absence of forces changing the climate, quite stable, he says. “These surges are interesting because they don’t do that,” he says, which raises some questions about how we currently model this kind of glacier. Current predictions about melting may not be taking these kinds of movements from the glacier into account—or, if they’re cyclical, Schoof says, can’t say what will happen to the cycle in the presence of climate change.