thwaites_nasa-800x449.jpg
The frozen continent of Antarctica holds a lot of ice. Like, a lot of ice. So finding out how much of it is melting into the oceans—raising sea level—and how much will melt in the future is a huge task. Not only do we need comprehensive measurements covering the most remote region on Earth, but even small measurement errors equate to very large differences in total ice mass.

This is largely a job for satellites, and we've got several different types of measurements being made from above, measuring things like ice surface elevation and even the gravitational attraction of the mass of ice. There are complicating issues, though, like the fact that the continent itself is responding to current and past ice loss by slowly rebounding upward. And we still need on-the-ground weather monitoring to track the accumulation of snow, among a host of other things.
The complex, multi-pronged nature of this effort means that researchers frequently publish separate estimates of change based on the type of data they are collecting, rather than integrating all sources of information. These numbers can naturally differ, making it hard to put your finger on one answer.

Enter the huge collaboration of Antarctic researchers called the IMBIE Project (the Ice Sheet Mass Balance Inter-comparison Exercise). This week, that collaboration published a new analysis of all the available data from dozens of studies, producing an overall best estimate of Antarctic ice loss between 1992 and 2017.

Jumping straight to that result, IMBIE finds that Antarctica lost 2,720 ± 1,390 gigatons of ice in that time period—enough to raise global sea level 7.6 millimeters on its own. The rate of ice loss has increased, though, averaging about 43 gigatons per year over the first 10 years and rising to 220 gigatons per year in the last 5 years.

Antarctica is far from uniform, and one number can’t tell the whole story. The ice is often separated into three regions—the East and West Antarctic Ice Sheets (separated by the Transantarctic Mountains), and the Antarctic Peninsula (the continent’s turtle tail). The West Antarctic Ice Sheet is far more vulnerable to melting because its bowl-shaped bedrock drops below sea level, and most of its glaciers contact the ocean.
Over the whole time period, the West Antarctic Ice Sheet has been losing ice at an average rate of 94 ± 27 gigatons per year. Despite the small size of the Antarctic Peninsula, the destabilizing breakup of ice shelves like the Larsen B has driven loss there at an average rate of 20 ± 15 gigatons per year. The huge but much more stable East Antarctic Ice Sheet, on the other hand, is estimated to have (probably) gained 5 ± 46 gigatons per year.

The biggest reason the error bars on the East Antarctic Ice Sheet trend are so large is that slow upward rebound of the bedrock beneath the ice. It’s extremely difficult to precisely work out what that hidden bedrock is doing compared to what the ice and snow at the surface are doing, complicating the interpretation of surface measurements.

As scientists usually do, the authors also discuss the ongoing developments that will improve future estimates, with that bedrock movement high on the list. Radar measurements from airplanes are also helping researchers map snow depth, as it compresses to denser firn and then ice. That’s important because you have to get from satellite measurements of surface elevation to estimates of the actual amount of frozen water present there.

Because the East Antarctic Ice Sheet generally sits on higher bedrock and isn’t as susceptible to warming ocean water, a warmer atmosphere can bring increased snowfall that temporarily outweighs losses. But elsewhere, Antarctic ice has been falling into the ocean rapidly, and the outlook is for more of the same.