This year, scientists reported alarming news about the huge continental ice sheet covering the western portion of Antarctica: It’s headed for collapse, due to rapid melting of some of its buttressing ice shelves. When it does, global sea levels will rise by several meters. It has long been suspected that warm ocean waters at the base of those floating ice shelves are responsible for hurrying things along. But with scant data from the waters around Antarctica, that has been difficult to prove. Now, a new study that pieces together 40 years’ worth of data collected in multiple regions around Antarctica suggests that scientists have found the smoking gun: Warming waters are indeed sneaking up under the floating ice in the regions of fastest melting.

Many questions about Antarctic climate-ocean-ice feedbacks remain, however. The places where warm waters are promoting melting are powerfully linked to current wind patterns around Antarcticaand these are also subject to change in a warming world.

Last May, a pair of studies suggested that melting has advanced enough that the ice sheet, called the West Antarctic Ice Sheet (WAIS), is well along a path to inevitable collapse. In several centuries’ time, its demise would raise global sea levels by about 3 meters, the studies concluded. This week, scientists offered more of that stark picture, noting that the melting rate for the WAIS has tripled over the last decade.

Ice and ocean scientists have long suspected “basal meltingmelting of the underbelly of some West Antarctic ice shelves by ocean waters, says climate scientist Eric Rignot of the University of California, Irvine, who was not involved in the new study. “We have known for about 20 years now that [the ice shelves are] bathing in warm waters, Rignot says. “But we did not know anything about the trend of the temperature of these waters.

The likely culprit behind basal melting is a relatively warm (about 2°C) water mass within the Southern Ocean called Circumpolar Deep Water (CDW), which rings the entire continent, flowing just at the edge of the continental shelf, where it merges into open ocean. But how, and why, portions of that water mass might be moving inland and circulating under some of the ice shelves had not been well documented.

“There are quite a few localized studies done with subs in canyons that show what is happening close to these ice shelves, says Sunke Schmidtko, an oceanographer at the GEOMAR Helmholtz Centre for Ocean Research Kiel in Germany and lead author of the new study. “But they’re all missing the link between what’s happening on the [continental] shelf and in the open ocean.

To understand how the CDW has been changing over time, Schmidtko and his colleagues turned to seven publicly available databases that aggregated 40 years’ worth of data on conductivity (a measure of salinity), temperature, and depth of the CDW and other (colder) water masses around the continent.

The team’s findings, reported online today in Science, showed that water temperatures in the CDW along the West Antarctic continental shelf have been on the rise since the 1970s. Moreover, the authors say, those warm waters have also been sneaking up higher along the shelf in some areas. The regions with the most rapid meltingparts of West Antarctica known as the Bellingshausen and Amundsen seasare also the ones receiving those warm water influxes, they showed. In West Antarctic regions that show far less melting, however, such as the Ross and Weddell seas, the warm CDWs have not intruded.

What’s driving these regional differences in intrusions of warm water, Schmidtko says, appears to be the current wind patterns.

But these, the authors note, are subject to change, due not only to greenhouse warming but also to the recovering ozone hole. That, in turn, could bring warmer waters to regions now dominated by cold waters, such as the southern Weddell Sea, he adds. “That could have consequences for glaciers that do not belong to West Antarctica and could be affected for the first time.

The results tell “a very compelling story, says Sarah Gille, an oceanographer at the Scripps Institution of Oceanography in San Diego, California. The study demonstrates that the water under the shelves is warming, she says. That answers one questionbut it raises many others about how to accurately model the dynamics of ocean-ice-atmosphere interactions around Antarctica and predict the future of the ice sheets, she says.

Some scientists sound a note of caution about the paper’s reported trends. “It’s an interesting paper in that it puts together a pan-Antarctic set of data, something that hasn’t really been done before, says glaciologist Ian Joughin of the University of Washington, Seattle. But the lack of data in many regionsparticularly in critical places like the Amundsen Seamakes it difficult to place strong confidence in the trends the authors report, he says.

Oceanographer Stanley Jacobs of the Lamont-Doherty Earth Observatory in Palisades, New York, offers a stronger critique. The authors used an algorithm to try to fill in the spatial gaps in water mass data around the continent. But given the observed large seasonal and spatial variability in water temperatures near Antarcticaeven in just the summer months“one should be wary of trend significance there, Jacobs says. The assumption that one can link ice shelf melting to water masses with temperatures that have been averaged on a regional scale “is not justified, he says.

This underscores “how little data has been collected, particularly in the critical Amundsen Sea, Joughin says. “How movement of the Circumpolar Deep Water evolves is crucial to understanding ice sheet stability, and much better sampling going forward is required.

2014 American Association for the Advancement of Science