30 Years of Antarctic Satellite Data Reveals 12,820 Square Kilometers of Lost Ice

Antarctica is not uniformly melting. That distinction - between the large stable majority and the acutely vulnerable minority - is one of the central findings from the most comprehensive ground-line mapping of the continent ever assembled. Published in the Proceedings of the National Academy of Sciences, the study draws on three decades of satellite data to produce a circumpolar record of exactly where Antarctica's grounded ice has moved and at what pace.

The answer, in aggregate, is 12,820 square kilometers lost since 1996. About 10 times the area of Greater Los Angeles. An average retreat rate of 442 square kilometers per year from the grounding line - the boundary where continental ice lifts off the bedrock and begins to float.

"We've known it's critically important for 30 years, but this is the first time we've mapped it comprehensively across all of Antarctica over such a long time span," said lead author Eric Rignot, UC Irvine Distinguished Professor and senior research scientist at NASA's Jet Propulsion Laboratory.

The Stable 77 Percent and the Vulnerable 23

Three-quarters of Antarctica's coastline showed no measurable grounding line change since 1996. That stability matters. It means most of the ice sheet is not yet responding to ocean warming in any detectable way - a finding that reconciles some conflicting estimates of Antarctic mass balance from different measurement methods.

The losses are concentrated. West Antarctica's Amundsen Sea and Getz sectors drove the majority of retreat, with individual glaciers moving back dramatically over three decades. Pine Island Glacier retreated 33 kilometers from its 1996 position. Thwaites Glacier - which has received intense scientific attention because of its potential contribution to sea level rise - pulled back 26 kilometers. Smith Glacier showed the largest single retreat: 42 kilometers.

"Where warm ocean water is pushed by winds to reach glaciers, that's where we see the big wounds in Antarctica," Rignot said. "It's like the balloon that's not punctured everywhere, but where it is punctured, it's punctured deep."

An Unexplained Exception on the Antarctic Peninsula

The warm-ocean explanation works for West Antarctica. It does not fully explain what is happening on the northeast Antarctic Peninsula, where the team documented substantial grounding line retreat despite a lack of evidence for warm subsurface ocean water in proximity.

Edgeworth Glacier lost 16 kilometers from its 1996 position. The Hektoria, Green, and Evans glaciers calved 21, 16, and 9 kilometers past their former grounding lines, respectively. Multiple ice shelves in the region collapsed before the study period even began, removing the buttressing support that slows glacier flow.

"On the east coast of the peninsula, there's substantial retreat, and we don't have evidence for warm water," Rignot said. "Something else is acting - it's still a question mark."

Atmospheric warming, surface meltwater infiltration into crevasses, and changes in wind-driven dynamics have all been proposed as contributing mechanisms. Disentangling them requires the kind of continuous monitoring this study was designed to enable.

Commercial Satellites Enter Polar Science

The study also represents a milestone for NASA's Commercial Satellite Data Acquisition program. Data from Airbus U.S. and ICEYE US - a commercial synthetic aperture radar provider based in Irvine - contributed alongside agency missions from the European Space Agency, Canada, Japan, Italy, and Germany.

"This work shows how commercial SAR data can be used to contribute to the virtual SAR constellation by augmenting the program of record from agency-run missions," said co-author Bernd Scheuchl. Daily observations of critical areas, made possible by commercial assets, enable the kind of temporal resolution that long-return-time agency satellites cannot provide alone.

What the Record Is For

Grounding line positions are direct inputs to ice sheet models used for sea level projections. A 30-year observational record provides something those models have lacked: a continuous, circumpolar benchmark against which model performance can be evaluated.

"Models have to demonstrate they can match this 30-year record to claim credibility for their projections," Rignot noted. "If a model can't reproduce this record, the modeling team will need to go back to the drawing board."

The fact that 77% of Antarctica is still stable is a meaningful constraint. It rules out model configurations in which warming-driven instability is already propagating widely across the continent. But it also defines the problem precisely: the instability that is occurring is concentrated, deep, and accelerating in at least some sectors.