A comprehensive new analysis of satellite data and direct ice measurement surveys published in Nature Climate Change confirms that ice loss from the Antarctic ice sheet is accelerating at a pace that exceeds the projections of most climate models used in the most recent Intergovernmental Panel on Climate Change assessment, raising questions about whether those models adequately capture the dynamic processes driving ice sheet instability and what the discrepancy implies for sea level projections in the decades ahead.

The study, conducted by a consortium of researchers from 11 institutions across the United States, United Kingdom, Australia, and France, analyzed 25 years of satellite elevation, velocity, and gravitational anomaly data alongside field measurements from ground-based monitoring stations on the Antarctic continent. The researchers found that the rate of ice mass loss from West Antarctica and parts of East Antarctica has roughly tripled over the period studied compared to the baseline rate observed in the late 1990s, when systematic satellite measurement of the ice sheet began.

The drivers of accelerated ice loss are multiple and interacting. Ocean warming is the primary physical mechanism: warmer water penetrating beneath floating ice shelves - the extensions of glaciers that extend over the ocean and help hold back inland ice - is melting them from below at rates that were not anticipated in early model scenarios. As ice shelves thin and break up, they exert less restraining force on the glaciers behind them, allowing ice to flow more rapidly toward the ocean. This dynamic is particularly pronounced in the Amundsen Sea sector of West Antarctica, where the Thwaites Glacier - sometimes called the "doomsday glacier" in popular science writing because of its potential contribution to sea level rise - has become a focus of intensive monitoring and research.

Sea level implications are the primary concern driving scientific and policy attention to Antarctic ice loss. The West Antarctic ice sheet alone contains enough ice to raise global average sea levels by approximately 3.3 meters if it were to fully collapse over a long period, though that scenario represents an extreme and very long-duration outcome. More near-term projections focus on the contribution to sea level rise this century. The IPCC's Sixth Assessment Report projected a likely range of global mean sea level rise of 0.3 to 1.0 meters by 2100 under high-emission scenarios, but noted that ice sheet instability processes could push that figure significantly higher in scenarios it characterized as low-probability but not negligible.

The new study suggests that the IPCC's ice sheet models may be structurally conservative in ways that make even their high-end projections potentially optimistic. Several of the feedback mechanisms that are driving the accelerated ice loss observed in the satellite data - including the interaction between ice shelf collapse and grounding line retreat, and the role of marine ice cliff instability in enabling rapid ice calving - are either poorly represented or absent from most operational climate models. Incorporating these dynamics could shift sea level projections toward higher values for the same emission scenarios.

Coastal planning and infrastructure investment decisions are already being complicated by uncertainty about sea level rise projections. Cities with significant low-lying coastal infrastructure - including Miami, New York, London, Bangkok, Jakarta, Mumbai, and Shanghai - are managing a planning environment in which the range of plausible outcomes over the 50-to-100-year planning horizons relevant for major infrastructure decisions is wider than earlier scientific assessments suggested. Engineers and urban planners increasingly need to design to a range of scenarios rather than a single expected outcome, which has significant implications for both cost and the political feasibility of securing public support for large-scale adaptation investments.

The research community emphasized that the findings do not change the fundamental policy conclusion that rapid and sustained greenhouse gas emission reductions are the most important lever for limiting Antarctic ice loss and its sea level consequences. The physical systems driving ice loss have significant inertia, meaning that some additional warming and ice loss is already locked in as a consequence of past emissions, but the magnitude of future ice loss and sea level rise over the rest of this century and beyond is strongly dependent on whether emissions trajectories are brought in line with the goals of the Paris Agreement in the coming decades.

Glaciologists said the monitoring infrastructure that has enabled the improved characterization of Antarctic ice dynamics needs to be sustained and expanded, noting that several satellite missions critical to ice sheet observation are approaching the end of their operational lifetimes and that replacement missions need to be funded and launched within the decade to maintain continuity of the observational record. A gap in the satellite data record would seriously impair the scientific community's ability to detect and characterize changes in ice dynamics in near-real time, which is increasingly important for both scientific understanding and operational sea level risk assessment.