Mapping tipping risks from Antarctic ice basins under global warming

nature climate change Article https://doi.org/10.1038/s41558-025-02554-0 Mapping tipping risks from Antarctic ice basins under global warming Received: 21 October 2024 Accepted: 30 December 2025 Published online: xx xx xxxx Check for updates Ricarda Winkelmann Torsten Albrecht 1,2 1,2,3 , Julius Garbe , Jonathan F. Donges 1,3 1,2,4 & The Antarctic Ice Sheet is subject to amplifying feedbacks which can accelerate ice loss and lead to effectively irreversible retreat. We here analyse the distinct nature and risk of long-term ice loss for each individual drainage basin under different levels of warming. Depending on topographic and climatic conditions, we find that ice loss in some basins unfolds gradually with warming, whereas other basins are characterized by a critical threshold or tipping point beyond which large parts eventually disintegrate. A first threshold, potentially as low as 1–2 °C above pre-industrial levels, triggers the long-term collapse of ~40% of marine ice volume in West Antarctica. Marine-based sectors in East Antarctica, representing ~5 m of potential sea-level rise, are at risk of losing stability at 2–5 °C. Our results imply that the Antarctic Ice Sheet does not act as one single tipping element, but rather as several tipping systems interacting across drainage basins. The Antarctic Ice Sheet is the largest ice sheet on Earth with a mass equivalent to nearly 60 m of global sea-level rise potential1. Its stability and future response to a warming climate is therefore highly relevant for coastal communities, infrastructure and ecosystems2. Under future anthropogenic climate change, the ice sheet is ‘projected to lose mass at an increasing rate throughout the twenty-first century and beyond (high confidence)’2, which could commit future generations to long-term sea-level rise3,4, with subsequent impacts including coastal erosion, ecosystem loss, human livelihood and infrastructure displacement, increased hazards from storm surges and potential groundwater salinification5. Ice loss from Antarctica would also affect the Southern Ocean and could lead to a weakening of Antarctic bottom water formation6, which would have cascading effects on the global ocean and climate7–9. Palaeorecords and modelling suggest that Antarctica has undergone periods of large-scale and abrupt ice loss in the past10–13. During past interglacial warm periods that were only slightly (~1–3 °C) warmer than today despite broadly comparable (~300–400 ppm) atmospheric CO2 concentrations, the Antarctic Ice Sheet probably contributed several metres to global sea level14,15, implying substantial retreat of marine ice-sheet regions in both West12,16,17 and East Antarctica18–20. In particular, meltwater pulses due to accelerated ice-sheet retreat in Antarctica during the last glacial termination might have caused sea levels to rise at rates of up to ~0.7 m per century (or ~7 mm yr−1) (ref. 21). On the basis of these palaeoreconstructions as well as modelling studies and process understanding, the Antarctic Ice Sheet is deemed a tipping element in the climate system22–25. This means that beyond a critical threshold (or several thresholds), self-sustaining feedbacks can lead to abrupt and often irreversible ice loss, with far-reaching impacts on the Earth system via global sea-level rise and changes in atmospheric and oceanic conditions and circulation patterns. Observations indicate that in particular the West Antarctic Ice Sheet has been losing mass at an accelerating pace over the last decades, leading to increasing contributions to global mean sea-level rise26,27. The Amundsen Sea Embayment sector in West Antarctica shows first signs of destablization in response to ocean-induced thinning that reduces ice-shelf buttressing28–30. Also, in Wilkes Land in East Antarctica, increased ice discharge has been observed in response to recent warming31. While palaeoreconstructions and climate modelling suggest that snowfall in Antarctica will probably increase with global warming32–34— which can mitigate some of the expected ice loss35,36—enhanced ablation, dynamical losses and amplifying feedbacks will probably dominate the overall mass balance in the future37,38. Earth Resilience Science Unit, Potsdam Institute for Climate Impact Research (PIK), Member of the Leibniz Association, Potsdam, Germany. 2Integrative Earth System Science, Max Planck Institute of Geoanthropology, Jena, Germany. 3Institute of Physics and Astronomy, University of Potsdam, Potsdam, Germany. 4Stockholm Resilience Centre, Stockholm University, Stockholm, Sweden. e-mail: 1 Nature Climate Change Article https://doi.org/10.1038/s41558-025-02554-0 0 –2 –4 –6 Topography (km) 2 km Brunt, Riiser-Larsen 0 1,000 Enderby Land Filchner Weddell Sea 500 Dronning Maud Land East AP (Larsen D-G) North AP (Larsen C) Ronne Antarctic Peninsula Recovery subglacial basin Amery West AP (George VI) West, Denman East Antarctica Bellingshausen Sea Abbot, West Antarctica Venable Thwaites, Pine Island Aurora subglacial basin Amundsen Sea Getz Totten, Moscow Ross West (Siple Coast) Ross Sea Wilkes subglacial basin Ross East (Byrd) Cook, Ninnis, Mertz Victoria Land 1 0 2 4 6 8 2 5 10 Threshold temperature (°C) Sea-level potential (m SLE) Fig. 1 | Risk map of Antarctic ice catchment basins. Map of Antarctica showing the 18 ice-sheet drainage basins as used in this analysis (thin black lines; ref. 84) as well as their sea-level potential (in metres sea-level equivalent, m SLE), illustrated by the size of the respective circles. Nested circles show the critical temperature levels at which the strongest ice loss occurs in the model simulations (circle colour) as well as the fraction of ice volume lost in the long term upon transgression of those thresholds with respect to the initial ice volume of the basin (circle size). Background shading shows the bedrock topography (tan– brown above sea level, white–blue below sea level); ice shelves are highlighted by grey shading. AP, Antarctic Peninsula. Observed Antarctic topography from the Bedmap2 dataset (ref. 95). Among the most prominent amplifying feedbacks are the surface melt–elevation feedback39,40, the melt–albedo feedback41, the marine ice-sheet instability42,43 (MISI) and the potential marine ice cliff instability11,44,45 (MICI); and further amplifying feedbacks have been suggested46–48. As surface melt is still very limited in Antarctica because of the cold surface temperatures, the melt–elevation and melt–albedo feedbacks are probably going to become more relevant under future, considerably warmer, conditions49. Current mass loss is dominated by ocean-driven subshelf melting50–53 and iceberg calving51,54. In marine ice-sheet regions, where the ice rests on bedrock below sea level, this can trigger MISI, an amplifying feedback between grounding-line retreat and the ice flux across the grounding line42,43. In fact, recent studies suggest tha (...truncated)