Antarctic ice sheet became more climate-sensitive after the Mid-Pleistocene Transition, study in Nature Geoscience finds
A team at the IBS Center for Climate Physics at Pusan National University reports in Nature Geoscience a three-million-year climate simulation showing that Antarctic ice responds in threshold fashion at carbon concentrations below about 240 ppm.
- A Pusan University team simulated Earth's climate over the past three million years.
- After the Mid-Pleistocene Transition around a million years ago, Antarctica responds in threshold fashion.
- The critical carbon dioxide concentration for sudden shifts sits at about 240 ppm.
- The paper was published in Nature Geoscience.
A team led by Kyung-Suk Jun at the IBS Center for Climate Physics at Pusan National University in South Korea published in Nature Geoscience a simulation showing that the Antarctic ice sheet became markedly more sensitive to climate change after the so-called Mid-Pleistocene Transition, a geological event that occurred about one million years ago.
The simulation covers three million years of continuous global climate integration, one of the longest such runs in the literature. It finds that when carbon dioxide concentration falls below 240 parts per million, Antarctic ice grows abruptly and dramatically, shifting into an expanded ice-sheet state.
What the Mid-Pleistocene Transition is
The Mid-Pleistocene Transition is a period in Earth’s climate history during which the rhythm of glacial cycles changed from cycles of about 41,000 years into cycles of about 100,000 years. Glacial periods became longer and more intense, with colder minima and warmer maxima. The exact mechanisms behind that transition remain debated in the geophysics community.
Antarctica was, until now, assumed to have responded gradually throughout this period, with proportional ice growth or melting. The new simulation finds otherwise. After the transition, the ice sheet behaves as a thermodynamic system with a threshold, capable of sudden, non-proportional changes.
Simulation methodology
The Jun team used a coupled climate model and ice-sheet model running in continuous integration. The difference from earlier approaches lies in the length of the simulation:
- a total of 3 million years of simulated time
- integration from scratch with all Earth’s orbital cycles affecting incoming solar energy
- atmospheric carbon dioxide inputs taken from polar ice-core records
- regular cross-checks against geological indicators of sea level and deep-ocean temperature
The simulation ran on South Korea’s national supercomputers, with processor resources equivalent to months of compute time.
What this means for future projections
After this transition, Antarctic ice responds much more strongly to changes in climate forcing. The system does not evolve gradually; it becomes significantly more responsive after a threshold is crossed in the climate system.
Kyung-Suk Jun stresses that the results change how future sea-level projections should be read. Climate models projecting the future usually assume linear sensitivity of Antarctic ice to temperature. The study suggests that assumption is likely wrong and that sudden, non-linear changes are possible.
What the study does NOT show
The authors stress that the finding does not mean Antarctica is about to be triggered in the near future. Today’s global carbon level is about 425 parts per million, well above the simulated 240 ppm threshold for rapid ice growth. The study identifies a general system sensitivity, not a present-day event.
A second limitation concerns model resolution. While the time scale is long, the spatial resolution of the ice model is coarse, so precise projections for specific parts of the Antarctic sheet require further simulations on finer grids.
A third point: the Mid-Pleistocene Transition itself remains debated. Different theories invoke different mechanisms, from ocean salinity changes to erosion of basal sediments under the northern ice sheets. The study does not resolve that question.
Link to Serbian climate science
The result is of interest to the Republic Hydrometeorological Service of Serbia and the Geophysical Institute of SANU, both of which already take part in international future-climate simulations under the CMIP process. Local researchers coordinated by Vladimir Djurdjevic work on regional Balkan models whose future sensitivity depends on how the global sea level behaves, and therefore on Antarctic ice.
Should the discovery of a threshold sensitivity in Antarctic ice change emissions policy?
Share your view: is it time to present climate projections explicitly as possible threshold jumps rather than only gradual changes? Tell us in the comments.
