New modelling provides more precise forecasts for Antarctic ice sheet melting


Image of Antarctic landscape. Credit: Jim Marschalek

New modelling has challenged some existing sea level projections, shedding new light on the potential impact of rising sea levels on coastal areas.

Using historical records of sea level rise, an international team of researchers at Imperial College London, University of New Hampshire, Australian National University, Columbia University, University of Victoria, and Harvard University has revealed the most precise estimate to date of past Antarctic ice sheet melt, which could provide a more realistic forecast of future sea level rise.

The Antarctic ice sheet is a key contributor to sea level rise globally, containing over 30 million cubic kilometres of water. Hence, its melting could have a devastating impact on future sea levels.

However, forecasts are not very precise – the most recent Intergovernmental Panel on Climate Change (IPCC) report suggested that sea levels could go up anywhere from 0.3 to nearly two metres by 2100.

By reassessing old records of our shorelines, new modelling has estimated that the future Antarctic contribution to global mean sea level rise may be less than previously thought. In a more likely worst-case scenario, there could be a rise of up to one metre by 2100 and up to three metres by 2300.

The findings, published in Science Advances, imply that the most concerning recent estimates, involving a rise of up to 1.6 metres by 2100 and 16m by 2300, are unlikely – providing a more optimistic outlook for coastal areas.

Studying the past to understand the future

Lead author Dr Fred Richards, a lecturer in the Department of Earth Science and Engineering at Imperial College London, said:

“By revealing how the Antarctic ice sheet behaved in the past, we can better understand how it might behave in the future. To accurately model how much sea levels could rise by in the next century, we sought to gain a deeper understanding of sea levels during the Mid-Pliocene era, around 3 million years ago, the most recent period in Earth history with climatic conditions similar to what we expect in the coming decades.”

“The traditional measurements used to gauge past ice volume and sea levels involve studying old shorelines – and we recently realised these measurements are not very precise. This is because the elevations of these features are affected by mantle flow – movements deep within the Earth, which can push the land up and down by hundreds of metres. Previous sea-level estimates have largely ignored this process, challenging our current understanding of ancient ice volumes and sea level rise.”

To determine historic sea levels, researchers first looked at the geological record of Australia to find fossilised corals and other sea-level markers that indicate how high the shoreline used to be when it was originally formed.

They compiled a detailed database of Mid-Pliocene shorelines across Australia and developed advanced simulations of mantle flow. Using machine learning, they made new estimates of past sea level based on these complex simulations, and fine-tuned their predictions about future sea levels by aligning them with their enhanced understanding of the Mid-Pliocene period.

New forecasts offer optimism, but researchers caution that further work is needed

Researchers warn that, while a lower estimated contribution by the Antarctic ice sheet is good news, there is still plenty of work to be done.

Co-author Dr Mark Hoggard, from the Australian National University, said:

“If you live in a Pacific Island nation like Tuvalu where the highest point of elevation is only 4.6 meters, small changes in the baseline sea level can have devastating impacts when disaster events like cyclones or storm surges hit.

“Ensuring we have more accurate models can help improve policy, especially when looking at coastal and low-lying communities which can be impacted by just centimetres of sea level change.”

The next steps for the international team will involve expanding the database of Mid-Pliocene shorelines globally, refining computational models, and continuing to explore the impact of mantle-driven changes on ice sheets.

“Geodynamically corrected Pliocene shoreline elevations in Australia consistent with midrange projections of Antarctic ice loss” by Richards et al., published 17 November 2023 in Science Advances.


Diana Cano Bordajandi

Diana Cano Bordajandi
Department of Earth Science & Engineering

Click to expand or contract

Contact details

Tel: +44 (0)20 7594 6445

Show all stories by this author


Climate-change, Engineering-Earth-Sci-and-Eng, Geology
See more tags