Scientists at the University of Lausanne (UNIL) have used AI to massively speed up computer calculations and simulate the last ice cover in the Alps. Much more in line with field observations, the new results show that the ice was thinner than in previous models. This innovative method opens the door to countless new simulations and predictions linked to climate upheavals. The research is published in Nature Communications.

25,000 years ago, the Alps were covered by a layer of ice up to 2 kilometers thick. For almost 15 years, this glaciation has been put into perspective by 3D digital models, based on climate reconstructions, thermodynamics and ice physics. However, these models have sparked debate in the scientific community, as until now there has not been a full correspondence between these simulations and the physical traces—rocks, moraines, etc.—found in the field, particularly erosion lines, which bear witness to past ice thicknesses.

A team of scientists from the University of Lausanne (UNIL) have just solved this persistent problem. For the first time, they have used artificial intelligence to massively boost their new glacial evolution model, generating a large series of simulations of unprecedented accuracy: they correspond much more closely to the physical traces left on the ground.

Their results show an average ice cover 35–50% thinner than in previous reference simulations. Model resolution has been increased from two kilometers to 300 meters, and it is only thanks to this precision that it is possible to describe the complex topography of the Alps numerically.

In line with the current state of scientific knowledge, based on field observations, it shows, for example, that certain peaks such as the Matterhorn and Grand Muveran were clearly protruding from the ice during the Ice Age.

The research is significant in more ways than one. Firstly, the ability to correctly model the glacial past is essential to understanding our environment. For over 2 million years, the Earth has experienced alternating glacial and warm cycles, which have profoundly shaped the landscape in which we live.

The new model now corresponds much more closely to the evidence left on the ground following the retreat of the glaciers, and make it possible to better quantify many natural phenomena, such as glacial erosion, which has largely contributed to sculpting the relief of the Alps.

On the other hand, the innovative methodology used in this research marks a new era in numerical modeling. “By using recent technology, and applying it to the last major glaciation in the Alps, we can finalize a 17,000-year simulation at very high resolution (300 m) in 2.5 days, whereas such spatial resolution would have taken 2.5 years to calculate using traditional methods, which are also extremely costly and energy-intensive,” explains Tancrède Leger, researcher at UNIL’s Faculty of Geosciences and Environment (FGSE), and first author of the study.

With this approach, the model first learns about the physics of ice flow, using Deep Learning methods. It then receives data on the climate of the period (temperature, precipitation, etc.), to calculate ice supply and melt.

Deep learning calculations are then performed not by the traditional central processing unit (CPU), but via a GPU (or graphics processing unit), which enables numerous operations to be performed in parallel, boosting the computer’s computing power phenomenally.

“It’s as if we once had six Ferraris at our disposal, and now we have ten thousand small cars. We’ve gone from very large machine clusters to a simple 30 cm graphics card,” illustrates Guillaume Jouvet, FGSE professor behind the AI model and co-first author of the study.

“We’re not doing anything new, but we’re doing it a thousand times faster, making it possible to achieve resolutions that were not even considered before.”

This progress will enable new research to be launched. In particular, a new project is about to get underway to use this revolutionary method to better predict the impact of the melting Greenland and Antarctic ice sheets on global sea level rise.