AI Predicts Major Ice Loss Even With Immediate Climate Action

Even if global warming were to cease entirely, the volume of ice in the European Alps is projected to decline by 34% by the year 2050. However, if the current trend observed over the last 20 years persists, scientists from the University of Lausanne (UNIL, Switzerland) demonstrate in a new international study that nearly half the volume of ice could be lost.

The Aletsch glacier in 2009, in Switzerland. Image Credit: UNIL - Guillaume Jouvet

By the year 2050, approximately 26 years from now, the projection indicates a minimum loss of 34% in the volume of ice in the European Alps. This prediction stems from a new computer model developed by scientists from the Faculty of Geosciences and Environment at the University of Lausanne (UNIL), in collaboration with the University of Grenoble, ETHZ, and the University of Zurich.

Utilizing machine-learning algorithms and climate data, the scenario assumes a halt in global warming in 2022; nevertheless, glaciers persist in experiencing losses due to inertia in the climate-glacier system. Despite being the most optimistic forecast, it remains far from a realistic future scenario, given the ongoing increase in worldwide greenhouse gas emissions.

In Reality, More Than Half the Volume of Ice Will Disappear

A more realistic projection from the study indicates that, without substantial changes or measures, the ongoing melting trend of the last 20 years would result in the disappearance of nearly half (46%) of the Alps' ice volume by 2050. Extrapolating data from the last ten years alone could elevate this figure to 65%.

2050: The Near Future

Diverging from conventional models that provide estimates for the end of the century, the new study, featured in Geophysical Research Letters, focuses on the shorter term. This approach facilitates a clearer understanding of the relevance within this lifetime, thereby promoting immediate action.

Considerations such as the ages of children in 2050 or the potential presence of snow in 2038, when Switzerland could host the Olympic Games, become pivotal. These estimates gain added significance due to the far-reaching consequences of the disappearance of kilometers of ice on the population, infrastructure, and water reserves.

The data used to build the scenarios stop in 2022, a year that was followed by an exceptionally hot summer. It is therefore likely that the situation will be even worse than the one we present.

Samuel Cook, Study First Author and Researcher, University of Lausanne

Artificial Intelligence Boosts Models

Artificial intelligence algorithms were employed to conduct the simulations. Scientists utilized deep-learning methods to train their model to comprehend physical concepts, incorporating actual climate and glaciological data into the training process.

Machine learning is revolutionizing the integration of complex data into our models. This essential step, previously notoriously complicated and computationally expensive, is now becoming more accurate and efficient.

Guillaume Jouvet, Study Co-Author and Professor, FGSE

The modeling was executed using the IGM model developed within the UNIL ICE group.

Journal Reference:

Cook, S., et al. (2023). Committed Ice Loss in the European Alps Until 2050 Using a Deep‐Learning‐Aided 3D Ice‐Flow Model With Data Assimilation. Geophysical Research Letters. doi.org/10.1029/2023gl105029.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.