In late October, a devastating storm struck northern Italy, causing Lake Como and the Seveso River to overflow this event served as a stark reminder of the increasing vulnerability of areas worldwide to extreme weather events driven by climate change. As global warming triggers more frequent and destructive storms, floods, heatwaves, and droughts, experts are turning to artificial intelligence (AI) as a potential solution to enhance forecasting and mitigation efforts.
AI’s role in climate forecasting
While AI has gained recognition for its text generation, personalized services, and artistic capabilities, it holds great promise in addressing the urgent challenge of climate change. Climate scientists and researchers are exploring the use of AI to improve climate and weather models, particularly in predicting extreme weather events.
Currently, climate models rely on large datasets and mathematical formulas to make predictions. However, these models have limitations and are not always accurate, especially for extreme weather events. AI offers the potential to address these weaknesses by leveraging its data processing capabilities and adaptability.
CLINT: An AI-enhanced climate forecasting project
One notable initiative at the forefront of AI-driven climate forecasting is the CLINT project, led by Professor Andrea Castelletti at the Polytechnic University of Milan. CLINT, funded by the European Union and set to run until June 2025, combines AI with Europe’s Copernicus satellite network to enhance climate forecasting. Researchers from various European countries are collaborating to determine how AI can enhance our understanding of extreme weather events.
Existing climate models often struggle to accurately predict certain extreme weather phenomena, such as the rapid increase in heatwaves in Europe. AI may hold the key to uncovering the underlying causes of these events and improving predictive accuracy. By doing so, it can provide more reliable and timely warnings, benefitting regions around the world facing disruptive weather patterns.
Focusing on causes of extreme weather
In parallel with the CLINT project, Professor Dim Coumou is leading the XAIDA initiative, funded by the EU and set to run until August 2025. XAIDA focuses not only on improving predictions but also on understanding the underlying causes of extreme weather events. Coumou’s research centers on the impact of global warming on the frequency and intensity of such events.
Challenges and opportunities for AI in climate science
Despite its potential, integrating AI into climate modeling presents challenges. AI systems operate by processing vast amounts of data through intricate networks of parameters, and interpreting the results can be complex. Researchers are working to identify the most influential parameters related to weather information to improve model accuracy.
Another challenge is the data scarcity for extreme climate events, which are, by nature, rare occurrences. To address this, the CLINT project is using data augmentation techniques, where AI systems generate synthetic data based on historical information. This augmented data can then be used to enhance predictions.
The path forward: AI’s growing role in climate models
Experts in the field believe that AI’s integration into climate models is on the horizon. The European Centre for Medium-Range Weather Forecasts is already experimenting with machine-learning models, and this trend is expected to grow exponentially in the coming years.
In a world where extreme weather events are increasingly frequent and severe, the potential for AI to enhance climate forecasting and mitigation efforts is a promising development. The United Nations climate change summit, COP28 focuses on addressing global warming and improving early