Beyond forecasting: Data-driven climate science
More than technological innovation.
More than the growing volume of data enriching Earth Observation Systems.
It is about how technology and data availability have brought unprecedented possibilities to climate science.
A methodological revolution that opens new possibilities in climate science, offering methods for a deeper and more robust understanding of the climate system and how to shape resilient and sustainable societies.
The CMCC Artificial Intelligence and Machine Learning program consists of an array of cross-institute activities that harness data-driven approaches and machine learning models to predict climate patterns, extreme weather events, and the effects of climate change on land use, agriculture, and society.
Data-driven methods have immense potential in climate science, utilizing vast datasets from numerical models, satellites, and ground-based sensors. These approaches improve model efficiency, enhance extreme event forecasting, and support early-warning systems. Machine learning also helps extract valuable insights to better represent physical processes and assess climate risks, impacts, and adaptation strategies.
Establishing CMCC as a global leader in machine learning and artificial intelligence for climate science
CMCC is pioneering initiatives centered on advanced downscaling techniques with regional applications and models addressing climate impacts across diverse contexts, spanning from soil and agriculture to coastal erosion. The program also uses machine learning to analyze satellite data for land use classification, aiding decision-making in environmental management.
Ongoing initiatives such as fire monitoring highlight CMCC’s commitment to integrating machine learning across diverse areas of climate research. Internationally, CMCC stands out by integrating human behavior and economic dynamics into its climate models, positioning itself as a leader in climate research, adaptation and resilience efforts.
Looking ahead, the program has ambitious goals to:
- Create an Earth System Model entirely driven by data.
- Develop Machine-Learning models that forecast the economic impacts of climate change.
- Expand efforts to predict effects on vegetation, land use, urban systems, and society.
These efforts aim to establish CMCC as a global leader in the application of machine learning for climate change research.
Know more about the CMCC strategic programs
CMCC builds its research agenda around a set of strategic programs that respond to frontier issues.
These are crucial to understanding the challenges facing socio-economic systems in an environmental and social context characterized by a changing climate.