Publication

AI and mitigation scenarios. Exploring climate futures with deep learning

Glancing forward to view alternative futures for limiting global warming requires understanding complex societal–environmental systems that drive future emissions. Now a study explores the potential, and limits, of deep learning to generate core characteristics of these futures. Al Khourdajie, A.

Al Khourdajie, A. Exploring climate futures with deep learning. Nature Climate Change (2025).
https://doi.org/10.1038/s41558-025-02350-w
https://rdcu.be/eqPpk

More publications

Reducing material requirements while decarbonising the Spanish economy: from a green growth to a postgrowth paradigm

30 July 2026

The transition to a low-carbon economy is expected to increase material requirements, as low-carbon technologies (LCTs) typically require more materials than fossil fuel-based energy systems. Here, we extend the MEDEAS-Spain Integrated Assessment Model (IAM) to improve the representation of material requirements, stocks, and flows. Key novelties are: (i) the use

Climate ambition explained: key indicators and net-zero target projections using machine learning

29 July 2026

This study uses advanced machine learning algorithms to analyze the indicators influencing national climate ambition in 2030 emissions targets and net-zero pledges. By focusing on technical and social feasibility, and political credibility, we assess climate ambition of countries through cumulative per capita emissions from 2021 to the net-zero target year.

A harmonised dataset for Earth system foundation models

28 July 2026

Foundation models for Earth systems have so far been trained primarily on physical climate and weather data, with limited representation of the human systems that both drive and respond to environmental change. The lack of a unified global training resource that combines climate, land, ocean, cryosphere, infrastructure, hazards, and socioeconomic