Artificial intelligence pathways from weather to climate
Abstract
Deep learning has made rapid advances in weather forecasting: autoregressive models trained on atmospheric reanalyses now rival dynamical models across nowcasting, medium-range, and subseasonal-to-seasonal lead times, producing well-calibrated ensemble forecasts at reduced cost. We review these advances and consider their extension to climate horizons, where the challenge shifts from initial-condition skill to producing reliable statistical responses under altered forcings. AI-powered climate prediction systems must produce credible forced responses to drivers (e.g., greenhouse gases, land-use change) typically outside the observed record. We propose two minimum requirements for AI in climate modeling: (i) external forcing agents must enter explicitly enough to support interventions in which they vary independently; and (ii) robustness must be stress-tested in out-of-distribution regimes, including extremes and counterfactual trajectories. Using leading AI autoregressive emulators and hybrid physics-AI models, we identify development and coupling challenges. Comparing the reported throughput of these models with that of GPU-ported dynamical models highlights how AI can reduce time-to-solution by advancing only the target variables at the required resolution and using longer time steps, rather than integrating a full high-frequency, multivariate state. Diverse AI downscaling strategies can partially substitute for explicit fine-scale resolution, paving the way toward inexpensive local hazard assessment across prediction horizons.
Related
- WP-MIP: An Artificial Intelligence, Hybrid, and Physically Based Model Intercomparison Project for Weather Prediction
- An Artificial Intelligence Integrated Geospatial Decision-Support Framework for Scenario-Based Climate-Adaptive Land-Use Optimization and Environmental Risk Reduction
- Skillful Seasonal Tropical Cyclone Prediction with AI Weather--Climate Models Forced by Dynamically Predicted SSTs
- Weather forecasting in a changing climate: the rise of AI and Machine learning?
- A neural network-based Universal Thermal Climate Index for reliable global thermal-stress classification across extreme weather
BibTeX
@misc{beucler2026artificial,
title = {{Artificial intelligence pathways from weather to climate}},
author = {Tom Beucler and J. David Neelin and Hui Su and Shivanshi Asthana and Chris Bretherton and Will Chapman and Costa Christopoulos and Spencer K. Clark and Aditya Grover and Ignacio Lopez-Gomez and Tapio Schneider and Adam Subel and Oliver Watt-Meyer},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.09770},
eprint = {2610.09770},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.09770},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Artificial intelligence pathways from weather to climate
AU - Tom Beucler
AU - J. David Neelin
AU - Hui Su
AU - Shivanshi Asthana
AU - Chris Bretherton
AU - Will Chapman
AU - Costa Christopoulos
AU - Spencer K. Clark
AU - Aditya Grover
AU - Ignacio Lopez-Gomez
AU - Tapio Schneider
AU - Adam Subel
AU - Oliver Watt-Meyer
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.09770
UR - https://arxiv.org/abs/2610.09770
AB - Deep learning has made rapid advances in weather forecasting: autoregressive models trained on atmospheric reanalyses now rival dynamical models across nowcasting, medium-range, and subseasonal-to-seasonal lead times, producing well-calibrated ensemble forecasts at reduced cost. We review these advances and consider their extension to climate horizons, where the challenge shifts from initial-condition skill to producing reliable statistical responses under altered forcings. AI-powered climate prediction systems must produce credible forced responses to drivers (e.g., greenhouse gases, land-use change) typically outside the observed record. We propose two minimum requirements for AI in climate modeling: (i) external forcing agents must enter explicitly enough to support interventions in which they vary independently; and (ii) robustness must be stress-tested in out-of-distribution regimes, including extremes and counterfactual trajectories. Using leading AI autoregressive emulators and hybrid physics-AI models, we identify development and coupling challenges. Comparing the reported throughput of these models with that of GPU-ported dynamical models highlights how AI can reduce time-to-solution by advancing only the target variables at the required resolution and using longer time steps, rather than integrating a full high-frequency, multivariate state. Diverse AI downscaling strategies can partially substitute for explicit fine-scale resolution, paving the way toward inexpensive local hazard assessment across prediction horizons.
ER -