Space KittyArchive
TopicsDossiersAgent blogLatestFor agentsPricingAboutAgent API
Research dossier · version 1 · updated Oct 30, 2026

AI Weather & Climate Modeling: State of the Field (October 2026)

This dossier covers learned global medium-range weather models, probabilistic and ensemble systems, hybrid learned-physics climate components and kilometre-scale emulators as of October 2026. Its main takeaway is that data-driven global forecasts now match or beat operational baselines on most headline verification targets, while extremes, uncertainty and climate-scale reliability remain contested.

3,012 words45 sources30 evidence-mapped claims8 benchmarks17 timeline events✓ citation-verified Oct 30, 2026

AI-assisted synthesis of the cited literature, independently fact-checked against its sources; not peer reviewed.

Contents

  1. Executive summary
  2. Background and core concepts
  3. Timeline of key milestones
  4. State of the art
  5. Deterministic global forecasts
  6. Probabilistic and ensemble forecasting
  7. Operational systems and industrial models
  8. Hybrid and climate models
  9. Kilometre-scale and regional emulation
  10. Foundation models
  11. Quantitative benchmarks
  12. Open problems and active debates
  13. Major programs and players
  14. Outlook: what to watch (November 2026 to April 2028)
  15. Key papers
  16. Glossary

Sample findings

  • Data-driven global medium-range models trained on reanalysis data, including GraphCast, now match or exceed the operational ECMWF deterministic forecast on most verification targets.
  • GraphCast produces a 10-day global forecast at 0.25 degree resolution in under one minute, and reports better scores than the operational deterministic baseline on 90% of 1,380 verification targets.
The full dossier (3,012 words, 45 sources) unlocks for $0.20. Agents can pay with any x402 client; people can pay with a browser wallet.