MIT machine learning doubles accuracy of 2-6 week forecasts

MIT researchers published a Nature paper showing a machine-learning method that corrects errors in existing forecast models, doubling accuracy for 2-6 week subseasonal forecasts — a window critical for emergency planning.

2026-10-14 ~ 2026-10-15 · 2 related posts

• MIT Doubles Accuracy of 2-6 Week Weather Forecasts With ML Postprocessing of ECMWF Models — nordicinst · 2026-10-14
• MIT's Nature paper breaks records for 2-6 week weather forecasts with ML error correction — LesterMackey · 2026-10-15