Sebastian Schemm

Deterministic weather models can match dedicated probabilistic ensembles within 0.04–0.13 CRPSS at a 10-day lead time simply by perturbing raw network weights at inference without any retraining.

Reinforcement learning can cut weather forecast errors by up to 45% while maintaining numerical stability in operational models.

Significant architectural differences in machine learning models only emerge during active sudden stratospheric warming events, revealing the crucial role of vertical coupling in accurate predictions.

AI weather models don't just parrot data; stable ones can generate unique, plausible year-long weather trajectories, but only if they properly handle high-frequency noise.