Demystifying Data-Driven Probabilistic Medium-Range Weather Forecasting
This simplifies medium-range weather forecasting by eliminating the need for tailored training recipes, benefiting meteorologists and climate researchers.
The paper tackles the complexity in data-driven weather forecasting by showing that state-of-the-art probabilistic skill can be achieved without intricate architectures or specialized training, achieving statistically significant improvements on most variables compared to existing systems.
The recent revolution in data-driven methods for weather forecasting has lead to a fragmented landscape of complex, bespoke architectures and training strategies, obscuring the fundamental drivers of forecast accuracy. Here, we demonstrate that state-of-the-art probabilistic skill requires neither intricate architectural constraints nor specialized training heuristics. We introduce a scalable framework for learning multi-scale atmospheric dynamics by combining a directly downsampled latent space with a history-conditioned local projector that resolves high-resolution physics. We find that our framework design is robust to the choice of probabilistic estimator, seamlessly supporting stochastic interpolants, diffusion models, and CRPS-based ensemble training. Validated against the Integrated Forecasting System and the deep learning probabilistic model GenCast, our framework achieves statistically significant improvements on most of the variables. These results suggest scaling a general-purpose model is sufficient for state-of-the-art medium-range prediction, eliminating the need for tailored training recipes and proving effective across the full spectrum of probabilistic frameworks.