WIND: Weather Inverse Diffusion for Zero-Shot Atmospheric Modeling

TL;DR

WIND is a versatile, pre-trained foundation model for atmospheric modeling that unifies multiple weather tasks without task-specific fine-tuning, using self-supervised video diffusion and inverse problem solving.

Contribution

Introducing WIND, a single pre-trained atmospheric foundation model that handles diverse weather tasks via inverse problems without fine-tuning, leveraging self-supervised video diffusion.

Findings

WIND can perform probabilistic weather forecasting and spatial downscaling.

It reconstructs atmospheric fields from sparse data effectively.

WIND explores extreme weather scenarios under out-of-distribution conditions.

Abstract

Deep learning has revolutionized weather forecasting, but many challenges remain, including climate modeling. Moreover, the current landscape remains fragmented: highly specialized models are typically trained individually for distinct tasks. To unify this landscape, we introduce WIND, a single pre-trained foundation model capable of replacing specialized baselines across a vast array of tasks. Crucially, in contrast to previous atmospheric foundation models, we achieve this without any task-specific fine-tuning. To learn a robust, task-agnostic prior of the atmosphere, we pre-train WIND with a self-supervised video reconstruction objective, utilizing an unconditional video diffusion model to iteratively reconstruct atmospheric dynamics from a noisy state. At inference, we frame diverse domain-specific problems strictly as inverse problems and solve them via posterior sampling. This unified approach allows us to tackle highly relevant weather and climate problems, including probabilistic forecasting, spatial and temporal downscaling, reconstruction of spatial fields from sparse observations and enforcing global dry air mass conservation. We further demonstrate how WIND can be applied to explore extreme weather events under prescribed out-of-distribution thermodynamic perturbations. By combining generative video modeling with inverse problem solving, WIND offers a computationally efficient alternative for AI-based atmospheric modeling.