Efficient Generative AI Boosts Probabilistic Forecasting of Sudden Stratospheric Warmings

2025-10-30

AI Summary

Subseasonal probabilistic forecasting of stratospheric sudden warming (SSW) events has long been hindered by physical representation biases, initial-condition sensitivity, and prohibitive computational costs in numerical weather prediction (NWP), while data-driven approaches remain inadequate for 3D dynamical modeling and uncertainty quantification. Method: We propose FM-Cast, the first generative AI model for SSW forecasting based on flow matching—integrating meteorological priors with data-driven learning to enable efficient ensemble prediction. Contribution/Results: Running on consumer-grade GPUs, FM-Cast generates 50-member, 30-day probabilistic forecasts in just two minutes. Evaluated on 18 major SSW events, it successfully predicts onset timing, intensity, and morphology 20 days in advance for 10 cases, achieving >50% ensemble accuracy—matching or surpassing state-of-the-art NWP systems. FM-Cast establishes a new paradigm for efficient, interpretable, and physically consistent spatiotemporal forecasting of atmospheric circulation.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Machine Learning: Large Multimodal Models (LMMs)
• Reasoning under Uncertainty: Probabilistic Programming

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Sudden Stratospheric Warmings (SSWs) are key sources of subseasonal predictability and major drivers of extreme winter weather. Yet, their accurate and efficient forecast remains a persistent challenge for numerical weather prediction (NWP) systems due to limitations in physical representation, initialization, and the immense computational demands of ensemble forecasts. While data-driven forecasting is rapidly evolving, its application to the complex, three-dimensional dynamics of SSWs, particularly for probabilistic forecast, remains underexplored. Here, we bridge this gap by developing a Flow Matching-based generative AI model (FM-Cast) for efficient and skillful probabilistic forecasting of the spatiotemporal evolution of stratospheric circulation. Evaluated across 18 major SSW events (1998-2024), FM-Cast skillfully forecasts the onset, intensity, and morphology of 10 events up to 20 days in advance, achieving ensemble accuracies above 50%. Its performance is comparable to or exceeds leading NWP systems while requiring only two minutes for a 50-member, 30-day forecast on a consumer GPU. Furthermore, leveraging FM-Cast as a scientific tool, we demonstrate through idealized experiments that SSW predictability is fundamentally linked to its underlying physical drivers, distinguishing between events forced from the troposphere and those driven by internal stratospheric dynamics. Our work thus establishes a computationally efficient paradigm for probabilistic forecasting stratospheric anomalies and showcases generative AI's potential to deepen the physical understanding of atmosphere-climate dynamics.

Problem

• Forecasting sudden stratospheric warmings accurately and efficiently remains challenging
• Data-driven probabilistic forecasting of complex 3D SSW dynamics is underexplored
• Understanding SSW predictability linked to physical drivers requires better tools

Innovation

• Flow Matching generative AI for stratospheric forecasting
• Achieves 50% ensemble accuracy in 2 minutes
• Links SSW predictability to physical drivers