Huge Ensembles Part I: Design of Ensemble Weather Forecasts using Spherical Fourier Neural Operators

2024-08-06

arXiv.org

AI Summary

Predicting low-probability, high-impact extreme weather events under climate change remains challenging due to the computational expense and limited ensemble size of traditional numerical weather prediction (NWP) models. Method: This work introduces the first application of the Spherical Fourier Neural Operator (SFNO) to ultra-large-ensemble forecasting—supporting 1,000–10,000 members—by integrating perturbed-parameter ensembles (to represent model uncertainty) and bred-vector initial-condition perturbations (to represent initial-state uncertainty). A multi-scale spectral diagnostic framework and extreme-event indices are incorporated to ensure physical consistency and forecast reliability. Contribution/Results: Experiments confirm spectral stability over time—validated via key spectral tests—and calibrated probabilistic forecasts of extremes exhibit discrimination and reliability comparable to ECMWF’s IFS. This establishes a scalable, physics-informed AI paradigm for modeling long-tail climate risks.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Reasoning under Uncertainty: Stochastic Optimization
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Studying low-likelihood high-impact extreme weather events in a warming world is a significant and challenging task for current ensemble forecasting systems. While these systems presently use up to 100 members, larger ensembles could enrich the sampling of internal variability. They may capture the long tails associated with climate hazards better than traditional ensemble sizes. Due to computational constraints, it is infeasible to generate huge ensembles (comprised of 1,000-10,000 members) with traditional, physics-based numerical models. In this two-part paper, we replace traditional numerical simulations with machine learning (ML) to generate hindcasts of huge ensembles. In Part I, we construct an ensemble weather forecasting system based on Spherical Fourier Neural Operators (SFNO), and we discuss important design decisions for constructing such an ensemble. The ensemble represents model uncertainty through perturbed-parameter techniques, and it represents initial condition uncertainty through bred vectors, which sample the fastest growing modes of the forecast. Using the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (IFS) as a baseline, we develop an evaluation pipeline composed of mean, spectral, and extreme diagnostics. Using large-scale, distributed SFNOs with 1.1 billion learned parameters, we achieve calibrated probabilistic forecasts. As the trajectories of the individual members diverge, the ML ensemble mean spectra degrade with lead time, consistent with physical expectations. However, the individual ensemble members' spectra stay constant with lead time. Therefore, these members simulate realistic weather states, and the ML ensemble thus passes a crucial spectral test in the literature. The IFS and ML ensembles have similar Extreme Forecast Indices, and we show that the ML extreme weather forecasts are reliable and discriminating.

Problem

• Enhancing extreme weather event predictions
• Overcoming computational limits with machine learning
• Calibrating probabilistic forecasts using large ensembles

Innovation

• Spherical Fourier Neural Operators
• Perturbed-parameter techniques
• Bred vectors for uncertainty