Climate Prompting: Generating the Madden-Julian Oscillation using Video Diffusion and Low-Dimensional Conditioning

TL;DR

This paper introduces a video diffusion model trained on atmospheric data to generate realistic Madden-Julian Oscillation sequences conditioned on key low-dimensional metrics, aiding understanding and prediction of tropical atmospheric phenomena.

Contribution

The study presents a novel video diffusion approach conditioned on low-dimensional metrics to synthesize and analyze MJO sequences, bridging theoretical models and atmospheric complexity.

Findings

Generated MJOs capture key features like spectra and multiscale structures.

Model can produce idealized MJOs based on specific low-dimensional conditions.

The approach aids in understanding physical drivers of the MJO.

Abstract

Generative Deep Learning is a powerful tool for modeling of the Madden-Julian oscillation (MJO) in the tropics, yet its relationship to traditional theoretical frameworks remains poorly understood. Here we propose a video diffusion model, trained on atmospheric reanalysis, to synthetize long MJO sequences conditioned on key low-dimensional metrics. The generated MJOs capture key features including composites, power spectra and multiscale structures including convectively coupled waves, despite some bias. We then prompt the model to generate more tractable MJOs based on intentionally idealized low-dimensional conditionings, for example a perpetual MJO, an isolated modulation by seasons and/or the El Nino-Southern Oscillation, and so on. This enables deconstructing the underlying processes and identifying physical drivers. The present approach provides a practical framework for bridging the gap between low-dimensional MJO theory and high-resolution atmospheric complexity and will help tropical atmosphere prediction.