Uncertainty-Aware Spatiotemporal Super-Resolution Data Assimilation with Diffusion Models

TL;DR

This paper introduces DiffSRDA, a diffusion model-based probabilistic data assimilation framework that enhances spatiotemporal super-resolution in chaotic systems, providing accurate, uncertainty-aware analyses with reduced computational costs.

Contribution

Develops DiffSRDA, a diffusion model-based probabilistic data assimilation method that achieves high-quality super-resolution and uncertainty quantification efficiently, without extensive retraining.

Findings

DiffSRDA achieves reconstruction quality close to high-resolution ensemble Kalman filter.

The ensemble from DiffSRDA captures meaningful uncertainty patterns in active regions.

Few reverse diffusion steps retain most of the full-chain accuracy, enabling practical cycling.

Abstract

Data assimilation (DA) improves prediction of chaotic systems by combining model forecasts with sparse, noisy observations. Many DA methods are inherently probabilistic, but accurate probabilistic DA is often computationally expensive because it requires repeated high-resolution (HR) forecasts and large ensembles. In this study, we develop DiffSRDA, a probabilistic spatiotemporal super-resolution data assimilation framework based on denoising diffusion models, and evaluate it on an idealized barotropic ocean jet instability testbed. DiffSRDA is trained offline to generate short HR analysis windows conditioned on (i) a time series of low-resolution (LR) forecast frames and (ii) sparse HR observations. Repeated reverse diffusion sampling then produces an ensemble of HR analyses, providing both point estimates and uncertainty information. Despite relying only on low-cost LR forecasts, DiffSRDA achieves reconstruction quality close to that of an Ensemble Kalman Filter (EnKF) driven by HR forecasts, while improving over deterministic CNN-based SRDA baselines. The sampled ensemble also yields physically meaningful uncertainty patterns, with spread concentrated in dynamically active regions similarly to EnKF. A key practical result is that accurate base DiffSRDA cycling does not require long reverse chains: most of the full-chain accuracy is retained with only a few reverse steps, making diffusion-based SRDA practical for repeated cycling. Finally, by exploiting the score-based structure of diffusion sampling, we demonstrate training-free observation-consistency guidance for deployment-time sensor-layout shifts, enabling improved use of changed observation configurations without retraining. Overall, diffusion models provide a practical, uncertainty-aware, and computationally efficient approach for spatiotemporal SRDA in chaotic fluid flows.