Hierarchy of extreme-event predictability in turbulence revealed by machine learning
Not on the map. Why?
Abstract (arXiv)
Extreme-event predictability in turbulence is strongly state dependent, yet event-by-event predictability horizons are difficult to quantify without access to governing equations or costly perturbation ensembles. Here we train an autoregressive conditional diffusion model on direct numerical simulations of the two-dimensional Kolmogorov flow and use a CRPS-based skill score to define an event-wise predictability horizon. Enstrophy extremes exhibit a pronounced hierarchy: forecast skill persists from ≈ 1 to > 4 Lyapunov times across events. Spectral filtering shows that these horizons are controlled predominantly by large-scale structures. Extremes are preceded by intense strain cores organizing quadrupolar vortex packets, whose lifetime sharply separates long- from short-horizon events. These results identify coherent-structure persistence as a governing mechanism for the predictability of turbulence extremes and provide a data-driven route to diagnose predictability limits from observations.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 8 of its 45 references that are in the map. First posted on arXiv 14 Mar 2026.
Builds on
8 of 45 references.
References · 8
- 2020Denoising Diffusion Probabilistic Modelscited by 21,076
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
- 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
- 2023Benchmarking Autoregressive Conditional Diffusion Models for Turbulent Flow Simulationcited by 71
- 2023Stochastic Latent Transformer: Efficient Modelling of Stochastically Forced Zonal Jetscited by 8
- 2025Dynamical errors in machine learning forecastscited by 6
- 2024Spatio-temporal Dynamical Indices for Complex Systemscited by 3
- 2025XAI4Extremes: An interpretable machine learning framework for understanding extreme-weather precursors under climate changecited by 3
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025Guided Unconditional and Conditional Generative Models for Super-Resolution and Inference of Quasi-Geostrophic Turbulence
- 2025Learning Temporally Consistent Turbulence Between Sparse Snapshots via Diffusion Models
- 2025Lazy Diffusion: Mitigating spectral collapse in generative diffusion-based stable autoregressive emulation of turbulent flows
- 2023Turbulence Scaling from Deep Learning Diffusion Generative Models
- 2026Event-Conditioned Diagnostics of Kinematic, Contact, and Object-Permanence Structure in Passive Object-State World Models
- 2023Bayesian Conditional Diffusion Models for Versatile Spatiotemporal Turbulence Generation
- 2025Autoregressive regularized score-based diffusion models for multi-scenarios fluid flow prediction
- 2026OceanCBM: A Concept Bottleneck Model for Mechanistic Interpretability in Ocean Forecasting