AI Emulation of Stochastic Sudden Stratospheric Warming with Interpretable Latent Structure
Abstract
Rare weather regime transitions pose a challenge for data-driven modeling due to class imbalance. In this study, we develop a probabilistic deep learning emulator for a prototypical system with regime transitions, the stochastic Holton--Mass model of stratospheric variability, and analyze the structure of its learned latent space. The Holton--Mass model exhibits two metastable regimes, a strong and a weak polar vortex, maintained by nonlinear wave--mean flow interactions, with weak stochastic forcing intermittently triggering rare transitions between these regimes that qualitatively represent SSW events. We employ a ResNet-inspired Conditional Variational Autoencoder with six-layer encoder and decoder layers and explicit current-state conditioning to model the distribution of the system's state at the next time step (one day). The emulator accurately reproduces short-term dynamics, steady-state probability distributions, regime persistence statistics, rare transition rates, the transition committor function, and the transition expected lead time of the physical model. Beyond emulation fidelity, we interrogate the learned latent representation to understand how the model internalizes the underlying metastable structure of the dynamics. Principal Component Analysis of the 32-dimensional latent space reveals a clear and unsupervised separation into four physically interpretable clusters corresponding to strong versus weak vortex regimes and stable versus transition-prone configurations. Such emergent regime separation in latent space is hard to identify for deep generative models applied to high-dimensional stochastic systems. Our results show that carefully designed probabilistic emulators can uncover physically meaningful manifolds governing extreme-event dynamics, potentially aiding the development of improved operational advanced warning systems.
Versions
Related
- Tropopause structure and biases in climate models related to stratospheric water vapour
- Diurnally asymmetric warming and hydroclimatic stress structure alpine ecosystem vulnerability in the Three-River headwaters region
- Resilient Stratospheric Aerosol Injection is Cooperative
- Uptake of Stratospheric Species on Minerals Proposed for Stratospheric Aerosol Injection
- Flood exposure and structure damage under deterministic design-storm and stochastic storm transposition rainfall scenarios: a case study in South Louisiana, USA
BibTeX
@misc{boscu2026emulation,
title = {{AI Emulation of Stochastic Sudden Stratospheric Warming with Interpretable Latent Structure}},
author = {C. Daniel Boscu and Daniel Hernandez and Fabio Alvarez Ventura and Justin Finkel and Ashesh Chattopadhyay and Pedram Hassanzadeh and Dorian S. Abbot},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.02069},
eprint = {2610.02069},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.02069},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - AI Emulation of Stochastic Sudden Stratospheric Warming with Interpretable Latent Structure
AU - C. Daniel Boscu
AU - Daniel Hernandez
AU - Fabio Alvarez Ventura
AU - Justin Finkel
AU - Ashesh Chattopadhyay
AU - Pedram Hassanzadeh
AU - Dorian S. Abbot
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.02069
UR - https://arxiv.org/abs/2610.02069
AB - Rare weather regime transitions pose a challenge for data-driven modeling due to class imbalance. In this study, we develop a probabilistic deep learning emulator for a prototypical system with regime transitions, the stochastic Holton--Mass model of stratospheric variability, and analyze the structure of its learned latent space. The Holton--Mass model exhibits two metastable regimes, a strong and a weak polar vortex, maintained by nonlinear wave--mean flow interactions, with weak stochastic forcing intermittently triggering rare transitions between these regimes that qualitatively represent SSW events. We employ a ResNet-inspired Conditional Variational Autoencoder with six-layer encoder and decoder layers and explicit current-state conditioning to model the distribution of the system's state at the next time step (one day). The emulator accurately reproduces short-term dynamics, steady-state probability distributions, regime persistence statistics, rare transition rates, the transition committor function, and the transition expected lead time of the physical model. Beyond emulation fidelity, we interrogate the learned latent representation to understand how the model internalizes the underlying metastable structure of the dynamics. Principal Component Analysis of the 32-dimensional latent space reveals a clear and unsupervised separation into four physically interpretable clusters corresponding to strong versus weak vortex regimes and stable versus transition-prone configurations. Such emergent regime separation in latent space is hard to identify for deep generative models applied to high-dimensional stochastic systems. Our results show that carefully designed probabilistic emulators can uncover physically meaningful manifolds governing extreme-event dynamics, potentially aiding the development of improved operational advanced warning systems.
ER -