Technical note: Integrating generative AI with physics-based simulations for probabilistic modelling of volcanic clouds
- 1Geociencias Barcelona (Barcelona, Spain)
Abstract
Ensemble-based modelling of the atmospheric dispersal of volcanic clouds enables more realistic forecasts by explicitly accounting for uncertainties in eruption source parameters, meteorological data, and systematic errors in transport models. Many ensemble applications, including forecast uncertainty quantification, data assimilation, and probabilistic hazard assessment, require large ensembles to reduce sampling errors and adequately represent probability distributions. However, simulating and processing large ensembles with atmospheric dispersal models can be computationally demanding, even on high-performance computing clusters. Generative AI offers a complementary approach by representing simulated fields in a lower-dimensional latent space, enabling new samples to be generated at low computational cost and facilitating the manipulation of large ensembles. A Variational AutoEncoder (VAE) is trained on ensembles of physics-based simulations and used to generate new samples. The learned latent-space representation is then used to test two ensemble-based data assimilation methods, demonstrating the potential benefits of performing data assimilation in a lower-dimensional space. As a proof of concept, the data assimilation problem is formulated in two dimensions. Two numerical experiments demonstrate that VAE-generated ensembles can reproduce key statistics of a larger reference ensemble and improve agreement with satellite retrievals. The VAE achieves up to a 1000-fold compression of the original fields into 16- and 32-dimensional latent spaces. The VAE-generated ensemble of 4096 members reduces RMSE by 4 % relative to the model prior, while latent-space data assimilation further reduces RMSE by 30 %. These results highlight the potential of generative AI to efficiently augment physics-based ensembles and improve forecasting skill through latent-space data assimilation.
Versions
Related
- Modelling Energy Transfer Efficiency in Collision Induced Dissociation and Unfolding Using Molecular Dynamics Simulations
- Integrating Generative AI and Data Science into General Chemistry Laboratories
- SurfGen: A generative framework for property-guided surfactant design by integrating inverse design and molecule optimization based models
- Integrating Predictive and Generative Capabilities by Latent Space Design via the DKL-VAE Model
- From Bench to Game Engine: An Instructor-Built Suite of 3D Laboratory Simulations Created with Generative AI
BibTeX
@article{mingari2026technical,
title = {{Technical note: Integrating generative AI with physics-based simulations for probabilistic modelling of volcanic clouds}},
author = {Leonardo Mingari and Arnau Folch and Heribert Pascual and Manuel Titos},
journal = {Atmospheric Chemistry and Physics},
year = {2026},
month = nov,
doi = {10.5194/acp-26-15953-2026},
url = {https://doi.org/10.5194/acp-26-15953-2026},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Technical note: Integrating generative AI with physics-based simulations for probabilistic modelling of volcanic clouds
AU - Leonardo Mingari
AU - Arnau Folch
AU - Heribert Pascual
AU - Manuel Titos
PY - 2026
DA - 2026/11/11
JO - Atmospheric Chemistry and Physics
DO - 10.5194/acp-26-15953-2026
UR - https://doi.org/10.5194/acp-26-15953-2026
AB - Ensemble-based modelling of the atmospheric dispersal of volcanic clouds enables more realistic forecasts by explicitly accounting for uncertainties in eruption source parameters, meteorological data, and systematic errors in transport models. Many ensemble applications, including forecast uncertainty quantification, data assimilation, and probabilistic hazard assessment, require large ensembles to reduce sampling errors and adequately represent probability distributions. However, simulating and processing large ensembles with atmospheric dispersal models can be computationally demanding, even on high-performance computing clusters. Generative AI offers a complementary approach by representing simulated fields in a lower-dimensional latent space, enabling new samples to be generated at low computational cost and facilitating the manipulation of large ensembles. A Variational AutoEncoder (VAE) is trained on ensembles of physics-based simulations and used to generate new samples. The learned latent-space representation is then used to test two ensemble-based data assimilation methods, demonstrating the potential benefits of performing data assimilation in a lower-dimensional space. As a proof of concept, the data assimilation problem is formulated in two dimensions. Two numerical experiments demonstrate that VAE-generated ensembles can reproduce key statistics of a larger reference ensemble and improve agreement with satellite retrievals. The VAE achieves up to a 1000-fold compression of the original fields into 16- and 32-dimensional latent spaces. The VAE-generated ensemble of 4096 members reduces RMSE by 4 % relative to the model prior, while latent-space data assimilation further reduces RMSE by 30 %. These results highlight the potential of generative AI to efficiently augment physics-based ensembles and improve forecasting skill through latent-space data assimilation.
SN - 1680-7324
SN - 1680-7316
ER -