Abstract
Rapid assessment of hazardous aerosol dispersion is critical for emergency response, yet operational dispersion workflows can exhibit end-to-end latency that is incompatible with the first minutes of decision-making. This study develops and validates a deep learning approach for near-real-time nowcasting of volcanic ash dispersion from geostationary observations. The model was trained on an archive of volcanic ash satellite imagery from EUMETSAT’s SEVIRI instrument (Ash RGB composite) and achieved a structural similarity index of 0.88 for 15-minute next-frame forecasts. The complete edge workflow, including data download and inference, runs in under five seconds on an NVIDIA Jetson AGX Orin. To illustrate how the same nowcasting pipeline can be used for hypothetical scenario exploration across particulate sources, a pixel-based event-injection algorithm is introduced to overlay synthetic plumes of varying sizes into real-time satellite frames before inference. Scenario demonstrations parameterized by nuclear-yield-inspired sizes (10 kt to 100 Mt) are presented at urban (Paris, London, Berlin), national (Iberian Peninsula), and continental (Europe-wide) scales. These scenario outputs are intended as illustrative, low-latency visualizations of kinematic transport patterns in the SEVIRI observation space, as validated predictions of nuclear plume morphology. The primary contribution is a fast, low-cost volcanic ash nowcasting system, complemented by a generalizable injection framework for rapid scenario visualization on edge computing.
Similar content being viewed by others
Data Availability
The code supporting this study is publicly available on Zenodo at: https://doi.org/10.5281/zenodo.18430542
References
Kristiansen, N. I. et al. Improving model simulations of volcanic emission clouds and assessing model uncertainties, in Natural Hazard Uncertainty Assessment: Modeling and Decision Support, ser. Geophysical Monograph. American Geophysical Union223, 105–124 (2017).
Ulfarsson, G. F. & Unger, E. A. Impacts and responses of icelandic aviation to the 2010 eyjafjallajökull volcanic eruption: case study. Transp. Res. Rec. 2214(1), 144–151 (2011).
Simon, S. L., Bouville, A. & Beck, H. L. Geographic distribution of radionuclide deposition across the continental united states. J. Environ. Radioact. 74, 91–105 (2004).
Takemura, T. et al. A numerical simulation of global transport of atmospheric particles emitted from the fukushima daiichi nuclear power plant. Sola 7, 101–104 (2011).
Horwell, C. J. & Baxter, P. J. The respiratory health hazards of volcanic ash: A review for volcanic health hazard research. Bull. Volcanol. 69, 1–24 (2006).
Mettler, F. A. Jr., Gus’kova, A. K. & Gusev, I. Health effects in those with acute radiation sickness from the chernobyl accident. Health Phys. 93(5), 462–469 (2007).
Darack, E. Nuclear sky: The atmosphere and the world’s most powerful weapons. Weatherwise 76(5), 45–57 (2023).
Brumfiel, G. Nuclear weapons physics: Welcome to the atomic weapons establishment, (2010).
Ingram, R. J. Emergency response to radiological releases: have we communicated effectively to the first responder communities to prepare them to safely manage these incidents?. Health Phys. 114(2), 208–213 (2018).
Stunder, B. J., Heffter, J. L. & Draxler, R. R. Airborne volcanic ash forecast area reliability. Weather Forecast. 22(5), 1132–1139 (2007).
Draxler, R. R. & Rolph, G. D. Evaluation of the transfer coefficient matrix (tcm) approach to model the atmospheric radionuclide air concentrations from fukushima. Journal of Geophysical Research: Atmospheres 117, D5 (2012).
Dreher, J. G. Configuring the hysplit model for national weather service forecast office and spaceflight meteorology group applications, NASA Contractor Report NASA/CR-2009-214764, Kennedy Space Center, Florida, Tech. Rep. NASA/CR-2009-214764, April 2009, prepared by ENSCO, Inc. for the Applied Meteorology Unit, Kennedy Space Center. [Online]. Available: https://ntrs.nasa.gov/citations/20090023414.
Zema, T., Kozina, A., Sulich, A., Römer, I. & Schieck, M. Deep learning and forecasting in practice: an alternative costs case. Procedia Comput. Sci. 207, 2958–2967 (2022).
Bowman, K. P. et al. Input data requirements for Lagrangian trajectory models. Bull. Am. Meteor. Soc. 94(7), 1051–1058 (2013).
Alves, D., Mendonça, F., Mostafa, S. S. & Morgado-Dias, F. Deep learning enhanced wind speed and direction forecasting for airport regions. Weather Forecast. 40(1), 207–221 (2025).
Soldatenko, S. A. Artificial intelligence and its application in numerical weather prediction. Russ. Meteorol. Hydrol. 49(4), 283–298 (2024).
Fang, S. et al. Coupled modeling of in- and below-cloud wet deposition for atmospheric \(^{137}\)cs transport following the fukushima daiichi accident using wrf-chem: A self-consistent evaluation of 25 scheme combinations. Environ. Int. 158, 106882 (2022).
Barsotti, S., Neri, A. & Scire, J. S. The vol-calpuff model for atmospheric ash dispersal: 1 approach and physical formulation. J. Geophys. Res. Solid Earth. 113, B03208 (2008).
Costa, A., Macedonio, G. & Folch, A. A three–dimensional eulerian model for transport and deposition of volcanic ashes. Earth Planet. Sci. Lett. 241, 634–647 (2006).
Folch, A. et al. Fall3d–8.0: A computational model for atmospheric transport and deposition of particles, aerosols and radionuclides—part 2: Model validation. Geoscientific Model Development 14, 409–436 (2021).
Folch, A., Mingari, L. & Prata, F. Ensemble–based forecast of volcanic clouds using fall3d–8.1. Front. Earth Sci. 9, 741841 (2022).
Osores, M. S., Ruiz, J. A., Folch, A. & Collini, E. Volcanic ash forecast using ensemble–based data assimilation: An ensemble transform kalman filter coupled with the fall3d–7.2 model (etkf–fall3d version 1.0). Geoscientific Model Development 13, 1–22 (2020).
Hammouti, M., Gencarelli, C. N., Sterlacchini, S., & Biondi, R. Volcanic clouds detection applying machine learning techniques to gnss radio occultations, GPS Solutions, 28, 116, (2024), open access. [Online]. Available: https://doi.org/10.1007/s10291-024-01656-0.
Adachi, K., Kajino, M., Zaizen, Y. & Igarashi, Y. Emission of spherical cesium-bearing particles from an early stage of the fukushima nuclear accident. Sci. Rep. 3, 2554 (2013).
Miura, H. et al. Characterization of two types of cesium-bearing microparticles emitted from the fukushima accident via multiple synchrotron radiation analyses. Sci. Rep. 10(1), 11421 (2020).
Abe, Y. et al. Widespread distribution of radiocesium-bearing microparticles over the greater kanto region resulting from the fukushima nuclear accident. Prog Earth Planet Sci 8, 71 (2021).
Pöllänen, R., Valkama, I. & Toivonen, H. Transport of radioactive particles from the chernobyl accident. Atmos. Environ. 31(21), 3575–3590 (1997).
Rolph, G., Ngan, F. & Draxler, R. Modeling the fallout from stabilized nuclear clouds using the hysplit atmospheric dispersion model. J. Environ. Radioact. 136, 41–55 (2014).
Zhuang, S., Fang, S., Goto, D. & Dong, X. Model behavior regarding in- and below-cloud \(^{137}\)cs wet scavenging following the fukushima accident using 1-km-resolution meteorological field data. Sci. Total Environ. 872, 162165 (2023).
Zhuang, S., Fang, S., Xu, Y., Goto, D. & Dong, X. Wet scavenging of multi-mode \(^{137}\)cs aerosols following the fukushima accident: Size-resolved microphysics modeling with observed diameters. Sci. Total Environ. 917, 170287 (2024).
Rolph, G. D., Ngan, F. & Draxler, R. R. Modeling the fallout from stabilized nuclear clouds using the hysplit atmospheric dispersion model. J. Environ. Radioact. 136, 41–55 (2014).
Sørensen, J. H. et al. Uncertainties in atmospheric dispersion modelling during nuclear accidents. J. Environ. Radioact. 222, 106356 (2020).
Beckett, F. M., Millington, S. C., Witham, C. S., Leadbetter, S. J. & Webster, H. N. Atmospheric dispersion modelling at the london vaac: A review of developments since 2010 eyjafjallajökull volcano ash cloud. Atmosphere 11, 352 (2020).
Folch, A. A review of tephra transport and dispersal models: Evolution, current status, and future perspectives. J. Volcanol. Geoth. Res. 235–236, 96–115 (2012).
Zhang, Y., Bocquet, M., Mallet, V., Seigneur, C. & Baklanov, A. Real-time air quality forecasting, part ii: State of the science, current research needs, and future prospects. Atmos. Environ. 60, 656–676 (2012).
Pasternak, F., Lorsignol, J., & Wolff, L. Spinning enhanced visible and infrared imager (seviri): the new imager for meteosat second generation, in Space Optics,. Earth Observation and Astronomy, 2209. SPIE1994, 86–94 (1994).
Patil, P., Watson, I., Mackie, S., Srivastava, P. K., Islam, T., & Sakhare, S. Assimilating seviri satellite observation into the name-iii dispersion model to improve volcanic ash forecast, Techniques for Disaster Risk Management and Mitigation, 151–170, (2020).
Oura, Y. et al. A database of hourly atmospheric concentrations of radiocesium (\(^{134}\)cs and \(^{137}\)cs) in suspended particulate matter collected in March 2011 at 99 air pollution monitoring stations in eastern japan. J. Nucl. Radiochem. Sci. 15, 15–26 (2015).
Kato, H., Onda, Y., Gao, X., Sanada, Y. & Saito, K. Reconstruction of a fukushima accident-derived radiocesium fallout map for environmental transfer studies. J. Environ. Radioact. 210, 105996 (2019).
Sato, Y. et al. Model intercomparison of atmospheric \(^{137}\)cs from the fukushima daiichi nuclear power plant accident: simulations based on identical input data. Journal of Geophysical Research: Atmospheres 123(20), 11748–11765 (2018).
Sato, Y. et al. Model intercomparison of atmospheric \(^{137}\)cs from the fukushima daiichi nuclear power plant accident: Simulations based on identical input data. Journal of Geophysical Research: Atmospheres 123(20), 11748–11765 (2018).
Sato, Y. et al. A model intercomparison of atmospheric \(^{137}\)cs concentrations from the fukushima daiichi nuclear power plant accident, phase iii: Simulation with an identical source term and meteorological field at 1-km resolution. Atmospheric Environment: X 7, 100086 (2020).
Dong, X., Zhuang, S., Fang, S., Li, H. & Cao, J. Validation and sensitivity study of micro-swift spray against wind tunnel experiments for small-scale air dispersion modeling between mountains and dense buildings at a nuclear power plant site. Prog. Nucl. Energy 142, 104007 (2021).
SEVIRI Ash RGB - Quick Guide, https://user.eumetsat.int/resources/user-guides/ash-rgb-quick-guide, EUMETSAT, 2011, accessed via EUMETSAT Training Library. [Online]. Available: https://user.eumetsat.int/resources/user-guides/ash-rgb-quick-guide.
EUMETSAT, Volcanic Ash Monitoring: Product Guide, EUMETSAT, Eumetsat-Allee 1, D-64295 Darmstadt, Germany, June 2015, doc. No. EUM/TSS/MAN/15/802120, Issue v1A. [Online]. Available: http://www.eumetsat.int.
Alves, D., Mendonça, F., Mostafa, S. S. & Morgado-Dias, F. A computer vision approach for satellite-driven wind nowcasting over complex terrains. Environ. Res. Commun. 6(5), 055014 (2024).
Desai, P., Sujatha, C., Chakraborty, S., Ansuman, S., Bhandari, S., Kardiguddi, S. Next frame prediction using convlstm, in Journal of Physics: Conference Series, 2161, 1. IOP Publishing, 2022, 012024, 1st International Conference on Artificial Intelligence, Computational Electronics and Communication System (AICECS 2021), 28–30 October (2021), Manipal, India.
Naz, F., She, L., Sinan, M. & Shao, J. Enhancing radar echo extrapolation by convlstm2d for precipitation nowcasting. Sensors 24(2), 459 (2024).
Glasstone, S., Dolan, P. J. The Effects of Nuclear Weapons, 3rd ed. U.S. Department of Defense / Energy Research and Development Administration, (1977). [Online]. Available: https://doi.org/10.2172/6852629.
Goffman, T. E. Nuclear terrorism and the problem of burns. Am. J. Emerg. Med. 29(3), 224–228 (2011).
Smith, J. & Smith, T. Nuclear war: The medical facts. Br. Med. J. 283(6294), 771–776 (1981).
Morton, B. R., Taylor, G. & Turner, J. S. Turbulent gravitational convection from maintained and instantaneous sources. Proceedings of the Royal Society A 234(1196), 1–23 (1956).
Soille, P. Morphological Image Analysis: Principles and Applications 2nd edn. (Springer-Verlag, Berlin, Heidelberg, 2003).
Gueffier, J. et al. Weather regimes and the related atmospheric composition at a pyrenean observatory characterized by hierarchical clustering of a 5-year data set. Atmos. Chem. Phys. 24, 287–316 (2024).
Funding
This work was funded by FCT (Fundação para a Ciência e a Tecnologia) projects: 10.54499/LA/P/0083/2020 & UID/50009/2025, and in part by Agência Regional para o Desenvolvimento da Investigação, Tecnologia e Inovação (ARDITI).
Author information
Authors and Affiliations
Contributions
D.A. designed the study, developed the deep learning model, conducted simulations, performed data analysis, and wrote the main manuscript text. F.M. contributed to the evaluation metrics, assisted in model validation, and helped write the manuscript. M.R. contributed to the simulation of nuclear detonation scenarios and supported manuscript writing. L.P. and F.M.D. reviewed and provided critical feedback on the manuscript. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Alves, D., Radeta, M., Mendonça, F. et al. A low-latency deep learning framework for volcanic ash cloud nowcasting using geostationary satellite imagery. Sci Rep (2026). https://doi.org/10.1038/s41598-026-42230-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-026-42230-7
