CloudViT: exploring cloud type classification  with vision transformers in global satellite data

Lenhardt, Julien; Quaas, Johannes; Sejdinovic, Dino; Klocke, Daniel

doi:10.5194/acp-26-5447-2026

Articles | Volume 26, issue 8

https://doi.org/10.5194/acp-26-5447-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-26-5447-2026

© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 26, issue 8

Research article

|

22 Apr 2026

Research article |

| 22 Apr 2026

CloudViT: exploring cloud type classification with vision transformers in global satellite data

Julien Lenhardt, Johannes Quaas, Dino Sejdinovic, and Daniel Klocke

Data sets

CloudViT - Method code and data for the article "CloudViT: exploring cloud type classification with vision transformers in global satellite data'' Julien Lenhardt et al. https://doi.org/10.5281/zenodo.12731287

MIDAS: Global Marine Meteorological Observations Data Met Office https://catalogue.ceda.ac.uk/uuid/77910bcec71c820d4c92f40d3ed3f249

LAND SYNOP reports from land stations collected by the Met Office MetDB System Met Office https://catalogue.ceda.ac.uk/uuid/9f80d42106ba708f92ada730ba321831

Short summary

Clouds come in various shapes and sizes and constitute a fundamental element of the Earth's climate system. Different cloud types show variable impacts on climate change. We present a new cloud type classification method called CloudViT (Cloud Vision Transformer) relying on spatial patterns of cloud properties obtained from satellite data using machine learning. We can thus help understanding the effects of different cloud types on climate change.