Representing subgrid-scale cloud effects in a radiation parameterization using machine learning: MLe-radiation v1.0

Hafner, Katharina; Shamekh, Sara; Bertoli, Guillaume; Lauer, Axel; Pincus, Robert; Savre, Julien; Eyring, Veronika

doi:10.5194/gmd-19-3875-2026

Articles | Volume 19, issue 9

https://doi.org/10.5194/gmd-19-3875-2026

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

https://doi.org/10.5194/gmd-19-3875-2026

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

Articles | Volume 19, issue 9

Development and technical paper

|

13 May 2026

Development and technical paper |

| 13 May 2026

Representing subgrid-scale cloud effects in a radiation parameterization using machine learning: MLe-radiation v1.0

Katharina Hafner, Sara Shamekh, Guillaume Bertoli, Axel Lauer, Robert Pincus, Julien Savre, and Veronika Eyring

Data sets

Code and Data for "Representing Subgrid-Scale Cloud Effects in a Radiation Parameterization using Machine Learning: MLe-radiation v1.0 " Katharina Hafner https://doi.org/10.5281/zenodo.18853569

Model code and software

Representing Subgrid-Scale Cloud Effects in a Radiation Parameterization using Machine Learning: MLe-radiation v1.0 Katharina Hafner https://doi.org/10.5281/zenodo.17280639

ICON release 2024.01 ICON Partnership et al. https://doi.org/10.35089/WDCC/IconRelease01

pyRTE-RRTMGP Robert Pincus et al. https://doi.org/10.5281/zenodo.16644555

Short summary

Most climate models cannot resolve clouds and cloud-radiation interactions at coarse horizontal resolutions of about 100 km, which introduces uncertainties. High-resolution models resolve clouds better but are expensive to run. We use short high-resolution simulations and artificial intelligence to learn the cloud-radiation interactions without making any assumptions about the small scales. We propose a new method that significantly reduces cloud related errors.