EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A generative likelihood framework for high-resolution climate model evaluation

Lilli Johanna Freischem, Tim Reichelt, Ronald Clark, Philip Stier, Hannah Christensen

Environmental Data Science · 2026 · 5

DOI: 10.1017/eds.2026.10054 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmental Data Science, online erschienen am 14.07.2026. Verfasst von Lilli Johanna Freischem und 4 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Environment Research Council.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Next-generation high-resolution (km-scale) climate models promise unprecedented accuracy in climate projections, but realizing their potential requires robust methods to quantify how well simulations align with real-world observations. Average-based metrics conventionally used for climate model evaluation ignore the physics encoded in the fine-scale structures of km-scale simulations. To overcome this limitation, we propose a novel, statistically principled evaluation methodology based on the likelihood function of a generative image model. Our method provides a continuous similarity metric derived from the likelihood distribution of observation and simulation snapshots, which can redefine the evaluation, intercomparison, and parameter tuning of high-resolution climate models. We demonstrate the applicability and interpretability of this method by evaluating convective clouds simulated by two state-of-the-art global km-scale models, using their outgoing infrared radiation fields. This work establishes a scalable pathway toward observation-based evaluation of next-generation climate simulations.

Förderung

  • Natural Environment Research Council

Zitieren

APA

Freischem, L. J., Reichelt, T., Clark, R., Stier, P., & Christensen, H. (2026). A generative likelihood framework for high-resolution climate model evaluation. Environmental Data Science, 5. https://doi.org/10.1017/eds.2026.10054

Vancouver

Freischem LJ, Reichelt T, Clark R, Stier P, Christensen H. A generative likelihood framework for high-resolution climate model evaluation. Environmental Data Science. 2026;5. doi:10.1017/eds.2026.10054

BibTeX
@article{Freischem2026,
  author = {Freischem, Lilli Johanna and Reichelt, Tim and Clark, Ronald and Stier, Philip and Christensen, Hannah},
  title = {A generative likelihood framework for high-resolution climate model evaluation},
  journal = {Environmental Data Science},
  year = {2026},
  volume = {5},
  doi = {10.1017/eds.2026.10054}
}
Kostenlos starten →