Sprecher
Beschreibung
The transformative power of ML unleashed in Earth System modelling is taking shape. Recent advances in building hybrid models combining mechanistic Earth system models grounded in physical understanding and machine learning models trained from huge amounts of data show promising results and are in the focus of international research initatives. However, the ongoing implementation of such models poses not only technical challenges, but also raises questions of trustworthiness and verifiability. The acceptance of hybrid approaches and full replacement models critically depends on the means scientists have for validating them. For further adoption of ML methods, it is crucial to establish the credibility of ML-enhanced Earth system models through statistical reproducibility. This contribution will structure the challenges and requirements, highlight emerging community approaches and provide a germination point for discussions between ESM and ML developers and downstream users.
