Toward Generative Machine Learning for Boosting Ensembles of Climate Simulations

Abstract Accurately quantifying uncertainty in predictions and projections arising from irreducible internal climate variability is critical for decision‐making. Such uncertainty is typically assessed using ensembles produced with climate models. However, computational constraints impose a trade‐off between generating large ensembles required for robust uncertainty estimation and increasing model resolution to better capture fine‐scale dynamics. Generative machine learning offers a promising pathway to alleviate these constraints. We develop a conditional Variational Autoencoder (cVAE) trained on a limited sample of climate simulations to generate arbitrary large ensembles. The approach is applied to output from monthly CMIP6 historical and scenario experiments produced with the Canadian Centre for Climate Modelling and Analysis' Earth system model CanESM5. We show that the cVAE model learns the underlying distribution of data and generates physically consistent samples that reproduce realistic low‐ and high‐moment statistics, including extremes. Compared with more sophisticated generative architectures, cVAEs offer mathematically transparent, interpretable, and c