This repository consists of a collection of Python scripts encapsulating the base functions needed to reproduce the results. Additionally, it includes a Jupyter Notebook illustrating how to apply those functions.
Both GCM and RCM data used in this study ---from the Coupled Model Intercomparison Project Phase 5 (CMIP5) and CORDEX, respectively,--- are openly available from the Earth System Grid Federation (ESGF) portal. The code needed to implement the DeepESD model can be found in a Zenodo repository (Baño-Medina et al. 2022a; https://doi.org/10.5281/zenodo.6828303) associated to a previous publication (Baño-Medina et al. 2022b).
Baño-Medina, J., R. Manzanas, E. Cimadevilla, J. Fernández, J. González-Abad, A. S. Cofiño, and J. M. Gutiérrez, 2022a: Repository supporting the results presented in the manuscript on Downscaling Multi-Model Climate Projection Ensembles with Deep Learning (DeepESD): Contribution to CORDEX EUR-44. Zenodo.
Baño-Medina, J., R. Manzanas, E. Cimadevilla, J. Fernández, J. González-Abad, A. S. Cofiño, and J. M. Gutiérrez, 2022b: Downscaling multi-model climate projection ensembles with deep learning (deepesd): contribution to cordex eur-44. Geoscientific Model Development, 15 (17), 6747–6758.
Baño-Medina, J., M. Iturbide, J. Fernández, and J. M. Gutiérrez, 2024: Transferability and explainability of deep learning emulators for regional climate model projections: Perspectives for future applications Geoscientific Model Development, Submitted to Artificial Intelligence for the Earth Systems.