Geoscientific Model Development
Geoscientific Model Development
GMD
Articles | Volume 19, issue 12
Articles | Volume 19, issue 12
https://doi.org/10.5194/gmd-19-5781-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-19-5781-2026
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
Articles | Volume 19, issue 12
Development and technical paper
 | 
01 Jul 2026
Development and technical paper |  | 01 Jul 2026

Pre-training for deep statistical climate downscaling: enhancing consistency and robustness across regional datasets

Jose González-Abad, Maialen Iturbide, Alfonso Hernanz, and José Manuel Gutiérrez

Data sets

Predictors for the manuscript "Pre-training for Deep Statistical Climate Downscaling: A case study within the Spanish National Adaptation Plan (PNACC)" Jose González-Abad https://doi.org/10.5281/zenodo.16687087

Predictands for the manuscript "Pre-training for Deep Statistical Climate Downscaling: A case study within the Spanish National Adaptation Plan (PNACC)" Jose González-Abad and Spanish Meteorological Agency (AEMET) https://doi.org/10.5281/zenodo.17338348

Model code and software

jgonzalezab/pretraining_PP_downscaling: v1.0 Jose González-Abad https://doi.org/10.5281/zenodo.18467873

Deep learning models used in the manuscript "Pre-training for Deep Statistical Climate Downscaling: A case study within the Spanish National Adaptation Plan (PNACC)" Jose González-Abad https://doi.org/10.5281/zenodo.18468086

Download
Short summary
We explore how deep learning can improve local climate projections by adapting a national model to regional data. By relying on a paradigm called pre-training, we show that models can produce more consistent and physically aligned results, even when data is limited. This helps make future climate projections more reliable and supports better planning at both national and local levels.
Read more
Share