Dataset for running the notebook: Robust Deep Learning Projections Reveal Intensification and Seasonal Reorganization of Fire Weather over the Iberian Peninsula

This repository contains the data required to run the Jupyter notebook main.ipynb from the GitHub project 2026_Mirones_DL_FWI_Projections. The materials provided here support the reproducibility of the analyses and results presented in the associated research workflow on deep learning (DL)–based fire weather index (FWI) projections through an adapted example. The repository includes all input data necessary for executing the notebook, including the preprocessed predictors and predictand to train the DL model, and pre-trained DL model .h5 file. These files are organized to facilitate direct integration with the notebook environment and to ensure consistency with the implementation hosted on GitHub. The purpose of this deposit is to provide a stable, citable, and long-term accessible source for the data dependencies of the project, in line with FAIR principles (Findable, Accessible, Interoperable, and Reusable). Users should download the contents of this repository and place them in the appropriate directory structure as specified in the project documentation before running the notebook. However, the reproducible notebook is prepared to download and allocate the required data in the corresponding directories automatically. Associated code and execution instructions are available in the GitHub repository: GitHub Project Repository This dataset is intended for research, validation, and reproducibility purposes. If used in derivative works, please cite both this Zenodo record and the associated publication/project repository.