Reproducibility package for a physics-deep learning daily drought forecasting framework

Reproducibility package for the manuscript 'Physics-Deep Learning Hybrid Framework for Multi-Station Daily Drought Indices and Flash Drought Forecasting in the Jinsha River Basin'. This record contains the GitHub release source archive, processed station data, manuscript-derived table and figure data, full one-step and recursive prediction tables, and trained model checkpoints. The source code is released under the MIT License; processed data and derived artifacts are released under CC BY 4.0. The GitHub repository is https://github.com/ARETE-zzwl/jinsha-daily-drought-hybrid.