NoahPy: A differentiable Noah Land Surface Model

NoahPy is a recently developed, differentiable version of the Noah Land Surface Model, designed to reconstruct the original model. The development focuses on enabling the application of advanced machine learning techniques, like physics-guided deep learning, to improve the accuracy and physical consistency of land surface and atmosphere interactions. NoahPy aims to achieve this by providing a physically consistent, differentiable framework for simulations, which allows for better prediction of various earth system processes such as permafrost thermo-hydrology.

Model description paper:

• Tian, W., Yu, H., Zhao, S*., Cao, Y., Yi, W., Xu, J., and Nan, Z*.: NoahPy: A differentiable Noah land surface model for simulating permafrost thermo-hydrology, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2025-4253, 2025.

Technical note:

• 技术文档 (Tech notes), v1.1, in Chinese
• Readme.pdf

Version history:

• v1.0.1: https://doi.org/10.5281/zenodo.16530326, August 30, 2025. (Codes & test data available in this Zenodo link, the version used to produce results in the model description paper.)

Code repository:

• https://github.com/wbtian/NoahPy
• https://github.com/nanzt/NoahPy