WOFOST-EW v1: Enhanced WOFOST for Extreme Weather

• WOFOST-EW v1: Enhanced WOFOST for Extreme Weather

WOFOST-EW v1 is an improved version of the WOFOST (World Food Studies Simulation Model) that improves crop growth simulations under extreme weather conditions. The model incorporates an Extreme Weather Function that integrates Extreme Weather Indices with an LSTM deep learning algorithm to improve prediction accuracy. In addition, the SCE-UA optimization algorithm is applied to achieve more efficient parameter calibration.

---

• Resources

WOFOST-EW v1 Source Code
- GitHub repository: https://github.com/zheng-jinhui/WOFOST-EW

WOFOST Model
The base model is part of the Python Crop Simulation Environment (PCSE):
- PCSE Documentation
- PCSE GitHub repository by Allard de Wit

SCE-UA Algorithm
For parameter calibration, the SCE-UA algorithm implementation is referenced in Spotpy:
- Spotpy Documentation

LSTM Implementation
The LSTM algorithm is implemented using the Keras library:
- Keras Documentation

---

• Publications

This version of the WOFOST model has been used in the publication:  

> Zheng, J., Yu, L., Du, Z., Xiao, L., and Huang, X.: Modeling wheat development under extreme weather with WOFOST-EW v1, Geosci. Model Dev., 18, 8379–8400, https://doi.org/10.5194/gmd-18-8379-2025, 2025.