Li et al. (2026) Improving Maize Yield Estimation Accuracy by Integrating Satellite Remote Sensing Data and the WOFOST Model Through Data Assimilation

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Agronomy
• Year: 2026
• Date: 2026-09-07
• Authors: Xuqing Li, Zekun Zhang, Zhihe Hu, Ligang Cui, Long Li, Tingxuan Wang, Tian Yang
• DOI: 10.3390/agronomy16171742

Research Groups

• Institute of Remote Sensing, Key Laboratory of Digital Earth Science, Chinese Academy of Sciences
• World Agroforestry Centre (ICRAF)
• University of Agricultural Sciences and Veterinary Medicine Cluj-Napoca

Short Summary

This study aimed to enhance summer maize yield estimation accuracy using data assimilation techniques and machine learning models. The results showed that data assimilation significantly improved WOFOST-based maize yield estimation, with ensemble Kalman filtering (EnKF) performing best.

Objective

• To investigate the effectiveness of data assimilation in improving summer maize yield estimation accuracy

Study Configuration

• Spatial Scale: Regional scale (study area not specified)
• Temporal Scale: Summer season (presumably June to September)

Methodology and Data

• Models used:
  • Partial least squares regression (PLSR)
  • Random forest (RF)
  • Extreme gradient boosting (XGBoost)
  • Support vector regression (SVR)
  • Convolutional neural network (CNN)
  • Categorical feature-enhanced gradient boosting (CatBoost)
• Data sources:
  • GF-1 imagery
  • Retrieved summer maize leaf area index (LAI) and measured soil moisture (SM)

Main Results

• All machine learning models effectively retrieved LAI, with CNN achieving the highest accuracy.
• Data assimilation significantly improved WOFOST-based maize yield estimation, with EnKF performing best for yield estimation.
• An ensemble size of 100 produced the lowest yield prediction error.
• Assimilation during the tasseling-to-grain-filling period resulted in lower yield prediction errors than continuous whole-growth-period assimilation.

Contributions

This study contributes to improving summer maize yield estimation accuracy using data assimilation techniques and machine learning models, supporting regional crop monitoring and precision agricultural management.

Funding

• This research was supported by the National Key Research and Development Program of China (Grant No. 2017YFC1502404)
• The authors acknowledge funding from the Chinese Academy of Sciences (CAS) Strategic Priority Research Program (Grant No. XDA19040201)
• ICRAF received support from the CGIAR Fund Donors

Citation

@article{Li2026Improving,
  author = {Li, Xuqing and Zhang, Zekun and Hu, Zhihe and Cui, Ligang and Li, Long and Wang, Tingxuan and Yang, Tian},
  title = {Improving Maize Yield Estimation Accuracy by Integrating Satellite Remote Sensing Data and the WOFOST Model Through Data Assimilation},
  journal = {Agronomy},
  year = {2026},
  doi = {10.3390/agronomy16171742},
  url = {https://doi.org/10.3390/agronomy16171742}
}