Hu et al. (2025) Modeling of Drought-Induced Crop Yield Loss Based on Solar-Induced Chlorophyll Fluorescence by Machine Learning Methods

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

Identification

• Journal: Atmosphere
• Year: 2025
• Date: 2025-12-28
• Authors: Han Hu, Minxue Zheng, Yue Niu, Qiu Shen, Qinyao Ren, Yanlin You
• DOI: 10.3390/atmos17010042

Research Groups

Not explicitly mentioned in the provided text.

Short Summary

This study developed a model integrating solar-induced chlorophyll fluorescence (SIF), vegetation indices, and meteorological data to quantify drought-induced yield reduction in winter wheat, finding SIF to be a superior indicator for accurate yield loss prediction.

Objective

• To quantify drought-induced crop yield loss in winter wheat for safeguarding national food security by developing an integrated evaluation model.

Study Configuration

• Spatial Scale: Not explicitly defined, but implied to be a "major grain production region."
• Temporal Scale: Not explicitly defined, but the study captures "interannual fluctuations" in yield, implying multiple years of data.

Methodology and Data

• Models used: Random Forest (RF), Extreme Gradient Boosting (XGBoost).
• Data sources: Solar-induced chlorophyll fluorescence (SIF), vegetation indices (VIs) including near-infrared reflectance of vegetation (NIRv), meteorological data. Data were processed as normalized anomalies.

Main Results

• SIF effectively captured interannual fluctuations in winter wheat yield and served as a reliable quantitative indicator of yield variation.
• Developed models, utilizing vegetation data (SIF, NIRv) and meteorological variables, could directly quantify drought-induced yield losses based on normalized anomalies without complex computations. SIF demonstrated superior performance among all tested variable combinations, yielding the most accurate predictions.
• Both Random Forest (RF) and Extreme Gradient Boosting (XGBoost) algorithms showed similar performance in evaluating drought-induced yield loss.

Contributions

• Developed a novel, simpler approach for directly quantifying drought-induced winter wheat yield loss by integrating SIF, VIs, and meteorological data, reducing the need for additional auxiliary data.
• Demonstrated the superior performance of SIF as a quantitative indicator for capturing interannual yield fluctuations and predicting drought-induced yield loss.
• Highlighted the advantages of combining normalized anomalies from multiple data sources for quick monitoring and early warning of crop yield loss in major grain production regions.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Hu2025Modeling,
  author = {Hu, Han and Zheng, Minxue and Niu, Yue and Shen, Qiu and Ren, Qinyao and You, Yanlin},
  title = {Modeling of Drought-Induced Crop Yield Loss Based on Solar-Induced Chlorophyll Fluorescence by Machine Learning Methods},
  journal = {Atmosphere},
  year = {2025},
  doi = {10.3390/atmos17010042},
  url = {https://doi.org/10.3390/atmos17010042}
}