Yan et al. (2025) Research on Grassland Classification Method in Water Conservation Areas of the Qinghai–Tibet Plateau Based on Multi-Source Data Fusion

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

Identification

• Journal: Agriculture
• Year: 2025
• Date: 2025-12-01
• Authors: Kexin Yan, Yueming Hu, Lu Wang, Xiaoyan Huang, Runyan Zou, Liangjun Zhao, Fan Yang, Taolue Wen
• DOI: 10.3390/agriculture15232503

Research Groups

[Not explicitly mentioned in the provided text.]

Short Summary

This study developed a novel grassland classification method for the Qinghai–Tibet Plateau by integrating multi-source remote sensing data with machine learning algorithms. The XGBoost model demonstrated the best performance (accuracy of 0.829), revealing that climate and topography are key drivers of alpine grassland distribution.

Objective

• To develop a novel grassland classification method that integrates multi-source remote sensing data with machine learning algorithms for the Qinghai–Tibet Plateau.

Study Configuration

• Spatial Scale: Hongyuan County, Sichuan Province, in the water conservation area of the Qinghai–Tibet Plateau.
• Temporal Scale: Contemporary, based on multi-source remote sensing imagery (specific dates not provided).

Methodology and Data

• Models used: Convolutional neural network (CNN), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), histogram gradient boosting (HistGradientBoosting), and random forest (RF).
• Data sources: Landsat 8 imagery (15 m and 30 m spatial resolution), MOD15A2 imagery (500 m spatial resolution), ALOS imagery (12.5 m spatial resolution), and 435 field survey samples.

Main Results

• The XGBoost model achieved the best classification performance with an accuracy of 0.829, Precision of 0.818, Recall of 0.829, weighted F1-score of 0.820, and an AUC value of 0.870.
• Key factors influencing alpine grasslands, identified by SHAP values, were mean annual precipitation (MAP, 0.675), >0 °C Accumulated Temperature (AT, 0.591), and aspect (ASPECT, 0.548).
• The driving mechanism of grassland differentiation is characterized by climate dominance, topographic regulation, soil support, and vegetation response.
• Pixel-by-pixel analysis showed that 75.82% of regions had no discrepancy, 23.63% had a minor discrepancy (absolute value = 1), and only 0.54% had a major discrepancy (absolute value = 2) between predicted and actual classifications.

Contributions

• Established a replicable paradigm for the precise management and conservation of alpine grassland resources.
• Generated superior baseline data through the synergistic application of deep learning and machine learning.
• Quantitatively uncovered a grassland differentiation mechanism dominated by hydrothermal factors and fine-tuned by topography in the complex Qinghai–Tibet Plateau.
• Delivered high-precision spatial distribution maps of grassland classes.

Funding

[Not explicitly mentioned in the provided text.]

Citation

@article{Yan2025Research,
  author = {Yan, Kexin and Hu, Yueming and Wang, Lu and Huang, Xiaoyan and Zou, Runyan and Zhao, Liangjun and Yang, Fan and Wen, Taolue},
  title = {Research on Grassland Classification Method in Water Conservation Areas of the Qinghai–Tibet Plateau Based on Multi-Source Data Fusion},
  journal = {Agriculture},
  year = {2025},
  doi = {10.3390/agriculture15232503},
  url = {https://doi.org/10.3390/agriculture15232503}
}