Chen et al. (2026) A spatial downscaling framework for vegetation optical depth based on geographically weighted stacking ensemble learning

Identification

• Journal: Ecological Indicators
• Year: 2026
• Date: 2026-09-11
• Authors: Xinru Chen, Jianhui Xu, Tian Zhao, Feng Hu
• DOI: 10.1016/j.ecolind.2026.115479

Research Groups

• School of Resources and Environmental Engineering, Anhui University
• School of Geographic Information and Tourism, Chu Zhou University

Short Summary

This study developed a geographically weighted stacking ensemble (GWSE) downscaling framework to generate high-resolution vegetation optical depth (VOD) products for the Yangtze River Economic Belt. The GWSE model achieved better downscaling performance than individual machine learning models, with an R2 of 0.702, an RMSE of 0.180, and an MAE of 0.124.

Objective

• To develop a high-resolution VOD downscaling framework for the Yangtze River Economic Belt.
• To investigate the relationships between VOD and auxiliary variables using machine learning models.

Study Configuration

• Spatial Scale: Regional scale (Yangtze River Economic Belt)
• Temporal Scale: Monthly time series

Methodology and Data

• Models used: Random Forest, Extreme Gradient Boosting, Light Gradient Boosting Machine, Geographically Weighted Regression, Stacking ensemble framework
• Data sources: AMSR2 VOD data, MODIS products (NDVI, EVI, LAI, LST), meteorological data, topographic data

Main Results

• The GWSE model achieved better downscaling performance than individual machine learning models.
• The R2 of the GWSE model was 0.702, with an RMSE of 0.180 and an MAE of 0.124.
• The spatial correlations between VOD and NDVI, EVI, LAI, and GPP were enhanced after downscaling.

Contributions

• This study provides a feasible approach for generating regional high-resolution VOD products.
• The GWSE model offers data support for vegetation dynamics monitoring and ecological environment assessment.

Funding

• National Natural Science Foundation of China (Grant No. 42141004)
• Anhui Provincial Natural Science Foundation (Grant No. 2008085QH213)

Citation

@article{Chen2026spatial,
  author = {Chen, Xinru and Xu, Jianhui and Zhao, Tian and Hu, Feng},
  title = {A spatial downscaling framework for vegetation optical depth based on geographically weighted stacking ensemble learning},
  journal = {Ecological Indicators},
  year = {2026},
  doi = {10.1016/j.ecolind.2026.115479},
  url = {https://doi.org/10.1016/j.ecolind.2026.115479}
}