Guo et al. (2026) Evaluating the correction methods for NDVI angular effect based on decametric resolution BRDF observations over cropland

Identification

• Journal: Remote Sensing of Environment
• Year: 2026
• Date: 2026-04-03
• Authors: Yuhan Guo, Xihan Mu, Xiang Lyu, Dasheng Fan, Chengzhuo Lei, Donghui Xie, Guangjian Yan
• DOI: 10.1016/j.rse.2026.115403

Research Groups

• State Key Laboratory of Remote Sensing and Digital Earth, Faculty of Geographical Science, Beijing Normal University, Beijing, China
• Beijing Engineering Research Center for Global Land Remote Sensing Products, Faculty of Geographical Science, Beijing Normal University, Beijing, China

Short Summary

This study comprehensively evaluated seven methods for correcting the Normalized Difference Vegetation Index (NDVI) angular effect using over 1,800 UAV-based BRDF observations and Sentinel-2 time series across seven crop types. It found that NDVI-Coupled methods, particularly the newly proposed mSVAC, significantly outperformed NDVI-Independent methods in reducing angular effects at decametric resolution.

Objective

• To comprehensively evaluate and differentiate the effectiveness of seven representative methods for correcting the NDVI angular effect in decametric resolution data, categorized as NDVI-Independent and NDVI-Coupled types.

Study Configuration

• Spatial Scale: Decametric resolution (e.g., Sentinel-2), with comparisons to 500-meter resolution (MODIS) for parameter transfer.
• Temporal Scale: Key crop growth stages, Sentinel-2 time series.

Methodology and Data

• Models used: Seven representative NDVI angular correction methods (categorized as NDVI-Independent and NDVI-Coupled), including the newly-proposed mSVAC method.
• Data sources:
  • Unmanned aerial vehicle (UAV)-based bidirectional reflectance distribution function (BRDF) observations (over 1,800 observations).
  • Sentinel-2 time series.
  • Field multi-angle data.
  • Moderate Resolution Imaging Spectroradiometer (MODIS) BRDF parameters (for comparison).

Main Results

• NDVI-Coupled methods generally outperformed NDVI-Independent methods in decametric resolution scenes.
• NDVI-Coupled methods reduced the mean absolute error (MAE) by approximately 38% during rapid vegetation growth.
• NDVI-Independent methods reduced the MAE by 9%.
• The newly-proposed mSVAC method achieved over 50% MAE reduction and demonstrated the best performance in improving the smoothness of Sentinel-2 NDVI time series.
• Transferring 500-meter MODIS BRDF parameters to decametric resolution introduced greater uncertainties in correction.

Contributions

• Provides a meaningful evaluation framework for NDVI angular correction methods.
• Offers valuable guidance for generating more stable NDVI products.
• Introduces the mSVAC method, which shows superior performance in mitigating backward NDVI correction issues.

Funding

• Not specified in the provided text.

Citation

@article{Guo2026Evaluating,
  author = {Guo, Yuhan and Mu, Xihan and Lyu, Xiang and Fan, Dasheng and Lei, Chengzhuo and Xie, Donghui and Yan, Guangjian},
  title = {Evaluating the correction methods for NDVI angular effect based on decametric resolution BRDF observations over cropland},
  journal = {Remote Sensing of Environment},
  year = {2026},
  doi = {10.1016/j.rse.2026.115403},
  url = {https://doi.org/10.1016/j.rse.2026.115403}
}