Mapping Aboveground Biomass and Spatial Uncertainty in Temperate Grasslands Using Landsat 8 and Sentinel‐2 Imagery
Land Degradation & Development · 30.07.2026 · 37 (12) · 8707-8724
DOI: 10.1002/ldr.70689 ↗Kurz gefasst
Fachartikel in der Zeitschrift Land Degradation & Development, online erschienen am 03.06.2026. Verfasst von Xifeng Zhang und 7 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China.
Abstract (Originalsprache)
Accurate estimation of grassland aboveground biomass (AGB) is essential for grazing regulation and ecological compensation. This study focuses on temperate grasslands in western Jilin Province, China, and develops an optical remote sensing AGB modeling framework that progresses from single‐source to multi‐source strategies. AGB models were independently constructed using vegetation indices derived from Landsat 8 and Sentinel‐2 as single‐source baselines. Building on this, two multi‐source fusion strategies were implemented: feature‐level fusion, which constructs a joint feature space from multi‐sensor vegetation indices, and decision‐level fusion, which integrates predictions from independently trained single‐source models. Based on data from 49 sampling plots (245 quadrats), the performances of partial least squares regression (PLSR), ElasticNet, and random forest models under different modeling strategies were systematically evaluated. The results show that under single‐source conditions, Sentinel‐2 outperformed Landsat 8 in AGB estimation, achieving a maximum coefficient of determination ( R 2 ) of 0.78. Multi‐source fusion further improved model accuracy and robustness, with decision‐level fusion performing best, increasing R 2 to 0.81 and reducing RMSE and MAE by approximately 7%–11% relative to the best single‐source model. Using the optimal model, a regional AGB map at 30 m resolution was generated, and pixel‐level prediction uncertainty was quantified using bootstrap resampling. Overall, this study proposes an optical multi‐source fusion AGB retrieval framework tailored to degraded grasslands that integrates multi‐sensor comparison, fusion strategy evaluation, and spatial uncertainty characterization, providing practical support for grassland management and risk identification.
Förderung
- National Key Research and Development Program of China
- National Natural Science Foundation of China
Zitieren
APA
Zhang, X., Xu, L., Li, Y., Li, S., Zhao, D., Li, J., Nyongesa, J. M., & Ma, H. (2026). Mapping Aboveground Biomass and Spatial Uncertainty in Temperate Grasslands Using Landsat 8 and Sentinel‐2 Imagery. Land Degradation & Development, 37(12), 8707-8724. https://doi.org/10.1002/ldr.70689
Vancouver
Zhang X, Xu L, Li Y, Li S, Zhao D, Li J, et al. Mapping Aboveground Biomass and Spatial Uncertainty in Temperate Grasslands Using Landsat 8 and Sentinel‐2 Imagery. Land Degradation & Development. 2026;37(12):8707-8724. doi:10.1002/ldr.70689
BibTeX
@article{Zhang2026,
author = {Zhang, Xifeng and Xu, Lu and Li, Yaxiao and Li, Shaoyang and Zhao, Dandan and Li, Jianguo and Nyongesa, John Maina and Ma, Hongyuan},
title = {Mapping Aboveground Biomass and Spatial Uncertainty in Temperate Grasslands Using Landsat 8 and Sentinel‐2 Imagery},
journal = {Land Degradation & Development},
year = {2026},
volume = {37},
number = {12},
pages = {8707--8724},
doi = {10.1002/ldr.70689}
}
EUVIMED