EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An RF-Guided Dual-Strategy Feature-Selection Framework for Multi-Source Remote-Sensing-Based Estimation of Forest Aboveground Carbon Stock in Mountainous Terrain

Yong Jiang, Jialong Zhang, Chenkai Teng, Yongming Ma, Zhixian Ding, Sha Li, Deguo Kong, Xinling Zhong

Remote Sensing · 29.05.2026 · 18 (11) · 1748

DOI: 10.3390/rs18111748 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 29.05.2026. Verfasst von Yong Jiang und 7 weiteren Autor:innen. Die Arbeit nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Yunnan Fundamental Research Projects, National Natural Science Foundation of China, Southwest Forestry University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study proposed an RF-guided heuristic feature-selection framework that integrates multi-source remote-sensing data for estimating Pinus densata aboveground carbon stock (AGCS) in Shangri-La, Yunnan Province, China. Compared with four baseline feature-selection methods, the Random Forest–Alpha Evolution (RFA) and Random Forest–Markov Chain Monte Carlo (RFM) algorithms generated more informative feature subsets and improved model performance, with the Optuna-optimized AdaBoost model based on RFM features achieving the highest accuracy (R2 = 0.71, RMSE = 10.53 t/ha). These results suggest that RF-guided heuristic feature selection can effectively improve AGCS estimation in complex mountainous environments. Vegetation indices and texture features were consistently prioritized across different feature-selection methods. Shapley Additive Explanations (SHAP)-based interpretation revealed that the most influential predictors were the Sentinel-2A green normalized difference vegetation index (S2_GNDVI) and precipitation of the wettest month (bio13) in the RFA Method, and the Sentinel-2A red-edge normalized difference vegetation index (S2_NDVI45) and bio13 in the RFM Method. These findings underscore the critical importance of canopy greenness, moisture availability, and structural complexity in regulating carbon accumulation in montane conifer forests. The final AGCS maps yielded total estimates of 9.83 Mt (RFA) and 10.46 Mt (RFM), and revealed a consistent spatial pattern, with moderate AGCS values dominating the landscape and a general tendency for higher values in the northwest and lower values in the southeast. In summary, the combination of RF-guided heuristic feature selection, Optuna-optimized machine learning and SHAP provides an effective and interpretable framework for AGCS estimation in mountain forests.

Förderung

  • Yunnan Fundamental Research Projects
  • National Natural Science Foundation of China
  • Southwest Forestry University

Zitieren

APA

Jiang, Y., Zhang, J., Teng, C., Ma, Y., Ding, Z., Li, S., Kong, D., & Zhong, X. (2026). An RF-Guided Dual-Strategy Feature-Selection Framework for Multi-Source Remote-Sensing-Based Estimation of Forest Aboveground Carbon Stock in Mountainous Terrain. Remote Sensing, 18(11), 1748. https://doi.org/10.3390/rs18111748

Vancouver

Jiang Y, Zhang J, Teng C, Ma Y, Ding Z, Li S, et al. An RF-Guided Dual-Strategy Feature-Selection Framework for Multi-Source Remote-Sensing-Based Estimation of Forest Aboveground Carbon Stock in Mountainous Terrain. Remote Sensing. 2026;18(11):1748. doi:10.3390/rs18111748

BibTeX
@article{Jiang2026,
  author = {Jiang, Yong and Zhang, Jialong and Teng, Chenkai and Ma, Yongming and Ding, Zhixian and Li, Sha and Kong, Deguo and Zhong, Xinling},
  title = {An RF-Guided Dual-Strategy Feature-Selection Framework for Multi-Source Remote-Sensing-Based Estimation of Forest Aboveground Carbon Stock in Mountainous Terrain},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {11},
  pages = {1748},
  doi = {10.3390/rs18111748}
}
Kostenlos starten →