EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainable Machine Learning Reveals Persistent Carbon Sink in Xishuangbanna Tropical Forests Under Future Climate Scenarios

Chenjia Zhang, Dingman Li, Luping Zhang, Yuxuan Zhu, Zhengquan Zhou, Daokun Ma, Yan Zhang, Feiri Ali, Yusheng Han

Forests · 06.04.2026 · 17 (4) · 456

DOI: 10.3390/f17040456 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Forests, online erschienen am 06.04.2026. Verfasst von Chenjia Zhang und 8 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 68 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Silk Road Economic Belt Innovation-Driven Development Pilot Zone, WuChangShi National Independent Innovation Demonstration Zone Project, Science and Technology Program of Jiangxi Province, Sichuan Provincial Science and Technology Program Project.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Tropical forests are predicted to become carbon sources by mid-century under climate change. However, this trajectory may not be inevitable for forests under long-term protection. Using 12 years of eddy covariance flux data from a long-term protected tropical rainforest site in Xishuangbanna, China, we develop an explainable machine learning framework (SHAP + structural equation modeling) to disentangle the environmental drivers of net ecosystem exchange (NEE) and evapotranspiration (ET), and project their future trajectories under four CMIP6 climate scenarios. We find a fundamental divergence: while conventional climate models predict a sink-to-source transition by 2050–2066, our data-driven model—trained on conservation-era observations—projects a persistent carbon sink through 2100 across all the scenarios. This divergence suggests that long-term protection may buffer tropical forests against climate-driven decline, challenging the prevailing narrative of inevitable carbon loss. We further identify critical environmental thresholds—solar radiation (~200 W m−2) and air temperature (~25 °C)—beyond which carbon uptake efficiency declines. Our findings provide empirical support for nature-based climate solutions and highlight the need to integrate conservation legacies into Earth system models.

Förderung

  • Silk Road Economic Belt Innovation-Driven Development Pilot Zone, WuChangShi National Independent Innovation Demonstration Zone Project
  • Science and Technology Program of Jiangxi Province
  • Sichuan Provincial Science and Technology Program Project
  • Liangshan Prefecture Science and Technology Program Project
  • Xichang University Doctoral Start-up Fund

Zitieren

APA

Zhang, C., Li, D., Zhang, L., Zhu, Y., Zhou, Z., Ma, D., Zhang, Y., Ali, F., & Han, Y. (2026). Explainable Machine Learning Reveals Persistent Carbon Sink in Xishuangbanna Tropical Forests Under Future Climate Scenarios. Forests, 17(4), 456. https://doi.org/10.3390/f17040456

Vancouver

Zhang C, Li D, Zhang L, Zhu Y, Zhou Z, Ma D, et al. Explainable Machine Learning Reveals Persistent Carbon Sink in Xishuangbanna Tropical Forests Under Future Climate Scenarios. Forests. 2026;17(4):456. doi:10.3390/f17040456

BibTeX
@article{Zhang2026,
  author = {Zhang, Chenjia and Li, Dingman and Zhang, Luping and Zhu, Yuxuan and Zhou, Zhengquan and Ma, Daokun and Zhang, Yan and Ali, Feiri and Han, Yusheng},
  title = {Explainable Machine Learning Reveals Persistent Carbon Sink in Xishuangbanna Tropical Forests Under Future Climate Scenarios},
  journal = {Forests},
  year = {2026},
  volume = {17},
  number = {4},
  pages = {456},
  doi = {10.3390/f17040456}
}
Kostenlos starten →