Quantification of the Influencing Factors of Stand Productivity of Subtropical Natural Broadleaved Forests in Eastern China Using an Explainable Machine Learning Framework

Forests, Год журнала: 2025, Номер 16(1), С. 95 - 95

Опубликована: Янв. 8, 2025

Natural broadleaf forests (NBFs) are the most abundant zonal vegetation type in subtropical regions. Understanding mechanisms influencing stand productivity NBFs is important for developing “nature-based” solutions climate change mitigation. However, minimal research has captured effects of nonlinearities and feature interactions that often have nonlinear impacts on factors. To address this gap, we used continuous forest inventory data, a machine learning model was constructed. Subsequently, through leveraging interpretable framework SHapley Additive explanation (SHAP) partial dependence plot, determined global local explanations factors productivity. Our findings indicate following: (1) The Autogluon performed strongest based R2, RMSE, rRMSE metrics. (2) basal area (BA), neighborhood comparison diameter at breast height (NC), age (AGE) were key Stand increased with increasing BA decreased NC AGE. maintained above 15 m2ha−1 below 0.45, which represent favorable conditions to maintain optimal growth. (3) SHAP interaction values calculated determine five major study provides reference sustainable management NBFs, thereby highlighting role mitigating change.

Язык: Английский

The negative effects of competition on tree resilience weakened as drought severity intensified

Agricultural and Forest Meteorology, Год журнала: 2025, Номер 368, С. 110544 - 110544

Опубликована: Апрель 9, 2025

Язык: Английский