Planted forests play a crucial role in addressing global climate change and are also valued globally for their numerous ecosystem services. Therefore, it is essential to understand how biotic and abiotic factors affect the carbon sequestration potential. This study focuses on quantifying the effects of 26 different variables on the carbon sequestration potential of Larix spp. plantations in northeast China, utilizing the random forest algorithm (RF). To eliminate the age-related tendency of stand carbon stock, a novel carbon sequestration index (CSI) was defined, which measures the ratio of actual to predicted stand carbon stocks for a stand at a specific age. The results indicated that the developed RF model explained approximately 64.75 % of the variations of CSI. Among the four categories of variables analyzed, stand variables (35.73 %) contributed significantly more than terrain variables (3.31 %), soil variables (3.68 %), and climate variables (9.06 %). The partial dependence analysis revealed that the Larix spp. plantations had a potential maximum carbon stock of approximately 73.34 t·ha-1. This potential was associated with certain attributes, including a stand mean diameter of 15 cm, a stand density of 1700 trees·ha-1, a stand basal area of 30 m2·ha-1, and a crown density of 0.7, respectively. These findings provide insightful information for plantation management to improve stand carbon stocks in northeast China with attempting to mitigate climate change.
Keywords: Carbon sequestration; Driving factor; Larix spp. plantation; Random forest algorithm.
Copyright © 2023. Published by Elsevier B.V.