低效林改培碳汇响应的阶段性转折点判别机制

Mechanism for Identifying Stage-Specific Turning Points in Carbon Sink Response during Low-Efficiency Forest Transformation

摘要: 目的：明确低效林改培过程中森林碳汇的阶段性演化特征，构建转折点识别机制，为碳储动态监测与碳管理提供技术支持。方法：以陕西商洛市中幼龄低效林为对象，整合2000~2023年多源数据(Landsat遥感指数、实地碳含量与生物量)，采用滑动窗口法与分段回归模型，量化碳储年增长斜率变化，进而识别碳储积累的关键转折点。结果：地上碳储在改培后第22年与第31年出现加速与趋稳转折，年增速由0.9 Mg∙ha⁻¹∙年⁻¹升至3.8 Mg∙ha⁻¹∙年⁻¹，后回落至1.0 Mg∙ha⁻¹∙年⁻¹；土壤碳储响应滞后约1.5年，相应转折点出现在第23.5年与32.6年。结论：构建的“滑动窗口 + 分段回归 + 多指标融合”机制，能有效捕捉低效林碳汇跃迁的阶段性信号。本研究的创新点在于提出了融合多源时序数据与统计模型的转折点判别框架，增强了动态监测的量化能力；明确了地上与土壤碳汇响应存在约1.5年的系统性滞后，为差异化碳管理提供了时序依据。该机制可为森林碳汇精准评估与生态服务管理提供科学支持。

Abstract: [Objective] This study aims to clarify the staged evolutionary characteristics of forest carbon sinks during the reconstruction and cultivation of low-efficiency forests, and establish a turning point identification mechanism to provide technical support for the dynamic monitoring and management of carbon storage. [Method] Taking young and middle-aged low-efficiency forests in Shangluo City, Shaanxi Province as the research object, multi-source data from 2000 to 2023 (including Landsat remote sensing indices, field-measured carbon content and biomass) were integrated. The sliding window method and piecewise regression model were adopted to quantify the annual change slope of carbon storage, and further identify the key turning points of carbon storage accumulation. [Result] The aboveground carbon storage showed accelerated and stabilized turning points in the 22nd and 31st years after reconstruction and cultivation, with the annual growth rate increasing from 0.9 Mg·ha⁻¹·yr⁻¹ to 3.8 Mg·ha⁻¹·yr⁻¹, and then decreasing to 1.0 Mg·ha⁻¹·yr⁻¹. The response of soil carbon storage lagged by approximately 1.5 years, with the corresponding turning points appearing in the 23.5th and 32.6th years. [Conclusion] The established mechanism of “sliding window + piecewise regression + multi-index fusion” can effectively capture the staged signals of carbon sink transition in low-efficiency forests. The innovation of this study lies in proposing a turning point discrimination framework integrating multi-source time-series data and statistical models, which enhances the quantitative capability of dynamic monitoring. It is clarified that there is a systematic lag of about 1.5 years between the responses of aboveground and soil carbon sinks, providing a temporal basis for differentiated carbon management. This mechanism can provide scientific support for the accurate assessment of forest carbon sinks and the management of ecological services.