Data Sheet 1_A novel light use efficiency model incorporating stand age to improve monitoring of mangrove productivity and biomass accumulation.docx

Accurate estimation of gross primary production (GPP) and above-ground biomass (AGB) is fundamental to assessing the carbon sequestration potential of artificial mangrove wetlands. However, pronounced spatiotemporal heterogeneity in stand structure, particularly in restored mangrove forests with diverse age compositions, introduces substantial uncertainty in GPP and AGB quantification. This study presents an innovative framework that explicitly incorporates stand age into the light use efficiency (LUE) model as a physiological constraint, thereby enhancing the accuracy of GPP and AGB estimations. Stand age was mapped using Landsat-7 and sentinel-2 time-series imagery and a random forest classification approach on the Google Earth Engine platform, providing high spatial resolution age distributions. Age-dependent productivity constraints, derived from net primary production–age relationships observed in evergreen broadleaf ecosystems, were incorporated into the LUE model to refine photosynthetic efficiency estimations. Application of this framework to mangrove plantations in the Luoyangjiang Estuary (2000–2022) yielded high accuracy in GPP (RMSE = 9.66 g d−1, R2 = 0.95) and AGB (RMSE = 1,051 g·m−2, R2 = 0.63) estimations. The results captured exponential AGB growth with stand development, and spatial analysis demonstrated a strong correspondence between biomass distribution and stand age, with mature stands (≥20 years) contributing disproportionately to carbon accumulation. This stand age–integrated approach delivers fine spatial and temporal resolution, offering a practical and transferable tool for monitoring carbon dynamics and informing adaptive management strategies in restored coastal wetlands, thereby supporting the long-term assessment of blue carbon projects.

精准估算总初级生产量（gross primary production, GPP）与地上生物量（above-ground biomass, AGB），是评估人工红树林湿地碳汇潜力的核心基础。然而，林分结构存在显著的时空异质性——尤其是林龄组成多样的恢复红树林，这为GPP与AGB的定量测算带来了极大不确定性。本研究提出了一种创新框架，将林龄作为生理约束显式融入光能利用率（light use efficiency, LUE）模型，从而提升了GPP与AGB的估算精度。研究基于谷歌地球引擎（Google Earth Engine）平台，利用陆地卫星7号（Landsat-7）与哨兵2号（Sentinel-2）时序影像，结合随机森林分类方法完成林龄制图，获取了高空间分辨率的林龄分布数据。研究从常绿阔叶生态系统中观测得到的净初级生产量-林龄关系中，推导得到林龄依赖的生产力约束条件，并将其融入LUE模型以优化光合效率估算。将该框架应用于洛阳江河口2000–2022年的人工红树林种植区，GPP估算精度达RMSE=9.66 g·d⁻¹、R²=0.95，AGB估算精度达RMSE=1051 g·m⁻²、R²=0.63，结果表现优异。研究结果揭示了随林分发育呈现指数增长的AGB变化规律，空间分析表明生物量分布与林龄存在极强的对应关系：成熟林（≥20年）对碳积累的贡献远超其面积占比。这种整合林龄的估算方法具备精细的时空分辨率，可为恢复型滨海湿地的碳动态监测提供实用且可推广的工具，为适应性管理策略制定提供支撑，进而助力蓝碳项目的长期评估。