基于不同数据源的福建省沿海防护林碳储量对比研究

Comparative Study of Carbon Stocks in Fujian Coastal Protection Forests Based on Different Data Sources

中文关键词:  沿海防护林  碳储量  生物量模型  样地数据  图斑数据  遥感数据

英文关键词:coastal protection forest  carbon stocks  biomass model  sample plot data  patch data  remote sensing data

中文摘要:

      在全球气候变化背景下，准确估算森林碳储量对评估沿海防护林固碳能力及制定适应性管理策略至关重要。本研究围绕福建省沿海防护林体系建设工程实施区域，利用森林资源清查样地、二类调查图斑以及遥感数据，采用生物量法，结合平均含碳系数法，估算了2014和2022年的森林植被碳密度和碳储量及其变化，并分析了不同数据源的估算差异。研究结果表明：（1）不同数据源估算结果存在差异，其中2014年的碳密度表现为图斑（31.29 Mg/hm2）<样地（33.34 Mg/hm2）<遥感（50.48 Mg/hm2），以样地数据为参考值，图斑偏低6.64%，而遥感数据偏高51.41%，差异可能源于图斑数据蓄积量测量误差以及遥感生物量数据准确性不足；2014和2022年的碳储量表现为图斑数据（35.33和44.93 Tg）低于遥感（60.53和65.84 Tg），但碳汇量表现为图斑（9.60 Tg）高于遥感（5.31 Tg），遥感数据源下生物量密度的误差和森林面积识别偏差是主要影响因素；（2）从不同森林类型来看，常绿阔叶林和常绿针叶林的面积、碳密度、碳储量均最高，是福建省森林碳汇的主要贡献者；遥感估算的大部分类型森林的碳密度都高于图斑估算结果，再次反映遥感生物量数据可能存在偏差；图斑和遥感数据估算的不同类型森林的碳储量也存在显著差异，这部分差距除了来自上述遥感估算的较高碳密度，还再次印证碳储量的估算差异与不同森林类型的面积识别有关。建议未来建立多源数据融合机制，优化碳汇监测体系，并通过差异化林分管理提升密闭林比例以增强碳汇功能。

英文摘要:

      Under the background of global climate change, accurately estimating forest carbon stocks is crucial for assessing the carbon sequestration capacity of coastal protection forests and formulating adaptive management strategies. This study focused on the implementation areas of the coastal protection forest system construction program in Fujian Province. Utilizing forest resource inventory sample plots, second-class survey patches, and remote sensing data, we estimated the forest vegetation carbon density and carbon stocks in 2014 and 2022, as well as their changes, by employing biomass model methods combined with average carbon coefficient methods. The differences in estimation results from different data sources were systematically analyzed. Our results show that: (1) differences were observed in the estimation results from different data sources. Carbon density showed a trend of patch data (31.29 Mg/hm2) < sample plot data (33.34 Mg/hm2) < remote sensing data (50.48 Mg/hm2). Using sample plot data as the reference value, patch data underestimated carbon density by 6.64%, while remote sensing overestimated it by 51.41%, likely due to measurement errors in patch-based volume estimation and uncertainties in remote sensing biomass data, Carbon stocks estimated from patch data (35.33–44.93 Tg) were lower than those from remote sensing data (60.53-65.84 Tg). However, the carbon sequestration from patch data (9.60 Tg) was higher than that of remote sensing data (5.31 Tg), primarily influenced by forest area recognition biases. (2) Among different forest types, evergreen broadleaf forests and evergreen coniferous forests had the highest area, carbon density, and carbon stocks, making them the primary contributors to forest carbon sequestration in Fujian Province. Remote sensing estimates of carbon density were generally higher than those from patch data across most forest types, further reflecting potential biases in remote sensing biomass data. In addition, significant differences in carbon stocks among different forest types were observed between patch and remote sensing data, which not only result from the higher carbon density estimates from remote sensing, but also highlight the impact of area recognition discrepancies among different forest types. We recommend to establish a multi-source data fusion mechanism in the future, optimize the carbon sequestration monitoring system, and enhance carbon sequestration capacity by increasing the proportion of closed forests through differentiated forest management.