林业科学2026，Vol.62Issue(6)：236-248,13.DOI:10.11707/j.1001-7488.LYKX20250770
从经验模型到智能估算:东北森林碳储量遥感监测技术革新路径
From Empirical Models to Intelligent Estimation:An Innovation Pathway of Remote Sensing Technology in Monitoring Forest Carbon Stocks in Northeast China
摘要
Abstract
Forest carbon storage plays a crucial role in regulating the global carbon cycle and mitigating climate change.Accurate estimation and dynamic monitoring of forest carbon stocks constitute a fundamental basis for advancing the understanding of terrestrial ecosystem carbon cycling processes and for achieving national carbon neutrality and carbon peaking goals.In recent years,development in remote sensing technology has substantially enhanced the capability for large-scale estimation of forest carbon storage.However,accurate carbon stock estimation in northeast China still faces challenging due to the region's high spatial heterogeneities in forest structure,species composition,origin,and age.This study systematically reviews research progress in forest carbon monitoring in northeast China based on 233 Chinese and English literature,and integrates key institutional milestones in global climate governance,including the establishment of the REDD+mechanism and the implementation guidelines of the Paris Agreement,together with major technological advances such as the emergence of spaceborne LiDAR-enabled global forest structural observations and the widespread adoption of cloud-computing platforms.The development process is divided into three stages:coarse-resolution empirical estimation(pre-2010),medium-resolution mechanistic modeling(2011-2018),and high-resolution intelligent estimation(since 2019).Based on the characteristics of data sources,modeling approaches,and major achievements at each stage,a comprehensive analysis was conducted on the key challenges in current research and future development directions.Literature analysis indicated that forest carbon stock monitoring in northeast China has evolved from sole reliance on coarse resolution optical remote sensing data to the deep integration of high-resolution multimodal,multi-temporal,and multi-scale remote sensing datasets.Concurrently,estimation methods have transitioned from empirical statistical models to process-based models and,more recently,to data-driven deep learning approaches.These advances have significantly improved the accuracy and spatiotemporal representativeness of forest carbon stock estimates,providing essential scientific evidence and technical support for assessing forest carbon sink capacity,strengthening the shields for ecological security,and promoting high-quality forestry development in northeast China.关键词
森林碳储量/多源遥感/人工智能/激光雷达/随机森林Key words
forest carbon storage/multi-source remote sensing/artificial-intelligence/light detection and ranging(LiDAR)/random forest(RF)分类
农业科技引用本文复制引用
全迎,邵国凡,李明泽..从经验模型到智能估算:东北森林碳储量遥感监测技术革新路径[J].林业科学,2026,62(6):236-248,13.基金项目
国家自然科学基金项目(32401568) （32401568）
国家重点研发计划课题(2023YFD2201704) （2023YFD2201704）
中央高校基本科研业务费专项资金(2572025DR01) （2572025DR01）
中国博士后科学基金(2024M760385) （2024M760385）
黑龙江省博士后资助经费(LBH-Z24051). （LBH-Z24051）
