文献类型：期刊文献

中文题名：基于集成学习和机载激光雷达的普文林场地上生物量遥感估测

英文题名：Remote sensing estimation of above ground biomass in Puwen Forest Farm based on ensemble learning and airborne LiDAR

作者：徐海龙[1,2] 杜黎明[2] 庞勇[2] 何云玲[1] 魏巍[3]

第一作者：徐海龙

机构：[1]云南大学国际河流与生态安全研究院,云南昆明650500;[2]中国林业科学研究院资源信息研究所国家林业和草原局林业遥感与信息技术重点实验室,北京100091;[3]云南省林业和草原科学院,云南昆明650204

年份：2026

卷号：48

期号：2

起止页码：297-305

中文期刊名：云南大学学报(自然科学版)

外文期刊名：Journal of Yunnan University(Natural Sciences Edition)

收录：;北大核心:【北大核心2023】；

基金：国家重点研发计划(2024YFD2201405-1);中央级公益性科研院所基本科研业务费专项资金(CAFYBB2024MA037);国家自然科学基金(42401491)。

语种：中文

中文关键词：激光雷达;地上生物量;普文林场;遥感估测;机器学习;集成学习

外文关键词：LiDAR;above ground biomass;Puwen Forest Farm;remote sonsing estimation;machine learning;ensemble learning

分类号：S757.2;S771.8

摘要：森林地上生物量是全球碳循环与气候变化研究中的关键变量,准确估算森林生物量对于碳排放量化和环境监测具有重要意义.机载激光雷达(LiDAR)提供了有效的手段来估测区域尺度的森林生物量.以普文林场为研究区,基于森林调查样地数据与机载激光雷达数据,通过支持向量机回归(SVR),包括随机森林(RF)、K近邻算法(KNN)和集成学习算法(Stacking)相结合,分析不同模型的估测精度.根据最优估测模型来反演研究区森林地上生物量并绘制空间分布图,结果表明:(1)高度特征在所有4种变量中对生物量估算的贡献最大,其次是强度特征和冠层特征;(2)集成学习算法Stacking在回归拟合过程中表现出了最优的预测性能,R2值均超过0.71,且在不同森林类型的估测精度上均表现优异;(3)机器学习算法对针叶林、阔叶林及针阔叶混交林3种不同森林类型生物量建模估测结果表明,针叶林优于阔叶林,阔叶林优于针阔混交林;(4)基于最优拟合模型,普文林场的森林地上生物量的平均值为213.3 t·hm^(-2),与林场实测生物量209.7 t·hm^(-2)具有较高的一致性.
Forest above ground biomass is a key variable in the study of global carbon cycle and climate change,and accurate estimation of forest biomass is of great significance for carbon emission quantification and environmental monitoring.Airborne LiDAR provides an effective means to estimate forest biomass at regional scales.Based on the forest survey plot data and airborne LiDAR data,the estimation accuracy of different models was analyzed by the combination of Support Vector Regression(SVR),including random forest(RF),K-nearest neighbor algorithm(KNN)and ensemble learning(Stacking).According to the optimal estimation model,the above ground biomass of the forest in the study area was inverted and the spatial distribution map was plotted,and the results showed that:①Height characteristics contributed the most to the estimation of biomass among all four variables,followed by intensity characteristics and canopy characteristics;②The stacking ensemble learning algorithm showed the best prediction performance in the regression fitting process,with R2 values exceeding 0.71,and performed well in the estimation accuracy of different forest types;③The results of machine learning algorithm modeling and estimating the biomass of three different forest types:coniferous forest,broad-leaved forest and mixed coniferous and broad-leaved forest,showed that the coniferous forest was better than the broadleaved forest,and the broad-leaved forest was better than the mixed coniferous and broad-leaved forest.④Based on the optimal fitting model,the average value of forest above ground biomass in Puwen Forest Farm was 213.3 t·hm^(-2),and has a high consistency with the measured biomass of the forest farm(209.7 t·hm^(-2)).