摘要
基于东北林区191个红松林(Pinus koraiensis)样地的机载激光雷达数据和地面实测数据,首先,通过多元线性回归和非线性回归估计方法,确定林分蓄积量及平均高、断面积的基础回归模型;然后,利用误差变量联立方程组方法,建立基于激光雷达变量的林分蓄积量与平均高、断面积的模型系统。结果显示:建立的多元线性、多元和二元非线性林分蓄积量回归模型,其确定系数R2分别为0.858,0.846和0.821,平均预估误差MPE分别为2.57%,2.66%和2.85%,平均百分标准误差MPSE分别为26.35%,16.35%和17.88%;利用模型系统对林分平均高、断面积和蓄积量进行估计,其R2分别为0.597,0.750和0.822,MPE分别为1.90%,2.52%和2.84%,MPSE分别为10.85%,15.28%和17.73%。结果表明:基于机载激光雷达数据估计林分蓄积量、平均高等主要森林参数,非线性模型优于线性模型,而且基于点云高度变量(中位数)和强度变量(75%分位数)的二元非线性模型就能达到比较理想的预估效果;误差变量联立方程组方法,是建立林分蓄积量与平均高、断面积回归模型系统的一种可行方法;所建立的东北红松林平均高、断面积和蓄积量联立模型,其预估精度达到森林资源调查相关技术规定要求,可以在实践中推广应用。
Based on the airborne laser scanning(ALS)data and field measurement data of 191 sample plots distributed across the Korean pine(Pinus koraiensis)forest stands in northeastern China,the base regression models for estimating forest volume,mean height and basal area were determined through multiple linear regression and nonlinear regression methods at first.Then,a model system including stand volume,mean height,and basal area based on the ALS data was developed,using the error-in-variable simultaneous equations approach.The results show that for the developed multiple linear,nonlinear and two-variable nonlinear stand volume regression models,the coefficients of determination(R2)are 0.858,0.846 and 0.821;mean prediction errors(MPEs)are 2.57%,2.66%and 2.85%;and mean percent standard errors(MPSEs)are 26.35%,16.35%and 17.88%,respectively.For the mean height,basal area and stand volume simultaneous models,the R2 are 0.597,0.750 and 0.822;the MPEs are 1.90%,2.52%and 2.84%;and the MPSEs are 10.85%,15.28%and 17.73%,respectively.For estimating the forest parameters such as stand volume based on ALS data,nonlinear model is better than linear model,and the two-variable nonlinear model based on height(the median)and intensity(the percentile 75)of point clouds performed well.The error-in-variable simultaneous equations approach is a feasible method for developing a model system for estimation of main forest parameters.The developed mean height,basal area and stand volume simultaneous models in this study meet the need of precision requirements to relevant regulations on forest resources inventory,indicating that the models can be applied in practice.
作者
曾伟生
孙乡楠
王六如
王威
蒲莹
ZENG Weisheng;SUN Xiangnan;WANG Liuru;WANG Wei;PU Ying(Academy of Forest and Grassland Inventory and Planning,National Forestry and Grassland Administration,Beijing 100714)
出处
《林业资源管理》
北大核心
2020年第2期79-86,共8页
Forest Resources Management
基金
国家自然科学基金项目(31770676)
中国国土勘测规划院招投标项目“主要树种航空林分材积表编制”(GXTC-A-19070081)。
关键词
机载激光雷达
森林蓄积量
非线性模型
误差变量
联立方程组
红松
airborne laser scanning
forest volume
nonlinear model
error-in-variable
simultaneous equations
Pinus koraiensis