【作者】 闵志强；

【导师】 孙玉军；

【作者基本信息】 北京林业大学 ， 森林经理学， 2010， 硕士

【摘要】 本文在东折棱河林场以长白落叶松林地上部分生物量为研究对象,利用样木收获法收集34个样地的林分地上部分生物量信息,选取其中29个样地生物量信息分别与样地林分因子信息和TM遥感影像信息拟合建立林分因子线性模型、遥感因子线性模型和林分因子非线性模型,利用剩下的5个样地进行模型精度检验和误差分析。结果表明：长白落叶松地上部分生物量均可以用遥感因子或林分因子模型拟合；林分因子线性模型对中幼林地上生物量的估测精度高(P林分=94.33%>P遥感=P非=92.32%)且检验误差小(MRE林分=6%<MRE非=7%<MRE遥感=31%),效果较好；若只考虑中龄林,则非线性模型与林分因子模型估测效果相当(误差之和E林分=366.7 t<E非=379.0 t<E遥感=515.4 t)。树皮、干材和总生物量部分的估测效果为林分因子模型估测最优、其次为非线性模型、最差为遥感因子模型,树冠部分则是遥感模型估测最好、其次为林分因子模型、最差为非线性模型。在树冠和地上部分总生物量模型中,非线性模型的不相容性最不明显。三种模型对长白落叶松碳密度估测结果为：林分线性模型为37.34 t／ha,遥感线性模型为35.96 t／ha,非线性模型为37.05 t／ha。通过三种生物量模型和碳储量估测效果的比较分析,希望为今后生物量模型的选择和研究提供一定的参考。更多还原

【Abstract】 Using above ground biomass of Larix olgensis plantation as study objects in DoZheleng Reiver forest farm, stand above ground biomass information of 34 plots were collected in the way of sample tree harvest. Biomass information of 29 plots were selected to establish biomass model with stand factor and TM RS image information and nonlinear factor respectively, and the rest 5 plots were used for model checking and error analysis. The results showed that:above ground biomass of Larix olgensis can be fitted by remote-sensing factor or stand factor model; The stand factor linear model in the estimation of above ground biomass Young was high accuracy (P stands= 94.33%>PRS=PNL=92.32%).while the test error is small (MRE stand=6%<MRE non-=7% <MRE RS=31%) and better; And if we only consider the middle-age forest, the effect of the non-linear model and stand factor model were estimated a considerable (errors, E stand=366.7t<E non=379.0 t<ERS=515.4t). The stand factor model for the bark, trunk and whole biomass estimation is best, followed by non-linear model, the worst factor model for the remote sensing, remote sensing model of the crown part of the estimation is the best, followed by factor model for the stand, the worst non-linear model. The results of three models about the carbon density estimation Larix olgensis is:linear model stand was 37.34t/ha, remote sensing linear model was 35.96t/ha, nonlinear model was 37.05t/ha. The comparing analysis of three biomass models will provide some references for model selection and research of biomass later.更多还原