巨尾桉人工林立木材积表的编制  

Compilation of Standing Timber Volume Table of Eucalyptus grandis×E.urophylla Plantation

作　　者：韦文长 高德祥[2] 刘世雄 刘明奎 WEI Wenchang;GAO Dexiang;LIU Shixiong;LIU Mingkui(Shuangjiang Bureau of Forestry and Grassland, Shuangjiang, Yunnan 677399, China;Yunnan Institute of Forest Inventory and Planning, Kunming 650051, China)

机构地区：[1]双江县林业和草原局,云南双江677399 [2]云南省林业调查规划院,云南昆明650051

出　　处：《林业调查规划》2022年第3期1-8,共8页Forest Inventory and Planning

基　　金：云南省临沧市科技创新人才培养项目(202004AC100001-B14).

摘　　要：根据国内现行立木材积表编制技术规程的相关要求,选择6~22 cm径阶巨尾桉人工林138株样木,按2 m(树高10 m以上)或1 m(树高10 m以下)区分段规范造材作为样木,以其中的103株样木作为建模样本,分别选择国内外有代表性的6个一元材积回归方程和9个二元材积方程,应用SPSS软件进行非线性回归分析,以Levenberg-Maquar迭代法对各方程进行拟合并求解各方程参数,以离差平方和、相关指数、总相对误差、相对误差平均值、相对误差绝对值平均值、预估精度和残差分布作为评价指标进行对比分析。结果表明,建模样本可选择的一元材积回归方程以中岛广吉式(V=a_(0)d^(a_(1))a_(2)^(d))为最佳,相关指数(R^(2))为0.95881,总相对误差(RS)为0.15488%;二元材积方程以迈耶(1949)式(V=a_(0)+a_(1)d+a_(2)d^(2)+a_(3)dh+a_(4)d^(2)h)为最佳,相关指数(R^(2))为0.99022,总相对误差(RS)为0.9926%,相对误差绝对值平均值(REAA)为7.97929%。以其余的35株样木作为检验样本,分别对中岛广吉式一元材积回归方程和迈耶(1949)式二元材积方程进行检验。中岛广吉式一元材积回归方程总相对误差(RS)为4.25673%,通过独立样本t检验,F<F_(0.05),表明模型适用;迈耶(1949)式二元材积方程总相对误差(RS)为0.32506%,相对误差绝对值平均值(REAA)为2.32762%,通过独立样本t检验,F<F_(0.05),表明模型适用。According to the current technical regulations for the compilation of standing timber volume table in China,138 sample trees of Eucalyptus grandis×E.urophylla plantation with a diameter of 6~22 cm were selected to make timber according to the 2 m(tree height above 10 m)or 1 m(tree height below 10 m)segmentation standard as the sample trees.103 sample trees were used as the modeling samples,and 6 representative unitary volume regression equations and 9 binary volume equations were selected for nonlinear regression analysis by SPSS software,and the Levenberg-Maquar least square method was used to fit the parameters of each equation and perform variance analysis.The sum of squared deviations,correlation index,total relative error,relative error average,relative error absolute value average,estimation accuracy and residual graph distribution were used as evaluation indexes for comparative analysis.The results showed the unitary volume regression equation that could be selected for modeling samples was the Nakajima Hiroyoshi formula(V=a_(0)d^(a_(1))a_(2)^(d))as the best,the correlation index(R^(2))was 0.95881,and the total relative error(RS)was 0.15488%;the binary volume equation based on Meyer(1949)formula(V=a_(0)+a_(1)d+a_(2)d^(2)+a_(3)dh+a_(4)d^(2)h)was the best,the correlation index(R^(2))was 0.99022,the total relative error(RS)was 0.9926%,and the average absolute value of the relative error(REAA)was 7.97929%.In addition,35 samples were used as test samples to test the Nakajima Hiroyoshi unitary volume regression equation and Meyer(1949)binary volume equation.The results showed that the total relative error of the Nakajima Hiroyoshi unitary volume regression equation(RS)was 4.25673%.Through independent sample t-test,F<F_(0.05),which showed that the model was applicable;Meyer(1949)binary volume equation total relative error(RS)was 0.32506%,the absolute value of the relative error average(REAA)was 2.32762%.Through independent sample t test,F<F_(0.05),which showed that the model was applicable.

关 键 词：立木材积表 编表模型 方程拟合 适用性检验 巨尾桉人工林 

分 类 号：S758.62[农业科学—森林经理学] S792.39[农业科学—林学]