广西桉树一元、二元林分形高表研建  

Development of Univariate and Bivariate Form Height Tables for Eucalyptus Plantations in Guangxi

作　　者：刘峰 徐庆玲 覃孙慧 莫焕雄 黎庆华 韦秋生 韩斐扬 LIU Feng;XU Qingling;QIN Sunhui;MO Huanxiong;LI Qinghua;WEI Qiusheng;HAN Feiyang(Guangxi Zhuang Autonomous Region Forest Inventory&Planning Institute,Nanning 530011,Guangxi,China;Guangxi Zhuang Autonomous Region Luzhai Forestry Bureau,Luzhai 545600,Guangxi,China;Guangxi Zhuang Autonomous Region Changzhou Forestry Bureau,Wuzhou 543003,Guangxi,China;Guangxi Zhuang Autonomous Region Liujiang Natural Resources Bureau,Liuzhou 545100,Guangxi,China)

机构地区：[1]广西壮族自治区林业勘测设计院,广西南宁530011 [2]广西壮族自治区鹿寨县林业局,广西鹿寨545600 [3]广西壮族自治区梧州市长洲区林业局,广西梧州543003 [4]广西壮族自治区柳州市柳江区自然资源局,广西柳州545100

出　　处：《桉树科技》2024年第3期27-33,共7页Eucalypt Science & Technology

基　　金：广西林业科研项目(桂林科研[2016]31号,桂林科研[2016]24号,桂林科研[2015]15号)。

摘　　要：为运用形高表准确测算桉树林分蓄积量,以期为采伐作业设计和森林资源规划设计调查等工作提供理论支撑。基于234块样地建模数据中林分平均胸径、平均树高和形高值数学关系,分别拟合得到一元形高模型{Hf=129.99766×[(1‒exp(‒0.0037249×H)]^(0.978321)}以及二元形高模型(Hf=0.577152×H^(1.20331)×D^(‒0.28581))。研究结果显示一元形高模型总相对误差TRE、平均系统误差ASE在±2%范围内,平均百分标准误差MPSE均小于4%,预估精度P值达98%以上;二元形高模型总相对误差TRE、平均系统误差ASE在±2%范围内,平均百分标准误差MPSE均小于3%,预估精度P值超过了99.5%。两个形高模型各项指标均达到了林分形高表编制技术规程的要求,二元形高模型的各项指标整体优于一元形高模型。To accurately calculate timber volume in Eucalyptus plantations using stand form height tables with the aim of providing theoretical support for harvesting operations and forest resource planning and design.This study developed both univariate and bivariate form height models based on the mathematical relationships between average diameter at breast height(DBH),average tree height,and form height values from 234 sample plots.The univariate form height model was expressed as{Hf=129.99766×[(1‒exp(‒0.0037249×H)]^(0.978321)},while the bivariate form height model was given by(Hf=0.577152×H^(1.20331)×D^(‒0.28581)).The results indicated that for various Eucalyptus species,production areas,and ranges of diameter and height,the univariate form height model achieved a total relative error(TRE)and average systematic error(ASE)within±2%,an mean percentage standard error(MPSE)below 4%,and a prediction accuracy(P value)exceeding 98%.Similarly,the bivariate height model had a TRE and ASE within±2%,an MPSE under 3%,and a prediction accuracy surpassing 99.5%.Both models met the technical requirements for compiling height tables,with the bivariate model demonstrating overall superior performance compared to the univariate model.

关 键 词：桉树 形高表 评价标准 

分 类 号：S758.51[农业科学—森林经理学]