摘要
为减轻森林资源调查的工作量,满足蓄积量调查的精度要求,利用广西276块杉木(Cunninghamia lanceolata)林标准地资料,选用理查德函数变型的固定参数式和可变参数式建立杉木林分相对树高曲线模型,综合考虑模型拟合决定系数(R^(2))、均方误差(MSE)、总相对误差(TRE)、平均相对误差(MRE)、平均百分比误差(MPE)和预估精度(P)等指标,选择表现较好的模型。结果表明,最佳的杉木林分相对树高曲线模型为理查德函数变型固定参数式模型。将利用模型估算的林分蓄积量与利用二元材积表估算的林分蓄积量形成成对数据,采用配对t检验法对二者均值差异进行检验;结果显示,二者无显著差异,精度达到99.46%。利用理查德函数变型固定参数式模型建立的杉木相对树高曲线方程预估精度高,可在广西区内应用,整体区域模型函数为Rhi=(1-e^(-1.97016))^(-1.55024)×(1-e^(-1.97016Rd_(i)))^(1.55024),中心产区模型函数为Rh_(i)=(1-e^(-1.95913))^(-1.56824)×(1-e^(-1.95913Rd_(i)))^(1.56824),一般产区模型函数为Rh_(i)=(1-e^(-2.10637))^(-1.63569)×(1-e^(-2.10637Rd_(i)))^(1.63569)。
In order to reduce workload of forest resource inventory and meet accuracy requirements of accumu-lation survey,based on data of 276 sample plots of Cunninghamia lanceolata stands in Guangxi,curve modelsof relative tree height for C.lanceolata stands were established by using fixed parameter formula and variableparameter formula of Richard function deformation.Model fitting determination coefficient(R^(2)),mean square er-ror(MSE),total relative error(TRE),mean relative error(MRE),mean percentage error(MPE)and predictionaccuracy(P)were considered,and better model was selected.Results showed that Richard function deformationfixed parameter model was the best curve model of relative tree height for C.lanceolata stands.Stand accumula-tion estimated by model and stand accumulation estimated by standard volume table were paired.By t-test,dif-ferences between means of two accumulations were tested.Results showed that there was no significant differ-ence between means of two accumulations,and accuracy reached 99.46%.Curve equation of relative tree heightfor C.lanceolata stands by using Richard function deformation fixed parameter model with high estimated accu-racy could be applied in Guangxi.Model function for whole region was Rh_(i)=(1-e^(-1.97016))^(-1.55024)×(1-e^(-1.97016Rd_(i)))^(1.55024).Model function for central region was Rh_(i)=(1-e^(-1.95913))^(-1.56824)×(1-e^(-1.95913Rd_(i)))^(1.56824).Model function for general region was Rhi=(1-e^(-2.10637))^(-1.63569)×(1-e^(-2.10637Rd_(i)))^(1.63569).
作者
曾春阳
糜新宇
张伟
徐庆玲
刘峰
岑巨延
王科
黄孝发
Zeng Chunyang;Mi Xinyu;Zhang Wei;Xu Qingling;Liu Feng;Cen Juyan;Wang Ke;Huang Xiaofa(Guangxi Forest Inventory&Planning Institute,Nanning,Guangxi 530011,China;Guangxi State-owned Gaofeng Forest Farm,Nanning,Guangxi 530025,China)
出处
《广西林业科学》
2023年第1期72-77,共6页
Guangxi Forestry Science
基金
广西林业科研项目(桂林科研[2015]第28号,桂林科研[2016]第24号,桂林科字[2022]第11号)。
关键词
相对树高曲线模型
杉木
T检验
立木材积
curve model of relative tree height
Cunninghamia lanceolata
t-test
standing volume
