旨在研究沿坝地区华北落叶松胸径-树高的生长状况,以北沟林场不同林龄的华北落叶松纯林为研究对象,采用实地测量和解析的方法获取华北落叶松的胸径、树高实测数据,利用SPSS分别对不同模型进行拟合.经各项指标检验,初步筛选拟合精度较好的曲线模型H=1.929D0.734(R2=0.939)和H=1.462+1.025D-0.012D2(R2=0.927)。将检验样木代回两个模型回归检验,进行树高的残差分析,其散点分布均匀,证明了两个模型的可靠性。运用各项误差分析指标判断,最终确定华北落叶松在本地域最佳胸径-树高模型为H=1.929D0.734。模型的建立为树高的测量提供了捷径,有利于森林资源的清查工作。
This article is aimed at studying DHB-tree height growth conditions of North China larch. The pure larch stands of different ages in Beigou Forest Farm were taken the object of study.The data of DHB and tree height were obtained through the ways of measurement and stem analysis and SPSS was used to fit different models. After the indicator test and,preliminary screening,models of the better fiting precision are H=1.929D0.734 (R2=0.939)and H=1.462+1.025D-0.012D2(R2=0.927).Using hights of standard tree regression test,the result showed the two models were reliable.According to the different indexes of error analysis the best model of North China larch is H=1.929D0.734.The establishment of the model provides a shortcut for tree height measurement .It is conducive to the forest resources investigation.
[1]孟宪宇.测树学 [M].2 版.北京:中国林业出版社,1996:66-95.
[2]赵敏,丁慧勇.上海水杉树高与胸径关系模式分析[J].上海师范大学学报:自然科学版,2009,10:531-535.
[3]戴继先,李信.林木“平均高径比”一元材积式的研制[J].山西林业科技,1994,9(3):48-50.
[4]冯仲科,隋宏大,邓向瑞,等.三角高程法树高测量与精度分析[J].北京林业大学学报,2007(S2):37-41.
[5]孙圆,王万江.江苏省杨树树高曲线模型的研制[J].应用研究,2005(5):34-37,
[6]刘云伟,冯仲科,邓向瑞,等.同一铅垂面两次设站法树高测量及其精度分析[J].北京林业大学报,2007(S2):63-66.
[7]侯红亚,王立海.目测法估测树高的误差分析[J].森林工程,2012,3(28):6-8.
[8]王成良.如何精确测定树高[J].林业勘查设计,1994(4):37-38.
[9]洪楠 .SPSS for Windows 统计分析教程 [M].北京 :电子工业出版社,2000.
[10]唐守正,郎奎建,李海奎 .统计和生物数学模型计算 [M].北京 :科学出版社,2008:291-293.
[11]钟义山,回归模型优劣评价的几个问题[J].中南林业调查规划,1992,11(4):10-14.
[12]刘璋温,吴国富.选择回归模型的几个准则[J].数学的实践与认识,1983,13(1):61-69.
[13]骆期邦,曾伟生,贺东北.林业数表模型—理论、方法与实践[M ].长沙:湖南科学出版社,2001.