摘要
以黑龙江省七台河市林业局金沙林场9株人工落叶松432个样品密度数据为例,利用逐步回归技术构建落叶松木材密度模型:WD=β1+β2RN+β3RN2+β4h。利用S-PLUS软件中的LME过程,分别考虑单水平和多水平效应,拟合线性木材密度混合效应模型。结果表明:基于单水平和多水平效应的混合模型拟合精度高于传统的基本模型,并且考虑单水平树高效应和2层次效应时的混合模型精度高于考虑单水平样木效应影响的混合模型。模型检验结果表明:混合效应模型不但能反映总体平均木材密度变化趋势,还能反映分组之间的差异。
In this study, the sample data was based on 432 samples of 9 trees from dahurian larch（Larix gmelinii）plantations located in Qitaihe Forest Bureau in Heilongjiang Province. The stepwise regression techniques were used to develop wood density model： WD=β1＋β2RN＋β3RN2＋β4h. Then, the developed model was fitted using single level and multilevel linear mixed-effects modeling approach based on LME procedure of S-PLUS software. The mixed effects models showed better model fitting results than basic model whatever considering single level and multilevel linear mixed effects. Moreover, the mixed effects model considering height effects and both effects showed more precision than that considering individual tree effects. Model test indicated that mixed effects models not only showed the mean trends of wood density, but also showed the variations among groups.
出处
《林业科学》
EI
CAS
CSCD
北大核心
2013年第7期91-98,共8页
Scientia Silvae Sinicae
基金
中央高校基本科研业务费专项资助(DL12DA01)
国家"十二五"科技支撑课题(2012BAD22B0202)
国家自然科学基金项目(31170591)
关键词
木材密度
两水平线性混合模型
落叶松
wood density
two-level linear mixed model
dahurian larch （Larix gmelinii）
