空间分析及其对杉木遗传试验效率的影响  

Spatial analysis and its effects on efficiency of genetic trial in Chinese fir

作　　者：边黎明[1] 郑仁华[2] 肖晖[2] 甘振栋 苏顺德[2] 施季森[1] 

机构地区：[1]南京林业大学南方现代林业协同创新中心,江苏南京210037 [2]福建省林业科学研究院,福建福州350012 [3]福建省邵武卫闽国有林场,福建邵武354006

出　　处：《南京林业大学学报（自然科学版）》2015年第5期39-44,共6页Journal of Nanjing Forestry University：Natural Sciences Edition

基　　金：福建省林木种苗科技攻关三期;四期项目(闽林科[2009]4号和闽林科[2013]1号);江苏高校优势学科建设工程资助项目(PAPD)

摘　　要：以位于福建卫闽国有林场10年生杉木80个半同胞家系的子代遗传测定林为材料,以ASReml程序进行林木遗传试验的空间分析,并分别利用试验设计模型和试验设计与空间分析整合模型对胸径生长进行遗传分析,检验空间分析对林木遗传试验效率的影响。模型选择结果表明,试验设计与空间分析整合模型中,空间效应完全取代了试验设计效应,模型可简化为空间分析模型。与试验设计模型相比,空间分析模型的AIC(赤池最小信息量准则)减少了129.2,模型的拟合程度更好;同时,加性遗传方差增加了16.5%,残差方差降低了11.6%,单株狭义遗传力增加了31.3%;行和列上的空间自相关参数分别为0.89和0.96,试验地点有明显的整体变异趋势。DBH at the age of 10 years in a progeny test of Chinese fir with 80 families from Fujian Weimin Forest Farm was examined by both experimental design model and combined design with spatial model,through which spatial analysis methods using ASReml was introduced,and efficiency of genetic trial was improved. All design effects was found replaced by spatial effects in combined model. AIC was reduced by 129. 2 in spatial model,which was more effectively modelled the spatial variation than the design model,and greatly increased additive genetic variance by 16. 5%,reduced residual variance by 11. 6%,and improved individual tree narrow sense heritabilities by 31. 3% correspondingly. Autocorrelations in row and column direction with 0. 89 and 0. 96 indicated that there are strong global trend in this site.

关 键 词：杉木 遗传试验效率 空间分析 空间变异 不完全区组 

分 类 号：S711[农业科学—林学] S722