中文摘要：

采用自组织特征映射网络（SOM）对芦芽山自然保护区青杄林进行数量分类和排序.结果表明：SOM将60个森林样地划分为8个群落类型,各类型在排序图上都有其分布范围和界限,且群落结构、物种组成差异明显,说明分类和排序结果合理,可较好地揭示群落间的生态关系;通过环境因子梯度的可视化方法,确定海拔、坡位和坡度是影响该地区青杄林生长和分布的主要因子,揭示群落、物种及植被分布和环境因子的关系.结果表明SOM网络适用于表征群落生态特征及探索群落和环境相互关系的研究,将其应用于自然保护方面,将有利于对不同植被类型进行分类和管理.

英文摘要：

The self-organizing feature map （SOM） was applied to the classification and ordination of Picea wilsonii forests in the Luya Mountain Nature Reserve.It was shown that 60 quadrats were classified into 8 groups,representing 8 types of plant communities.These communities had obvious differences in species composition and structure.Each type had its boundary and distributional range on the ordination map.The SOM clustering showed significant ecological meanings and revealed the relationships among communities.Additionally,by using a visualizing method of environmental factor gradient on topology map,SOM indicated that elevation,slope position and slope were the main environmental factors affecting the distribution of P.wilsonii forests.The SOM ordination clearly showed ecological gradients and revealed the relationships among communities and environmental factors.This study suggests that SOM is a very effective quantitative technique in plant ecology,and it is applicable to describe community characteristics and to analyze relationship between communities and environmental variables.