摘要
物种多样性是研究森林群落结构、组成和空间特征的重要指标。本研究通过样地植被调查,采用物种多样性分析和相关分析方法,研究了影响东大河林区青海云杉林(Picea crassifolia)物种多样性的因素。结果表明,影响青海云杉林物种多样性的因素有6个,其重要程度依次为坡向】苔藓盖度】土壤容重=土壤孔隙度】群落类型】郁闭度;群落特征直接影响着青海云杉林物种多样性,坡向通过影响林内环境、土壤类型及其理化性状,间接影响群落的物种多样性;群落类型苔藓―青海云杉林和较高的郁闭度、苔藓盖度、土壤孔隙度不利于增加青海云杉林的物种多样性;西坡土壤容重为0.73~0.77 g·cm-3、土壤孔隙度为68%~70%、郁闭度为0.6~0.7、苔藓盖度为14%~40%时,青海云杉林有较高的物种多样性;在不同森林群落类型中,草类―青海云杉林物种多样性高于灌木―青海云杉林和苔藓―青海云杉林。本研究建立了物种多样性与主要影响指标的多元回归模型,为青海云杉林物种多样性的初步计算提供了依据。
Species diversity is an important index of forest community structure,composition and spatial feature.In the present study,the influence factors of species diversity of Picea crassifolia forest in Dongdahe forest area were analyzed with vegetation survey,species diversity analysis and correlation analysis.There were total 6 factors influence species diversity of P.crassifolia forest and their importance decreased in the following order:aspect,bryophyte coverage,soil bulk density and soil porosity,community types,canopy coverage.These factors had different ways to influence species diversity,for examples,the community characteristic directly affected the species diversity whereas the aspect indirectly affected species diversity by affecting the forest environment,soil type and characters.Bryophyte-P.crassifolia forest,higher canopy coverage,bryophyte coverage,soil porosity were not favor to increase species diversity.P.crassifolia forest had higher species diversity with 0.73 ~ 0.77 g · cm-3soil bulk density,68% ~ 70% soil porosity,0.6 ~ 0.7 canopy coverage and 14% ~ 40% bryophyte coverage.For the different forest community types,species diversity of herb-P.creassifolia forest was higher than shrub-P.creassifolia forest and bryophyte-P.creassifolia forest.The multivariable regression model has been established for preliminary calculation of species diversity and the influence factor.
出处
《草业科学》
CAS
CSCD
北大核心
2015年第1期28-35,共8页
Pratacultural Science
基金
甘肃省金昌市星火计划(2009008)
关键词
东大河林区
青海云杉林
α-多样性
物种丰富度
多元回归模型
Dongdahe forest area
Picea crassifolia forest
α-diversity
species richness
multivariable regression model
