北京山区防护林优势树种分布与环境的关系

Relationships between dominant arbor species distribution and environmental factors of shelter forests in the Beijing mountain area

在北京山区,以该区防护林主要树种为依据,选择在典型天然林中设置26块具有代表性的样地,并计测样地中乔木树种的重要值以及包括海拔、坡度、土壤含水量、土壤养分等14个环境因子,采用除趋势典范对应分析(DCCA)方法对样地乔木和环境因子进行分析,以揭示该区防护林优势树种与环境因子间的关系。根据样地树种组成及树种重要值首先将26块样地划分为以油松(Pinus tabuliformis)、侧柏(Platycladus orientalis)、栎类(Quercus spp.)、杨桦(Populus davidiana-Betula platyphylla)及落叶松(Larix principis-rupprechtii)为优势种的5个典型群落类型,然后应用DCCA方法进行分析,结果表明,在各个环境因子中,海拔作为大多数环境因子的综合反映,是影响植被群落分布最重要的环境因子,第一排序轴主要反映海拔、水分、土壤有机质等的变化,第二排序轴主要反映坡度、坡向的变化,各植被群落类型沿着第一排序轴呈有规律的分布;将环境因子对物种分布格局的影响进行定量分离,结果显示,14个环境因子解释了43.19%的物种分布格局,其中地形因子解释部分为20.36%,土壤因子解释部分为11.9%,土壤和地形的交互作用解释部分为10.93%。

Shelter forests of the Beijing mountain area are a major ecological barrier of Beijing.Forest cover protection and construction,and health and stability of forest ecosystems are critically important to the ecological security of Beijing.This study aims to provide a scientific basis for the restoration of degraded vegetation,protection of natural vegetation,and construction and management of shelter forests.Based on the dominant species of the shelter forest in the Beijing mountain area,26 representative natural forest plots were established across 13 natural reserves or forest management areas in 7 counties of the Beijing mountain area.To reveal the distribution pattern of arbor species in the area and explore how environmental factors influence the distribution pattern,Detrended Canonical Correspondence Analysis（DCCA） was used to analyze the relationships between the distribution of dominant arbors and 14 environmental factors.Environmental factors tested included soil infiltration rates（IR）,pH value（pH）,alkali-hydrolyzable nitrogen（AN）,available phosphorus（AP）,available potassium（AK）,organic matter（OM）,total nitrogen（TN）,total phosphorus（TP）,total potassium（TK）,elevation（ELE）,slope（SLO）,aspect（ASP）,thickness of soil（TOS）,and soil water content（SWC）.Twenty six sample plots were divided into 5 communities： Pinus tabulaeformis（I）,Platycladus orientalis（Ⅱ）,Quercus（Ⅲ）,Populus-Betula（IV） and Larix principis-rupprechtii（V） according to the dominant arbor species in the area and their importance values.Analysis of environmental factors revealed that elevation was significantly positively correlated with OM（0.6）,AN（0.683）,SWC（0.598）（P0.001）,TN（0.556）（P0.01）,TOS（0.477） and AK（0.419）（P0.05）,and negatively correlated with TK（-0.423）（P0.05）.Results of DCCA ordination showed that elevation was dominant among the 14 environmental factors that affected community distribution.The first DCCA axis accounted for the largest fraction of the variation and showed gradients of elevation,soil water content and organic matter.Along the first axis,community distribution ranged from broad-leaved forest to coniferous forest.The second DCCA axis showed variation due to slope and aspect.On the whole,the five types of communities coincided with the distribution of DCCA ordination diagram demonstrating the importance of environmental gradients in shaping community distribution.From the left of the first DCCA axis community types were Ⅳ,Ⅴ,Ⅰ,Ⅱ,whereas community type Ⅲ was widely distributed along the elevation gradient.The DCCA ordination of main arbor species with importance values ≥15 displayed two distribution patterns.First,an arbor species was dominant or subdominant in a community and rarely emerged in other communities,the arbor species similarly distributed in the community.Second,an arbor species was dominant or subdominant in many communities,the arbor species distributed among these communities.Furthermore,the influence of environmental factors was quantitatively assessed.Fourteen environmental factors were divided into topographic and soil factors,and the effects of environmental factors on community distribution were quantified.The results indicated that together,the tested factors could explain 43.19% of the distribution.Within that,topographic factors accounted for 20.36% of the variation,soil factors for 11.9%,and topographic-soil-factor interaction for 10.39%.Other,undetermined,factors accounted for an additional 56.81% of the variation.