Abstract: Species abundance distribution(SAD)combines species richness with species abundance in a community and is an important indicator of community structure. Although many studies have applied mathematical models to fit the SADs of plant communities, there are still few studies on the changes in the shape of SAD(i.e., the skewness of the SAD curves)along the environmental gradient. Especially for forest communities, it is still unclear whether the model fitting and shape change of SADs are consistent in different vegetation types. Here, 28 plots of 20 m × 20 m were set up in Hynobius amjiensis National Nature Reserve according to the main forest vegetation types. The species composition and abundance of each species were surveyed in each plot. The logseries model and the lognormal model were fitted to the SAD in each plot, and then the best-fit model was selected based on the corrected Akaike's Information Criteria(AICc). The α value in Gambin model and the η value in Weibull model were calculated to reflect the shape of SAD in each plot. The λ value in Weibull model was calculated to reflect the scale of SAD(i.e., the range observed in abundance). The relationship between altitude and the shape and scale of SAD was analyzed. The results were as follows:(1)The logseries model fitted SAD better than the lognormal model.(2)When all plots were included, there was no significant correlation between SAD shape(α and η)and altitude, but there was a significant positive correlation between the λ value and altitude.(3)There was a negative correlation between altitude and α and η values in the mixed evergreen and deciduous broad-leaved forests, and there was a significant positive correlation between the α value and altitude in the deciduous broad-leaved forests, while there was no significant correlations between both α value and η value and altitude. The results indicate that the model fitting and shape change of SADs along the environmental gradient are related to the vegetation types of the forest community. Therefore, it is necessary to consider the vegetation types when analyzing the relationship between SAD and impact factors in plant communities.

Key words: subtropical forests, community structure, species abundance distribution, models, vegetation types, Longwangshan

摘要： 虽然大量研究已利用模型拟合的方法对植物群落的物种多度分布(SAD)进行了不同数学模型拟合,但对SAD形状(曲线的偏斜度)如何在环境梯度上连续变化的研究仍然不足; 尤其是森林群落,同一地区不同植被类型群落SAD的模型拟合和形状变化是否一致,仍无明确定论。该研究针对安吉小鲵国家级自然保护区中分布的主要森林植被类型,采用样方调查法,记录了28个20 m × 20 m样方中的物种组成及其个体多度。通过对数级数和对数正态模型对样方中的SAD曲线进行拟合,选择最优模型,并通过Gambin模型中的α值和Weibull模型中的η值反映SAD的形状,以及Weibull模型中λ值反映SAD的变化尺度(物种间个体多度的差异程度),分析海拔高度与SAD的形状和变化尺度之间的关系。结果表明:(1)该地区的森林群落物种多度分布主要符合对数级数模型。(2)当包含所有样方时,α值和η值与海拔高度无显著相关性,λ值与海拔呈显著正相关。(3)针对不同的植被类型,常绿与落叶阔叶混交林中α值和η值与海拔高度呈负相关,但在落叶阔叶林中λ值与海拔高度之间具有正相关关系,而α值和η值与海拔高度之间均无显著相关性。该研究结果表明,不同植被类型SAD的形状变化与海拔之间的关系存在差异,说明海拔对不同植被类型中各物种多度分布具有不同影响。因此,在关于植物群落的物种多度分布及其与影响因子关系的研究中,需要考虑区分不同的植被类型。

关键词: 亚热带森林, 群落结构, 物种多度分布, 模型, 植被类型, 龙王山