种-面积关系是生态学中的基本问题, 其构建方式对种-面积关系的影响以及最优种-面积模型的选择仍然存在争议。该文利用吉林蛟河42 hm²针阔混交林样地数据, 分别采用巢式样方法和随机样方法建立对数模型、幂函数模型和逻辑斯蒂克模型, 并通过赤池信息量准则(AIC)检验种-面积模型优度。结果表明, 种-面积关系受到取样方法的影响, 随机样方法的拟合效果优于巢式样方法。采用随机样方法构建的幂指数模型(AIC = 89.11)和逻辑斯蒂克模型(AIC = 71.21)优于对数模型(AIC = 113.81)。根据AIC值可知, 随机样方法构建的逻辑斯蒂克模型是拟合42 hm²针阔混交林样地种-面积关系的最优模型。该研究表明: 在分析种-面积关系时不仅应考虑尺度效应, 还需注意生境变化及群落演替的影响。

Aims The Species Area Relationship (SAR) is a fundamental pattern in ecology. Recent analyses have often demonstrated substantial uncertainty in selecting the best SAR model for a data set. Our objective was to understand the effects of sample design on species-area relations, in order to suggest a more appropriate SAR model for the given plot data.
Methods A long-term 42 hm²forest plot was established in conifer and board-leaved mixed forest in Jiaohe, China. All trees with diameter at breast height (DBH) > 1 cm were tagged and their height, DBH and crown diameter recorded. We propose three SARs models (logarithmic function, power function and logistic function) to compare SARs constructed from nested design and random design. We use Akaike Information Criterion (AIC) to compare the quality fit of each SAR model given the data.
Important findings The way of constructing SARs influences the outcome. The random design showed significantly better goodness of fit of SARs model than the nested design. Among the three SAR models, Logistic function model from the random design was the best, suggesting it provided a reasonable description of the species-area relationship in the plot. This study demonstrates the significance of scale variance in species-area relationships; the effects of area on species richness are variable and can be scale dependent. However, because the species distribution patterns and spatial scale vary greatly, further work is needed to consider environmental effects and the community succession on different spatial scales.