Abstract: The variations of the Shannon diversity index and Patrick richness index with different sampling scale were analyzed to study the effect of special scale on species diversity in Inner Mongolia typical steppe using the transect line survey method. The study educed the fractal dimensions (D) of two α-diversity indexes using fractal theory combining with fractal methods. Results showed that Shannon diversity index first increased rapidly then decreased with a sampling scale increasing, while Patrick richness index held the line after reaching a peak value. Five square meters was an optimal scale to study α-diversity in this area. Two α-diversity indexes had significant correlation with sampling scale in bi-logarithm coordinates of five-meter sampling range (P<0.01), and the correlation coefficients (R) were more than 0.99, but there was no obvious linear regularity beyond five-meters. D value indicated that the species composition in the researched region was not complicated. Therefore, smaller single sampling squares or more sampling squares should be applied carefully to study community diversity. The sampling scale should be introduced basing on study purpose and the characteristics of diversity indices.

Key words: Typical steppe, Species diversity, Spatial scale, Fractal analysis

摘要： 为研究空间尺度对物种多样性的影响,以内蒙古典型草原为对象,采用样线调查法分析Shannon指数和Patrick 丰富度指数2种α多样性指数随取样尺度的变化特征,结合分形理论和方法,得出2种α多样性指数的分形维数D。结果表明:随着取样尺度的逐渐增加,Shannon指数呈先迅速增加后降低的规律,而Patrick 丰富度指数达到最大值后则保持不变;5 m²(1 m×5 m)是该区域α多样性研究的最佳尺度;在5 m范围内,2种α多样性指数与取样尺度在双对数坐标上均呈极显著线性相关(P<0.01),相关系数(R)都在0.99以上,5 m后未表现出明显的线性规律;D值表明该区域物种组成简单。因此,在植物群落多样性研究中,采用较小的单个或多个样方要格外谨慎,应根据研究目的和多样性指标反应的特征来综合考虑取样尺度。

关键词: 典型草原, 物种多样性, 空间尺度, 分形分析