LAUSR

Abstract Understanding how species sort themselves into communities is essential to explain the mechanisms that maintain biodiversity. Important insights into potential mechanisms of coexistence may be obtained from observation of non-random patterns in community assembly. The spatial niche overlap (Pianka index) and co-occurrence (c-score) patterns in carabid species in three types of steppes (desert steppe, typical steppe, and meadow steppe) in China was investigated. Non randomness was tested using null models. Niche overlap values were significantly higher than expected by chance in the desert steppe, where vegetation cover is less abundant and less uniformly distributed, which possibly forces species to concentrate in certain places. In the typical and meadow steppes, results were influenced by the scale of the analysis. At a broad scale, niche separation was found as a result of species segregation among different sectors (habitats) within these steppes, but when the analysis was conducted at a finer scale, species appeared to be no more segregated than expected by chance. The high co-occurrence averages found in the meadow and typical steppes indicate that the distributions of the species found in a site may be negatively affected by the presence of other species, which suggests that some species tend to exclude (or reduce the abundance of) others. The very low c-score average observed in the desert steppe suggests that competition is not involved there. Thus, in more homogeneous landscapes (such as the typical and meadow steppes), competition might play some role in community structure, whereas spatial variation in the abundances of species is more driven by the uneven spatial distribution of vegetation in the landscape where productivity is lower and less uniformly distributed.