Abstract
Global patterns of organisms have long been investigated by calculating the (dis)similarity among geographical units followed by multivariate analysis. Beta-diversity-related structural characteristics of world-scale data, such as nestedness or species replacement may also be considered as an additional tool in revealing distributional patterns more accurately. To achieve this objective, our study combines cluster analysis and ordination based on Jaccard and Simpson dissimilarity with the decomposition of beta diversity into meaningful fractions. As a model group, the oribatid mite fauna of the seven biogeographic realms was analysed at three taxonomic levels, i.e. species, genus and family. The highest overall similarity was obtained between the Palaearctic and Nearctic realms and the lowest richness resulted for the Antarctic realm. The classifications and ordinations usually differed with the two dissimilarity indices. Beta diversity decomposition showed that these discrepancies were caused by different patterns of nestedness and taxon richness. Our study is the first to demonstrate that such a complex approach may disclose several features of biogeographic data not apparent otherwise and therefore may improve our understanding of inter-regional relationships.
Acknowledgements
The authors gratefully acknowledge the contribution of Prof. János Podani for his professional and linguistic help and for his valuable ideas in data analyses. We would like to thank two anonymous reviewers for their valuable and helpful comments.