Dissimilar biodiversity data sets yield congruent patterns and inference in lichens

Abstract

Large-scale efforts to aggregate and promote the re-use of biodiversity data are leading to novel insights into biogeography and macroecology. However, secondary analyses must account for the tradeoffs and limitations of the original studies. Studies of speciose and taxonomically complex groups often utilize morphospecies or functional subsets as proxies, potentially complicating data re-use. We evaluated whether lichen biodiversity patterns are robust to differences in sampling methodology, utilizing parallel analyses to compare species richness, regional species pool variation, species probabilities of occurrence, and correlation of those three with environmental variables in data sets that cover the same geographic region. Our analyses revealed that, although individual species distributions sometimes differed in idiosyncratic ways, inference based on the aggregated response of multiple species was generally robust across the two datasets, despite differences in observer expertise and functional and taxonomic scope. This suggests that biodiversity data assembled from disparate sources could be used to evaluate biogeographical and macroecological hypotheses in understudied groups such as lichens, particularly at larger spatial scales.