Bioregions of Eastern Brazil, Based on Vascular Plant Occurrence Data

Abstract

The geographical division of the Earth into meaningful biodiversity units (e.g., biomes, areas of endemism, ecoregions or bioregions) is a necessary step for the study of biodiversity and its conservation. Eastern Brazil harbors a significant proportion of the Earth’s terrestrial biodiversity in a geographically complex area. However, the delimitation of biogeographical areas in eastern Brazil has relied on the simultaneous use of biotic (animal and plant distributions and vegetation physiognomy) and abiotic factors, often without an explicit methodology. Here we take advantage of the availability of large numbers of vascular plant specimens and their digitized data, the existence of a well curated taxonomy for the plants that occur in the area, and the emergence of new biogeographic tools in order to identify bioregions (geographic areas that contain similar taxa) of eastern Brazil. To provide a classification scheme suitable to studies of lineages that differ in dispersion, species richness, and endemism, we provide three levels of grouping. The dataset analyzed here had a comparable number of species across eastern Brazil relative to the recent taxonomic synthesis of the Brazilian flora. Maps of richness and endemism are provided for the region, and confirm regions of eastern Brazil recognized for both highly diversity and endemism across both coastal (Serra do Mar and Mantiqueira) and inland mountain ranges (Campos Rupestres), as well as in southern Bahia. The first network analysis divided eastern Brazil into 10 bioregions, which were clustered in five super-bioregions and divided in 23 sub-bioregions in the two additional network analyses. The super-bioregions recovered correspond to the Southern Atlantic Forest/Paraná Forests, the Northern Atlantic Forest, the Espinhaço/Mantiqueira, the Cerrado, and the Caatinga/Diamantina. To some extent, these areas present some congruence with domains, where a major incongruence is distinctiveness of the Espinhaço/Mantiqueira super-bioregion, while the Caatinga/Diamantina super-bioregion presented the highest congruence. Comparisons of species richness, endemism and overlapping of the bioregions with three other classifications (domains, ecoregions and biogeographical provinces) are presented. The shapefiles of the recovered bioregions are available for public use.