In megadiverse regions and countries, gaps in biological sampling can make traditional mapping approaches for conservation difficult. Strategies developed for well-sampled regions may produce unsatisfactory results when applied in countries with less even sampling—such as Brazil.
In this study, authors developed a new comprehensive spatial framework for mapping highly biodiverse areas in Brazil using species occurrences as mediated by GBIF and applying special techniques to reduce the effect of irregular sampling. From thoroughly cleaned species records and a phylogenetic supertree, the authors derive a range of quantitative biodiversity variables, including composition, richness and endemism—both at the species and phylogenetic level.
Summarizing the quantitative variables at a regional scale, the model identifies the smallest possible areas with the most unique biodiversity, producing a map of relevant priority areas for biodiversity conservation. In the optimal solution, the model is able to encompass ~90 per cent of known biodiversity in only 10 per cent of the country's area.