Abstract

Understanding why some regions have higher levels of diversity and which factors are driving the occurrence of species in a particular area is crucial for environmental management and for the development of species conservation strategies. In this study, we studied seven species of the Caesalpinioideae that are endemic in Central Africa (Democratic Republic of the Congo, Burundi and Rwanda). The objectives of this study were to identify the environmental factors that constrain their distribution, to determine the potential areas where each species could be present, to assess the current conservation status of each species and to evaluate how well the species are protected by the protected areas in the region. Distributions were analyzed and potential distributions predicted using the Maxent species distribution algorithm with climatic (precipitation and temperature) and non-climatic predictor variables (soil, elevation, and slope). Environmental variables and species occurrence data were obtained respectively from the WorldClim database and from herbarium specimens kept at the National Botanic Garden of Belgium and the Université Libre de Bruxelles. Our results suggest that the distribution of endemic species is influenced by a combination of climatic and non-climatic variables. Soil type, temperature annual range and precipitation of the driest month were the most important predictor variables. Overlaying the potential distributions of the seven selected species indicated three areas of concentration of endemic species which should be given particular conservation attention. Comparing the potential distributions to the current Central African protected areas showed that the endemic species are not well protected, as 97% of their potential habitat is localized outside protected areas. Hence, additional reserves should be created to improve the protection of these endemic plant species.

Key words: Caesalpinioideae, maxent, species distribution model, species response curve, protected area.