Deficiency in African plant distribution data - missing pieces of the puzzle

Biodiversity is spatially unevenly distributed and so is the information on its spatial patterns. This uneven distribution of information on species occurrences is an important impediment to the conservation of biodiversity. Based on 185 427 collection records of 5873 plant species in sub-Saharan Africa, we analyse the availability of distribution data suitable for the GIS-based mapping of plant diversity patterns at a one-degree resolution. Using the bioclimatic model GARP, distribution ranges for each species were modelled. In order to identify data-deficient areas, the documented and modelled diversity patterns were compared. Only for a few, well-known centres of plant diversity are there comparatively many data collection records available. For several of the areas with very few collection records, such as the Guinean montane forests, the north-western Congolian lowland forests, and the southern Albertine Rift montane forests, the model predicts a species richness much higher than currently documented. In many of the data deficient areas, difficult conditions for scientific research appear to have limited collection activities for decades. Only strategic field collections can fill these gaps. Another cause of data deficiency is that data collected and digitized do not match the quality requirements for GIS-based work at the super-regional scale. In particular, regional databases documenting partial ranges of species are rarely connected. One challenge for the Global Strategy for Plant Conservation is therefore to establish international collaborative structures and technical standards that will allow analysis of biogeographical patterns across political boundaries. © 2006 The Linnean Society of London, Botanical Journal of the Linnean Society, 2006, 150, 355–368.