How many herbarium specimens are needed to detect threatened species?

The distribution, ecology and conservation status of the majority of plant species are poorly known. One of the challenges ahead is to address this knowledge gap and give more emphasis to this important group of species that represents a critical component of earth’s biodiversity. Full conservation assessments require expert knowledge of the group concerned, but for the majority of plant species, especially those from the tropics, the best source of knowledge is specimens housed within herbaria. Digitisation projects are underway to render information from this important global biodiversity resource more accessible; the next step is to assemble and utilise these data to make better informed conservation decisions. One crucial question is: how many herbarium specimens are needed to detect threatened species? Such information would inform and help to prioritise digitisation efforts. Using 11,461 herbarium records we assessed species geographic range to determine a preliminary conservation status of 661 endemic species of Leguminosae and Orchidaceae from Madagascar, following the IUCN criteria. By capturing 15 georeferenced specimens per species we produced range estimates for use in conservation assessments consistent with estimates based on all known specimens, for more than 95% of species. None of the threatened species were misclassified as not threatened, and less than 3% of species would receive conservation support as a result of being falsely identified as threatened. This approach can therefore help progress towards the Global Strategy for Plant Conservation target of a conservation assessment for each plant species, while reducing digitisation effort by up to half.