Subpopulations, locations and fragmentation: applying IUCN red list criteria to herbarium specimen data

Despite the ecological and economic importance of plants, the majority of plant species and their conservation status are still poorly known. Based on the limited knowledge we have of many plant species, especially those in the tropics, the use of GIS techniques can give us estimates of the degree of population subdivision to be used in conservation assessments of extinction risk. This paper evaluates how best to use the IUCN Red List Categories and Criteria to produce effective and consistent estimates of subpopulation structure based on specimen data available in the herbaria around the world. We assessed population structure through GIS-based analysis of the geographic distribution of collections, using herbarium specimen data for 11 species of Delonix sensu lato. We used four methods: grid adjacency, circular buffer, Rapoport's mean propinquity and alpha hull, to quantify population structure according to the terms used in the IUCN Red List: numbers of subpopulations and locations, and degree of fragmentation. Based on our findings, we recommend using the circular buffer method, as it is not dependent on collection density and allows points to be added, subtracted and/or moved without altering the buffer placement. The ideal radius of the buffer is debatable; however when dispersal characteristics of the species are unknown then a sliding scale, such as the 1/10th maximum inter-point distance, is the preferred choice, as it is species-specific and not sensitive to collection density. Such quantitative measures of population structure provide a rigorous means of applying IUCN criteria to a wide range of plant species that hitherto were inaccessible to IUCN classification.