Island biogeography from a Grinnelian niche perspective: An example using swallowtail and sulphur butterflies from the Caribbean Islands

Island Biogeography is one of the most important theories in Ecology. Most of it is based on numerical estimates, such as total number of species, and dispersal, extinction, and speciation rates. Despite the undeniable success of this theory, species identities, history, and environmental conditions are disregarded. In this study, we propose an approach to address one of these factors, by moving from predicting species numbers to predicting the identity of species in the islands. Using our protocol, we were able to predict which species could be present on the islands, using information on the size and isolation of the island, as well as the presence of suitable environmental conditions for species establishment. We tested this approach with data of Papilionidadae and Pieridae butterflies of the Caribbean Islands (our target area). First, we defined a pool of species from a source pool area. Then, we characterized the ecological niches of the species of interest in the source pool and predicted suitability for such species in the target area. Omission errors were rare in our findings (the ability to correctly predict a species to be present in a focal region) as we were successful in predicting most of the butterfly’s species to be in the target region based on the species present in the source pool area and the environmental conditions in both regions. Commission errors (species predicted but not present in the target area) were higher, but these errors were associated with accessibility and other geographical properties of the target area.