Transferring and implementing the general dynamic model of oceanic island biogeography at the scale of island fragments: the roles of geological age and topography in plant diversification in the Canaries

Abstract

Aim The general dynamic model (GDM) of oceanic island biogeography integrates rates of immigration, speciation and extinction in relation to a humped trajectory of island area, species carrying capacity and topographic complexity through time, based on a simplified island ontogeny. In practice, many islands have more complex ontogenies, featuring surfaces of varying age. Here, we extend the GDM to apply at a local scale within islands, and test the predictions analytically within individual islands. Location El Hierro, La Palma and Tenerife (Canary Islands). Methods Following the GDM logic, we derive predictions for the distributions and richness of single island endemics (SIEs) across island landscapes of different age. We test these predictions by means of generalized linear models and binominal tests using gridded species occurrence data for vascular plant SIE species and a set of climatic, topographic and terrain age variables. We also examined phylogenetic divergence times for a subset of endemic lineages. Results Geological age, in interaction with slope, and topographic variables, best explained SIE richness at the landscape scale. About 70% of SIEs had ranges strongly biased to, or largely restricted to old terrain. Available phylogenetic divergence times of SIEs of radiated plant lineages suggested an origin on the older parts of the islands. Metrics of anthropogenic disturbance and habitat availability were unrelated to the observed SIE pattern. Main conclusions Our findings support the hypothesis that SIEs have evolved and accumulated on older and topographically complex terrain, while colonization processes predominate on the youngest parts. These results imply that evolutionary processes shape species distributions at the landscape scale within islands. This opens the perspective of extending the GDM framework to understand processes at a local scale within individual islands.