ORIGINAL ARTICLE: A general dynamic theory of oceanic island biogeography

Abstract Aim MacArthur and Wilson’s dynamic equilibrium model of island biogeography provides a powerful framework for understanding the ecological processes acting on insular populations. However, their model is known to be less successful when applied to systems and processes operating on evolutionary and geological timescales. Here, we present a general dynamic model (GDM) of oceanic island biogeography that aims to provide a general explanation of biodiversity patterns through describing the relationships between fundamental biogeographical processes – speciation, immigration, extinction – through time and in relation to island ontogeny. Location Analyses are presented for the Azores, Canaries, Galápagos, Marquesas and Hawaii. Methods We develop a theoretical argument from first principles using a series of graphical models to convey key properties and mechanisms involved in the GDM. Based on the premises (1) that emergent properties of island biotas are a function of rates of immigration, speciation and extinction, (2) that evolutionary dynamics predominate in large, remote islands, and (3) that oceanic islands are relatively short‐lived landmasses showing a characteristic humped trend in carrying capacity (via island area, topographic variation, etc.) over their life span, we derive a series of predictions concerning biotic properties of oceanic islands. We test a subset of these predictions using regression analyses based largely on data sets for native species and single‐island endemics (SIEs) for particular taxa from each archipelago, and using maximum island age estimates from the literature. The empirical analyses test the power of a simple model of diversity derived from the GDM: the log(Area) + Time + Time 2 model (ATT 2 ), relative to other simpler time and area models, using several diversity metrics. Results The ATT 2 model provides a more satisfactory explanation than the alternative models evaluated (for example the standard diversity–area models) in that it fits a higher proportion of the data sets tested, although it is not always the most parsimonious solution. Main conclusions The theoretical model developed herein is based on the key dynamic biological processes (migration, speciation, extinction) combined with a simple but general representation of the life cycle of oceanic islands, providing a framework for explaining patterns of biodiversity, endemism and diversification on a range of oceanic archipelagos. The properties and predictions derived from the model are shown to be broadly supported (1) by the empirical analyses presented, and (2) with reference to previous phylogenetic, ecological and geological studies.