How to explain bird diversity on islands?

Species distribution on the surface of the earth is neither regular nor uniform. Some geographic regions are richer in species than others, or harbour a specific pool of species. Understanding the nature and causes of this variation is one of the major challenges of biogeography, a discipline at the interface between ecology, evolutionary biology and geology. Since the pioneering work of Darwin and Wallace, oceanic volcanic islands have been the favorite study areas for biogeographers.

In 1963, Robert MacArthur, a mathematician and naturalist, and Ed Wilson, an entomologist, became interested in the problem and proposed one of the most important theories of modern biology, i.e. the theory of island biogeography (also known as the equilibrium theory). Based on the observations that large islands contain more species than small islands, and that isolated islands contain fewer species than islands close to continents, MacArthur and Wilson proposed a simple mathematical model where the number of species on an island results from a dynamic equilibrium between the colonization process of the islands and the extinction of the species present on the islands. This model predicts the variation in the number of species according to the area and the isolation of the islands. However, one observation could not be taken into account in the model of MacArthur and Wilson : the observation that species evolve and that over time  new species that differ from their relatives on the continents emerge on islands.

As surprising as it may seem, no study so far has been able to provide a clear understanding of the causal relationships between the area and isolation of islands and the probabilities of colonization, formation of new species, and extinction of species present on the islands. The estimation of these relationships even seemed out of reach for decades, in particular because of the difficulties linked to the estimation of the evolutionary history of species.

In a new study published in the journal Nature, a group of researchers including ornithologists, evolutionary biologists and mathematicians, led by Luis Valente (Museum für Naturkunde Berlin / Naturalis Biodiversity Center, Leiden), and including researchers from the National Museum of Natural History (Institute of Systematics, Evolution, Biodiversity; UMR 7205 CNRS MNHN UPMC EPHE) and Paul Sabatier University of Toulouse (Laboratory of Evolution and Biological Diversity; UMR 5174 CNRS UPS IRD) quantified, for the first time, the dependence of colonization, speciation, and extinction processes on the area and isolation of the islands.

Using a set of new data obtained on birds from oceanic islands around the world, researchers proved that the colonization rate decreases with the degree of isolation of the islands. They also proved that the risk of extinction decreases with area, and the probability of the formation of new species increases with area and degree of isolation. While these results may seem intuitive at first glance, appropriate data and appropriate statistical methods have so far been lacking to rigorously test the theory of MacArthur and Wilson. The study of Valente and colleagues, in addition to confirming the theoretical predictions, provides statistically robust estimates of the precise form of the relationships between these key parameters underlying the process of establishing fauna and flora, which are colonization, speciation, and extinction.

Using genetic data obtained from blood samples of hundreds of species of birds  during numerous expeditions in dozens of oceanic archipelagos during the last 20 years, the authors have estimated the evolutionary history of the lineages of birds that colonized the islands. They developed and then implemented a new model that makes it possible to predict not only the number of species but also the evolutionary characteristics of island faunas on a global scale. This model remarkably well predicts the number of species of certain archipelagos, such as Hawaii or the Canary Islands. On the other hand, some archipelagos contain a greater number of species than that predicted by the model, notably the Comoros and São Tomé & Príncipe, two archipelagos whose avifauna - still poorly studied - deserves more attention from ornithologists.

A result of this study is completely surprising and almost unexpected. Oceanic islands are well known to be suitable places  for spectacular radiations of species, such as the Finches of Darwin where an ancestral species diversified in more than 15 species after having colonized the Galapagos archipelago. However, in a large majority of cases in this study, species richness on islands is mainly explained by the evolution of colonizing species without increasing the number of species within archipelagos or islands. This study constitutes a theoretical as well as a methodological breakthrough in the analysis of the fundamental processes that govern global biodiversity.