Journal of Biogeography Aguilée R, Pellerin F, Soubeyrand M, Choin J and Thébaud C
Aim: Accounting for geo-environmental dynamics is crucial to understand commu-nity assembly across islands. Whittaker et al. (J Biogeogr, 35:977–994, 2008)’s General Dynamic Model (GDM) aims towards this goal. Yet, it does not explicitly consider that most islands belong to archipelagos. We examined how island biodiversity dynamics are influenced by the interaction of eco-evolutionary processes acting at the archi-pelago level with each island's geo-environmental dynamics.
Location: Hypothetical archipelagos.
Taxon: Any.Methods: We used an individual- based model, ecologically neutral within the archi-pelago. Several islands emerge in succession with a typical volcanic ontogeny. We considered both mainland and inter-island dispersal. Geographically isolated lineages diverged over time, possibly speciating.
Results: We found diversity to be at dynamic equilibrium. In an archipelago, islands hosted more diversity and more endemic species, at both island and archipelago levels, than an equivalently-sized single isolated island. This was due to an ‘archipelago effect’: inter- island dispersal increased within-island diversity through species occurrence on multiple islands; species may undergo anagenetic changes on the colonised islands, eventually speciating, thereby increasing archipelago diversity. Biodiversity dynam-ics of different islands may differ even on islands with identical geo-environmental dynamics because the archipelago effect varied over time and affected each island differently (‘history effect’). By accounting for these effects, we predicted detectable deviations from the GDM predictions, which are largest for remote archipelagos, with islands located close together and with an intermediate time of island emergence. In linear stepping-stone archipelagos, we predicted higher diversity on centrally located islands.
Main conclusions: Our results demonstrate that analyses of insular biodiversity data would greatly benefit from explicitly accounting for both archipelago and his-tory effects. We suggest incorporating variables characterising the spatio-temporal
Comments