A global test of the subsidized island biogeography hypothesis

AIM: The decreasing capacity of area to predict species richness on small islands (the small‐island effect; SIE) seems to be one of the few exceptions of the species–area relationship. While most studies have focused on how to detect the SIE, the underlying ecological factors determining this pattern remain largely unexplored. Here, we evaluate one of the few mechanisms proposed to explain the SIE, the subsidized island biogeography hypothesis, which posits that marine productivity around small islands may alter their expected species richness. LOCATION: Seven hundred and ninety islands worldwide, including 420 islands < 1 km². TIME PERIOD: Present. MAJOR TAXA STUDIED: Angiosperms. METHODS: We applied iterative partial regression to determine the effects of island area and marine productivity on plant species richness for islands of varying sizes. We also employed geographically weighted regression to account for non‐stationarity in the marine productivity effects. Lastly, we used estimates of ammonia emissions based on nutrient excretion by seabird colonies from a subset of 66 islands to evaluate the effects of marine resources deposition on angiosperm species richness. RESULTS: We found no effect of marine productivity on insular species richness, at both regional and global scales. In all models, area emerged as the only predictor of plant species richness. A weak contribution of marine productivity was only detectable in models with a low number of islands, but this effect was independent of island size. Although nutrient deposition significantly contributes to explaining plant diversity, this effect was also independent of island size. MAIN CONCLUSIONS: Our study demonstrates that marine productivity has no general effect on plant species richness of small islands worldwide. Although marine‐derived resources may still contribute to species richness variation, this effect does not seem to be restricted to small islands. Overall, our results do not provide support for the subsidized island biogeography hypothesis.