“Beta diversity patterns derived from island biogeography theory”

Posted on

Muyang Lu, David Vasseur, and Walter Jetz (Sep 2019)

Read the Article (Just Accepted)

Beta diversity (Jaccard) is independent of the size of mainland species pool in the Island Biogeography Theory

Island biogeography theory links beta diversity to colonization and extinction rates

The island biogeography theory pioneered by MacArthur and Wilson has served as a foundation to understand the drivers of species richness for more than 50 years, but never tells us about what shapes beta diversity, i.e. the spatial variation of community composition among islands. A theoretical synthesis of island biogeography theory and beta diversity finally emerged at the call of a unified metacommunity framework to link ecological processes to more ecological patterns. Following MacArthur and Wilson’s original formulation (assuming species-equivalency), Muyang Lu, David Vasseur, and Walter Jetz first show that pairwise beta diversity (Jaccard dissimilarity) of two random islands is only determined by colonization and extinction rates of a species and independent of the size of species pool. This result runs counter to the prevailing belief that beta diversity increases with the size of species pool, with simulations further corroborating the conclusions under non-neutral scenarios. They further use an empirical bird dataset in Thousand Island Lake, China to demonstrate that beta diversity patterns is more powerful in detecting non-neutral processes than species richness. By adding a new dimension to the predictions of the half-century old island biogeography theory, this study opens the gate to investigate community assembly processes in a more systematic way and paves the way for a unified metacommunity theory.


Metacommunity theory and its constituent Theory of Island Biogeography (TIB) have the potential to unify ecology across different scales. TIB has been successful in predicting alpha diversity patterns such as species-area relationships and species-abundance distributions, but lags behind in predicting spatial beta diversity patterns. In this study we use island biogeography theory as the starting point to integrate spatial beta diversity patterns into metacommunity theory. We first derive theoretical predictions for the expected beta diversity patterns under the classic MacArthur and Wilson framework where all species have equal colonization and extinction rates. We then test these predictions for the avian community composition of 42 islands (and 93 species) in the Thousand Island Lake, China. Our theoretical results corroborate that longer distance and smaller area lead to lower beta diversity, and further reveal that pairwise beta diversity is independent of the size of mainland species pool. We also find that for the partitioned pairwise beta diversity components, the turnover component increases with the ratio of extinction rates and colonization rates, while the nestedness component is a monotonic function of the ratio of extinction rates and colonization rates. For the empirical island system, we find that beta diversity patterns better distinguish a species-equivalent model from a species-nonequivalent model than alpha diversity patterns. Our findings suggest that beta diversity patterns provide a powerful tool in detecting non-neutral processes and our model has the potential to incorporate more biological realism in future analyses.