Data from: Functional traits explain ecosystem function through opposing mechanisms

The ability to explain why multispecies assemblages produce greater
biomass compared to monocultures, has been a central goal in the quest to
understand biodiversity effects on ecosystem function. Species
contributions to ecosystem function can be driven by two processes: niche
complementarity and a selection effect that is influenced by fitness
(competitive) differences, and both can be approximated with measures of
species’ traits. It has been hypothesised that fitness differences are
associated with few, singular traits while complementarity requires
multidimensional trait measures. Here, using experimental data from plant
assemblages, I show that the selection effect was strongest when trait
dissimilarity was low, while complementarity was greatest with high trait
dissimilarity. Selection effects were best explained by a single trait,
plant height. Complementarity was correlated with dissimilarity across
multiple traits, representing above and below ground processes. By
identifying the relevant traits linked to ecosystem function, we obtain
the ability to predict combinations of species that will maximise
ecosystem function.