Functional diversity predicts overyielding effect of species combination on primary productivity

The effect of biodiversity on ecosystem functioning has proven variable both within and among manipulative studies. Species richness is the most commonly used measure of biodiversity in such studies, but the range of species' functional traits (functional diversity), not the number of species per se, likely underpins a key mechanistic link between species richness and ecosystem functioning. However, the majority of experiments that have examined the effect of functional diversity have manipulated functional group richness, an approach recognised to suffer numerous limitations. Continuous measures of functional diversity avoid many of these limitations, but the relationship between continuous functional diversity and the magnitude of ecosystem processes has been largely untested. Using one vs two-species mixtures of rock pool macroalgae as a model, we conducted a field experiment to determine the effect of a continuous measure of functional diversity (functional attribute diversity, FAD, the degree of functional differentiation based on four functional traits) on the magnitude of net primary productivity and overyielding, based upon two alternative null-models. The total magnitude of productivity was largely determined by the identity of species present, not FAD. However, FAD proved to be a good predictor of overyielding (variation in productivity after the dominant effects of species identity had been accounted for). Furthermore, despite differences in the mean magnitude of the effect of combining species, the positive relationship between FAD and overyielding was consistent according to both additive and substitutive null-models. Our findings imply that whilst knowledge of species' independent contributions remains indispensable in the prediction of biotic effects on ecosystem functioning within a trophic level, continuous measures of functional diversity should be used as a supplementary tool to predict the magnitude of overyielding, thereby refining predictions.