Sown species richness and realized diversity can influence functioning of plant communities differently.

Biodiversity-ecosystem functioning experiments (BEF) typically manipulate sown species richness and composition of experimental communities to study ecosystem functioning as a response to changes in diversity. If sown species richness is taken as a measure of diversity and aboveground biomass production as a measure of community functioning, then this relationship is usually found to be positive. The sown species richness can be considered the equivalent of a local species pool in natural communities. However, in addition to species richness, realized diversity is also an important community diversity component. Realized diversity is affected by environmental filtering and biotic interactions operating within a community. As both sown species richness and the realized diversity in BEF studies (as well as local species pool vs observed realized richness in natural communities) can differ markedly, so can their effects on the community functioning. We tested this assumption using two data sets: data from a short-term pot experiment and data from the long-term Jena biodiversity plot experiment. We considered three possible predictors of community functioning (aboveground biomass production): sown species richness, realized diversity (defined as inverse of Simpson dominance index), and survivor species richness. Sown species richness affected biomass production positively in all cases. Realized diversity as well as survivor species richness had positive effects on biomass in approximately half of cases. When realized diversity or survivor species richness was tested together with sown species richness, their partial effects were none or negative. Our results suggest that we can expect positive diversity-productivity relationship when the local species pool size is the decisive factor determining realized observed diversity; in other cases, the shape of the diversity-functioning relationship may be quite opposite.