Background/Question/Methods
The importance of biodiversity for maintaining function in ecosystems has been demonstrated by a large number of experiments. Increased species richness has also been shown to reduce temporal variability in biomass production suggesting that different species are required to sustain productivity in different years. We present a new approach designed to test for temporal multifunctionality, i.e. if more species are required to maintain functioning across several subsequent years?
First, we present results of a simulation study to show that the correlations of species contributions to a particular function, between years, lead to a predictable increase in the number of species needed to maintain function. We analyse the effects of species richness, rank-abundance relationships among species, and temporal variability on temporal multifunctionality. Second, we analyse a longer-term data series on aboveground productivity from the Jena biodiversity experiment in central Germany.

Results/Conclusions
Larger numbers of species were required to ensure biomass production across increasing numbers of years, although the relationship was saturating. The proportion of species needed to maintain function was higher and more variable in plots with lower species numbers. These results strongly suggest that biodiversity experiments may underestimate the number of species needed to maintain function across longer time scales.