Below-ground resource partitioning alone cannot explain the biodiversity-ecosystem function relationship : A field test using multiple tracers

Below-ground resource partitioning is among the most prominent hypotheses for driving the positive biodiversity-ecosystem function relationship. However, experimental tests of this hypothesis in biodiversity experiments are scarce, and the available evidence is not consistent. We tested the hypothesis that resource partitioning in space, in time or in both space and time combined drives the positive effect of diversity on both plant productivity and total community resource uptake. At the community level, we predicted that total community resource uptake and biomass production above- and below-ground will increase with increased species richness or functional group richness. We predicted that, at the species level, resource partition breadth will become narrower, and that overlap between the resource partitions of different species will become smaller with increasing species richness or functional group richness. We applied multiple resource tracers (Li and Rb as potassium analogues, the water isotopologues-H2 18O and 2H2O, and 15N) in three seasons at two depths across a species and functional group richness gradient at a grassland biodiversity experiment. We used this multidimensional resource tracer study to test if plant species partition resources with increasing plant diversity across space, time or both simultaneously. At the community level, total community resource uptake of nitrogen and potassium and above- and below-ground biomass increased significantly with increasing species richness but not with increasing functional group richness. However, we found no evidence that resource partition breadth or resource partition overlap decreased with increasing species richness for any resource in space, time or both space and time combined. Synthesis. These findings indicate that below-ground resource partitioning may not drive the enhanced resource uptake or biomass production found here. Instead, other mechanisms such as facilitation, species-specific biotic feedback or above-ground resource partitioning are likely necessary for enhanced overall ecosystem function