Community niche predicts the functioning of denitrifying bacterial assemblages.

Predicting biodiversity effects on ecosystem functioning requires adequate evaluation of the mechanisms explaining why more diverse systems could perform better than less diverse ones. In this context, tackling functional diversity has become an important issue. Even though the aggregation of species into functional groups supposes niche differences among groups, the concept of niche has not been fully exploited in the context of the biodiversity-ecosystem functioning research. Here we report the results of microcosm experiments where we used bacteria as a model to explore whether niche differences among species provide a good estimation of community functioning. For that we used experimental communities of denitrifying bacterial species and investigated the effects of bacterial diversity on two community processes, denitrification and anaerobic CO2-production. We first measured the activities of 16 bacterial species grown individually on six different carbon sources. We then used the same set of species to assemble communities varying in both species richness and composition in microcosms containing a mixture of all six carbon sources. The performances of individual species on individual carbon sources were used to calculate, for each process measured, an a priori index called "community niche" that accounted for the performances of the species present in a given community across the entire range of the six resources. We found that species richness had a positive but small effect on both community processes whereas community niche explained a much larger proportion of the variation. According to the results of a path analysis, community niche was the main driver for the corresponding community process, but species richness affected community niche and thus had an indirect effect on denitrification and CO2 production. In addition to community niche, the presence of particular bacterial species also influenced community functioning, indicating that other effects than the capacity to use carbon sources played a, albeit minor, role in our experiment. Our study provides evidence for the importance of resource niches in shaping biodiversity-ecosystem functioning relationships of bacterial communities.