Stability of ecosystem properties in response to above-ground functional group richness and composition

While there has been a rapidly increasing research effort focused on understanding whether and how composition and richness of species and functional groups may determine ecosystem properties, much remains unknown about how these community attributes affect the dynamic properties of ecosystems. We conducted an experiment in 540 mini-ecosystems in glasshouse conditions, using an experimental design previously shown to be appropriate for testing for functional group richness and composition effects in ecosystems. Artificial communities representing 12 different above-ground community structures were assembled. These included treatments consisting of monoculture and two- and four-species mixtures from a pool of four plant species; each plant species represented a different functional group. Additional treatments included two herbivore species, either singly or in mixture, and with or without top predators. These experimental units were then either subjected to an experimentally imposed disturbance (drought) for 40 d or left undisturbed. Community composition and drought both had important effects on plant productivity and biomass, and on several below-ground chemical and biological properties, including those linked to the functioning of the decomposer subsystem. Many of these compositional effects were due to effects both of plant and of herbivore species. Plant functional group richness also exerted positive effects on plant biomass and productivity, but not on any of the below-ground properties. Above-ground composition also had important effects on the response of below-ground properties to drought and thus influenced ecosystem stability (resistance); effects of composition on drought resistance of above-ground plant response variables and soil chemical properties were weaker and less consistent. Despite the positive effects of plant functional group richness on some ecosystem properties, there was no effect of richness on the resistance of any of the ecosystem properties we considered. Although herbivores had detectable effects on the resistance of some ecosystem properties, there were no effects of the mixed herbivore species treatment on resistance relative to the single species herbivore treatments. Increasing above-ground food chain length from zero to three trophic levels did not have any consistent effect on the stability of ecosystem properties. There was no evidence of either above-ground composition or functional group richness affecting the rec