Introduction

Functional diversity schemes: are they useful?

Recent years have seen great interest in the importance of species richness for the functioning and stability of ecological communities (Ives and Carpenter 2007). Empirical examinations of richness effects typically vary the number of species in experimental treatments and measure resulting ecosystem functions such as biomass accumulation or resource uptake (Naeem et al. 2009). Across trophic levels and communities of many types, a clear pattern has emerged from these experiments: community processes (biomass accumulation, resource uptake, etc.) generally become more efficient when more species are present (Hooper et al. 2005; Cardinale et al. 2006). This pattern is generally attributed to resource partitioning among species, where species differ in ecologically significant ways such that they complement one another (Hooper et al. 2005). For example, in English meadow communities multiple plant species coexist, because different plant species exploit different hydrological conditions (Silvertown et al. 1999). The plants that dominate drought-prone areas are different from those that thrive in flood-prone areas and, presumably, total plant biomass is greatest when both plant groups (drought tolerant and flood tolerant) are present.

A remaining challenge is to effectively predict, a priori, the particular species (or groups of species) that will complement one another. One simplifying scheme that has received considerable attention is the lumping of species into ‘functional groups’. In this functional-group approach, species within a group are relatively similar to one another, and considered ecologically redundant, whereas species in different groups are distinct and complementary (Hillebrand and Matthiessen 2009). This approach gained support from studies suggesting that plant species can be classified into such functional groups (grasses, forbs, legumes and woody plants), and that the number of functional groups is a more effective predictor of ecosystem function than species richness (Diaz and Cabido 2001). For example, in savannah grasslands, plant communities that included C3 grasses, C4 grasses, forbs, legumes and woody plants had greater biomass and plant nitrogen accumulation, and reduced light penetration, than those communities lacking one or more of these groups (Tilman et al. 1997). These authors suggested that competition was greater within than between functional groups, consistent with niche similarity within, but niche differentiation among, groups.