Functional richness: Overview of indices and underlying concepts
Introduction
Assessing functional diversity, which is now recognized as a main driver of the functioning of ecosystems and their responses in both terrestrial and aquatic environments, has become a crucial challenge in community ecology. This broad concept is now usually split into three major components: functional richness, functional evenness and functional divergence (Mouillot et al., 2005, Mouillot et al., 2013, Mason et al., 2005). The assessment of functional richness is probably the oldest and the most widely used way of investigating functional diversity (see Blondel, 2003). However, behind this term, simple in appearance, lies a much more complex component which may vary widely in its definitions according to the authors. Some approaches define the concept of functional richness as the number of species sharing the same characteristics, whereas other approaches have used the distribution of species trait values in the functional space. Consequently, combining all these definitions, and the associated indices, within a single general concept (i.e. functional richness) can cause confusion for field ecologists, and misinterpretation of the results obtained (Battisti and Contoli, 2011, Contoli and Luiselli, 2015). However, there has to date been no survey reviewing all methods and indices used to assess functional richness and what facet of this concept they really measure (and how).
Initially, functional richness was mainly defined as the number of species sharing similar taxonomic, phylogenetic or morphological characteristics (Blondel, 2003). The first method used to estimate this concept of functional richness was based on the number of predefined functional groups in a given assemblage (Magurran, 2013). This approach consisted of grouping together all species sharing similar characteristics (Blondel, 2003). It was then assumed that all species belonging to the same functional group have similar effects on or responses to ecosystems (Chapin et al., 1996, Petchey and Gaston, 2002, Lavorel and Garnier, 2002). Although some studies have demonstrated that the number of functional groups was highly correlated with the ecosystem productivity (Tilman et al., 1997, Hector et al., 1999, Reich et al., 2001), and might be a good proxy of ecosystem functioning (Petchey et al., 2004), assessing the functional richness through the assessment of pre-defined functional groups can be problematic. The methods used to assign species to each group are in fact highly subjective because they are based on the user's choices (Holmes et al., 1979, Chapin et al., 1996, Diaz and Cabido, 2001). The definition of the boundaries between each group determines the number of functional groups and by consequence, may impact which species are included within each group. If we assume that the number of functional groups is in direct relation with the ecosystem functioning, defining this number in a non-standardized way can produce undesirable variability in the results obtained (Loiseau and Gaertner, 2015). Furthermore, the use of functional groups ignores all possible differences between species belonging to a single group, whereas these differences can be important notably when the functional traits are quantitative (Petchey and Gaston, 2002, Petchey et al., 2004, Ricotta, 2005). Due to these shortcomings, new approaches based on analyses of the trait values of species have been developed to assess the functional richness of communities. The aim of these approaches is to evaluate the functional richness according the position of species and the distribution of species trait values in the functional space. The first approach developed along these lines refers to the FR index of Mason et al. (2005) and describes the amount of niche filled by the community (Mason et al., 2005). This index considers only a single functional trait in its computation. However, many studies have demonstrated that considering just one functional trait may induce oversimplified conclusions because the major part of the ecosystemic functions cannot be summarized by a single functional trait (Villéger et al., 2008, Schleuter et al., 2010, Hortal et al., 2015).
For these reasons, the new approaches developed in recent years for assessing the functional richness of communities are based on simultaneous consideration of several functional traits. Following the definition offered by Mason et al. (2005), the great majority of studies consider functional richness as “the amount of niche space occupied by the species within a community”. These approaches take into account the degree of difference among species with reference to functional variables (i.e. functional traits, Petchey and Gaston, 2002, Mason et al., 2003, Mouillot et al., 2005), and aim to estimate their distribution within the community studied. For that purpose, various indices have been developed, based on different mathematical formulas and concepts (e.g. FD index of Petchey and Gaston, 2002, FRic index of Villéger et al., 2008, TOP index of Fontana et al., 2015). In this context, the aim of this paper is to clearly identify which index characterizes what (in accordance with their respective mathematical construction), and to highlight how we can use this information to achieve a better – and sometimes more complete - view of the functional richness of a community. On this basis, we propose some guidelines to help users in selecting indices according to the concept of functional richness on which they wish to focus and to the properties of the available data.
More specifically, we have focused our discussion on the indices respecting three essential properties in the assessment of the functional richness of communities. Firstly, the index should conserve the continuous values of functional traits (because using functional groups would implicitly assume that species within a group are functionally identical and that the functional differences of species pairs drawn from different groups are equivalent, Ricotta, 2005). Secondly, the index should consider several functional traits (to limit the risk of oversimplification and misrepresentation arising from analyses restricted to a single functional trait, Villéger et al., 2008). Thirdly, the index should not consider the species abundance (to be consistent with the one of the criteria of functional richness index as defined by Mason et al., 2005: “a section of niche space is considered to be occupied even if only very little abundance occurs within it”, i.e. a section of niche space is considered as occupied even if only one individual is present in this place).
Consequently, we focused our study on the FD P&G index (Petchey and Gaston, 2002), the FRic index (Villéger et al., 2008), the TOP index (Fontana et al., 2015) and the N-hypervolume index (Blonder et al., 2014), hereafter referred to as ’inter-entity approaches’. Furthermore, some recent studies have demonstrated the importance of taking into account the intraspecific trait variability (ITV) in the assessment of the functional richness of communities, because traits vary among genotypes within species and according to the environmental conditions (Messier et al., 2010, Albert et al., 2011, Clark et al., 2011, Laughlin and Laughlin, 2013, Auger and Shipley, 2013). Thus, we also included in our analyses the FRim index of Schleuter et al. (2010) and the TPDRich index of Carmona et al. (2016a), which are two recent indices taking into account the ITV.
Section snippets
Inter-entity approaches
The indices presented in this section consider as basic unit in their computation the species or the functional entities (i.e. individuals sharing the same combination of trait values) but not the probabilistic nature of intraspecific variability.
Considering the intraspecific trait variability: the emergence of new approaches
In this work, we take consideration of intraspecific trait variability (ITV) as the fact that a functional trait is considered as a random variable drawn from a probability distribution (Albert et al., 2011). Following this definition, none of the four indices considered above really takes into account the intraspecific trait variability. The use of functional richness indices taking into account ITV in the scientific literature is still very scarce, in particular because of the poor
Discussion
Mason et al. (2005) originally defined functional richness, which is one of three major components of functional diversity, as “the amount of functional trait space occupied by the species”. On the basis of this definition, a range of indices has been developed with the aim of assessing this component. However, although the authors argued that their indices describe the functional richness, we have demonstrated here that each of them describes only one part of it. A detailed survey clearly
Author contribution
G.L., N.L., J-C. G. conceived the ideas; G.L., N.L. make the simulations and analysed the results; G.L. led the writing. All authors contributed to the final draft.
Acknowledgments
This work was supported by the University of French Polynesia (ED 469) and funded by the Research Ministry of French Polynesia (Délégation à la Recherche de Polynesie Française). We are grateful to M. Paul for improvements to the English text. We thank the Assistant Editor and the two anonymous referees whose comments have contributed to improving the previous version of this manuscript.
References (67)
- et al.
When and how should intraspecific variability be considered in trait-based plant ecology?
Perspect. Plant Ecol. Evol. Systemat.
(2011) Pushing past boundaries for trait hypervolumes: a Response to Carmona et al
Trends Ecol. Evol.
(2016)- et al.
Traits without borders: integrating functional diversity across scales
Trends Ecol. Evol.
(2016) - et al.
The density awakens: a reply to Blonder
Trends Ecol. Evol.
(2016) - et al.
Vive la difference: plant functional diversity matters to ecosystem processes
Trends Ecol. Evol.
(2001) Conservation evaluation and phylogenetic diversity
Biol. Conserv.
(1992)- et al.
Advances in modeling trait-based plant community assembly
Trends Plant Sci.
(2013) - et al.
Fitness of multidimensional phenotypes in dynamic adaptive landscapes
Trends Ecol. Evol.
(2015) - et al.
Trait variation and functional diversity maintenance of understory herbaceous species coexisting along an elevational gradient in Yulong Mountain, Southwest China
Plant Divers.
(2016) - et al.
A functional approach reveals community responses to disturbances
Trends Ecol. Evol.
(2013)