Intraspecific diversity has been identified as an important component of functional diversity and is hypothesized to influence community assembly and structure, and species coexistence. Measuring intraspecific diversity requires specialised indices that can measure trait diversity among individuals. For an index to be optimal in measuring any facet (richness, evenness and divergence) of intraspecific diversity, it must; (i) be mathematically and conceptually consistent (ii) reflect intuition, and (iii) not depend on the number of individuals. We reviewed the literature and identified multi-trait indices: TOP (Trait Onion Peeling), FRic (Functional Richness), TED (Trait Even Distribution), FEve (Functional Evenness), FDis (Functional Dispersion) and Rao (Rao’s quadratic entropy). Therefore, we evaluated multi-trait richness using FRic and TOP, evenness with FEve and TED and divergence with FDis and Rao. We evaluated these indices based on the afore-mentioned criteria using simulated individual-level data in a 2-dimensional trait space. To evaluate criterion (ii), we tested how these indices responded to a random shift of 20% of the individuals in trait space (scenario one), or randomly deleting 10%-50% of the individuals at the centre (scenario two). To evaluate criterion (iii) we gradually increased the number of individuals (scenario three).
Results/Conclusions
We found two (TED and TOP) of the initial six indices to violate criterion (i). We developed modified versions of these indices (TEDM and TOPM) to correct for this. TEDM fufiled this criterion while TOPM still violated the criterion. Three of the eight indices (FEve, TOPM and FRic) were found to violate criterion (ii). FEve did not respond to small changes in location, TOPM increased while richness was originally decreasing, and FRic did not respond to changes at the centre of the trait space. Criterion (iii) was found to be violated by five (TOP, FRic, TED, TEDM and TOPM) of the eight indices. TOP and FRic continually increase richness as the number of individuals increased. TED, TOPM and TEDM also responded to changes in the number of individuals but became stable as the number of individuals increased. We conclude that criterion (ii) and (iii) are presently main challenges for the quantification of intraspecific diversity, and new indices need to be developed to address these challenges.