Integrating intraspecific trait variability in functional diversity: an overview of methods and a guide for ecologists

Abstract

The variability in traits within species (intraspecific trait variability; ITV) has attracted an increased interest in functional ecology, as it can profoundly influence the detection of functional trait patterns, calculation of functional diversity (FD), and assessments of ecosystem functioning. This renewed focus stems from the recognition that species are not homogeneous entities but rather mosaics of individuals with varying traits. Researchers dealing with FD have increasingly recognized this issue, and consequently, multiple methods have emerged to explicitly incorporate ITV into FD calculations. Some methods treat individuals as the unit of analysis, while others characterize trait distributions around their means. Ecologists navigating this landscape of methods may face challenges in selecting the most appropriate approach to address their research questions. This synthesis provides an overview and guidelines on how and when to use the different methods available to quantify ITV in biological assemblages and integrate it into FD. As a case study, we computed correlations on simulated assemblages with varying degrees of trait variability. Our findings suggest that the choice of FD metric should be guided primarily by the ecological question being addressed and, to a lesser extent, by the number and types of traits. Simulations revealed strong correlations among FD metrics that account for ITV, particularly those indicating the size of the occupied functional trait space. As evidence accumulates and shows how ITV is key to shaping species’ fitness and distribution as well as affecting ecosystem functioning, this synthesis will serve as a conceptual and practical tool inspiring and guiding researchers to integrate ITV in functional diversity analyses.

Intraspecific variability of leaf form and function across habitat types

“…All the analyses were carried out at the 0.99 and the 0.50 quantile of the probability distribution of species in the trait space to better characterize differences when targeting the whole target dataset (0.99 quantile) or the most likely trait combinations (0.50 quantile). All the kernel-density-based analyses and visualization were implemented by combining the functionalities of the 'TPD' (Carmona et al, 2019) and the 'funspace' (Carmona et al, 2024) R packages. The differences between the BTV and ITV level metrics were tested using the as.randtest function, as specified above.…”

“…ITV is a relevant facet of trait diversity because individual-level trait variability ultimately represents the raw material for natural selection (Bolnick et al, 2011;Des Roches et al, 2018;Palacio et al, 2024;Westerband et al, 2021). However, ITV has been generally overlooked due to its smaller magnitude compared to between-species trait variation (BTV; Violle et al, 2012) as well as for practical reasons (e.g., sampling effort; Puglielli et al, 2022), especially in studies involving many taxa across large spatial scales.…”

Calculating functional diversity metrics using neighbor‐joining trees

“…In addition, if intraspecific data are available, one can build trees using individuals instead of species, by-passing this issue. Intraspecific trait data are increasingly seen as being crucial to understanding how organisms interact (Tautenhahn et al 2019, He et al 2021, Wong and Carmona 2021, Palacio et al 2024. Given that intraspecific trait data are not always available at the community level, one workaround is to simulate intraspecific variability from compound measures such as the standard deviation of a given trait, which could approximate the kernel-density approach using trees.…”