Assessing vulnerability of functional diversity to species loss: a case study in Mediterranean agricultural systems

1. Increasing land-use intensification is leading to biodiversity losses world-wide, which can reduce the functioning of ecosystems. However, it is increasingly clear that not all species are equally important for ecosystem processes: whereas the loss of a functionally unique species may reduce the capacity of the community to perform some functions, losing a functionally redundant species should have a much smaller impact. 2. Assessing the vulnerability of functional diversity (FD) to species extinctions can help to predict the impacts of land-use intensification. This approach consists in ranking species according to their risk of extinction and then estimating the trajectory followed by FD as species are lost from local communities. 3. However, the most widely used FD indices are not independent of species richness, being much more sensitive to the loss of species in species-poor than in species-rich sites. This may result in misleading interpretations, affecting our ability to rank communities according to the vulnerability of their FD to species loss, by confounding it with the initial level of species richness. Here, we propose comparing the trajectory of FD under the most plausible order of species loss with that followed under random species losses as an effective way to remove the trivial effect of species richness in the assessments of vulnerability to species loss. 4. After decoupling vulnerability from species richness, we used it to analyse the effect of agricultural intensification on the vulnerability of arable plant communities in Mediterranean agricultural fields. Our results show that management strategies aiming to increase the functionality of these systems should focus on intermediately intensified fields, where small reductions in the level of intensification are likely to benefit arable plant diversity, increasing the number of species and FD and decreasing the vulnerability of FD to species losses. 5. Removing the effect of species richness is essential to attain unbiased estimations of the vulnerability of communities to species loss, especially when species-poor communities are considered. Combining vulnerability with information on taxonomic and functional diversity appears as a promising tool to inform decision-making processes, anticipating the effects of local extinctions.