A trait‐based approach to understand the consequences of specific plant interactions for community structure

Question In plant communities, the presence of a species has consequences for other species, with some being competitively excluded while others benefit from the close vicinity of neighbours. Even though such specificity in plant interactions is common and known, there is no empirical assessment of the mechanisms that would help us understand its importance for plant diversity. Here we asked whether analysing spatial associations between plant traits known to affect the environment (i.e. effect traits) and those known to respond to the environment (i.e. response traits) might explain plant–plant interactions and their role in community assembly. Location Sierra Nevada Mountains, Spain. Methods In a field study, we addressed the specificity of plant–plant interactions by quantifying effect traits of three co‐occurring cushion‐forming species and response traits of their associated plant assemblages. Traits were measured at the individual level and then aggregated to trait metrics (mean, range, dispersion) at the plot level. Finally, plot‐level metrics of effect traits were related to response traits and the species composition of plant communities. Results Each cushion‐forming species had a distinctive combination of effect traits and harboured a unique plant community with an exclusive composition of response traits. With multivariate statistics we showed that differences in effect traits (branch density and canopy height) among and within cushion species significantly explained response traits (specific leaf area, leaf dry matter content) of associated species and the local‐scale species composition. Conclusions Using effect and response traits measured at the individual level, we provide a mechanistic understanding of the species specificity of plant interactions and demonstrate how important such specificity is for species diversity in an ecosystem.