Trait filtering in island floras: A conceptual framework

Abstract

Aim Dispersal and environmental filtering processes affect plant species colonisation success on islands and can be identified by functional traits. However, the lack of synthesis about the different methodological approaches in functional ecology hampers generalisation of filtering processes across island systems. Location Seventy islands of the Houtman Abrolhos archipelago, Western Australia. Major Taxa Studied Angiosperms. Methods We (i) apply a simple, conceptual framework based on the mean and variability of individual functional traits in plant assemblages to identify species filters on islands, (ii) illustrate how trait distributions of island assemblages change in relation to island area and their source pool, (iii) compare distributions of individual traits to multivariate functional diversity indices and trait spaces and (iv) provide guidelines to detect a signal of trait filtering in island floras. Results The island assemblages showed evidence for selective filters operating on seed mass and marginally on leaf area but not on plant height. Mean and variability of seed mass differed to those of the source pool indicating selective forces operating between source pool and island assemblages, especially on smaller islands. Multivariate functional diversity indices and trait spaces failed to reveal filtering processes acting on the island assemblages and insights into the putative processes. Main Conclusions Using the mean and variability of individual traits in plant assemblages provides direct information on the trait composition of island floras and the processes involved beyond what can be inferred from multivariate functional diversity indices or trait spaces. We used islands as their distinct boundaries and relatively simple sets of species provide good research models, but joint analyses of trait means and variability should also be applicable to understand filtering processes in isolates and habitat fragments on mainlands.

Bibliographic Information