A test of the explanatory power of plant functional traits on the individual and population levels

An important issue in plant ecology is the extent to which functional traits that explain patterns at one organizational level serve as explanatory variables at higher levels, while a related subject is the universality of relationships between traits and responses to environmental variables even at a given level. We addressed both questions experimentally by either mowing or abandoning plots within two meadows, one wet and one dry, and then measuring the performance of 42 species in terms of shoot growth and flowering (individual level) and species cover (population level), and relating these performance measures to traits assessed either directly on the individual level or indirectly (mostly using databases) on the species level. Of particular interest were traits thought to confer competitive advantages on individual shoots, to see if these traits were especially useful in predicting early population level responses to changed management. Our study found that (1) only one trait had predictive value for responses at both the individual and population level for the environmental alteration consisting of mowing vs. abandonment; (2) traits important to competitive ability were not particularly good predictors for responses to abandonment at the individual or population level; (3) the predictive value of particular traits was greater earlier than later after abandonment, with this relationship often site-specific; and (4) the number of significant results increased after phylogenetic correction. The limited ability of the predictive power of traits to transcend organizational level and ecological milieu suggests that trait function is highly context dependent, and implies the need for mechanistic examinations of interactions between traits and perturbation.