Functional traits are weak predictors of tree sapling growth worldwide

Functional traits, well-defined characteristics of organisms that influence or coordinate with ecological performance, promise that one may predict vital rates, such as survival and growth, by measuring a few aspects of anatomy and physiology. Functional traits have been shown to correlate significantly, though weakly, with the relative growth rate (RGR) of trees in five Neotropical forests. Here, we estimated size-standardized height RGR for 26344 saplings of 240 species from 23 sites around the world using nonlinear models of growth. We then predicted the contribution of three functional traits (specific leaf area, wood density and seed mass) to individual RGR with a hierarchical Bayesian model. The three functional traits, alone and in combination, explained a relatively small amount of variance in RGR, with variation among continents and between temperate and tropical regions. We echo the calls for researchers to shift toward the measurement of 'hard' functional traits, with clear linkages to realized vital rates, rather than 'soft' functional traits, which are easier to measure but may only indicate the theoretical maxima of vital rates. Doing so would advance ecology onto a more solidly functional foundation.