Power laws and plant trait variation in spatio-temporally heterogeneous environments

A challenge: Variation is ubiquitous in nature across all spatial and temporal scales and underlies prominent ecological and evolutionary theories. Although understanding the causes and consequences of trait variation is a central goal of trait-based ecology, the scaling of trait variance across space and time (variance scaling) is unresolved. A solution: We argue that characterizing trait variance across spatio-temporal scales using a combination of prominent power laws can elucidate the role of environmental variability in trait variation and potential mechanisms driving trait patterns. In particular, the species-time-area relationship and Taylor's power law help to establish a generalizable framework for developing and testing variance scaling theory. Finally, we outline priority research questions and tractable systems for answering them. Successional forests, long-term forest monitoring networks and censuses of short-lived taxa are ideal for coupling high-resolution environmental data with measurements of trait variance across scales to test the models proposed here. Main conclusions: Characterizing the behaviour of variance across spatio-temporal scales is feasible and a prerequisite for developing a predictive theory of trait-based ecology.