Search

Summary Studies have explored how traits separate plants ecologically and the trade‐offs that underpin this separation. However, uncertainty remains as to the taxonomic scale at which traits can predictably separate species. We studied how physiological traits separated three Pinus (Pinus banksiana, Pinus resinosa, and Pinus strobus) species across three sites. We collected traits from four common leaf and branch measurements (light–response curves, CO2–response curves, pressure–volume curves, and hydraulic vulnerability curves) across each species and site. While common, these measurements are not typically measured together due to logistical constraints. Few traits varied across species and sites as expected given the ecological preferences of the species and environmental site characteristics. Some trait trade‐offs present at broad taxonomic scales were observed across the three species, but most were absent within species. Certain trade‐offs contrasted expectations observed at broader scales but followed expectations given the species' ecological preferences. We emphasize the need to both clarify why certain traits are being studied, as variation in unexpected but ecologically meaningful ways often occurs and certain traits might not vary substantially within a given lineage (e.g. hydraulic vulnerability in Pinus), highlighting the role a trait selection in trait ecology. [ABSTRACT FROM AUTHOR]

Summary: Functional traits offer a promising avenue to improve predictions of species range shifts under climate change, which will entail warmer and often drier conditions. Although the conceptual foundation linking traits with plant performance and range shifts appears solid, the predictive ability of individual traits remains generally low. In this review, we address this apparent paradox, emphasizing examples of woody plants and traits associated with drought responses at the species' rear edge. Low predictive ability reflects the fact not only that range dynamics tend to be complex and multifactorial, as well as uncertainty in the identification of relevant traits and limited data availability, but also that trait effects are scale‐ and context‐dependent. The latter results from the complex interactions among traits (e.g. compensatory effects) and between them and the environment (e.g. exposure), which ultimately determine persistence and colonization capacity. To confront this complexity, a more balanced coverage of the main functional dimensions involved (stress tolerance, resource use, regeneration and dispersal) is needed, and modelling approaches must be developed that explicitly account for: trait coordination in a hierarchical context; trait variability in space and time and its relationship with exposure; and the effect of biotic interactions in an ecological community context. [ABSTRACT FROM AUTHOR]