UC Santa Barbara, Ecology, Evolution, and Marine Biology, United States
Abstract: For long-lived organisms like trees, many species will likely be unable to migrate or adapt quickly enough to keep pace with climate change. Yet individuals may be able to acclimate to rapid environmental change through phenotypic plasticity. However, the acclimation potential of tree species’ is poorly understood, and this knowledge is crucial for predicting forest responses to climate change. Estimating phenotypic plasticity in mature trees can be achieved by measuring temporal (i.e., interannual) variation in functional traits. Measuring spatial trait variation is easier (or at least faster) and may be a reasonable proxy for temporal acclimation. However, spatial variation may overpredict acclimation potential by including ecotypic variation and ontogenetic plasticity or underpredict acclimation potential by capturing spatial variation in mean climate that is smaller than interannual climatic variation. The goals of this meta-analysis are to investigate tree species' acclimation potential via spatial and temporal intraspecific variation in functional traits. Our approach used multi-year trait datasets from published journal articles, National Ecological Observatory Network (NEON), and Long Term Ecological Research (LTER) sites to address two questions: 1) What is the extent of spatial versus temporal variation in functional traits?; 2) Can temporal trait variation be predicted by trait change along geographic environmental gradients? Results from our meta-analysis indicate strong but context-dependent variation in both spatial and temporal scales according to geographic scale, species, and clade (gymnosperm or angiosperm). In general, temporal coefficients of variation (CV) for traits were higher when spatial gradients were smaller but were often high even when spatial gradients covered a species’ entire range (e.g., elevation distribution). For co-existing species, there were distinct interspecific differences in spatial and temporal trait CVs which indicate the importance of genetics and life history on functional trait expression. For example, in the San Juan Mountains of Colorado, USA, aspen (Populus tremuloides) exhibited much higher temporal trait CV than co-occurring ponderosa pine (Pinus ponderosa). Overall, deciduous angiosperms tended to exhibit higher temporal trait CV, yet disentangling the effects of average versus current year leaf traits for gymnosperm species could alter these findings. Overall, species’ spatial gradients were poor predictors of the magnitude of temporal trait variation. Continually improving trait datasets with more years of observations may reveal clearer emergent patterns, but multi-year trait datasets remain extremely rare. In conclusion, these results emphasize the importance of long-term monitoring and inclusion of relevant physiological traits in measurement protocols for assessing tree response to climate change.