Plant functional traits predict vegetation response to two decades of simulated climate change
Plant strategy theory has emerged as the key link between plant evolution and ecosystem functioning, but it remains unclear whether primary axes of functional differentiation are relevant to the continuing shift of vegetation in a warmer world. We analyzed shifts in species composition resulting from two decades of simulated winter warming and summer drought in northern England, UK, initiated in 1993 on steep limestone grassland. Previous studies at this site have demonstrated a surprising level of fine-scale spatial shifts among species across shallow and deep soil profiles over two decades in response to climate manipulation, but the physiological basis for such shifts, and whether species are responding primarily to the direct effects of environment or to changes in competitive intensity, remain unknown. We subjected 23 species to lab-based functional trait assays and used a hierarchical model to analyze species responses as a function of traits across treatments. We hypothesized that drought responses would be associated with a general stress tolerance strategy and more apparent in shallow soils, while heating responses would be associated with competitive strategies that are more advantageous in deeper soils and in plots experiencing a longer growing season.
Species were well differentiated with respect to an overall ‘resource axis’, with species of high tissue investment on one end and high growth capacity on the other. Species that increased after two decades of extended growing seasons due to winter warming were strongly associated with the competitive end of the resource axis, and this effect was amplified in deeper soils. Conversely, species that increased after two decades of chronic summer drought were associated with high tissue investment and low growth potential, and were among those most likely to decrease in the warming treatment. Soil depth had less of an effect on the responses of these conservative species. We conclude that species responses to shifts in temperature and precipitation are surprisingly predictable from their functional traits, but only in the context of both stress tolerance and competitive ability. Such changes are likely to have strong ramifications for ecosystem carbon and nutrient budgets.