Understanding how forests respond to a changing climate is necessary for predicting future forest composition and for improving forest conservation practices. As tree species are expected to migrate to match their physiological tolerances with changing climate, tree species characteristic of lower elevations are expected to move to higher elevations to keep pace with regional warming trends; yet, the evidence from field observations is mixed. Utilizing a tree demography approach, we constructed logistic regressions based on field observations from ten mountains to calculate the spatial distributions of tree seedling versus adult conspecifics to determine potential range shifts along elevational climatic gradients for multiple tree species dominant in the northeastern United States. We tested if tree seedling distributions were generally shifted upslope relative to the conspecific adults. In addition, we tested if these mismatches differed for seedlings found in canopy gaps relative to seedlings found under closed canopies in order to gain a better understanding of how gaps may facilitate tree species distributional shifts. We used linear mixed models to quantify the proportion of variance explained by broad potential drivers of these elevational range shifts, including climate, soils (physical and chemical properties), and landform (slope, aspect, and light availability).
Results/Conclusions
The northern hardwood species Fagus grandifolia and Acer saccharum displayed significant retractions from their upper elevation range limit, although the magnitude of change in F. grandifolia was far less than A. saccharum. The high elevation conifer species Picea rubens and Abies balsamea showed significant retractions from their lower range limit. These results are consistent with previous observations that reveal increased recruitment of Fagus grandifolia in areas of low A. saccharum density. Interestingly, P. rubens seedling distribution displayed an opposite trend to previously observed downslope shifts of sapling distributions relative to conspecific adults, consistent with the idea that seedlings have been responding to more contemporary (and warmer) climate than saplings. For both conifer species the leading edge of their elevation shift was supported by gaps, possibly as a result of increased light availability. Climate was consistently a major driver of seedling spatial dynamics, but soil factors also explained a large portion of spatial variation in seedling distributions. Given the inconsistency between expected trends under ongoing climate change and trends from field observations, this study underlines the importance of underappreciated abiotic and biotic interactions to alter species ranges in ways not predictable by examining climate alone.