Stacked distribution models predict climate-driven loss of variation in leaf phenology at continental scales

Integrating stacked distribution models with genetically-based trait diversity of Populus angustifolia reveals how phenological variation can change differently across populations in response to climate change. Climate change is having profound effects on species distributions and is likely altering the distribution of genetic variation across landscapes. Maintaining population genetic diversity is essential for the survival of species facing rapid environmental change, and variation loss will further ecological and evolutionary change. We used trait values of spring foliar leaf-out phenology of 400 genotypes from three geographically isolated populations of Populus angustifolia grown under common conditions, in concert with stacked species distribution modeling, to ask: (a) How will climate change alter phenological variation across the P. angustifolia species-range, and within populations; and (b) will the distribution of phenological variation among and within populations converge (become more similar) in future climatic conditions? Models predicted a net loss of phenological variation in future climate scenarios on 20-25% of the landscape across the species' range, with the trailing edge population losing variation on as much as 47% of the landscape. Our models also predicted that population's phenological trait distributions will become more similar over time. This stacked distribution model approach allows for the identification of areas expected to experience the greatest loss of genetically based functional trait variation and areas that may be priorities to conserve as future genetic climate refugia.

Automated summary

Integrating stacked distribution models with genetically-based trait diversity of Populus angustifolia reveals different impacts of climate change on phenological variation across populations. Results indicate that climate change will lead to a reduction in phenological variation and convergence of phenological trait distributions among populations.