Integrating environmental and genetic effects to predict responses of tree populations to climate

Climate is a major environmental factor affecting the phenotype of trees and is also a critical agent of natural selection that has molded among-population genetic variation. Population response functions describe the environmental effect of Planting site climates On the performance Of a Single population, whereas transfer functions describe among-population genetic variation molded by natural selection For climate. Although these approaches are Widely used to predict the response of trees to climate change, both have limitations. We Present a novel approach that integrates both genetic and environmental effects into a single universal response function (URF) to better predict the influence of climate on phenotypes. Using a large lodgpole pine (Pinus controta Dougl. ex LOUd.) field transplant experiment composed of 140 populations planted on 62 sites to demonstrate the methodology, we show that the URF makes full use of data from provenance trials to: (1) improve predictions of climate change impacts oil phenotypes (2) reduce the size and cost of future provenance trials without compromising predictive power' (3) more fully exploit existing, less comprehensive provenance tests: (4) quantify and compare environmental and genetic effects of climate oil population performance; and (5) predict the performance of any population growing in any climate. Finally, we discuss how the last attribute allows the URF to be used as a mechanistic model to predict population and species ranges for the future and to guide assisted migration of seed for reforestation, restoration, or afforestation and genetic conservation ill a changing climate.