Genomic Prediction Enables Provenance-Aware Selection in 1 Sessile Oak (Quercus petraea) using Foliar Physiological Traits

Climate change is reshaping the adaptive landscape of forest ecosystems, demanding more efficient strategies to identify and deploy resilient tree genotypes. Genomic prediction offers a powerful framework to accelerate selection for complex physiological traits underlying climate adaptability in long-lived species such as sessile oak (Quercus petraea (Matt.) Liebl.). Here, we conducted genomic prediction for three key physiological traits carbon isotope composition, nitrogen isotope composition, and the carbon-to-nitrogen content ratio (C/N ratio) measured in 746 trees genotyped with dense genome-wide markers (~580,000 SNPs). High genomic heritabilities were estimated across traits, with within-year prediction accuracies (Pearsons r between genomic estimated values and observed phenotypes) reaching 0.77. Notably, across-year and across-provenance predictions remained substantial (0.41 < r < 0.82), with predictability declining with increasing genetic distance (FST) between training and test provenances for nitrogen isotope composition and C/N ratio. In addition, GWAS-guided SNP preselection increased heritability capture by ~15% relative to random SNP subsets. And, the pronounced provenance-by-environment interactions observed indicated substantial phenotypic plasticity in these traits. These results demonstrate the strong potential of applying genomic prediction to foliar physiological traits as polygenic predictors for climate adaptation in plants, support provenance-aware breeding to improve forest establishment, and provide practical strategies for deploying genomic prediction in long-lived species.