Genomic diversity, trait–ancestry associations and core subset optimization of Pinus radiata germplasm in New Zealand

Radiata pine ( Pinus radiata D. Don.) comprises 90% of New Zealand’s exotic plantation forests, backed by 70 years of breeding history. Maintaining large in-situ archives generated by decades of breeding is resource intensive and with fixed rotations makes germplasm management more complex than in traditional agricultural crops. Genomics tools offer a pathway to target resource management and reducing redundancy. In this study, we assembled 1,573 accessions from in-situ archives, spanning founders, landraces, and advanced lines and which are genotyped using 8,243 markers. Population structure analysis (STRUCTURE and PCA) identified an optimal cluster of K = 3, revealing low to moderate genetic differentiation ( \\(\\:{F}_{ST}\\) = 0.021–0.045), and increasing admixture across generations. Two clusters correspond to known Californian mainland (Monterey) and Island provenances, the third cluster likely represents an under-characterized mainland lineage (Año Nuevo) within NZ landraces. While mainland ancestry remains dominant, signals from Guadalupe and Cedros Islands are evident in recent breeding series. Ancestry-weighted analysis revealed contrasting trait associations among clusters, with cluster 1 favouring growth traits, cluster 2 associated with improved Dothistroma resistance, and cluster 3 exhibiting superior wood quality traits but reduced growth performance. These results may offer the opportunity to incorporate ancestry via metafounders in single-step evaluations to reduce bias and support balanced genetic gain. Furthermore, Corehunter-derived subsets captured near complete allelic representation and high diversity with minimal redundancy. This study demonstrates the value of genomics-enabled core selection for streamlining germplasm conservation in both in-situ and ex-situ archives, ensuring a resilient foundation for ongoing radiata pine improvement.