Abstract: To investigate the genetic variation patterns of multiple traits in Pinus sibirica half_sibling families introduced to the Greater Khingan Range, this study aims to establish a comprehensive trait evaluation system based on genetic parameters and identify fast-growing, high-quality genetic materials. The findings provide scientific support for advanced-generation seed orchard development and multi-objective genetic improvement. A total of 66 half-sibling families from a 26-year-old trial plantation of the Tomsk seed source were evaluated. Eleven traits were measured, including growth traits (tree height, diameter at breast height, volume, and crown width), morphological traits (lateral branch diameter), wood quality traits (Pilodyn value), and needle traits (fresh weight, dry weight, moisture content, needle length, and needle width). Genetic parameters were estimated using variance component decomposition. Superior families with favorable performance in both growth and wood density traits were identified using Best Linear Unbiased Prediction (BLUP) weighted by genetic correlation coefficients. Additionally, individual tree selection was conducted based on growth traits using the index selection method. Significant genetic differences among families (Z ratio > 1.50) were observed for 10 traits, including growth, wood density, and needle traits. The phenotypic coefficient of variation (PCV: 5.05–62.50%) and genetic coefficient of variation (GCV: 2.19–13.81%) exhibited substantial heterogeneity. Volume displayed the highest variation (PCV = 62.50%, GCV = 13.81%), while the highest family heritability was observed for the needle length-to-width ratio (h² = 78.30%), and the highest individual heritability was recorded for needle moisture content (H² = 95.22%). Genetic correlations analysis revealed strong positive associations between volume and tree height (r = 0.880), diameter at breast height (r = 0.968), and Pilodyn value (r = 0.508). Using the BLUP method, 13 superior families (e.g., Families 59, 11, and 51) were identified, with an average genetic gain in volume of 13.377% and a family retention rate of 70%, significantly exceeding the population mean (65.10%). Through index selection, 94 elite individual trees were selected, 52.14% of which originated from superior families. The genetic gain in individual tree volume reached 26.80%, with the within-family gain for elite individuals increasing to 28.47%. This study establishes the first multi-trait genetic evaluation model for P. sibirica and proposes a “family-individual” collaborative selection strategy. The selected superior families achieved a volume genetic gain of 3.864–32.307% and an overall genetic gain of 2.729–20.069%, while elite individual trees exhibited a volume genetic gain of 16.328–52.716%. These findings would provide critical technical support for multi-objective breeding and seed orchard development in cold-temperate coniferous species.

Key words: Pinus sibirica, Half-sib family, Genetic parameter estimation, Selection of families and single plants