Patterns of genetic parameters for height in field genetic tests of Picea abies and Pinus sylvestris in Sweden

Abstract

We reviewed the genetic parameter estimates carried out from 1992 to 2006 for height increment in genetic tests of Norway spruce and Scots pine, to describe patterns of genetic variation, heritability, and genetic correlations. The material included seedling and clonal tests in Sweden, aged between 5 and 20 years. Multiple regression was used to explore relationships between parameter values and test environments. Results showed moderate narrow-sense heritabilities (\( {\hat{h}^2} \): mean =0.29 in Norway spruce; mean =0.23 in Scots pine) that decreased with test site latitude for both species. In Norway spruce, \( {\hat{h}^2} \) increased with better growth and decreased with tree age, while for Scots pine, \( {\hat{h}^2} \) increased with tree age and southward transfer. The additive genetic coefficient of variation (Open image in new window; mean 15%), in Norway spruce, decreased with growth as well as site latitude. Open image in new window in Scots pine (mean =8.5%) increased with southward transfer and more southerly test latitude. Additive and genotypic within-site genetic age-age correlations in Norway spruce were high, with mean rA and rG of 0.92 and 0.85, respectively. Corresponding across-sites estimates were on average lower. Genetic parameters were better expressed on favorable sites, at younger ages in Norway spruce and at older ages in Scots pine. The results imply that gain calculations should be based on different parameters in the two species. For maximizing genetic gain in the Swedish breeding program, testing times could be shorter for Norway spruce than for Scots pine. The investigation showed a large variation in parameter estimates from different field experiments, highlighting the importance of testing over multiple sites.