Genetic parameters and predicted selection responses for timber production traits in a Castanea sativa progeny trial: developing a breeding program

Abstract

Total height, diameter, index volume, stem straightness, apical dominance, and survival were assessed at 8 years from seed in an open-pollinated progeny test of 36 families of European chestnut (Castanea sativa Miller) established at two sites in the Atlantic area of Galicia, Spain. Iterative spatial analysis was applied to eliminate the effect of the spatial dependence in the original data and to estimate accurately genetic parameters for evaluating the potential for selection of the measured trees. Spatial analysis was very beneficial for growth traits and survival, but less so if at all for form traits. Estimated individual heritabilities ranged from moderate to high for growth traits (\( \widehat{h}_i^2 = 0.29 - 0.42 \)) and stem straightness (\( \widehat{h}_i^2 = 0.{24} - 0.{42} \)). High coefficients of additive genetic variance were obtained for volume (\( \widehat{\text{C}}{{\text{V}}_{\text{A}}} = {36}.{5} - {41}.{5}\% \)) and straightness (\( \widehat{\text{C}}{{\text{V}}_{\text{A}}} = {44}.{26} - {53}.{84}\% \)). Phenotypic and estimated genetic correlations between growth traits were very high, and correlations between sites indicated that there was no important family × site interaction. No adverse correlations between traits were evident. The results indicate the ample potential for selection in the current progeny trial, where responses to within-family and combined selection for growth traits may be high. Accordingly, three selection scenarios were addressed with the aim to initiate the selection of individuals for implementing the Forest Breeding Plan of Galicia for European chestnut.