Evaluation of an international series of Pinus kesiya provenance trials for growth and wood quality traits

• Published in: Forest ecology and management 2008-05, Vol.255 (8), p.3477-3488
• Main Authors: Costa e Silva, João, Graudal, Lars
• Format: Article
• Language: English
• Subjects: Animal and plant ecology; Animal, plant and microbial ecology; Biological and medical sciences; BLUP; branches; branching; correlation; environmental factors; equations; forest trees; Forestry; Fundamental and applied biological sciences. Psychology; Generalities. Production, biomass. Quality of wood and forest products. General forest ecology; genetic variation; Genotype by environment (G × E) interaction; genotype-environment interaction; growth traits; Linear mixed models; linear models; mixed models; Pinus kesiya; plant architecture; provenance; Provenance variation; Quality of wood and forest products; stem form; Synecology; Terrestrial ecosystems; Trait correlations; tree breeding; wood quality
• ISSN: 0378-1127; 1872-7042
• DOI: 10.1016/j.foreco.2008.02.027

This study presents the results from seven international provenance trials of Pinus kesiya established in Colombia, South Africa, Swaziland, Vietnam and Zimbabwe. The traits evaluated were total volume per hectare, stem straightness, branch diameter, forking and foxtailing, at ages 5, 6, 7 or 8 years from planting. The tested material included 28 seedlots, comprising 24 provenances from the natural range of the species distribution and four local seed sources from Madagascar and Zambia. Survival was generally high, with percentages exceeding 85% in most of the trials. Significant provenance variance was found for total volume per hectare, stem straightness, branch diameter and foxtailing in all of the trials where these traits were measured. Forking exhibited significant provenance variance in only four of the seven trials measured. Across-site analysis for single traits used a factor analytic structure to model the provenance effects in each trial, and indicated that the practical importance of genotype by environment interaction was not great for the examined traits, as the majority of the provenances did not display relevant ranking changes across trial sites. Bivariate analysis across sites indicated that provenance correlations involving total volume per hectare, stem straightness and forking were statistically significant, moderate to high, and their signs suggested that simultaneous improvement could be achieved in these traits following provenance selection. However, these traits were significantly and adversely correlated with foxtailing. Provenance correlations between branch diameter and the other traits were generally low and not significantly different from zero. Vietnamese provenances, as well as local seed sources from Madagascar and Zambia, provided favourable combinations of growth and wood quality traits, and thus could have an important value for provenance research, seed supply and breeding.