Genetic analysis of fiber dimensions and their correlation with stem diameter and solid-wood properties in Norway spruce

Zhi-Qiang Chen,Sven-Olof Lundqvist,María Rosario García Gil,Lars Olsson,Bo Karlsson,Tommy Mörling,Ewa J Mellerowicz,Harry X Wu

doi:10.1007/s11295-016-1065-0

Abstract

Adverse genetic correlations between growth traits and solid-wood, as well as fiber traits are a concern in conifer breeding programs. To evaluate the impact of selection for growth and solid-wood properties on fiber dimensions, we investigated the inheritance and efficiency of early selection for different wood-fiber traits and their correlations with stem diameter, wood density, modulus of elasticity (MOE), and microfibril angle (MFA) in Norway spruce (Picea abies L). The study was based on two large open-pollinated progeny trials established in southern Sweden in 1990 with material from 524 families comprising 5618 trees. Two increment cores were sampled from each tree. Radial variations from pith to bark were determined for rings 3–15 with SilviScan for fiber widths in the radial (RFW) and tangential (TFW) direction, fiber wall thickness (FWT), and fiber coarseness (FC). Fiber length (FL) was determined for rings 8–11. Heritabilities based on rings 8–11 using joint-site data were moderate to high (0.24–0.51) for all fiber-dimension traits. Heritabilities based on stem cross-sectional averages varied from 0.34 to 0.48 and reached a plateau at rings 6–9. The “age-age” genetic correlations for RFW, TFW, FWT, and FC cross-sectional averages at a particular age with cross-sectional averages at ring 15 reached 0.9 at rings 4–7. Our results indicated a moderate to high positive genetic correlation for density and MOE with FC and FWT, moderate and negative with RFW, and low with TFW and FL. Comparison of several selection scenarios indicated that the highest profitability is reached when diameter and MOE are considered jointly, in which case, the effect on any fiber dimension is negligible. Early selection was highly efficient from ring 5 for RFW and from rings 8–10 for TFW, FWT, and FC.

Highlights

All the variables calculated using cross-sectionally weighted averages for rings from pith to bark were larger than those calculated using arithmetic averages from rings 8–11, except for the coefficient of phenotypic variation (CVp) for tangential fiber width (TFW) and heritability for radial fiber width (RFW) (Table 1)
High type-B genetic correlations between sites were found for fiber-dimension traits, indicating that G × E interactions of these traits are unimportant for Norway spruce breeding in southern Sweden
Genetic correlations observed between stem diameter or solid-wood properties and RFW, fiber wall thickness (FWT), or fiber coarseness (FC) were moderate to high, and the correlations between these traits and the rest of the fiber dimensions were low

Summary

Introduction

Karst.), stem growth, straightness, and branch angle are commonly assessed as breeding targets based on their economic value (Rosvall et al 2011). Breeding for growth is a concern, as it could result in adverse effects on solid-wood and fiber dimensions (Rozenberg and Cahalan 1997). Negative genetic correlations are commonly observed between annual ring width (as a measure of stem diameter growth) and wood density or modulus of elasticity (MOE) in Norway spruce (Chen et al 2014; Hannrup et al 2004; Gräns et al 2009; Rozenberg and Cahalan 1997) and other conifers (Baltunis et al 2007; Hong et al 2014; McLean et al 2016; Wu et al 2008).

Objectives

Methods

Results

Discussion

Conclusion