Early genetic selection for wood stiffness in juvenile Douglas-fir and western hemlock

Abstract

Wood stiffness, or modulus of elasticity (MOE), is an important property of structural wood products. The ability to genetically improve wood stiffness has increased because of the availability of tools that can measure acoustic velocity (AV, an indirect measure of MOE) on standing trees. Our goals were to optimize methods for measuring AV on young (6- to 12-year-old) Douglas-fir and western hemlock, and then use these methods to understand the genetics of wood stiffness. In Phase I, we evaluated four standing-tree tools (TreeSonic-TS, TreeSonic-SD02, Microsecond Timer, and Ultrasonic Timer) by comparing AV2 measured on standing trees to AV2 measured on logs cut from these trees (i.e., using the HM200). We found that measuring across a whorl of branches had little adverse effect on AV measurements, and that placing the acoustic sensors on the same-face of the tree was slightly better than placing them on opposite faces. In Phase II, we used full-sib heritabilities (hfs2) to compare the TreeSonic-SD02 and Microsecond Timer, sensor orientations (same-face versus opposite-face), and flight path adjustments. Because the effect of standing-tree tool and sensor orientation was non-significant, we recommend that breeders use the TreeSonic-SD02 and same-face orientation for practical reasons. Using this approach, hfs2 for AV2 was 0.90 to 0.56 in Douglas-fir and 0.51 in western hemlock, and we concluded that genetic gains of 7.0–17.2% can be obtained by measuring 10 or more trees per full-sib family, depending on species and seed production strategy. Because we found no strong statistical support for substantial non-additive genetic variation, the current approach of breeding for additive genetic value, and then collecting open-pollinated seed from orchards seems appropriate for improving wood stiffness. AV2 had weak genetic correlations with growth traits (height, diameter at breast height, and volume) in Douglas-fir (−0.21 to 0.33), but moderately positive correlations in western hemlock (0.36–0.54). Therefore, there is ample opportunity to improve the stiffness of young trees without adversely affecting growth. Although mean stiffness was higher for Douglas-fir, about 10% of the western hemlock families (7 out of 80) had higher AV2, and about 28% of the families (22 out of 80) had higher dynamic MOE than the mean for Douglas-fir. This suggests that intensive selection and breeding in western hemlock could be used to create varieties of trees that compare favorably with unselected Douglas-fir. This may improve the ability of western hemlock to compete with Douglas-fir where stiffness is important.