Heritability estimates for wood stiffness and its related near-infrared spectral bands in sugi (Cryptomeria japonica) clones

Abstract

Rapid and cost-effective methods of evaluating the properties of wood are required for forest-tree breeding programs. In this work, we first attempted to develop a calibration model for predicting the stiffness of wood using near-infrared (NIR) spectroscopy. Wood samples were collected from three stands of sugi (Cryptomeria japonica) plus-tree clones aged from 34 to 36 years old. Two to three sample trees were harvested from each clone, yielding a total of 129 sample trees. Partial least-squares analysis was performed to predict the dynamic modulus of elasticity (Efr). A reasonable calibration model that allowed Efr to be predicted with a correlation coefficient of 0.69 and a root mean square error of prediction of 0.82 GPa was obtained. Second, in a novel approach, we tried to select the NIR spectral bands that were the most strongly related to wood stiffness, and we examined the degree to which these bands were genetically controlled. The heritability estimates (h2) for the NIR absorbance values at 7,320 and 6,281 cm−1, which were found to be the most important bands for predicting Efr, were 0.48 and 0.57, respectively. Although the genetic control of the NIR spectral bands was weaker than the genetic control associated with the actual measurement of Efr (h2 = 0.74), the results imply that the NIR spectral bands are under moderate genetic control. Assuming that the NIR spectral bands are the dissected components of the target trait (Efr), the results should provide useful information when attempting to perform genomics-based breeding.