Can wood properties be predicted from the morphological traits of a tree? A canonical correlation study of plantation-grown white spruce

Abstract

There is a growing interest in predicting wood quality from tree morphology traits, which can be measured using remote sensing techniques such as LiDAR, to enhance forest inventory for operational planning. In this study, we investigated the correlation structure between these two categories of traits in white spruce (Picea glauca (Moench) Voss) using canonical and multiple regression analyses with the objective of identifying key morphology variables that are predictive of wood quality. For 495 trees from a 30-year-old plantation, we obtained measurements of tree height and dimensions of the living crown, as well as the number and diameter of live branches at selected whorls. Wood traits were assessed from wood cores with SilviScan technology. Morphological traits explained almost 29% of the overall variation observed in wood traits. However, the magnitude of the correlations and the ability of crown morphological traits to predict wood traits differed widely among the latter. Average ring width and radial cell diameter, both related to increment, were well correlated with tree morphology, whereas traits related to subcellular structure, for instance, microfibril angle, were poorly correlated. These results could guide the choice of wood traits to improve inventory techniques aiming to optimize the forest product value chain.