What do ecological regions tell us about wood quality? A case study in eastern Canadian white spruce

Abstract

There is a growing interest in improving knowledge on wood quality of forest resources to allow for an end-product oriented supply for the wood-processing industry. We investigated the differences of white spruce (Picea glauca (Moench) Voss) wood traits from 13 ecological regions of the eastern Canadian province of Quebec and identified climatic and geographic variables and tree characteristics that may help predict wood traits in novel approaches to forest inventories. One hundred and eleven wood disks were used to determine wood traits that are related to wood anatomy and to mechanical use of wood. Our results indicate significant differences between ecological regions for most of the investigated wood traits. Samples from several coastal regions showed, for example, wood anatomy that is advantageous for some pulp and paper grades, but they offer comparatively weak estimates for mechanical applications. Ordination and projection to latent structures allowed us to identify maximum temperature, degree days, geographic location, tree height, and tree diameter as important predictor variables for wood traits. The quality of models was variable and R2 values ranged from 0.13 to 0.59. Prediction errors in an independent sample were mostly below 10%, and bias was below 5%. Although the prediction of wood mechanics was less precise, the presented results are an encouraging step toward obtaining a wood trait signature for different environments.