Predictions of wood density and module of elasticity of balsam fir (Abies balsamea) and black spruce (Picea mariana) from near infrared spectral analyses

Abstract

The predictions of properties for wood disc average are seldom reported, and they are important for sorting out logs based on their quality. The minimum near infrared (NIR) spectra required to predict wood disc average properties would also be of critical importance. In this study, calibration and prediction models for wood disc average properties were developed using NIR spectral data for balsam fir (Abies balsamea (L.) Mill.) and black spruce (Picea mariana (Mill.) B.S.P.) samples collected from 14 different sites across Newfoundland, Canada. The calibration was done against area-weighted average wood properties determined by SilviScan. NIR spectra were collected in 18 mm increments from the radial–longitudinal face of green and oven-dried samples. Results showed that using NIR spectra from three spots per wood strip was sufficient for the modeling and prediction for density and module of elasticity (MOE). The coefficients of determination ranged from 0.76 (MOE of green wood samples) to 0.88 (density of oven-dried wood samples). However, the microfibril angle (MFA) cannot be well predicted from either green wood or oven-dried wood NIR spectra. Our results further showed that the NIR spectra collected from oven-dried wood samples gave better calibration and prediction than those collected from green wood samples.