Assessment of variations in air-dry wood density using time-of-flight near-infrared spectroscopy

Non-destructive measurement of wood density using reflectance and transmittance near-infrared spectroscopy (NIRS) has already been reported by several researchers. However, these methods do not allow an independent determination of the spectral contributions of the absorption and scattering effects. In this study, time-of-flight NIRS (TOF-NIRS) was used to measure wood density of seven softwood and hardwood species. A curve fitting procedure was used to resolve the diffusion equation and compute the absorption (mu(a)) and reduced scattering coefficients (mu'(s)) from the time-resolved profiles (TRPs) acquired from the wood samples. Statistical parameters were also computed from TRPs and correlated with the wood airdry density as well as with both optical coefficients. A simple linear regression model was built to relate air-dry wood density to mu(a) (R-2 = 0.56, root mean square error [RMSE] = 0.047 g/cm(3), p value < 0.05). Results obtained were in good agreement with the modified Beer-Lambert law for turbid media. The square root of the mu'(s)/mu(a) ratio (which was well correlated with the photons mean path length) was used to correct for the distortion due to scattering in absorbance NIR spectra acquired using reflectance spectroscopy. This research demonstrates the potential of TOF-NIRS to assess wood density and to correct reflectance spectra for scattering effects.