Effect of density and drying methods on the spatial variation of moisture in Eucalyptus sp. wood estimated by NIR spectroscopy

Portable near-infrared (NIR) spectroscopy enabled the nondestructive estimation of internal moisture gradients in Eucalyptus sp. wood subjected to natural and kiln drying. Freshly cut wood was processed into beams, and samples were collected radially, from the heartwood to the sapwood. NIR spectra were obtained using a portable spectrometer, while moisture content was determined gravimetrically at levels of approximately 40%, 30%, and 20%. Partial least squares regression (PLS-R) models were developed to predict the moisture distribution in the wood. The models successfully detected the spatial variation of moisture in the radial surfaces, revealing persistent internal moisture gradients in both drying methods. Wood density strongly influenced prediction performance; denser samples showed greater agreement between NIR estimates and laboratory measurements. The results demonstrate that portable NIR spectroscopy is a promising tool for the rapid, nondestructive monitoring of internal moisture distribution during wood drying, with potential applications in process control and quality assessment.