Near infrared spectroscopy as a tool for in-field determination of log/biomass quality index in mountain forests

Current in-field methods for grading logs are based on visual rating scales, which are subjective, operator-dependent, and time-consuming. Different wood defects such as knots, resin pockets, rot, compression wood, amongst others, affect the quality and potential usage of the log. Early detection of these defects and an adequate wood quality classification helps to optimize the resource use along the whole production chain. Therefore, the specific target for the development of an efficient in-field grading approach was defined within the project Integrated proceSsing and controL systems fOr sustainable forest Production in mountain areas - SLOPE. The grading is conducted by means of automatic measurements of selected wood properties with diverse sensors, including NIR spectrometers. A series of studies was conducted on wooden discs using laboratory equipment and a portable NIR spectrometer. In-field measurements on standing trees and harvested logs were also performed using a portable instrument. PCA models for identification of logs defects were developed using the spectra collected with both instruments. Such models will serve for the automated determination of quality indexes to be used for log grading. It is foreseen that the NIR-based quality indexes will be integrated with the expert system under development within the SLOPE project and combined with quality information derived from other sensors. The overall goal is to provide a reliable technology for automatic log quality grading in the forest industry.