Near-Infrared Spectroscopy for Discrimination of Charcoal from Eucalyptus and Native Cerrado Species-Contribution to a Database for Forestry Supervision

Abstract

The illegal charcoal trade is an internationally well-known forest crime. In Brazil, government agents try to control it using the document of forest origin (DOF). To confirm a load's legality, the agents must compare it with the declared content of the DOF. However, to identify charcoal is difficult even for specialists in wood anatomy. Hence, new technologies would facilitate the agents' work. Near-infrared spectroscopy (NIR) provides a rapid and precise response to differentiate carbonized species. Considering the rich Brazilian flora, NIR studies are still underdeveloped. Our work aimed to differentiate charcoals of seven eucalypts and 10 Cerrado species based on NIR analysis and to add information to a charcoal database. Data were collected with a spectrophotometer in reflectance mode. Partial least square regression with discriminant analysis (PLS-DA) and a linear discriminant analysis (LDA) was applied to confirm the performance and potential of NIR spectra to distinguish native Cerrado species from eucalyptus species. Wavenumbers from 4,000 to 6,000 cm-1 and transversal surface presented the best results. NIR had the potential to distinguish eucalypt charcoals from Cerrado species and in comparison to reference samples. NIR is a potential tool for forestry supervision to guarantee the sustainability of the charcoal supply in Brazil and countries with similar conditions.