Abstract
This study focused on the use of Near InfraRed (NIR) Spectroscopy to address the lack of tools and skills for wood identification of Dalbergia species from Madagascar. Two sample sets of 41 wood blocks and 41 wood cores belonging to four Dalbergia species (D. abrahamii, D. chlorocarpa, D. greveana, and D. pervillei) were collected in the northern and western regions of Madagascar. Sapwood and heartwood NIR spectra were measured on wood at 12% moisture content by using a portable VIAVI MicroNIR 1700 spectrometer. Four discrimination models corresponding to sapwood and heartwood of the two sample forms were developed using Partial Least Square Discriminant Analysis (PLSDA). Good accuracy of 83.3% and 81.8% were obtained from the heartwood-based PLSDA models respectively for wood blocks and wood cores samples. All D. chlorocarpa samples were well-classified by the two models. Results highlighted the potential of portable NIR Spectroscopy as a helpful tool to support sustain-able management and trade of Madagascar's Dalbergia species. Further studies are, however, needed for its operational use in identification routine.