1：Experimental Design and Method Selection:

The study used near infrared (NIR) spectroscopy calibrated with extractives content (EC) determined by accelerated solvent extraction (ASE). Partial least squares regression (PLSR) and PLS-discriminant analysis (PLS-DA) models were employed for prediction and discrimination.

2：Sample Selection and Data Sources:

216 heartwood samples from Eucalyptus argophloia, E. bosistoana, and E. globoidea were collected from various locations in New Zealand. Samples were processed into powder with a Wiley mill and conditioned to constant moisture content.

3：List of Experimental Equipment and Materials:

NIR spectrometer (Tensor 37, Bruker), fiber-optics probe (Model N-500, Bruker), Wiley mill (Thomas Scientific), accelerated solvent extractor (ASE, Thermo Scientific), ethanol, aluminum containers, stainless-steel cells, desiccators, oven.

4：Experimental Procedures and Operational Workflow:

NIR spectra were recorded from 9000 to 4000 cm?1 with 32 scans per sample. Samples were then oven-dried, extracted with ASE using ethanol, and EC was calculated on a dry mass basis.

5：Data Analysis Methods:

Data analysis was performed using R software with packages for spectral manipulation (prospectr), variable selection (plsVarSel), PLSR and PLS-DA (caret and pls). Kennard-Stone sampling was used to split data into calibration and validation sets, and Savitzky-Golay algorithm for spectral derivatives.