Development and evaluation of models for specific gravity of Eucalyptus tereticornis wood by fourier transformed near infrared spectroscopy and partial least squares regression analysis

Specific gravity is very important factor in determining the economics of wood and wood products. This study is aimed to apply the near infrared spectroscopy technique with integrating sphere accessory for fast prediction of specific gravity of dry timber (moisture content 12 %) of Eucalyptus tereticornis from varying age group. Specific gravity was determined by conventional method and correlated with near infrared spectra using partial least square regression. The coefficient of determination of cross validation (R cv 2 ) is in the range of 0.80-0.91 for radial face and 0.88-0.94 for tangential face. The calibration equations when applied to test set resulted in coefficient of determination (R TS 2 ) ranging from 0.68 to 0.93 for radial face and 0.90 to 0.94 for tangential face. The ratio performance of deviation was calculated as the ratio of the standard deviation of the prediction set to the root mean square error of prediction and it was found to be between 2.1 and 3.8 for radial face and 3.2-4.1 for tangential face. The values were found identical when groups i.e., calibration and test set were interchanged. The range error ratio for both faces models were also identical and indicating that models are quite satisfactory for screening purpose. © 2012 Indian Academy of Wood Science.