To validate multivariate regression models, the performance of regression models was evaluated according to the ratio of prediction to deviation (RPD), the root mean square error of calibration (RMSEC) and prediction (RMSEP), and the coefficient of determination of calibration (R2c) and prediction (R2p). The RPD is defined as the ratio of standard deviation (SD) of the reference values over the root mean square error of prediction (RPD = SD/RMSEP). If the RMSEP is smaller than SD, a relatively higher RPD value appears, indicating that the calibration model is considerably robust