The number of latent variables (LV) for PLSR models were identified from the calibration data using 10-fold cross-validation with 100 Monte Carlo replicates for cross-validation. Models were constructed using a range of LV from 1 to 50, and to avoid model overfitting, the optimal number of latent variables was determined by selecting the model with the smallest root mean square error (RMSE). The optimal model was then applied to the validation set to generate the coefficient of determination (R2) and RMSE performance statistics (Cheng and Sun, 2015). The validation slope (b), intercept (a), and bias (expected predicted − observed) were also calculated based on the regression Ymeasured = a + b × Ypredicted.