Effect of preparation method on composition and NIR spectra of forage samples

To evaluate the effects of freezing and drying methods on chemical composition and NIR spectral features of pasture samples, 20 samples of different swards were frozen by liquid nitrogen (LN) or conventional freezer and subsequently dehydrated by freeze drying or by forced-air oven under different conditions of time and temperature explaining seven experimental treatments. Near infrared spectra were taken and principal components (PC) computed to plot samples according to scores and visualise potential segregating effects of treatments. Several chemical constituents relevant to nutritional quality of forages were determined and effects of treatments were evaluated by one-way ANOVA. Two-way ANOVA was applied to evaluate effects of freezing (LN or freezer) and drying method (freeze drying, minimum 72 h, or oven drying at 60 °C, 48 h). One-way ANOVA detected significant effects of treatments (P<0.05) for soluble protein (SCP), neutral detergent fibre (NDF) and insoluble nitrogen in the NDF (NDFIN), but no significant effects were apparent (P>0.05) for dry matter (DM), crude protein (CP), crude fibre (CF), acid detergent fibre (ADF) and in vitro digestible organic matter content (DOMD). Two-way ANOVA showed that in comparison with freeze drying, oven drying at 60 °C for 48 h significantly reduced (relative change) SCP and DOMD, by 27 and 2%, respectively, while there was a significantly relative increase in NDF and NDFIN, by 10 and 247%, respectively. No significant effects of freezing methods were apparent. Plotting spectra according to the first two principal components showed a clear separation of freeze dried samples with respect to oven dried samples, indicating relevant spectral differences and suggesting that sample manipulation in the laboratory prior to spectral collection should be consistent to avoid errors that could reduce prediction accuracy in the NIRS analysis. It is concluded that drying method and conditions (temperature and time) can lead to changes in composition, particularly with reference to the protein fraction and digestible organic matter content. Although freeze drying could be considered the best method, cost and time will probably preclude its massive application, except for research purposes or eventually in the reference method for the development of NIR calibrations for predicting composition of undried samples.