Prediction of the feeding value of grass silages by chemical parameters, in vitro digestibility and near-infrared reflectance spectroscopy

Abstract

Sixty-four grass silages with known in vivo organic matter digestibility (VOMD) were used to compare the potential of chemical parameters (dry matter (DM), crude protein (CP), crude fibre (CF), crude fat, ash, neutral detergent fibre (NDF), acid detergent fibre (ADF), acid detergent lignin (ADL), water-soluble carbohydrates (WSC)), in vitro digestibility (with rumen fluid (RFOMD) or commercial cellulases (COMD)) and near-infrared reflectance spectroscopy (NIRS, 1100–2500 nm) in predicting VOMD and calculated metabolizable energy (ME) and net energy for lactation (NEL). Further, the possibilities of NIRS to predict chemical composition and calculated protein values (digestible protein in the intestine and degraded protein balance) were investigated. NIRS equations, developed by partial least-squares analysis, were tested on a second set of 36 grass silages with digestibility and energy values based on COMD. VOMD showed the highest correlation with NIRS-predicted VOMD ( r = 0.89), followed by COMD (0.83), RFOMD (0.81) and ADL (−0.73). Combining RFOMD with DM and CP could explain 78% of the variance in VOMD, whereas COMD in combination with CP, CF and DM accounted for 84% of the variance. Best multiple linear regressions based on RFOMD or COMD had residual standard deviations of 0.32 MJ kg −1 DM for ME and 0.23 MJ kg −1 DM for NEL. About the same accuracy for the prediction of the energy value could be reached with NIRS-predicted VOMD and determined ash content. The potential of NIRS to predict the chemical composition of grass silages was highest for CP, followed by WSC, moisture, CF, ash, NDF, ADF and ADL. NIRS also seems to have prospects for protein evaluation.