Prediction of fiber fractions, ash and organic matter digestibility in untreated maize stover by near infrared reflectance spectroscopy

Fassio, A., Restaino, E., Cozzolino, D.


Original Research Article,
Agriculture Economics
and Rural Development


Near infrared reflectance (NIR) spectroscopy was used to predict acid and neutral detergent fibre (ADF and NDF), ash, and organic matter digestibility (OMD) in maize stover samples. Samples were analyzed by reference laboratory methods and spectra collected using a NIR spectrophotometer in reflectance (1100 – 2500 nm). Predictive equations were developed using partial least squares (PLS) with full cross validation using second derivative. The coefficient of determination in calibration (Rcal) and the standard error in cross validation (SECV) were for ADF 0.85 (SECV: 16.5 g kg-1 DM), for NDF 0.92 (SECV: 58.4 g kg-1 DM), for ash 0.90 (SECV: 1.9 g kg-1 DM) and for OMD 0.85 (SECV: 10.4 g kg-1 DM). The results from this study suggested that maize stover samples might be analysed by NIR spectroscopy in order to determine ADF, NDF and ash. However, the prediction accuracy is less than desirable to be used for routine analytical purposes.