NIRS Analysis of As-Fed and Prepared Samples of Corn Silage.

Near Infrared Spectroscopy (NIRS) has a wide variety of uses including many in agriculture. NIRS is being used to analyze the organic contents of forages in a non-destructive, chemical-free manner, using reflected light in the infrared range. Samples are analyzed dried and ground. Results of as-fed samples are timely but moisture hinders the accuracy of the spectra. The objective of this study was to develop a predictive equation for as-fed samples using spectra from dried and ground samples. Approximately 100 samples of corn silage were collected from the upper Midwest (NE Minnesota, NW Wisconsin and central Iowa). Samples were scanned using a bench top near-infrared diffuse reflectance spectrometer (2600XT operating in the wavelength range 680-2600mm/1mm increment). A predictive model was developed by collecting spectra from as fed samples and then again from those same samples after drying and grinding. The two sets of spectra were then compared and a predictive equation generated for each main constituent [dry matter (DM), crude protein (CP), Phosphorus (P), Acid Detergent Fiber (ADF), Digestible Neutral Detergent Fiber at 30 hrs (DNDF30), DNDF at 48 hrs (DNDF48), Ash, Calcium (Ca), Fat, In Vitro True DM Digestibility at 30 hrs (IVTDMD30), IVTDMD at 48 hrs (IVTDMD48), Potassium (K), Lignin, Magnesium (Mg) and Starch].