Equine Forage Analysis by NIR SpectroscopyBy Kentucky Equine Research Staff · October 20, 2014
Horse owners may want to have samples of their hay analyzed to determine carbohydrate level or other nutritional parameters. Various techniques are used to perform this analysis, one of which is near infrared reflectance (NIR) spectroscopy. This is a sophisticated analytical technique used for determining the chemical characteristics of agricultural and food products, pharmaceuticals and beverages. It is based on the fact that each of the major chemical components of a sample has near infrared absorption properties that can be used to differentiate one component from another. Once a feed or forage sample has been dried and ground, a NIR analysis can be completed in about 60 seconds, yielding results for up to 18 nutrients.
NIR forms the basis for most commercial feed analyses. Advantages of this technique are:
1. Accuracy – advancements in computer software have provided the tools to take full advantage of the technology.
2. Speed –NIR analyses can be completed in under 24 hours.
3. Cost – analyses are typically half the price of wet chemistry.
4. Labor efficiency – more analyses can be completed in a shorter period of time with less labor.
5. Safety – eliminates the use of hazardous chemicals.
6. Environmentally friendly.
Several factors influence the accuracy of NIR measurements. These include:
1. Does the component have NIR reflectance properties? Each of the major organic feed components has absorption characteristics (due to vibrations arising from the stretching and bending of H bonds associated with C, O, N) in the near infrared region that are specific to that component. NIR is most sensitive to organic compounds. Compounds lacking the above properties will not calibrate as well.
2. Robustness of the calibration set. There must be adequate variation in the population of samples used for developing the calibration. The variation must be inherently reflective of the sample population as a whole. The more closely the calibration set resembles the sample population, the better the performance of the calibration.
3. Accuracy of the reference method. A NIR measurement can only be as accurate as the reference method used to develop the calibration. For example, crude protein can be accurately measured and calibrated. The detergent fiber methods of analysis are not as precise, and therefore are less well determined by NIR.
4. Calibration updates. New varieties and hybrids are introduced every year. In order to keep calibrations current, they must be continually expanded to include new genetics. Software routines ease the process of identifying new samples for calibration expansion. Each sample has its own spectral fingerprint (spectra). Software comparisons of new spectra to existing spectra in the calibration database identify samples to add for expansion.
NIR does not measure minerals directly. Values are indirect measurements based on relationships with other components. Predicted mineral values will be better than average tabular values, but they may not be the absolute values. NIR mineral results are routinely used by the feed industry for ration formulation. Wet chemistry minerals should be substituted when precise formulation is required for exceptional circumstances. This could be for rations where the animals are not performing as expected or for high performance rations where fine tuning to the highest degree is desired.