In the first two parts of this series we discussed the effects of moisture, particle size and instrument type on the accuracy of near infrared (NIR) predictions. We also established that current on-farm NIR technology can measure moisture values and detect changes in other nutrients. In the third and final part of this series we’ll look at where it may or may not make sense to utilize on-farm NIR

Obvious wins

Obvious “win” senarios for on-farm NIR are cases where there is a lot of variation, and great value in quantifying that variation in a short amount of time. A prime example would be a dairy feeding a high haylage diet out of bags.

Is there a role for IR on the farm?

Ohio State University recently published findings from a study concerning the variation of feeds within a farm (see Chart 1). It clearly shows that day–to-day variation in haylage dry matter can be very high.

Here, measuring moisture values using an on-farm NIR instrument seems to make a lot of sense, because the values can be obtained and used in real time. While there are other methods available for measuring moisture, most take an hour or per sample. In contrast, on farm NIR could allow a feeder to measure several samples within minutes before feeding or even make adjustments between batches.

Obvious fails

Obvious “fail” senarios for on-farm NIR are cases where there is very little variation or precise nutritional values are required. For example, a dairy feeding corn silage from a large bunker that is faced from top-to-bottom prior to feeding. In this case, most of the variation in the bunker is eliminated when the various layers on blended together during facing.

Is there a role for IR on the farm?

The University of Minnesota published a trial (see Chart 2) from a similar senario and compared the results of an on-farm NIR system to oven drying samples once per week. Over 9 weeks, this study showed no statistical difference in any animal performance parameters.

In hindsight, these results are not surprising, because the primary forage in the diet – corn silage – showed very little variation throughout the study. In fact, the dry matter content from the weekly oven drying measurements always fell within about a 2% range, which is approximately the same as the accuracy limitations for this measurement by on farm NIR

These are just two examples of where application of on-farm NIR may and may not make sense. Some other cases that are worth consideration include monitoring incoming wet commodities like wet distillers, beet pulp and gluten feed. How often do we currently test the moisture content of these feeds, and how often should we?

In addition, mounting NIR instruments on forage harvesting equipment allows for the creation of yeild maps that reflect both the total tonnage and the yield of individual nutrients like starch and protein. If we could cross reference our soil maps with protein yield maps what could we learn?

Finally, what if we could segregate each truckload of silage or bale of hay depending on the nutrient qualities of that individual lot?

As the scale of individual farms continues to increase we may need to re-think some of our beliefs about how feedstuffs should be tested. Under the right circumstances on farm NIR may be a viable option. Like any tool, it will require a user with some basic knowledge of its capabilities and acceptance of its limitations.