top of page

Feed Testing is Critical to Manage Winter Feeding Programs

In dry years, forage yield is lower than normal because the plants are shorter than average. This results in a higher leaf to stem ratio. With leaves containing more protein and energy than stems; the quality of forages are typically higher in a dry year.

Photo credit: Barry Yaremcio, Yaremcio Ag Consulting Ltd.

To take advantage of the higher quality, a representative sample needs to be taken and submitted to an accredited laboratory for analysis. There are many different types of sampling tools, but the key is to get a representative sample. If sampling baled feed, probe the bales between the twines or wrap. This provides material across the many layers or flakes within the bale. Take 20 core samples from each different type of forage. If there are 100 bales off one field, go along the stack and sample every fifth bale to get a representative sample.

Taking grab samples results in some of the fine stems, leaves, and flowers to be lost during collection. This can potentially reduce protein results 10 to 15% (11 to 9.3% protein), and TDN dropping from 62 to 57% or lower. If inaccurate information is used when balancing a ration, it is possible to spend an additional $0.50 per head per day when balancing a lactating cow ration.

There are two different techniques available to analyze feeds. The wet chemistry method is considered to be the standard. Near Infrared Spectrometry (NIRS) is a newer technique. This is what is done when grain samples are checked for protein at the elevator. Either method of analysis is accurate when determining protein, acid detergent fibre (ADF), and neutral detergent fibre (NDF). Testing for macro and trace minerals is more accurate by wet chemistry. If minerals are done by NIRS, results can be off by as much as 300% depending on feed type, growing conditions and the equations used to predict the nutrient content.

For more information on sampling feeds and feed analysis, contact Barry at 403-741-6032 or


bottom of page