UW–Madison Smart Restart: For information about fall semester instruction and campus operations, please visit For COVID-19 news updates, see

During this time, please contact us at

Digestion Kinetics: UW-Madison Researchers Developing “Smart” Forage Analysis

Custom-tailored forage digestibility predictions may soon replace “one-size-fits-all” forage analysis in the upper Midwest. Digestion kinetics adjusts digestibility predictions based on a cow”s forage intake, and more accurately measures the energy value of forages for cows at various intake levels, according to Dave Combs, a dairy nutritionist at the University of Wisconsin-Madison”s College of Agricultural and Life Sciences. Digestion kinetics formulas, coupled with near-infrared reflectance spectroscopy, should help dairy farmers to fine-tune ration formulations and forage-purchasing decisions.

“Digestion kinetics will more accurately predict how much per mouthful a cow will digest at a given intake level. It”s more of a dynamic number,” Combs says. “We can predict how fast and completely a forage is digested, for use in ration software. If you”re buying forage for cows at 100 pounds milk per day, we could predict the energy value for that forage, helping you fine-tune your purchase decisions.

“For example, you may have two lots of hay with the same acid detergent fiber value. Currently, both look OK, but our system may say that one lot is much better for high producers.”

There”s a wide range in relative feed values, but two samples that look very different on paper may behave the same in a cow, and vice-versa. The new approach ranks forages differently, and should reduce these discrepancies, according to Combs.

Current forage-analysis procedures are based on Relative Feed Value equations developed at CALS 20 years ago. They were a giant step forward at the time, but have serious limitations today, Combs says.

The original equations were developed just for alfalfa. However, most forage analyzed today contains grass as well as alfalfa. The old equations weren”t developed for grass/alfalfa mixtures, and can over- or underestimate the digestibility of such a mix by 25 percent or more.

We feed cows differently now, and there is much more variability between farms, and on farms between cows, Combs says. For example, a good milking cow may eat 60 pounds of dry matter a day, while a dry cow may eat just 20 pounds per day. Intake has a huge impact on the energy value of hays, but current equations give us just one value, he says.

Digestion is a process over time – as intake increases, the cow pushes feed through her digestive system more quickly. The less time the feed is in the cow”s system, the less time she has to extract energy from the feed.

“Digestion kinetics gives us an estimate of how completely digestible the feed is, and how quickly it digests. Using this information, we can predict how a forage will be used by animals at different production levels.”

The old system would predict digestibility based on fiber content. There”s a relationship, but it”s not perfect, Combs says. “Digestion kinetics identifies components of a feed that digest immediately, slowly, and not at all, and how fast they do it. We couple that with the cow”s intake, which tells us how long she”ll hang onto the forage. This gives us an estimate of the energy value of that particular forage for a cow at a particular level of production.

When Combs used the new and old equations on forage samples from Wisconsin labs, he got different results with about 30 percent of the samples. Subsamples run through lab and rumen tests showed that digestion kinetics predictions were more accurate than predictions from the old equations.

Most labs analyze forages using near-infrared reflectance spectroscopy, because it”s cheap and quick. Wet chemistry takes longer and costs more. “Unless we can predict digestion kinetics with NIRS, it won”t be practical to apply,” Combs says. “We feel we can predict those digestion kinetics parameters with NIRS almost as well as we can predict the NDF and ADF content of a forage with NIRS.”

Combs plans further validations and feeding trials to compare the old and new systems. He”s also working with other CALS dairy scientists to incorporate the new system into current UW ration software. The test will be available at forage testing labs by the end of March.

CALS dairy scientists work with a consortium of NIRS labs in the upper Midwest. Combs hopes to release these equations through the consortium, where they can be downloaded to anyone”s NIR machine. This will ensure that everyone gets the same equations, and help researchers follow problems that may arise as the equations are used.

Combs started this project in 1996, working with CALS agronomist Dan Undersander and dairy nutritionist Pat Hoffman, along with researchers at the University of Minnesota. Hoffman has done similar work with bypass protein values, and has licensed equations through the Wisconsin Alumni Research Foundation. The Marshfield Agricultural Research Station”s forage lab and some private labs are using Hoffman”s protein equations.