Current Progressive Dairy digital edition

New milk analysis tools to improve animal performance

Progressive Dairyman Editor Karen Lee Published on 19 January 2018

Since the introduction of the Babcock butterfat tester almost 130 years ago, dairy producers and milk processors have been analyzing the content of milk to improve quality and profitability.

More recently, tools have been developed to monitor pregnancy-associated glycoproteins (PAGs), beta-hydroxybutyrate (BHB) and milk urea nitrogen (MUN) to better manage the health status and nutrition of dairy cows.



Dr. Dave Barbano, professor in the department of food science at Cornell University, and his colleagues are bringing together many years of work to new milk analysis tools that he said could soon be applied on the farm.

At the Vita Plus Dairy Summit on Dec. 6 in Madison, Wisconsin, Barbano explained how they have been analyzing bulk tank milk samples through milk cooperatives in the Northeast, DHI labs in Minnesota, Cornell University, the William H. Miner Agricultural Research Institute and a few other locations to study the fatty acid composition in milk and how it impacts milk and component production.

Thus far they have gained an understanding of how to use the fatty acid data for whole herd and for milking group diagnostics. However, they are still working on how these parameters are affected by days in milk, so they are holding off on applying this information to individual cow management.

In addition, hardware and software need to be developed to integrate this approach into the milk system. “The ultimate goal is this is going to be a sensor in the milking system to test milk continuously,” Barbano said.

Meanwhile, researchers are focusing on the metrics they have collected, what they mean, how to use them in the evaluation of the performance of the herd and how the cows are interacting with feed.


Since fatty acids vary in chain length and degree of unsaturation, the test measures this variation to determine the origin of the fatty acid:

  • De novo – made from within the mammary cells, and the fuel for this comes from the digestion of forage in the rumen, usually 18 to 30 percent of the fatty acids found in milk
  • Preformed – already made and come directly from feeds in the diet, 30 to 45 percent
  • Mixed – fatty acids that could be classified as de novo or preformed, 35 to 40 percent

“These new metrics let me tell the balance of those and how it’s changing as to what I’m feeding or the stage of lactation. That gives me insight as to how well or how poorly the cows are using the feed or interacting with the management practices,” Barbano said. “It tells me how the cow is digesting forage, how the rumen is working, and that’s really important information to determine if feeding strategy is working well.”

He explained the preformed fatty acids come in through the diet and are carried in the bloodstream to the mammary tissue and incorporated in milkfat. In late-lactation cows and during the dry period, they are stored as body fat, which is then mobilized at the beginning of the cow’s next lactation when she is in negative energy balance.

“From these metrics, I can tell how fast the cow is mobilizing fat and how much. If she does it too fast and too much, she’s going to get sick,” he said.

In running models on bulk tank milk samples, the researchers quickly noticed bulk tank fat and protein seemed to be correlated with certain fatty acid compositions. They selected 10 Jersey farms with low de novo percentages and 10 Jersey farms with high de novo percentages and did the same with Holstein farms. For one month, they did on-farm surveys, measurements of the facilities and feed samples. They also went back a second year to 40 Holstein farms, half high and half low, and focused on metrics they hadn’t done the first time, as well as looked at average milk composition for a full year.

The high de novo Holstein herds averaged 0.4 percent higher milkfat and 0.15 percent higher milk protein than the low de novo herds. For Jerseys, it was 0.9 percent higher milkfat and 0.5 percent higher protein on average for the whole year compared with the low de novo Jersey herds.


Since the high de novo farms were producing more milk per cow and they had higher components, it translated to more pounds of protein and more pounds of fat output from these farms. These differences for milk, fat and protein between high de novo and low de novo herds at 55 pounds of milk per 100 cows per year resulted in a gross income difference of nearly $30,000 more for high de novo herds when calculated in 2014.

The next year with the 40 Holstein herds, there was no difference in milk production between the high and low de novo groups, but the high group had 0.2 percent more fat and 0.1 percent more protein. The gross income difference was half of what it was the previous year, but that had more to do with how milk prices changed in 2015, Barbano said.

They looked at the farm survey information to find out what factors can be attributed to higher de novo levels. “We analyzed ourselves to death in terms of feed and didn’t come up with any smoking gun,” he said.

However, what stood out consistently both years was herds that had less bunk space per cow and more cows per pen with less resting space per cow to get good rumination had lower de novo fatty acids and lower fat and protein tests.

“The low de novo herds tended to be doing more bypass fat feeding since they tended to be on the low end of the test value, but they weren’t buying the highest quality fat supplements either. We think that had something to do with it too,” Barbano said.

When analyzed another way, preformed fatty acids had little effect on fat tests. These are fed in the form of bypass fat, which he said is still an important part of the ration, but mainly for the purpose of proper nutrition and energy input.

Barbano also mentioned that herds that had higher de novo and higher fat tests were consistently running higher effective fiber in the ration.

Age, stage of lactation matters

In order to start analyzing fatty acids in milk on different pens of cows or individual cows, it is important to understand the effect of days in milk.

“Fatty acid composition changes very rapidly in cows less than 80 days in milk,” Barbano said.

Typically at day four postpartum, the percentage of de novo fatty acids is going to be around 18 percent. If everything is working right in the cow, that percent is going to climb as she enters positive energy balance and level off around 25 percent for Holsteins and 27 percent for Jerseys.

What he’s found is that cows that are going to be in trouble with ketosis or a displaced abomasum will see a decrease in de novo, while the blood NEFA level stays high. “Any cow where de novo goes down is heading for a train wreck. It picks them out right away,” he said.

When comparing first-lactation heifers to older cows, the first-lactation heifer is going to have lower de novo fatty acids because she still needs energy for body growth. Barbano said he would like to sort animals by percentage of mature bodyweight at first calving to start to understand the impact of that metric with how much milk and components she is producing.

Other metrics

Another metric Barbano’s team is studying is trans-fatty acids as they interfere with the synthesis pathway in the mammary cell that’s producing de novo fatty acids. As trans-fatty acids go up, de novo fatty acids go down.

He also noticed that protein goes up when de novo goes up. Barbano said that if de novo fatty acids are being maximized, it is a sign the rumen is working well. A strong rumen biomass produces a lot of essential amino acids that support protein synthesis.

A third correlation he noticed was grams of lactose output per day and pounds of milk per cow per day. “You can’t make milk unless you make lactose,” he said. To achieve high milk output, glucose supply, transport and metabolism come into play because it is needed for lactose synthesis. In addition, anything causing an immune response will divert glucose to the higher priority activity and hinder milk production.

Through this extended analysis of de novo fatty acids, Barbano was able to draw correlations to milk and component production, and recognized age and stage of lactation as factors in the amount of de novo fatty acids produced. Further development of this test for milking system application could lead to its use for herd management as well as linking the performance of additional metrics.  end mark

Karen Lee
  • Karen Lee

  • Editor
  • Progressive Dairyman
  • Email Karen Lee