"Good ration formulation is a form of risk management. Diets are formulated with a safety factor to ensure nutrient needs are met and toaccount for feed variability," noted Bill Weiss, with The Ohio State University, at the Tri-State Dairy Nutrition Conference.
Instead of looking at the range in a feed's nutrient value, Weiss proposed a shift to focusing on the mean and standard deviation of feed to quantify a safety factor.
Elevated levels of variation in a feedstuff means your feeding at the extremes; cows are inconsistently feeding off both ends of the bell curve. With the nutrient swings, the rumen is kept in a constant flux. When you know the feed's variation, though, you can take corrective measures and decide which end of the curve to balance toward.
Weiss used an example of forage with a wide variation in neutral detergent fiber (NDF). When formulating, a low NDF diet puts cows at a risk for acidosis. A high NDF diet could lead to drops in dry matter intake and yield. On farm, you need to decide which way to shift the mean so the risk of low or high NDF diets is reduced.
But when you receive an analysis, are the lab's standard deviations useful as an on farm tool? Weiss concluded not really. In the labs population, there are years of data and likely foreign data. There is a good deal of year-to-year variation in the data set and environmental factors that may not be relevant to your operation. Knowing the variation in feed composition at the farm level allows you to fine tune ration safety levels and develop sampling schedules.
To look at the variation on farm, Weiss along with members of the Ohio State Extension team collected samples of corn silage and haylage from eight Ohio dairies for 14 or 28 days.
For corn silage, NDF and starch values were most variable while dry matter tended to be more consistent. For hay crop silage, dry matter, and crude protein, values were the most variable. Compared to the national data set, within farm variation was lower. Farms tended to experience a quarter of the variation in dry matter and half of the variation in starch and NDF compared to the national data set.
He added that often it takes three to four samples to accurately define a corn silage or haylage population. Single samples should not be used to provide an accurate description of the feed or be the base for substantial ration changes. Additionally, when you switch cuttings and enter a new population you need to resample.
To manage variance, Weiss recommended either limiting the inclusion of variable feeds or using multiple ingredients. "Two highly variable feeds can be turned into a reasonably consistent feed," he concluded.
Instead of looking at the range in a feed's nutrient value, Weiss proposed a shift to focusing on the mean and standard deviation of feed to quantify a safety factor.
Elevated levels of variation in a feedstuff means your feeding at the extremes; cows are inconsistently feeding off both ends of the bell curve. With the nutrient swings, the rumen is kept in a constant flux. When you know the feed's variation, though, you can take corrective measures and decide which end of the curve to balance toward.
Weiss used an example of forage with a wide variation in neutral detergent fiber (NDF). When formulating, a low NDF diet puts cows at a risk for acidosis. A high NDF diet could lead to drops in dry matter intake and yield. On farm, you need to decide which way to shift the mean so the risk of low or high NDF diets is reduced.
But when you receive an analysis, are the lab's standard deviations useful as an on farm tool? Weiss concluded not really. In the labs population, there are years of data and likely foreign data. There is a good deal of year-to-year variation in the data set and environmental factors that may not be relevant to your operation. Knowing the variation in feed composition at the farm level allows you to fine tune ration safety levels and develop sampling schedules.
To look at the variation on farm, Weiss along with members of the Ohio State Extension team collected samples of corn silage and haylage from eight Ohio dairies for 14 or 28 days.
For corn silage, NDF and starch values were most variable while dry matter tended to be more consistent. For hay crop silage, dry matter, and crude protein, values were the most variable. Compared to the national data set, within farm variation was lower. Farms tended to experience a quarter of the variation in dry matter and half of the variation in starch and NDF compared to the national data set.
He added that often it takes three to four samples to accurately define a corn silage or haylage population. Single samples should not be used to provide an accurate description of the feed or be the base for substantial ration changes. Additionally, when you switch cuttings and enter a new population you need to resample.
To manage variance, Weiss recommended either limiting the inclusion of variable feeds or using multiple ingredients. "Two highly variable feeds can be turned into a reasonably consistent feed," he concluded.
