1. All Articles
  2. Publications
  3. Noble News and Views
  4. 2001
  5. June

Evaluation of Methods Used to Predict Supplemental Feeding Needs for Cattle

  Estimated read time:

One of the most common requests we get from ranchers is, How can I run more cattle? Other than buying or leasing more property, there are about five ways to do this: increase forage production, improve nutrient conversion, understand and use animal behavior as a management tool, increase forage use, and increase managerial efficiency. Oftentimes, these improvements may require additional inputs such as fertilizer, fencing, and water points. Ranchers often deal with small profit margins and, because of cost constraints, sometimes can't incorporate the required elements to increase stocking rates.

This article is a summary of two years of information gleaned from a demonstration project at the Red River Demonstration and Research Farm in Burneyville, Oklahoma. The project's objective was to help ranchers improve their managerial efficiency by using technology to control the cost of one of the most expensive production items: supplemental feeding. Feed and grazing costs can represent 50 to 70 percent of total annual production costs in typical cow-calf management systems in Oklahoma (Lalman et al., 1997). If you take out the cost for pasture, total raised and purchased feed can account for $110 to $151 per cow (Doye and Northcutt, 1997). These significant costs can sometimes leave the rancher with less money to make necessary ranch improvements. Being able to understand your cattle's nutritional requirements and then being able to address nutritional deficits economically can drastically improve your managerial efficiency and leave you with more money to meet other goals for the ranch.

This demonstration project evaluated available methods for predicting supplemental feeding requirements for cattle. The objectives were to introduce nutrition profiling technology to ranchers, determine the reliability of each method for predicting livestock performance, and provide recommendations for the least expensive supplemental feeding.

These are the methods we evaluated:

  1. fecal sampling via near infrared reflectance spectroscopy (NIRS),
  2. vegetative sampling via wet chemistry, and
  3. traditional supplemental feeding procedures based on visual inspection of manure piles, pasture conditions, and body condition score (BCS) of cattle. Fecal and vegetative samples were analyzed for crude protein (CP) and digestible organic matter (DOM).

We began the project by separating ninety-six Brangus cows into three groups based upon common frame score, BCS, and physiological stage. Each group was assigned a paddock with a bermudagrass forage base for the winter. The Nutritional Balance Analyzer (Nutbal 1.18) software created by the Ranching Systems Group at Texas A&M University was used to predict supplemental feeding requirements according to forage quality estimates from fecal and forage samples. We considered five locally available supplemental feeds for this demonstration because of their common use in this region.

Evaluations were made during normal feeding months, November through early March. Livestock were weighed biweekly, and the average composite weight change for each group was compared with that predicted by the Nutbal software. If feeding recommendations were necessary, the least expensive ration was fed and reported for the three herds.

Colored ear tags identified cattle by herd. Nutbal was used to determine supplemental feeding for the orange herd according to fecal sample results and the blue herd according to forage sample results. The green herd was supplemented according to visual inspection of body condition, manure, and forage conditions. A farm technician supplemented the green herd without knowing forage or fecal sample results for the other two herds.

Average weights and BCS were recorded biweekly for each group of cattle. Actual change in body weight and observed BCS was compared with that predicted by Nutbal. Livestock were to begin the project at a BCS of 5.0 to 6.0 and end it within one-half score of their beginning BCS, without dropping below a BCS of 5.0.

Differences in predicted and measured weight gains and losses (individual herd averages), CP, and DOM were documented for each method during both years of the study. Feeding costs for the evaluation periods were also compared.

Contrary to our expectations, preliminary results from the first two years indicated that forage-predicted (wet chemistry) CP values tended to be higher than fecal predicted (NIRS) CP values from 1998—1999. Fecal samples represent forage that was actually consumed by livestock and generally is higher in nutritive quality than handpicked forage. Our only explanation for the higher value is that these results could be due to forage degradation during the seven-day gap between collecting forage and fecal samples. However, this gap was needed so that we could receive forage sample results and carry out supplementation recommendations within the observation periods. Results of CP analysis from 1999-2000 were inconsistent with data collected from 1998-1999.

Adjustments to the sampling protocol, including taking an additional forage sample on the same day that the fecal sample is collected and verifying measured forage degradation, will be made during the remainder of this demonstration. Fecal-predicted DOM values were consistently higher than forage-predicted DOM values from 1998-1999 and 1999-2000. Again, we expected this outcome because cattle diet selectivity was reflected in the fecal samples.

During both years of data collection, there was a tendency for fecal (NIRS) analysis to overpredict and forage (wet chemistry) analysis to underpredict measured average daily gain (ADG) in all three herds. More specifically, in 1998-1999 and 1999-2000, ADG was overestimated within the fecal-predicted (orange) herd by 1.19 and 1.37 pounds per day, respectively, while ADG in the forage-predicted (blue) herd was underestimated by 0.70 and 0.83 pound per day, respectively. These differences were not statistically significant but could prove relevant, depending upon management objectives and time of year. There were no differences in ADG from 1998—1999 for the green herd. However, from 1999—2000, fecal-predicted (NIRS) ADG was significantly higher than forage-predicted (wet chemistry) ADG, but neither was statistically different from the measured ADG.

The blue herd (forage/wet chemistry) was wintered most economically from 1998—1999. However, the difference between the blue herd and the orange herd (fecal/NIRS) was only $0.53 per head. The orange herd wintered most economically from 1999—2000. The difference between the orange herd and the blue herd that winter was $11.05 per head. All three of the methods accomplished the winter BCS goals of the project.

We will continue this project for at least one year. A much-improved version of the Nutbal software (Nutbal Pro) was available two years after we started the project. We, along with Texas A&M, are using the new software to analyze the last two years' worth of data. We want to find out whether there's any disparity in the results. We'll do another report on this project after we analyze next year's data.

Other staff responsible for this project include Clay Wright, Dan Childs, Evan Whitley, Jeff Ball, Tim Stokes, and Rob Self.

Literature Cited
Doye, D. and S. Northcutt. (1997)
Cow/Calf Financial and Production Performance: What We Are Learning from Standardized Performance Analysis (SPA) Data. Oklahoma State University Extension Service, F-231.

Lalman, D., D. Gill, and B. Johnson. (1997)
OSU Cowculator: Beef Cow Nutrition Evaluation Software. Oklahoma State University Extension Service, CR-3280.