Abstract #M13

# M13
Using 30-h in vitro NDF digestibility of feedstuffs in ration formulation: evaluation of predictions for milk and methane production in lactating dairy cows.
K. C. Krogstad*1, D. L. Morris1, P. J. Kononoff1, K. J. Herrick2, 1Department of Animal Science, University of Nebraska-Lincoln, Lincoln, NE, 2POET Nutrition, Sioux Falls, SD.

In vitro NDF digestibility (NDFd) is often used to determine forage quality, and its increased availability has spurred interest for use in ration formulation procedures. The objectives of this study were to compare the predictions of metabolizable energy (ME) allowable milk and methane production when default NDFd values or lab-determined 30-h NDFd were used in the Cornell Net Carbohydrate and Protein System (CNCPS; v 6.5). We hypothesized that the use of 30-h NDFd would improve ME-allowable milk and methane predictions from CNCPS. The 30-h NDFd was measured on forages and byproducts included in diets fed to lactating dairy cows. Predictions of carbon dioxide production and DMI were also evaluated. Treatment means (n = 32) originating from 8 energy balance studies were assembled to evaluate the prediction of ME allowable milk and methane production when either a default value or observed 30 h NDFd was entered into the feed library. Of the 32 means, 26 were made on Jersey cows and 6 were made on Holstein cows. Our data set averaged 18.8 ± 2.26 kg of DMI, 25.8 ± 6.00 kg of milk, and 15.1 ± 2.65 L of methane/kg of milk. Headbox-style indirect calorimeters were used to measure methane and carbon dioxide production in all studies. Comparison statistics were generated using R (v 3.5.2) and predictions were evaluated based upon the root mean square prediction error (RMSPE). A concordance correlation coefficient (CCC) was used to evaluate the agreement between the predicted and observed values. Carbon dioxide and the mean of the upper and lower bounds of predicted DMI resulted in CCC of 0.48 and 0.75 and RMSPE of 8.8% and 20.5% of the mean, respectively. The use of lab-determined NDFd resulted in a decreased CCC (0.87 to 0.82) and an increased RMPSE (10.7% to 12.5% of the mean) for ME allowable milk but an increased CCC (0.55 to 0.62) and reduced RMSPE (18.4% to 15.7% of the mean) for methane production. These results suggest that determination of NDFd may not be advantageous in predicting milk yield but may improve prediction of methane production.

Key Words: neutral detergent fiber, digestibility, Cornell Net Carbohydrate and Protein System (CNCPS)