Skip to main content
ARS Home » Northeast Area » Ithaca, New York » Robert W. Holley Center for Agriculture & Health » Research » Publications at this Location » Publication #131569
Title: IMPROVEMENT OF THE CNCPS MODEL TO BETTER PREDICT THE MILK PRODUCTION OF DAIRY CATTLE FED ALFALFA SILAGE

Author
item AQUINO, DANIEL - CORNELL UNIVERSITY
item TEDESCHI, LUIS - CORNELL UNIVERSITY
item LEE, SANG - CORNELL UNIVERSITY
item Russell, James

Submitted to: Cornell Nutrition Conference for Feed Manufacturers
Publication Type: Proceedings
Publication Acceptance Date: 9/5/2003
Publication Date: 10/21/2003
Citation: AQUINO, D.L., TEDESCHI, L.O., LEE, S.S., RUSSELL, J.B. IMPROVEMENT OF THE CNCPS MODEL TO BETTER PREDICT THE MILK PRODUCTION OF DAIRY CATTLE FED ALFALFA SILAGE. CORNELL NUTRITION CONFERENCE FOR FEED MANUFACTURERS. 2003. p. 137-150.

Interpretive Summary:

Technical Abstract: The Cornell Net Carbohydrate and Protein System (CNCPS) is a mathematical model that uses feed digestion rates and composition to predict the end products of rumen digestion and nutrients that are available for absorption. The rumen sub-model balances ruminal digestion rates with the passage rate of under graded feed to estimate metabolizable energy and protein. Validations of the model indicate that it can provide realistic estimates of the animal performance over a wide range of dietary ingredients. The CNCPS, however, requires methods of feed analysis that are not always currently available (e.g. digestion rates). When standard CNCPS feed library values for alfalfa silage are used, the predicted milk can be significantly lower than the actual production (bias of 9.1 kg/d). However, it appears that relatively simple adjustments can be made to improve this prediction. These adjustments involved the movement of the peptides and amino-N (approximately 70%) from the NPN to the B1 protein fraction and increasing the NDF (B2) digestion rate from 5.5%/h to 11%/h. When these adjustments were employed, predicted and actual milk production were highly correlated (r2 = 76.7%) and virtually all of the bias is eliminated.