Using the Cornell Net Carbohydrate and Protein System to predict ruminal pH in dairy cows grazing high quality pasture

Abstract

Predicting milk production is complex due to interactions between the requirement for nutrients by the cow, the characteristics of different feeds and the contributions from body reserves towards the animal's requirements. Grazing introduces more complexity because of the variation in the characteristics of the forage, variation in nutrient intake between cows, diurnal patterns of eating, and further energy demands due to the effort of grazing. Ruminal pH is a critical determinant of metabolisable energy supply because of effects on the growth of different groups of microorganisms in the rumen. Our objective was to use the Cornell Net Carbohydrate and Protein System model to determine whether it was possible to predict ruminal pH for a range of pasture-based diets, based on inputs of nutrient intake, milk composition, liveweight and body condition. There was no (P>0.05) relationship between observed and predicted ruminal pH. The average daily ruminal pH for observed and predicted pH were similar and the mean bias was small and not different to zero. However, when the residuals were regressed against the predicted pH, a positive (P<0.05) systematic bias was demonstrated with pH over-predicted at low pH (5.7 to 6.0) and under-predicted at higher pH (6.1 to 6.2).