Evaluation of the NRC (2000) Beef Model for Predicting Performance and Energy Requirements of Cattle Fed under Western Canadian Environmental Conditions

The NRC (2000) beef model is widely used to evaluate nutrient requirements and feeding programs for cattle. The objectives of this study were to assess the accuracy and precision of the NRC (2000) beef model in predicting dry matter intake (DMI), shrunk weight gain (SWG), net energy of maintenance (NEm) and gain (NEg) requirements and also to determine the relationship of body condition score (BCS) and ultrasound subcutaneous fat thickness (USF) to total body fat of steers fed under western Canadian environmental conditions. Data used for this study was from Basarab et al. (2003). The study was conducted over two years using a total of 176 steers. The DMI, SWG, NEm and NEg for each steer were modeled using the NRC (2000) beef model under actual environmental and thermoneutral conditions. Retained energy (NEg) was calculated for each animal based on initial and final body composition. Actual NEm utilized was calculated by subtracting NEg adjusted for the efficiency of metabolizable energy used for gain (kg) from total metabolizable energy consumed and by adjusting for the efficiency by which metabolizable energy is used for maintenance (km). The accuracy of predicted values was evaluated by means comparison, regression and residual analysis, concordance correlation coefficient (CCC) and reliability index methods. Dry matter intake was over predicted (P0.05) under all methods investigated with a general over prediction for NEm and under prediction for NEg under actual environmental and thermoneutral conditions. Cattle NEg was under predicted, possibly explaining why SWG was under predicted. Potential reasons for this inaccuracy includes failure to account for specific physiological and behavioral adjustments such as changes in organ size, passage rate, hide thickness influencing the NEm calculation and in the case of NEg due to the lack of precise knowledge of actual composition of gain by growing steers due to lack of specificity of initial body composition. Body condition score and USF had a comparably strong relationships to total body fat (P<0.05, adjusted r2= 0.55 or 0.56, respectively), suggesting potential for their use in improving composition of gain predictions (P