Predicting Forage Intake by Sheep through the Pampa Corte Model or NRC

Predicting Forage Intake by Sheep through the Pampa Corte Model or NRC

The aim of the present study was to evaluate the precision and accuracy of the Pampa Corte and National Research Council (2007; NRC) models for predicting forage intake (FI) by sheep. Individual data (n = 213) of observed FI, body weight and chemical composition of consumed diet were taken from fifteen indoor digestibility trials conducted with male sheep housed in metabolic cages and fed only forage ad libitum. The diets were composed of tropical grasses, temperate grasses and legumes. Individual observations of FI were averaged by treatment (n = 32) into each experiment which were then compared to FI values predicted by Pampa Corte model or NRC using concordance correlation coefficient (CCC) and regression analysis. The average value of observed FI was 847 (± 241) whereas those predicted by Pampa Corte model and NRC were, respectively, 826 (± 230) and 987 (± 208) g DM/day. Observed values of FI were linearly related (P < 0.01) to those predicted through either Pampa Corte or NRC. However, the Pampa Corte resulted in higher CCC than NRC. Also, through the Pampa Corte model, the linear regression presented a slope not different from 1 and an intercept not different from 0. The NRC model, however, resulted in a slope of the linear regression lower than 1 despite the intercept was not different from 0. The Pampa Corte model was more precise and accurate in predicting FI by sheep fed only forage than NRC.