Evaluation of the Cornell Net Carbohydrate and Protein System on the prediction of dry matter intake of grazing lactating Holstein cows.

The aim of this work was to evaluate the ability of the Cornell Net Carbohydrate and Protein System (CNCPS) version 5.0 to estimate the dry matter intake (DMI) of grazing lactating Holstein cows, grazing coast-cross pasture supplemented with corn silage and concentrate (3 or 6 kg/cow/day). Six experiments were carried out, with 12 cows each. The cows were fed 17 kg/cow/day of corn ensilage in three experiments. The chemical composition of extrusa samples of forage was determined, obtained using an esophageous fistulated cow. The intake estimates were obtained using 5 g of chromium oxide (Cr2O3) methodology supplied two times a day. Data corresponding to animals (body weight, age, milk yield and composition and racial type), environment (temperature, air relative humidity and management condition) and the feed chemical composition in each experimental period were provided to the program. The DMI values predicted by the CNCPS model were close to those obtained by the chromium oxide methodology.