Evaluation of dry matter intake and daily weight gain predictions of the Cornell Net Carbohydrate and Protein System with local breeds of beef cattle in China

Three feeding trials with different dietary treatments were conducted to evaluate predictions of dry matter intake and average daily gain of Chinese local beef cattle by Cornell Net Carbohydrate and Protein System (CNCPS) version 5.0. In Trial 1, 60 growing Jin Nan yellow steers (local yellow cattle breed originated from Shanxi Province) were assigned to two treatments with five pens in each treatment and six animals in each pen. Trial 2 was conducted with 72 Lu Xi yellow growing steers (local yellow cattle breed originated from Shandong Province) in 12 conventional pens. These animals were assigned to four treatments with six animals in each pen. In Trial 3, 60 Lu Xi yellow steers were allocated to four treatments with five animals in each pen. Average daily dry matter intake (DMI) for each steer was measured as a mean of each pen. All the data about animals, environment, management and feeds required by the CNCPS model were collected, and model predictions were generated for animals of each pen. Model predictions were evaluated by mean bias, mean square prediction error (MSPE) and regression between the observed and predicted values. Animals in 9 of 10 treatments showed a very close model-predicted DMI values to the observed. In another treatment, animals had a mean bias of −0.72 kg DM/day (TRT 10 in Trial 3). With overall data included, the regression equation between observed and predicted DMI was: YOBS = 1.10XCNCPS − 0.78 (R2 = 0.82; P<0.0001), with an intercept not different (P=0.32) from zero, but a slope still different (P<0.0001) from 1. Although the model-predicted ADG values were very close to the observed values in all trials, most of the predicted values were always lower than the observed values with absolute mean bias less than 0.2. When all data were pooled, the regression equation between observed and predicted ADG was: YOBS = 1.07XCNCPS + 0.02 (R2 = 0.73; P<0.0001), with an intercept not different from zero (P=0.88), but a slope that differed from unity (P<0.0001). Results suggest that the CNCPS model is an acceptable model to predict dietary DM intake and average daily gain of Chinese local beef cattle breeds. Further studies are warranted to introduce model adjustment for Chinese conditions.