Australian grass-fed beef for Japan. 1. Carcass quality characteristics, saleable meat yield, and the value of various carcass measurements for predicting yield

Abstract

Fifty-six carcasses from grass-fed steers of 4 breed types, purchased for the Japanese market, were evaluated for quality and yield characteristics. The mean (¦ s.d.) carcass weight was 324 ¦ 28.3 kg, and P8 fat depth 16.5 ¦ 5.27 mm. Based on the Japanese grading system. 52 of the carcasses were considered acceptable for marbling, 54 for fat colour, and all carcasses for meat colour. The mean marbling, fat colour, and meat colour scores were 3.3, 3.3, and 3.5 respectively. The rank correlation between P8 fat depth and marbling score was 0.09, and a similar level of association was found between marbling score and fat depth measurements at other positions on the carcass. Measures of M. longissimus thoracis et lumborum (LD) area, using different techniques at the same rib, were significantly (P<0.001) correlated (5th rib, r = 0.87; 10th rib, r = 0.80); however, there was no association between LD area at the 5th and 10th ribs. Models were developed to predict the weight of saleable meat (SM) and fat trim (FT) based initially on hot standard carcass weight (HCW) and P8 fat depth as the independent variables. For all models HCW was the important predictor, with various measures of subcutaneous fat depth increasing the accuracy of estimates by a small amount. Fat depth measurements from the loin region reduced the error of prediction of SM more than fat depth measurements from the rump region, irrespective of the method used to measure fat depth. Replacing fat depth measures with the weight of trimmed fat significantly improved the accuracy of the predictions. The best model for predicting FT weight was based on the predictors HCW and fat depth at the 12th113th rib. Breed type was found to be not significant (P = 0.07) if included in a model with P8 fat depth and HCW. When breed type was included, P8 fat depth was no longer significant. When LD area at the 10th rib was included with HCW and P8 fat depth as a predictor of SM, P8 fat depth was no longer significant (P>0.05), but this was not the case when estimates of LD at the 5th rib were used. Regardless of this, estimates of LD area at both ribs showed themselves to be useful predictors of SM additional to HCW and P8 fat depth. Breed type was not a significant (P>0.05) contributor to the model based on the predictors HCW, P8 fat depth, and LD area at the 10th rib. If fat depth at the 10th rib was used with HCW and LD at the 10th rib instead of P8 fat depth, then more of the variation (R2 = 0.88 v. 0.85) in SM was explained and the error of the prediction was less (3.34 v. 3.75). Also, when fat depth at the 10th rib was used, the rib site used to estimate LD area was less important. The models showed that prediction of SM in heavyweight carcasses was marginally aided by using estimates of LD area.