INFLUENCES OF TIME AND COMBINATIONS OF EQUIPMENT AND OPERATOR ON ULTRASONIC RIB FAT MEASUREMENTS FOR PREDICTION OF BEEF CARCASS COMPOSITION
Rib fat measurements, collected on 81 crossbred steers during three sampling periods using three operators of whom operators 1 and 2 operated a Krautkramer USM2 and a Scanoprobe 731A and operator 3 operated only a Scanogram 722, were used to study the patterns of subcutaneous fat deposition in live cattle and to assess the differences in the accuracy of predicting carcass composition among machine-operator combinations. The effects of breed of sire, animals within breed and machine-operator combinations were all significant (P < 0.01) sources of variation affecting ultrasonic measurements in each of the three periods, which were at least 22 days apart. The within-breed-of-sire linear regression of ultrasonic rib fat measurements on age (291–472 days) and on half liveweight (226–274 kg) were all positive and significantly different from zero (P < 0.01), but they were not consistent among machine-operator combinations, which limits the potential usefulness of ultrasonic measurement to characterize the rate of subcutaneous fat deposition in live cattle. However, repeated measurements over time on the same cattle improved the accuracy from 10% to 50% of four machine-operator combinations in predicting carcass dissectible muscle. The ultrasonic measurements by various machine-operator combinations differed in the accuracy with which carcass composition was predicted. In decreasing order they ranked Krautkramer-operator 1, Scanogram-operator 3, Scanoprobe-operator 2. Krautkramer-operator 2, and Scanoprobe-operator 1. Errors in ultrasonic measurements were identified as factors contributing to the relatively low efficiency of prediction equations for ultrasonic prediction of live animal carcass composition. Key words: Ultrasonic measurements, time, machine-operator differences, carcass composition