Submitted to: Journal of Animal Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/1/2008
Publication Date: 9/1/2008
Citation: Leeds, T.D., Mousel, M.R., Notter, D.R., Zerby, H.N., Moffet, C.A., Lewis, G.S. 2008. B-mode, real-time ultrasound for estimating carcass composition in live sheep: Accuracy of ultrasound measures and their relationships with carcass composition. Journal of Animal Science. 86:3203-3214. Interpretive Summary: Lamb carcasses are the initial (raw) product of sheep meat production. Sheep breeders and producers in the United States are challenged with improving carcass characteristics, such as lean meat yield and quality, to better meet the demands of lamb consumers and to improve production efficiency. Improving carcass characteristics is challenging because carcass data are difficult and expensive to obtain, and animals, once harvested, are no longer available for breeding. To help circumvent this challenge, breeders and producers need methods to estimate carcass measures in the live animal that are inexpensive and give a meaningful description of the carcass. Ultrasound is a practical option because of its low relative cost and ease of portability, and it has been used successfully to estimate carcass measures in the beef and swine industries. The research reported here used current ultrasound technology, and market lambs representative of current U.S. production, to evaluate the accuracy of live-animal ultrasound measures, and to determine the ability of these ultrasound measures to characterize carcass yield and value. Using criteria established for the beef and swine industries, we were able to achieve acceptable accuracy in estimating carcass measures with live-animal ultrasound. Additionally, these ultrasound measures were predictive of carcass yield and value. Our data indicate that sheep with larger loin muscle area, when measured using ultrasound, will yield larger and more valuable carcasses. Thus, breeders and producers who use ultrasound to select sheep with larger loin muscle area and minimal fat thickness will be able to improve carcass characteristics, and generate more revenue per lamb marketed.
Technical Abstract: The accuracy and repeatability of live-animal ultrasound measures, and the relationships of these measures with carcass yield, composition, and value, were investigated using data from 172 wethers. Wethers were F1 progeny from the mating of 4 terminal sire breeds to Rambouillet ewes, and were finished on a concentrate diet to a mean BW of 62.9 kg (SD = 9.5 kg). Before transport to harvest, LM area, LM depth, and backfat thickness were measured from transverse ultrasound images taken between the 12th and 13th ribs. After harvest, these measures were taken on the carcasses. Carcasses were fabricated into subprimal items, and weights were recorded. Ultrasound accuracy and repeatability was assessed using bias, SE of prediction, SE of repeatability, and r statistics. Relationships among ultrasound and carcass measures, and between these measures and carcass yield, composition, and value, were evaluated using residual correlations and linear prediction models. Ultrasound bias approached 0 for LM area, and backfat thickness was overestimated by only 0.69 mm. The SE of prediction and r were 1.55 cm2 and 0.75 for LM area, and 1.4 mm and 0.81 for backfat thickness, respectively. The SE of repeatability was 1.31 cm2 and 0.75 mm for LM area and backfat thickness, respectively. At a standardized BW and backfat thickness, wethers with larger LM area and LM depth yielded larger and more valuable carcasses, and these relationships were detectable with ultrasound. For each SD increase in carcass LM area, dressing percentage increased 1.57 percentage points, gross carcass value increased US$5.12, and boxed carcass value increased US$6.84 (P < 0.001). For each SD increase in ultrasound LM area, dressing percentage increased 0.95 percentage points, gross carcass value increased US$3.15, and boxed carcass value increased US$3.86 (P < 0.001). The response in boxed carcass value attributed to disproportionate increases in high-value subprimal item weights was small. Responses in dressing percentage and carcass value were significant (P < 0.01) for ultrasound and carcass LM depth, but were smaller in magnitude when compared with LM area. These data indicate biological and economical incentives for increasing LM area in wethers, and live-animal ultrasound can provide reliable estimates of carcass measures. These results are applicable to terminal sire breeders and producers who market sheep using carcass-merit pricing systems.