Analysis of the variances of South African pig carcass classification parameters

Abstract

This study analysed pig carcass records from a commercial abattoir to determine the existence of variance between/within the parameters, warm carcass mass (WCM) and back fat thickness (Fat), of the South African pig carcass classification system between class, gender, mass cluster and producer. A second part included taking the carcass measurements back fat thickness, eye muscle length (EML), depth (EMD) and area (EMA) and analysing the records to determine the relationship between the parameters and if a predictor could be identified to predict EMA and lean meat percentage (LM%). A data set of 65 788 pig carcass records from a commercial abattoir was analysed and it was determined that the relationship between Fat and WCM was strong across all the data. When the data was divided into males and females across all the data the relationship between Fat and WCM was stronger (P < 0.0001) in females which was also seen within the Sausagers class. However, when the carcasses that are classified into the P, O, R, C, U or S classes were examined the relationship between Fat and WCM was stronger (P < 0.0001) for males. Overlaps were found between the mass categories, Porkers, Cutters, Baconers and Heavy Baconers, which is an area of concern as these categories are used along with the P, O, R, C, U and S fat divisions to determine the monetary value of the carcass. However, the amount of variation found within class was low, with the exception of the PP class and S classes, therefore showing that the PORCUS carcass classification system in South Africa can still reliably describe carcass composition. Lastly, the data was analysed to determine the existence of a producer effect. Fifteen producers were chosen on the basis of having produced 200 or more carcasses. The linear and quadratic relationships between Fat and WCM across the 15 producers improved by 10%. Within the 15 producers the R2 values ranged from 0.12 to 0.72, showing that producer had a significant effect (P < 0.0001). The second part of the study included measuring the carcass characteristics, fat thickness (Sfat), eye muscle length (EML), eye muscle depth (EMD) and eye muscle area (EMA) at the point between the 2nd and 3rd last ribs, 45 mm from the dorsal midline using a calliper, on 87 carcasses and analysing the recorded data. Fat thickness (HGPFat) recorded by the Hennessy Grading Probe (HGP) was also recorded and analysed. Differences were found between the HGPFat and Sfat measurements which may be attributed to the different apparatus being used to take the measurements or because the HGPFat was taken on a warm carcass and Sfat was measured on a cold carcass or a combination of the two factors may have contributed to the difference. Warm carcass mass and EML were found to have the strongest relationship (P < 0.0001) out of all the carcass characteristics measured where WCM could explain 41% of the variation seen within EML and had a Pearson’s correlation coefficient of 0.64. Eye muscle area was able to explain 25% of the variation within WCM.