Improving selection on udder health by using different trait definitions of somatic cell count

Two sets of somatic cell count (SCC) traits were defmed in this study: (1) traits derived from the proportion of test-day SCC above 150,000 cells/ml, (2) patterns of peaks in SCC. Genetic parameters were estimated for these SCC-traits, as weil as their genetic correlation with mastitis-traits. Mastitis-traits were clinical and subclinical mastitis (CM and SCM, respectively) and were both scored as binary traits. Data was available from farms with a PC Management Information System, and the dataset contained 56,726 lactations of 30,145 cows on 272 herds. Variance components for sire and permanent animal effects were estimated using ASREML. The estimated heritabilities for CM and SCM were both low at 3%. HeritabilitieS for SCC-traits ranged from 1 to 11%; 10west for the patterns of peaks in SCC, and highest for SCC-traits describing the dynamics in SCC. A range of genetic correlations was estimated between SCC-traits and.CM or SCM, vary41g from 0.60 to 0.93 for CM, and from 0.68 to 0.94 for SCM. Strongest genetic correlation was estimated between CM and the pattern of peak that describes a quick recovery in SCC (0.93), and between SCM and the presence of at least one test-day SCC above 150,000 cells/ml (0.94). Therefore, to improve the overall udder health, information from a combination of SCC-traits is expected to be most successful.