Genetics evaluation for longevity in Slovenian Simmental cattle

Abstract

A maternal grandsire model (sire-mgs model) using survival analysis was used to estimate fixed and to predict random effects in genetic evaluation for longevity of Slovenian Simmental population. Longevity was described as length of productive life (LPL). The culling frequency increases with lactation duration with the culling peak at the lactation end. At the end of second lactation, most cows were culled, while the number of culled cows decreases in subsequent lactations. Most culled cows first calved at the age of 26 to 30 months and the relative culling risk increases with the increase of age at first calving. The cows that are breed in herds with decreasing size are more likely to be culled. Determined minimal differences in estimated breeding values between analysed birth years indicate that in the period from year 1965–1998 selection for longevity was not conducted. In the period from year 1998 to year 2004 relatively high differences between average estimated sires BV for longevity per birth year were observed which could not be taken as accurate presentation BV because of the small number of data per sire. As the genetic evaluation accuracy highly depends on ratio of censored and uncensored records and also on number of daughters per evaluated sire, further research with the purpose to detect the impact of censored records proportion on accuracy of genetic evaluation as well as the determination of number of daughters per sire sufficient for accurate evaluation is necessary.