Estimation of genetic parameters for cow age at last calving under different censorship criteria

Abstract

Cow longevity measured by age at last calving was evaluated using a censorship criterion that consisted of the difference between the dates of age at last calving and the last calving on the farm. If this difference was greater than certain value, the cow failed, indicating that it should be discarded. Otherwise, the cow was censored indicating the possibility of future calving. The aim of this study was to estimate heritability and breeding values of bulls for longevity considering three different censorship criteria, 16, 26 and 36 months, using Weibull proportional hazard sire model. The 16-month criterion was proposed because it is the estimated average interval between births in Nellore. The 26-month criterion was proposed because it is an average value between 16 and 36 months. Lastly, the 36-month criterion was considered a long time interval for the cow have a new calving. The Spearman correlation test was used to compare the rankings of the bulls regarding the estimates of breeding values for longevity considering the different censorship criteria. The records of 21996 Nellore cows were used. The cows were the daughters of 2113 bulls from 13 farms that participate in the Nellore breeding program of the National Association of Breeders and Researchers (ANCP). Age at first calving was considered a fixed effect while the random effect was the contemporary group (season, year of birth, and herd) and sire. Heritability estimates for cow age at last calving were 0.1020, 0.1002 and 0.0871 for the 16, 26 and 36-month censorship criteria, respectively. The Spearman correlation estimates of sires’ rankings were −0.2124, 0.1348 and 0.1211 (P > 0.05) for the censorship criteria of 16–26 months, 16–36 months, and 26 and 36 months, respectively. Despite of the little genetic variance to age at last calving, this values were higher than those reported in the literature. The accuracy of the selected bulls varies depending on the criteria adopted. The Weibull proportional hazard sire model predicted the highest reliabilities for the 16-month criterion, compared to other censorship criteria studied, which can lead to increasing of reproductive and productive efficiency of cows in the herd, since the lower open days of cows and higher number of calves per cow in her productive life.