Estimation of environmental sensitivity of genetic merit for milk production traits using a random regression model.

The objective of this study was to estimate effects of environmental sensitivity of milk production traits for several environmental parameters and to investigate the impact of combining traits with different environmental sensitivity in an economic index. Variance components and breeding values were estimated for milk, fat, and protein yield, and fat and protein percentage by applying a random regression on values of an environmental parameter for each sire. Fourteen environmental parameters were defined and fitted to data consisting of 151,696 heifers in 6780 herds in The Netherlands with first-lactation records for milk production, somatic cell count, body condition score, and number of inseminations. Milk, fat, and protein yield showed environmental sensitivity in combination with 12 environmental parameters. Herd-year averages of protein, body condition score, age at calving, calving interval, and peak date of calving explained most genotype by environment interaction, mainly resulting from scaling effects. Almost all genetic correlations across environments were 0.99 or higher. Although heterogeneity of genetic variances was considerable, heterogeneity of heritabilities was limited. Scaling had a large effect on the weights of the economic index, but environmental sensitivities of milk, fat, and protein yields were approximately of equal magnitude. Consequently, very little reranking occurred based on the economic index.