THEORETICAL BASIS FOR GENETIC-PARAMETERS OF HERD LIFE AND EFFECTS ON RESPONSE TO SELECTION

A theoretical model of herd life was used to quantify biases in genetic parameter estimates from culling on production and to study effects on response to selection. Herd life was modeled as a linear function of production and ability to survive regardless of production (survival). Genetic improvement of survival is of interest. Results of analytically derived formulas showed that estimates of heritability obtained from analysis of herd life are biased for survival. For moderately negative genetic correlations between production and survival, biases tended to be upward. The genetic correlation between production and survival was severely overestimated when based on production, and herd life. Adjustment of phenotypic herd life for phenotypic production removed some of the biases in genetic parameter estimates unless little direct culling for production occurred. Incorporation of estimated breeding values for herd life that were adjusted for production in a selection index, along with estimated breeding values for production, resulted in more response in a breeding goal consisting of production and survival than did inclusion of estimated breeding values for herd life when the standardized direct effect of production on herd life was larger than .15. For smaller values, adjustment of herd life for production reduced response to selection. Given current levels of culling on production, measures of herd life should be adjusted for production when included in selection strategies.