Genetics of feed intake and efficiency in grazing dairy cows : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Animal Science

Abstract

Feed efficiency in dairy cows is widely acknowledged as a highly desirable characteristic to improve because of its well-documented impact on production costs. Traditional measures of feed efficiency have used ratio traits, specifically energy conversion efficiency, but these have undesirable statistical properties. Alternative measures of feed efficiency are those based on the residuals from regression-type statistical models, the most common of which is residual energy intake (REI). Residual energy intake is defined as the difference between actual and predicted intake and is usually derived from least squares regression models. The general objective of this thesis was to quantify phenotypic and genetic (co)variances between the feed intake complex, performance, and fertility traits in lactating Holstein-Friesian dairy cows. A total of 8,199 feed intake records from 2,693 lactations on 1,412 grazing lactating Holstein-Friesian dairy cows from experimental farms were used. Several alternative efficiency definitions were developed, each with their own respective strengths and weaknesses. Exploitable genetic variation was demonstrated to exist for the range of alternative efficiency traits, and the magnitude of this variation was sufficiently large to justify consideration of the feed efficiency complex in future dairy breeding goals. The heritability estimates for the different efficiency traits estimated using repeatability models varied from 0.06 to 0.21. Variance components, however, differed across lactation when estimated using random regression models; for example, the heritability of REI varied from 0.04 (34 DIM) to 0.11 (280 DIM) across lactation. Phenotypic correlations among many traits including REI and energy balance (EB) differed not only by stage of lactation but also by cow parity. Moderate to strong genetic correlations existed between REI and EB across lactation (ranging from 0.45 to 0.90). Albeit associated with large standard errors, estimated genetic correlations between feed efficiency and reproductive performance were either neutral or favourable suggesting greater genetic merit for feed efficiency does not appear to be antagonistically genetically correlated with reproductive performance. Selection index calculations using the current economic weights in the Irish Economic Breeding Index, and genetic (co)variances estimated in this thesis, indicate that the inclusion of REI in the index with an economic weight of €0.078/UFL will generate animals with improved REI.