Skip to main content
ARS Home » Research » Publications at this Location » Publication #126566

Title: WITHIN HERD EFFECTS OF AGE AT TEST DAY AND LACTATION STAGE ON TEST-DAY YIELDS

Author
item BORMANN, J - GEMBLOUX AGRIC UNIV
item Wiggans, George
item DRUET, T - GEMBLOUX AGRIC UNIV
item GENGLER, N - GEMBLOUX AGRIC UNIV

Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/1/2003
Publication Date: 11/1/2003
Citation: Bormann, J., Wiggans, G.R., Druet, T., Gengler, N. 2003. Within-herd effects of age at test day and lactation stage on test-day yields. Journal of Dairy Science. 86(11):3765-3774.

Interpretive Summary: Differences in lactation curves due to age and stage of lactation within herd were investigated. Test-day data starting with 1990 calvings for the entire Jersey population and Holsteins from California, Pennsylvania, Wisconsin and Texas were analyzed. Test-day yields were adjusted for the across herd effects using solutions from regional analysis. Effect of age was small, indicating that regional adjustments for age were adequate. The effect of lactation stage was large enough to indicate that accuracy of genetic evaluations could be improved by including this effect for milk, fat, and protein, but not for somatic cell score. By assuming the lactation-stage effect to be random, the effect is regressed toward the regional effects for small herds, but in large herds, lactation curves are herd specific. Including random stage effects increased heritability, and reduced prediction error variance thereby giving more accurate genetic evaluations.

Technical Abstract: Variance ratios were estimated for within herd effects of age at test day and lactation stage, on test-day yield and somatic cell score. Test-day data starting with 1990 calvings for the entire Jersey population and Holsteins from California, Pennsylvania, Wisconsin and Texas were analyzed. Test-day yields were adjusted for the across herd effects using solutions from regional analysis. Estimates of the relative variance (fraction of total variance) due to age were small, indicating that regional adjustments for age were adequate. The relative variances for lactation stage were large enough to indicate that accuracy of genetic evaluations could be improved by including this effect for milk, fat, and protein, but not for somatic cell score. When the within-herd lactation-stage effect is assumed to be random, the effect is regressed toward the regional effects for small herds, but in large herds, lactation curves are herd specific. Including random stage effects reduced relative-residual variance and increased heritability. Model comparisons demonstrated the greater explanatory power of the random stage model. A without stage effects model had increased prediction error standard deviations.