Assessing and explaining trends in dairy cattle herd performance variables while using three-breed rotational...

Three-breed rotational crossbreeding can improve the functional traits of dairy cows, but few farmers in most Western countries use it. This could be due to a lack of knowledge of its medium- and long-term effects on animal performance at the herd level, regardless of the initial states of the farm and herd, the crossbreeding programme(s) used and changes in farm management and structure while transitioning to it. We aimed at assessing changes in animal performance of dairy cattle herds transitioning towards three-breed rotational crossbreeding and at identifying factors that explained them. We sampled 13 French dairy cattle farms for which animal performance data were available for a 9-year period (2009-2017) and described their trends in herd management and the farming system when introducing crossbreeding. We calculated five herd performance variables: mean milk productivity per cow-year (MilkCow), mean Milk solids content (FatProtCont), and the percentages in the herd of cows with no more than 117 days open (FertileCow), with mean somatic cell score less than 4 over the lactation period (HealthyLact) and with four lactations or more (L4+). We defined 15 variables related to herd and farm structure and management. We applied linear regression to each variable over the study period and used the intercept and slope to describe their initial values in 2009 and their trends, respectively. We performed partial least squares (PLS) regression to assess relationships between the trends in herd performance variables and potential explanatory variables, including the initial values of herd performance variables and the initial values of and trends in herd and farm structure and management variables. Linear regressions showed that within herds, median MilkCow decreased (-177 kg/cow per year), FertileCow increased (+1.5%/year) and FatProtCont either remained stable or increased (+0.1 g/kg per cow). Conversely, changes were less distinct for HealthyLact and L4+. From 2009 to 2017, herd size increased in most farms (med = +1.2 LU/year). The PLS regression highlighted that herds in which MilkCow decreased the most over time were those in which HealthyLact and L4+ decreased and increased the most, respectively. Moreover, for herds in which FertileCow increased the most, FatProtCont either decreased, or increased less. Based on association scores from the PLS, changes in the genetic composition of the herd influenced changes in herd performance variables more than did changes in overall farm management (|0.28|-|0.59| vs |0.26|-|0.50|). Our findings suggest that dairy crossbreeding is relevant to rebalance herd performance between production and functional traits.