Influence of foreign stallions on the Swedish Warmblood breed and its genetic evaluation

Abstract

The objective of this study was threefold: 1) To survey the use and impact of foreign stallions in the Swedish Warmblood (SWB) breed 2) To investigate the impact of including genetic groups in the genetic evaluation of the SWB and 3) To estimate the impact of incomplete pedigree information on the estimated breeding values (EBVs) for stallions. Data consisted of 189,330 horses of which 50,907 had performance information. Studbook of origin was determined for stallions with at least 5 progeny tested in Sweden. Those 757 stallions had together 116,505 progeny registered in the SWB. Genetic groups were formed according to preliminary analyses of the genetic levels in dressage and show jumping of the different populations. During the first studied period, comprising horses born before 1980, about 80% of the foals were sired by Swedish born stallions, whereas these only sired 40% of the horses born 1990–2003. In this period Holstein stallions had sired 21% of the foals and Dutch stallions (KWPN) 10%. The proportion of mares covered by foreign stallions increased after the early 1980s to about 80% in 2006. Some of the foreign populations, e.g., Holstein, Selle Francais and KWPN, have significantly affected the show jumping performance level of the SWB breed in the past although the deviation from SWB stallions has become smaller over time. Regarding dressage, no particular foreign population has consistently shown such an impact on the performance although Oldenburg stallions seem to be influential lately. The pedigree completeness (PEC values) of the horses included amounted on average to 0.845 for horses with performance data and to 0.907 for stallions with at least 15 progeny. Foreign stallions had an average PEC value of 0.858. When simulating no pedigree information available for foreign stallions, the EBVs of these stallions changed on average by 19–22 index units, which equals about one genetic standard deviation. Thus, keeping complete and correct pedigree registers is crucial for the assessment of reliable EBVs of SWB horses. Including genetic groups in the genetic evaluation of the SWB resulted in hardly any re-ranking of horses. Correlations between EBVs were about 0.99. Inclusion of genetic groups caused larger standard errors of the EBVs. Furthermore, as the pedigrees on average have great depth, and partially also because Warmblood sport horses have become a mix of many different populations, it was not recommended to include genetic groups in the model for future routine genetic evaluations of SWB.