Improving offspring survival outcomes for mammalian livestock species

The objective of this paper is to discuss strategies to select for improved offspring survival in pigs, sheep and cattle. Poor offspring survival is a welfare issue eroding social license and high survival is important from production and sustainability perspectives. However, genetic selection for improved offspring survival creates unique challenges. Survival of young animals is a lowly heritable binary (0/1) trait, with significant genetic variation influenced by both direct and maternal effects. Since all parents are survivors, the relative influence of direct and maternal effects can be difficult to disentangle to inform trait selection. Maternal death or lactation failure is accompanied by death of progeny. However, causes of poor survival in young animals also varies with other maternal factors, like dam parity or birth type, along with age and environmental challenges, such as the thermal environment, nutrition, disease, or management activities. Potential explanatory factors, survival outcomes and the birth type and pedigree of all individuals, dead or alive, are often poorly recorded.Natural selection favours the survivor. In contrast, modern production systems can mask the inherent differences amongst animals in their ability to survive, limiting useful selection criteria. It is advantageous to counteract undesirable correlated trends affecting offspring survival, such as very high or low and variable birthweights resulting from selection for increased growth or larger litter sizes. Additionally, selection for better maternal characteristics is of increasing importance as litter size increases and also offers the opportunity of a genetic evaluation with repeated dam records, even if sire parentage of progeny is unknown. Gestation length and litter size are characteristics influencing progeny survival, along with maternal attributes affecting ease of parturition, maternal behaviour, lactation or udder characteristics, maternal physiology and body condition. Several of these attributes are applicable to both extensive and intensively raised livestock, but their relative importance may differ and not all attributes are easily recorded.Genetic evaluation models at the offspring level (fitting additional maternal effects) offer opportunities for gains connected to traits which can be recorded directly on offspring, particularly where variation exists between litter mates or full-sibs. Several physiological traits have been proposed, but these are typically costly to record and can be biased by failing to phenotype non-surviving progeny. Further, non-surviving progeny are not selection candidates, such that selection decisions must be at a different (eg sire) level. Genotyping offers improved opportunities in this regard, particularly if dead progeny are also genotyped.We conclude that selection for improved offspring survival is likely to require a multi-pronged approach, using both 0/1 survival data (direct selection) and specific characteristics (indirect traits) of dams or their offspring. For national genetic evaluation systems, robust trait choice and achieving quality data at scale may be difficult, but is achievable.