Effect of short and long periods of separation on agonistic behaviour, injuries and stress in Hérens cows kept in loose housing

Abstract

Hérens cows are typically not dehorned and are therefore housed in tie-barns during winter. Recently, however, farmers have started to also use loose housing systems. They separate single cows from their herd for periods of a few days to ensure undisturbed calving and to avoid excessive activity caused by animals in oestrus. As these cows are highly motivated to fight, the re-introduction of individuals after separation is usually associated with increased agonistic behaviour. The aim of this study was to compare the effects of the length of separation with regard to behaviour, injuries and stress response. The study was conducted with 45 focal horned Hérens cows kept in four groups of 10, 20, 18 and 21 cows respectively on two Swiss farms: two groups were kept as suckler cows and two groups as dairy cows. Each focal cow was separated on two occasions, once for a short (median: 1 day; high oestrus in dairy, late pregnancy or lactation in suckler cows) and once for a longer (median: 2 days; high oestrus in dairy, calving in suckler cows) period. During the 6 h following the re-introduction, agonistic behaviour was recorded. Occurrence of fresh wounds was noted 10 h after re-introduction. As an indicator of physiological stress, faecal samples were collected and concentrations of the cortisol metabolite 11,17-dioxoandrostanes (11,17-DOA) were determined. All data were analysed using generalised linear mixed-effects models. Frequency and duration of agonistic interaction increased with longer separation ( p < 0.02) and quickly decreased during the hours following the re-introduction of the focal cow ( p < 0.001). The risk of injuries did not significantly change with duration of separation. Separating the cow from her group brought about an increase in concentration of 11,17-DOA ( p = 0.012) that decreased over time ( p < 0.001). Interestingly, concentrations during the re-introduction phase were lower in comparison with the separation phase ( p < 0.001). The concentration of 11,17-DOA was lower during the longer separations ( p = 0.018). We did not find any difference of practical importance in how cows reacted to long compared to the short separation related to the different hormonal states of dairy and suckler cows during separation, nor related to the changes in hormonal state within the suckler cows. In conclusion, the duration of a separation of a single Hérens cow from her herd should be minimised so as to reduce negative effects resulting from agonistic interactions.