Effects of social dominance on milk production and grazing behaviour of lactating dairy cows : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

Abstract

In this study, three experiments were conducted to determine factors influencing social dominance in lactating, grazing dairy cows. Additionally, we investigated the effect of separating cows, based on social dominance, on milk production and grazing behaviour. Dominance in all experiments was quantified by calculating a dominance value (DV), which was measured through observation of wins and losses between cows in social interactions. A dominant cow which won all, or most, of its interactions would have a DV range between 60-90, a mid-ranking cow a DV between 30-60 and a subordinate cow which lost all, or most, of its interactions having a DV of 0-30. In New Zealand pastoral-based dairy farm systems, no information exists on the impact of social hierarchy, and its disruption, on animal productivity. In experiment 1 (Chapter 3), an observational study was carried out to identify factors determining social dominance of grazing dairy cows and the subsequent relationship with milk production. Recognition of dominance among peers was evaluated for three groups of cows differing in stocking rate and herd size. The three groups of Friesian × Jersey cows used in this study were, a large group of 189 cows stocked at a medium stocking rate (4.2 cows/ ha; MSR), a small group of 34 cows stocked at a high stocking rate (5.0 cows/ ha; HSR), and another small group of 29 cows stocked at a low stocking rate (3.5 cows/ ha; LSR). All cows (n=252) ranged in age from 2 to 11 years old. Cow liveweight (LW) ranged from 340 kg to 648 kg. In each of the three groups, LSR, MSR and HSR, the DV was positively correlated with age (r = 0.646, 0.349, and 0.442 respectively, P<0.05) and liveweight (r = 0.472, 0.166 and 0.487, respectively, P < 0.05). Dominance value was more strongly and positively correlated with milk production for the LSR group (r = 0.476, P <0.05), but less so for the MSR and HSR groups (r= and 0.291, P<0.05 and r=0.289, P<0.10 respectively). It is likely that dominant cows in the LSR group had higher milk yield because they were more successful when competing for feed, whereas in the higher stocked groups individual feeding of grain supplements in the shed probably buffered the competitive effect of dominance in the paddock. During the first experiment the stability of dominance in the large MSR group was also investigated by removing and returning 40 cows after three weeks. Observations before and after removal of cows showed that although the DV changed, the value stayed within the same range throughout the entire separation and re-grouping process indicating a stable hierarchy. Interestingly in the smaller HSR and LSR groups, (where no regrouping occurred) the social hierarchy was less stable as cows shifted in and out of DV range. Given that MSR remained socially stable when a random group of cows were removed, a second experiment (Chapter 4), was carried out which was designed to test whether social disruption by separating dominant and subordinate cows would affect their milk production. Over 28 days, from 4 to 31 March 2015, 48 multiparous, late-lactation and pregnant Friesian × Jersey cows, were allocated to 6 treatment groups, based on observations of dominance recorded in MSR group. The groups were dominant only (n=12), subordinate only (n=12) or dominant and subordinate mixed together (n=24). To enhance competitive interactions each group was further divided and offered either high (target 16 kg DM/ cow/ day; 65 m²/ cow) or low (target 12 kg DM/ cow/ day; 52 m²/ cow) herbage allowance above 3.5 cm as grazed pasture. The results again showed higher milk yield and milksolids (MS) production in dominant compared with subordinate cows (16.5 vs 13.7 L of milk/ cow/ day; 1.58 vs 1.33 kg MS/ cow/ day). However, there was no effect of grouping based on social dominance in milk production of dominant or subordinate cows when they were mixed or kept apart (16.76 vs 16.34 L of milk/ cow/ day and 14.22 vs 13.24 kg MS/ cow/ day, respectively). Further, herbage allowance, did not affect milk production between dominant and subordinate cows when grouped apart or together. However, when subordinate cows were grouped apart from the dominant cows, they achieved better liveweight gain (0.36 vs 0.05 kg/ day, respectively) than subordinate cows in the mixed rank group. Overall, there was no benefit of separating cows based on social dominance on milk production even when a low herbage allowance enhanced the social dominance interactions between cows. More information on competition and social stability was needed to explain the lack of milk yield response and sensitivity of subordinate cows to mixing in experiment 2. Therefore experiment 3 (Chapter 5), was carried out to determine the effect of separating subordinate cows on milk production and behaviour under restricted feeding of pasture and fodder beet (FB) supplementation. The investigation was carried out over 19 days from 18 April to 6 May 2016. A total of 54 multiparous, late lactation and pregnant Friesian × Jersey cows were classified as dominant or subordinate prior to the experiment. A replicated factorial design was used whereby subordinate cows grazed in a mixed group together with dominant cows (n = 36) or in a separate group of subordinate cows (n = 18). The diet consisted of a supplement of 3 kg DM/ cow/ day of fodder beet (FB) grazed in situ on a 4.2 ± 0.16 m strip of FB/ cow offered from 0830 to 1130 h, followed by a herbage allowance of 12 kg DM/ cow/ day above 3.5 cm as grazed pasture, with mean space allocation of 94.3 ± 4.21 m²/ cow/ day to be grazed throughout the day. The provision of feed supplement functioned to enhance the aggression between cows as a result of competition. The total number of agonistic interactions between cows was greater when cows were on FB than when they were on pasture (428 vs 16 interactions, respectively). The increase in interactions on the small area of FB showed that feeding supplement creates competition. Regardless of the variation in interaction on pasture or FB, there was no difference in milk yield of subordinate cows grazed as single rank or mixed rank group (9.2 ± 0.49 vs 8.3 ± 0.89 kg/ cow/ day). The results suggest there was no benefit in separating cows based on their social dominance even in an intense strip grazing of high dry matter yield crop situation was created to enhance competition. In conclusion, dominance in grazing dairy cattle in a New Zealand pasture-based system was found to be most closely associated with age, LW and to a lesser extent milk production. Dominant cows typically had greater milk production than subordinate cows but dominance value and milk production were more closely linked when individuals were not supplemented. However, there was no effect on milk production of separating cows according to their social rank, even under competitive conditions.