Objectives: The objectives of this project were to: 1. identify characteristics of low-ranking sows in group housing systems; 2. assess effects of group size on well-being and performance of low-ranking sows; 3. determine effects of feeding stalls on well-being and performance of low-ranking sows; 4. assess effects of social rank of sows on well-being and performance of offspring.
Procedures: The project consisted of two studies conducted at two sites: the University of Minnesota’s West Central Research and Outreach Center (WCROC), and a commercial farm with a 5,000-sow unit. At the WCROC, 150 sows (Landrace x Yorkshire, parity 1 to 9) that were group-housed with individual feeding stalls were used to accomplish objective 1, 3, and 4. Sows were assigned to pens (15 sows/pen, 2.2 m2/sow of floor space) at weaning. The control pen allowed sows to access feeding stalls only for feeding during the initial 48 h of mixing. The treatment pen allowed sows to access feeding stalls continuously. From the third day of mixing through the remainder of the gestation period, all sows in both control and treatment pens had continuous access to feeding stalls. To determine sow rank and stall usage, fighting among sows and sow staying in stalls were recorded by video-cameras for 48 h after mixing. Social rank was determined for each sow based on outcomes (wins vs. losses) of fights. Skin lesions were assessed before and 48 h after mixing, and before farrowing. Fear response, salivary cortisol concentrations, and heart rates were measured 5 weeks after mixing. To examine effects of sow rank on offspring, high ranking sows and low ranking sows farrowed in a group-farrowing system where 8 sows farrowed in individual pens and shared a communal area in each room. Sows and piglets commingled into a large group within each room when pens were removed at 10 d after farrowing. Farrowing performance, aggression among piglets at farrowing pen removal, and weight gain of piglets during lactation were recorded. At the commercial farm site, 152 confirmed pregnant sows (35 d post-mating, Camborough, PIC USA, parity 1 to 6) were allocated to 4 large pens (26 sows per pen) and 8 small pens (6 sows per pen) to accomplish objective 2. Both large and small pens provided the same floor space allowance (1.5 m2 per sow) and sows were fed on the solid portion of partially slatted floors. Aggression among sows at mixing and during the first two meals after mixing was recorded, and social rank was determined for each sow based on outcomes of aggression. Skin lesions were assessed at 24 after mixing and before farrowing. Salivary cortisol concentrations were measured at 5 weeks after mixing. Sows farrowed in farrowing crates and the reproductive performance of sows was recorded.
Findings: Results from the WCROC demonstrated that low ranking sows were younger (parity 1.5 vs. 3.9) with lower body weight (221 vs. 242 kg) at mixing than high ranking sows. Low ranking sows fought less frequently (40 vs. 60 fights/sow/6h), and lost more fights (76 vs. 19% of total fights) than high ranking sows, but sustained similar injuries caused by fighting as high ranking sows. Low ranking sows were more fearful at 5 weeks after mixing, indicating compromised welfare. In addition, low ranking sows farrowed smaller litter sizes (12.0 vs. 13.4 total born, 11.8 vs. 12.8 born alive) compared with high ranking sows, which may be attributed to poor welfare. The smaller litter sizes farrowed by low ranking sows were associate with fewer piglets dead (1.2 piglets/litter) in the group-farrowing system. As a result, low ranking sows did not wean smaller litter sizes (9.0 vs. 8.4) than high ranking sows. Piglets that were born to low ranking sows were similar in birth weight and weaning weight to piglets born to high ranking sows, indicating that maternal social rank did not affect the growth performance of offspring during the lactation period. Piglets born to low ranking sows fought frequently as and won more fights than piglets born to high ranking sows at 10 d of age, indicating that the behavioral characteristics of low ranking sows did not reflect on offspring. No difference in condition scores, backfat thickness, weight gain during gestation, cortisol concentrations, heart rates, and farrowing rates between high ranking and low ranking sows was observed.
On average, sows in pens with continuous access to feeding stalls spent 10% of their time in stalls during the first 10 h after mixing. Low ranking sows spent more time (13.4 vs. 5.8%) in stalls than high ranking sows. The difference in stall usage between low ranking and high ranking sows was more significant during the initial 4 h after mixing (27.2% vs. 6.9% of total observation time). This suggests that low ranking sows use feeding stalls as hiding spaces to avoid attacks from unfamiliar sows when fighting was intense. As a result, sows in pens with continuous access to feeding stalls fought less (38 vs. 50 fights/sow/6h), and low ranking sows sustained fewer skin lesions than their counterparts in control pens. These results suggest that access to feeding stalls during the mixing period improved the welfare of low ranking sows, as indicated by reduced injuries caused by fighting.
Results from the commercial farm confirmed the finding at the WCROC that low ranking sows fought less frequently and lost more fights than high ranking sows at mixing, but had similar skin lesions caused by fighting as high ranking sows. Low ranking sows and high ranking sows entered gestation pens with similar body weights and condition scores, but low ranking sows gained less weight (33 vs. 50 kg), resulting in lower body weight (251 vs. 268 kg) and poor conditions before farrowing than high ranking sows. This suggests that low ranking sows were less competitive for feed than high ranking sows in the group housing system with floor feeding. In addition, low ranking sows sustained more skin lesions than high ranking sows before farrowing, which might be attributed to fights received by low ranking sows during feeding. Small groups appear better than large groups for sow welfare and performance because all sows had fewer skin lesions and higher farrowing rates (97.9 vs. 87.4%) in small pens than in large pens. Consistent with the study at the WCROC, we did not observe difference in cortisol concentrations, condition scores of sows, litter sizes and litter weight weaned between low ranking and high ranking sows. We also did not observe difference in litter size farrowed between low ranking and high ranking sows, which was inconsistent with the results of the WCROC study.
Conclusions: Low ranking sows were more fearful, fought less frequently and lost more fights than high ranking sows in group-housing systems. Low ranking sows were less competitive for feed in the group housing system with floor feeding and had poor welfare compared with high ranking sows, as indicated by more skin lesions, less weight gain, and lower body weight before farrowing. Small group size appears better than large group size for sow well-being and performance because sows in small groups had fewer skin lesions and higher farrowing rate. In the group-housing system with individual feeding stalls, low ranking sows used stalls as hiding spaces to escape from fighting during the initial mixing period, which reduced skin lesions caused by fighting and improved the welfare of low ranking sows. Sow rank did not affect the growth performance of their piglets during lactation, and behavioral characteristics of low ranking sows did not reflect on their offspring.
Contact details: Yuzhi Li, [email protected], office: 320-589-1711
