Epsilon Archive for Student Projects

Abstract

This master thesis aimed to investigate variation in the sow and piglet behaviour and welfare over time during a 5-week nursing period in the loose housing pen. This study was performed through analysis of video-recorded behaviours in 17 Swedish and Dutch Yorkshire sows and their piglets (284 piglets, pure-bred Yorkshire × Hampshire, average litter size with standard error = 16.7 ± 0.56). All behaviours in the sows and their piglets were recorded by using a continuous sampling method in every second and 30-minute interval scan sampling. The behaviours were recorded from 09.00 to 20.00 for two consecutive days each week postpartum (pp) during the 5-week nursing period.
According to the results of behavioural observation in the sows and piglets, the sows showed a gradual decrease of nursing behaviour such as the number of nursing occurrence (from 18.5 in week 2 pp to 12.3 in week 5 pp), nursing duration (from 7 minutes in week 1 to 5.6 minutes in week 5 pp), and total nursing duration (from 111.9 minutes in week 1 pp to 68.1 minutes in week 5 pp). On the other hand, the piglets became more active and vigorous over time during the nursing period with showing the highest percentage of exploration (10.3 % in week 3 pp, 12.1 % in week 4 pp, and 12.5 % in week 5 pp vs 6.6 % in week 1 pp and 6.3 % in week 2 pp) and the percentage of contact to the littermates gradually increased week by week during the nursing period, from 0.2 % in week 1 pp to 2.5 % in week 5 pp.
This thesis used the sow location in the pen and piglet mortality as indicators to evaluate the sow and piglet welfare because the slatted floor can cause severe injuries to limbs and udder in sows and the piglet mortality has been well known as the piglet welfare indicator. The sow welfare might be impaired after week 1 pp because the percentage of lying and staying on slatted floor in the sows was significantly increased from 17.8 % in week 1 pp to 40.3 % in week 5 pp. The piglet welfare might be impaired during week 1 pp because the piglet mortality (the percentage of piglets crushed by the sow, euthanized and total mortality) were significantly higher in week 1 pp than the other weeks with reaching respectively 7.20 %, 7.89 %, and 18.15 % in the week 1 pp.
In conclusion, the piglets became more active and vigorous while the sows showed more avoiding behaviours related to nursings over time during the 5-week nursing period. Also, the sows showed an increased preference for lying and staying on the slatted floor from week 2 pp which could cause severe injuries to their legs and udder and irritated respiration caused by ammonia gas from the slatted floor. For the reasons above, the sow welfare might be impaired from week 2 pp. Regarding the high piglet mortality with piglets crushed by their sows in week 1 pp, the piglet welfare might be impaired during week 1 pp because crushed piglets by a sow are hardly seen in natural environments. However, there were no piglets crushed by their sows and the piglet mortality significantly decreased from week 2 pp. For the reasons above, the piglet welfare might become better from week 2 pp. However, missing data, imbalanced sample size in each week, and the absence of fixed effects of sow’s parity and batch and litter size in the continuous and scan sampling data might affect the assessment of the sow and piglet behaviour and welfare in this master thesis. Hence, to improve the accuracy of welfare assessment and behaviour assessment for the sows and piglets, non-missing data, larger same sample size and improved balance between multiparous and primiparous sows considering their parities and batches and litter size are needed.