Effects of floor-feeding and the presence of a foraging substrate on the behaviour and stress physiological response of individually housed gilts

Both restricted feeding and barren housing have a negative influence on sow welfare. The aim of this study was to test whether sows that have to search for their feed in a substrate on the floor show less stereotyped (and other abnormal) behaviour and have a lower physiological stress response. In three batches, 96 gilts were housed individually in two rooms in 3.1 m² pens with 1.9 m² solid floor. In a 2×2 factorial design either wood shavings (S) or no substrate (NS) were provided on the floor, and 900 g of feed was provided twice daily (06:30 and 15:00 h) either in a trough (T) or on the floor (F). In weeks 8–12, behaviour was scan-sampled once in the periods 07:00–09:00 h (P1), 10:00–12:00 h (P2), and 13:00–15:00 h (P3). Data from the 5 weeks were pooled per animal. Video recordings (24 h) in week 12 or 13 were scan-sampled for ‘standing’. Saliva samples were taken in week 11 at 2 h intervals during 24 h and measured for cortisol. Spontaneously voided morning-urine was sampled in weeks 2, 7 and 12 or 13 for determination of ratios of adrenaline (A) and noradrenaline (NA) with creatinine (CR). Most effects that were found were due to substrate presence. Main findings were that compared with S-animals, NS-animals stood more during the dark period (4.8% versus 3.0%; P<0.05) and showed more (visible) oral behaviour in P1 (56.8% versus 47.6%; P<0.05), P2 (35.8% versus 30.8%; tendency) and P3 (44.1% versus 33.8%; P<0.05). This included more sham chewing in P1 (tendency), P2 and P3, more pen manipulation in P3, and more other oral behaviour (e.g. teeth grinding) in P1 (tendency) and P2. They had higher cortisol levels before feeding (peaks) and in the early evening (24 h average: 1.49 ng/ml versus 1.02 ng/ml; P<0.05) and higher NA/CR ratios in weeks 7 (tendency) and 12 (5.5 ng/mg versus 3.7 ng/mg; P<0.05). No treatment effects were found on A/CR ratios. Substantial interactive effects of feeding method and substrate were only found in floor manipulation. In all three periods NS-T-animals manipulated the floor less than other animals, probably because they had no attraction to the floor and were not rewarded for searching. Results imply that the presence of a substrate on the floor improves welfare, whereas provision of feed in the substrate in order to stimulate foraging behaviour is of less importance.