Impact of daylength on behavioural output in commercial broilers

Monitoring broiler behaviour is a helpful tool in assessing the impact that management systems, such as lighting programs, have on bird welfare. This study examined broiler behaviour at two ages in two separate flocks (27/28 days of age (27d) in flock one and 42/43d (42d) of age in the second flock) as affected by one of four lighting programs: 14Light (L):10Dark (D) (14L), 17L:7D (17L), 20L:4D (20L) and 23L:1D (23L). These lighting programs were chosen so relationships between the percentage of time spent on particular behaviours and daylength could be examined using regression analyses. Data were recorded with an infrared camera system, and then observed using a scan sample technique. The number of birds performing particular behaviours was recorded every 10min for a 24-hour (h) period (6 measurements×24-h per d) in two replicate pens per lighting treatment per age. The data, as a percentage of time, were tested for normality, and (log+one transformed) prior to analysis. Photoperiod length altered behaviour regardless of age during both the photophase and when averaged over the 24-h period. A negative linear or quadratic reduction in percent of time spent standing, walking, feeding, preening, stretching, dustbathing and litter pecking was found with increasing daylength at both ages in both the photoperiod and 24-h average. A similar reduction for running behaviour was found in the photoperiod and 24-h period at 27d, with the behaviour eliminated under 23L. Dustbathing was no longer present in the behavioural repertoire of the older 23L birds. Long photoperiods resulted in a reduced reactivity of birds to the presence of an observer, and of sound level within a room. Differences in percent of time spent performing all behaviours were non-existent or minor in birds raised on 14L and 17L, and reactivity was similar at 27d, but higher in 14L birds at 42d. Sound levels within these rooms were similar. Birds raised on 20L were intermediate in behavioural expression. In conclusion, the decrease or elimination of behaviours, including those that are highly motivated or necessary for health, indicated poor welfare for birds raised on near-constant light. There appeared to be no behavioural advantages for the use of a long scotoperiod such as in the 14L:10D program as compared to a 17L:7D program. Based on behavioural expression, optimal welfare was achieved under 16L to 17L.