Experimental light at night explains differences in activity onset between urban and forest great tits.

Artificial light at night (ALAN) is rapidly increasing and so is scientific interest in its ecological and evolutionary consequences. In wild species, ALAN can modify and disrupt biological rhythms. However, experimental proof of such effects of ALAN in the wild is still scarce. Here, we compared diel rhythms of incubation behaviour, inferred from temperature sensors, of female great tits () breeding in urban and forest sites. In parallel, we simulated ALAN by mounting LED lights (1.8 lx) inside forest nest-boxes, to determine the potentially causal role of ALAN affecting diel patterns of incubation. Urban females had an earlier onset of activity compared to forest females. Experimentally ALAN-exposed forest females were similar to urban females in their advanced onset of activity, compared to unexposed forest birds. However, forest females exposed to experimental ALAN, but not urban females, were more restless at night than forest control females. Our findings demonstrate that ALAN can explain the early activity timing in incubating urban great tits, but its effects on sleep disturbance in the forest are not reflected in urban females. Consequently, future research needs to address potential effects of ALAN-induced timing on individual health, fitness and population dynamics, in particular in populations that were not previously affected by light pollution.