Food availability limits avian reproduction in the city: An experimental study on great tits Parus major

1. The altered ecological and environmental conditions in towns and cities strongly affect demographic traits of urban animal populations, for example avian reproductive success is often reduced. Previous work suggests that this is partly driven by low insect availability during the breeding season, but robust experimental evidence that supports this food limitation hypothesis is not yet available.

2. We tested core predictions of the food limitation hypothesis using a controlled experiment that provided supplementary insect food (nutritionally enhanced mealworms supplied daily to meet 40%–50% of each supplemented brood's food requirements) to great tit nestlings in urban and forest habitats.

3. We measured parental provisioning rates and estimated the amount of supplementary food consumed by control and experimental nestlings, and assessed their body size and survival rates.

4. Provisioning rates were similar across habitats and control and supplemented broods, but supplemented (and not control) broods consumed large quantities of supplementary food. As predicted by the food limitation hypothesis we found that nestlings in (a) urban control broods had smaller body size and nestling survival rates than those in forest control broods; (b) forest supplemented and control broods had similar body size and survival rates; (c) urban supplemented nestlings had larger body size and survival rates than those in urban control broods; and crucially (d) urban supplemented broods had similar body size and survival rates to nestlings in forest control broods.

5. Our results provide rare experimental support for the strong negative effects of food limitation during the nestling rearing period on urban birds' breeding success. Furthermore, the fact that supplementary food almost completely eliminated habitat differences in survival rate and nestling body size suggest that urban stressors other than food shortage contributed relatively little to the reduced avian breeding success. Finally, given the impacts of the amount of supplementary food that we provided and taking clutch size differences into account, our results suggest that urban insect populations in our study system would need to be increased by a factor of at least 2.5 for urban and forest great tits to have similar reproductive success.