Testing habitat copying in breeding habitat selection in a species adapted to variable environments

The habitat copying hypothesis states that animals use the reproductive performance of conspecifics to evaluate habitat quality and choose their future breeding site. We used data from Audouin’s Gull Larus audouinii (1992–2003), a species adapted to unpredictable environments, to analyse subcolony (as patch) choice within a colony (small spatial scale). We also assessed the suitability of alternative hypotheses to the habitat copying hypothesis. The probability of subcolonies being reoccupied annually increased with their size (as number of nests), which suggests the existence of group adherence effects. Subcolony growth rate was related to its average reproductive success (or patch reproductive success) in the previous year: the higher the reproductive success in a colony, the higher the probability of growth the following year. However, this last result was obtained without considering the effect of colony size on the response variable because colony size is related to it. Therefore, results suggest at the population level that in this system habitat copying might either be one of the strategies used by the species in selecting its breeding habitat, or one of the possible strategies operating alone. The other strategies are group adherence mechanisms, and also the effect of conspecific attraction. At the individual level we failed to find evidence of habitat copying and only the previous success of an individual affected its fidelity to a subcolony. The importance of the lack of environmental predictability in the system is discussed, as predictability is a prerequisite of habitat copying.