Résumé
The conservation of grasslands is a concern worldwide as they are threatened by climate change and the expansion of intensive agricultural practices. The management of these areas must take into account the decisional process of habitat selection by individual organisms to identify potential ecological traps or underused habitats. Organisms that live in heterogeneous environments must select their breeding habitat based on cues that reflect habitat quality. In dynamic ecosystems such as grasslands, environmental cues used by individuals should show a strong temporal autocorrelation, such that their characteristics during breeding can be predicted earlier in the season. Our objective was to test if habitat features that explain grassland birds' distribution during the nesting and chick-rearing period could be predicted from the habitat features available on territory settlement. In western France, we analysed the relationships between the occurrence, richness and abundance of four passerine species, and vegetation structure and composition during chick-rearing period. We then analysed the temporal autocorrelation of vegetation features to determine whether the cues used during the settlement period reliably predicted the vegetation features encountered at later stages of breeding. We found that birds selected habitats characterized by a low cover of grasses, but did not respond to the physical structure of vegetation. The composition of vegetation was also the only variable that exhibited temporal autocorrelation over the course of the season, suggesting that individuals may rely on this feature to select optimal breeding habitats. Our results suggest that in dynamic environments, and in the absence of breeding experience or public information, animals can choose their breeding habitat based on a simple assessment of vegetation composition. A detailed knowledge of the underlying drivers of habitat selection is essential to manage habitats, identify potential ecological traps, and enhance the attractiveness of areas especially those under agri-environmental schemes.