Large‐scale habitat selection by agricultural steppe birds in Spain: identifying species–habitat responses using generalized additive models

1 Predictive models of species’ distributions are used increasingly in ecological studies investigating features as varied as biodiversity, habitat selection and interspecific competition. In a pilot study, we based a successful model for the great bustard Otis tarda on advanced very high resolution radiometer (AVHRR) satellite data, which offer attractive predictor variables because of the global coverage, high temporal frequency of overpasses and low cost. We wished to assess whether the approach could be applied at very large spatial scales, and whether the coarse resolution of the imagery (1 km 2 ) would limit application to those bird species with large home ranges or to simple recognition of broad habitat types. 2 We modelled the distributions of three agricultural steppe birds over the whole of Spain using a common set of predictor variables, including AVHRR imagery. The species, great bustard, little bustard Tetrax tetrax and calandra lark Melanocorhypha calandra , have similar habitat requirements but differently sized home ranges, and are all species of conservation concern. Good models would reveal differences in distribution between the species and have high predictive power despite the large geographical extent covered. 3 Generalized additive models (GAMs) were built with the presence–absence of the species as the response variable. Individual species’ responses to the habitat variables were identified using partial fits and compared with each other. We found that this modelling framework could successfully distinguish the habitats selected by the three species, while the response curves indicated how the habitats differed. Model fits and cross-validations assessed using receiver operating characteristic (ROC) plots showed the models to be successful and robust. 4 We overlaid the predictive maps to identify key areas for agricultural steppe birds in Spain and compared these with the present network of protected sites in two sample regions. In Castilla León the provision of protected sites appears appropriate, but in Castilla La Mancha large areas of apparently suitable habitat have no protection. 5 These results confirm that large-scale models are able to increase our understanding of species’ ecology and provide data for conservation planning. AVHRR imagery, in combination with other variables, has sufficient resolution to model a range of bird species, and GAMs have the flexibility to model subtle species–habitat responses.