Commonness of not-so-common birds: the need for baseline knowledge of actual population size for the validation of population size predictions

Capsule The hypothesized regional population size of a common species in Spain was found to be unreliable when compared with a complete and simultaneous single-species direct census.Aims To provide a regional census of Eurasian Jackdaws Corvus monedula as an example to highlight the crucial importance of baseline knowledge of actual population size of common species for validation and calibration of population size predictions derived from extrapolation.Methods Population size was measured by means of simultaneous counts of winter communal roosts after determining the best census date and other relevant information on flock routines, foraging areas and potential sources of count variability.Results A large discrepancy between hypothesized (∼330 000 individuals) and directly censused population size (∼15 000 individuals) was recorded.Conclusion Programmes aimed at establishing population size of wildlife by statistical inference should attempt to explicitly test their predictions by comparison with directly censused population sizes of particular species in control areas.