Résumé
Biodiversity monitoring is central to conservation biology, allowing the evaluation of the conservation status of species or the assessment of mechanisms of biodiversity change. Birds are the first taxonomic group to be used to build headline indicators of biodiversity due to their worldwide spatial and temporal coverage and their popularity. However, the landscape of bird-monitoring practices has never been characterized quantitatively. To objectively explore the strengths and weaknesses of the massive bird-monitoring effort in Europe we assessed the bird-monitoring practices, acquired with a questionnaire-based survey, in a sample of monitoring programs. We identify major correlates of among-program variability and compare monitoring practices from our database to recommendations of best monitoring practices. In total, we obtained responses from 144 bird-monitoring programs. We distinguish three types of monitoring programs according to the number of people that they involve: small, local-scale programs (56%), medium or regional programs (19%), and large-scale, national and international, programs (23%). In total, the programs in our sample involved 27941 persons, investing 79298 person days per year. Our survey illustrated that 65% of programs collected quantitative indices of abundance (count data). The monitoring design in a majority of the programs could be improved, notably in terms of unbiased spatial coverage, sampling effort optimization, replicated sampling to account for variations in detection probability, and more efficient statistical use of the data. We discuss the main avenues for improvement in bird-monitoring practices that emerge from this comparison of current practices and published methodological recommendations.