Assessing the completeness and consistency of global-scale avian datasets

The advancement of ecological research has increasingly focused on comparative methods and global-scale analyses. Birds are frequently selected for such studies for their widespread distribution and sensitivity to environmental changes. Given the need to integrate fragmented avian data, we aim to collect global-scale avian datasets and evaluate the completeness and consistency of them. We reviewed literature published in Springer Nature and American Association for the Advancement of Science (AAAS) from 2020 to 2024 and identified 90 articles with large-scale coverage and broad taxonomic representation. From these, we compiled 123 datasets covering environmental factors, avian traits, and ecosystem-level data. These were categorized into species-level and spatial data for completeness assessment, while consistency was evaluated across three modalities: species lists, nominal variables, and continuous variables. The database encompassed 12 distinct data types, including morphological, physiological, behavioural, and life-history traits, niche characteristics, evolutionary data, and environmental factors. Our assessment revealed that species distribution and abundance data were most frequently utilized in avian research worldwide. Both the species-level and spatial data exhibited high completeness. Continuous variables showed high consistency, whereas nominal variables and species lists exhibited moderate consistency with notable variability. Therefore, the supplementation of low-integrity data such as reproductive and behavioural datasets remains imperative. Systematic reconciliation between legacy and revised avian taxonomic inventories are critical for ensuring the consistency of data. We hope that the work can advance understanding of macroecology and providing guidance for global avian research.