Central Asia Bird Occurrence Database — curated, openly-licensed layer

A standardized, taxonomically harmonized compilation of bird occurrence records for the five countries of Central Asia (Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, Uzbekistan), reconciled to the AviList v2025 taxonomy and bridged to the BirdTree phylogeny. This deposit is the openly redistributable layer of a larger database described in the accompanying Data Descriptor. It contains 219,611 records of 518 species in five CSV files, each holding the records released under a single licence (CC BY 4.0, CC BY-NC 4.0, CC BY-SA 4.0, CC BY-NC-SA 4.0, CC0 1.0), all sharing one 33-column schema and data dictionary. The files are deliberately not merged: a combined file could only be offered under the most restrictive licence it contained. Records from sources whose terms do not permit redistribution – eBird, the Macaulay Library and the birding.day portals – are cited rather than re-hosted. The full published snapshot is 911,836 records of 547 species, so this deposit holds 24.1% of the records but 94.7% of the species; the 29 species absent from it are held only by the cite-only sources and are disproportionately scarce or vagrant. Reusers computing regional richness should combine these files with the cite-only components obtained from their own sources. See README_DEPOSIT_SCHEMA for the schema, licensing logic and attribution requirements. The file-to-licence mapping is: occurrences_ca_ccby.csv — CC BY 4.0 (109,545 records); occurrences_ca_ccbync.csv — CC BY-NC 4.0 (103,070); occurrences_ca_ccbysa.csv — CC BY-SA 4.0 (5,978); occurrences_ca_ccbyncsa.csv — CC BY-NC-SA 4.0 (101); occurrences_ca_cc0.csv — CC0 1.0 (917). Zenodo attaches licences at record level rather than per file, so all five are listed together above; the binding mapping is the filename, the deposit_licence field carried on every record, and DEPOSIT_MANIFEST_2026-08-12.csv, which also gives the MD5 checksum of each file.