GBIF occurrence data
Global Biodiversity Information Facility (GBIF Secretariat, Copenhagen)
Nearly 3.9 billion records of where and when a living thing was seen or collected.
GBIF aggregates species occurrence records from museums, herbaria, citizen-science platforms, surveys, and eDNA studies worldwide into one standardized index. Each record says which taxon was observed or collected, where, when, and by whom, with a license from the publishing institution. It is the starting point for species distribution modeling and range maps.
Access
- Cost
- FreeFree; a free account is needed for large downloads, which get a citable DOI.
- License
- Open licensePer data set: CC0, CC BY 4.0, or CC BY-NC 4.0
- Sign-up
- Free account
- How you get it
- API and Bulk downloadDarwin Core Archive, CSV, parquet
What's in it
- One row is
- One occurrence record (an observation, specimen, or sample of a taxon)
- Covers
- Worldwide land and sea
- Years
- Specimens from the 1700s to the present
- Updated
- Daily
- Size
- About 3.9 billion occurrence records (November 2026 API count)
Quality
The global standard for occurrence data with flags for common errors. Records are heavily biased toward rich countries, birds, and roadsides.
- Strong taxonomic and geographic sampling bias
- Some records carry non-commercial licenses
- Coordinate and identification errors remain
Connects to
- eBird Basic Dataset: eBird is the largest single contributor of GBIF records
- OBIS (Ocean Biodiversity Information System): Marine records shared between the two
- IUCN Red List of Threatened Species: Compare occurrences with assessed ranges and threat status
Questions it could answer
- Which regions have the fewest biodiversity records per square kilometer?
- How have recorded ranges of a species shifted northward over decades?