eBird Basic Dataset
Cornell Lab of Ornithology
Over a billion bird sightings logged by birders, with checklists that record what they did not see.
eBird is a citizen-science platform where birders submit checklists of species seen and heard, with time, location, effort, and whether the list is complete. The eBird Basic Dataset is the full download of those observations. Complete checklists let analysts infer absences, which makes eBird usable for abundance and trend modeling, and Cornell's Status and Trends products are built from it.
Access
- Cost
- By applicationNo fee, but you need an eBird account and an approved data request; terms bar commercial use without permission.
- License
- Provider's own termseBird Data Access Terms of Use
- Sign-up
- Application required
- How you get it
- API and Bulk downloadTab-delimited text
What's in it
- One row is
- One species observation on one checklist
- Covers
- Worldwide, densest in North America and Europe
- Years
- 2002-present, plus historical records entered later
- Updated
- Monthly
- Size
- Well over a billion observations; the full file is hundreds of gigabytes
Quality
Effort fields and automated plus expert review filters make it the strongest citizen-science bird data set. Effort is concentrated near cities and birding hotspots.
- Observer skill varies
- Spatial bias toward hotspots and roads
- Sensitive species locations are withheld
Connects to
- North American Breeding Bird Survey: Structured survey trends to compare with eBird-based trends
- GBIF occurrence data: eBird observations are shared to GBIF
- xeno-canto bird sound recordings: Song recordings for the same species
Questions it could answer
- How have spring arrival dates of migrant warblers shifted?
- Which species have expanded their winter range northward?