Where this data comes from
Every trait, interaction and measurement in THATSBees is traceable to a published dataset. This page starts with the totals and gets more detailed as you scroll down. Counts are live.
What the records are
Records in each part of the database, and the datasets they come from. A bee counts once per part however many records it has there.
Datasets
Every dataset that contributes, largest US coverage first. These are the datasets the species list can filter by. Species counts overlap between datasets, so they do not add up.
Floral data tiers
A species page's Floral Associations table counts plant records from every source, one column each, in three tiers:
- Published datasets: host lists and interaction compilations from peer-reviewed papers and data publications: Noori et al. (GloBI), Big-Bee, Bee-Gap, Fowler, BeeFunc. Noori and Big-Bee count records; Bee-Gap, Fowler and BeeFunc list host plants without a count, shown as a tick.
- iNaturalist: research-grade observations from the Pollinator Interactions on Plants (PIP) project that name the flower the bee was on. Crowdsourced, identified from photos.
- GBIF records: specimen and observation records whose associated-taxa field names a plant (iNaturalist excluded, since PIP covers it). Most say only that the bee was taken on or associated with the plant; a GBIF or iNaturalist association counts whether or not it says "on flower".
The tiers overlap. Noori compiles GloBI, which includes some iNaturalist and GBIF records, so a plant's Total can count one record twice.
How to cite
If you use data from THATSBees, cite the datasets it came from. Most licenses here (CC BY, MIT) require that credit; the rest ask for it.
Dataset detail
The full citation for each dataset and how its records are split. The code under
“Stored as” is the source_dataset value those records carry in the database.