How these pages are made

Where the data comes from, and where it stops being able to tell you anything.

What a record is

A georeferenced occurrence record: a herbarium sheet, a museum specimen, or a photograph submitted with a location. All of them come from GBIF, which aggregates iNaturalist, university herbaria and government collections. Records based on a living specimen, a material sample or a fossil are excluded: a plant in a botanical garden is a place somebody put it.

The map

A shaded square means at least one usable record falls inside it, and darker squares hold more records. A filled dot is a herbarium specimen: a pressed sheet somebody can re-examine. A hollow ring is a field observation, identified by community agreement. The buttons show either alone.

Of 171,896 marks, 24,726 are specimens and 147,170 are observations.

Three things the map does not say:

• Darker is not more plant. It is more collecting. The squares over Calgary and Edmonton are dark for nearly every species here.
• Empty is not absent. Unsurveyed and absent look identical from here.
• Recorded is not native. A garden escape can accumulate hundreds of records.

A record is drawn only if its coordinate is precise to within 10 km, it falls inside Alberta or Saskatchewan, and its dataset licence permits redrawing the coordinate. We accept non-commercial licences for coordinates but not for photographs: a photograph is somebody's work, a coordinate is a fact about a place. Records closer together than 0.01 degrees are drawn once, which is under half a pixel.

"Native to", and when we leave it blank

Which provinces a species is native to comes from VASCAN, the Database of Vascular Plants of Canada. 414 of 430 species carry an answer from it.

The other 16 read Not established and show no province list. VASCAN records some of them as introduced here, carries some only under a name this catalogue does not use, and does not carry a few at all. Each needs a second source before we can say anything.

We would rather show you nothing than show you a guess.

The ecoregions

Below its range map, each species lists the ecoregions its records fall in, which is how you get from one plant to what else grows alongside it. A region needs at least 3 records to be listed: two can be a misidentified sheet and a garden escape. This cuts against rare species, which is a real cost. A region missing from a list means nobody has recorded the plant there.

How current this is

A snapshot, taken once and shipped. Retrieved on:

GBIF occurrence search, retrieved 2026-08-25

GBIF changes daily and this page does not. The GBIF and iNaturalist links on every species page are live; follow those for the current picture.

Known limits

• Misidentifications are in this data. We do not require records to be identification-verified: that field is absent from most herbarium material, so requiring it would discard the best-determined specimens to exclude very little.
• Where two regions meet, a few records are counted in both. The outlines are simplified to roughly 900 m and each is simplified on its own, so neighbours overlap by a sliver along a shared border. That inflates the region totals by about eight records in a thousand, most of it around Calgary where Aspen Parkland and Fescue Grassland cross.
• Flower colour is mostly unverified. 36 of 430 species have a colour checkable against the plant's own name; the rest are a genus-level default and say not verified on the page.
• Coverage stops at the Saskatchewan border. Manitoba shares several of these ecoregions and is not included.

Corrections

These pages are wrong in places, and the useful thing to do with an error is report it. The catalogue and its data are public, including the record of species removed for not being native here and the authority for each removal.