Platform (Android, iOS, Website): Website
Browser, if a website issue (Firefox, Chrome, etc) : Firefox
URLs (aka web addresses) of any relevant observations or pages: https://www.inaturalist.org/observations/116156134
Screenshots of what you are seeing (instructions for taking a screenshot on computers and mobile devices: https://www.take-a-screenshot.org/):
Description of problem (please provide a set of steps we can use to replicate the issue, and make as many as you need.):
Step 1: Go to observation
Step 2: Request help from CV with ID by putting cursor in species name field. Note suggestion Eleocharis quinqueflora noted as “visually similar/expected nearby”
Step 3: View observation map of this species. The nearest observation is many thousands of miles away.
See the Geomodel for this species: https://www.inaturalist.org/geo_model/76785/explain
For some reason, it includes part of SA.
Probably not a “bug” since the site is just pulling from the geomodel nad the geomodel expects this species in the area, so I’m going to move this to General. But it’s definitely evidence that the geomodel needs improvement. We took at look and there were no incorrectly IDed observations of this taxon in Africa in the training set, so that wasn’t the cause.
This is a prediction based on altitude. The model is selecting cells with an altitude of over ~1500m in Africa and South America and the Himalayas - even though there are no records in any of these areas. .
I’ve already stated elsewhere that I’m not a fan of the initial Geomodel. I have to be patient and view it an infant just beginning to take its first steps. Because of its simplicity (elevation), it is falling down a lot. IMHO it has been implemented too early in its development.
Elevation certainly seems to be a part of this, but there are probably other factors as well.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.