Hello, I’ve been following Fagus sylvatica observations and quite often this species has been reported from the USA. I’ve been puzzled by this and today I checked out several such observations where either leaves or nuts were well visible to identify the species and many of them were actually Fagus grandifolia, a native American species (F. sylvatica is native to Europe).
My question is thus for those who know more about US forestry: Is the European F. sylvatica cultivated in the USA (in a larger scope) in forests? Is it commonly planted in parks? How likely is it many observations from the USA where only the tree is visible (without a detailed view of the leaves/nuts) are actually F. grandifolia and F. sylvatica was perhaps only suggested by the auto ID? If it is the case, can the auto ID be tweaked to suggest F. grandifolia for observations from the USA if the exact species cannot be clearly identified from a photo?
I’m just curious. I’ve never met F. grandifolia in Europe in a forest or a park, only in specialized arboreta.
AI suggestions can’t change, other than not show “seen nearby” if you clean the taxon, if you’d check spiders, insects or shrooms for an example you’ll see hundreds of american species recorded in Eurasia when there’s no data for them being there, it just happens with similar looking taxons when one is over-presented on iNat, like New Zealand species with 3k records will be suggested all over the world.
You can see if it was a suggested id by a little symbol in the upper right corner of the id.
I only ever see it as an ornamental or specimen tree out here. I think you’d be pretty safe correcting F. sylvatica to F. grandifolia for any wild specimen in eastern NA. People do plant F. sylvatica in their yards occasionally, but as far as I know, it doesn’t escape from cultivation.
Most likely, any wild plants in the USA are misidentified. You may want to check the IDs via the website, because there is a small icon which designates IDs chosen from AI suggestions.
You can change the AI suggestions for identifiable specimens by correcting incorrect IDs and adding IDs for yet-unidentifed observations. The next AI cycle (which is about twice a year) will use the new data.
Thanks. I don’t use the AI suggestions myself, I only put an ID to other people’s observations when I can recognize the species. I was asking this question as I suspected F. grandifolia ID is put there by iNaturalist because it was suggested by the AI.
Thank you, that answers my question whether it’s planted in US forests. I’ve been changing the ID only when I was sure from detailed photos they were F. grandifolia so I assume I can do the same for observation coming from “wild” forests (as determined from the map).
OK, thanks. I can see the observers often change their ID to what I suggest so if I continue doing so I might trigger the AI to start correctly suggesting F. grandifolia as the most likely taxon for US beeches.
The only valid Fagus sylvatica observations that I’m aware of in my area (Western NC/Southern Appalachians) is a planted tree on our campus and I made sure it’s marked cultivated. I’ve never come across one growing “wild” in the woods.
Yes, we have lots of F. grandifolia here in eastern North America, but I have never once seen an ‘escaped’ F. sylvatica, not including root suckers from planted trees. As others have said, any wild tree, particularly away from extremely disturbed suburban and urban areas where the other species has been planted, is virtually guaranteed to be the native F. grandifolia.
@fffffffff Perhaps I’m not being very observant: Could you offer a screen capture of an example with the “little symbol in the upper right corner of the id” where AI was used? I’ll look for examples but I’m not seeing it.
p.s. That would seem to be a useful check box to have on the Filters page:
" ___ ID based (solely) on iNat Computer Vision" or something like that.
If a user has selected a computer vision suggestion for their identification, a small icon will appear in the upper right of their ID when viewing iNaturalist through a web browser. Clicking the icon reveals additional information about computer vision.
Note: It is not possible to tell whether the user selected a computer vision suggestion because they are following the suggestion versus whether they are simply using the tool as an “autofill” to save time and effort typing out species names.
I occasionally see a similar issue with Platanus, where you might see the European one as a street tree in US but not on a random riverbank in the wild. In those cases I just put the “typical” localized id with the comment, “might it be this one instead”? Often there’s a quick fix, or the obs can safely stay at genus for a while.
