In the past few days I’ve come across two situations where iNat users have misidentified hundreds of records because taxonomists split a species and naturalists in the field have not kept up with the changes:
The Blue Metalmark (Lasaia sula) had the subspecies peninsularis elevated to species status. This new species L. peninsularis includes the entire US population of the genus. However, the English-language common name has remained with the original species L. sula. The two species are difficult, perhaps impossible, to separate in macroscopic photos. As a result, American butterfly-watchers - many of whom are unfortunately oblivious to scientific names - have continued to identify, and elevate to “research grade”, US observations as L. sula. iNat now shows over 500 US records of L. sula, a species that does not occur in the USA, and only one of L. peninsularis, the species which makes up the US population and is actually featured in the aforementioned 500+.
The Two-barred Flasher (Astraptes fulgerator, in the process of being switched to genus Telegonus) was determined to involve a complex of several species which differ not only genetically, but use different host plants and look different as caterpillars but are visually identical as adults. The resulting taxonomic split assigned the US population to the new species Qian’s Flasher (Telegonus tongae); the name Two-barred Flasher still applies to species fulgerator, which is found only in South America. Similiar to the Blue Metalmark situation, butterfly-watchers have continued to identify their and others’ records as Two-barred Flashers in ignorance of their actually being Qian’s, such that iNat currently shows over 250 records of the South American species in the USA, and I think only 3 records of the species that actually occurs in the US. There appears to be an even bigger mess to resolve in Mexico and Central America…
Is there any quick fix with this? Some way to tag all of the misidentified records of a species in one geographic area, and re-identify them en masse to the correct taxon? Or is the only option to slog through the hundreds of erroneous IDs and try to fix them all one at a time?
When taxonomic changes are committed, some automatic changes of IDs can be done via substitutions/atlasing the split, though depending on the situation, this isn’t always possible and/or may not address all records (eg, in cases where species overlap). I’m not familiar enough with the situation to know whether this was possible or done for these taxa.
At this point, “manual”/individual IDs are the way to proceed. Depending on the issue, it may also be important to update common names for the taxa.
This should have been addressed during the taxon split using altasing. It may be possible to re-merge and re-split them, but this could be disruptive. My suggestion would be to raise this issue with prominent butterfly curators on iNaturalist. As cthawley mentioned, the usefulness and feasibility of this approach also depends on how much overlap there is between the species. And it sounds like someone needs to update the common names for the new species as well.
Has the taxon split been done in iNat? If not that would be the first step. As you noted though, you will still have the older species being ID’d for a while until the new intel becomes common knowledge.
My suggestion would be to start slamming out ID’s in the problem area and post a link to the paper’s that made the splits. Also, reaching out to the major butterfly identifiers in the region and asking for help in correcting may mean you’re not trying to do everything yourself. We had to do similar things with bumble bees due to the splits that have happened in the last decade.
But, when it really comes down to it, your best option in any case of mis-ID’s, is to put out a bunch of correct ID’s.
For the metalmark, that revision from subspecies to species was published less than a year ago, I believe and iNat made the switch this summer. So not surprising the records do not yet reflect that taxonomic revision.
BugGuide, which has the advantage of only having to deal with taxa found in the USA and Canada, already has all of its flasher records transferred to the new taxon. However, it still shows the metalmark as a subspecies, so its data is split up between records submitted at the species level (under the now-incorrect name) and those submitted at the subspecies (to which, AFAIK, all records north of the Mexico-USA border belong).
I can envision trouble from this. Picture a range map in which the boundary between two species exactly follows a geopolitical border. A butterfly who flies over the border wall thereby changes species.
Even if they don’t overlap in the sense of sympatry, a pair of parapatric species probably don’t observe their borders perfectly.
We are often advised not to identify based solely on range, yet that seems to be what atlasing entails.
I would argue that, whether we’re aware of it or not, a lot, if not most of the species-level IDs of small arthropods on iNat are range-based at some level. For many moth taxa, for example, there is some externally identical taxon found somewhere else in the world. They might differ in phenology, host plants, behavior, DNA, internal anatomy, larval characteristics, etc., but if all we have is a photo of an adult moth, range is a huge part of what name we apply to it on iNat. The species might not be split by taxonomists based on range, but if there are no differences visible in photos, yet the ranges are known to differ, I’d say it’s fair to make the call based entirely on the location of the observation. Otherwise, the majority of photos of adult moths would be unidentifiable due to sister taxa in different parts of the world looking the same in photos. Even in big showy taxa, this is the case. Would this photo not have been identified as a luna moth if it were submitted from Canada rather than Mexico? https://www.inaturalist.org/observations/231858337
I can see using location if the two very similar species are allopatric, or the location is from deep in the range of one but a long way from the other. But in parapatric species where it’s really impossible to say in the zone of contact which one it could be, I’d probably punt it to genus or species group/complex.