iNat used to id rare blue moth and paper published!

iNaturalist is truly amazing and users helped me identify a rare/rarely seen blue moth that hasn’t been recorded in Sumatra for 130 years.

I thought that forum members might be interested in the paper I published recently…which would not have been possible without iNaturalist and its amazing community. Thank you to to everyone who contributed to the id.



Congrats on the paper and thank you for sharing! This is a great story to show the value of iNat to science and conservation research.


Congratulations! I added the paper to the Publications wiki:


Thank you!

iNat is a truly amazing resource isnt it!


Thank you!

Im not v au fait with how to promote and publicise such things…other than writing a paper on it​:grin::joy:


“Species ARE nowadays being identified without voucher specimens and it is recommended that traditional taxonomic scientists engage with this ever expanding network and community (Marshall, 2018; Marshall & Evenhuis, 2015; Wilson et al., 2020).”… I have seen countless “research grade” observations which were completely wrong. Congrats on being published, but in many cases, species cannot be ID’d through photos, such as the bee genus Melissodes.


And I have seen countless “research grade” observations that were completely correct.

It’s not clear what contrast you are trying to establish. The fact that many species can’t be identified by photo doesn’t devalue iNaturalist as a tool for studying taxonomy. Furthermore, scientists using iNaturalist to study species that can’t be identified by photograph would almost certainly understand that they can’t rely on “research grade” images of such species for identification. And yet these photos may legitimately lead to a discovery about how to differentiate species previously thought to be indistinguishable by photos.


I have personal experience where iNat observations have lead to (or facilitated) species discovery and increased understanding of known species. It is a tremendously valuable resource. However, I agree with @cgwilson25’s point that this publication is attempting to swing the pendulum in one direction–in opposition to collecting specimens. So much so that it’s difficult to figure out what the purpose of the paper is–to add to our understanding of this rarely seen species or to dissuade the collection of voucher specimens (the majority of the paper focuses on the latter). Just as there are instances where we can learn about a species without collecting vouchers, there are many times more instances where we cannot learn about them without collecting vouchers. Do scientists now have to publish papers similar to this one but describing each and every instance in which an in-hand voucher is necessary? Will we be pressured to avoid collecting altogether out of fear of public outrage in the event that we might collect something that could be identified without a voucher? How will that impact our understanding of the millions of invertebrates that cannot be identified without dissection or microscopic examination? The value of iNaturalist should not be used as evidence for the devaluing of scientific vouchers–which appears to be the argument of this paper.


I don’t have an issue with collecting and I don’t think the authors do either. Voucher specimens and photos are both valuable and both should be used.

I read the paper and I don’t see anywhere that they make a blanket statement against collecting. They’re simply demonstrating that iNaturalist facilitated rediscovery of this species. They discuss the controversy surrounding collection and note that collection in this case may have interfered with rediscovery because part of the identification was based on behavior, which of course can only be observed while the specimen is alive. They encourage traditional taxonomic scientists to use iNat, but they don’t say they should stop collecting.

And yes, I do think scientists should have to justify collecting, especially for species known to be rare.


True, I suppose I overinterpreted. The practical problem with the “to collect or not to collect” debate is that before embarking on a field trip, a biologist would need to need to know, in advance, which species must be collected to ID and which species can be photographed. And there are thousands of species that a biologist may encounter. With rare exceptions, it’s just not feasible to have this knowledge before embarking on a field trip to study invertebrates. First hand experience with this…one day I was out photographing to post on iNat (I don’t always collect) and saw a wasp. I had to chase it down to get a photo. A year later an icheumonid specialist asked me about it, saying that it is likely a new species. But a photo isn’t sufficient to identify or describe a new species of ichneumonid wasp. That was the one that got away.

The point that iNat is valuable in understanding biodiversity could have been made very clearly without rehashing the “to collect or not to collect” debate.

One thing to keep in mind with photo-vouchers posted on iNat (and similar platforms), and which isn’t addressed in the paper, is that the person posting the observation can take it down at any time by just clicking a button. In this particular case, the person making the observation is the same person publishing the paper. And the photographs are reproduced in the paper (though perhaps not a high enough resolution to be ideal for identification). So at least there’s a “backup”.

1 Like

It seems to me that this is true now, with regards to all species. (I work with plants.) It’s simply real. When in doubt, I think we should collect (in most cases) if we have a reason to collect, but some people will disagree.

Agreed. I’ve been in Arkansas for a little under a year, and have collected around 15 new bee records just for the Ozarks region. In most cases, many small invertebrates cannot be IDed without a specimen in hand. I probably did over analyze the paper a bit, but this “to collect or not to collect” business is just foolish. Insects reproduce at such a high rate, taking 10 individuals from a population probably would do no harm to the species, and with how few people collect insects, there are likely more populations that haven’t been found.

1 Like

In most cases, at least in bees, you don’t know what you have until you get it under a scope. I’ve collect maaaany “rare” species that no one knew were in certain areas. I agree that I probably over analyzed the paper in regards to photographing vs collecting, but just because they don’t say something, doesn’t mean they don’t elude to it. Honestly, they probably didn’t even need to bring up the “to collect or not to collect” controversy in this paper, as its more interesting left as an cool new occurrence point for a “rare” species.

Requiring justification for collecting insects, in most cases, would just make it impossible to get any distribution work done. I’m finding that many species of bee have wider distributions than originally thought, and if I needed justification to collect specific bees, I wouldn’t be able to justify the collection of those “rare” bees because I didn’t know they were there. ← not sure how to word this last sentence lol, but hopefully it makes sense. Also, in most cases, insects reproduce quickly and easily, so removing 10-20 individuals would probably do nothing to a population.

1 Like

You’d hope that scientists would double check their data from iNat, but I feel most people just get their occurrences from gbif, and I am not sure how stringent they are with their double checking of a species level ID. Pictures and videos are great, but using voucher specimens in museums should be preferred by most scientists, at least if they’re serious.

As a side thought, a few bad occurrences might not mess up a regular SDM, but in maxent, which is presence only, if the person coding doesn’t prune out unlikely data points, their whole model, and thus our understanding of a species distribution, could be completely thrown off.

At the end of the day, I just took issue with their idea that photos could be just as good as a physical specimen. Sorry if this makes no sense its pretty late haha

It sounds like we’re on the same page. What you’re saying makes sense.

Most negative reviews I hear regarding “photo-only” observations are from experts in select groups, especially invertebrates, who are upset with how many misidentifications there are in their respective field on iNaturalist. This is reasonably fair, but it is not a portrayal of the platform or citizen science in general. And if you are an expert in this field, then the community would appreciate it if you helped out and corrected the mistakes. Just as any museum or herbarium welcomes experts to fix labels on their collections.

Issues with identification are not new to a citizen science platform and it is disruptive to me to see it used as a widespread argument against them. The point of the matter is that iNaturalist has enabled a far greater audience to access, post, document and record wildlife, and this greater audience by virtue contributes greater “incorrect” identifications just by the process of mathemetical growth.

Obviously, there are select areas where this is a greater issue, such as species that cannot be identified from photos alone, which can result in a larger proportion of incorrect or overly confident identifications. But in my experience, I’ve seen better identifications in most groups in general than I can find in a herbarium set from 20, 30 or 40 years ago.

As some would say, a museum collection is only as good as its curator and the experts that pay visits to the collections to evaluate and correct specimen labels. Sound familiar? It should; that’s what the entire iNaturalist platform runs on, community input and attention to working out identification.


I run into this often for species that can easily be identified with a photograph. And it’s a great argument for funding to digitize museum collections into a platform like iNat where identifications can be improved (and collections used) by a much larger community. But this would only help species that can be identified by photographs…and the photographs would need to be of high resolution.

1 Like

Only if you subscribe to the idea that every genetic lineage is a separate species. Since even the splitters will admit that there is no agreed-upon number of base pairs to define a species-level distinction, the premise is open to interpretation.

I was trying to say something that doesn’t seem dependent on whether every genetic lineage is considered a separate species, or not. (The identification problems are certainly worse if one takes that approach, of course!) However, the issue I was trying to respond to was the collect / don’t collect debate. I myself don’t know before a field trip what I will want to collect, photo, both, or neither. True, if I’m doing a research or monitoring problem, I know the main species I want to collect and/or photo, but there are always other interesting species out there that I might decide in the field to collect, and I may not know whether those species could be ID’d with photos alone or not. I would likely collect if I didn’t know.

Note: I collect a lot less than I used to! Why? Photos are a great record most of the time, and processing plant specimens and depositing them in appropriate herbaria takes more time than I like to spend. Yeah, photos!

1 Like

Congrats @greenhillsumatra on this important and exciting observation!
I found your paper to be well-done and especially the Conclusion. The Curators or staff may be interested in it, to aid in supporting the growth of iNat.