So I’ve just started identifying and I have a question concerning identification and species ranges. I take issue with the act of identifying cryptic species to species level just because of the location, which is something i’ve seen many identifiers and observers do. I get the temptation - especially for hobbyists who just want to fill out a personal checklist - but to me iNat is much more than that. Species ranges change over time, with climate change and human introductions now faster than ever. iNat should be a platform to challenge known species ranges, not be bound by them. I think IDing something based on location defeats the whole purpose of iNaturalist as a citizen science platform, so I’d really like to know what others think about this…
That issue seems to split across animal or plant. From many heated previous discussions. Botanists do. Zoologists don’t. In a nutshell (or whatever zoologists use).
If you have a foto from Nepal of an animal vanishing into the forest and all you can clearly see is an elephants tail. What species would you suggest?
Exactely. Location is important in order to ID iNat observations and there is nothing wrong with it.
Certainly as soon as two or more similar species have been found in the area where the foto was made (and the foto doesn’t allow for telling them apart) than species level ID based on location is no longer possible.
Oh they do it even worse. Nobody can identify a few coloured blurry pixels faster than ornithologists… and yes they use location to do this, else they wouldn’t have a chance.
The more observations we have from a certain area or species, the more likely are we to find these range expansions.
And for the suggestion that this elephant in the forest of Nepal is an African species, we use Occam’s razor.
I agree that there can be issues with using location to ID, but this is a matter of degree, not black or white. For instance, if there is a blurry picture of a lizard from an area where only one species of lizard is known to occur, it’s reasonable to ID as that species. If there are many lookalike/cryptic species and an observation of one of them is from some arbitrarily small distance where another of the cryptic species is known to occur (e.g., five miles from the putative break between the two), I don’t think it’s reasonable to ID to species - there’s a reasonable chance that the organism could be multiple species.
But this will always be a judgment call about what “reasonable” means. IDers will have to take into account how well we know species boundaries, how vagile the organism is, etc.
All that said, on iNat, I’ve generally seen more specific IDing based on location that I felt was unjustified than people being too conservative. A personal pet peeve is subspecies IDs based on solely on location, as these IDs don’t effectively add any information and can obscure novel changes (i.e., cases where subspecies overlap or boundaries don’t actually align with current understanding or are in a state of flux).
In some cases location is the only way to differentiate species without capturing and examining them up close. Take the Amur and Siberian Stonechats (split into different species a few years ago), the physical differences are really difficult to tell if you’re not seeing them side-by-side, and even then it’s still difficult, so location is one of the best ways of quickly telling them apart. It’s not perfect, but it works most of the time. That’s a mild example, but there are more extreme ones.
It’s absolutely true that species to change range. At a population level this usually means expanding into a new area that’s relatively adjacent, but there can be outliers, especially in flying (including aerial seed dispersion in this category) and aquatic organisms, and there is also the human mediated range shift. iNat helps track this in part because of the mix of location based aspect of identifications and the community based identification system.
Another instance when morphology isn’t always helpful but location is is when there are geographic gaps. In the Western Ghats (India), we’re often forced to ID some reptiles down to species almost solely based on location/elevation as different species share some morphological traits that are visible in photos (scalation). We ID to genus only when ranges overlap. But that’s the best we’ve got at the moment, or we lose the ability to differentiate many species.
As others have said, it’s subjective even if not perhaps the most accurate means of identifying species.
Absolutely no birder in the history of birding has ever distinguished that snipe from every snipe species in the entire world every time they identify one in the field.
Okay, maybe you could find one person who does that, but I bet their life list is really short.
We all make some reasonable assumptions based on location. It’s just a matter of where you draw the line. For me, personally, it’s a matter of knowing the organism, and knowing how likely it is for its range to be expanding in that way.
For example, nobody in North America needs to worry about distinguishing Wilson’s Snipe from African Snipe. That’s not happening. But if you’re in Alaska, Common Snipe might be a real possibility.
Hi all,
Thank you everyone for your replies! I guess the topic is nuanced and there is no right answer to cover all cases. i appreciate the fun examples, though they aren’t exactly comparable to the case i was thinking of (which I should have specified): For an observation in Czechia, someone identified a blurry image on the basis that it’s the only species of that genus to exist in czechia, despite there being a couple of observations of the alternative species just 220km to the west in southern Germany. to me, that seems just too close to rule out the alternative species, so I wanted to check with the forum to see what the standard is for identifying based on location. Now I realise though that since there is no standard, I should probably just take this up with the identifier instead of posting here.
perhaps “In a tortoise shell”? 
After fires we see too many sad tortoise shells
For your linked obs it is the only sp on the checklist.
Timothy,
Have you read through this thread? https://forum.inaturalist.org/t/why-identifying-by-range-is-a-problem/
@petezani started it in 2024. Pete was actually talking about something else:
But the thread turned into a discussion about identifying based on location, and it gathered a lot of thoughtful comments from experienced identifiers.
when you provide an ID, you aren’t saying “this is definitively what this organism is”. That’s not the case with photo ID and it’s not the case with experts in a lab with a specimen in front of them. Rather, when you make an ID you are saying “based on the available evidence and our current understanding of what defines this taxon, it is probably this”. Location is a valid form of evidence that can be useful in delineating species which would otherwise be tricky. Is location-based ID foolproof? No, but as long as you keep in mind the possibility of vagrants and a willingness to reconsider an ID when additional evidence is presented there shouldn’t be a problem
Не вижу проблемы.
Когда я вижу наблюдение за растением на клумбе или в горшке я просто прокручиваю предложения ИИ в самый низ и нажимаю на “Включать предложения, не встречающиеся поблизости” - и выбираю из нового списка вид, если он в нём есть)
А вот когда другой человек ставит таксон от балды, зачастую в принципе не встречающийся в моём регионе - это дополнительная раздражающая работа.
Like others, I consider location when identifying, with discretion of course. I work with ladybugs and there are many identical species that are broadly separated geographically or by habitat, allowing for reasonable species ID.
That said, relying only on distribution can be risky. There are gaps in knowledge in the literature and many range maps are at least somewhat approximate. Occasionally, species are shown to exist in areas where there are no formal records.
So, whenever using distribution, be aware of the possible limitations in what we currently know. Species may inhabit regions not currently documented. It also helps to know a bit about the species’ biology to accurately assess its range and possible new locations for it to appear.
Having this belief that there is more to be discovered allows us to expand our knowledge of how species are distributed. And iNaturalist is key for doing this!
Continuing the discussion from Why identifying ‘by range’ is a problem:
I stand by this statement, and to echo what others have said, it’s a nuanced question without a black-and-white answer. Ignore ranges entirely, and virtually nothing is identifiable to species. Rely too dogmatically on published ranges, and you’ll never find anything new and unexpected. Neither of those takes is helpful, but everyone has their own line as to where to draw the line between them.
As someone who attempts to ID subspecies when possible, range is often important to be able to do so at all: many subspecies are described based on range, plus a few factors that can’t possibly be determined by a photo (especially if it’s just a matter of genetics).
This thread seems pretty positive about IDing using location when it comes to the species level, so I don’t see why the same can’t be applied to subspecies. It’s of course reasonable to avoid IDing to subspecies if there’s a reasonable doubt, but there are a lot of cases where people will disagree with a subspecies ID because it hypothetically could be some other subspecies, even if the other subspecies isn’t known to be found any closer than hundreds or even thousands of miles away.
What ends up happening is that some observations randomly get to subspecies while some randomly get disagreed with, so the subspecies data still isn’t useful for anything.
This
is essentially my point. If the ID to subspecies is made solely based on location, i.e. Species X in Location Y = Subspecies XY, then the subspecies ID adds no information. We already know that the observation is Species X in Location Y - calling it Subspecies XY instead of Species X adds no information.
But subspecies boundaries are often imperfectly known. Known boundaries change over time (e.g., after a guidebook was written). Individuals disperse or are moved by humans. In situations where location is the only factor in IDing to subspecies, there’s not really any benefit (as above) and there is a clear potential cost - ignoring the potential that another subspecies might be correct and IDing incorrectly. Observations of an unexpected (based on range/location) subspecies would be much more valuable - they could indicate a previously incorrect understanding of ranges, that ranges are shifting with climate change, that species have dispersal vectors we don’t understand, that species are being trafficked by humans, etc. But if IDers reflexively identify based solely on location, then these potentially valuable observations may just be lost in a sea of assumptions.
I think the difference is that checklists, leaderboards, etc. are largely “species”- based. So arguably, there’s not much of a point to distinguishing between geographically-distinct subspecies. i.e. we already know that all the Silver-spotted Skippers in Maine are the “Eastern” subspecies based on location, so labeling them all “Eastern” based on them being in Maine doesn’t add anything to the observation, nor does it “add” any species to anyone’s lists or impact the CV training. I don’t see a problem with it, but I can see an argument being made that it’s not really useful.
… until a taxon change happens, and curators have to move said subspecies to a species of its own. Oh, wait… nobody cared to ID to subspecies because it didn’t “add anything” back then. Then : atlasing needed (which amounts to a postponed, massive “ID based on location”).
In the world of birds/ornithology, subspecies are often designated where there are specifically phenotypic differences regardless of what the genetics say. These differences can be related to diet, behavior, sound… even if they all LOOK identical. In which case, identifying these subspecies by photos and range alone would potentially obscure relevant data about vagrants or limit our understanding of subspecies ranges.