Using range to identify species
While knowledge of the ranges of various species is no doubt a useful aid in identification, I worry some that using it too much might lead to reduced knowledge of the true ranges of species. If we say "this is such-and-such species because it is the most common species in this area," we produce a self fulfilling prophecy. The most common species becomes even more common (at least according to our identifications). This reduces the likelihood that we gain knowledge about the ranges of similar looking (at least from photographs) species. Yes, I am also guilty of using this crutch. And, I know contributors of observations prefer knowing what species they've observed, even if the identification is tentative and based primarily on range. It's one thing to use range to identify observations of species that have widely non-overlapping ranges, say a species common to the montane west, versus a species of the Appalachians. Or highly local endemics, for instance salamanders endemic to specific caves. I'm more concerned about "the most likely of a few similar species in an area" identifications.
I don't have any good resolution to this dilemma (at least, I see it as a dilemma). Especially since contributors are free to use their best judgement when offering identifications. I certainly don't want to stifle use of the site, through over-thinking this thing.
Has this bugged anybody else, or it just me?
Comments
Great point, Lee. I have done this same thing SO many times -- I look at a map and see all of the dots and say, "Well, it's bound to be that..." And I don't even go through the keys...
I've found that when I ID something based primarily on the range, I see someone else's observation of a similar but different species, and I go back and modify my ID. Nonetheless, I still like to try to give an ID, as well as I can. I learn as I go, and "as I go" includes a lot of errors! :)
I am kinda weird -- I think it's good to make a lot of these sort of errors. Personally, it keeps me modest, of course. But I also think it's a good way to say, "This is what I think... what do you think?" with a verification or ID. It's a more inclusive ID process. It's better than a "just trust me -- I'm an expert" sort of explanation. :)
My two cents are worth 1.5 cents, by the way.
Yes, it is a potential issue, but it would be equally problematic not to take distribution into account. Judgement must be used, and all available evidence considered. Few, if any, identifications are ever 100% certain. In cases of supposedly allopatric cryptic species (which are fortunately rare), the only evidence other than place would be DNA!
I guess it only makes a difference if anyone uses iNat observations to try to delineate species' ranges, and has any faith that they might resemble reality. Certainly that is an issue with any data gathered by such a range of observers. I guess I must have been considering using the data in such a way. Otherwise, why would I care? Maybe "Research Grade" is a bit pretentious. Perhaps "More Research Required" would be a better tag. I'm not really complaining (all that much). It's just something that has been bugging me.
Anyway, thanks for the feedback!
OK, here's an example that makes my concern seem somewhat silly. Someone posts a photo of Couch's Kingbird from Corpus Christ, TX. Yes, it could be a Tropical Kingbird, but it's almost certainly a Couch's Kingbird. I would probably offer an ID of Couch's Kingbird without commenting on the fact that it could be Tropical Kingbird. I should probably at least comment about the similarity of the two species, and the likelihood of encountering one of the two species. Or just leave it at genus, and say "Couch's or Tropical" in the comment.
Hey Lee,
I'm curious -- which taxa do you 'subscribe to'? I notice you ID a LOT of the TX plants and critters -- are you subscribed to all observation from TX?
I subscribe to TX plants (although, I really don't know the western plants...or southern... or many eastern). :)
Or do you browse through the ID please section like me every now and then?
Think about it this way, using known range to ID species doesn't result in any falsely claimed range extensions, it might just fail to notice a new range extensions. So, it is a "conservative approach", which can be expected to work out just fine in most cases.
Sam, I set up a search that covers the ID Please observations for the midwest from South Dakota east to Indiana, south to the coast and west to about Coahuila.
Good point Stephen. I'm starting to calm down now.
Hi Lee, thanks for bringing up this point. I have had similar "concerns". I have wondered how often more cryptic species (I am thinking sedges) would get mis-identified under the research grade. I am subscribed to the bulrushes, but I am hesitant to make an ID because without certain features photographed or described I really would be making a guess and I know the bulrushes fairly well.
Here is how I have "calmed" down:
I like how Stephen put it...its a "conservative approach"
I have been asking those who made the observations more more details, to see if they remember seeing a key character trait, or sometimes they may have more pictures that they didn't think would be helpful with the ID
If the taxa is not something I am too confident in, sometimes I don't suggest an ID, but in the comments I suggest they look into a certain species
I have been encouraged by others on iNat (like Sam) to be more ok with best guesses to be part of a collaborative effort (so maybe something like Collaborated Grade would be better than Research Grade, but I am ok with Research Grade at this point)
I see an iNat record kind of like an herbarium record. If I were to go to a site (that I got off of an herbarium sheet) I would need to make sure of the identification at the site before I made a collection. Or put it this way, I have volunteers survey invasive plants for me on preserves, but before I do any treatment I always make sure of the ID, I have the final responsibility for identification. This would be the same for some instances of professional use of iNat data.
Lastly, iNat is great for getting anyone and everyone out making observations of the natural world, but it doesn't replace the need for professional field biologists or highly qualified naturalists to do surveys and inventories (this is the focus of a project of mine, www.RateMyLand.com).
My few cents!
- Derek
Add a Comment