I’m trying to ID all the birds that need ID in California, a Sisyphean task in which the queue seemingly gets bigger as I ID birds…
One thing I struggle with is that a decent percentage of the observations more than a few months old are simply unidentifiable. Often these are cell phone shots of a bird over 50 meters away, showing a small blob that is identifiable as a bird only because it’s in the air, and probably not an insect. Or a blurry shot of a tree across a field with a single pixel that’s likely a bird. Sometimes there will be a specific ID and a comment to the effect of “I saw it better, but this is the only photo I got”.
No one can confirm the ID just based on the photo, so the observation will sit forever in the ID queue unless someone A) gives a general ID of “bird” and checks the “no it can’t be improved box” B) blindly trusts the original ID and agrees with it or C) checks the No option for “evidence of organism” or D) checks the “Reviewed” box and moves on
Option A means the observation will be casual at order level. Option B makes it research grade. Option C makes it casual, but with the original ID still attached. Option D means that the review queue will get longer and longer, and identifiable observations needing ID will be harder to find.
Personally I would prefer option C, but is it appropriate when there is a photo that shows AN organism, it’s just impossible to say what that organism is?
If it can be reasonably assumed that the observation does indeed does show a bird, use option A. If it cannot, use option C. [Update based on tiwane’s response: If it cannot, do something similar to option A, but ID to “Life” instead of “Birds”.] I almost always leave a note explaining why I’m doing it, explaining that iNat is a community platform and observations should be reasonably verifiable by the community.
Option A, placing a higher level ID like “Birds” or “New World Warblers” or genus “Egretta” is an appropriate response, but I always urge caution about checking the “No, it cannot be improved” option. IMHO, that is a loaded decision that should only be made be experts with years of experience in field and photo identification. It’s not meant as a general inquiry, “Can you, as a casual observer or non-expert, identify this organism based on the evidence?” That’s not the metric it intends to assess.
Thanks, I forgot about that meaning of the word placeholder in iNaturalist.
I used that phrase because the type of observation I’m referring to when I consider using option C are ones where the observer is confident about the ID of the organism, and is probably correct, but the photo doesn’t show anything beyond a blurry dot or a branch where the organism is hiding. Marking is as just “Birds” defeats the entire purpose of the user uploading the observation. I guess those users could simply opt out of community ID for those observations, but most casual users probably don’t even know that’s an option.
I agree, “No, it cannot be improved” is an option I very rarely resort to. Even in cases where species are considered inseparable based on photos, sometimes further research reveals some previously unknown character that allows them to be separated. And I’m always impressed by experts’ abilities to come up with IDs based on very little information.
Definitely don’t do B or C, unless it’s clearly not evidence of an organism. That DQA should only be used if the photo is of a rock or a sunset, or something like that.
I think specifying a higher level taxon and disagreeing with the original ID is the best option if you believe there is insufficient evidence in the photo to get it down to species. Doing that by itself will still leave open the possibility that someone with more expertise than yourself will come along and be able to identify it. Whether to check the “cannot be improved” DQA is a tricky question, and in general I agree that it’s best done only sparingly by experts-- however there are some cases (single pixel or trust-me-it-was-here type of observations) that might warrant a “cannot be improved” annotation by non-experts.
For photos that are identified as a bird but don’t show a bird, but do show a tree that formerly had a bird in it, or an empty field, should those be IDed as “Plant”? I’ve been tagging them as not containing evidence of organism, since the OP clearly didn’t intend the observation to be of the plant, but there is no evidence of a bird in the photo, as far as I can tell.
For example, here’s an observation that I just created to illustrate the issue. I come across observations like this fairly often, especially around the start of spring and fall semester when classes require students to start an iNat account. There’s a Hooded Oriole in this tree, but it’s not visible in the photo: https://www.inaturalist.org/observations/237013887
if the evidence of the organism is not visible/audible in the provided media, the observation should be marked as not containing evidence of the organism
Agreed, but it is inefficient for identifiers. I wonder if the best course might be to ask observers to mark it “cannot be improved” when they think there isn’t enough evidence, and let experts come through these as they see fit. (Maybe they can filter the search to “cannot_be_observed flags == 1” vs “cannot_be_observed flags > 1”? Not sure.)