How do multi-species observations get made?

As has been discussed often on this forum, a fair number of new users will make an observation that has photos of a bunch of different photos of different organisms in it (via the mobile apps), like this example observation.

I suspect a decent number of people here might have done that when they first started out on iNat, which is a totally understandable mistake. As we’re working on a new mobile app, I’m trying to think about ways to reduce the occurrence of this issue, so I’m curious how people end up with these observations. It seems to me that there would be two common ways for this to happen:

  1. You make a new observation by taking a photo, then while still on the Observation Details screen, you tap on the + symbol to add more photos to the same observation as you’re walking around.

  2. You import a bunch of photos of various different organisms and the app turns them into one observation, which you then post.

Here’s a poll for you to fill out, if you like, and I’d love to hear comments or stories if you or someone you watched use iNat have done this. Thanks!

How did you make multiple-species observations?
  • I imported mulitple photos into the mobile app at once and they became one observation.
  • I took a photo and then kept pressing the “+” button to add more photos to the observation.
  • Some other process.
0 voters

How people do it, I don’t know, but why people do it: I always thought it was an Instagram habit. On that platform it’s normal to make one post about your outing and add several photos from various parts of it.


I answered the first option but my observation was corrected as soon as I saw the result. I assumed that the app worked the same way as the browser uploader. I figured out that it didn’t, fixed the observation and learned a lesson. All good.


I started with the desktop, rather than mobile. First time or two I made mobile observations I had selected multiple photos at once, thinking it’d work like it did on the PC and default to each one being a separate observation. Oops.


Yeah, me too. I think a lot of (or at least some) people expect it to create multiple separate observations at a time.


The web app on an iPad makes it fairly easy to do this during the drag-and-drop process for uploads.

1 Like

I’m on the same page as some of the other users in this discussion right now - people come from places like Instagram or Facebook where there is absolutely no problem including pictures of different things together in one big post. On iNat, of course, that is not the case because of how data compilation functions on the site. In the future, I think a mandatory tutorial that pops up as users start up the iNat app for the first time describing how they can avoid making those and similar mistakes would be amazing and save potentially thousands of observations from falling into either the void of the ‘Unknown’ or into improper categories (wrong locations, captives marked as wild etc.)


Some other process: I took photos of what I assumed were one species of parrot but were in fact a flock of parrots that fly together but are not all of the same species. Every photo is co-mingled. (They are impossible to separate, really, since I still cannot distinguish them. I am laughably terrible at birds.)


in this situation, this is fine, no need to separate out any photos. The best thing to do here if you wanted would be to duplicate the observation with the same photo.

What Tony is angling for is an observation where each photo shows something totally different, without the same species appearing across photos


Sometimes when I’m scrolling through the photo library in the app (iOS), I will select non-target photos by accident while searching for a specific photo. I’m not sure how common this is with other users.


This is exactly what I was going to say! (also iOS)


I edited to add in that last sentence and in doing so have misled (sorry!). I meant that I could not sort them out. That said, in some photos only one parrot is shown, in others two, but as I cannot suss out which is which, it is a multi-species observation albeit accidentally and due to user idiocy, not app error, which is why I replied.

It was three years ago, one of my first observations, though I could not say my parrot skills have evolved.

Sorry if I’m not meant to have answered. : /

1 Like

Could the CV generate a popup ?

You seem to have multiple species here …

(I think it is mostly an honest mistake, before they understand iNat. Usually new users, or perhaps someone struggling with internet connection while hiking)


I’ll confess that my bird recognition skills are equally poor and I have inadvertently done this as well.

Unintentional conflating different organisms happens not infrequently with hymenopterans – it can be quite tricky to keep track of who is who when there is a lot of activity on a particularly popular plant.

I fear this is not an error that can be prevented by better app design, however.

Users often do seem to have difficulties deleting photos using one or both of the current apps. I don’t know whether this is user error, non-obviousness of how to do so using the interface, or the app not working correctly, but this seems like addressing this end of the equation (making it easy to fix mistakes) is an important complement to reducing the number of mistakes made in the first place.


I still use a flip phone (I know) and do everything for iNat on the computer. So I think the only times I’ve ever uploaded multiple species to an observation was on like CNC’s where I’ve got 50 plus photos being uploaded and combined the wrong ones.

But another issue is the fact that splitting already combined photos is very difficult. It was quite a learning process for me and I constantly see issues where someone wants to remove problematic photos from an observation and has to be talked through step by step. It happens all the time where one of us tells someone that a photo needs to be removed and later they will say something like “thanks, I removed it”, and the photo is still there. So making that process more user friendly would also help.

Finally, there are times where the user doesn’t believe us when we tell them that there are multiple species in the observation. We had an incident where an observer had a multi-observation, and three identifiers all said to remove a particular photo, the observer called in a friend and the two of them started insulting everyone. I fired back and the observer blocked me. Allowing a way for identifiers to ID individual photos for cases like this would be useful.


I’ve had this happen also. Fortunately, this upload process was limited to 4 pictures, so it is relatively easy to catch.

But, this is a possibility for creating an observation with multiple photos of different organisms.

1 Like

Thanks for the replies, this is really insightful!

That’s totally understandable.

Yeah, the photo picker (which is the native iOS photo picker) is pretty sensitive and this happens to me sometimes. For what it’s worth, you can always choose to see the photos you’ve selected by tapping this at the bottom:

The good thing about this photo picker is that it’s at the system level, meaning that what you’re seeing is coming from Apple. It’s only when you tap on Add that iNat sees the photos you’ve picked. iNat itself doesn’t access your photo library, so it’s a method that protects your privacy.

Currently CV only checks one photo at a time, it won’t be checking all photos in an observation on its won. Maybe something that could eventually be used, but not anytime soon.


Then we need a new DQA Multiple species, which human identifiers can tick. And which will tip it out of Needs ID until the observer splits it.


Funny thing about that example observation: the highlighted photo, by itself, has multiple species.

I didn’t get as far as completing the upload, but yes, I had that same confusion.

1 Like

I suspect that the case of multiple photos of different organisms is most likely very new users who have not yet been educated about how iNat works, and that an ‘observation’ should only contain a single organism. Perhaps it’s just a matter of clearer explanation that targets new users. Otherwise it’s likely accidental and can happen to anyone.

Secondly, and I don’t think this relates directly to your question, but in relation to the above example with two species in the one photo, I would like to see the image crop tools being used much more. While photos of interactions between species are often interesting, perhaps the user could be encouraged to crop, or maybe just place a marker/box on the image of the one they want ID’d.