Lost my 1st observation in iNat

Yesterday, I was asked by friends when I did enter my first observation in iNaturalist and as I didn’t know the exact date but remembered clearly what species it was (a rare species of dragonfly from Senegal), I did a quick search to find it and what was my surprise as I wasn’t able to find it.
The species well exists in iNat and is illustrated by my picture but shows an observation count of zero !

Here’s the species : https://www.inaturalist.org/taxa/109412-Phyllogomphus-pseudoccidentalis.

The observation was made on 28th november 2015 and it will be quite simple to put it again if nothing is done to call it back.

Is it the only lost observation ?? Having more than 23000 observation to date, it will be quite hard to check it out for me.

1 Like

Sounds odd all right. Did you try your Calendar? From your profile icon on iNat choose Calendar and try that date

Hi Benoit. Are you looking for the first observation that you added to iNaturalist? Or for your earliest observation on iNaturalist?

From what I can see, the first observation that you uploaded was this Toucan in Honduras, on 18 March 2021. The earliest observation date for any of your observations is for this Warbler on 28 August 2001.

Your profile shows that your account was created on 3 May 2016, so there is a gap of nearly five years between that date and the first upload currently associated with your account. I guess it’s possible that intervening observations were deleted by you or by some bug.

Alternatively, did you perhaps use a different account to upload observations prior to March 2021?


Early on with iNat (maybe 8-9 years ago), I had two accounts for a while and didn’t know it. When I figured that out I asked the staff if they could merge the two and they did.

1 Like

For a long time, I had duplicate accounts without realizing it, one with a Google-sign in and one with a traditional username and password

There was also a bug in the iOS app a while back that could cause observations to be accidentally deleted: https://github.com/inaturalist/INaturalistIOS/issues/499.

My first thought would also be a weird account issue
However, the page for the photo is linked to the current account:
Which would suggest (but perhaps not prove) that the observation was uploaded by the current account.

1 Like

looking at old copies of the AWS open data set metadata that i have on my machine, i can see that the observation (UUID dd61138f-b5a5-44ea-8483-ce29ad01a856) did in fact exist as of 2021-04-15 but was gone by 2022-07-27.

the only other information captured in the AWS metadata is:

  • lat = 13.350038
  • long = -13.375983
  • positional accuracy (m) = 258
  • taxon = 109412 (Phyllogomphus pseudoccidentalis)
  • quality grade = research
  • observation date = 2015-11-28
  • observer = 229502 (benoit_segerer)

the associated photo was the photo record that is used for the taxon page that cthawley pointed out:

i can’t tell you how the observation was deleted though. i don’t see a copy of it in GBIF either. it’s possible that iNat staff might be able to look at old transaction logs to see what triggered the deletion of the observation, though they might need to narrow down possible dates to something smaller than a 15-month date range. if someone wanted to do the research, they could download old copies of the observation metadata file from the AWS data set to see if it exists at, say, 2021-12, then (2021-08 or 2022-03), and so on…


I would also recommend that @benoit_segerer emails help@inaturalist.org with a link to this discussion thread so that iNat staff are aware of the issue.

I brought this up to our developers, I’ll see if they can find anything. Our logs are only stored for about 45 days, though.

1 Like

@benoit_segerer according to our devs, observations 3089428 and 116480471 were deleted on May 26, 2022. 3089428 is a siginificantly lower number than your other observations’ IDs, and if I go to the observation that has an ID number that’s 1 higher (3089429 - https://www.inaturalist.org/observations/3089429), it looks like it was uploaded on May 3, 2016, meaning that 3089428 was almost certainly uploaded that day as well. Your account was created on May 3rd, 2016, which seems like the date when you’d make your first observation.

So I think the evidence points to the observation being deleted on May 26th of this year. Our logs don’t show how it was delteted, but we don’t have any reason to believe it was deleted by anyone but you. Do you think it might have been done by mistake?

If an observation’s photo is being used for a taxon, then the photo will remain on iNat if the observation associated with it is deleted.


Thank you for your precise answer.
The fact that I have accidentally deleted the observation is the most probable scenario but how I managed to do that is a mystery.
Is there a way to undelete it in order to keep the upload date ?
I clearly remember to have created my account just to upload this particular species and I started to use iNat more intensively only last year.


No, there isn’t.

not sure if this helps, but if you were to create a new observation without a photo, there is a way to reassociate the existing photo record to your new observation. so even if you didn’t have the old submission date on the observation to signify that it was your first observation, the observation would be associated with your first uploaded photo record.


is it / can it be known how far in time the deletions were separated, down to the millisecond? that might provide an idea of whether the deletions could have been done by the same action / process, or whether they were done individually. (if they were done together, it seems like it’s more likely that that some sort of bug or user error could have resulted in the deletion of that first observation, in the process of trying to delete the other observation.)

1 Like

Ok. It’s not so much a problem as I will re-upload the observation and I still will have the date in mind as it is the date of the creation of my account.
Thanks for your help anyway.

1 Like

My new “1st observation in iNaturalist” is here : https://www.inaturalist.org/observations/143353782


@tiwane I think this same thing just happened to me a few minutes ago. I deleted a Tragopogon dubius observation and who knows what else. Here are the steps as best as I can recall.

I was deleting some older Glechoma hederacea observations via the Android app. I’ve found that in general when I delete an observation, it still shows up on the page (and other pages) and in my counts until I close out the app and relaunch it. Well, this time, a few of the observations I had just deleted still showed up even after relaunching. I went to delete them (again) and after redeleting a few I started to notice that when I went to the edit page for these previously deleted observations, the photos that were appearing were for observations other than what I’d selected. When I noticed this, Tragopogon dubius, which I think was my very first observation, is what came up, but I moved ahead anyway and deleted the observation. For the next observation I edited in order to redelete, an Oxalis observation came up (which happens to now by my first observation in place of T. dubius) and I decided to stop. I checked a few minutes later on my PC and sure enough my first Tragopogon dubius observation is gone.

Unfortunately, I don’t know what all I ended up actually deleting, nor do I likely have the photos any longer. :(

I understand there’s likely nothing that can be done to salvage whatever I just deleted, but hopefully this helps to replicate further, as it appears that this issue (or something like it) still exists.

We might be able to get some of the observations back from our saturday backup. I do see the Tragopogon dubius observation there.

can you please send log files from the day you did this? To send log files, go to the About tab in the app and tap three times on the version number. You will then have the option to email the log files to us.

1 Like

Really? That’s great news! I just emailed you all the log files from today. Anything you can do is appreciated. Otherwise, even if they can’t be recovered, at the very least hopefully this helps pin down the root cause.