Several of my observations have lost their photo

Bug Reports
1

Platform: Web

URLs (aka web addresses) of any relevant observations or pages:

https://www.inaturalist.org/observations/identify?reviewed=any&quality_grade=needs_id%2Ccasual&sounds=false&photos=false&user_id=756324&place_id=any

Screenshots of what you are seeing:

Screen Shot 2026-05-15 at 9.25.04 PM
Screen Shot 2026-05-15 at 9.25.04 PM2360×1462 327 KB

Description of problem:

I’ve never intentionally uploaded observations without media, so when I searched for my medialess observations and found several, I at first assumed that I must have accidentally deleted their photos. However, some of these observations previously had media, such as this one: https://www.inaturalist.org/observations/49725542

which has an agreeing identification and was added by someone to a project for observations that “must have media.”

I can see a few possibilities here. One is that I have on several occasions accidentally deleted all media from individual observations. This seems unlikely, as I almost never delete media, and can’t remember ever deleting all media either accidentally or intentionally. Another possibility is that through some bug the media from these observations was lost. It is also possible that for some of these the media failed to upload but the location, date, time, and identification was recorded.

Please advise. Thank you.

2

They do seem to have occurred in a few batches (ie, observations from same day/place that were likely uploaded together), which suggests that there may have been something with the upload process in those cases. You have a lot of observations, so there could be multiple causes (upload issues for some, something else for others)

There are some related bug reports you could check:

https://forum.inaturalist.org/t/media-missing-from-my-observation/28408

https://forum.inaturalist.org/t/photos-from-observations-dissapeared/11310

https://forum.inaturalist.org/t/photos-disappeared-from-research-grade-observation/14673

3

I had one do it before I submitted the observations. When I checked it had disappeared and had to drag it in.

I have has it on other occasions but switched to the other photo.

4

only this one out of the 12 has an identification other than yours, and it’s also only 1 of 2 that had an update some time after upload. see https://jumear.github.io/stirfry/iNatAPIv1_observations?user_id=dlevitis&photos=false&sounds=false&options=idextra,

i would guess that in most of these cases, you never had a photo on these observations at all.

i have copies of the AWS Open Dataset metadata from April 2021 and July 2022. unfortunately, it doesn’t look like you provided a default photo license until after April 2021, and you didn’t apply the license choice retroactively until sometime after July 2022. so only a small set of your observations created between April 2021 and July 2022 are represented in the July 2022 copy.

here’s one of your observations missing photos created on 6 May 2021: https://www.inaturalist.org/observations/77707436. here’s another observation created the same day that represents a miner in the plant of the previous observation, i think: https://www.inaturalist.org/observations/77707434. what i see in the July 2022 AWS Open Dateset metadata is that the miner observation exists, but the plant does not. so then that supports the idea that the photos for the plant observation were already missing by July 2022.

one possible thing that could have happened is that you dragged the photo from your plant observation over to your miner observation without duplicating it. i’d guess it might also be possible that you had a photo that was duplicated attached to both observations, and you wanted to delete the link to the photo on the miner photo but unintentionally deleted the photo altogether, which made it unavailable for both observations. or maybe the photo failed to load, and you never noticed.

currently Lathyrus vestitus observations are about 99% verifiable, and 75% of the verifiable are research grade. so i would have expected someone to come along relatively quickly to add an identification if the photo had been there.

at this point, i would guess that the turkey observation (the one with the identification) is the only one really worth trying to dig into, but unless anyone has copies of data from back in the day, i doubt anyone can realistically find out anything about it. my guess is that it’s 50/50 that the observation never had a photo to begin with. looking at other observations in the set loaded that day, it looks like 2 that were created from photos from a smartphone or tablet device are missing locations and datetimestamps. so if you missed that those observations were malformed, you probably missed that the other one was missing a photo. see: https://www.inaturalist.org/observations?user_id=dlevitis&place_id=any&verifiable=any&created_on=2020-6-15.

but if you can find a copy of the iNat to GBIF DWCA export file from just after the second identification was added, then you can see whether it contained your observation or not. if not, that’s more evidence that the observation just never had a photo. if so, then that’s evidence that the observation did have a photo, but i’m not sure there’s much that anyone could do to get additional meaningful information beyond that. in other words, even if there was a bug – which it’s not clear there was – it’s not clear that anyone could figure out anything useful about the bug at this point.