"Gravediggers" annotation tips

Issue: Many people don’t like stumbling across observations of dead organisms.
Solution: Such observations can be annotated as Dead so that people can screen them out. Some people may be interested in doing that annotation step.

Issue: It is typically difficult to find such observations given the large background of living-specimen observations.
Solution: Figure out ways to find un-annotated deceased-specimen observations, to label them. This is a wiki for that, currently a stub.

Potential approaches:

  • Some existing projects may focus on dead specimens
  • Some existing observation fields may flag dead specimens in different ways
  • Some users’ tags may flag dead specimens
  • Some users’ description fields may describe specimens as dead

Projects with unannotated dead animals:

Unannotated vertebrate observations tagged “dead”:

Unannotated vertebrate observations with “dead” in the description:


There are lots of projects that have dead animal observations, is that the kind of thing you’re looking for? In theory, you can filter out obs from those projects, but filtering them all out via a single annotation filter is definitely easier.

Yep! Above, relevant projects could be added as listings or something like that. It’s truly a stub so structure of potential listings is TBD right now.


Could the above projects could be rolled into an umbrella project? Annotations could then be done on that project from identify without having to go to each project.

1 Like

This feature request, Add tags, observation fields, threatened (status), description choice(s) to populate Collection Projects, would allow your last two queries to be used and then from there that project could be added to the greater umbrella project.

Is this the sort of thing you could crawl & mark through the API? I keep meaning to poke around the iNat API but haven’t yet.

1 Like

The issue is that not everything in those links is actually dead, e.g. https://www.inaturalist.org/observations/71775383. I think the projects are pretty reliable (and I previously suggested that they map to the Dead Annotation), but the tags and descriptions especially are likely to have some alive observations.

1 Like

I rolled all the project links into one, which isn’t exactly the same as an umbrella project, but does mean only one click.


If there are more observation fields that cleanly map to Alive/Dead, I think tiwane can add them to the existing list.

Annotating as dead, is part of my daily ID routine (for Unknown or Needs ID in Cape Town)


I’ll add it to my daily routine as well.


I annotate for dead, but alive is something I think is just assumed if the dead annotation is not set. It just feels super wrong to be marking annotations of “alive”, unless it is something that is playing dead but very much alive…


Same here, I might only mark alive if there’s something noticeable about that state in context (like, prey seen escaping).


I feel that if I’m annotating an observation anyway and the subject is alive, I should annotate it as alive. I don’t go out of my way to do it, though. I do usually make a point of annotating as dead the observations of dead mammals and birds (and sometimes other vertebrates) that I see. At least that’s my theory. In practice? Oh, well.


Some people might not want to see dead stuff so they use the filter alive so there may be some point in adding alive.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.