Often I see a photo of an ant nest. There are larvae, eggs, adult ants, all together. This is great since it’s the kind of evidence that helps with other IDs.
I’ve taken to tagging the most rare one when I get to filling out those fields on the posts I ID. But with “Evidence of Presence” I can pick multiple things, both “construction” and “organism”
Why not allow for multiple entries on these other fields? That would make it easier to find photos of all stages of life. This could apply to birds and larger animals too. What do you do if you have a photo of a bird standing on their nest by their eggs?
The option to select multiple life stages for animals is intentionally unavailable because of the definition of an observation on iNaturalist:
An observation records an encounter with an individual organism, or recent evidence of an organism, at a particular time and location
Some observers like to duplicate their observation and use the same photo to annotate each life stage separately. It’s definitely not required to do so though.
I would also like to have a built in solution for this. But as there are none currently and doesn’t seem to be likely in the future, I’ve made some projects to at least help sort these kinds of observations:
Agreed. Yet strangely, if you look at this other thread, someone on the staff almost seems to be contradicting the one organism = one observation rule:
My solution for multiple sexes is to add the “sex” observation field, which allows the value “mixed”.
There’s probably a similar solution to multiple life stages. I vaguely recall that there are life stage fields for insects where you can add separate fields for each stage that’s present. So you could have something like larva=yes, adult=yes, etc. I’ve never used that approach myself, so I can’t remember the exact fields. I think it was something like “Insect Life Stage Larva”, “Insect Life Stage Adult”, etc.
Of course, the problem with observation fields is that they are only useful to those who know to look for them.
I wouldn’t suggest duplicating the observation for every ant, but making one duplicate per life stage/sex/caste seems reasonable. It would technically be acceptable to make 200 duplicates of a photo with 200 ants in it, but I don’t think anyone would ever do such a thing.
I can’t help but feel as though the “one observation = one organism” justification to keep this from being added seems a little arbitrary. The site already lets us go against that given our ability to pick multiple evidences of presence—I could mark an observation of a nest full of eggs as both “egg” and “construction”, both valid, yet it’s indisputable that those tags would represent separate organisms.
I think the idea behind it though is that the ability to mark multiple evidences of presence is that (even if only for taxa other than ants) there can be multiple evidences that are attributed to the same individual.
And yet, there are folks who go to a location and photograph every butterfly of a given species (or every monarch caterpillar in a milkweed patch). I’m not sure I’ve ever seen 200 “duplicates”, but I’ve definitely seen 50+.
It’s pretty common with ants for a species to only be able to be ID’d from a photo with one particular kind of caste member, often the queen or a major. For example, in photos Camponotus decipiens and Camponotus discolor are very easy to confuse. But, there are some more noticeable differences that are easier to see on majors. And, anyway, we tend to think of each colony as an individual. This is kind of what it means to be eusocial.
The perfect series of ant photos shows several individuals of different casts from many angles and their nest.
The current system on iNaturalist is somewhat awkward and certainly arbitrary, but I think it is valuable because it can head off problems that are fairly common, especially with new users (of whom there are many). Fortunately, there are work-arounds so we can be as detailed as we want to be.
First, if the photo includes two genders, two age classes, and a nest, and you make only one observation of it, annotate it with all the relevant terms. Why? Because the purpose of the annotations is to help people find observations they’re interested in.
Second, you can post the photo 4 times, once for adult female, once for adult male, once for juveniles, and once for the nest (construction). If I have a series of photos of this group, I prefer to use a different photo for each observation, just so I don’t confuse myself as I’m working, but you can post exactly the same photo for all these classes if you want to.
What about posting the photo multiple times for each of the dozens or hundreds of individuals? Please don’t. For the sake of us already overburdened identifiers, please don’t.
I think having an option for multiple attributes for identifiers would be more beneficial for scientific data. Most casual observers won’t care to single out individuals for this purpose.
Annotations are applied to the observation, not to individual photos. So if an observation contains multiple individuals who are not all present in all of the photos, you end up with photos that are associated with an annotation that does not apply. This seems problematic, because people may be viewing the photos in a context where it is not clear that the annotation was intended for a different individual or type of evidence than the one shown in the photo.
I see the issue of potentially mislabelled photos as also being the main reason for the definition of an observation as being for a specific individual. It is not uncommon for people to observe multiple individuals and put them in the same observation because they think they are all the same species, but sometimes they are not, or sometimes the individuals are of different sexes.
However, this criterion does seem more applicable to animals with clearly bounded bodies than to other types of organisms such as plants or fungi where it may be less clear what constitutes a single individual. I don’t think plant people generally fuss much about “one observation = one individual” too much for precisely this reason, though if there are plants at a site at very different phenological stages it might in some cases be worthwhile to make separate observations to document (for example) that some specimens have just started blooming while others are finished.
This morning I ID’d a photo. It’s a nice photo of Camponotus pennsylvanicus in a natural setting showing the queen with pupae, larvae of several sizes, and eggs. Since they are all together I have the highest confidence that these are all one species. Sometimes carpenter ant eggs are yellow, other times they are white. They vary in size. I might like to show someone a group of photos showing this diversity. So, I could tag this photo as “egg” as there are 1000s of photos of queens of this very common species. I could tag it as “larvae” or “pupae” for similar reasons.
But, I have to pick just one.
I can write a note to the observer asking them to duplicate this observation three times, and then tag each of those but that seems like clutter to me. Further, one might want to search for photos with BOTH larvae and an adult.
I deeply value photos that show BOTH the ant and their nest “constructions” – this is the only way to have confidence they are linked.
For now I just pick one thing to tag in such photos, so at least one can find them some of the time. Feels janky.
I understand that, given the diversity of living things, a database system will never be a perfect fit for everything. And making changes can be hard.
This seems like a perfect use of Observation Fields. If you want to keep track of something that’s important to your particular organism of interest but doesn’t apply to most observations, you can make a field for that. Something like “Multiple Ant Stages Visible” and have options for each combination visible in the photos (egg/larva, egg/queen, egg/larva/queen, etc.) This way you can keep exact track of what’s visible where without needing an overhaul of the annotations. There’s always been a tension between what things are general enough to make an annotation vs what things are niche enough to solve using fields.
There is a series of existing fields that would cover this. I mentioned them earlier, but just now I verified the exact format. They are:
insect life stage: egg(s)
insect life stage: larvae
etc. It appears they can take 3 values: yes, no and “not recorded”
So if you wanted to indicate that there are both eggs and larvae, you could add 2 fields:
insect life stage: egg(s) = yes
insect life stage: larvae = yes
My guess is that somebody was trying to account for observations that contain multiple life stages. I’ve been using “Insect Life Stage”, but that only allows one value for an observation. There’s also “Life Stage (Insecta)”. It allows all the same values as “Insect Life Stage” except for Teneral.
This wouldn’t allow you to record the presence of a Queen - there’s just “insect life stage: adult(s)”. But a corresponding field could be added. Perhaps something like “insect life stage: adult (Queen)”.
There’s probably some other existing observation field that already covers the Queen/worker distinction. I do a fair amount of annotation using observation fields either because something isn’t covered by the existing annotations, or because I need a way of “overriding” an incorrect annotation. I don’t think I’ve had to create an observation field so far - in every case there was already at least one existing observation field that covered whatever it was I was trying to annotate. Usually there are several to choose from.
Adding a field to an observation is a bit more tedious than using an annotation, but by using the metadata tool, you can create keyboard shortcuts for your most frequently used field/value combinations (or just use the buttons that the tool superimposes on the identify screen).
I do think it’s good to have multiple entries in some cases. Definitely I don’t want to split every ant into an entry. But also some great observations out there are things like insects hatching, coming out of a cocoon, molting, etc. Those clearly have multiple life stages of a single individual. It meet’s inat’s stated rules for an observation but we can’t fully document what’s in the observation due to only being able to have one life stage. You could duplicate it, but I think it’s odd because it’s still a single individual seen in a single encounter, and a duplication to me would be worse data because it suggests you saw two individuals
It’s the exact scenario I made the Multiple Life Stages project for. When you add an observation to it, it will prompt you with an observation field for each life stage. Add a yes value for the ones present and add to the project. This topic has been brought up enough that I don’t expect built-in functionality to ever be added here. The project simplifies adding the needed observation fields without having to memorize all the different ones.