Let's Talk Annotations

Yes, should happen soon. We’re not going to be able to include all of them, but I’ve made a list of some of the more oft-used ones for a start.

I can bring that up with our team.

Yeah, I think that’s something in the offing.


OK, the following Observation Fields should map to the Alive or Dead annotation:


Certain values won’t affect the annotation, like Moribund for Dead or alive and No for Roadkill. The site is also auto annotating observations that currently have these fields, it might take a little while to go through all of them.

I’m sure we missed a few, so if there are any well-used fields that could work here, please let me know, especially ones that are not in English.


Any possibility to auto-annotate observations in specific traditional projects, like https://www.inaturalist.org/projects/dead-herps?


This one is Alive or Dead but in spanish: https://www.inaturalist.org/observation_fields/7381


12/17/2019 I was so delighted today when I noted the annotation revision. Thank you, thank you for adding the “flower budding” annotation. I hope the change will add to seasonal plant life cycle data. Much appreciation.


Topic title: Life stage teneral for true bugs annotation

Is there a reason it is unavailable, and if not, can it please be added.

I’m not sure I understand your request…

Pretty sure it’s supposed to be “teneral” :)

Is this about instars? It would be nice but I think you need professionals for proper ids

It was and corrected, silly autocorrect on my phone and no coffee yet…


As I remember after-moult state was requested aready, but not for bugs specifically, I guess all arthropods need it, but have no idea if the term applies to non-insects.

applies for arachnids… many of us have just been assigning a field “teneral=yes” if we encounter them.

in fact, many of the annotations started out as fields…


@bouteloua @janetwright @sedgequeen I still have to disagree here. Though there are gradations and often few or no clear dividing lines for various levels of dormancy, identifying things at various stages, including dormant/dead instead of living/active, are usually very different processes. Also, the identification process of dead plants is similar to nearly identical to identification process for dormant plants (unless the plant died sterile, in which case there is little hope for any kind of ID). If you like, perhaps even completely dormant (no active above ground growth)/dead would provide the clearest distinction possible. There are multiple ways that we can try to group this to make it work, but I think there are ways to make this work. Just so we’re clear, my argument is for utility, not absolute data. Though this category might not be useful from a data perspective, it would be very useful from a data cleaning (i.e., identification) perspective even if improperly applied. For examples of this kind of information being utilized, I have project specifically devoted to providing references for ID of plants during the most challenging times of the year. I specifically use three observation fields to organize this information: Winter/Young Plant Type, Seedling?, and Evergreen. The first has the utilizes “only dead material visible”. Are the distinctions exact? Not at all. Are they useful categorizations? Absolutely. Lastly, I would argue that the distinction between dormant/dead and active/alive are actually no more arbitrary than the juvenile/adult values for life stages in animals.

I find these helpful in the case of dioecious plants. More selective implementation would be very nice, but I’d hate to get rid of it altogether.


Generally, yes, there are a lot of useful observation fields and other data that can be added to observations that help particular use cases or projects.

Since by far most plants are monoecious, it’s a small enough number of observations where sex is useful for plant observations that I don’t think it merits an annotation to be displayed at all times for all of Plantae.

This is possibly region or climate specific, but in the upper Midwest, I often find plants to be not actually completely dead or dormant aboveground, even though they seemed to be at first glance. And I’m looking at deadish and dormantish looking plants for a pretty big portion of the year. Determining whether an organism is “completely dormant or dead” is a task not possible based on most iNat photos of plants, and I definitely know that most of the photos that I’m uploading to iNat that “look dormant” do not contain detailed enough level of information to make that assessment. For example, a brown and crispy fruiting stem where all the cauline leaves have long since fell off, but peeking out from under the snow are the plant’s green basal leaves. Often I don’t take a photo of the green leaves but just the fruiting stem.

There are grey areas for almost all of the annotations, but the grey area is pretty big here. If you can confirm based on your photos and in-person observations that an organism is completely dead or dormant, that’s great and definitely continue your work/project because these tend to be some of the most useful photos on iNat to me.

But it sounds like you’re referring to and most interested in the compartmentalization of tissue and the appearance of different aboveground plant parts at different times of the year, which is, in most cases a different question to ask and answer than whether an entire plant is dead or dormant. Maybe it’s less useful for areas that have less severe growing seasons than mine, but I usually start by sorting by month if I want to view crispy or vegetative plants. Maybe the “vegetative, not flowering, or not possible to tell” option, which should be added to the flowering phenology one, would help with this use case.


Is there a list of plant taxa for which this would be useful? Happy either add only taxa for which it would be useful, or exclude taxa for which it’s not, whichever one is smaller. Are entire families dioecious, for example?


There is a list here, though it is definitely incomplete (e.g., none of the dioecious Crotons are included). Here is a list of all the species on iNaturalist that have used the sex annotation. This too is almost certainly incomplete despite the over 1,600 species using it. Euphorbiaceae is one of those families that has many dioecious members (mostly monoecious, but a fair number of dioecious species too). Ephedra and Baccharis are two very common dioecious genera that come to mind. There are also examples of functional dioecy (e.g., Echinocereus coccineus) where a sex annotation would be useful to have available.

@bouteloua over 12,000 plant observations currently use the sex annotation. I guess this is relatively small compared to the vast numbers on iNaturalist, but still sizable.


Oof, yeah, 0.1% of plant obs. (As a comparison, 7.4% of flowering plant obs have a flowering phenology annotation.)

I wonder how many were incorrectly/accidentally annotated too, e.g. this (hastily filtered) subset with ~100 obs at the time of posting https://www.inaturalist.org/observations?place_id=any&subview=grid&taxon_id=47604&term_id=9&view=species&without_taxon_id=51901,55850,52111,52834


Are there any families (or higher) that can be totally excluded?


I appreciate your comments, and I do know what you’re talking about (plants in West Texas behave similarly unless you get so far south that the Annuals start to overwinter).
I still think it useful, but I guess it’s really not as important to me as a lot of other things like adding a sterile option (maybe even a leaf bud option?). So, I guess it would be better for me to drop the issue and move onto something else.

Ultimately, the ideal for me would be to try to implement life stage annotations for plants like there are for animals, but I realize there are difficulties. I occasionally produce sets of observations for plants that show seedlings to adulthood and frequently photograph identifiable rosettes (as do a lot of naturalists in my area, though many not so identifiable, much to my annoyance). I guess in my mind, the difficulties are perhaps no more complicated than the difficulties for Animals (think of insects and salamanders in comparison with mammals; Plantae and Animalia are both kingdoms after all), though perhaps more ecosystem controlled differences to add to the phylogenetically based differences.


I don’t know any fully monoecious families offhand, I’m sure someone could find a few, but like Nathan says there are mostly monoecious families with some dioecious exceptions. :\ Even still, there is the question as to whether the annotation is intended to refer to the whole plant, or just part of it, in which case you could have an observation of a monoecious species but have photos depict only male or only female flowers. I guess photo-level annotations would help solve that.