Let's Talk Annotations

Sprouting is a young plant. No longer a seed.

2 Likes

Thought it would be something like that. I was just having a moment of “there are seeds visible -”

But yes, sprouts aren’t fruiting, this is true. Some of them bud before they even have true leaves though, which startled me. Seedlings with buds!

2 Likes

Everyone knows that conifers bear cones. Surely taxonomy will let us use an annotation of some sort. One day?

Apologies, I have been using Seek for a few years, but am new here and unfamiliar with… everything. I have two features that I would like to see that seem closely related to this topic.

  1. Diseased, infected, etc. with additional description of severity i.e. life threatening.

Example I:

(Image1 removed due to limitations of new user.)
Image1. Healthy Monarch larva.


Image2. Monarch larva infected with parasitic (Tachinid Flies?) larva. Life threatening (destroy).

Example 2:


Image3. Healthy monarch egg.


Image4. Mottled monarch egg infected with parasitic (trichogramma?) wasp. Life threatening/dead (destroy).

  1. Color coded toxicity severity in plants and risk of contracting a pathogen from in insects. This would be greatly beneficial to carry forward to identification apps for those in the field (hikers, foragers, farmers, gardeners, children etc.) with limited time to research each observation encountered.

observation fields will do in the meantime; I’ve been applying “diseased/infected: yes” to fungus-smutted inflorescences

separately, I have a question: do teeth (NOT in a skull) count as bones? for the purposes of the annotation anyway, they’re not literally bones

I would certainly annotate teeth as bones.

2 Likes

Alright good, I shall leave my annotation there until further notice then. I was mildly concerned as to whether it might be an owl pellet situation, which are apparently not scat

3 Likes

Need to include damage as evidence of presence
Including “Damage” as a new annotation under iNaturalist’s “Evidence of Presence” would greatly enhance the documentation of species that are cryptic, nocturnal, or rarely observed directly. Damage, such as leaf chewing damage, chewed bones, boreholes, claw marks, or bark scoring, can serve as reliable, species-specific evidence, especially for caterpillars, hidden insects like borers, or elusive mammals such as rare big cats.

Unlike existing categories, damage is neither a “Construction” (which implies purposeful building) nor a “Track” (impressions left by movement), but an incidental result of behaviour such as feeding or territory marking. It also differs from “Leafmine,” which refers to a specific feeding pattern within leaf layers by certain insect groups. A broader “Damage” annotation would fill this gap, supporting more accurate identification, early pest detection, and improved biodiversity monitoring across a wider range of species.

This has been discussed in this thread but I don’t feel a satisfactory answer was reached.

Examples of reports where “Damage” as an annotation would be appropriate:
Eucalyptus Longhorn Beetle report: https://www.inaturalist.org/observations/260873608
Eastern Black Bear: https://www.inaturalist.org/observations/285632888
Pileated Woodpecker: https://www.inaturalist.org/observations/283928260
Red Rover Case Bearer Moth: https://www.inaturalist.org/observations/269865178
American Beaver: https://www.inaturalist.org/observations/280913908
Yellow Bellied Glider: https://www.inaturalist.org/observations/280468361

3 Likes

Damage sounds anthropocentric.

Evidence of feeding - for most of your examples?
Marking territory for leopard claw marks on trees.

The bear example annotations are odd. Must have been alive when clawing the tree, and claw marks is not a construction.

‘Feeding damage’ is the term I commonly use in applied entomology for insect chewing marks, and I thought that ‘Damage’ would be broad enough to cover this as well as claw marks, teeth marks, digging / excavation and bore holes etc. However, I agree, if the option is available: “Scratches” and “Chewing marks” could be separate annotations instead of “Damage” that would be valuable? It can be difficult to know intention, claw marks may not always be for territory, and could be for fun or for finding food. Also scratches can be with hooves, eg. wild pig digging marks in the ground.

I am not sure about the bear one, I just quickly found the most relevant examples, maybe because the tree is dead?. Yeah, not having an appropriate category for claw marks is causing confusion I think haha.

1 Like

I disagree that

iNat doesn’t include “purposeful building” in its requirement for annotating as “Construction” and some of these examples (specifically boring by beetles) are specifically mentioned in the documentation as things that should be annotated as “construction”, see:
https://www.inaturalist.org/blog/91456-new-evidence-of-presence-values-added

““Construction” is meant for observations of anything an animal has constructed or excavated out of other materials, such as nests, burrows, hives, spider webs, beetle galleries, egg sacs, and the like. Rather than make separate values for all of those things and assign them to specific taxa, we decided to make a more general annotation for ease of translation and maintenance. It’s not meant for coral reefs, galls, or mollusc shells, which are either part of the animal’s body or are instigated but not built by the animal. Note that an observation can have multiple Evidence of Presence annotation values, so a bird in a nest could be annotated as both “Organism” and “Construction”.”

That said, some type of “Damage” might be a useful annotation, though I think something even broader might be more useful. “Damage” is a bit of a loaded term - it implies loss, harm, degradation, etc. I wouldn’t describe chewed bones as damage (the living thing that once had the bone is no longer there to experience damage). Marks might also be on non-living material (rocks, salt) or dead trees, etc. and I don’t think “damage” is the most appropriate there. “Damage” could also cause some confusion with existing annotations - for instance, a woodpecker nest burrowed out of a living tree is definitely “damage” but also a construction. Same with a leafmine. Translating “damage” might also be challenging as this term varies a lot in what it implies in different languages (at least those I’m familiar with). It feels like what this is asking for is some type of catchall category for a physical sign that isn’t encompassed by other annotations.

2 Likes

Good points. Do you have any alternative suggestions? ‘Affected’ maybe?

Any necessary elaboration or context can then go in the notes.

Applied entomology explains why you see damage. iNat focuses more on the creature, and the ‘damage’ as evidence of said creature.

Cape leopard needs claw marks / territorial marking

1 Like

I view anything the animal made, including claw marks or feeding damage, as a “construction.” I usually don’t annotate it, though, unless the organism and its tracks are absent

1 Like

My feature request was rejected so I’m copying the text here to request a “not enough evidence to assess/determine” for various fields to avoid cases (including ones I’ve seen myself) of people annotating, for example “no flowers or fruits” on a close up image of a single leaf.

Details below:

Need a new field for Annotations “Not enough evidence to be determined”

Platform(s), such as mobile, website, API, other: Website and mobile

URLs (aka web addresses) of any pages, if relevant: N/A

Description of need: When making annotations for plants and animals, I often find that either the parts that need to be photographed are missing, or there isn’t enough effort (number of photos and survey of canopy in the case of trees) to assess, for example, whether there are flowers or fruits on a tree, or sex of an organism.

I think we need an option for “not enough evidence to be assessed/determined” as there are often inaccurate annotations (especially for plants), e.g. “no flowers or fruits” but there’s only a single blurry photo of a few leaves.

Feature request details: It’s pretty straightforward, just an option at the bottom of each existing field that allows for annotation as “not enough evidence to be assessed/determined” for each field.

2 Likes

The annotation applies the observation, not to the organism. Therefore, a close up image of a single leaf indeed contains “no flowers or fruits.”

4 Likes

Thank you for the clarification, it’s helpful to understand that iNaturalist’s definition of “Construction” can include beetle galleries and similar features. However, I believe there’s still a meaningful gap for a general category to capture incidental signs of presence—particularly those that don’t fit neatly into “Organism”, “Track”, “Leaf mine”, or “Construction”.

Neutral term suggestions such as “Sign” or “Mark” to describe physical evidence of an organism’s presence would be great.

I feel that this wouldn’t seek to replace or compete with existing annotations like “Construction”, “Leaf mine”, “Gall”, or “Track”.

It’s more that the term “Damage” was what I am familiar with, however I am very happy for other suggestions that fill the same gap. @cthawley Proposed “Mark”, and I also suggested “Sign” which would be great and could work well.

1 Like

From what I can tell, this isn’t really true when it comes to research applications. It’s easy to imagine how most observations don’t necessarily attempt to document or sample a representative part of the canopy, so annotating all of those on the basis of what’s visible will have a negative impact on the phenology charts at least from my understanding of it.

As an identifier, we can only use the info available in the pictures, or notes from the observer.

there was fruit visible, but too high to photograph

3 Likes