Life stage annotations for “Recent Evidence” observations (e.g. leafmines, galls, juvenile feathers)

Well i would say a a juvenil feather is evidence for a juvenil bird, an eggshell is evidence for an egg and an hatched pupae is evidence for a pupae. And the date of observation is the date i found this evidence. A leaf mine is certainly evidence for a mining larvae within the leave. Neither pupae nor adults produce leaf mines … so how can a leaf mine be evidence for pupae or adults?.

If i find evidence for the existence of a certain animal at a certain life stage then i record this life stage and animal at the date i observe it. What the animal that left the evidence does at the time of my observation is highly speculative … and largely unimportant. Data collection for science is about reporting facts, not imaginary stories.

I dont think that there is any way that data from museum specimens or community science websites can be used in scientific works, without the researcher looking at the specimen or the photo evidence given on the website. At least unless the introduced error is acceptable. Just my 10 cents.

5 Likes

@tiwane are observations marked “recent evidence” included in the annotation graphs?

1 Like

I am sorry to lead the discussion further away from the specific title and more into the general aspects of ‘life stage’ annotation, but I feel the need to answer here.

On the taxa pages, iNat has built in these wonderful phenology diagrams, with ‘life stages’ in insects indicating when, during the year, the different stages occur. These graphs are fed by the annotations given on specific observations. As the intrinsic meaning of these graphs refers to living individuals found at that time, annotating traces of living organisms (such as leaf mines without the insect still in there) would impact those graphs in an unintended way.

So whenever I find on, say, an uncleaned window sill a dead moth in winter (without evidence of it being alive shortly before), or a mantis ootheca in autumn or empty exuviae from cicadas or dragonflies, I refrain from annotating them. Yes, they indicate a certain life form was there, but not at the time of observation.

Would be great to annotate them still, with a toggle to opt-out of contributing to the phenology, but this is a rather niche request, I’d say :slightly_smiling_face:

3 Likes

This is why I queried @tiwane about whether those phenology graphs filter out or include “recent evidence” observations. If they filter them out, then the graphs are not impacted, as most situations would have that DQA set.

1 Like

If they are impacted, then we would need to look at setting “date inaccurate” to effectively detach the connection between lifestage and date… Not ideal at all…

Most leaf miners pupate within the mine, and for many it is not detectable if it is larva or pupa. Also consider your cicada exuviae example, if the leafminer exuviae is present it would, according to your comments above, be considered adult.

Lots of interesting discussion here. I think discussions of juvenile feathers are relevant too. I might update the subject if I can think of something suitable.

That was why I haven’t tagged them*.

The disadvantage of not tagging them is that there is no easy way if finding photos of the juvenile live stage. That could be useful for many reasons including the possibility of training the image recognition to distinguish life stage in future. If someone wanted to fine leaf mines most photos are likely to be of the mines (so filtering out adult would work) but finding photos of a juvenile bird feather would be tricky.

I can see both sides of the argument for tagging things like empty exuviae from cicadas or dragonflies. It is evidence that an adult was there but it is also evidence that the species is breeding there.

I think not tagging is probably best in all these cases unless there is a way of excluding them from the graphs (or there will be in the near future and we know how it will work now).

* With the exception of my own observation where I could see larvae moving plus I think a couple where the observer had indicated that larvae were present. I also tagged a pupa where the butterfly had clearly emerged (as a pupa).

1 Like

not helpful…

An observation is for an individual at a time and place. Observations showing evidence that an individual has been there in the past are NOT observations of an individual at a place and time. No imaginary story involved here. iNat allow them by having a DQA setting that segregates that data from all the observations that DO represent an individual in a place at a time. The question here is whether that DQA setting prevents your observation of a cicade exuviae in winter showing up as a data point of an adult but that died (as did every other adult of that species) in the autumn. If it doesn’t, then you are not recording facts, you are recording false data… not what good data collection for science is about at all!

1 Like

How about:
Annotations in the context of “Recent Evidence” observations (eg leafminers)

1 Like

Life stage annotations aren’t too hard to do. Butterflies & moths are very easy if you ignore the small number of leaf mines and occasional. Take a look at the tutorial below. If you want to make it even easier then remove “needs id” from the search as you’re less likely to get dubious quality photos that way. That’s what I’ve been doing.

Using Identify to Annotate Observations

Hi kiwifergus

It seems we disagree largely here.

Which translates in case of traces, feathers, exuviae, eggshells and similar to:

If i find a cicada exuvia in winter, then i annotate it “nymph” as a exuvia is the skin of a nymph, and i give it the date when i found it. Well and if some graph then shows that nymphs can be found long after adults have emerged … then we can at least learn how long exuviae last, which is an interesting question on its own.

4 Likes

For such finds you should add the “dead” field and still add an annotation, it will show on the diagram that you can find an adult at this time of the year, yes, dead, but still it can be there. If work with annotations is done well for the species it’s not affected by one dead individual, if there will be many dead found by many people that means they should affect the diagram, cause the latter helps you showing when the animal is present, alive or dead.

4 Likes

Considering that observations of evidence of organisms are valid Research Grade observations, and not discouraged in any way, what is it that makes the graphs implicitly referring strictly to living organisms? All they do is describe what the set of observations indicate, and there’s nothing stopping anyone from adding evidence of an organism as an observation. If a researcher wants to use the data making up the graphs, he or she should have a look through the observations to make sure they’re all helpful for the purposes of the research (as I assume should be done for any data set for research).

What if, hypothetically, you had a species where males and females are active and visible at different times of the year, with males appearing in one month and females appearing the next month. However, a male dies and you find it on the ground a month later. That woud technically “mess up” the graphs, but I don’t think it’s reasonable to prevent people from marking the sex of the organism. However, it would be easy to find and filter out.

Or if a Type of Evidence annotation is added, then a researcher could filter out any observations that don’t have the value “Photo of a living organism”.

3 Likes

As I said, we can go through and mark all those as “date inaccurate” and then they will become casual. I’d much rather wait for @tiwane to let us know if it is an actual problem.

1 Like

I don’t get the point of making valid observations casual because they had annotations (and those were right).
Both species with strict limits of appearing of one sex or another and species without such should be treated the same in this situation, both dead bodies will be the evidence of organism, both died who knows when, so no reason to make one casual and the other one not, cause the second has some alive specimens at the moment. You (observer) don’t know when this individual died, and who said it died within described limits? I focus on invertebrates now and see how lots of data about timings isn’t that accurate as it seems.

1 Like

If you put the annotations, then the observation is for the live animal at the time it lost the feather or made the footprint or left the exuviae as it eclosed. The date would be inaccurate if you put it as the date you saw the evidence. If you DON’T put the annotation, then it is effectively just an observation of a feather, so won’t affect phenology graphs.

I’ll repeat myself… again! I think we should wait for @tiwane to let us know if this is an actual problem!

I wouldn’t consider myself to be the final word on this at all, and to be clear I’m not a researcher, but personally I don’t see a problem with annotating evidence of a certain sex of life stage. If a leaf mine is presented as evidence for the observation, then I think it’s fine to annotate it as “larva” for life stage. As @mreith said, any researcher using the data should look into it.

That being said, I love me some galls but I’m don’t annotate them as either larva or pupa since I’m not sure if the organism is in the larval or pupal stage inside the gall. But if someone adds an annotation of either stage to an observation of mine I wouldn’t really mind.

3 Likes

@tiwane
more specifically, are observations marked as “recent evidence” included in the phenology graphs?

For example, is an observation of a juvenile feather found in December with a December observation date but which came off the bird when it WAS a juvenile back in June, included in the phenology graph, making a data point for a juvenile in December when there are never juvenile birds present in December.

Substitute any one of the other scenarios raised above if it helps, eg for cicada exuviae marked as adult in winter making adult data points when there are no adults around…

All observations are by default assumed to be recent evidence of an organism – that’s how the Data Quality Assessment works – so yes, they’re all included in the phenology graphs.

Another case would be observing, in the spring, the fruits of a flower from the previous season. There’s no problem adding an annotation for fruiting because you are observing fruits. It’s understandable that the phenology and life stage graphs will be a bit messy from datapoints like these.

6 Likes

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