When we add new observations, if we know the same individuals appear in mutiple days, should we merge in the same observation? In the same day, if many individuals are very simliar, but not sure if they are the same species, should we merge to one observation or separate to many observations?
1 Like
For me I would attribute each individual and each appearance of an individual its own observation. Ultimately the content of your observations should reflect the data you’re interested in documenting and sharing.
2 Likes
If it’s a different day, you should make a new observation.
If you’re sure they’re separate individuals, it’s best practice to make a separate observation. That said, if you see an entire colony or flock, most people keep it in one observation rather than make dozens/hundreds of separate observations :)
From the Help docs:
Observations are the basic units of iNaturalist. An observation records an encounter with an individual organism, or recent evidence of an organism, at a particular time and location. This includes encounters with signs of organisms like tracks, nests, scat, or things that just died. iNaturalist provides a place to add this information along with associated text, photos, and tags.
You should make separate observations for each separate organism you encounter. But if you take multiple pictures of the same organism during the same encounter, please combine them into a single observation. If you revisit that organism later, such as returning to a plant on a later date when it’s in bloom, you should make a separate observation because it was observed on a different date.
9 Likes
In the “Observation Fields” box you can add the field “Related observation” and enter the url of the previous observation of the same individual.
4 Likes
had the same doubt thanks :)
I have been wondering about this too and now it appears I have been doing it wrong all the time. Sorry for that. Problem is, I like to observe slime molds, and also take pictures of the various stages of maturity. Sometimes they even change in a matter of hours. If I would make a different observation of every encounter with the same sporangi or group of sporangi, I feel I would inflate the national observation database of my country, because it is connected to iNat. Another problem is, that the slime molds are usually only identifiable at certain stages, so I would be posting a lot of basically unidentifiable observations. I do not see the joy in this, and it would probaly lead to me not posting any of these immature stages. And what, btw would be the acceptable time interval between observations to be allowed to be combined. 1 minute? 1hour? 12 hrs?
3 Likes
I would take photos whenever it looked different, but combine all of one day’s photos into one observation. You can write in the notes that the 2nd photo is 2 hours later, 3rd is 30 minutes after that, and so on.
4 Likes
Depends on how standardized and uniform your uploads are. I’m not familiar with slime molds but if there are observable phenological changes in such a short amount of time I feel multiple instances of observation in one upload could be advantageous.. considering there are detailed notes correlating with individual photos and their features. Keeping each instance separate is still an option if you’re tracking the same individual sporangi or grouping.. keeping obs. connected through links in the notes. Once an ID can be made on an individual or grouging at a certain stage go back and revise the related prior obs. As long as your uploads are accurate I don’t think inflating the database is a concern either, especially if you’re recording phenology in detail. Posting those immature stages is just as valuable data as any other so don’t be discouraged if you have detailed obs. that don’t reach RG any time soon.
2 Likes
First example: In a time during fall season, I continue see a Red-shouldered Hawk at the same spot, maybe should be the same individual, but not very sure, so I upload many observations. Someone suggest I merge into one observation. Recently I deleted some of them because I upload too many observation of the same species/individual.
Second example: In the same spot, I saw some large flocks of sandpiper. However, I’m not very sure how to distinguish sandpiper. I worry mistake identify, so I separate to many different observations. But most of them should be the same species.
I agree #3 answering my original question good enough. but if other people have more questions related to this topic, they can still discuss. Once no more new discussion, I will recommend this answer and set up as “resolve”.
1 Like
You can use the field “Similar Observation Set” to link repeated observations of the same organisms, and add a comment that they are of the same individual organism through time. Then anyone can see the sequence of “observations with this field and value”. Examples:
3 Likes
Thank you very much, this is very helpfull!!
2 Likes
I have seen nest site observations being updated with new photos/notes as time progresses; the same observation then includes all the information over a span of weeks/months. It seems to me that this is the most valuable way of documenting this organism(s), as the important information is the progression of development (nest was built dd/mm/yyyy, eggs laid dd/mm/yyyy, chicks hatched dd/mm/yyyy, fledged dd/mm/yyyy, etc).
When it comes to birds out and about though it seems better to record as different observations, as I feel like presence on a specific time/date is what is valuable. Then linking to the past observations to create a contiguous timeline of that specific individual.
For me, I mostly upload plants, so I have certainly made an observation and then updated it months or even years later with new photographs/notes to indicate some specific thing (for example, that it produced male cones, whereas the initial observation it couldn’t be determined whether it was male or female). It seems silly to me to have a cluster of observations all of one single organism that doesn’t move around at all, and much more valuable to have a single observation that includes all the details about it.
3 Likes
Why would it be silly? The problem with putting all the photos in the same observation is that various annotations will be impossible to add accurately because they won’t apply to all photos in the observation, or if added, will be wrong (e.g. life stages, phenology) because those will vary through time but are being condensed to a single date. Better to use the “Similar Observation Set” field so that all important information is easily accessible. You can also summarise this in the Notes of the first observation of the set.
1 Like
For something like slime molds, I personally would not see a problem with putting all observations from the same day in the same observation according to the reasoning that this represents a single encounter (one extended observation session).
I think the usual rule of thumb cited for the question of how often one should observe a particular individual is something like one observation per day, unless something significant changes, but the guidelines weren’t really designed for microorganisms.
There are no annotations for slime molds, so concerns about phenology or putting different life stages of the same individual in different observations don’t apply the same way. I’ve encountered slime mold colonies that included sporangi at different developmental stages, so it doesn’t seem to me that you would be overly distorting the data by putting photos taken over the course of a couple of hours in the same observation.
However, there may also be good reasons for putting each distinct stage in a new observation – for example, to clearly document the progression over time. As noted above, it is possible to link observations using observation fields or by adding links in the notes, and IDers are often willing to confirm linked observations that would not otherwise be identifiable (e.g. immature stages) based on the subsequent ones if it is clear that it is the same organism.
2 Likes
I was always wondering what the rationale for the last sentence is. We have many young plants that are doomed to stay unidentified forever because nobody (including the AI) knows what young specimens of a known species look like. If the rules were changed to allow adding later images of the blooming plant (allowing it to be ID’d), people and CV could learn what the young plant looks like and use that knowledge later. I don’t even know how the project “Rosettes of Dicots” is to gather data (that project is intended to collect observations of young plants ID’d to research grade so CV can learn what they look like, I even added my baby Digitalis there for that purpose).
If you take a photo on August 28th and you attach it to an observation dated June 1st, that makes it seem like the species looks like that on June 1st, which is wrong.
It’s pretty quick to link to new observations of the same organism on a different date in the notes, comments, or observation fields.
4 Likes
Please don’t do this (adding photos from different days to the same observation, especially after initial upload). As other users have noted, it breaks iNat’s definition of what an observation is “an encounter with an individual organism, or recent evidence of an organism, at a particular time and location.” It can cause all kinds of issues with phenology, annotations, etc. It can also lead to very confusing situations when evidence in an observation changes over time and so different identifiers are acting with different sets of information - I’ve seen many strange situations arise with IDs, DQAs, annotations and (more rarely) observation fields and project inclusion because of actions like this.
As @deboas noted above, it’s quite easy to keep the same information in multiple observations in ways that will avoid issues and generally be more useful to iNat users.
2 Likes
Thank you so much for understanding:)
At the risk of going off topic here, I would like to ask this. When trying to learn how to use the above mentionened “observaton fields”, I noticed that any possibillity to filter for observation fields does not seem to appear in my query menu. Am I doing something wrong here? Or are they available only to certain users?
There are far too many search options available to be able to fit them all into an easy-to-use menu. Many observation qualities are only searchable by manually altering the URL. Here is a link to the forum post explaining how to do that. For observation fields there is one other option: go to an observation that has the field entered and click on the name of the field. That brings up a menu where you can search for “Observations with this field and value”, “Observations with this field”, or “Observations without this field”.
2 Likes