I have questions for researchers and data users on iNaturalist about descriptions on observations. Is there a lack in descriptive descriptions and is there a need for them for specific taxon observations or observations in general? Is there specific criteria wanted or needed for some species or taxon? Are annotations and/or observation fields better than the descriptions in some cases? For an example, I just posted tons of marine low-tide observations. Would it be beneficial to include the level of low tide (-3.7 ft) in the descriptions or observation fields? Felt like I am lacking in descriptive posts.
I search using observation fields way more often and find observation fields more helpful and easy to use/search for. With the taxon I look at (Erotylidae) I need host fungi information which can be hard to find if it’s in a description vs just using the “Feeding on” or “Associated with” observation fields.
2 Likes
Descriptions are good for things that aren’t easily standardized (such as descriptions of animal behavior), but for most other things, observations fields are far better. Observation fields are much more searchable/filterable, making them easier for researchers to find/pull the data they are interested in.
If you like, you could always put it in both places, but an observation field is likely the most beneficial for this type of data.
2 Likes
I think a researcher will likely only use descriptive fields (notes) for one-off observations (new behaviors, interactions, etc.) just because they aren’t standardized. It would be very difficult to process them from across iNat for a large scale analysis. For an analysis, observation fields or annotations are standardized and can be used “out of the box.” If you want to maximize the usefulness of descriptive notes to potential researchers, I would be sure to standardize formatting, etc. so that they are easy machine readable. If there is a specific project or researcher though, they could ask people to use descriptions for some data in a standardized way which could end up being useful (like asking users to always enter tide data as “Low Tide: -3.7 ft”)
2 Likes
Okay. The general answer is use of primary obs fields and annotations. My next question is what fields that are needed to be filled out more or lacking per say? Thank you for your input on best methods of observation postings so far.
Filling in the phenology graph for sp that interest you?
Add Annotations for …
1 Like
sure, anything will great! is there a specific methods that I can try?
You choose something that interests you - then click the cogwheel icon on the phenology chart.
Say plants - Add Annotations for Flowers and Fruits.
Takes you to Identify where there are keyboard shortcuts (see the keyboard icon bottom left)
Plant and fLower = p + l
Plant and fRuit = p + r
If in doubt, skip that obs.
Or caterpillars?
Life and Larva = l + l
https://forum.inaturalist.org/t/using-identify-to-annotate-observations/1417
1 Like
Data in standard observation fields are easier to access. I’ve recently been annotating Australian mistletoe observations and adding, for example, “Host Plant ID” even where the same information was captured under “Associated Species” (but as text input, not taxon input). This would allow any future interested parties to access the data in a single search. I also transcribe the host plant from the observation notes if present.
Separately, I use the “ripe fruits” observation field because “fruits or seeds present” and the “fruiting” observation field are ambiguous and used differently by different observers. I can’t otherwise capture the long period in which mistletoes retain unripe fruit or analyse its occurrence by region or species.
I have also created “Many plants on one host” and “Endocortical growth” observation fields to be able to find relevant observations later to analyse extrinsic and intrinsic factors in mistletoe growth. I also use the “Gall present” field to track interesting cases of flower bud deformation; this might later connect to identification of the insect responsible.
2 Likes
i find the using the taxon input easiest for adding observation fields on observations and wondering what is the difference between the taxon input and the text input in data analysis as in the pro and cons of each?
I can’t think of any points in favour of text input. If a species classification changes, the taxon input will presumably update, but the text definitely won’t. Misspellings and wrong common names can also populate the text input.
I don’t imagine there is any way to convert an observation field from text input to taxon input. The only way would probably be to create a new field and manually enter data for each instance of interest.
1 Like