Tips on data entry (that will help the TEA's Ontario Butterfly Atlas)

While processing iNat observations from 2020/2021 for inclusion in the Ontario Butterfly Atlas, I'm finding that folks are adding all kinds of information to the "Notes" section of their iNat observations. This is great - better that the information be included somewhere than omitted completely. However, it isn't necessarily the "best" place for some pieces of information. If observers are going to go to the trouble of adding this information, they can greatly help the Ontario Butterfly Atlas by putting the information in places where our software can find it more easily.
One of the things I see most frequently is a location name (eg. "Ganaraska Forest"). iNaturalist enters a default location name based on the Lat/Long of the observation. This default name is often vague, so this additional information can be very important. However, when you put it in the Notes, it means that someone has to manually copy it into the location field in our database, which is a lot of work, and is error prone. iNat users should note that they are free to edit/replace the default location name that iNaturalist assigns to their observations. It's pretty straightforward, and you can apply the same location name to a group of observations (at the time you are entering them). I'll create a separate post with instructions on how to do this, but our suggested format for the location format is to start with a town/city/region name, and then append something more specific like a park or trail. In other words, go from a general area to something more specific.
For example:
Toronto - Edward's Garden
Algonquin PP - Old Airfield
Khartum - Doorley Creek Rd - km 2

Sometimes the location that iNat inserts is fine. You can use your judgement as to whether or not it needs to be "tweaked". There's no need to pinpoint the location down to the meter - just something that would allow someone to visualize where on the map the observation occurred. It can also be a useful check that the lat/long coordinates are correct.

Other pieces of information that folks often put in the Notes are things like:
number of individuals observed*
life stage (adult, larva, etc.)*
additional observers*

The items marked with an asterisk are of particular interest to the Atlas project (especially the life stage!!!), but when these pieces of information are all crammed into the Notes, it means that somebody has to separate them out and copy that information from the Notes into their proper fields of the Atlas database.

If observers would like to help the Ontario Butterfly Atlas, one thing that would be a huge help would be to include these pieces of information in specific Observation Fields that are dedicated to them. When they are included in their proper Observation Fields, the software we use to process the iNat data can put this information into the proper fields automatically. This saves a huge amount of time, and is far less error prone. We get somewhere in the range of 40,000 observations from iNat every year. That's a lot of notes to parse by hand.

Please note that over the years, hundreds (possibly thousands) of Observation Fields have been defined in iNat by different people. When we download the iNat data, we have to specify which fields we want to include, and we can only download and process a limited number of them. So if you put your additional information in a field other than the one we have chosen, the Atlas won't receive it - it would be better to just leave it in the Notes.

The fields we have chosen for the Atlas project are the following:
Count (the number of individuals you are reporting - you can give a total of those seen/photographed)
Insect Life Stage (you don't have to specify that it was an adult. we will assume it's an adult unless this field is set to some other value)
Additional Observers (give a list of observers other than yourself, separated by commas)
Habitat - any notes about the habitat you feel are important

Using these Observation Fields is completely optional. If you feel it is too much trouble, please continue to include the information in the Notes. Please note that these fields have to be added to your observations - they are not there by default (some projects automatically add fields to your observations, but the Ontario Butterfly Atlas doesn't do that). I am not suggesting that anyone should go back and add these Observation fields to their existing observations. You can if you want, but this is mostly for observations you are going to add in the future.

In addition, the Atlas captures the following Observation Fields:
Nectar Plant (self explanatory)
Insect Host Plant (please use this for larvae/eggs - not for nectaring adults)

However, whatever we find in those fields gets copied into the Notes field of the Atlas, so there is no need to add these fields if you prefer to just put the information in the Notes of your iNat observation. But if you are going to add these fields, then there's no need to duplicate the information in the Notes as well (some people have been doing this).

In Summary:
Instead of putting a location name in your observation Notes, you can simply EDIT the default location name that iNat assigns to your observation.

The Ontario Butterfly Atlas captures and processes the following Observation Fields, so it is preferable to include the pertinent information in these fields rather than the Notes:
Insect Life Stage (very important, but optional for adults)
Count (optional, and only really necessary if you're reporting more than one individual)
Additional Observers (only add the field if you need to report an additional observer)
Nectar Plant
Insect Host Plant ( use for larval host plants )

Feel free to add other Observation Fields, but please be aware that the Atlas won't "see" the information you put there.

Posted on December 18, 2021 02:00 PM by rcavasin rcavasin


Hi, Rick,
If you have any influence with iNat, it would be easier if, when uploading and posting, these Annotation fields come up right at the start.

Are you happy with "Toronto - Edward's Garden" or do you want to know where in the park?

Our issue is we go out and take a "few" photos of multi subjects. All in the same park or ravine. But at the upload end i do not want to take the time to pin point the butterfly to a plant or smaller area. I thought "Toronto - Edward's Garden" or in our case Bayview Village would be ok.

Walking daily and making regular postings, I do not want to go back into each post and update them further. Let us do it up front.



Posted by gary-james over 2 years ago

Hi Gary,

I think some projects trigger the addition of their own set of fields, but I'm not sure how that is done. We could probably do it very easily, but my concern would be that some folks would find it intimidating to be faced with a long list of extra fields, and it would have the opposite effect from the one we're aiming for. I see a lot of observations with a long list of fields that were automatically added (presumably by some project), and most of the time, they are left blank. For the 2021 data, my software ran into trouble because there were a bunch of observations where the "Insect Life Stage" field had been added, but the value was "unknown". It looks as though some project added the field automatically, and then the observer never changed the value from "unknown". So I had to tweak my software to assume that all these records are "adults".

As far as place names go, names like "Toronto - Edward's Garden" are fine. That's exactly the point I was trying to make. Those 3 examples I gave are all perfectly acceptable. If everyone were using place names like those, we wouldn't be having this conversation! I was just trying to suggest that you start with a general location (like 'Toronto'), and then add something more specific (like "Edward's Garden"). It's really up to you how specific you want to be, but I think that is plenty specific. The idea is just to give people a rough idea of where you were geographically, not to pinpoint the location within a few meters. My third example is a bit more specific because most people have no idea where Khartum is (google maps will show you it's along Hwy 41), and because Dooley Creek Road is quite long. So for that example, I said "Khartum - Dooley Creek Rd - km 2", to indicate that the location is at about Km 2 along that road (as opposed to 10 km further down the road). Algonquin Park is quite large, so just saying "Algonquin PP" doesn't narrow things down very much. But when you add "Old Air Field", most Ontario butterfliers know roughly where you are talking about. Generally, we're looking for combinations like "Town - road" or "Park - Trail".

The problem with the default location supplied by iNat is that it might be something like "Toronto, Ontario, Canada", which isn't horrible, but it still covers a lot of ground. Or it might be as specific as a street address - it's very unpredictable. The real problem is with rural areas, where the default locations are extremely vague, especially in Northern Ontario. We sometimes get default locations like "Sudbury District", which is larger than some countries. This is why I suggested that folks use their judgement. Look at the default location that iNat has supplied, and consider whether it gives the average person a reasonable idea of where in the province you made your observation. Giving a more specific name is completely optional, but we really would appreciate it. I mostly posted this because some people are trying to give more specific location names in the notes. I just wanted them to be aware that there's a better place for it.

Another tip: If you replace the default location with one of your own, please don't put "ON", "Ontario", "CA", or "Canada" at the end of your location name. That's something that iNat typically does for their default locations, and we want to be able to distinguish the default locations from those supplied by observers (there may come a day when we figure out a clever way to replace all the default locations, so we need to be able to easily identify the locations supplied by iNat).

And yes, I'm a big believer in the "checklist" approach to reporting butterflies, where you go to a location, and you record a bunch of species in the general area, and then your report all those species as if you found them at one set of coordinates - in the approximate center of the area you explored. Then you set the "accuracy" figure to a radius that encompasses the area you covered. I just took a look at Edward's Garden on google maps. If you were to go there and walk around the Botanical Gardens, I would take all my photos, and do a simultaneous edit to apply the following coordinates:43.733221, -79.359064 to all of them, with an accuracy of about 250m, and I would give the location name as "Toronto - Edward's Garden". See my follow-up post on changing locations for more tips.

Cheers, Rick

Posted by rcavasin over 2 years ago

thanks. Maybe iNat can figure out a level for beginners and advanced and your level for tracking.

Posted by gary-james over 2 years ago

Hi Gary,

I'm not sure what you mean. Perhaps we should take the discussion to private message so that I can better understand your concerns.

Cheers, Rick

Posted by rcavasin over 2 years ago

Add a Comment

Sign In or Sign Up to add comments