Suggestions for determining which species have not been previously documented (on iNat) at a particular place

I’m wanting to make a list of species that are “first time observed” at a place. For example, we host a bioblitz on a particular weekend. Which observations made that weekend are of species that had not been observed previously?

Could anyone offer some suggestions?

[Clarification added after original post: Just looking for “first documented on iNat” rather than first documented in general.]

1 Like

I made a tool for doing exactly this earlier in the year for the CNC. It’s probably a bit tricky to run for people without programming skills, but when the data is available I’m happy to run it for you and send you the results.

Here are some examples of the kind of report it makes:

https://waterpigs.co.uk/articles/citynaturechallenge-2021-wien-uniquely-observed-species/ (the better example as it includes a table of contents)

https://waterpigs.co.uk/articles/citynaturechallenge-2021-notable-species-in-graz/

And here’s the code if you’re interested: https://github.com/barnabywalters/inaturalist-project-stats/

I’m also happy to make small modifications if there are different features you need. Just let me know!

2 Likes

It would be interesting if something like this could be integrated into the site somehow too.

@pfau_tarleton Currently if you weren’t to use an external tool, the best manual way to do this is by searching the location and taxon in Explore and clicking on Species to see which have been observed and how many each. Finding which date and observer were earliest per species is easiest when there’s a low number of obs. Finally one thing to also remember is species may have been earlier recorded in a location in records on external databases, museum collections, publications, etc., so first to iNat is sometimes different from first recorded in general. In other cases, the first iNat locality record is also the first one in general too. I tend to check as many sources as I can to determine that.

That looks fantastic! I’d LOVE to take you up on that offer. Here is the link to the observation period and location in question.
https://www.inaturalist.org/observations?d1=2021-06-11&d2=2021-06-13&place_id=118103

Sorry for kind of co-opting this thread but it got me thinking. Is there already a code floating around that shows the first record for every species in a particular location? I thought it would be really cool to see if any of the observations in a project or for a particular user were the first in the area (a particular county for example).

1 Like

look at: https://forum.inaturalist.org/t/several-external-tools-for-inat-data-by-kildor/19906.

the first item listed in the original post might work, depending on whether or not you had a corresponding project set up for the place.

or you could ty the thing that i mention in the subsequent post in that thread.

these and additional options are also discussed here: https://forum.inaturalist.org/t/how-to-search-for-1st-records-in-a-region/25114/8. moderators might actually want to consider merging this thread into that thread.

Very often, indeed, most of the time, not just sometimes.

Here you go!

https://waterpigs.co.uk/static/2021-06-timberlake-report.html

I didn’t include county or national level notable species analysis, as USA-level analysis takes too long and I wasn’t sure how useful the county-level view would be.

In case it isn’t obvious: the Observers section lists, for each observer, the taxa for which either only they observed during the event, or only they have RG observations during the event. I’m happy to run the analysis again at a later date if you still have ongoing ID efforts.

Feel free to restructure and reuse the report however you want.

1 Like

My place is an 800 acre property. So every first record on iNat is a real first record in general :)

1 Like

Thanks!!

Is this the definition of notable observations for this output?
“These species have either been observed ≤ 5 times within Timberlake [the place], or ≤ 10 times on iNaturalist globally.”

if you’re just looking for taxon-level counts of observations in your set being equal or close to the same-taxon count of observations recorded in all of iNaturalist, then i think this is the easiest way to get that: https://jumear.github.io/stirfry/iNatAPIv1_observations_species_counts.html?d1=2021-06-11&d2=2021-06-13&place_id=118103&per_page=500. just look for cases where the count for your set is equal or close to the taxon total count.

this method wouldn’t get:

… but (assuming there aren’t a lot of observations elsewhere occurring since the d1 start date, or anywhere outside of d1 to d2 date range) it would approximate:

Yes, exactly, ≤5 for Timberlake and Texas, and ≤10 globally. I can configure either threshold differently if necessary.

Possibly for that specific parcel of land, but unlikely for the larger area that your land falls in, or of the species.

If you pick an arbitrarily small chunk of land then technically almost every observation will be the “first” record for that area.

Of course, that goes without saying. Our purpose is to take a biodiversity inventory of this particular parcel of land. I’m going to be using the “first observation” information to track the number of newly-documented species on this parcel annually. We expect the plants and charismatic fauna to level off first and the more secretive taxa to continue increasing over time. It’s for publicity, in part. To inform participants that their efforts are still productive–and it can help to focus our efforts as we continue to take inventory and monitor biodiversity on the property. It has also already proven useful to find observations that were misidentified.

1 Like