Research grade observations in GBIF

General
1

What is the the best way to make a list all of the research grade observations from a project that are also listed in GBIF without having to individually check each observation? Thank you.

2

There is no way to find records in GBIF as far as I know. If the parameters are precise enough you can of course query the GBIF database for records whose source is inat and do it that way.

3

Thanks, I’ll start working on that.

1 Like
4

You can also make research parameters as uploaded before X as GBIF gets data each week and new observations can’t physically be there.

5

welcome to the forum.

if there are a lot of observations in the project (thousands) and/or if there’s no obvious way to filter a similar set over in GBIF, then one way to approach this would be:

  1. export your project observations, including the observation id as a field.
  2. download the iNat to GBIF interface file, which is very large file (several GB: https://www.inaturalist.org/observations/gbif-observations-dwca.zip)
  3. compare your export vs the interface file.
    (the above process assumes that everything in the interface file will be loaded successfully into GBIF.)

other notes:

  • theoretically, every research grade observation with licenses that are permissive enough should end up over at GBIF.
  • there is an indicator on each observation that should tell you whether it made it to GBIF (shows up near the bottom of the observation on the observation page on the website), but that indicator is not updated very frequently.
6

If you do this, you want to do updated before not uploaded before as an observation can sit in a state that makes it ineligible for GBIF for some time, the key is when it achieves eligibility to go to GBIF, not when it was uploaded to the site.

There is also no guarantee how often GBIF process the file they receive, even if we know they get one weekly.

7

How would you set it, is it something you change in URL? Cause there’re only options for uploaded/made before/after. Anyway it’s an option to cut out big part of observations.

8

Rather than downloading this enormous file, I recommend searching GBIF just for observations from the iNaturalist dataset with geographic/taxonomic/temporal parameters matching the scope of your project and just downloading those instead for comparison.

Or, you could focus your efforts outreach to all participants in your project about licensing their observations such that they will be automatically shared with GBIF if they are research grade. Then you can just keep tabs on the research grade observations in your project and not worry about a tedious comparison between the two.

Occasionally our weekly export fails to generate for some reason or GBIF fails to ingest it, but in general it is safe to assume that it’s updated once a week. Worst case scenario is probably once a month.

1 Like
9

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