This is interesting for another reason, if the observation that was corrected a few years later happens to be originally incorrect, what happens to the information that has already been transmitted to GBIF?
It’s always updated with new info upload to gbif (1-2 weeks), same as when obs are deleted from gbif if they’re deleted on iNat.
Thank you for the explanation!
I met the woman who actually processes the uploads from iNaturalist to GBIF, sitting at the GBIF table in the vendor’s section at the Ecological Society of America meeting in 2019. I asked her the same question. She said that they do not have a process to go back and change IDs on observations that had already been uploaded to GBIF. I explained that that was a problem, as there are many research grade IDs that are incorrect, which often get changed eventually. She said that maybe they could work that into their process, to do every year or so. I don’t know if that has happened. This conflicts with @marina_gorbunova’s answer, so I don’t know if @marina_gorbunova is describing a new process, or if the 1-2 week uploads represent new research-grade observations only.
I am totally sure all the new differences (RG > Needs id or obs deleted = obs gone from GBIF) actually happen each time new data set is getting to GBIF.
Thanks, @marina_gorbunova - if you don’t mind me asking, how do you know?
You can check your observations in GBIF, and when you change RG obs id there’s an indication that you need to wait for synchronization (sorry, not English version):
Cool, I didn’t think of checking my own observations. Thank you, @marina_gorbunova.
Either that has changed since you talked to her, or she had incomplete info, as another GBIF employee has posted the opposite here
Based on this and what @marina_gorbunova was saying, it sounds like records are getting corrected at this point in time - which is great!
Hi folks, just a quick note that I split out this separate question about GBIF to its own topic.
This is really cool @marina_gorbunova , would you be able to explain how you get that?
Well, just click on GBIF sign on observation to look at it or search by your name.
That was easy, thank you!
A related observation, in case anyone wonders if data seems less reliable if it can “change,” is that it updating based on further verification or revision improves overall accuracy. By contrast a misidentification “permanently” added to a physical museum specimen label or book is much harder to undo, and may mislead many for a long time. (Speaking of, GBIF does have some issues pertaining to certain records it receives from non-iNat sources, such as adding now-invalid names or taxonomic statuses from the labels of older museum specimens.)
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.