Yes I believe it is. I made a note internally earlier this year when the iNaturalist GBIF dataset passed the Smithsonian (all of its GBIF datasets together) in number of citations. The current tallies are 1905 and 1767 citations, respectively. (They were 965 and 956 when I noted that in January).
In 2019 the GBIF Data Blog had a post on citizen science on GBIF that show some stats on taxonomic breadth which is correlated with citations (that’s my inference). Here’s the top 3 most speciose from that list with their current GBIF citations:
I think the taxonomic spread of iNat data is key here. Despite its vast numbers, eBird represents a single class of organisms, where iNat (in GBIF at least) has more than 200!
The majority of the papers I checked through this link don’t explicitly cite iNaturalist, but are checked on the left column as being publisher = iNaturalist.
What does this mean exactly?
Following the 2nd link you provided then clicking on citations, the first paper I see is https://www.biorxiv.org/content/10.1101/2021.10.26.465896v1
Downloading the full pdf and searching for the string “inaturalist” within the pdf I see no instances. I checked 10-15 others and found most to be the same. I guess I must be misunderstanding what you / others are referring to here. … .I expected there to be direct and visible citations in the papers referring to iNaturalist.
GBIF is not scanning the scientific literature to draw and represent on its website how often all its data partners have been cited.
The citation numbers shown on the respective GBIF-contributors are referring to those papers, where the GBIF-dataset is referenced. If that dataset contains observations from iNaturalist, it counts as a citation. So, in each paper you should be successful by searching for ‘GBIF’ and check in that dataset for iNat observations. At least that’s how I understood.
That’s correct. When users download data from GBIF—for example, for a given species or a country—the returned data is often derived from dozens if not hundreds of datasets and thereby publishers. We assign a single unifying DOI for the download that summarizes the contributions of each dataset.
In the case of the palms preprint, a GBIF download is cited (https://doi.org/10.15468/dl.rjmqfy) containing 800k palm records of which around 6k came from iNaturalist.
eBird has a different system than iNaturalist. All observations are self-reported and a relatively small number of official reviewers check for mistakes or questionable sightings. So there are no identifiers confirming each observation, although you can flag a photo as misidentified (most observations don’t have photos).
Thanks, my “comment” was a round about way of asking about ebird “identifiers”. Because @vihaking says to quote “ebird is more successful”
Am a bit unsure about the ratio of “identifiers” to “observations” and how that can be used as a metric for success.
That ebird has a lot more individual species observations there is no doubt - but then another metric is how many of them are incontrovertibly proven (inat is based on pictures and / or sound) while ebird is based on trust, statistical possibility, and the knowledge base on a handful of experts (who make up the filters etc) and only in a fraction of cases are the observations based on video, audio or pictures.