Unusual date on 6 observations exported to GBIF

[pisum]

not sure what’s going on here or if the weirdness originates in iNaturalist or GBIF, but i wanted to make a note of it here.

i was looking at iNaturalist metrics over at GBIF (https://www.gbif.org/dataset/50c9509d-22c7-4a22-a47d-8c48425ef4a7/metrics), and the Occurrences by Year graph at the end of the page seems to start in year 1601:

Sure enough, it looks like there are 6 occurrences that have been loaded with a date of January 1, 1601:
https://www.gbif.org/occurrence/search?dataset_key=50c9509d-22c7-4a22-a47d-8c48425ef4a7&year=1601.

looking at the original iNaturalist record for one of these records, it looks like the submitted and observed dates are both in 2020.
iNat record: https://www.inaturalist.org/observations/47718591
iNat API response: https://api.inaturalist.org/v1/observations/47718591
GBIF record: https://www.gbif.org/occurrence/2634052212

[pisum]

looking at the comments in another observation from the same set (https://www.inaturalist.org/observations/47718596), it looks like these observations may have originally been created with 1601 as the observation date iNaturalist. so i think that’s the issue.

[carrieseltzer]

Written before pisum’s reply: Hmm, that is strange. The error could either be in the Darwin Core Archive iNat prepares for GBIF: http://www.inaturalist.org/observations/gbif-observations-dwca.zip (clicking will start a huge download—I don’t think my computer has enough free space for it at the moment!) or GBIF’s ingestion of it.

Good to know the likely source of the problem. There are often weird early dates in the GBIF dataset that seem to be corrected quickly enough that I can’t find them on iNaturalist when I search, e.g. 0002