Insect is from Kansas which is USA not China

gbif-portal commented 3 months ago

Insect is from Kansas which is USA not China

https://www.gbif.org/occurrence/2235721586

Locality does not match coordinates

User: See in registry - Send email System: Chrome 126.0.0 / Windows 10.0.0 Referer: https://www.gbif.org/occurrence/2235721586 Window size: width 2124 - height 1073 API log&_a=(columns:!(_source),filters:!(),index:'3390a910-fcda-11ea-a9ab-4375f2a9d11c',interval:auto,query:(language:kuery,query:''),sort:!())) Site log&_a=(columns:!(_source),filters:!(),index:'5c73f360-fce3-11ea-a9ab-4375f2a9d11c',interval:auto,query:(language:kuery,query:''),sort:!())) System health at time of feedback: OPERATIONAL datasetKey: e6137ccd-15b7-47ca-bb81-91dc3003a9b0 publishingOrgKey: 73fb29e5-c684-4edf-b2ee-41b13a4f329f

Node handles: @sformel-usgs

Mesibov commented 3 months ago

decimalLongitude for this should be negative. Interesting this error wasn't picked up in GBIF processing. In this dataset https://www.gbif.org/dataset/e6137ccd-15b7-47ca-bb81-91dc3003a9b0 there are 235 other records from Kansas (according to locality) with the longitude mistakenly entered as positive.

It would be nice to think that having been notified of a single-record error (as above), a data publisher would check their dataset for other records with the same error. This dataset has many other data problems, though.

sformel-usgs commented 2 months ago

@CatChapman I agree with @Mesibov about the flipped longitudes. You're listed under georeferencedBy, could you please take a look when you have a moment? The occurrenceIDs for which the longitudes should be negative, instead of positive, are:

CatChapman commented 2 months ago

Hi there, thanks for tagging me in this. I indeed georeferenced this many years ago, using the built-in GeoLocate tool in Symbiota. At the time, I reviewed batch of records I georeferenced to avoid issues like this, and it passed "QAQC" at that time. It looks like the longitude value was flipped from negative to positive between then and now - maybe something in the code is returning the absolute value instead of the negative value?

I'm not sure who manages the K-State data on SCAN these days. In my free time I can see if I still have edit privs on SCAN for these data, and if so, if these data points can be easily corrected on the Symbiota frontend.

Mesibov commented 2 months ago

@CatChapman, there are quite a few other data problems in this dataset. Do you think it would be worthwhile contacting the current data manager, or is that unlikely to lead to corrections due to time/funding/staff/resources issues?

CatChapman commented 2 months ago

@Mesibov - I don't think it would hurt to reach out, but I also would be realistic about expectations given your latter clause... only one way to find out, as they say. :)

sformel-usgs commented 2 months ago

I wrote an email to Greg Zolnerowich, the dataset contact.

gbif / portal-feedback

Insect is from Kansas which is USA not China #5408