search

gbif-norway / helpdesk

Please submit your helpdesk request here (or send an email to helpdesk@gbif.no). We will also use this repo for documentation of node helpdesk cases.
GNU General Public License v3.0
3 stars 0 forks source link

DNV GL dataset -- OpenPSD report feedback on geo-coordinate issues #11

Closed dagendresen closed 2 years ago

dagendresen commented 3 years ago

From the GBIF CESP feedback to the OpenPSD final report:

The DNV dataset is associated with a number of flags relating to coordinates. For example, around 20,000 records have invalid coordinates such as this record (https://www.gbif.org/occurrence/2623177439) where the coordinates fields contain the value “inf”. Recommended action: Would it be possible to check coordinates for this dataset?

rukayaj commented 3 years ago

When they say "check", do you understand what they mean? I am not sure if they mean exclude/don't publish records without coordinates, or that we should ask DNV to georeference these records?

dagendresen commented 3 years ago

I guess they mean doing something with the geocoordinate strings reported as "inf". (Simply making these "inf" values NULL).

Screenshot 2021-02-26 at 15 33 17

Maybe these are some (error) output from a function converting from the UTM zone 33N (EPSG:32633) geocoordinates indicated as the verbatim source format.

However, would we dare to do something like "translating" "inf" to "NULL"??? (Maybe test on a test IPT first)

Screenshot 2021-02-26 at 15 39 13
dagendresen commented 3 years ago

I added occurrenceStatus = present

Doing something with the country is more demanding because there are multiple countries that the points in the sea are interpreted as... maybe be bold and just "claim" all the respective waters as Norway :-D

Screenshot 2021-02-26 at 15 56 21

Maybe GBIF is wrong...? Why would the oil companies monitor the environmental impact in the waters of UK and Denmark :-D 1-s2 0-S0308597X15001700-gr1

rukayaj commented 3 years ago

Fairly soon it will be time for me to publish this year's set of data from them, I plan to look at this issue at the same time. I guess it's easiest to map "inf" to NULL. I think the country issue isn't something we should tackle...

dagendresen commented 3 years ago

I had a meeting with OBIS after the previous GBIF Community webinar -- and will attend a meeting with OBIS and Colombia organized by the secretariat soon. The topic of "oceans" as more relevant than "countries" is sure to be relevant.

Recall that also Kjell Bjørklund was much more interested in "oceans" than "countries" for the Radiolaria dataset.

rukayaj commented 2 years ago

Republished with NULL instead of inf