Closed peterdesmet closed 5 years ago
Dropping filter on status present
adds 2592 -> 2638 = 46 new taxa overall
Or perhaps NOT ABSENT would be more future proof.
Brilliant!! Requires an update of the gbif_has_distribution()
function, so cannot test right away. Would you also exclude distributions that are defined as DOUBTFUL
?
Yes, ignore DOUBTFUL
Implemented in #39. It's an improvement overall, the only negative effect is that two distributions now lost their years (they had it in RINSE pathways, but not in WRIMS and WRIMS takes precedence).
From https://github.com/trias-project/wrims-checklist/issues/2#issuecomment-483296534, these are the values WRIMS has for:
WRiMS value | interpretation by GBIF | in unified |
---|---|---|
Absent | ABSENT | excluded by filter π |
Detected in invasion pathway | - | included π |
Detected in the wild | - | included π |
doubtful | DOUBTFUL | excluded by filter π |
Eradicated | - | should be excluded π, but no records for Belgium π |
Established | - | included π |
Established and expanding | - | included π |
Established and stable | - | included π |
excluded | EXCLUDED | excluded by filter π |
In captivity/cultivated | - | Β included π |
Not specified | - (https://github.com/gbif/parsers/issues/16: DOUBTFUL ) |
should be excluded π, but no records for Belgium π |
present | PRESENT | included π |
Recorded in error | - (https://github.com/gbif/parsers/issues/15: DOUBTFUL ) |
should be excluded π, but no records for Belgium π |
Reported | EXCLUDED (https://github.com/gbif/parsers/issues/15: PRESENT ) |
excluded by filter π |
Sometimes present | - | included π |
Uncertain | - (https://github.com/gbif/parsers/issues/16: DOUBTFUL ) |
should be excluded π |
WRIMS value | GBIF interpretation | in unified |
---|---|---|
Origin unknown | Β UNCERTAIN | not included π |
Origin uncertain | UNCERTAIN | not included π |
Alien | INTRODUCED | Β included π |
Native | Β NATIVE | not included π |
Native Endemic | - (https://github.com/gbif/parsers/issues/17: ENDEMIC ) |
not included π |
Native Non-endemic | - (https://github.com/gbif/parsers/issues/17: NATIVE ) |
not included π |
So most of it good, but will report the bold ones to GBIF for better mapping.
@qgroom, are there any other values in https://github.com/trias-project/unified-checklist/issues/37#issuecomment-493006247 that you would map to GBIF vocab (those with a -
in column 2)? E.g.:
Eradicated -> ABSENT ?
Established -> ??
If so, I will report them to GBIF
I'm not total clear on the question, but here is my response anyway...
detected in wild -> PRESENT Established -> PRESENT Established and expanding -> PRESENT Established and stable -> PRESENT
Sometimes present -> IRREGULAR
In captivity/cultivated -> dwc:degreeOfEstablishment category B
https://github.com/gbif/parsers/issues/15 is now resolved, so taxa that are "reported" are now longer excluded. Will rerun pipeline.
@qgroom, your suggestions seem ok, but less important to report to GBIF as they don't effect the pipeline.
WRIMS distributions need a combination of:
... to be selected. Ideally they have temporal information:
I notice many distributions only have part of the information: e.g. year, but not status. Search for example for
"BE"
in https://api.gbif.org/v1/species/157131005/distributions: there are 8 distributions for Belgium, but none have all 3 properties and year.I wonder if we should drop the
status
field in our selection.