Brazilian Flora contains far too many unmatched names

18480
Reporter: mdoering
Assignee: mdoering
Type: Bug
Summary: Brazilian Flora contains far too many unmatched names
Priority: Critical
Status: InProgress
Created: 2016-05-19 10:24:39.856
Updated: 2016-08-09 17:47:40.971
        
Description: The Brazilian flora has more than 100.000 unmatched names:
http://www.gbif.org/species/search?dataset_key=aacd816d-662c-49d2-ad1a-97e66e2a2908&issue=BACKBONE_MATCH_NONE

This seems wrong as the flora is used to build the backbone and random name test do actually exist in our backbone, e.g the genus Piper:

Brazilian Flora: http://www.gbif.org/species/114761199
Backbone: http://www.gbif.org/species/3075433

https://github.com/gbif/checklistbank/blob/master/checklistbank-nub/nub-sources.tsv#L24

So really pretty much all names should match.]]>
    

Author: mdoering@gbif.org
Created: 2016-07-04 16:10:49.226
Updated: 2016-07-04 16:10:49.226
        
An example that has not matched but should according to the service is Chamaecrista desvauxii:
http://api.gbif.org/v1/species/match?verbose=true&strict=true&family=Fabaceae&name=Chamaecrista%20desvauxii%20(Collad.)%20Killip

http://www.gbif.org/species/114721344
    

Author: mdoering@gbif.org
Created: 2016-08-04 13:31:25.88
Updated: 2016-08-04 22:16:20.16
        
there are only 2609 unmatched in the august 2016 backbone and it appears they all are from unsupported or higher ranks like TRIBE, CLASS:
http://api.gbif-uat.org/v1/species/search?dataset_key=aacd816d-662c-49d2-ad1a-97e66e2a2908&issue=BACKBONE_MATCH_NONE

still 1546 species:
http://api.gbif-uat.org/v1/species/search?dataset_key=aacd816d-662c-49d2-ad1a-97e66e2a2908&issue=BACKBONE_MATCH_NONE&rank=SPECIES
    

Author: rdmpage
Comment: At first glance lots of species with the name "inaccurate cedes", in other words identified to genus level only, and lots of _Euglena_ which have been lost with the removal of Algaebase (sigh).
Created: 2016-08-04 23:10:50.935
Updated: 2016-08-04 23:10:50.935

Author: joaomlanna@gmail.com
Created: 2016-08-08 22:40:30.111
Updated: 2016-08-08 22:40:30.111
        
Dear Markus,

I suppose that our list are already on the GBIF backbone. Am I right?

And about the 73.365 "Nomenclatural status invalid" in our statistics? What that means? Is it related to this POR-3107?

Regards,
    

Author: mdoering@gbif.org
Comment: [~joaomlanna@gmail.com] these "invalid" nomenclatural status flags mean that GBIF could not interpret them into our enumeration. This is nothing important and I will need to make sure we better understand them. But it's a bit down the priority line right now I am afraid. Nothing to worry about on your side
Created: 2016-08-09 17:47:40.971
Updated: 2016-08-09 17:47:40.971