search

trias-project / ad-hoc-checklist

🍀 Ad hoc checklist of alien species in Belgium
https://trias-project.github.io/ad-hoc-checklist
MIT License
1 stars 2 forks source link

Empty field date first observation in ad-hoc raw data #34

Closed damianooldoni closed 1 month ago

damianooldoni commented 5 years ago

While working on pathway indicator, I found that Rattus Norvegicus is published in unified checklist with temporal: 2018/2018. See distribution here: http://api.gbif.org/v1/species/152543293/distributions.

This is due to the fact that while choosing a single distribution across checklists the Ad-hoc checklist is more trustworthy than RINSE. Distribution from ad-hoc: http://api.gbif.org/v1/species/148437763/distributions Distribution from RINSE: http://api.gbif.org/v1/species/146042175/distributions

I don't understand why this species is in the ad-hoc checklist. Maybe for the description? http://api.gbif.org/v1/species/148437763/descriptions where invasion stage: established? If it should be present in ad-hoc, the temporal window should be changed. Otherwise we are stating the following: the alien species Rattus norvegicus is established AND it has been for the first time observed in 2018, quite contradictory and not true. @timadriaens : any idea?

timadriaens commented 5 years ago

date first observation in add hoc is empty so something happens in flow to unified

damianooldoni commented 5 years ago

In src/dwc_mapping.Rmd-line 494-502

It seems a problem in interpretation of the empty field.

timadriaens commented 5 years ago

don't know why we chose to fill this field with checklist publication date in case it was NA, but clearly for many species this is just wrong.

damianooldoni commented 5 years ago

Still, I don't understand why this specie is in ad-hoc as we have both distribution AND description about pathways in RINSE: http://api.gbif.org/v1/species/146042175/distributions http://api.gbif.org/v1/species/146042175/descriptions If we need ad-hoc just to have regional level information, then why not use same temporal information from RINSE?

damianooldoni commented 5 years ago

Same for Myocastor coypus, which is one of the EU concern species. In our unified checklist, we assess: temporal: 2018/2018. Again, this is due to the interpretation of empty field date first observation in ad-hoc raw data and higher priority of ad-hoc checklist than RINSE. Distribution in RINSE (Belgian level): temporal: "1963/2017". See: http://api.gbif.org/v1/species/146042157/distributions. Distribution in ad-hoc checklist (Flemish level): temporal: "2018/2018". See: http://api.gbif.org/v1/species/148437776/distributions. A solution would be to set the ad-hoc checklist to lowest priority. As we are going to introduce regional distributions in unified, the flemish distribution in ad-hoc will still pop up in the unified checklist. But, still, that field should not be empty otherwise we cannot define the temporal window on presence in Flanders, or we define it wrongly.

LienReyserhove commented 1 month ago

I see two questions here (if these questions are still valid)

1: should we give priority to the RINSE checklist in favor or the ad-hoc checklist --> I don't think so 2: what should we do with an empty value in the date of first introduction? This is covered here: https://github.com/trias-project/ad-hoc-checklist/issues/42

I think we can close this issue?

LienReyserhove commented 1 month ago

This question is answered here: https://github.com/trias-project/ad-hoc-checklist/issues/42#issuecomment-2166039019