search

trias-project / wrims-checklist

🐡 World Register of Introduced Marine Species (WRIMS)
MIT License
0 stars 0 forks source link

occurrenceStatus - establishmentMeans - Invasion Stage mapping #2

Open LienReyserhove opened 6 years ago

LienReyserhove commented 6 years ago

approach WRIMS

Currently, the following information is available in the distribution extension of WRIMS (only specified for Belgian species!):

approach TrIAS

For occurrenceStatus and establishmentMeans, we strive to use GBIF vocabulary. Information on Invasiveness is integrated under Invasion Stage in the description extension (for more information, see https://github.com/trias-project/alien-plants-belgium/issues/60 and https://github.com/trias-project/alien-fishes-checklist/issues/6).

Below, I give a full overview of all combinations of occurrenceStatus, establishmentMeans and Invasiveness in WRIMS, with an attempt to translate it to the TrIAS vocabulary:

(OS = occurrenceStatus, EM =establishmentMeans, I = invasiveness, IS = invasion stage)

WRIMS OS WRIMS EM WRIMS I TrIAS OS TrIAS EM TrIAS IS
uncertain origin unknown not specified doubtful uncertain
uncertain alien uncertain doubtful introduced
doubtful alien NA doubtful introduced
doubtful NA NA doubtful
excluded alien NA excluded introduced
excluded NA NA excluded
present alien NA present introduced
present origin uncertain NA present uncertain
present NA NA present
established alien invasive present introduced established
established alien not specifed present introduced established
established alien uncertain present introduced established
reported alien not invasive present introduced
reported alien not specified present introduced
reported alien of concern present introduced
reported alien uncertain present introduced
qgroom commented 6 years ago

in the Alien Plants Belgium list you use the term "invasion stage", but here you say "invasiveness". These are not the same thing. I assume invasiveness comes from WRIMS and TrIAS I is invasion stage. If this is the case then your mapping is fine.

LienReyserhove commented 6 years ago

Yes, invasiveness applies to WRIMS and invasion stage applies to TrIAS. I agree this is confusing in the table. I will change this.

LienReyserhove commented 6 years ago

This list will probably be longer. I filtered on the Belgian locations, but this by using the prefix "belg". This excludes locations like Westerschelde, De Panne, etc. (see #3 ). I will update this issue once we have the subset for all Belgian locations.

LienReyserhove commented 5 years ago

After selecting on the Belgian locations (country= BE), the issue remains more or less the same. But: to generate the unified checklist, we make a selection based on the following criteria:

occurrenceStatus= present, common, rare, irregular establishmentMeans= introduced, naturalized, invasive, assisted colonization

This has implications for our suggestions to map the WRIMS data to a controlled vocabulary. Some of the taxa in WRIMS will not be selected anyway (e.g. as they have an uncertain occurrenceStatus), even if we change the current WRIMS vocabulary. Other taxa however, should be integrated in the unified checklist but will not be selected unless we change the current WRIMS vocabulary. An overview:

WRIMS OS WRIMS EM WRIMS I needed in unified checklist? selected based on criteria?
uncertain origin unknown not specified no no
uncertain alien uncertain no no
doubtful alien NA no no
doubtful NA NA no no
excluded alien NA no no
excluded NA NA no no
present alien NA yes no
present origin uncertain no no
present NA NA no? no?
established alien invasive yes no
established alien not specifed yes no
established alien uncertain yes no
reported alien not invasive yes no
reported alien not specified yes no
reported alien of concern yes no
reported alien uncertain yes no

This means that we need to change the following vocabs:

occurrenceStatus established --> present occurrenceStatus reported --> present establishmentMeans alien --> introduced

With respect to invasiveness: this field should be in description. I will make another issue for that

qgroom commented 5 years ago

Perhaps you could explain the problem on Tuesday?

peterdesmet commented 5 years ago

The required mapping I see is:

  1. occurrenceStatus uncertain --> doubtful (or the alternative presence uncertain)
  2. occurrenceStatus established --> present
  3. occurrenceStatus reported --> present
  4. establishmentMeans alien --> introduced (or the alternative alien => no need to update)
  5. establishmentMeans origin unknown --> uncertain (or the alternative unknown)
  6. establishmentMeans origin uncertain --> uncertain
LienReyserhove commented 5 years ago

@bart-v --> what do you think about this suggestion?

bart-v commented 5 years ago

I'm OK with this.

But it's a pity you loose so much valuable information, just to make GBIF happy... i.e. "established" says so much more than "present" ...

peterdesmet commented 5 years ago

Thanks @bart-v, and I agree that we loose some information that way. @qgroom is leading a paper to suggest changes to Darwin Core to capture this information better. For established it would have:

LienReyserhove commented 5 years ago

I guess we can close this issue then Thanks for the discussion!

LienReyserhove commented 5 years ago

To summarize:

  1. Translation of occurrenceStatus in WRIMS to GBIF vocabulary:
WRIMS GBIF
uncertain doubtful
doubtful doubtful
excluded excluded
present present
established present
reported present
  1. Translation of establishmentMeans in WRIMS to GBIF vocabulary:
WRIMS GBIF
origin unknown uncertain
origin uncertain uncertain
alien introduced
  1. Invasiveness

This is a special case as we do not integrate this term in the distribution, but in the description extension. But this is how we translate Invasiveness in WRiMS to invasion stage in the description extension on GBIF:

LienReyserhove commented 5 years ago

The latest update of WRiMS did not contain the suggested changes. Rather, it includes even more terms for occurrenceStatus and some native taxa. @bart-v , could you please go through the suggestions, comment on this and integrate this in the next update of WRiMS?

  1. occurrenceStatus

New terms since the latest updates are indicated in bold

WRiMS GBIF
Absent absent
Detected in invasion pathway present
Detected in the wild present
doubtful doubtful
Eradicated absent
Established present
Established and expanding present
Established and stable present
excluded absent
In captivity/cultivated present
Not specified doubtful
present present
Recorded in error doubtful
Reported present
Sometimes present present
Uncertain doubtful
  1. establishmentMeans

No new terms for introduced species:

WRIMS GBIF
origin unknown uncertain
origin uncertain uncertain
alien introduced

But: some taxa have status native

WRIMS records
Native 113
Native Endemic 2
Native Non-endemic 3

Unless I'm missing something these records should be removed for WRiMS

  1. Invasiveness:

This term is not yet integrated into the description extension. This is the suggested content (with changes relating to the latest update in bold):

bart-v commented 5 years ago

I think TrIAS should do (script) this mapping, if really wanted.

We cannot publish WRiMS data with a GBIF vocab that does not include all the important detailed information provided by our editors. DwC is made for this: extensible via vocabs: so better use them...

LienReyserhove commented 5 years ago

OK, we can do the scripting if we need the information. These records will have a distribution:invalid flag This is how the information will look like on GBIF for Claviceps purpurea:

Input record:

term value
eventDate 1960
occurrenceStatus Established
locality Irish part of the Irish Sea and St. George's Channel
countryCode IE
establishmentMeans Alien
locationID http://marineregions.org/mrgid/25211
Invasiveness Of concern
source Nehring, S.; Boestfleisch, C.; Buhmann, A. (...)
occurrenceRemarks In the winter of 1960-1961 (...)

Record on GBIF:

verbatim information:

term value
locationID http://rs.tdwg.org/dwc/terms/locationID http://marineregions.org/mrgid/25211
locality Irish part of the Irish Sea and St. George's Channel
source Nehring, S.; Boestfleisch, C.; Buhmann, A.; Papenbrock, J. (2012). The North American toxic fungal pathogen G3 Claviceps purpurea (Fries) Tulasne is established in the German Wadden Sea. BioInvasions Records. 1(1): 5-10.
establishmentMeans Alien
occurrenceRemarks In the winter of 1960-1961, C. purpurea was detected in Townsend's cord-grass at Dublin Sites
eventDate 1960
occurrenceStatus Established
countryCode IE
Invasiveness Of concern

interpreted information:

PRESENT IN IRELAND, HTTP://MARINEREGIONS.ORG/MRGID/25211, IRISH PART OF THE IRISH SEA AND ST. GEORGE'S CHANNEL

Record available though the API:

{ "http://rs.tdwg.org/dwc/terms/locationID": "http://marineregions.org/mrgid/25211", "http://rs.tdwg.org/dwc/terms/locality": "Irish part of the Irish Sea and St. George's Channel", "http://purl.org/dc/terms/source": "Nehring, S.; Boestfleisch, C.; Buhmann, A.; Papenbrock, J. (2012). The North American toxic fungal pathogen G3 Claviceps purpurea (Fries) Tulasne is established in the German Wadden Sea. BioInvasions Records. 1(1): 5-10.", "http://rs.tdwg.org/dwc/terms/establishmentMeans": "Alien", "http://rs.tdwg.org/dwc/terms/occurrenceRemarks": "In the winter of 1960-1961, C. purpurea was detected in Townsend's cord-grass at Dublin Sites", "http://rs.tdwg.org/dwc/terms/eventDate": "1960", "http://rs.tdwg.org/dwc/terms/occurrenceStatus": "Established", "http://rs.tdwg.org/dwc/terms/countryCode": "IE", "http://www.marinespecies.org/traits/Invasiveness": "Of concern" }

peterdesmet commented 5 years ago

@LienReyserhove could you update your previous comment with actual data now that it is published?