new dwc_occurrence mapping Part1

[LienReyserhove]

This pull request contains several (major) changes to the dwc_occurrence mapping. This is because we decided to go back one step and to extract the raw columns in SSMS to generate the new denormalized_observations.csv (see #17 ).

As a consequence, we had to:

• remove the old dwc_occurrence.R and dwc_occurrence.md files and rewrite the script.
• carry out the mapping to DwC terms in R instead of in SQL.

Information to perform the mapping was extracted from SSMS and took into account the suggestions from #18

• [x] type
• [x] language
• [x] license: http://creativecommons.org/publicdomain/zero/1.0/
• [x] rightsHolder: Ghent University Aquatic Ecology
• [x] accessRights
• [x] datasetID: updated to https://doi.org/10.15468/xjtfoo
• [x] institutionCode: updated
• [x] datasetName: updated to Alien macroinvertebrates in Flanders, Belgium (was macro-invertebrates before)
• [x] basisOfRecord
• [x] occurrenceID
• [x] recordedBy: Van de Neucker T → Van den Neucker T, rest ok
• [x] otherCatalogNumbers
• [x] eventDate: use startDate
• [x] verbatimEventDate: populate for ranges only
• [x] continent
• [x] countryCode
• [x] municipality: not a vocabulary --> contains codes (e.g. "ES75D") --> remove these to verbatimLocality? not worth it
• [x] verbatimLocality: contains NULL. Translate all NULL to NA when reading dataset in R
• [x] decimalLatitude:change comma to point and rounded
• [x] decimalLongitude:change comma to point and rounded
• [x] geodeticDatum
• [x] coordinateUncertaintyInMeters: always 30 (assumed GPS for all)
• [x] verbatimCoordinates: remove
• [x] verbatimLatitude: see comment below
• [x] verbatimLongitude: see comment below
• [x] verbatimCoordinateSystem: decimal degrees or Belgian Lambert 72
• [x] verbatimSRS: WGS84 or Belgian Datum 1972
• [x] identifiedBy: Van de Neucker T → Van den Neucker T, rest ok
• [x] scientificName: Dreissena (Dreissena) polymorpha was changed to Dreissena polymorpha
• [x] kingdom
• [x] taxonRank: changed from "Spp" and "SubSpp"
• [x] scientificNameAuthorship: = from raw data, looks more or less OK
• [x] nomenclaturalCode

[peterdesmet]

How to map the verbatim coordinate information: use sample_spatial_ref_system to differentiate and do:

166956.8952,214232.8477 (BD72)

• verbatimLatitude: 214233
• verbatimLongitude: 166957
• verbatimCoordinateSystem: Belgium Lambert 72
• verbatimSRS: Belgian Datum 1972

51.12857222N,4.35763056E (LTLN)

• verbatimLatitude: 51.12857222N
• verbatimLongitude: 4.35763056E
• verbatimCoordinateSystem: decimal degrees
• verbatimSRS: WGS84

[LienReyserhove]

I integrated all earlier comments in the new dwc_occurrence mapping. Please check off in earlier PR body when considered OK

• license: http://creativecommons.org/publicdomain/zero/1.0/
• rightsHolder: Ghent University Aquatic Ecology
• recordedBy: Van de Neucker T → Van den Neucker T
• eventDate, not sure if GBIF can handle the "/" already: eventDate and verbatimEventDate were adapted:
• eventDate = start date
• verbatimEventdate = interval when start date != end date
• verbatimEventdate = empty when start date == end date
• verbatimLocality: all NULL translated to NA when reading dataset in R
• coordinateUncertaintyInMeters: always 30 (assumed GPS for all)
• verbatimCoordinates: removed and changed to verbatimLatitude and verbatimLongitude
• verbatimLatitude: corrected, based on comments given in "How to map the verbatim coordinate information"
• verbatimLongitude: corrected, based on comments given in "How to map the verbatim coordinate information"
• verbatimCoordinateSystem: decimal degrees or Belgian Lambert 72
• verbatimSRS: WGS84 or Belgian Datum 1972
• identifiedBy: Van de Neucker T → Van den Neucker T; Rik Clicque" = "Clique R", --> "Cliqcue R"

[peterdesmet]

Verified open items on todo list. Looks all good. 👏 👏 👏