Skip to contents

Darwin Core Output for LTER-MareChiara Legacy Data

Source: vignettes/darwin-core.Rmd
darwin-core.Rmd

Overview

Darwin Core is the international data standard required by OBIS, EMODnet Biology, and GBIF to publish biodiversity occurrence records. It structures data into three linked tables — sampling events, species occurrences, and associated measurements — allowing any repository worldwide to ingest and cross-reference the records. Converting the LTER-MareChiara dataset to this format is the step that makes it ready for submission to these repositories.

This vignette shows how the package builds those Darwin Core tables from the LTER-MareChiara zooplankton dataset. The conversion is handled by format_to_dc(), which downloads the merged tidy dataset from SharePoint (produced by format_to_tidy()), adds the required Darwin Core fields, and returns the three tables as an R list. The function does not upload anything — that is handled by format_to_DC_archive(), which calls format_to_dc() internally and then assembles and uploads the Darwin Core Archive zip. The function uses data that already contains WoRMS LSIDs and does not perform additional taxonomic queries.

Prerequisites

  • inst/config.yml configured with SharePoint credentials and bucket names.
  • The merged dataset must exist in SharePoint (produced by running format_to_tidy() earlier in the pipeline). This dataset covers 1984-2024 and contains preprocessed records with WoRMS LSIDs.
  • Optional: set verbose = FALSE to silence log messages.

Expected Input Structure

format_to_dc() expects the tidy dataset to contain at least:

  • eventID: sampling event identifier (for example "mc_1").
  • eventDate: sampling date (Date column).
  • scientificname: WoRMS validated taxon name.
  • lsid: WoRMS Life Science Identifier.
  • Abundance: abundance measurement (individuals per cubic meter).
  • lifeStage: life stage codes (f, m, j, fm, fmj).

Any extra columns are carried through and pivoted into the measurement table.

Running the conversion 

# Convert tidy data to Darwin Core format
# Tables are uploaded to SharePoint automatically
format_to_dc()

The function downloads the merged dataset from SharePoint and builds the three Darwin Core tables (Event, Occurrence, eMoF) in memory. It does not upload anything to SharePoint.

To build a Darwin Core Archive and upload it to SharePoint, call format_to_DC_archive(). This function calls format_to_dc() internally, so in the automated pipeline it is the only function you need to run:

# Build DwC-A zip (runs format_to_dc() internally) and upload to SharePoint
format_to_DC_archive()

Darwin Core tables

Event

format_to_dc() builds one row per unique event with fixed station metadata.

# After downloading the DC tables from SharePoint:
# event_table %>%
#   select(eventID, eventDate, decimalLatitude, decimalLongitude, samplingProtocol) %>%
#   head()

Columns include:

  • eventID, eventDate
  • decimalLatitude = 40.81, decimalLongitude = 14.25
  • locality, country, stateProvince, waterBody
  • maximumDepthInMeters = 50, minimumDepthInMeters = 0
  • samplingProtocol, sampleSizeValue (= 1), sampleSizeUnit (= “sample”)

Occurrence

Occurrences are derived from the tidy data with an automatically generated occurrenceID and presence flag.

  • occurrenceStatus is set to "present" when Abundance > 0, otherwise "absent".
  • scientificName and scientificNameID come directly from the input (no new validation is run).

eMoF (Extended Measurement or Fact)

eMoF is the Darwin Core extension used to store quantitative measurements alongside each occurrence record. For the LTER-MareChiara dataset this includes abundance (individuals per m³), sex, and life stage. Each measurement is linked to a controlled-vocabulary identifier so that the values can be interpreted unambiguously by any international biodiversity database.

Mapping logic implemented in format_to_dc():

  • Measurements with values f, m, fm, fmj are labelled as measurementType = "sex" with measurementTypeID = http://vocab.nerc.ac.uk/collection/P01/current/ENTSEX01/.
  • Measurements with value j are labelled measurementType = "lifeStage" with measurementTypeID = http://vocab.nerc.ac.uk/collection/P01/current/LSTAGE01/.
  • Abundance keeps measurementType = "Abundance" with measurementTypeID = https://vocab.nerc.ac.uk/collection/S06/current/S0600002/ and measurementUnitID = http://vocab.nerc.ac.uk/collection/P06/current/UPMM/.
  • measurementValueID is set for the coded values above (S10/S11 URIs); other measurements keep NA.

Processing summary

format_to_dc() prints a processing summary on completion: the number of sampling events, occurrences, and measurements produced, the date range covered, and the number of unique taxa included.

Publishing (optional)

If you need to publish the archive to GBIF:

Limitations and next steps

  • The function assumes WoRMS identifiers are already present; it does not call external taxonomic services.
  • No automated quality control is applied beyond field renaming and simple presence/absence derivation. Add your own checks before publication.