Lewati ke konten

Basic tutorial 7: Review and export a Darwin Core Archive

Konten ini belum tersedia dalam bahasa Anda.

Darwin Core is the shared vocabulary that museum databases and portals such as GBIF use for specimen data. Turning a paper catalog into Darwin Core usually means retyping it into a spreadsheet, then fixing the spellings, units, dates, and missing localities by hand. Because you fixed those problems as you entered the data, NAHPU can build the archive from the linked records directly.

In this tutorial you review the seven specimens, explore their statistics, export them as a Darwin Core Archive, and read the result.

Check the records against the paper one last time. Tutorials 3–5 removed most problems at entry; this review looks for the rest.

CheckWhereWhat to look for
Field IDsSpecimens listHH436, HH437, HH442, HH453, HH456, HH523, HH529, each once
NamesEach specimenOne spelling per name; Maxomys (genus) only for HH 523
PlacesEach specimenPondok Merah for HH 436–456; Pasir Putih for HH 523 and HH 529
MethodsEach specimenPitfall for HH 453 and HH 529; Snap trap for the others
Explained choicesRemarksHH 437, HH 442, HH 456, HH 523, and HH 529 say what was changed and why
GapsSites and eventsNo coordinates for Pasir Putih; no pitfall count for Line 4

The gaps are not errors to fix before export. They are honest records of what the 2014 paper did not capture.

Before you export, look at the data as a whole. Exploring the record statistics is a NAHPU best practice. It shows imbalance in what was collected, shows how consistently parts were sampled, and, in the field, helps you decide which parts to prioritize when supplies such as tubes or ethanol run low.

On the dashboard, open the Record statistics panel and select Explore more stats. The Summary view counts the whole project:

NAHPU Record statistics summary on an iPad: 7 specimens, 19 specimen parts, 5 narratives, 7 recorded sites, 6 events, and a taxonomic breakdown of 2 orders, 3 families, 4 genera, and 5 species

Select Explore > Counts > Specimens. Group by Species, break down by Sex, and switch to Table:

NAHPU Explore view on an iPad showing specimens by species and sex as a table

Five of the seven specimens are male, and Maxomys hylomyoides is the only species with both sexes. Seven hand-picked records are not a sample, so read this as a demonstration. With a full catalog, the same table shows whether a skew is real, for example from trap behavior or season, and whether to note it before analysis.

Select Part quantity and group by Part type:

NAHPU Explore view on an iPad showing part quantity by part type: Liver 6, Whole body 4, Skeleton 3, Skin 3, Skull 3

Six of the seven specimens have a liver tissue. The exception is HH 442, whose form has no tissue circled. During fieldwork, check this table at the end of each day: it shows whether every specimen got the parts your protocol requires, and which part types use the most tubes, ethanol, and formalin. When supplies run short, it helps you decide what to keep collecting.

Every table here has an Export table button for reports. See Export statistics.

To see the records themselves, open Data table from Tools at the bottom of the dashboard. Choose what to show, such as Specimen records, Specimen parts, Sites, Events, or Narrative, then sort, search, or filter. This is the parts table for the seven specimens:

NAHPU Data table on an iPad listing specimen parts with type, count, treatment, cataloger, and field number

The Data table is a quick way to scan for gaps, such as a part with no treatment or a specimen with no tissue, without opening each record.

  1. Open the Dashboard, open the project menu, and select Bundle records.
  2. Choose Darwin Core Archive as the Bundle format. It always uses ZIP.
  3. Choose Selected taxa and select Mammals.
  4. Enter the file name mt-tujuh-2014 and choose a directory.
  5. Review Package contents. Expand occurrence.csv to see its fields. The list also shows any recorded fields that are withheld because they have no Darwin Core term.
  6. Read any warnings, then select Create bundle.

NAHPU writes mt-tujuh-2014.dwca.zip.

Unzip the file. It contains:

  • meta.xml
  • eml.xml
  • occurrence.csv
  • material.csv
  • measurement_or_fact.csv

occurrence.csv is the core table, with one row per specimen. The other tables hold parts and measurements and point back to their specimen. meta.xml describes the columns, and eml.xml holds the dataset metadata.

This is part of the HH 436 row. Your UUID and Event ID will differ.

Darwin Core termValueWhere it came from
occurrenceIDA UUIDAssigned by NAHPU when the record was created
basisOfRecordPreservedSpecimenSpecimen record
catalogNumberHH436Field ID
eventIDThe TJHH-L1 eventEvent
eventDate, eventTime2014-09-12, 16:00Specimen
scientificNameLariscus insignisTaxon registry
familySciuridaeTaxon registry
country, stateProvince, countyIndonesia, Jambi, KerinciSite
localityGunung Tujuh, Sungai Jernih, Pondok Merah, beside Lake Gunung TujuhSite
decimalLatitude, decimalLongitude-1.70274, 101.39032Site coordinate
minimumElevationInMeters2004Site coordinate
habitatType, condition, and descriptionSite
recordedByHeru HandikaEvent personnel
identifiedByHeru HandikaSpecimen
sex, lifeStageMale, SubadultSpecimen

On paper, many of these values lived in a sheet header, the journal, or the collector's memory. Here each comes from the record that owns it: the site, the event, the taxon registry, or the people you entered once.

PaperIn the archive
Five spellings of Maxomys hylomyoidesOne scientificName value
01.70274° S, crossed-out feet-1.70274; elevation in meters
Blank locality headersEvery row has its own locality
Pasir Putih filed under Pondok MerahHH 523 and HH 529 have the Pasir Putih locality and no coordinates
Line 1 for two methodsPitfall and snap-trap captures have separate events
Heru as collector on every formrecordedBy names the person from the event
SA and ALRows in material.csv with a type and treatment, such as Skull, Whole body in Formalin, or Liver in ETOH
TTL 238, TL 84, … with no unitsRows in measurement_or_fact.csv with names and units, such as total length, 238, mm

Across the seven Basic tutorials you:

  • read a paper catalog critically and listed its problems;
  • created a project with its people and a clean taxon registry;
  • turned one locality header into a camp and six line sites, and the journal's trap notes into six events;
  • entered seven specimens, choosing the best-supported value where the paper disagreed with itself and explaining each choice;
  • corrected and entered the field journal, linked to the camp where it was written;
  • checked sample balance and part sampling with record statistics and the Data table;
  • exported a standard Darwin Core Archive from the linked records.

Young Heru's catalog was better than most: it has standard measurements, trap lines, and a journal. NAHPU does not replace that care. It makes the links between place, time, method, people, and specimen part of the record, so they are not lost when the notebook is closed.