Data Export

Three export types are available. Choose the one that matches your use case.

1 — GBIF / IPT Darwin Core Archive Export

Generates a strict Darwin Core Archive (DwC-A) zip package suitable for IPT submission and GBIF review. Internal-only fields are remapped to accepted DwC terms or packed into dynamicProperties. Formal GBIF publication still requires completing eml.xml placeholders and IPT registration.

Latest Export Package

Loading…

Package Contents
  • occurrence.csv — 28-column DwC occurrence records
  • meta.xml — DwC-A descriptor with term URIs
  • eml.xml — EML 2.1.1 metadata (complete TODOs before submitting)
  • validation_report.json — field coverage and QC checks
  • export_manifest.json — provenance and field mapping summary
Reads current DB records (or master CSV if DB unavailable). ~30 s.
Field mapping — internal → DwC-A
Internal fieldDwC-A outputRationale
sourceURLreferences (dcterms)URL of the source platform record
sourceRecordIDdynamicPropertiesPlatform ID — not a DwC term
qualityGradedynamicPropertiesiNaturalist-specific flag
vitalitydynamicPropertiesTreeKeeper tree health
iucnRedListCategorydynamicPropertiesIUCN enrichment — not a core DwC term
Before formal GBIF submission: open eml.xml, replace PLACEHOLDER_CONTACT_EMAIL, fill in <resourceLogoUrl>, confirm CC-BY 4.0 with your institution, and register via IPT.
2 — Research Data Export — by Source

Downloads a zip of pipeline outputs for one or all data sources. Includes the standardized CSV, extended CSV, QC exclusion logs, and the most recent run manifest. All 30 internal fields are preserved — this is a research/internal format, not a GBIF publication format.

SourceStandardized CSVExtended CSV QC logsRun manifest
iNaturalist
community observations		 ✓✓ (+9 annotation cols)✓✓
eBird
bird occurrence records		 ✓✓ (+6 or +19 cols)✓✓
TreeKeeper
campus tree/shrub inventory		 ✓✓ (+11 horticultural cols)✓✓
External
researcher uploads		 ✓✓ (+11 upload-provenance cols)✓✓

Raw API dump files are stored on the server under raw/{source}/ and are not bundled here due to variable file sizes.

3 — Filtered Occurrence Export

Filter the combined occurrence dataset and download a subset as CSV (30-column research format, all sources combined) or GeoJSON (geolocated records only). All filters use parameterized SQL queries — no arbitrary SQL is accepted. Leave all fields blank to download the full dataset.

CSV includes all 30 internal fields (research format). GeoJSON includes only records with coordinates.