Galiano Island BC Canada Marine Zoology 1893–2021

Study Extent

Study extent was roughly delimited by bathymetric and geographic boundaries, including: Porlier Pass (north of Galiano Island), Active Pass (south of Galiano Island), the Trincomali Channel (west of Galiano Island), and the Outer Island fault line, which lies in the Strait of Georgia (east of Galiano Island). For a GeoJSON file precisely designating the study area please see: https://github.com/IMERSS/imerss-bioinfo/blob/data-paper-i-final/data/Galiano/Galiano_Island_Project_Boundary_Chu_final_2021-02-23.json

Sampling

Sampling procedures vary according to each data source, including systematic dive records (Pacific Marine Life Surveys), ecological data collected by ROV (Chu and Leys 2010, Chu and Leys 2012), museum voucher specimens (Canadian Museum of Nature, Royal British Columbia Museum), crowd-sourced citizen science observations (British Columbia Cetacean Sightings Network and iNaturalist observations on the Biodiversity Galiano Island Project), and additional reports from the literature (Agassiz 1862, McMurrich 1921, Erickson 2000). Museum specimens may represent collections made through heterogenous methods. All raw catalogues, processing scripts, and processed catalogues contributing to this paper have been tagged in GitHub at https://github.com/IMERSS/imerss-bioinfo/tree/data-paper-i-final .

Quality Control

Curation of this dataset was facilitated through a rigorous review of taxonomic summaries and catalogues of occurrence data based around each phylum and data source. The algorithms we designed summarized all taxa represented in source catalogues by phylum. These summaries were then made available in Google Sheets for expert review. Based on the critical remarks added by experts to these taxonomic summaries, catalogues were reviewed and revised as necessary. This iterative process was continued until there was a one-to-one correspondence between taxonomic summaries and catalogues of occurrence records. Our algorithms conserved memory of all modifications, including typographic errors, and taxonomic and nomenclatural changes. iNaturalist observations were thoroughly reviewed and identifications added on the iNaturalist platform. Other sources were modified in collaboration with contributing authorities. Where disagreements arose between our critical review process and occurrence records published by sources such as the Royal BC Museum and Canadian Museum of Nature, we have added critical annotations identifying the discrepancy between species names reported in this dataset vs those reported by institutions. We also reported these discrepancies directly to museum curators. Georeferencing was also reviewed and corrected where appropriate based on the best available metadata. These change are recorded in our finalized catalogue of occurrence records.

Method steps

1. The data catalogues contributing to this dataset have been normalised, aligned, corrected, and rendered into visualisations by a collection of open source data processing scripts written in JavaScript. These scripts operate in the following stages: 1. The columns of each source catalogue, imported as CSV, are mapped onto a common core of fields drawn from a subset of the Darwin Core standard, as well as other project-specific fields 2. The taxon name is mapped onto a core backbone by means of a taxon resolution file which resolves preferred taxon names and accounts for typographical errors 3. A dataset id is assigned to every source catalogue, and they are then combined into a single master catalogue  4. This catalogue is filtered to include only the taxa of interest—marine fauna 5. Private or obscured coordinates held in project-specific fields are copied into the principal georeferencing fields 6. A patch of georeferencing corrections is then applied to the resulting coordinates, together with curational notes motivating the corrections 7. The resulting observations are then filtered by the polygon representing the project area 8. The resulting output produces two consolidated CSV files, a catalogue of all observations, and a master summary file 9. The master summary file is then divided into phyla according to the checklist divisions in this paper, for curation by subject matter experts—these are exported into a Google Sheets representation where they may edit them live 10. The subject matter experts add and check authorities for the taxa, add curation notes and resolve taxonomic discrepancies 11. After curation, the Google Sheets are then re-ingested, combined, and converted back into CSV, and compared with the original summary produced at Step 8 12. Any discrepancies between these summaries are fed into the taxon resolution file at Step 2, and, where appropriate, circulated amongst the managers of the source catalogues to incorporate corrections they find desirable 13. The process is then rerun from Step 1 until repeated passes of curation and reconciliation give rise to no further discrepancies at Step 12 The files output at Step 8 of the pipeline form the basis of the map-based data visualisations referenced from this paper, as well our our Darwin Core data submission to the Global Biodiversity Information Facility.

Taxonomic groups covered include: Porifera, Cnidaria, Ctenophora, Nemertea, Platyhelminthes, Chaetognatha, Mollusca, Annelida, Sipuncula, Arthropoda, Entoprocta, Brachiopoda, Bryozoa, Phoronida, Echinodermata, Chordata

1. Porifera
   rank: phylum
2. Cnidaria
   rank: phylum
3. Ctenophora
   rank: phylum
4. Nemertea
   rank: phylum
5. Platyhelminthes
   rank: phylum
6. Chaetognatha
   rank: phylum
7. Mollusca
   rank: phylum
8. Annelida
   rank: phylum
9. Sipuncula
   rank: phylum
10. Arthropoda
    rank: phylum
11. Entoprocta
    rank: phylum
12. Brachiopoda
    rank: phylum
13. Bryozoa
    rank: phylum
14. Phoronida
    rank: phylum
15. Echinodermata
    rank: phylum
16. Chordata
    rank: phylum

Location: Galiano Island, British Columbia, Canada. For a GeoJSON file precisely designating the study area please see: https://github.com/IMERSS/imerss-bioinfo/blob/8df8a3847aa71e5c28a57f558204ea58e42c15c2/data/Galiano/Galiano_Island_Project_Boundary_Chu_final_2021-02-23.json