Marie-Victorin Herbarium (MT) - Plantes vasculaires

Study Extent

Digitization and photography of vascular plant specimens from the Marie-Victorin Herbarium (MT, including WAT)

Sampling

Originally, data were manually digitized as text from specimen labels, without photography. Currently, digitization still occurs as before for specific projects, but a new process based on specimen photography has been initiated, with the aim of increasing the pace of data capture. The process is: minimal digitization of specimen, photography (photos stored on photo server accessible via URL address), publication of minimally digitized data with photo. This will be followed eventually by full digitization either of targeted collections or via crowd sourcing (in development). All specimens newly acquired by MT and returned loans are digitized (when data are not provided by donor; MT encourages donors to provide digital data used to generate labels) and photographed.

Quality Control

Initially, legacy data had been formatted to fit the new database structure and cleaned as well as possible, though they still contain inconsistencies. They are being corrected both opportunistically and in a systematic manner for determinations and geography data. Inaccurate georeferenced data are currently being corrected. New records are entered using controlled vocabularies where relevant and should contain fewer errors. Data from donations are processed via a workbench and compared with the controlled vocabularies to ensure consistency. Specimens are validated by a taxonomist before databasing to ensure identification accuracy.

Method steps

1. Examine specimen and annotations labels. Verify and update determination if qualified. Search specimen record in collection database and update where necessary, or create new record and populate fields with the information on the specimen label. Use controlled vocabularies where relevant. Determinations are recorded in a related table (included in this dataset as http://rs.gbif.org/extension/dwc/identification.xml), as are related projects and specimen exchange history.
2. Verify locality provided on specimen labels with online databases to ensure quality of geographic data. Georeferences provided on labels are validated using online maps and GIS tools and corrected at need, with the addition of a measure of inaccuracy.
3. Print the newly generated specimen number onto the physical specimen, to be used as reference to the datase and as accession number in the collection.

This dataset includes all digitized vascular plants and bryophytes in the collection. The top 3 families are Cyperaceae (15%), Asteraceae (9%) and Poaceae (5%). For more detailed information about the whole collection, see our inventory at http://dataset.canadensys.net/mt-inventory.

1. Plantae
   common name: Plants rank: kingdom
2. Equisetopsida
   common name: Terrestrial plants rank: class
3. Lycopodiidae
   common name: Lycopods rank: subclass
4. Psilotidae
   rank: subclass
5. Ophioglossidae
   rank: subclass
6. Marattiidae
   rank: subclass
7. Polypodiidae
   common name: Ferns rank: subclass
8. Equisetidae
   common name: Horsetails rank: subclass
9. Cycadidae
   common name: Cycads rank: subclass
10. Ginkgoideae
    common name: Ginkgo rank: subclass
11. Gnetidae
    rank: subclass
12. Pinidae
    common name: Conifers rank: subclass
13. Magnoliidae
    common name: Angiosperms rank: subclass

Global terrestrial coverage (except Antartica). Over 88% of the specimens in this dataset are collected in Canada, of which 70% in the province Quebec and 19.5% in Newfoundland. 5% of the specimens are from the United States and 3.5% from Cuba. For more detailed information about the whole collection, see our inventory at http://dataset.canadensys.net/mt-inventory.