The Fungal Literature-based Occurrence Database for the Southern West Siberia (Russia )

Study Extent

The project was aimed at mobilization of species records accumulated in the course of previous mycological studies and published in peer-reviewed scientific literature from the beginning of research up to date. The geography extended the Sorthern part of West Siberia, in the administrative borders of ten regions. About 900 publications were compiled in a bibliography list and the species occurrence records were extracted from about 112 selected works by the time of the lase dataset update. The initiative will be continued in the nearest years to extract the records from all published sources.

Sampling

Methods of sampling vary in different reviewed publications, but generally follow the protocols of Mueller et al. (2004) for different taxonomical and ecological groups. The majority of the records were made using direct observation of fruiting structures. The exception are a few studies of micromycetes and yeasts where cultivation techniques were applied. No molecular (environmental sampling) methods were used until present to reveal molecular diversity of fungi in the region. The majority of fungal occurrence records were accompanied by accessioning of specimens in fungaria, the specimen numbers are rarely reported in publications, however. The specimens are stored in different collections within and outside the region where the researchers working (mentioning the main collections of NSK - Central Siberian Botanical Garden, Novosibirsk; National Research Tomsk State University, and others).

Quality Control

The original species identifications from the published works were recorded in the database as published, no attempt was made for the revision of the species identification accuracy. The incorrect spelling of taxa was verified using GBIF Species Matching tool at the later stage of the database compilation.

Method steps

1. The bibliography of related publications was compiled using Zotero bibliographic manager. Only published works (peer-reviewed papers, conference proceedings, PhD theses, monographs or book chapters) were selected. If possible, the sources were scanned and added to the library as PDF files.
2. The template of the FuSWS database was made with Google Sheets and simple Microsoft Excel template. The Darwin Core standard was applied to the database field structure to accommodate the relevant information extracted from the publications. Totally 29 fields were selected to describe the literature-based occurrence data in the most detail.
3. From the available bibliography of publications related to the region, the works with species occurrence reports were selected for the databasing purpose. The main source of occurrences were annotated species lists with exact localities of the records. However, different sorts of other species citations were also included, provided that they had the connection to any geography and could be georepherenced at least to the regional level.
4. Most of the occurrences were georeferenced, either from the coordinates provided in the paper, or from the verbatim description of the field work locality. The georeferencing of the verbatim descriptions was made using Yandex or Google map services. Depending on the quality of georeference provided in publications, the uncertainty was estimated as follows: 1) the coordinate of a fruiting structure or a plot provided in the publication gives the uncertainty about 3-30 meters; 2) the coordinate of the field work locality provided in publication gives the uncertainty between 500 m to 5 km; 3) the report of the species presence in a particular region gives the centroid of the area with the uncertainty radius to include its borders.
5. The locality names were reserved in the field «verbatimLocality» for accuracy.
6. When possible, the «eventDate» was extracted from the annotation data. Whenever this information was absent, the date of the publication was used instead with the remarks in the «verbatimEventDate» field about the origin of the date. 
7. The ecological features, habitat or relief were written in the «habitat» field and reserved in Russian.
8. The substrate is important feature of fungal occurrences and was extracted in the «fieldNotes» field.
9. Other annotation records, including the abundance, etc., was accomodated in the «occurrenceRemarks» field.
10. The original scientific names reported in publications were filled in the «verbatimScientificName» field. Correction of spelling errors was made using the GBIF Species Matching tool. This tool was also used to create the additional fields of taxonomic hierarchy from species to kingdom, to fill in the «taxonRank» field and to synonymize according to the GBIF Backbone Taxonomy.
11. To track the digitization process, a metedata worksheet was maintained. Each bibliographic record had a series of fields to describe the digitization process and its results: the total number of extracted occurrence records, general description of the occurrence quality, presence of the observation date, details of georeferencing and the name of a person responsible for the digitization.

,

1. Fungi
   rank: kingdom
2. Chromista
   rank: kingdom
3. Protozoa
   rank: kingdom

The dataset is limited by the administrative borders of ten regions (Tyumen, Sverdlovsk, Omsk, Chelyabinsk, Kurgan, Tomsk, Novosibirsk, Kemerovo, Altay, and Gorny Altay).