The study site – the Mukhrino field station and the Mukhrino Bog – is located in the middle taiga zone of Western Siberia, near the regional capital city of Khanty-Mansiysk (60.89N, 68.68E) (Fig. 1). The Mukhrino Bog is an ombrotrophic landscape entity covering an area of about 10 by 15 km, located along the northern edge of a larger paludified area, the Konda Lowlands (Russian Кондинская низменность), on the left terrace of the Irtysh River close to its confluence with the Ob'. The vegetation of the raised bog is represented by the typical ombrotrophic or oligo-mesotrophic communities from the geobotanical classes Scheuchzerio-Caricetea nigrae, Oxycocco-Sphagnetea and Vaccinio-Piceetea. Two major vegetation types dominate: treed Scots pine – dwarf shrubs – Sphagnum bogs dominated by Pinus sylvestris, Chamaedaphne calyculata, Rhododendron groenlandicum, Rubus chamaemorus and Sphagnum fuscum) and open graminoid-Sphagnum lawns (dominated by Scheuchzeria palustris, Carex limosa, Eriophorum russeolum, Vaccinium oxycoccos and Sphagnum balticum).

Figure 1.  

Layout of the Mukhrino field station infrastructure and position of plots (red dots) alongside the walking board (blue dashed line), the star marking the position of an automated meteorological complex; in lower right insert: layout of half-circles on each side of the boardwalk in the treed bog; in lower left insert – counting event in the graminoid-Sphagnum bog, using an arch.

About 500 specimens were collected during plot-based observations from 2014 through 2022. The specimens are accessible through the Fungarium YSU collection database at the biological portal of the Yugra State University (http://bioportal.ugrasu.ru/; https://fungariumysu.org) or through the collection's dataset in GBIF (Filippova 2023). The collections were processed as described by Lodge and Ammirati (2004). Fresh fruit-bodies were wrapped in aluminium foil and transported to the laboratory to be processed on the day of collection. The processing of specimens included photography in situ and ex situ, description of vital characters, preliminary microscopy and identification, entering new data into the database, labelling and drying at 40°C for packaging and storage in the YSU Fungarium.

For the purpose of barcoding, the collection was revised and specimens of good quality with good photographs and descriptions were selected. Each morphologically-defined taxon was represented by 2-3 specimens on average, with up to 10 specimens for some cryptic taxa.

Сommon and easily recognisable species were identified in the field. Thorough identification of collections impossible to identify with certainty in the field was done in the laboratory. Samples from dry specimens were rehydrated in tap water or potassium hydroxide (KOH) (10%), dyes and other chemicals (Congo Red, Melzer's Reagent, ammonia) were applied when necessary. A Zeiss Axiostar microscope with Achromat 5/0.12, 10/0.25, 40/0.65 (dry) and 100/1.25 (oil immersion) objectives were used for microscopical examination. Most of the specimens were identified using Funga Nordica keys (Knudsen and Vesterholt 2018); a number of monographs on particular taxa were used when necessary.

We relied primarily on the ITS region using the ITS1-F and ITS4 primers (Gardes and Bruns 1993); for 10 specimens, TEF1 region using the primers EF1-983F and EF1-1567R (Rehner and Buckley 2005) and for three specimens, LSU using the primers LR0R, LR5 (Vilgalys and Hester 1990) were amplified. DNA extraction was done from dry exsiccata or freshly-collected material using Diatom DNA Prep 200 or TransDirect® Plant Tissue PCR Kit following the standard protocols of these kits. The PCR was made using ready mix for PCR 5X ScreenMix (Evrogen). PCR and sequence reaction products were purified using Cleanup Standard (Evrogen), CleanMag DNA (Evrogen) and Dynabeads™ Sequencing Clean-Up kits. Sequencing was performed with BrilliantDye™ Terminator (v.3.1) Cycle Sequencing Kit (NimaGen) using Applied Biosystems® Sanger Sequencing 3500 Series Genetic Analyzer. Sequences were processed with Sequencing Analyst 7 and MEGA 11 software. The resulting sequences were uploaded to GenBank, PlutoF and BOLD (Filippova et al. 2023c) and as a DNA-derived extension dataset to GBIF (Filippova et al. 2023b). The closest or matching taxa were chosen by BLAST at NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and massBLASTer SH matching (BLAST+ 2.13.0) at PlutoF (https://plutof.ut.ee/en).

The following approach was used to search for the correct taxon name using sequence alignment. The NCBI BLAST search was performed to find the nearest sequences from a type specimen or an authentic specimen with a percentage identity conventionally accepted (for example, 99% for Cortinarius, Liimatainen et al. (2020)). In case no type or authentic specimen exists in NCBI, any other reliable sequence was chosen. The sequences of each group were aligned with the nearest sequence of a type (or an authentic specimen) from this group, trimmed for maximum overlap and the ITS1 and ITS2 regions with the number of nucleotides conventional for this group (including 5.8S) were left. The chosen taxon name with final species assignment, percentage identity and nearest sequences in GenBank are presented in Suppl. material 1. Most of the sequences (107) had 100% similarity and an additional 33 sequences had 99% percentage similarity. However, 15 sequences had much lower threshold (98% and less) and identifications were left at genus level. These specimens should be studied in detail by taxon-specific phylogenetic analyses.

The dataset was analysed and visualised using a collection of packages (Wickham et al. 2019) for R ver. 4.1.2 (RCoreTeam 2021) in RStudio ver. 2021.09.1+372. The «metacoder» package was used to visualise the taxonomic coverage (Foster et al. 2017). The packages «tsibble», «tidyverse», «feasts» and «ggplot2» were used for data analysis and visualisation (Hyndman and Athanasopoulos 2021). The estimation of species diversity was carried out using the «iNEXT.3D» and «iNEXT.4steps» packages (Chao et al. 2020, Chao et al. 2021). The ordination of fungal communities was made using t-SNE ordination (Laurens and Geoffrey 2008).

The plot-counts dataset includes two related tables of sampling event format: the basic Event table and the related Occurrence table. The Event table includes 18 fields and about 25 thousand records (263 plots and 96 visits): habitat, geography, date and plot size fields. The Occurrence table includes seven fields and 8284 records of observations with fruit-body counts within each plot on a particular day. The absence of occurrence records corresponding to the record in the Event table denotes an absence of fungi within a plot on a particular day (an approach recommended in GBIF IPT user manual). The database is attached as Suppl. material 2 and can be accessed through GBIF dataset (Filippova et al. 2023a).

Since different number of surveys were carried out in different years, we used a standardisation procedure before any comparative analyses could be performed. We transformed the actual data to a matrix of 9 years per 13 decade counts, where each year had an equal number of counts from the end of May to the end of September. Empty decades (without a single count) were interpolated by calculating the mean value between adjacent counts.

Since many morphological species were re-identified into several taxa, based on molecular analysis, we used composite names for these species (e.g. Cortinarius armeniacus/kauffmanianus or adding s.l. in some cases).

The sequenced specimens dataset includes two connected tables: the occurrence table and the DNA-derived data extension table (https://rs.gbif.org/extension/gbif/1.0/dna_derived_data_2021-07-05.xml#DNA_sequence). The Occurrence table contains the descriptions (in 22 fields) of 149 sequenced specimens, including images of fresh fruit-bodies. The related DNA-derived data extension table provides 155 sequences with descriptions of their parameters in eight fields. The dataset can be accessed through GBIF (Filippova et al. 2023b).

The literature-based dataset contains roughly 5000 citations of data on peatland fungi globally. Additional information includes bibliographic citation, date, country where the species was studied, habitat and substrate. The database can be accessed through GBIF (Filippova and Rudykina 2023). To avoid a mix-up of taxonomical concepts in literature datasets, names were synonymised using the GBIF species matching tool following the GBIF Backbone taxonomy.