We analysed plant-pollinator networks in 48 landscapes over 4 countries (France, Germany, Poland, Switzerland), during three sampling rounds (144 visits in total), and looked at plant-pollinator network structure and how this affected the presence of three types of RNA virus in wild pollinators (DWV-A, DWV-B, and BQCV). We analyzed plant-pollinator network structure, the presence of viruses, landscape composition, and vegetation. The dataset does not contain any threatened species. All missing data represented as NA.

Description of the data and file structure

This dataset contains information on raw landscape composition, plant-pollinator interactions, vegetation composition and viral analysis of pollinators. All data is contained in the zipped folder "Raw_data_Proesmans_ea_EL_v2.zip" (Raw_data_Proesmans_ea_EL.zip contains an older, less well-annotated version of the data, lacks two scripts on nectar caclulation and has the floral data divided in 4 different files each representing one of the four sampled countries, but is otherwise identical).

Contains following datasets (each with associated .xlsx metadata file):

• Landscape_centroids.csv: File with the GPS coordinates of all site centroids. Used to account for spatial autocorrelation.
• Pollinator_masterfile.csv: File with all plant-pollinator interactions observed in all landscapes

• Landscape_composition.csv: File containing the landscape composition (EUNIS level 3) within a 1 km radius from the landscape centroids

• Flower_data.csv: File containing raw vegetation data from 0.5 x 2 m plots in each landscape, during each sampling roun

• Connectancelist.csv: File containing connectance values for each plant-pollinator network

• Niche_overlap.csv: File containing for each individual network the niche overlap (Horn-Morisita) of all pollinator species with honeybee (Apis mellifera). If honeybees are absent from networks, Niche overlap is given as NA.

• European_networkmetrics.csv: File containing metrics at the level of all plant-pollinator networks. This includes network structure (web asymmetry, modularity,...) but also nectar diversity and nectar abundance and information on viral prevalence in honeybees.

• Viral_masterfile.csv: File containing information on the presence and concentration of the analyzed viruses in honeybees and wild pollinators.

Furthermore, following R scripts were used in the analysis:
(each of them is annotated, so no further dedicated metadata).

• Nectar_per_site_script_v1.1.R: This script calculates the amount of nectar in mg of sugar per m² at the site-visit level. Not necessary to run before the overarching analysis as the European_networkmetrics.csv file already contains this information.
• Calculate_nectar_diversity_v1.1.R: This script calculates the mean plot-level Shannon nectar diversity at the site-visit level. Not necessary to run before the overarching analysis as the European_networkmetrics.csv file already contains this information
• Calculate_networkmetrics_v1.2..1R: This script calculates connectance, modularity and niche overlap with honeybees. Not necessary to run this before the other script, as the csv files already contain this data. Script is purely informative and added for transparency concerning calculation of these metrics.

• Overarching_analysis_v1.8.1.R: This script runs the complete analysis, from compiling the final dataset to running models and creating figures. If all files are available, it will run just by copy-pasting the full script, although some models take several hours to run. Requires installation of Stan to run MCMC and the library rethinking, which can be a bit of a pain to install.

Sharing/Access information

This dataset is provided under a CC0 1.0 Universal (CC0 1.0) Public Domain Dedication

Code/Software

The data was analysed in R. See above for a further explanation on the different scripts used. The scripts are extensively annotated.