Data from: Landscape management can foster pollinator richness in fragmented high-value habitats
Data files
Jan 07, 2025 version files 18.74 KB
-
Biegerl_data_Rcode_ProceedingsB.zip
14.32 KB
-
README.md
4.41 KB
Abstract
Pollinator diversity is declining due to habitat loss, low habitat quality, limited habitat connectivity, and intensification of agriculture in remaining high-value habitats within human-dominated landscapes, such as calcareous grasslands. Options to increase the local area of protected habitats are often limited. Therefore, we asked how local habitat quality as well as agri-environmental schemes (AES) and configuration of the surrounding landscape can contribute to the preservation of pollinator diversity. We sampled bees, butterflies, and hoverflies in 40 calcareous grasslands in Germany and assessed the effects of calcareous grassland area, quality and connectivity, agricultural configuration, and AES on species richness and abundance. While calcareous grassland area was an important predictor for bee and butterfly species richness, with strongest effects sizes for endangered species, local flower resources and nesting sites, and landscape characteristics such as small field size, high proportion of organic fields and connectivity with other grasslands significantly enhanced pollinator richness with responses differing among the three studied taxa. In contrast to expectations, AES flowering fields did not benefit pollinator communities in grasslands. We conclude that improving local habitat quality in combination with targeted landscape management are effective measures to promote pollinator richness in highly fragmented protected grassland.
README: Landscape management can foster pollinator richness in fragmented high-value habitats
https://doi.org/10.5061/dryad.dncjsxm90
Description of the data and file structure
Data and code for the manuscript: "Landscape management can foster pollinator richness in fragmented high-value habitats"
DOI: 10.1098/rspb.2024.2686
Files and variables
File: Biegerl_data_Rcode_ProceedingsB.zip
Description:
File 1: Biegerl_R_code_ProceedingsB.Rmd: R script with the analyses to reproduce the results in this study.
File 2: Landscape_data.csv: Dataset with the response variable (pollinators species richness and abundance) and the predictors used in this study:
Variables:
Site_ID = Calcareous grassland/Study site identifier
long = Longitude
lat = Latitude
region = Region of study site (Lower or Upper Franconia)
WB_abun = Solitary bee abundance
WB_spec = Solitary bee species richness
WB_abun_endang = Endangered Solitary bee abundance (according to the Red List of Bavaria)
WB_spec_endang = Endangered Solitary bee species richness
BB_abun = Bumblbebee abundance
BB_spec = Bumblebee species richness
SF_abun = Hoverfly abundance (SF = syrphid fly)
SF_spec = Hoverfly species richness
BF_abun = Butterfly abundance
BF_spec = Butterfly species richness
BF_abun_endang = Endangered Butterfly abundance
BF_spec_endang = Endangered Butterfly species richness
habitat_area = Area/size of study site/calcareous grassland in ha
habitat connec = Cover of calacareous grassland in a 2 km radius around study site (excluding study site area itself) in %
annualcrop_cover = Cover of annual crop fields in a 2 km radius around study site in %
organic_farming = Cover of organic farming in relation to the total annual crop cover in % (2 km radius)
flowering_fields = Cover of flowering fields in relation to the total annual crop cover in % (2 km radius)
mean_field_size = Mean size of annual crop fields in a 2 km radius around study site in ha
flower_cover_bee = Cover of flowers on bee/hoverfly transect in %
flower_cover_butter = Cover of flowers on butterfly transect in %
nesting_sites = Cover of potential nesting sites for solitary bees on bee/hoverfly transect in %
flower_diversity_bee = Richness/Number of flowering plant species on the bee/hoverfly transect
flower_div_butter = Richness/Number of flowering plant species on the butterfly transect
annual_mean_temperature = Mean annual temperature from 1970 to 2010 in Celsius
Missing values = NA
File 3: DAG_data.csv: Dataset with the response variable (pollinators species richness and abundance) and the predictors used in this study:
Variables:
Region = Region of study site (Lower (0) or Upper (1) Franconia)
Wild pollinator richness/abundance = Solitary bee abundance (example for pollinator species richness and abundance)
Calcareous grassland area = Area/size of study site/calcareous grassland in ha
Habitat connectivity = Cover of calcareous grassland in a 2 km radius around study site
Annual crop cover = Cover of annual crop fields in a 2 km radius around study site in %
Organic farming = Cover of organic farming in relation to the total annual crop cover in % (2 km radius)
Flowering fields = Cover of flowering fields in relation to the total annual crop cover in % (2 km radius)
Mean field size = Mean size of annual crop fields in a 2 km radius around study site in ha
Nesting resources = Cover of potential nesting sites for solitary bees on bee/hoverfly transect in %
Flower resources = Richness/Number of flowering plant species (example for Flower resources)
MAT = Mean annual temperature from 1970 to 2010 in Celsius
Missing values = NA
Code/software
R version and list of R packages used for the analyses and plots:
We used R version 4.3.1 and RStudio version 2023.06.2
marginaleffects package v 0.20.1 (Arel-Bundock et al. 2024)
modelsummary package v 2.1.0 (Arel-Bundock et al. 2024)
DHARMa package v 0.4.6 (Hartig & Lohse 2022)
performance package v 0.11.0 (Lüdecke et al. 2024)
MASS package (Venables & Ripley 2002)
tidyverse package v 2.0.0 (Wickham et al. 2023)
ggeffects package v 1.6.0 (Lüdecke et al. 2024)
ggplot2 package v 3.5.1 (Wickham et al. 2024)
DAGitty package v 0.3-4 (Textor et al. 2023)
car package v 3.1-3 (Fox et al. 2024)
ncf package v 1.3-2 (Bjornstadt & Cai 2022)