Data from: Responses of insect pollinators to habitat fragmentation: A global meta-analysis
Data files
Aug 26, 2025 version files 63.43 KB
-
Abundance_20250630.csv
24.56 KB
-
README.md
2.34 KB
-
Richness_20250627.csv
36.54 KB
Abstract
Insect pollinators are declining globally, with habitat fragmentation recognized as an important driver. While these declines have garnered widespread public and policy attention, current evidence remains largely limited to certain taxa, geographic regions, and ecosystems, thereby resulting in diverse outcomes.
Here, we aimed to present a global meta-analysis using a dataset comprising 80 studies across 28 countries on all continents, excluding Antarctica. We used Pearson’s correlation coefficient (r) as an effect size to measure the association between habitat fragmentation and the abundance and species richness of wild bees, butterflies, and wasps. Subsequently, we performed categorical meta-analyses that fragmentation indices, climates, and taxonomic groups as moderators to investigate the dependency of habitat fragmentation effects on these factors.
Increasing habitat fragmentation was significantly associated with reductions in both the abundance (r = -0.26, CI: -0.39, -0.12) and species richness (r = -0.46, CI: -0.55, -0.35) of insect pollinators. The magnitude of these associations varied depending on the evaluated fragmentation index. Specifically, pollinator abundance (r = -0.43, 95%CI: -0.56, -0.29) and species richness (r = -0.62, 95%CI: -0.71, -0.52) significantly decreased with reduced habitat area.
Synthesis and Applications. Our results show that insect pollinators have declined globally due to habitat fragmentation. Reduced habitat area and increased edge density were the main drivers of pollinator decline. The findings indicate that preserving large habitats can support more insect pollinators. This implies that policies must target the landscape scale to mitigate the negative effects of habitat fragmentation.
Dataset DOI: 10.5061/dryad.dz08kps9p
Description of the data and file structure
We aimed to present a global meta-analysis using a dataset comprising 80 studies across 28 countries on all continents, excluding Antarctica. We used Pearson’s correlation coefficient (r) as an effect size to measure the association between habitat fragmentation and the abundance and species richness of wild bees, butterflies, and wasps.
Files and variables
File: Richness_20250627.csv
Description: This dataset includes 109 data observations from 63 publications.
Variables
- Publication: Publication identity;
- Reference: Authors and publication year;
- Title: Title of publication;
- Site: The locations/areas where experiment was carried out;
- Sampling_method: the method used for sampling pollinators;
- Latitude: Latitude of study site;
- Longitude: longitude of study site;
- Exp_year: The year when study was carried out;
- Study: A unique combination of publication and the year of study conducted;
- Country: Country where study was carried out;
- Continent: Continent where study was carried out
- Climate: According to Köppen Geiger Climate Classification;
- Dominant_matrix: The main matrix in study area;
- Sampling_matrix: The habitat type where insect pollinators sample from;
- Reported_fragment_index: The fragmentation index evaluated in the publication (description in original paper);
- Fragment_index: The fragmentation index evaluated in the publication;
- Reported_pollinators: Taxa of pollinating insects (description in original paper);
- Taxonomic_group: Taxa of pollinating insects;
- Richness_level: The level that study reported the abundance data, i.e. species level, functional level;
- Sample_size: Sample size;
- r_effect_size: Pearson’s r (effect size).
File: Abundance_20250630.csv
Description: This dataset includes 78 data observations from 40 publications.
Variables: The variables and their meanings are the same as those in richness dataset.
The empty cells in both datasets indicating information not available in original study.
