Global patterns and drivers of urban biotic homogenization: A meta-analysis
Data files
Mar 17, 2026 version files 290.44 KB
-
Appendix_S2._Standardization.csv
89.74 KB
-
beta_diversity_partitioning.csv
46.58 KB
-
README.md
5.43 KB
-
Script_UBH-meta_analysis.R
29.07 KB
-
UBH-Nestedness_Sorensen.csv
39.86 KB
-
UBH-Total_Sorensen.csv
39.92 KB
-
UBH-Turnover_Sorensen.csv
39.85 KB
Abstract
Aim: Urbanization is a major driver of biotic homogenization (BH), reducing biodiversity through species invasions and extinctions. However, the drivers and mechanisms remain poorly understood, and urban ecosystems may also exhibit biotic differentiation (BD). This study uses a meta-analysis to examine BH and BD patterns, underlying processes and key drivers across regions and taxa.
Location: Global.
Time Period: 2004–2023.
Major Taxa Studied: Arthropods, amphibians, birds and plants.
Methods: We conducted a global meta-analysis integrating 110 effect sizes from 45 publications to evaluate urban biotic homogenization within and between cities. We quantified β-diversity and its turnover and nestedness components across urbanization levels. Within cities, we evaluated changes in β-diversity along urban levels, whereas between cities we assessed similarity among urban assemblages. We then examined how environmental, urban, and biotic factors (e.g., climate, impervious cover, non-native richness) influenced homogenization.
Results: Across studies, β-diversity responses to urbanization were heterogeneous but revealed clear patterns. Birds showed the strongest within-city homogenization, particularly in North America, whereas amphibians, arthropods, and plants exhibited weaker or inconsistent trends. Turnover was the dominant β-diversity component (> 65%), yet declined with increasing urbanization, especially in birds and plants. Nestedness increased with urbanization, reflecting species loss in some contexts. Among potential drivers, non-native species richness consistently predicted homogenization, whereas climatic and geographic variables had context-dependent or negligible effects. Between-city analyses revealed biotic homogenization only in certain taxa and regions, whereas cities remained more similar within regions than among regions.
Main Conclusions: Urbanization influences β-diversity through context-dependent processes, with turnover decline emerging as the primary mechanism of homogenization. Non-native species richness acts as a consistent driver, whereas other factors show idiosyncratic effects. Although homogenization is evident in some taxa and regions, it is not universal. This highlights an opportunity to conserve the remaining biogeographic distinctiveness of some urban assemblages and prevent further homogenization.
Dataset DOI: 10.5061/dryad.0p2ngf2c7
Description of the data and file structure
This dataset accompanies the manuscript on urban biotic homogenization (UBH). It compiles standardized effect sizes and β-diversity partitioning results across taxa, regions, and urbanization gradients. Data were extracted from published studies (2004–2023) following systematic inclusion criteria.
The dataset consists of five .csv files:
- UBH_Total_Sorensen,
- UBH_Turnover_Sorensen,
- UBH_Nestedness_Sorensen,
- Beta_diversity_partitioning, and
- Appendix_S2_Standardization.
Additionally, we provide the R script used to perform all analyses reported in the original publication, allowing full reproducibility of the results.
Files and variables
File: UBH-Total_Sorensen.csv
Description:
Content: Effect sizes and metadata for total Sørensen dissimilarity (β_sor).
Variables (columns):
- PID: Unique identifier for each publication.
- Publication: Title of the publication.
- Year_pub: Year of publication.
- Study_year: Year in which the study was conducted.
- Country: Country where the study was conducted.
- City: City where the study was conducted.
- Region: Geographic region of the study.
- Biol_group: Focal biological group.
- Taxa: Taxonomic group to which the focal biological group belongs.
- Mean_control, SD_control, n_control: Mean, standard deviation, and sample size of β-diversity values for sites with lower urbanization.
- Mean_treat, SD_treat, n_treat: Mean, standard deviation, and sample size of β-diversity values for sites with higher urbanization.
- n_total: Total sample size from all pairwise β-diversity comparisons.
- Contrast: Urbanization levels compared (e.g., Rural–Urban, Rural–Suburban, Suburban–Urban).
- d: Effect size (Hedges’ d).
- r: Effect size (Pearson’s correlation).
- Zr: Fisher’s z-transformed correlation coefficient.
- VarZ: Variance of Zr.
- delta_imper: Change in impervious surface cover between sites.
- delta_alien: Difference in non-native species richness between sites.
- delta_s: Difference in total species richness between sites.
- delta_size: Difference in the size of the study sites.
- avg_dist: Average geographic distance between sites.
- city: Area of the city (km²).
- pop: Human population size of the city.
- temp: Mean annual temperature (°C).
- prec: Mean annual precipitation (mm).
- lat: Latitude of the study area.
- Max_alien: Maximum number of non-native species recorded across urbanization contrasts.
- Country_alien: Number of non-native species reported for the country where the study was conducted.
- Where it was not possible to retrieve a value from the articles included in the meta-analysis, we left the spaces as n/a.
File: UBH-Turnover_Sorensen.csv
Description:
Content: Effect sizes and metadata for turnover Sørensen dissimilarity (β_sor).
Content: Effect sizes and metadata for turnover (β_sim).
Variables: Identical structure to SOR, but all values correspond to the turnover component of β-diversity.
File: UBH-Nestedness_Sorensen.csv
Content: Effect sizes and metadata for nestedness (β_sne).
Variables: Identical structure to SOR, but values correspond to the nestedness component of β-diversity.
File: beta_diversity_partitioning.csv
Content: Partitioning of β-diversity by urbanization category.
Variables (columns): Identical structure to UBH-Total Sorensen.csv, but values correspond to Partitioning of β-diversity
File: Appendix_S2._Standardization.csv
Description: This dataset provides the standardized classification of study sites used in the meta-analysis of urban biotic homogenization (UBH). It ensures consistency in the categorization of urbanization levels across studies, based on impervious surface cover, geographic coordinates, and supporting metadata. The file allows replication of the site-level standardization procedure applied prior to calculating β-diversity metrics.
Variables:
- Publication: Unique identifier of the publication (matches PID in main dataset).
- Original_site_name: Site name as reported in the original publication.
- Analysis: Type of analysis conducted (e.g., Within, Between).
- Standard_cat: Standardized urbanization category assigned to the site (e.g., Rural, Suburban, Urban, Natural).
- City: City where the study site is located.
- imp_cover_data: Impervious surface percentage estimated for the site.
- Scale: Spatial buffer (in meters) used to calculate impervious surface.
- Method: Source or method used to estimate impervious cover (e.g., Google Earth, remote sensing).
- Latitude: Latitude of the study site (decimal degrees, WGS84).
- Longitude: Longitude of the study site (decimal degrees, WGS84).
- Year_study: Year when the field study or data collection was conducted.
Script_UBH-meta_analysis.R
This R script contains all analyses associated with the manuscript on urban biotic homogenization (UBH). It is designed to reproduce the results presented in the study and relies on the input datasets Data_set_UBH.xlsx
