Socio-economic status and non-native species drive bird ecosystem service provision in urban areas
Data files
Jul 04, 2025 version files 632.28 KB
-
calculating_CFD.R
56.34 KB
-
CFD_data_1percent.csv
110.64 KB
-
CFD_data_difference_1percent.csv
19.25 KB
-
CFD_data_difference.csv
19.38 KB
-
CFD_data.csv
110.85 KB
-
CFD_data.xlsx
201.80 KB
-
Data_species_traits.xlsx
70.60 KB
-
median_treePortugal.tree
8.03 KB
-
median_treeSpain.tree
12.13 KB
-
Models.R
3.70 KB
-
README.md
19.57 KB
Abstract
Areas of higher socio-economic status within urban areas generally have higher levels of taxonomic diversity than poorer areas, representing a form of environmental injustice. This inequality may result in lower-income areas experiencing both lower cultural (e.g., non-material services provided by bird aesthetics) and regulating (e.g., pest control by insectivorous birds) ecosystem service provision. Furthermore, urban areas are hotspots for the introduction of non-native species, which can potentially alter the functional space of invaded communities, and hence their ecosystem service provision. However, the relationship between socio-economic status and the ecosystem services that both native and non-native urban biodiversity provide has yet to be fully addressed. We tested whether functional diversity associated with avian cultural and regulating ecosystem services varied along the socio-economic gradient of functional urban areas (FUAs) in the Iberian Peninsula. We used bird atlases from Spain and Portugal to characterise the bird communities of all FUAs across each country, calculating both species richness and functional dispersion of traits related to cultural and regulating ecosystem services. We then applied generalized linear mixed models to understand the relationships between these diversity metrics and the socio-economic status of each community, measured as annual median household income. Furthermore, we assessed whether the presence or absence of non-native species moderated community responses along the socio-economic gradient. Both cultural and regulating ecosystem services were negatively affected by socio-economic status, while species richness followed the opposite trend. However, invaded communities showed higher species richness and cultural ecosystem services than non-invaded communities. Our results suggest a negative relationship between socio-economic status and ecosystem service provision, with non-native species playing a major role in shaping this relationship. Our counterintuitive findings highlight the need for targeted management strategies that address both socio-economic and ecological inequalities in urban planning, in addition to non-native species management.
Authors: Marcolin, F.; Mammola, S.; Alba, R.; Segurado, P.; Reino, L.; Chamberlain, D.
DOI: 10.5061/dryad.vx0k6dk3z
Description of the data and file structure
This repository contains data and scripts used in the analyses of functional diversity and corrected functional dispersion (CFD) of birds in Spanish and Portuguese urban areas. Traits list, community matrices, and income data are included, along with the R scripts used for CFD calculation and statistical modeling.
Files and variables
File
Data_species_traits.xlsx: This file include only traits from Storchová & Hořák (2018), Spreadsheets Spain traits and Spreadsheets Portugal traits, which are openly available under a CC0 license (https://doi.org/10.5061/dryad.n6k3n). Traits from other sources (i.e. AVONET, EltonTraits, and Santangeli et al., 2023) are not included due to licensing restrictions (CC-BY) and must be retrieved independently from their respective repositories. The full list of traits used and their sources is provided in the file (Metadata spreadsheet).
contain species-by-trait matrices used for calculating standardized effect sizes of functional dispersion (CFD) in the script calculating_CFD.R.
- Tail.Length: Tail length (distance between the tip of longest rectrix and the point at which the two central rectrices protrude) relative to size (body mass). Continuous variable (mm) used only as cultural trait.
- n.loci: Number of cells occupied by colours retrieved from a grid on bird plumage (see Santangeli et al., 2023 for details) used only as cultural trait.
- elaboration: Average distance between all colours found in a species and the global average colour across all species (see Santangeli et al., 2023 for details) used only as cultural trait.
- crest: Presence of a crest with four discrete categories (see Santangeli et al., 2023 for details) used only as cultural trait. Four discrete categories: 0 = no crest, 1 = crest follows shape of head, 2 = short crest, and 3 = long crest
- Inverse_Mass: Inverse body mass in grams based on body mass measured in grams (Tobias et al., 2022). Inverse of body mass was used to avoid correlation with relative tail length and thus retain it in the analyses. Used only as cultural trait.
- Mass: Body mass Body mass given as species average (incorporating both male and female body mass) in grams. Used only as regulating trait.
- AssociationS: Association of adults during nesting. S = solitary. Used only as regulating trait. Binary
- AssociationSC: Association of adults during nesting period SC = semicolonial
- AssociationC: Association of adults during nesting period. C = colonial. Used only as regulating trait. Binary variable.
- NestCA: Nest type: CA = closed-arboreal. Used only as regulating trait. Binary variable.
- NestG: Nest type: G = ground. Used only as regulating trait. Binary variable.
- NestGC: Nest type: GC = ground close. Used only as regulating trait. Binary variable.
- NestH: Nest type: H = hole (in tree, bank, ground, crevice). Used only as regulating trait. Binary variable.
- NestOA: Nest type: OA = open-arboreal (cup in bush, tree, or cliff ledge). Used only as regulating trait. Binary variable.
- Folivore_B: At least 10% of diet during breeding season composed of grass, leaves, small plants. Used only as regulating trait. Binary variable.
- Frugivore_B: At least 10% of diet during breeding season composed of fruits. Used only as regulating trait. Binary variable.
- Granivore_B: At least 10% of diet during breeding season composed of grains, seeds, and nuts. Used only as regulating trait. Binary variable.
- Invertebrates_B: At least 10% of diet during breeding season composed of invertebrates. Used only as regulating trait. Binary variable.
- Vertebrates_B: At least 10% of diet during breeding season composed of vertebrates (excluding carrion). Used only as regulating trait. Binary variable.
- Carrion_B: At least 10% of diet during breeding season composed of carrion. Used only as regulating trait. Binary variable.
- Vertebrates_Y: At least 10% of diet throughout the year composed of vertebrates (excluding carrion). Used only as regulating trait. Binary variable.
- ForStrat-wataroundsurf: Prevalence of foraging on or just (<5 inches) below water surface. Used only as regulating trait. Binary variable.
- ForStrat-ground: Prevalence of foraging on ground. Used only as regulating trait. Binary variable.
- ForStrat-understory: Prevalence of foraging below 2m in understory in forest, forest edges, bushes, or shrubs. Used only as regulating trait. Binary variable.
- ForStrat-midhigh: Prevalence of foraging in mid to high levels in trees or high bushes (2m upward), but below canopy. Used only as regulating trait. Binary variable.
- ForStrat-canopy: Prevalence of foraging in or just above the tree canopy. Used only as regulating trait. Binary variable.
- ForStrat-aerial: Prevalence of foraging well above vegetation or any structures. Used only as regulating trait. Binary variable.
- Species names follow national breeding bird atlases.
BirdTree_namematches names at http://birdtree.org/ (Jetz et al., 2012).
CFD_data.xlsx: Model variables for the analyses (both predictor and response variables) with metadata (Metadata spreadsheet), including all species (CFD_data spreadsheet) and including in the analyses only those species that were observed in at least one percent of atlas grid cells (CFD_data_1percent spreadsheet).
- CODE: Unique spatial unit identifier for the 10×10 km grid cell. UTM for Spain; ETRS89 for Portugal.
- Country: Country where the grid cell is located. Values: Spain or Portugal.
- fuaname: Name of the Functional Urban Area (FUA) associated with the grid cell.
- X: X-coordinate in UTM projection (EPSG: 25830). For Portugal, coordinates were transformed to match this projection.
- Y: Y-coordinate in UTM projection (EPSG: 25830). For Portugal, coordinates were transformed to match this projection.
- Scaled Median Income: Median household income of the grid cell, scaled (z-transformed) across all sampled communities.
- Species richness: Total number of bird species recorded in the community. Includes or excludes non-native species depending on community type.
- CFDCultural: Corrected Functional Dispersion (CFD) for cultural ecosystem services. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- CFDRegulating: Corrected Functional Dispersion (CFD) for regulating ecosystem services. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- Community type: Type of bird community considered:
invaded= at least one non-native species present,non-invaded= no non-native species presentinvaded_noAlien= only native species retained in an invaded community (aliens excluded)
CFD_data.csv: Model variables for the analyses (both predictor and response variables), including all species.
- CODE: Unique spatial unit identifier for the 10×10 km grid cell. UTM for Spain; ETRS89 for Portugal.
- Country: Country where the grid cell is located. Values: Spain or Portugal.
- fuaname: Name of the Functional Urban Area (FUA) associated with the grid cell.
- X: X-coordinate in UTM projection (EPSG: 25830). For Portugal, coordinates were transformed to match this projection.
- Y: Y-coordinate in UTM projection (EPSG: 25830). For Portugal, coordinates were transformed to match this projection.
- Scaled Median Income: Median household income of the grid cell, scaled (z-transformed) across all sampled communities.
- Species richness: Total number of bird species recorded in the community. Includes or excludes non-native species depending on community type.
- CFDCultural: Corrected Functional Dispersion (CFD) for cultural ecosystem services. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- CFDRegulating: Corrected Functional Dispersion (CFD) for regulating ecosystem services. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- Community type: Type of bird community considered:
invaded= at least one non-native species present,non-invaded= no non-native species presentinvaded_noAlien= only native species retained in an invaded community (aliens excluded)
CFD_data_1percent.csv: Model variables for the analyses (both predictor and response variables), including in the analyses only those species that were observed in at least one percent of atlas grid cells.
- CODE: Unique spatial unit identifier for the 10×10 km grid cell. UTM for Spain; ETRS89 for Portugal.
- Country: Country where the grid cell is located. Values: Spain or Portugal.
- fuaname: Name of the Functional Urban Area (FUA) associated with the grid cell.
- X: X-coordinate in UTM projection (EPSG: 25830). For Portugal, coordinates were transformed to match this projection.
- Y: Y-coordinate in UTM projection (EPSG: 25830). For Portugal, coordinates were transformed to match this projection.
- Scaled Median Income: Median household income of the grid cell, scaled (z-transformed) across all sampled communities.
- Species richness: Total number of bird species recorded in the community. Includes or excludes non-native species depending on community type.
- CFDCultural: Corrected Functional Dispersion (CFD) for cultural ecosystem services. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- CFDRegulating: Corrected Functional Dispersion (CFD) for regulating ecosystem services. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- Community type: Type of bird community considered:
invaded= at least one non-native species present,non-invaded= no non-native species presentinvaded_noAlien= only native species retained in an invaded community (aliens excluded).
CFD_data_difference.csv: Model variables for the analyses (both predictor and response variables), including all species.
- CODE: Unique spatial unit identifier for the 10×10 km grid cell. UTM for Spain; ETRS89 for Portugal.
- Country: Country where the grid cell is located. Values: Spain or Portugal.
- fuaname: Name of the Functional Urban Area (FUA) associated with the grid cell.
- diff_SpRich: Species richness difference values between invaded and invaded-no-alien communities.
- diff_CFDCultural: Corrected Functional Dispersion (CFD) for cultural ecosystem services difference between invaded and invaded-no-alien communities. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- diff_CFDRegulating: Corrected Functional Dispersion (CFD) for regulating ecosystem services difference between invaded and invaded-no-alien communities.. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
CFD_data_difference_1percent.csv: Model variables for the analyses (both predictor and response variables), including in the analyses only those species that were observed in at least one percent of atlas grid cells.
- CODE: Unique spatial unit identifier for the 10×10 km grid cell. UTM for Spain; ETRS89 for Portugal.
- Country: Country where the grid cell is located. Values: Spain or Portugal.
- fuaname: Name of the Functional Urban Area (FUA) associated with the grid cell.
- diff_SpRich: Species richness difference values between invaded and invaded-no-alien communities.
- diff_CFDCultural: Corrected Functional Dispersion (CFD) for cultural ecosystem services difference between invaded and invaded-no-alien communities. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
- diff_CFDRegulating: Corrected Functional Dispersion (CFD) for regulating ecosystem services difference between invaded and invaded-no-alien communities.. Calculated as the standardised effect size (SES) from residuals of a linear regression between functional and phylogenetic dispersion, corrected for phylogeny and species richness.
Accessing Median Household Income Data
Spain: Median income data can be downloaded from the Instituto Nacional de Estadística (INE) website: https://www.ine.es/dynt3/inebase/index.htm?padre=12385&capsel=8752. Select “Resultados por municipios, distritos y secciones censales”, then “Total Nacional” and finally “Indicadores de renta media y mediana”
Portugal: Median income data is available from the Instituto Nacional de Estatística (INE): https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_indicadores&indOcorrCod=0012674&contexto=bd&selTab=tab2. Click “Alterar condições de seleção” then select all geographic units from the cascading menu then click “Visualizar quadro” to display the table and finally export using icons below.
Phylogenetic trees
median_treeSpain.tree: Median phylogenetic tree (Spain) used for calculating phylogenetic diversity. Trees can be retrieved at http://birdtree.org/ (Jetz et al., 2012).
median_treePortugal.tree: Median phylogenetic tree (Portugal) used for calculating phylogenetic diversity. Trees can be retrieved at http://birdtree.org/ (Jetz et al., 2012).
Scripts
This repository includes all scripts necessary to reproduce the functional diversity (FD) and corrected functional dispersion (CFD) analyses presented in the manuscript. The scripts are fully annotated to describe each step of the workflow. We provide R code to calculate community-level functional and phylogenetic dispersions, extract regression residuals, and derive final SES values (CFD) using the BAT R package.
Due to restrictions on the use of the original bird atlas data—provided by third parties (Equipa Atlas 2022; Molina et al. 2022) the community matrices shared here have been anonymised and replaced with generic names for illustrative purposes only. The raw data on species occurrences that underpin the analyses are not publicly available but can be requested directly from the data providers. Spain (Third Spanish Breeding Bird Atlas), requests for data should be addressed to SEO/BirdLife, the environmental NGO that compiled and curated the atlas. Contact: seguimientodeaves@seo.org.
Portugal (Third Portuguese Breeding Bird Atlas), requests for data should be addressed to SPEA. The data (species occurrence or abundance) are available in .csv or shapefile format in 10x10 km resolution. Contact: spea@spea.pt. Moreover, due to licensing restrictions (CC-BY), trait data for SpeciesTraits_Spain.csv and SpeciesTraits_Portugal.csv should be retrieved indipendently from the respective datasets reported in the Data_species_traits.xlsx file. Therefore we provide the processed data (e.g. CFD_data.csv) for running the models.
calculating_CFD.R: the script is fully annotated to explain each step. It calculates Functional Dispersion (FD) and Phylogenetic Dispersion (PD), fits a linear model to relate FD to PD, extracts model residuals, and then calculates Standard Effect Sizes (SES) using 999 null models. The outputs are corrected SES residuals representing community-level Corrected Functional Dispersion (CFD) for both cultural and regulating ecosystem services. Output files are used to generate CFD_data.csv, CFD_data_1percent.csv, CFD_data_difference.csv and CFD_data_difference_1percent.csv
Models.R: this script loads the processed CFD datasets (e.g. CFD_data.csv) and runs linear mixed models (LMMs) to test the relationship between socio-economic status and ecosystem service provision. It also includes intercept-only models to compare invaded vs. invaded_noAlien communities (CFD_data_difference.csv and CFD_data_difference_1percent.csv). Multiple versions of the CFD datasets allow the user to test full vs. filtered (≥1% occurrence) communities and compare differences between community types.
Code/software
All analyses were done in R (R version 4.3.2 (2023-10-31 ucrt) — “Eye Holes”).
Data was derived from the following sources
*Equipa Atlas (2022). III Atlas das Aves Nidificantes de Portugal (2016-2021). SPEA, ICNF, LabOr/UÉ, IFCN. Portugal. https://www.listavermelhadasaves.pt/atlas/
*Jetz, W., Thomas, G.H., Joy, J.B., Hartmann, K., & Mooers, A.O. (2012). The global diversity of birds in space and time. Nature, 491(7424), 444-448.
*Molina, B., Nebreda, A., Muñoz, A. R., Seoane, J., Real, R., Bustamante, J., & Del Moral, J. C. (2022). III Atlas de las aves en época de reproducción en España. SEO/BirdLife: Madrid.
*Santangeli, A., Haukka, A., Morris, W., Arkkila, S., Delhey, K., Kempenaers, B., ... & Mammola, S. (2023). What drives our aesthetic attraction to birds?. npj Biodiversity, 2(1), 20
*Statistics Portugal(INE) (2024). Database. Retrieved from https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_indicadores&indOcorrCod=0012674&contexto=bd&selTab=tab2
*Statistics Spain(INE) (2024). Database. Retrieved from https://www.ine.es/dynt3/inebase/index.htm?padre=7132
*Storchová, L., Hořák, D. (2018). Life‐history characteristics of European birds. Global ecology and biogeography, 27(4), 400-406. (https://doi.org/10.1111/geb.12709)
*Tobias, J.A., Sheard, C., Pigot, A.L., Devenish, A.J., Yang, J., Sayol, F., ... & Schleuning, M. (2022). AVONET: morphological, ecological and geographical data for all birds. Ecology Letters, 25(3), 581-597. (https://doi.org/10.1111/ele.13898)
*Wilman, H., Belmaker, J., Simpson, J., de la Rosa, C., Rivadeneira, M.M., Jetz, W. (2014). EltonTraits 1.0: Species‐level foraging attributes of the world's birds and mammals: Ecological Archives E095‐178. Ecology, 95(7), 2027-2027. (https://doi.org/10.1890/13-1917.1)