Patterns and drivers of range filling of alien mammals in Europe
Data files
Jun 04, 2025 version files 1.55 MB
-
README.md
22.91 KB
-
Supplementary_Data_2025_05_06.xlsx
1.53 MB
Abstract
Biological invasions are major drivers of biodiversity change. Alien mammals are particularly concerning in Europe, where their expansion remains unabated, though the driving factors are still unclear. Well-documented introductions and distributions in this continent provide a unique opportunity to understand how human activities influenced this expansion. We modelled the potential alien ranges of 46 established alien mammals in Europe using species’ introduction localities, residence time, dispersal ability, generation length, and climatic suitability. We compared potential and observed ranges through three range indices: range filling (portion of potential distribution occupied), overfilling (portion of observed distribution unexpectedly occupied), and unfilling (portion of potential distribution currently unoccupied), and we investigated the effects of native range size, introduction pathways (species’ sum of the known pathways of introduction across the European alien range, spanning 1492–2020), and socio-economic variables (spanning 1980–2017) on uncovered patterns.
We show that the median range overfilling value was high (22%), suggesting that alien mammals are substantially spreading outside expected distribution areas. Conversely, median values of range filling (14%) and unfilling (17%) were lower, suggesting recorded introductions inadequately explain alien mammals’ distributions. Range patterns were strongly shaped by human population density, which positively influenced all three range indices, driving range patterns and influencing alien mammals’ introduction and establishment. Contrary, roads and railways were negatively related to range overfilling and unfilling, as was the number of introduction pathways to range filling and overfilling.
Ultimately, the role of these socio-economic factors depends on human behaviour rather than environmental characteristics or species’ ecology. We confirm human agency as an important driver of alien mammals’ distribution and spread in Europe, highlighting that modifying human attitudes and regulations towards these taxa is key to limiting further spread.
Dataset DOI: 10.5061/dryad.wstqjq2zh
Description of the data and file structure
Supplementary Data - Patterns and drivers of range filling of alien mammals in Europe
Files and variables
File: Supplementary_Data_2025_05_06.xlsx
Description:
This supplementary dataset is in Excel format (.xlsx) which can be explored and manipulated with programs like Microsoft Office, OpenOffice Calc, LibreOffice Calc, or imported into Google Sheets. The following text describes the information contained in its twenty-two sheets.
Legend: contains names and affiliations of all authors, and a description of the supplementary data.
Species_variables: Species’ traits used in the calculation of the potential alien ranges, species-specific variables used to explain the different range patterns in the GLMMs, their rationale, and the sources, for each study species (n = 46). The sheet consists of:
- 48 columns: 2 columns for the variables, plus 46 columns for the species.
- 22 rows: 2 rows for the taxonomic order and species' name, plus 20 rows for the variables.
Variables description:
- Order: the taxonomic order of the species, following the PHYLACINE 1.2 taxonomy (Faurby et al. 2018).
- Binomial: the scientific name of the species, following the PHYLACINE 1.2 taxonomy (Faurby et al. 2018).
- Median and maximum dispersal: respectively, the median and maximum distance between species’ birth and breeding site (both in km). Median dispersal is derived from the COalesced Mammal dataBase of INtrinsic and Extrinsic traits (COMBINE; Soria et al. 2021), which follows the methodology described by Santini et al. (2013). We calculated maximum dispersal following the same approach as described by Santini et al. (2013).
- Generation length: the average age of parents of a cohort (in days). Generation length is derived from COMBINE (Soria et al. 2021).
- Age at first reproduction: the age at which females give birth to their first litter or their young attach to teats (in days). Age at first reproduction is derived from COMBINE (Soria et al. 2021).
- Residence time: the number of days since the first successful introduction in Europe (Biancolini et al. 2021, Melone et al. 2021).
- Pathways of introduction: species' known pathways of introduction across the European alien range and across the different pathways, spanning 1492–2020, from Biancolini et al. (2021). In Biancolini et al. (2021), they are described as follows:
- Biological Control: the release of alien mammals to serve as control agents for undesirable species and pests.
- Conservation: introduction carried out to support species recovery, including Assisted Colonization programs following the IUCN Guidelines for Conservation Translocations (IUCN 2013).
- Farming: introduction of alien mammals used as working animals or as a food source in captivity or left under limited control.
- Fauna Improvement: introduction of alien mammals to modify the landscape species composition and improve its aesthetics, e.g. efforts carried out by the Acclimatization Societies, species used as tourism attractions or to populate city parks.
- Fur Farming: alien mammals breed in captivity for fur production that subsequently escaped or have been released into the wild.
- Hunting: introduction of alien mammals as game species in the wild or hunting enclosures.
- Ornamental Purposes: alien mammals kept in captivity as ornamental species that subsequently escaped or have been released into the wild.
- Pet: alien mammals traded and kept in captivity as domestic pet animals that subsequently escaped or have been released into the wild.
- Research: alien mammals used in medical or ecological research kept in captivity or released into the wild.
- Stowaway: unintentional introduction of alien mammals present in cargo generally transported by ship.
- Wild Fur: purposeful introduction of alien mammals to create a wild population of fur- bearers.
- Zoo: alien mammals used in zoological gardens as public display that escaped or have been released into the wild.
- TOTAL: total number of introduction pathways for each species is also indicated.
- Native range extent: from IUCN (2023) (number of 10 x 10 km cells).
- Points of introduction: sum of the species' known (successful and unsuccessful introductions at the administrative area level) are compiled by Blackburn et al. (2017), with additional literature from Biancolini et al. (2021), and Melone et al. (2021).
GLMM_variables: Variables used in the Generalised Linear Mixed Models (GLMMs). Expected influence on each range pattern, rationale, time span, measurement unit, original variable’s resolution, processing methods, processed data resolution, statistical processing, and source are indicated. Variables available for several individual years were averaged. Variables description:
- Number of introduction pathways: the species’ sum of the known introduction pathway across the European alien range, spanning 1492–2020, from Biancolini et al. (2021).
- *Native range size: *from IUCN (2023) (number of 10 x 10 km cells).
- Human population density: data was downloaded from the History Database of the Global Environment (HYDE) 3.1 (Klein Goldewijk et al. 2011), for the decades 1980-2000, and the year 2005. Data were in the form of raster maps, with each decade represented as inhabitants/grid cell. We log-transformed population density, since it was strongly skewed.
- Anthropic land-use: data was downloaded from the History Database of the Global Environment (HYDE) 3.1 (Klein Goldewijk et al. 2011), for the decades 1980-2000, and the year 2005. We selected only categories of anthropic environments (pasture, cropland, and urban area), excluding natural land, and produced a raster per each category with the area covered per cell in km2. We then summed all categories to obtain a single anthropic land-use variable describing the amount of anthropized environments per cell.
- Roads and railways: were selected from Williams et al. (2020) for the year 2017.
ODMAP: Overview, Data, Model, Assessment and Prediction (Zurell et al. 2020) protocol for SDMs
single_models_statistics: Single models statistics for the 4,140 SDMs models (six algorithms x three runs x five pseudo-absences sets x 46 species) of the study. The sheet consists of:
- 9 columns: Species (species name in the format Genusspecies), PA (pseudo-absence set, from 1 to 5) , run (run, from 1 to 5), algo (the six algorithms used for the ensemble modeling; acronyms as follows: Generalised Linear Models GLM, Multivariate Adaptive Regression Splines MARS, Flexible Discriminant Analysis FDA, Generalised Boosting Models GBM, Artificial Neural Network ANN, Random Forest RF), metric.eval (evaluation metric used; acronyms as follows: True Skill Statistics TSS, Cohen's kappa KAPPA, receiver operating characteristic curve ROC), Testing.data (value of the testing data), Cutoff (cutoff value), Sensitivity (sensitivity value), and Specificity (specificity value) for each model are indicated.
- 12409 rows.
OAR_Range_ove_climatic_suitab: the sheet consist of:
- 7 columns: Binomial (scientific species' name), Range_overfilling (number of 10x10 raster cells for range overfilling), Range_overfilling_suit (number of 10x10 raster cells within the climatically suitable areas projected by the SDMs for the range overfilling), Range_overfilling_suit_perc (proportion of the raster cells occupied within the climatically suitable areas projected by the SDMs for the range overfilling), OAR (number of 10x10 raster cells for the Observed Alien Range OAR), OAR_suit (number of 10x10 raster cells within the climatically suitable areas projected by the SDMs for the OAR), OAR_suit_perc (proportion of the raster cells occupied within the climatically suitable areas projected by the SDMs for the OAR).
- 47 rows.
Disp_gen_correlations: the sheet contains the correlation matrix of the variables for the three (one for each filling pattern: range filling, overfilling, and unfilling) Generalised Linear Mixed Models, for the median dispersal and generation length (Disp_gen) scenario. Variables are as follows:
- Pathways_tot: represents the species’ sum of the known pathways of introduction across the European alien range, spanning 1492–2020, from Biancolini et al. (2021).
- Native_range: represents the native range extent from IUCN (2023) (number of 10 x 10 km cells)
- PopdensityAvg_3035_average: Human population density data was downloaded from the History Database of the Global Environment (HYDE) 3.1 (Klein Goldewijk et al. 2011), for the decades 1980-2000, and the year 2005. Data were in the form of raster maps, with each decade represented as inhabitants/grid cell. We log-transformed population density, since it was strongly skewed.
- landuseHAvg_3035_average: Anthropic land-use data was downloaded from the History Database of the Global Environment (HYDE) 3.1 (Klein Goldewijk et al. 2011), for the decades 1980-2000, and the year 2005. We selected only categories of anthropic environments (pasture, cropland, and urban area), excluding natural land, and produced a raster per each category with the area covered per cell in km2. We then summed all categories to obtain a single anthropic land-use variable describing the amount of anthropized environments per cell.
- infrastructuresAvg_3035_average: Roads and railways were selected from Williams et al. (2020) for the year 2017.
NB: The same variables are repeated in the sheets Disp_age_correlations, MaxDisp_gen_correlations, and MaxDisp_age_correlations.
Disp_age_correlations: the sheet contains the correlation matrix of the variables for the three (one for each filling pattern: range filling, overfilling, and unfilling) Generalised Linear Mixed Models, for the median dispersal and age at first reproduction (Disp_age) scenario.
MaxDisp_gen_correlations: the sheet contains the correlation matrix of the variables for the three (one for each filling pattern: range filling, overfilling, and unfilling) Generalised Linear Mixed Models, for the maximum dispersal and generation length (MaxDisp_gen) scenario.
MaxDisp_age_correlations: the sheet contains the correlation matrix of the variables for the three (one for each filling pattern: range filling, overfilling, and unfilling) Generalised Linear Mixed Models, for the maximum dispersal and age at first reproduction (MaxDisp_age) scenario.
Disp_gen_Range_filling: this sheet contains, for the median dispersal and generation length scenario, the following information for range filling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_filling (number of 10x10 raster cells for range filling), Filling_percentage_range_filling (filling percentage of range filling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs: Pathways_tot (represents the species’ sum of the known pathways of introduction across the European alien range, spanning 1492–2020, from Biancolini et al. (2021), Native_range (represents the native range extent from IUCN (2023); number of 10 x 10 km cells), PopdensityAvg_3035_average (Human population density data was downloaded from the History Database of the Global Environment (HYDE) 3.1 (Klein Goldewijk et al. 2011), for the decades 1980-2000, and the year 2005. Data were in the form of raster maps, with each decade represented as inhabitants/grid cell. We log-transformed population density, since it was strongly skewed), landuseHAvg_3035_average (Anthropic land-use data was downloaded from the History Database of the Global Environment (HYDE) 3.1 (Klein Goldewijk et al. 2011), for the decades 1980-2000, and the year 2005. We selected only categories of anthropic environments (pasture, cropland, and urban area), excluding natural land, and produced a raster per each category with the area covered per cell in km2. We then summed all categories to obtain a single anthropic land-use variable describing the amount of anthropized environments per cell), infrastructuresAvg_3035_average (Roads and railways were selected from Williams et al. (2020) for the year 2017).
- 365 rows.
NB: The same polygon-level variables are repeated in the sheets Disp_gen_Range_overfilling, Disp_gen_Range_unfilling, Disp_age_Range_filling, Disp_age_Range_overfilling, Disp_age_Range_unfilling, MaxDisp_gen_Range_filling, MaxDisp_gen_Range_overfilling, MaxDisp_gen_Range_unfilling, MaxDisp_age_Range_filling, MaxDisp_age_Range_overfilling, MaxDisp_age_Range_unfilling.
Disp_gen_Range_overfilling: this sheet contains, for the median dispersal and generation length scenario, the following information for range overfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_overfilling (number of 10x10 raster cells for range overfilling), Filling_percentage_range_overfilling (filling percentage of range overfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 638 rows.
Disp_gen_Range_unfilling: this sheet contains, for the median dispersal and generation length scenario, the following information for range unfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_unfilling (number of 10x10 raster cells for range unfilling), Filling_percentage_range_unfilling (filling percentage of range unfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 298 rows.
Disp_age_Range_filling: this sheet contains, for the median dispersal and age at first reproduction scenario, the following information for range filling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_filling (number of 10x10 raster cells for range filling), Filling_percentage_range_filling (filling percentage of range filling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 486 rows.
Disp_age_Range_overfilling: this sheet contains, for the median dispersal and age at first reproduction scenario, the following information for range overfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_overfilling (number of 10x10 raster cells for range overfilling), Filling_percentage_range_overfilling (filling percentage of range overfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 1059 rows.
Disp_age_Range_unfilling: this sheet contains, for the median dispersal and age at first reproduction scenario, the following information for range unfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_unfilling (number of 10x10 raster cells for range unfilling), Filling_percentage_range_unfilling (filling percentage of range unfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 897 rows.
MaxDisp_gen_Range_filling: this sheet contains, for the maximum dispersal and generation length scenario, the following information for range filling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_filling (number of 10x10 raster cells for range filling), Filling_percentage_range_filling (filling percentage of range filling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 475 rows.
MaxDisp_gen_Range_overfilling: this sheet contains, for the maximum dispersal and generation length scenario, the following information for range overfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_overfilling (number of 10x10 raster cells for range overfilling), Filling_percentage_range_overfilling (filling percentage of range overfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 1198 rows.
MaxDisp_gen_Range_unfilling: this sheet contains, for the maximum dispersal and generation length scenario, the following information for range unfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_unfilling (number of 10x10 raster cells for range unfilling), Filling_percentage_range_unfilling (filling percentage of range unfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 1062 rows.
MaxDisp_age_Range_filling: this sheet contains, for the maximum dispersal and age at first reproduction scenario, the following information for range filling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_filling (number of 10x10 raster cells for range filling), Filling_percentage_range_filling (filling percentage of range filling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 682 rows.
MaxDisp_age_Range_overfilling: this sheet contains, for the maximum dispersal and age at first reproduction scenario, the following information for range overfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), *Observed_range *(number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_overfilling (number of 10x10 raster cells for range overfilling), Filling_percentage_range_overfilling (filling percentage of range overfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 1582 rows.
MaxDisp_age_Range_unfilling: this sheet contains, for the maximum dispersal and age at first reproduction scenario, the following information for range overfilling:
- 12 columns: Order (species' taxonomic order), Binomial (scientific species' name), Observed_range (number of 10x10 km cell for observed range), Potential_range (number of 10x10 km cell for potential range), POAR (number of 10x10 km cell for Potential and Observed Alien Range POAR), Range_unfilling (number of 10x10 raster cells for range unfilling), Filling_percentage_range_unfilling (filling percentage of range unfilling). The last 5 columns contain polygon-level values for the variables considered in the GLMMs.
- 3229 rows.
Code/software
All the scripts are openly available here: https://github.com/Lisa-Tedeschi/Range_patterns_alien_mammals/tree/main
Lisa Tedeschi1,2,3,*, Bernd Lenzner1, Anna Schertler1, Johannes Wessely4, Dino Biancolini2,5,6, César Capinha7,8, Beatrice Melone9,10, Carmen Soria11, Carlo Rondinini2+, Franz Essl1+
1 Division of BioInvasions, Global Change & Macroecology, Department of Botany and Biodiversity Research, University of Vienna, Rennweg 14, 1030 Vienna, Austria.
2 Global Mammal Assessment Programme, Department of Biology and Biotechnologies “Charles Darwin”, Sapienza University of Rome, Viale dell’Università 32, 00185 Rome, Italy.
3 Vienna Doctoral School of Ecology and Evolution, University of Vienna, Vienna, Austria.
4 Division of Biodiversity Dynamics and Conservation, Department of Botany and Biodiversity Research, University of Vienna, Rennweg 14, 1030 Vienna, Austria.
5 National Research Council of Italy - Institute for Bioeconomy (CNR-IBE), Via dei Taurini 19, Rome, Italy.
6 IUCN SSC Invasive Species Specialist Group, Rome, Italy.
7 Centre of Geographical Studies, Institute of Geography and Spatial Planning, University of Lisbon, Rua Branca Edmée Marques, Cidade Universitária, 1600-276 Lisboa, Portugal.
8 Associate Laboratory TERRA, Lisbon, Portugal.
9 European Commission – Joint Research Centre Unit D.02, Via Enrico Fermi 2749, 21027 Ispra, Italy.
10 FINCONS SPA, Via Torri Bianche 10, Pal. Betulla, 20871 Vimercate, Italy.
11 Department of Spatial Sciences, Faculty of Environmental Sciences, Czech University of Life Sciences Prague, Prague, Czech Republic.
* Corresponding author: Lisa Tedeschi (lisa.tedeschi@univie.ac.at)
Franz Essl + and Carlo Rondinini + should be considered joint senior authors.
Published in Oikos