You are viewing the Initial Version, the most recent version of this dataset.
1 version(s) available
Date of publication: October 23, 2024

Version 1

Date of publication: October 23, 2024

Type of change:

Description:

preview image

Species habitat suitability of European terrestrial vertebrates for contemporary climate and land use

by Sara Si-Moussi

Predicted habitat suitability of most terrestrial vertebrate species (1,207 species) in Europe at 1km resolution under current conditions (1990-2020). Predictions for each species were obtained from an ensemble of species distribution models (ensemble SDMs). The provided data are the raw weighted mean of habitat suitability through the ensemble SDMs, the rescaled weighted mean of habitat suitability through the ensemble SDMs but spatially constra ...(continue reading)

DOI 10.25829/wpfn43
Citation
Si-moussi, S., Thuiller, W. (2024). Species habitat suitability of European terrestrial vertebrates for contemporary climate and land use (Version 1) [Dataset]. German Centre for Integrative Biodiversity Research. https://doi.org/10.25829/wpfn43

Terrestrial VertebratesEurope UnionSpecies DistributionsPredicted Distributions

84
The title of the dataset. Species habitat suitability of European terrestrial vertebrates for contemporary climate and land use
The date on which this version of the data was created in YYYY-MM-DD format.
A paragraph describing the dataset.
Hover to see a suggestion for a good description. Allowed:
Predicted habitat suitability of most terrestrial vertebrate species (1,207 species) in Europe at 1km resolution under current conditions (1990-2020). Predictions for each species were obtained from an ensemble of species distribution models (ensemble SDMs). The provided data are the raw weighted mean of habitat suitability through the ensemble SDMs, the rescaled weighted mean of habitat suitability through the ensemble SDMs but spatially constrained by expert range maps, the committee averaging score (% of predicted presence across models), and the coefficient of variation of predicted habitat suitability across models.
Provide the DOI number of associated publications. Click Plus to add DOIs.

N/A

The method of production of the original data.
Hover to see a suggestion for a minimum description. Each vertebrate species was modeled using an ensemble of machine learning algorithms. These algorithms were fitted on curated presence-only data from GBIF with various sets of pseudo-absences, in function of selected environmental data including climate (CHELSA), soil (Soilgrids), terrain (EU-DEM), hydrography (EU-Hydro) and land use (Dou et al 2021) variables. In total, we used 3 algorithms (Random Forest, XGBoost, Neural Networks), 5 pseudo-absence repetitions and 5 spatial cross-validation partitions (i.e., 75 models). We kept only the models that reached a TSS > 0.4 on the cross-validation set. From this, we provide different ensemble predictions.

1- The weighted mean habitat suitability, i.e., the mean probability of occurrence across the models weighted by their TSS predictive accuracy. This map then shows the potential suitable areas across the whole extent.

2- Coefficient of variation of suitability scores. This map shows where predictions tend to diverge or converge.

3- The mean probability of occurrence across the models weighted by their TSS predictive accuracy but rescaled with a spatial constraint layer derived from a probabilistic combination of IUCN / Bird Life expert range maps or known occurrence records. This map is then close to the ‘realized’ distribution of the species since it is spatially constrained by expert range maps.

4- Committee averaging presence, which represents the percentage of time the species is predicted to be present across the 75 predictions (after having binarized the probability of presence into presence-absence using a threshold maximising the TSS). This map shows the agreement and disagreement of the predictions across the algorithms-pseudoabsence sets-cross-validation sets. Values close to 0 means that all predictions converge to predict the absence of the species, values close to 1 means the way around.

The name of the Project.
The URL from the project website. https://naturaconnect.eu/
The name of the person or other creator type principally responsible for creating this data.
The email of the person or other creator type principally responsible for creating this data. sara.si-moussi@univ-grenoble-alpes.fr
sara.si-moussi@univ-grenoble-alpes.fr
The names of the co-creators responsible for creating this data. Click Plus to add co-creators.
Select between Creative Commons (CC) or Non-CC license.
Please select the CC license from the list. We recommend the use of CC BY 4.0

Essential Biodiversity Variables

Select the EBV class and the EBV name for the dataset. For cross-cutting use the comment at the bottom of the page for further information.
Genetic composition
Intraspecific genetic diversity
Genetic differentiation
Effective population size
Inbreeding
Other
Species populations
Species distributions
Species abundances
Other
Species traits
Morphology
Physiology
Phenology
Movement
Other
Community composition
Community abundance
Taxonomic and phylogenetic diversity
Trait diversity
Interaction diversity
Other
Ecosystem functioning
Primary productivity
Ecosystem phenology
Ecosystem disturbances
Other
Ecosystem structure
Live cover fraction
Ecosystem distribution
Ecosystem Vertical Profile
Other
Ecosystem services
Pollination
Other
Cross-cutting

Biological entity

Select the entity type of the dataset.
Species
Communities
Ecosystems
Other
None
A description of the range of taxa or ecosystem types addressed in the dataset. E.g. "300 species of mammals”, “Forests”, etc.
The reference as a URL. N/A

Metric

Provide the name, description, units of minimum 1 metric. Click Plus to add metrics.

Scenario

Spatial domain

Global
Continental/Regional
National
Sub-national/Local
Name of the continent/region/country/area, separated by comma
meter
southWest lat: 160000, lon: 723000
northEast lat: 6615000, lon: 7700000

Temporal domain

The targeted time period between each value in the dataset.
decadal
annually
monthly
weekly
daily
Other
Irregular
Single time
Select the temporal extent of the dataset.
When the dataset represents a Single time, then use the same start and end date.
__

Environmental domain *

Terrestrial
Marine
Freshwater
Miscellaneous information about the data, not captured elsewhere. N/A