Skip to main content
Dryad

Integrating animal tracking datasets at a continental scale for mapping Eurasian lynx habitat

Oeser, Julian1

Published Oct 05, 2023; Updated Oct 19, 2023 on Dryad. https://doi.org/10.5061/dryad.z8w9ghxhn

Cite this dataset

Oeser, Julian (2023). Integrating animal tracking datasets at a continental scale for mapping Eurasian lynx habitat [Dataset]. Dryad. https://doi.org/10.5061/dryad.z8w9ghxhn

Abstract

Aim:
 
The increasing availability of animal tracking datasets collected across many sites provides new opportunities to move beyond local assessments to enable detailed and consistent habitat mapping at biogeographic scales. However, integrating wildlife datasets across large areas and study sites is challenging, as species’ varying responses to different environmental contexts must be reconciled. Here, we compare approaches for large-area habitat mapping and assess available habitat for a recolonizing large carnivore, the Eurasian lynx (Lynx lynx).
 
Location: Europe
 
Methods:
 
We use a continental-scale animal tracking database (450 individuals from 14 study sites) to systematically assess modeling approaches, comparing (1) global strategies that pool all data for training vs. building local, site-specific models and combining them, (2) different approaches for incorporating regional variation in habitat selection, and (3) different modeling algorithms, testing nonlinear mixed effects models as well as machine-learning algorithms.
 
Results:
 
Both global and local modeling strategies allowed building transferable habitat models with overall similar predictive performance. Model performance was the highest using flexible machine-learning algorithms and when incorporating variation in habitat selection as a function of environmental variation. Our best-performing model used a weighted combination of local, site-specific habitat models. Our habitat maps identified large areas of suitable, but currently unoccupied lynx habitat, with many of the most suitable unoccupied areas located in regions that could foster connectivity between currently isolated populations.
 
Main conclusions:
 
We demonstrate that global and local modeling strategies can achieve robust habitat models at the continental scale and that considering regional variation in habitat selection improves broad-scale habitat mapping. More generally, we highlight the promise of large wildlife tracking databases for large-area habitat mapping. Our maps provide the first high-resolution, yet continental assessment of lynx habitat across Europe, providing a consistent basis for conservation planning for restoring the species within its former range.

README: Integrating animal tracking datasets at a continental scale for mapping Eurasian lynx habitat

This dataset is a product of the Conservation Biogeography Lab at Humbolt-Universität zu Berlin (hu.berlin/biogeography) and is a result of the following publication:

Oeser, J., Heurich, M., Kramer-Schadt, S., Mattisson, J., Krofel, M., Krojerová-Prokešová, J., Zimmermann, F., Anders, O., Andrén, H., Bagrade, G., Belotti, E., Breitenmoser-Würsten, C., Bufka, L., Černe, R., Drouet-Hoguet, N., Duľa, M., Fuxjäger, C., Gomerčić, T., Jędrzejewski, W. … Kuemmerle, T. (2023). Integrating animal tracking datasets at a continental scale for mapping Eurasian lynx habitat. Diversity and Distributions, 00, 1–15. https://doi.org/10.1111/ddi.13784

Description of the data and file structure

GEOTIFF files contain habitat suitability raster maps of the best-performing global and local habitat models, respectively, using an algorithm ensemble of the best-performing modeling algorithms (Maxent and random forest). Maps represent scale-integrated predictions obtained by combining (i.e., multiplying) habitat models built at the landscape and home range levels of selection (selection of home ranges in the wider landscape and selection of locations within home ranges). This means that areas with suitable habitat too small for the establishment of lynx home ranges will be 'masked out' in the scale-integrated prediction. Predictions represent a relative habitat suitability index rescaled to a 0-1000 scale, best understood as a relative ranking of raster cells in terms of their habitat suitability. Raster maps are in EPSG 3035 projection and have 100m spatial resolution.

CSV tables contain data frames used for building habitat models. Column description:

"animals_id" - ID of the lynx individual
"site" - name of the study site
"occ" - occurrence status: 1 for presence locations and 0 for available locations (background points)"sex" - sex of the individual"acquisition_time" - date and time of observation
"ruggedness" - terrain ruggedness
"forest" - forest cover"forest_integrity" - forest integrity
"human_modification" - human modification index
"accessibility" - accessibility (travel time to cities)
"snow_cover" - snow cover frequency
"road_density" - road density
"greenness_median" - median Landsat tasseled cap greenness
"greenness_variability" - amplitude of Landsat tasseled cap greenness
"brightness_median" - median Landsat tasseled cap brightness
"brightness_variability" - amplitude of Landsat tasseled cap brightness
"wetness_median" - median Landsat tasseled cap wetness
"wetness_variability" - amplitude of Landsat tasseled cap wetness

The numbers following variable names indicate spatial scale of the model variables (e.g., _22000 for 22km diameter moving window).

Funding