Example dataset for N-SDM workflow

This dataset is an example for modeling habitat suitability using the N-SDM workflow. It focuses on three species (Larix decidua, Capra ibex, and Cantharellus cibarius) at 100-meter resolution across Switzerland, covering both the current period (1980–2021) and a future period (2070–2100). The dataset is intended for demonstration and testing, enabling users to reproduce the workflow setup, follow the folder and naming conventions, and explore the functionality of N-SDM on a case study.

The archive contains a structured set of folders that mirror the expected N-SDM data architecture:

• covariates: environmental and climatic predictor layers.

• species: occurrence data for model training and evaluation.

• masks: optional spatial masks used to constrain predictions.

• background: optional raster bias (weight) files for constraining background point sampling.

Data sources:

• Global-level occurrence data (varying spatial precision) were obtained from GBIF https://www.gbif.org/ (DOI: 10.15468/dl.fktyas).

• Regional-level occurrence data (aggregated at a 100 m resolution) were provided by the Swiss Species Information Center, InfoSpecies https://www.infospecies.ch (DOI: 10.15468/htjezm), and are shared after agreement with InfoSpecies.

• Climatic covariates (~1 km resolution) were sourced from CHELSA (DOI: 10.1038/s41597-021-01084-6).

• Environmental covariates (resampled at a 100 m resolution) were sourced from the SWECO25 database (DOI: 10.1038/s41597-023-02899-1).

# set your target nsdm directory
TARGET_DIR="/cluster/work/eawag/p01001/nsdm"

# go there
cd "$TARGET_DIR"

# download the archive (resume if interrupted)
wget -c "https://zenodo.org/record/17177174/files/data.zip?download=1" -O data.zip

# unzip and overwrite existing files if needed
unzip -o data.zip

# optional: clean up
rm data.zip