Woodlark presence data with corresponding environnemental variables and code used for the model of Habitat selection

Sandrine Wider
Supervision: Prof. Clara Zemp
Université de Neuchâtel


Contact : sandrine.wider@vogelwarte.ch
clara.zemp@unine.ch

Presence Data collected by Vogelwarte between years 2020 and 2022 (see publication for precise protocol)
in the Jura mountains in the canton of Neuchâtel

Environmental data provided by:
- Canton of Neuchâtel : Données cartographiques du SITN © [2022] / Service de la Géomatique et du Registre Foncier / Service de l'agriculture
- Federal inventory of dry pastures
The use of the data is restricted to this specific project and should not be distributed further. 

Work financed by the Conservation Biology Lab, university of Neuchâtel

Files explanation: 

GISdata.gpk cointains Vector data:
- pres : presence data of Woodlark
- transect : transects used to collect presence data
- stone : surfaces of rocky outcrops, computed using Aerial images by Nica Humber (see related paper for Detail About protocol)
- pps : surfaces of dry pastures from the national inventory in canton of Neuchâtel - 2025
- forest : surfaces of Forests, computed from land use map from the canton of Neuchâtel
- fer_esti : fertilized surfaces in summer pastures, provided by canton of neuchâtel
- edges : lines representing all forest and Group of trees edges, computed from a MNC derived from Lidar data belonging to canton de Neuchâtel, point density: 30 pts/m2, spatial resolution: 1m )
- bareground : surfaces of bareground, computed using Aerial images by Nica Humber (see related paper for Detail About protocol)

Raster_data.tiff conatains raster:
- mnc : used to extract isolated shrubs and trees (derived from Lidar 	data belonging to canton de Neuchâtel, point density: 30 pts/m2, spatial resolution: 1m )
- tri : each cells has the value of tri at a 10m x 10m spatial Resolution, computed using an DEM (Digital Elevation model derived from a Lidar data belonging to canton de Neuchâtel, point density: 100 pts/m2, resolution: 1 m
- exp : each cells has the value of exposure at a 10m x 10m spatial Resolution, computed using the same DEM

Results_all.Rdata:

This is the data frame where each line/observation is a presence or a generated pseudo-Absence. Columns are the environemental variables dervied from GISdata.gpk and Raster_data.tiff, measured at ten different radius around Observation (50,100,150,200,250,300,350,400,450,500m)

R_code :

1. GIS data workflow: getting pseudoabsences, wrangling and extracting GIS covariates
2. Select best spatial scales for each predictors
3. Habitat selection functions: main analysis
4. Plotting results: Model estimates and effect plots