Data and code from: Why wolves walk – drivers of daily movement in a social large carnivore
Data files
Jul 31, 2026 version files 304.24 KB
-
dataset_summer_dmd_linearity.csv
69.81 KB
-
dataset_winter_dmd_linearity.csv
196.23 KB
-
DMD_D_analysis.R
16.62 KB
-
Linearity_D_analysis.R
17.27 KB
-
README.md
4.32 KB
Aug 25, 2026 version files 301.52 KB
-
dataset_summer_dmd_linearity.csv
67.08 KB
-
dataset_winter_dmd_linearity.csv
196.23 KB
-
DMD_D_analysis.R
16.62 KB
-
Linearity_D_analysis.R
17.28 KB
-
README.md
4.32 KB
Abstract
Animal movement patterns are influenced by a range of extrinsic and intrinsic factors, including the distribution of resources, human disturbance, landscape connectivity, species interactions, and individual traits. We examined factors associated with daily movement distance (DMD) and linearity in 20 territorial Scandinavian wolves (Canis lupus) during summer (n = 406) and winter (n = 614) by analysing hourly GPS locations (2001-2019). We used generalised linear mixed-effects models to relate DMD and linearity to explanatory variables including indicators of road infrastructure (gravel road density, paved road density), environmental quality (forest fragmentation), prey availability (moose (Alces alces) and roe deer (Capreolus capreolus) densities), intra- and interspecific density (number of neighbouring wolf territories, brown bear (Ursus arctos) density), individual characteristic (breeding status, inbreeding coefficient), and annual home range size. DMD was shorter during winter than summer, but linearity was similar. In summer, DMD was positively associated with the number of neighbouring territories, home range size, and brown bear density, but linearity was not related to any of the included variables. A higher number of neighbouring territories and larger home range size likely require longer daily movements to cover and defend the territory. The relationship with brown bear density supports previous research showing increased search time for ungulate prey for wolves in areas with high brown bear densities. In winter, DMD was negatively associated with forest fragmentation, suggesting reduced movement when forests are patchy. DMD was also negatively associated with roe deer density, indicating reduced movement when preferred prey is abundant. Contrary to summer, linearity in winter decreased with increasing number of neighbouring territories. Alternative models also indicated reduced DMD with increasing paved road density and increased movement linearity with increasing gravel road density during winter. Our results showed that wolf movement in Scandinavia is shaped by extrinsic factors, with limited influence from individual-level traits.
Description of the data files and structure:
The data are organized into two seasonal datasets (summer and winter), each containing the same set of variables. Some variables are only used in one season due to ecological or contextual reasons: brown bear density and breeding status are included only in summer, moose and roe deer densities only in winter.
Below is a description of the columns:
- wolf_id: Unique identifier for each individual wolf in the study.
- breeding_status: Whether there have been confirmed pups that year or not.
- neighbouring_territory: Number of neighbouring territories surrounding the wolf individual's territory.
- paved_density_km: Density of paved roads (km/km2) within the wolf's study area.
- gravel_density_km: Density of gravel roads (km/km2) within the wolf's study area.
- moose_density: Winter density estimate of moose (individuals/km2) within the wolf's study area, estimated from pellet group counts.
- roedeer_density: Winter density estimate of roe deer (individuals/km2) within the wolf's study area, estimated from pellet group counts.
- bear_density: Relative density of brown bears (individuals/km2) within the wolf's territory, estimated from official harvest statistics and spatial distribution data.
- fi_highest: The inbreeding coefficient for each individual wolf.
- date: Date of the wolf's daily movement measurement.
- sum_distance_km: Daily movement distance of the wolf (in kilometers).
- linearitet: Index of straightness of the wolfves daily movement paths, ranging from 0 to 1, with values closer to 1 indicating more directed movements.
- fragmentation_index: Index of forest fragmentation within the wolves study areas, with higher values indicating a more fragmented forest structure.
- mcp_100_km2: 100% MCP for yearly home range size (km2).
Analysis scripts
Two R scripts are included.
DMD_D_analysis.R
This script contains the analyses of daily movement distance (sum_distance_km) for summer and winter. It reads the seasonal datasets, prepares the data for analysis by removing incomplete observations and standardizing continuous predictors, and fits generalized linear mixed-effects models with wolf identity included as a random effect. Candidate models are compared using AICc, and the script outputs model summaries, confidence intervals, R² values, effect estimates, and the figures presented in the manuscript.
Summer models use a Gaussian error distribution, whereas winter models use a Gamma distribution with a log link.
Linearity_D_analysis.R
This script contains the analyses of daily movement linearity for summer and winter. As in the daily movement distance analysis, the data are prepared by removing incomplete observations and standardizing continuous predictors before fitting beta mixed-effects models with wolf identity as a random effect. The script performs model selection using AICc and outputs model summaries, confidence intervals, R² values, effect estimates, and the figure included in the manuscript.
Required input files
Both scripts require the following semicolon-delimited data files:
dataset_summer_dmd_linearity.csvdataset_winter_dmd_linearity.csv
The scripts assume that these files are located in the working directory. If they are stored elsewhere, the file paths in the scripts must be updated accordingly. ## Software
The analyses were performed using R version 4.3.2 on Windows 11 (64-bit).
The following R packages were loaded and used:
readr2.1.5dplyr1.1.4glmmTMB1.1.8MuMIn1.47.5performance0.16.0DHARMa0.4.7broom.mixed0.2.9.6ggeffects1.5.0ggplot24.0.2patchwork1.3.2
The complete R session information, including packages loaded through dependencies, can be obtained by running:
sessionInfo()
Running the scripts
Set the working directory to the folder containing the scripts and input data files, then run:
source("DMD_analysis.R")
source("Linearity_analysis.R")
The scripts can be run independently and do not require the output from one another.
Changes after Jul 31, 2026:
2026/08/23: Updated brown bear density data and subsequent changes in the R scripts for DMD and linearity.
