Data from: Habitat selection of moose in Sweden in managed boreal forests with Pinus contorta and Pinus sylvestris
Data files
Mar 25, 2026 version files 25.22 MB
Abstract
Human land use can leverage exotic species to increase financial benefits. However, the use of exotic tree species may affect ecosystem functioning, including the habitat use and movement behaviour of animals, their ecological impacts, and their interactions with human land use. The native pine of Sweden is Scots pine (Pinus sylvestris). In 1920, the Swedish forestry introduced the North American Lodgepole pine (Pinus contorta) to Sweden to increase timber production. We used a two-year data set from GPS-collared adult moose (Alces alces) to examine the annual and seasonal habitat selection of this large keystone herbivore in a managed forest landscape with a relatively higher percentage of Pinus contorta compared to other Swedish boreal regions. Compared to other studies, a central departure point in this work was our focus on whether moose exhibited different selection between the native Pinus sylvestris stands and the exotic P. contorta stands.
We found no evidence that moose avoided stands of P. contorta within forest landscapes. Instead, our results show that young stands of both P. sylvestris and P. contorta were a highly preferred habitat for moose year-round, and that selection for P. contorta forests depended on the proportion of P. contorta at the landscape level and the season. Our work also reveals a diverse selection of forested habitats in almost all seasons, reflecting the differing needs of moose during different times of the year. Despite being rare in our system, we suggest our findings can have important implications for forestry – P. contorta is often planted in the belief that these forests are less likely to be browsed by moose than are native P. sylvestris forests. Our results do not indicate that the current distribution of different P. Contorta stands have a limiting effect on how moose distribute across the landscape matrix.
Dataset DOI: 10.5061/dryad.6djh9w1gt
Description of the data and file structure
Data and R-code for the publication Habitat selection of moose in Sweden in managed boreal forests with Pinus contorta and Pinus sylvestris
The present data set consists of one zipped folder that includes one documentation file (READ_ME), one data file, and one associated code file.
Objective
Quantification of annual and seasonal habitat selection by adult moose (Alces alces) in a managed forest landscape in Sweden with a relatively higher percentage of Pinus contorta compared to other Swedish boreal regions. Compared to other studies, a central departure point in this work was our focus on whether moose exhibited different selection between the native Pinus sylvestris stands and the exotic P. contorta stands.
Description
The data files and code files are connected to the analyses and findings in the present open-source publication (Neumann et al. 2026). A detailed description of the methodology and processing of the data can be found in the Materials and Methods section of the publication.
Files and variables
File: Data_and_R-code_for_the_publication_Habitat_selection_of_moose_in_Sweden_in_managed_boreal_forests_with_Pinus_contorta_and_Pinus_sylvestris.zip
Data file: Data_WBL370036.tsv
Information on selected habitat features at the end of observed and random steps by 29 adult GPS-marked moose in Northern Sweden (counties of Västernorrland, Jämtland, and Västerbotten, March 2022-2024). Land cover information comes from the Swedish maps on National Land Cover (Swedish EPA 2019), tree height (Swedish EPA 2019), and Pinus contorta proportion (SLU Forest Map 2018). NA represents missing data.
11 columns, 1065427 rows
- case_: Binary, observed (TRUE) and random (FALSE) step. Each observed step is coupled with 10 random steps
- step_id_: Numeric, identifier of the step, coupling one observed and 10 random steps
- cos_ta: Numeric, cosine of the turning angle, belonging to each step
- log_sl: Numeric, logarithm of the step length, belonging to each step
- seasons: Factor, season a given step is falling into, four alternatives (Winter, Spring, Summer, Autumn)
- habitat: Factor, final habitat classes used for the analysis; 15 classes.
- Pine 5-15m (Minimum crown cover of 10%, crown cover of at least 70% of pine, tree height 5-15 meters)
- Spruce 5-15m (Minimum crown cover of 10%, crown cover of at least 70% of spruce, tree height 5-15 meters)
- DecMixed 5-15m (Minimum crown cover of 10% where coniferous crown cover can reach up to 70%, but deciduous crown cover can reach >70%, tree height 5-15 meters)
- Contorta 20-40% 5-15m (Forestland with the proportion of 20-40% P. contorta, tree height 5-15 meters)
- Contorta >40% 5-15m (Forestland with the proportion of >40% P. contorta, tree height 5-15 meters)
- Pine >15m (Minimum crown cover of 10%, crown cover of at least 70% of pine, tree height above 15 meters)
- Spruce >15m (Minimum crown cover of 10%, crown cover of at least 70% of spruce, tree height above 15 meters)
- DecMixed >15m (Minimum crown cover of 10% where coniferous crown cover can reach up to 70%, but deciduous crown cover can reach >70%, tree height above 15 meters)
- Contorta 20-40% >15m (Forestland with the proportion of 20-40% P. contorta, tree height above 15 meters)
- Contorta >40% >15m (Forestland with the proportion of >40% P. contorta, tree height above 15 meters)
- Young <5m (Temporary non-forest land with a tree height below 5 m that is expected to grow into a forest, for example, clearcuts, fire-damaged forests, wind-damaged forests)
- Contorta 20-40% <5m (Forestland with the proportion of 20-40% P. contorta, tree height below 5 meters)
- Contorta >40% <5m (Forestland with the proportion of >40% P. contorta, tree height below 5 meters)
- Wetlands (Open land where the water for a large part of the year is close by, in or just above the ground surface)
- Others (Not forest-related habitat: Arable land, Inland/marine water, open land, roads, and buildings.)
- habitat0: Factor, habitat classes, similartos ‘habitat’, including height information, but deciduous (dec) and mixed forest separated, 17 classes.
- habitat00: Factor, habitat classes, similar to ‘habitat’ but without height information, 11 classes.
- landcover.cl: Factor, land cover classes as given by the Swedish National Land cover map (Swedish EPA 2019), without any information on Pinus contorta
- pinusContorta_share: Numeric, share of Pinus contorta as given by the standing volume map (SLU Forest map 2018).
- height_5_45m: Numeric, forest height class as given by the additional height map connected to the Swedish National Land cover map (Swedish EPA 2019)
References
Bodlund, M., Stenbacka, F., Widemo, F., Ball, J.P., Ericsson, G. & Neumann, W. Habitat selection of moose in Sweden in managed boreal forests with Pinus contorta and Pinus sylvestris. Wildlife Biology DOI: 10.1002/wlb3.01662.
Swedish EPA 2019. Maps on official open-source National land cover and vegetation height data, 10x10 m pixel, 2018. https://geodata.naturvardsverket.se/nedladdning/marktacke/NMD2018/NMD_Produktbeskrivning_NMD2018Basskikt.pdf [In Swedish], https://geodata.naturvardsverket.se/nedladdning/marktacke/NMD2018/NMD2018_ProductDescription_ENG.pdf,
SLU Forest Map 2018. Map on pthe ercentage share of Pinus contorta produced through co-processing of field data from the Swedish National Forest Inventory, as well as aerial images and satellite images. https://www.slu.se/en/environment/statistics-and-environmental-data/environmental-data-catalogue/slu-forest-map/
Code/software
Code_WBL370036_final.R: This R script cleans the workspace, loads movement ecology data, creates detailed habitat categories based on land cover and vegetation height, and computes habitat availability percentages.
It then fits integrated step selection function (iSSF) models (overall and seasonal) to analyze habitat selection and visualizes the results using regression plots.
Coding language is linked to the programming software R (version 4.4.1, https://cran.r-project.org/)
R (version 4.4.1) was used for spatial analyses, data handlin,g and visualization.
Used packages:
· library(amt)
· library(dplyr)
· library(readr)
· library(sjPlot)
· library(egg)
