Spatio-Temporal Modelling of Moose Browsing Damage in Young Swedish Forests

Description:

This project aims to improve the understanding of moose browsing damage in young Swedish forests by developing a spatio-temporal statistical model based on national ABIN monitoring data from 2016–2025. Moose browsing is an important challenge in Swedish forestry because it can reduce tree growth, damage young stands, affect species composition, and delay forest regeneration. Since browsing pressure varies across both space and time, general regional summaries are often not enough to support local management decisions. More detailed information is needed to identify where damage risk is high, how risk changes between years, and where the available data are uncertain.

The project will use plot-level ABIN data on the number of available stems and the number of stems damaged by browsing. These data will be analysed using a model-based approach that can estimate browsing damage risk not only at sampled plots, but also at unsampled locations across Sweden. The main goal is to produce annual, wall-to-wall risk maps for young forests, together with uncertainty maps showing where predictions are reliable and where more information may be needed. This will make it possible to identify persistent hotspots, temporary high-risk areas, and spatial patterns that may be hidden in aggregated management-unit averages.

The statistical model will combine spatial and temporal information in a single framework. It will account for yearly changes in browsing pressure and for spatial clustering of damage, while also allowing risk patterns to persist or change over time. The model will be evaluated using spatial and temporal validation methods, so that its ability to predict new areas and new years can be assessed. This is important for ensuring that the final maps are not only visually informative, but also statistically reliable.

The expected results will support both forestry and wildlife management. By identifying areas with high browsing risk, the project can help guide targeted regeneration measures, such as tree species choice, plant protection, and planning of young stand management. The uncertainty maps will also help decision-makers understand where conclusions are strong and where caution is needed. In this way, the project can improve communication between forest managers, wildlife managers, and conservation stakeholders by providing a shared spatial picture of browsing risk.

Overall, the project contributes to evidence-based Swedish nature conservation by providing better knowledge of how browsing pressure affects young production forests. Since conservation outcomes depend not only on protected areas but also on how managed forests are regenerated and maintained, improved information about browsing damage is highly relevant. The project will deliver annual risk and uncertainty maps, a scientific manuscript, reusable R code, and a final report, making the results useful for both research and practical forest management.