Abstract
Moose (Alces alces) in Scandinavia rely on commercially valuable Scots pine (Pinus sylvestris) as a winter food source. Browsing impacts to forest resources are labeled as “damage” and have become especially important along the border between Norway and Sweden, where a semi-migratory moose population moves seasonally between two management regimes. As part of the EU-funded GRENSEVILT project, we studied multiple factors thought to drive browsing damages within a single sub-population in a cross-border context. We combined elements of two national methods of assessing browsing damage: Solbraa (Norwegian) and Äbin (Swedish). We analyzed four damage indicators (stem breakage, bark browsing, browsing pressure, and number of winter-browsed top shoots) by grouping over 20 predictor variables into three categories, using data collected on 3,033 individual pine trees. GLMM model variable selection was completed using AIC. Covariates that measured severity of prior damage (such as cumulative impacts on plant architecture) were included in the top models for all four indicators. Notably, the covariates for pine density and snow depth, factors previously found to be important in predicting browsing damage, were not present in any of the top models. GAM landscape models revealed that distributions of the four damage types are quite different – with bark browsing illustrating an isolated “hotspot” and browsing pressure showing more widespread prevalence. There is some evidence to suggest that different damage types represent a natural progression of moose foraging, with stem breakage and bark browsing signaling more severe damage and potential over-use with time.