Unveiling the patchwork: landscape arrangement drives moose winter utilization of forests

Context Landscape heterogeneity is a key driver of ecological processes, yet how spatial structure influences animal space use across scales remains insufficiently understood. In managed boreal forests, the composition and configuration of forest patches in the landscape may constrain how large herbivores use their winter habitat. Objectives We examined how multi-scale structural metrics of young forest patches shape utilization intensity within moose ( Alces alces ) winter ranges and core areas. Methods Using GPS data from 176 adult moose across seven populations in Sweden (2016–2024; 62–68°N), we estimated winter ranges with Biased Random Bridge (BRB) models to quantify utilization distributions (UDs). We distinguished intensively used core zones (50% UD) from peripheral zones (95–50% UD) within a given winter range. The landscape was classified into three habitat classes biologically relevant for moose during winter: young forest (tree height < 5 m), forest (tree height ≥ 5 m), and non-forest. For each moose and winter season, UDs were linked with multi-scale landscape metrics characterizing patch area, patch cohesion, fragmentation, and edge density. Linear Discriminant Analysis was conducted to identify structural differences between core areas and peripheral zones across three spatial levels, and mixed models were used to determine whether utilization intensity varied with patch structure, connectivity, snow depth, and soil moisture. Results Patches differed structurally between core areas and peripheral zones. Core areas were characterised by fewer, larger, and more consolidated patches with higher young forest (< 5 m) representation, while peripheral zones displayed greater patch fragmentation, higher patch cohesion, and larger patches of forest with larger trees (≥ 5 m) across all habitat classes. Individual patch geometry did not differ between core areas and peripheral zones at any site. These patterns were consistent across populations. Within core areas, moose utilization intensity was positively associated with edge density and cohesion of forest (≥ 5 m), and negatively associated with young forest patch area, suggesting that moose concentrate their winter space use in well-connected, edge-rich forest interspersed with small young forest patches. Conclusions Landscape arrangement exerts a strong influence on the spatial distribution of a large herbivore such as moose within managed boreal forest matrices. Structural features of the landscape mediate behavioural responses to resource distribution and environmental constraints, shaping how animals organise and concentrate their winter habitat use. This highlights landscape structure as a behavioural filter linking forest management patterns to animal movement, with direct implications for forest planning aimed at redistributing winter space use and possibly reducing localized browsing damage.