Forage availability and moose winter browsing in forest landscapes

In a large-scale survey, comprising 38 landscapes throughout Sweden, we measured winter forage availability and use by moose within different forest- and land types, and also at three different spatial scales: plot, tract and area. Measured as percent cover of browse species within moose browsing range, young forests contained significantly more browse (16.5 +/- 1.20%: mean +/- SE) compared with older forests (6.7 +/- 0.57%) or non-forest land (5.9 +/- 0.67%). Total cover of browse species increased with increasing mean tree height in forests, and peaked at around 22% when the trees were 3-5 m. The relative consumption was highest in young forests (9.2 +/- 0.98%) compared with non-forest land (7.8 +/- 1.40%) or older forests (5.8 +/- 0.78%). The selection order did not vary much among forest- or land types, with aspen, willows, rowan and oak being the most selected species. A multiple regression, including four independent variables (moose index, cover of Scots pine, cover of other species and temperature sum) explained 45% (young forests), 18% (older forests) and 36% (non-forest land) of the variation in relative consumption. In young forests, moose index and cover of Scots pine accounted for 40 and 44%, respectively, of the explained variation. Model selection, using AICc, further emphasized the importance of moose index and cover of Scots pine, as these variables were included in the most parsimonious models for all forest- and land types. At the tract level, we recorded a proportional relationship between total cover of browse species and absolute consumption, suggesting an ideal free distribution. Taking into account the availability of forage, the degree of browsing and the proportion of each land type in the landscape, we calculated that 44.3% of the total forage consumption occurred in young forests, 42.3% in older forests and 13.4% in non-forest land. Our main conclusions are that moose conform to an ideal free distribution based on availability of forage within the landscape, and that all forest- and land types, not just young forests, are important as sources of browse for moose. Thus, it may be possible to release the browsing pressure on damage-sensitive young forests by increasing the food supply in other forest- and land types. Furthermore, managers needs to consider both the size of the moose population and the amounts of browse in the landscape when deciding on prudent management actions.