Grazing and warming effects on shrub growth and plant species composition in subalpine dry tundra : An experimental approach

Questions Vegetation in the forest–tundra ecotone faces changes in both climate and land‐use. While climate warming is an important driver of vegetation growth and composition, herbivory may have opposing effects. In the present study, we experimentally test how removal of sheep herbivory affects the vegetation in an alpine forest–tundra ecotone, and how responses are manifested at higher temperatures. Location Dovre Mountains, Central Norway. Methods Shrub growth (height and cover) and ground layer composition were analysed each third year over an 18‐year period in a nested, three‐factorial experiment (ambient temperature and herbivory; ambient temperature and no herbivory; increased temperature and no herbivory). Fencing and open‐top‐chambers were used as expedients. Treatment effects and interactions over time were analysed using linear mixed effects models and ordination. Results Shrub height and cover increased over time due to reduced herbivory, but without additional warming effect. Lichen cover declined in all treatments over time, but more rapidly and earlier under warming treatment (significant after three years). Contrary to expectations, there was no statistically significant increase in woody species due to warming, although evergreen woody species displayed a trend shift after six years, comprising a sharp decline towards year twelve. Litter accumulated in all treatments, but at higher rates under warming (significant after nine years). Conclusions Our results disclose removal of sheep herbivory as a prominent driver of shrub growth, with warming as a subordinate driver in the studied alpine vegetation. The warming‐driven increased litter abundance may, however, be caused by the decrease of wind inside chambers and the subsequent absence of wind‐driven removal of litter. This chamber effect and the displayed timing differences in vegetation responses call for the critical use of short‐term experimental data in predictions of long‐term consequences of environmental change