Disentangling the impacts of herbivory and climate on ecological dynamics

Climate change is one of today's biggest environmental challenges. However, in many parts of the world, ecosystems are changing because of variation in grazing and browsing by large herbivores. Densities of large herbivores are affected by human management and land-use and are also linked to climate. This is particularly relevant in tundra and boreal parts of the world. Since herbivory and climate can interact to affect ecosystem dynamics, it may be possible to manage large herbivores to counteract the effects of climate change. In this project we investigated how climate and herbivory together affect ecological processes in northern ecosystems. These processes were researched at scales from individual plants to whole biomes. At the smallest scale, we developed a network of boreal and tundra sites in the North Atlantic region to investigate how the growth of woody plants in northern ecosystems responds to climate and herbivory and to highlight ecosystems that are vulnerable to changes in climate or herbivore density. Our results have shown that both red deer and moose can cool down the warming response of trees in multiple contexts including the Scottish Highlands and the boreal forests of Canada and Norway. However, the magnitude of the cooling effects of herbivores depends on the relationships between herbivores and vegetation, and other factors such as snow cover. In the Arctic tundra, the effect of herbivory on shrub growth was moderate, but peaked at intermediate temperatures. At larger scales, we investigated the impact of herbivores on plant species distributions, as well as how climate affects the interactions between plants and herbivores. Across Norway we have shown that herbivore communities have become wilder as livestock are replaced by increasing populations of cervids. The change has been particularly evident in the lowland forested regions of Norway. In these ecosystems, we have found that rare plants have wide ranging responses to wild herbivore densities; with some rare plants responding positively, others negatively and a third group with a hump-backed response with peak habitat suitability at intermediate herbivore densities. We have identified that epiphytic lichen species are particularly sensitive to herbivore densities, most likely due to herbivore impacts on tree composition. Finally, in order to identify regions that are susceptible to future environmental change due to changes in herbivore species composition, we have mapped patterns of herbivore diversity and community composition across global arctic and boreal biomes in terms of species richness, evolutionary relatedness and functional diversity finding lower phylogenetic diversity and functional diversity than expected in the Arctic. This shows that herbivore communities in northern ecosystems are under strong environmental constraints. Furthermore, threshold changes in the composition of herbivore communities was observed across the biome boundary between the boreal forest and Arctic tundra. Herbivore communities were directly affected by temperature, rather than through vegetation. This indicates that herbivore communities will respond to warming in northern ecosystems, with limited opportunities for trophic interactions to preserve current dynamics. Through the DISENTANGLE project, we have started to understand how the interactions between herbivores and climate jointly affect ecological dynamics in northern ecosystems. While herbivory can often counter warming climates, there are additional interacting factors that shape the overall ecological responses.

The project's outcomes include increased international collaboration, in particular with Canada and Iceland. The Disentangle project has had an impact on research in Iceland, where a new project uses the techniques developed to assess herbivore diversity across space and time in Iceland. The quantification of temperature and deer densities in determining pine growth in Scotland will be used to set target deer densities for reforestation. The preponderance of epiphytic lichens species, for which habitat suitability was associated with Cervidae densities, calls for field studies to focus on Cervidae impacts on forest lichens.

Understanding the basic ecological interactions between abiotic and biotic drivers of ecological processes is paramount in a time of global environmental change. Trophic dynamics regulate most global ecosystems, yet multiple shifts in ecological dynamics have been attributed to climate change. Densities and distributions of large-herbivores are highly dynamic, driven by management and land-use; the interactions between herbivores and climate will determine ecosystem states under future global environmental change. The DISENTAGLE project will address how climate and herbivory together affect ecosystem processes across spatial scales, from individual plant-herbivore interactions to broad geographical distributions of functional groups of herbivores. This will be achieved through a range of approaches spanning spatial scales from individual sites to global biomes. First, the nature of interactions between herbivory and climate in determining ecosystem dynamics in northern ecosystems will be investigated by partitioning variance in growth of woody plants between climate and herbivory across multiple species and a network of sites. Next, the ability of herbivores to determine plant species distributions and the influence of environmental drivers on the nature of plant-herbivore interactions will be investigated. Finally, the functional and phylogenetic ecology of herbivores will be assessed across the world's arctic and boreal ecosystems in order to identify assemblages of herbivores vulnerable to future environmental change. The outcomes of DISENTANGLE will push the boundaries of our understanding of abiotic and biotic determination of ecosystem dynamics by disentangling the role that climate and herbivores play in regulating ecological processes and ecosystem states in northern ecosystems, and by developing our ability to scale up these interactions.