Ungulate herbivores are key components of terrestrial ecosystems. However, anthropogenic transformation of landscapes has altered how ungulate populations interact with vegetation communities. While many large ungulate species are at risk of extinction, deer (Cervidae) populations have increased rapidly across the temperate zone over recent decades. Overbrowsing from expanding deer populations can reduce the structural complexity of the woodland understory, prevent regeneration, and reduce tree growth, which can be detrimental to woodland wildlife and reduce the profitability of forestry operations. Expanding deer populations present a significant barrier to the goals of the UK and devolved governments to increase forest cover for carbon capture, biodiversity conservation and economic prosperity. Studying the habitat use and diet of individual deer species can provide a better understanding of their effects on woodland environments. This thesis assesses the scientific knowledge regarding the influence of deer on woodland habitats and explores the development of novel methodologies for studying deer behaviour and diet.
A systematic map collated current evidence for the ecological effects of seven deer species – six of which are present in the UK – on woodland and forest vegetation (Chapter 2). The review highlighted discrepancies in coverage between species and geographical regions and showed that many studies have not separated the effects of different deer species on the environment. A field study was conducted in the Elwy Valley, North Wales, a region occupied by a growing fallow deer (Dama dama) population (Chapter 3). Using mobile Terrestrial LiDAR (TLS) and forest inventory surveys, I characterised habitat structural components that may influence deer behaviour (Chapter 4). Horizontal visibility at deer eye-height was significantly reduced with higher densities of small-diameter tree stems, indicating that woodlands with this type of structure may provide safe refuges for deer under hunting pressure. This visibility metric was applied to behavioural data in a diel occupancy modelling approach (Chapter 5), using motion-activated camera data from a landscape-scale study of the Elwy Valley fallow deer. Diurnal occupancy was lower in hunted than non-hunted woodlands, which demonstrated the capability of fallow deer to respond to fine-scale temporal variation in risk. Using DNA metabarcoding to study the seasonal diet of this population (Chapter 6), I revealed that bramble (Rubus fruticosus agg.) was a key forage species, especially in winter, while deciduous browse accounted for a high proportion of the diet in the spring, summer and autumn. Counter to expectations, grasses formed a comparatively small proportion of the diet, indicating that this population mainly relies on woodland resources.
This thesis demonstrates the ability of fallow deer to respond to seasonal variation in resource availability and daily fluctuations in human disturbance, illustrating how their behavioural flexibility has facilitated them becoming one of the most widespread deer species in the world. The UK is lacking an effective landscape-scale deer management plan, with very little information on how individual deer species are affecting vegetation communities and wider ecosystems. This thesis exhibits use of novel technologies to refine the way we study deer herbivory and the ecological role of ungulate species in human transformed landscapes, to focus more on the individual species and design management practices that consider animal behaviour and trophic ecology.