Behavioural changes of prey towards large carnivores can have profound impacts on prey populations and the ecosystems they inhabit. However, European studies have thus far only found limited support for prey behavioural changes to returning large carnivores. We examined the trade-off between forage quality and perceived predation risk of an ambush (Eurasian lynx; Lynx lynx) and cursorial predator (wolf; Canis lupus) on the behavioural response of red deer (Cervus elaphus) in an experimental set-up within semi-natural grasslands and forest patches. We hypothesized that red deer would i) reduce visitation frequency and duration but increase their time spent vigilant in the presence of olfactory cues of large carnivores, ii) show a stronger response towards the scent of wolf with increasing distance from the forest edge in open areas and a stronger response towards lynx in forested areas independent of distance, and iii) show a stronger response in areas with lower forage quality. We tested these using camera trap video data on 76 plots at different distances within the forest and open areas (mown and unmown) in the Grafenwöhr military training area, Germany. Within three trials, we applied each of the following scent treatments to each plot: wolf urine-scat, lynx urine-scat, and horse urine-dung. To analyse forage quality, we determined crude protein concentration from the herbaceous vegetation by near-infrared spectroscopy. Red deer avoided open areas further away from the forest edge when wolf scent was present, whereas an increase in vigilance was found in forested areas towards lynx scent. In contrast to plots with horse and lynx scent, red deer visitation frequency on plots with wolf scent remained low despite increasing crude protein concentration. Our research is the first to experimentally show that large carnivores with different hunting mode induce different changes in the foraging behaviour of free-ranging red deer.

Predators can affect ecosystems through non-consumptive effects on their prey, which can lead to cascading effects on the vegetation. In mammalian communities, such cascading effects on whole ecosystems have mainly been demonstrated in protected areas, but the extent to which such effects may occur in more human-dominated landscapes remains disputable. With the recolonisation of wolves (Canis lupus) in Europe, understanding the potential for such cascading processes becomes crucial for understanding the ecological consequences of wolf recovery and making appropriate management recommendations. Here, we investigate the evidence for non-consumptive effects of wolves on their wild ungulate prey and cascading effects on the vegetation in European landscapes. We reviewed empirical studies reporting wild ungulate responses to wolves involving spatio-temporal behaviour at large and fine spatial scales, activity patterns, vigilance, grouping, physiological effects, and effects on the vegetation. We reveal that non-consumptive effects of wolves in Europe have been studied in few regions and with focus on regions with low human impact and are highly context-dependent and might often be overruled by human-related factors. Further, we highlight the need for a description of human influence in NCE studies. We discuss challenges in NCE research and the potential for advances in future research on NCE of wolves in a human dominated landscape. Further, we emphasise the need for wildlife management to restore ecosystem complexity and processes, to allow non-consumptive predator effects to occur.

Quantifying how multiple ecosystem services and functions are affected by different drivers of Global Change is challenging. Particularly in African savanna regions, highly integrated land-use activities created a landscape mosaic with flows of multiple resources between land use types. A framework is needed that quantifies the effects of climate change, management and policy interventions on ecosystem services that are most relevant for rural communities, such as provision of food, feed, carbon sequestration, nutrient cycling and natural pest control. In spite of progress made in ecosystem modelling, data availability and stakeholder interactions, these elements have neither been brought together in an integrated framework, nor evaluated in the context of real-world problems. Here, we propose and outline such framework as developed by a multi-disciplinary research network, the Southern African Limpopo Landscapes network (SALLnet). Components of the framework such as the crop model APSIM and the vegetation model aDGVM2 had already been parameterized and evaluated using data sets from savanna regions of eastern, western and southern Africa, and were fine-tuned using novel data sets from Limpopo. A prototype of an agent-based farm household model was developed using comprehensive farm survey information from the Limpopo Province of South Africa. A first test of the functionality of the integrated framework has been performed for alternative policy interventions on smallholder crop-livestock systems. We discuss the versatile applicability of the framework, with a focus on smallholder landscapes in the savanna regions of southern Africa that are considered hotspots of global change impacts.