As well as abiotic factors, biotic factors can also affect ecosystem processes at the forest edge. Ecological generalists are more able to adapt to the varied abiotic factors of edges, while edge effects are generally considered to have negative effects on existing specialist biological communities \cite{Harrison_1999}. An increased proportion of these generalists can affect overall ecosystem functioning through degradation of underlying ecological processes, such as nest predation \cite{Ries_2004}. Increased nest predation as an edge effect was initially suggested as a general rule in 1978, and was hypothesised to be driven by increased bird population density at edges \cite{Gates_1978}. Previous studies had already shown predator population density can increase at edge habitats \cite{Bider_1968}, which indicates edges may form a so-called "ecological trap" for nesting birds. This ecological trap is formed of 2 parts: i) The already observed effect of nest predation increasing with nest density \cite{Krebs_1971}, and ii) increased nest density at edges \cite{Gates_1978}, which together form the cumulative edge effect on nest predation. While many previous studies have shown nest predation can increase with proximity to the forest edge, the universality of this rule has been challenged \cite{Huhta_1999}. In a meta-analysis, Lahti (2001) found the hypothesis had been rejected in a majority of experiments, and the underlying causes of such an effect are not as simple as the ecological trap hypothesis suggests. It's been subsequently suggested that a nest predation edge effect is mediated by abiotic edge factors, such as land use of the bordering habitat. For example, forest fragments bordering open grassland has been hypothesised to assist predators using linear geographic features such as edges as a foraging route \cite{Rodewald_2002}. Geographic orientation is another edge factor which can exacerbate edge effects, and is rarely controlled for in edge effect studies. There is strong evidence to support the hypothesis that south-facing edges in the Northern hemisphere suffer larger edge effects, due to increased insolation \cite{Ries_2004a}. Similarly temporal factors, such as seasonality and annual changes in temperature have been seen to play a role in edge effects, and some edge responses can even reverse depending on the season, although this is rare \cite{Ries_2004a}. All of these abiotic edge factors interact with the native biota and can create unique combinations of edge effects. Given the vast variability in these edge factors worldwide, and that predator composition and behaviour differs between regions, this suggests previous findings on edge-mediated nest predation may not be applicable to new experimental environments \cite{THOMPSON_2007}. Particularly given that most nest predation experiments do not identify predators responsible for effect \cite{article}.
Edge effects may have significant impacts for birds in the United Kingdom (U.K.). Native forest cover in the U.K. has been declining since the early Holocene, and the remaining areas are increasingly fragmented \cite{Whitehouse_2010}. Historical forest cover in the U.K. is inherently difficult to calculate, as undocumented prehistoric land conversion from forest to subsistence farming destroyed most of the native climax community and replaced it with artificial habitats, such as heath and moorland \cite{PAULEIT_2005}. From the 19th century onward, land use change in temperate regions has trended away from subsistence farming towards urbanisation and intensive agriculture. These trends have only spread to tropical regions in the last 50 years, which has resulted in a loss of 1/3 of the world's forested areas \cite{Haddad_2015}. Urbanisation has been responsible for vast deforestation in sub-Saharan Africa, as urbanisation drives up demand for fuel wood \cite{n1984}. While intensive agriculture is the current main driver of deforestation in the Amazon rainforest, due to the economic incentives of cattle ranching \cite{Sparovek_2012}. Despite these global trends, forest cover in the U.K. has been increasing since the 1920s \cite{defra2013}. However, the majority of original forest cover has been replaced with non-native conifer monocultures. These monocultures do not provide the same ecological function as native forests and further fragments the U.K. deciduous climax community \cite{Liu_2018}. This high level of fragmentation is particularly problematic for the contained biological communities, as edge effects increase as a function of habitat fragmentation \cite{Vetter_2013}. In addition, U.K. forest fragments often border open farming landscapes due to economic incentives for intensive agriculture \cite{PAULEIT_2005}. These open landscapes provide a "sharp" border to the forest, and are capable of supporting high densities of opportunistic generalist predators, which can then prey on the exposed edges \cite{Lahti_2001}. These risks are particularly high for increasingly endangered U.K. ground nesting birds, given the predominantly terrestrial morphology of opportunistic generalist predators.
Research on the effects of edges on nest predation in temperate ecotones have largely been performed in North America \cite{Lahti_2001}. As U.K. predator composition and behaviour differs greatly from North America, this suggests findings from these studies may not be applicable to U.K. deciduous woodland \cite{THOMPSON_2007}. This study aims to examine whether the previously hypothesised generality of edge effect mediated nest predation applies to a U.K. mixed deciduous forest fragment, at both ground and arboreal levels. It was hypothesised that predation rates would increase with proximity to the forest edge regardless of height. Given the previous mixed effects of edge on nest predation, a large scale study seemed appropriate to examine the effect on the forest fragment and this experiment is the first of its scale to study these effects in the U.K.
Many previous studies on nest predation using artificial nests have been unable to identify predators responsible for the predation, despite predator identity being hypothesised as a significant driver of nest predation at forest edges \cite{article}. Therefore, in this study, predators were identified, where possible, using bite marks left on plasticine eggs. This should provide insights into the composition of predators that prey on forest edges in deciduous woodland and elucidate pest control priorities around these fragmented areas.
Methods