Discussion
At the experiment site there was an overall significant effect of nest predation at the forest edge compared to the interior, however our hypothesis was incorrect that this would be effect would be present on ground nests. As previously mentioned, the universality of a linear edge effect on nest predation has been effectively disproved. As Lahti (2001) showed, the majority of nest predation experiments show no significant difference between edge and interior, which may be because the gradient of predation rates is not always linear. Previous research has indicated that nest predation edge effects can follow unusual, eratic gradients at intermediate distances from the edge The varied nature of these previous findings is suggested to be due to interactions between multiple abiotic edge factors, causing unique microclimates at various distances from the edge. Hence the focus should be on the abiotic and biotic factors specific to this edge and interior which may interact to cause such an edge effect on arboreal nests.
Abiotic Edge Factors Affecting Predation
Edge contrast has previously been hypothesised to exaggerate edge effects \cite{Ries_2004}. The bordering habitat to the examined forest fragment is open grassland, which has a comparatively low productivity compared to forests \cite{Lahti_2001}. Previous studies have indicated that the steepness of the gradient in productivity over an ecotone may play a role in nest predation, and that such an effect is exaggerated by high levels of habitat fragmentation . Lahti (2001)'s meta-analysis did not show such an effect, however it did show that a edge effect on nest predation is significantly less likely where the gradient in productivity is shallow (e.g grassland and agricultural land). This suggests steepness of ecotone productivity gradients may only increase nest predation conditionally, subject to other factors. The large productivity contrast in this study between woodland and open grassland may partially explain increased nest predation at the edge, however other U.K. forest fragments with smaller productivity contrasts will need to be examined before a casual effect can be established.
The effect of orientation may not have played a significant role on the observed edge effect in this study. The experimental forest edge is east facing and previous studies on the effect of orientation on edges focus on North-South orientation. Current understanding of orientations impact on edges suggests east-west orientation does not significantly influence edge effects \cite{Ries_2004a}.
There may however be a temporal impact on this experiment, as the study was only carried out for 2 weeks post-spring, where U.K. bird populations are near their peak Previous studies in Scotland have shown significant seasonality in nest predation rates around spring. Nest predation rates were shown to peak in late April, and then slowly decline until late June \cite{Malzer_2015}. If this seasonal predation effect is also seen in England, it may help account for the unusually high levels of predation seen in this experiment. However it should be noted that although Malzer & Helm (2015) found an overall effect of season on nest predation, it did not influence the edge effect, which was constant year round.