Double-edged effect? Impact of dual edge proximity on the distribution of ancient woodland indicator plant species in a fragmented habitat

Abstract

The influence of edge proximity on woodland plants is a well-established research area, yet the influence of dual edge exposure has rarely been investigated. This novel research aims to establish whether proximity to two edges has any additive influence on Ancient Woodland Indicator (AWI) species presence relative to proximity to a single edge. Several AWI species are threatened and thus specific conservation priorities, while Ancient Semi-Natural Woodland (ASNW) itself is often highly fragmented: almost half of remnant patches are less than 5 ha, which increases the potential for dual edge effects. Here, systematic mapping of herbaceous AWI species was conducted in 310 vegetation plots in two formerly-connected ASNW fragments in South-West England. Linear regression modelling revealed that distance to nearest edge and distance to second nearest edge were both univariately positively correlated with AWI species richness. After distance from nearest edge was entered into a multivariate model first, distance from second edge was entered in a second optional step after meeting stepwise criteria. The resultant multivariate model was more significant, and explained more variance, than either variable in isolation, indicating an additive effect of dual edge exposure. Likewise, binary logistic regression modelling showed presence of individual AWI species (Anemone nemorosa, Hyacinthoides non-scripta, Lamiastrum galaeobdolon and Paris quadrifolia) was significantly related not only to distance from the nearest and second nearest edges in isolation, but significantly more strongly by the additive effect of distance from both edges in a single model. We discuss the implications of these findings from community ecology and conservation perspectives.