Although invasive plants are recognized as a major ecological problem, little is known of the relative importance of plant community characteristics versus landscape context in determining invasibility of communities. We determined the relative importance of community and landscape features of 30 woodlots in influencing the invasion of Lonicera maackii. We sampled woodlots using the point-quarter method and calculated canopy openness and basal areas and densities of shrub, sapling and tree species, as well as woody species richness. We used aerial photos and ArcView GIS to calculate landscape parameters from the same woodlots using a buffer distance of 1500 m. We used logistic and linear regression analyses to determine the community and landscape factors that best explain L. maackii presence and density. We also tested whether woodlot invasion by L. maackii begins at woodlot edges.
Presence of L. maackii was significantly explained only by distance from the nearest town (logistic regression, p=0.017); woodlots nearer town were more likely to be invaded. Among invaded woodlots, density of L. maackii was positively related to the amount of edge in the landscape (partial R2=0.592) and negatively related to total tree basal area (partial R2=0.134), number of native woody species (partial R2=0.054), and sapling shade tolerance index (partial R2=0.054). Lonicera maackii in woodlot interiors were not younger than those on the perimeters, leading us to reject the edge-first colonization model of invasion.
Our findings reveal that landscape structure is of primary importance and community features of secondary importance in the invasion of L. maackii. This shrub is invading from multiple foci (towns) rather than an advancing front. Connectivity in the landscape (i.e. the number of corridors) did not promote invasion. However, edge habitat was important for invasion, probably due to increased propagule pressure. The community features associated with L. maackii invasion may be indicators of past disturbance.