TWEL Kathi Borgmann Thesis

Invasion of riparian forests by exotic shrubs: Effects of landscape matrix and implications for breeding birds
Kathi L. Borgmann, MS
Advisor: Amanda Rodewald
Landscape-level studies of avian communities have moved beyond investigating simple area and isolation relationships to addressing the complex role of the landscape matrix. The landscape matrix can directly influence habitat availability, dispersal and colonization of organisms, metapopulation dynamics, nest predation, and species abundance and diversity patterns. The landscape matrix also can indirectly affect habitat suitability within a patch by mediating microclimatic conditions and floristic composition. In particular, invasion of forest patches by exotic plants may be closely tied to surrounding land uses, and exotic plants can alter floristic composition and structure. These changes also can affect predator-prey relationships, such as between nesting birds and nest predators, but our understanding of these interactions is limited. I used a multi-scale approach to determine (1) how changes within the landscape matrix influence the extent of exotic shrubs in riparian forests and (2) how exotic shrub species [honeysuckle (Lonicera spp) and rose (Rosa multiflora)] affect avian nest success.
Twenty sites were selected in mature riparian forests ranging from 63-200 m wide along five major waterways (ca. 20-40 m wide) in central Ohio. In July 2001, site width, volume of exotic shrubs, percent canopy cover, and aspect of forest edge (N/E or S/W) were measured at each site along 3 transects extending from the forest edge to river edge. Within 1 km of each site, coverage by urban development (i.e., paved surfaces and buildings) was determined from 1994 Thematic Mapper Imagery data. Sites occurred along a rural-urban gradient, ranging from < 1-47% urban land cover within 1 km. From May to August in 2001 and 2002, 231 open-cup nests within the understory (≤ 5 m tall) were monitored at a subset of 12 sites to estimate daily mortality rates. Nest substrate species, nest height, and distance from the forest edge were recorded for each nest, whereas, nest-placement and nest-patch characteristics (0.04 ha circular plot) were measured only for Northern Cardinal (Cardinalis cardinalis) nests (n = 68). An artificial nest experiment also was conducted using old Northern Cardinal and American Robin (Turdus migratorius) nests baited with one clay egg. Artificial nests (n = 79) were placed in honeysuckle, rose, and native substrates along a transect 50-75 m from the forest edge at two rural sites.
Potential factors promoting invasion by exotic shrubs were investigated using the information-theoretic approach and Akaike’s Information Criterion. I developed 19 a priori models consisting of local and landscape level characteristics likely to influence exotic shrub invasion. Exotic shrub volume was best described by the amount of urban land cover within 1 km of my sites. In particular, honeysuckle volume within riparian forests increased with increasing urbanization within the landscape. These results suggest that consideration of landscape matrix characteristics may improve our ability to predict or reduce invasion.
Both natural and artificial nests placed in exotic shrubs were 2-4 times more likely to fail than nests in native nest substrates, regardless of distance from the edge. Moreover, nests in exotic shrubs were especially vulnerable to predation in urban landscapes. These patterns may be explained by differences in nest-placement and nest-patch characteristics among nest substrates. In particular, nests in native substrates were nearly 1.5-2 m higher than nests in exotic shrubs and higher nests were less likely to fail than nests placed closer to the ground. Northern Cardinal nests in exotic shrubs had fewer and smaller diameter support branches, and were positioned closer to the central axis of the nest substrate. Nests in exotic shrubs also had 6-9 times more exotic shrub volume surrounding the nest (i.e., in the nest patch) than nests in native substrates. Together, these changes may facilitate a predator’s ability to locate nests. Of these characteristics, only the number of support branches was directly associated with nest fate, such that failed nests had fewer supporting branches. These findings demonstrate that exotic shrubs can reduce nest success of forest birds. However, the negative effect of exotic shrubs depended on the landscape matrix such that differences in daily mortality rates were most extreme in urban landscapes. This pattern combined with the fact that exotic shrubs were more pervasive in urban areas suggests that exotic shrubs may serve as an underlying mechanism of increased nest failure in fragmented landscapes.
This study is the first to (1) establish a link between the landscape matrix and invasion by exotic shrubs and (2) demonstrate that exotic shrubs impair reproductive success of forest-nesting birds while accounting for edge and landscape matrix effects. These findings suggest that restoring native shrub communities within riparian forests, especially in urban environments may improve avian nest success. An important step in restoring the native shrub community is the identification of land uses that can increase the risk of invasion.