Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.



Degree Program

Biology, PhD

Committee Chair

Carreiro, Margaret

Committee Co-Chair (if applicable)

Barton, Christopher

Committee Member

Barton, Christopher

Committee Member

Cobbs, Gary

Committee Member

Emery, Sarah

Committee Member

Zipperer, Wayne

Author's Keywords

invasive species; urban wetlands; honeysuckle; Lonicera maackii


Globally, wetlands are known for providing important ecosystem services that enhance the quality of human life and regulating global biogeochemical cycles. Despite the wide recognition of their value, temperate forested wetlands are the least protected type of ecosystem world-wide, and are threatened by human activities such as logging and development. The ecology of forested wetlands remaining in urbanized areas is impacted by a multitude of anthropogenic threats, including fragmentation (which decreases the amount of interior habitat and increases edge habitat), hydrologic modification (ditching and draining of wetlands) and the incursion of invasive species (which are frequently introduced by human activities). In the first study presented in this dissertation (Chapter 2), I examined how woody plant communities of three urban wetlands—the dominant biota of these ecosystems—changed along edge-to-interior and hydrologic gradients. Detailed measurements of elevation, surface water levels, and ground water levels were made to estimate the number of days each transect was flooded. The three study forests were surveyed in transects along edge-to-interior gradients (0-60 meters from the forest edge) and were found to exhibit a gradient of flooding (measured as the number of days flooded in sampled areas). Ordination with non-metric multidimensional scaling (NMS)—using Importance Values (IVs) of adult trees, saplings, tree seedlings, shrubs and vines in transects at 1, 5, 10, 30 and 60 m from the edge—was used to see if patterns in the woody plant community related to distance from edge, number of days flooded and other environmental variables. Distance from the forest edge and number of days flooded were the two variables shown to be most correlated with ordination axes generated from the species matrix (r2≥0.15), and each was associated with a different axis. The shrub community was most indicative of community differences along the hydrologic gradient (Axis 1); Lindera benzoin (a facultative shrub) and Cornus foemina (a facultative wetland shrub) were the species most associated with drier and wetter transects, respectively. The invasive shrub species, Lonicera maackii, was present at all sites, but more important at the two driest sites. Because the relative elevation (within a transect) of L. maackii plants increased with the number of days flooded, it appeared that higher elevation microsites may have provided refugia for the establishment and/or persistence of this invasive species in flooded areas. Fraxinus pennsylvanica (green ash) was a major component of the tree canopy species at all sites, and the imminent threat posed by the emerald ash borer (Agrilus planipennis) will most certainly result in the formation of large canopy gaps. Following the findings of Chapter 2, two manipulated experiments were conducted to determine the degree to which L. maackii was impacted by flooding at early life history stages. In the first experiment (Chapter 3), seedlings of L. maackii were subjected to factorial treatment combinations of simulated canopy cover (2 levels), flooding duration (3 levels) and flooding depth (4 levels) in experimental mesocosms for 40 days. Flooding events were followed by a recovery period. Inundation and saturation of seedlings for 10 to 17 days had a markedly negative effect on honeysuckle seedlings as shown by slow vertical growth or dieback and loss of leaves. No seedlings inundated for 17 days survived. In the second experiment (Chapter 4), I measured the germinability and mean germination time (MGT) of L. maackii seeds stratified under cool-moist, cool-saturated, or cool-inundated conditions and transitioned to a second set of conditions intended to mimic natural drawdown or flooding events. A constantly moist control was included. Only MGT, but not germinability, of moist-to-inundated transitions (a simulated flooding treatment) differed significantly from moist controls. Seeds subjected to simulated drawdown conditions did not differ from moist controls, indicating that seeds of L. maackii seeds can remain viable in flooded areas, and that temporary, winter flooding of wetland microsites may not impose a major limitation on L. maackii germination. The results of these three studies underscore the threats posed by invasive plant and insect species to the native woody plant community of these study systems. Evidence from the field and mesocosm studies suggest that flooding does pose a barrier to L. maackii colonization, but that higher elevation “safe sites” (with shorter flooding durations and/or depths) may enhance the ability of this species to invade wetter areas. At the time of field data collection (2010), emerald ash borer was largely absent from these forests, but has now caused widespread tree death. Because Lonicera maackii is known to benefit from increased light availability, the opening of the tree canopy could have a beneficial effect on the spread of this invader within wetland sites.