Water chestnut, Trapa natans, is a non-native aquatic plant with floating rosettes of leaves that has established and continues to spread in northeastern North America. Growing in calm freshwater environments, T. natans can grow densely on water surfaces, transforming open waters into thick floating beds of vegetation. Replacement of open waters with T. natans can, in turn, shift fundamental ecosystem processes, with implications for water quality and biological communities. In the first chapter of this thesis, I present seasonal dissolved oxygen and nitrate concentrations of water samples collected from within and outside of three T. natans beds in New York State, US. These environmental characteristics are known to be impacted in other habitats invaded by T. natans and can prompt cascading changes in ecosystem dynamics. Dissolved oxygen concentrations within Trapa beds were consistently at hypoxic levels throughout summer months, and nitrate concentrations were extremely low relative to samples taken upstream of Trapa beds. I explored denitrification as a mechanism for nitrate removal in Trapa beds with laboratory mesocosm incubations using field collected water and sediment. Concentrations of N2, a denitrification product, were higher in sites within Trapa beds, but rates of changes in N2 did not differ in the presence of Trapa, suggesting that we did not fully capture denitrification dynamics in the experiment. Due to the ecological transformations that can occur when Trapa establishes in a waterbody, the plant is a target for environmental management. Despite intensive efforts, attempts to control the plant with chemical and mechanical means have largely been unsuccessful. Biological control is the only feasible control method capable of landscape-level and long-term control of the plant. Research on the biological control of Trapa with Galerucella birmanica, a specialist leaf-feeding beetle species was re-established in 2016. In the second chapter of this thesis, I present host-specificity data of G. birmanica, a required test protocol used to gauge risks to native species from any potential biological control insect. We tested 57 plant species in the presence of G. birmanica and recorded host-acceptance behaviors in no-choice and multiple choice tests. With these data, we test the prediction that non-target feeding damage is more likely on plant species phylogenetically related to water chestnut. We find evidence in support of that prediction, but find that the host-associations of congeneric beetles are a better predictor of G. birmanica feeding behavior, an uncommon result for host-specificity testing and biological control research, but not without precedent in plant-insect interaction science. In addition, we tested G. birmanica performance and preference on multiple populations of North American Trapa, including two distinct morphotypes, and find evidence that G. birmanica can utilize all populations tested, raising hopes for effective biological control of the diverse populations of Trapa in North America.