Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Christopher R. Kincaid


The subestuaries in the northern reaches of Narragansett Bay (Rhode Island, USA) are susceptible to summertime water quality impairment. High nutrient loading of these systems leads to eutrophication and recurrent hypoxia, which cause ecosystem degradation and incidents of organism mass mortality. Management efforts to improve water quality include reductions in nutrient input to Narragansett Bay. We hypothesize that nutrient reductions may not sufficiently improve the health of severely impacted regions due to physical processes that contribute to water quality deterioration, including high vertical density stratification and low horizontal exchange. Dispersion depends on vertical turbulent mixing and advection. Circulation and dispersion are assessed using high-resolution observational and numerical modeling techniques. A realistic Regional Ocean Modeling System (ROMS) model for Narragansett Bay is validated by comparison with data and by doing a turbulence closure scheme assessment. It is found that the model provides an accurate representation of Narragansett Bay hydrodynamics when a k − ! scheme is implemented. High agreement exists between the model and data of tidal velocity and sea surface elevation, as well as subtidal velocity, temperature, and salinity. Future optimization of the model involves refinement of some forcing specifications. Observations of hydrodynamics in Greenwich Bay, one of the most severely impaired subestuaries of Narragansett Bay, and the adjacent Warwick Neck channel show the response of the subsystem to different forces. The data show that the wind exerts a dominant influence on the subtidal flow. Wind directed predominantly eastward improves exchange between the basins of Greenwich Bay relative to northward wind. Vertical turbulent mixing by tidal shear is generally low in Greenwich Bay indicating that bottom water may become isolated from the atmosphere for prolonged periods of time. This situation is conducive to hypoxia. Diurnal wind can cause an increase in vertical turbulent mixing and can force intratidal residuals by its interaction with the semidiurnal tide. The numerical model is used to carry out semi-idealized scenario experiments to determine the influence of specific forcing conditions on circulation, dispersion, and flushing of Greenwich Bay. Model experiments show that residence time depends on wind direction, since flushing occurs largely by wind-driven advective dispersion. The dominant sustained and diurnal wind conditions during the summer months lead to retention by weak flow and horizontal recirculation structures.