Date of Completion

January 1980


Geological Survey|Engineering, Civil




Long Island Sound is a long, relatively narrow body of water situated between the urbanized shores of Long Island, New York, and Connecticut. The predominant cause of water movement in Long Island Sound is the semidiurnal lunar tide. The features of this tide induced motion may be qualitatively described as being those of a damped co-oscillating tidal system. That is, the tidal wave enters the eastern end of the Sound as a progressive wave and is reflected from the 'closed' western end, thus setting up a stationary wave. The dimensions of the Sound are such that the wave length of this standing wave is approximately four times the length of the Sound. The outcome is large tidal ranges and lessened current strengths at the head of the Sound, and small tidal ranges and considerable current strengths near the Race.^ The theoretical treatment of tidal motion in Long Island Sound in terms of damped co-oscillating tidal system is useful in achieving an overall understanding of the phenomena. However, providing the quantitative analyses which would be needed in connection with water quality studies, environmental impact statements, and coastal engineering construction projects requires more comprehensive mathematical models. To this end, a hierarchical sequence of numerical models has been developed.^ The first of these is a one-dimensional model in which the Sound is divided into 25 sections. Starting from prescribed 'no flow' boundary conditions at the western end, the tidal equations are solved section by section with the harmonic analysis method. By means of the one-dimensional model, information on appropriate bottom roughness coefficients for usage in subsequent more complex models was obtained.^ Following this, a two-dimensional, vertically averaged model of the Sound (the LIS model) was implemented. The LIS model employs the multioperation method solution technique. The LIS model implementation process involved a detailed evaluation of the approximation characteristics of the multioperation method. In particular, the stability and wave deformation properties of the scheme were examined. This was done to insure the choice of time-step and grid size for the LIS model leads to stable and accurate computations.^ To expedite verification, the first LIS model application was to a reduced area of the Sound generally corresponding to the broad central basin (the CB model). The results of the CB model verification run confirm the accuracy and stability of the model solution technique with the chosen time-step and grid size. The CB model was also used to study the effects of initial conditions and variable bottom roughness coefficients.^ The LIS model for the whole Sound was then utilized in numerical experiments designed to study the effects of initial conditions, horizontal diffusion of momentum, and density gradients. Finally, the LIS model was applied to a succession of hindcast-type simulations with forcing functions which ranged from mean tides to extreme storm surge tides. The LIS model computational results for all the simulations attempted are in good agreement with the available prototype data.^ In addition to the development of the numerical models, an attempt was made in the present work to discuss the tidal behavior of Long Island Sound in a thorough and complete fashion. It is hoped that this effort will have the consequence of this dissertation being a worthwhile reference source for future investigators. ^