Open AccessESTIMATION OF BOUNDARY CONDITIONS FOR COASTAL MODELS
Author(s) -
S.K. Liu,
Jan J. Leendertse,
James A. Voogt
Publication year - 1974
Publication title -
proceedings of conference on coastal engineering/proceedings of ... conference on coastal engineering
Language(s) - English
Resource type - Journals
eISSN - 2156-1028
pISSN - 0589-087X
DOI - 10.9753/icce.v14.123
Subject(s) - interpolation (computer graphics) , transfer function , swell , boundary (topology) , function (biology) , computer science , grid , field (mathematics) , frequency response , noise (video) , geodesy , remote sensing , meteorology , geology , mathematics , telecommunications , geography , mathematical analysis , engineering , oceanography , frame (networking) , evolutionary biology , artificial intelligence , pure mathematics , electrical engineering , image (mathematics) , biology
In this study, frequency response and transfer function techniques are used together with cross-spectral and fast Fourier transform methods to determine the proper boundary values for computing the flow field of a coastal sea. Tide data containing considerable perturbations from swell and meteorological disturbances are analyzed. In computing the frequency response estimates, the effect of noise in the input is treated by a cancelling technique and by the choice of a reference station to evaluate the interdependencies among the other stations at the boundary. The usefulness of the network frequency response function is threefold: (1) future conditions can be simulated using observed water levels at any single location, (2) boundary information for models of different grid size can be obtained by interpolation, and (3) missing data at a given location can be estimated optimally using data at neighboring stations and the network response function. The paper discusses an example of such an application, the determination of a boundary of a two-dimensional model of Jamaica Bay, New York City, U.S.A.