the location of intermediate- and long-period wave gages can be obtained 

 from mathematical models where the modes of oscillation of the harbor 

 areas for different wave periods are plotted by digital computer 

 techniques. 



The resonant wave periods of vessels that frequent a harbor can be 

 obtained from model studies of individual vessels in which the vessel's 

 shape, size, mass distribution, and weight, and the characteristics of 

 the mooring- line assembly, are reproduced to acquire dynamic similarity, 

 model-to-prototype. This information should be complemented by prototype 

 wave data correlated with actual vessel observations with respect to the 

 severity of surge oscillations. Both types of data are difficult, expen- 

 sive, and time consuming to obtain, and sufficient data of these types 

 can seldom be acquired to allow accurate analysis of harbor wave action 

 problems. Without this information, the harbor wave action model will 

 reliabily show which of several plans is the best for intermediate- and 

 long-period wave action in the different basins of the harbor, but it 

 cannot ensure that the best of several plans will be satisfactory for 

 the surge and sway of vessels moored elastically at docks within the 

 harbor. This situation is also true for short -period waves in small- 

 craft harbors, at least in principle. However, experience has shown 

 that small-craft harbors are usually satisfactory, with respect to the 

 breaking of lines and boat damage, if the wave heights in the harbor 

 basins can be reduced to about 1 foot or less. Raichlen (1968) has 

 stated that the motions of small craft may be reduced by the proper 

 design of mooring systems and the imposition of certain mooring restric- 

 tions, rather than the drastic and expensive reduction of wave energy 

 that is allowed to enter the harbor basins. 



The tsunami data needed for the conduct of a harbor wave action model 

 are essentially the same as those needed when the waves are short- or 

 intermediate-period waves. Design and operation of the model and the 

 interpretation of model test results require that the frequency of occur- 

 rence of tsunamis with different periods and wave heights, from the dif- 

 ferent directions of approach, be available for locations in the ocean 

 corresponding to' the shallow-water positions of the wave generator. How- 

 ever, tsunamis in the open ocean are so long and their heights so small 

 that deepwater tsunami data are exceedingly difficult to obtain and such 

 data, sufficient for harbor model studies, are not presently available. 

 Since the tsunami problem areas of the United States are limited to the 

 west coast and to some coasts of Hawaii, considerable data are available 

 for these areas. These data consist of the frequency of occurrence for 

 tsunamis of different heights and periods as measured in bays, coastal 

 zones, and harbors in the problem areas (Cox, 1964; Wiegel, 1965). The 

 directions of the wave fronts at the ocean limits of a proposed model 

 are determined by wave refraction studies in which the wave rays are 

 projected from the earthquake epicenters where the major tsunamis origi- 

 nate. A digital computer program, written specifically for tsunami re- 

 fraction, has been used in a study concerning the design of a proposed 

 tsunami model of Crescent City Harbor, California (Keulegan, Harrison, 

 and Mathews, 1969) . 



214 



