It will also be necessary to vary the distance (D) incremented 

 between successive calculation points along a ray in order that the ray 

 may ^proach as close to the shoreline as possible. As an example, the 

 value of D could be assigned such that D - 0.5 when d/Lo>0.5 and D = d/LQ 

 when d/Lo<0.5. 



Wave Ray Representation 



At present, output must be plotted by hand; and, if there are a 

 large number of rays to be constructed, this can be a tedious and time- 

 consuming process. The use of an X-Y plotter, if adaptable to changing 

 grid scales, would be an ideal scheme for rapid plotting of wave rays. 

 Another scheme might involve the hand or machine plotting of the (numbered) 

 terminal points of the wave rays, along the shore or grid margin. 



Testing of Ray Constructions 



The absolute validity of the ray constructions (Figures 3-8) is 

 impossible to evaluate without a test in nature. It would be desirable 

 to test the ray constructions in an area where significant refraction of 

 relatively constant-period wave trains occur and where the bathymetry is 

 well known. The use of aerial photography would be an invaluable aid in 

 conducting such a test. 



It is possible to test ray constructions by comparison with analytic 

 solutions for wave rays passing over algebraically-described surfaces 

 (Pocinki, 1950). Griswold (1953), in testing the Griswold-Mehr program, 

 used both a straight uniformly-sloping beach and a parabolic bay and found 

 little variation between computed and analytic rays. However, such a 

 theoretical test is not sufficient reason for acceptance of a computer 

 program utilizing an interpolation scheme. In these theoretical tests, 

 an algebraically-described surface of velocity values is input to the 

 computer. The close agreement then noted between computed and analytic 

 rays is due to the fact that the given interpolation scheme can easily and 

 accurately represent the velocity surface when interpolating within any 

 grid cell. In order to adequately test such a computer program, an 

 irregular surface of velocity values must be provided. This necessity 

 supports the need for a test in nature, as mentioned in the previous 

 paragraph. 



CDNCLUSION 



The procedure described in this report, when fully improved using 

 the accompanying suggestions, will constitute a rapid and accurate method 

 of wave ray construction. At the present stage of development, however, 

 the procedure must be accepted with reservation. 



