where T^ is the primary period, Lj> the length of the inlet, and 

 d the average depth of the inlet. Values of length, depth, width, 

 period, and relative intensity of secondary oscillations of the water 

 level, as given by Murty, Wigen, and Chawla, for inlets on the coast of 

 Alaska and British Columbia, and for Puget Sound, are given in Table 4. 

 These values are only approximate because variations in inlet cross 

 section, restricted entrances, and the effects of branched inlets are 

 not considered. 



Referring to the work of Nakano (1932) which showed secondary undu- 

 lations to be proportional to the length of an inlet, L^, and inversely 

 proportional to the width, B, and to d^ /2 , Murty, Wigen, and Chawla 

 (1975) proposed that the relative intensity, I, of the secondary undu- 

 lations could be given as 



I = 



Bd 3/2 



a 



(294) 



Values of I for inlets of Alaska, British Columbia, and for Puget Sound 

 are shown in Table 4. Inlets with higher relative intensities, I, would 

 be expected to excite larger amplitudes of oscillation. As indicated by 

 Murty, Wigen, and Chawla, some bays which have small ratios of L^/B have 

 large secondary oscillations. They point out that equations (293) and 

 (294) are based on a one-dimensional theory which is not valid for low 

 ratios of L^/B, and that transverse motion is important in these cases. 



Fukuuchi and Ito (1966) consider a tsunami passing from a larger bay 

 or inlet into a smaller inlet. Where the larger inlet has a width B 1} 

 and the width narrows to a width B 2 in the smaller inlet (see Fig. 40), 

 they give the amplitude a 2 at the head of the smaller inlet as 



B, 



2/2 a, 3- 



.,,.,,_ 



\B 2 



1/2 



(295) 



where a, is the incident tsunami amplitude in the larger inlet, T the 

 period of the tsunami, and Tj the period of the smaller inlet as given 

 by equation (293). The maximum amplitude a 2 will occur when T-j/T = 

 1, 3, 5. . . while the minimum amplitude would be at Tj/T = 2, 4, 6. . . 

 etc. Equation (295) would predict very high values of a 2 /a I where BjyB 2 

 is large. This is not consistent with the work of other investigators. 



Ippen, Raichlen, and Sullivan (1962) carried out a hydraulic model 

 investigation of an inlet connected to an "infinite ocean." The ocean 

 was simulated in a wave basin, using wave absorbers to minimize reflected 



132 



