The bow wave extends laterally about two to three radii away from 

 the test pile (Figs. 10, 12, 14, 15, 16, and 35). Hallermeier (1976] 

 noted that blockages of one-sixth or one-twelfth of the tank by a circu- 

 lar pile resulted in W(a) patterns not revealing different confinement 

 effects; the side slope in W(a) depends on F^ but not on Z. Con- 

 sidering these facts, Z <_ (1/6) seems a reasonable criterion for minor 

 confinement effects. This requirement is satisfied for every test sit- 

 uation except the 3x3 H-pile and the 6- inch-diameter circular pile in 

 the 96-foot tank. 



Figure 35, 



Photo from rear of disturbance at 2x2 H-pile during crest 

 flow in the 1 .5-foot-wide tank. 



The tests with the 6-inch circular pile in the 96-foot tank 

 (Z = 0.33) were discussed in Hallermeier (1976). Effects of pile con- 

 finement were negligible because linear wave scattering was the dominant 

 effect in those tests, rather than unidirectional peak crest flow. 

 According to the theoretical results by Spring (1973) for linear wave 

 scattering, situations with certain values of ka and b/L and 

 significantly large Z can be equivalent to an unconfined pile. The 

 tests with the 6-inch circular pile in the 96-foot tank (ka = 0.5, 

 b/L = 0.24) constitute one of these situations. 



2. Inaccuracies in Modeling the Prototype . 



The Reynolds number, R = 2Ua/v, characterizes the important flow 

 effects occurring at a circular pile due to fluid viscosity. Because 

 stagnation effects were of primary importance in the present tests with 

 surface-piercing piles, the laboratory situation was scaled to have the 

 same Froude number as important prototypes; the Reynolds number was lower 

 than in corresponding prototypes by a factor of the length scale to the 

 power 1.5. This factor is between 6 and 60 in the present tests (see 

 Table 3 for the length scales) . High ocean waves at typical field piles 

 give 10** < R < 10^, approximately. Even with the slight exaggeration in 

 pile diameter (Table 3) , the present laboratory tests were such that 

 103 < R < 10^. 



61 



