in the structure is accomplished with a variety of manual and automatic tech- 

 niques designed to simulate the realities of full-scale field placement. The 

 placement tolerances of model rubble-mound breakwaters often are smaller than 

 their prototype counterparts, however. The hydraulic characteristics of un- 

 derlayers and the core can be especially difficult to model by the Reynolds 

 criteria since the shape of the units, their surface friction (particularly 

 between layers), and the shape of the interstices are critical. Erosion of 

 fine foundation material at the toe of breakwaters is also a problem, gener- 

 ally yielding only qualitative conclusions. An account of these and other 

 scale effects is necessary for reliable interpretation of model results 

 (Jensen and Klinting 1983). 



47. Scale modeling operational procedures associated with the design 

 of rubble-mound breakwaters can be classified in three general groups: 

 (a) cross-section design tests run in two-dimensional flumes (Figure 8), 



Figure 8. Scale model testing in 

 two-dimensional wave flume 



(b) tests of heads, elbows, transitions, offshore hydrographic effects, and 

 oblique waves in three-dimensional wave basins, and (c) tests of breakwaters 

 at various stages of construction (in either flumes or wave basins). The 

 first of these is of primary interest to discussions of analytical optimiza- 

 tion, since it is this type of scale model testing which has generated most of 

 the analytical relations used by designers. These tests of proposed cross- 

 section designs are intended to verify the predictions of analytical proce- 

 dures and to refine detailed features of the cross section. They are often 



30 



