layers such that their number, complexity, and associated construction quality 

 control requirements are minimized. Unfortunately, breakwater specialists do 

 not agree on a precise filter criterion for rubble-mound breakwater underlay- 

 ers (Jensen, Graveson, and Kirkegaard 1983), and physical modeling of scour of 

 core material is complicated by scale effects (Hedges 1984). 



38. A densely packed core can reflect a significant amount of wave en- 

 ergy back through the underlayers and reduce the stability of the armor or in- 

 crease scour near the toe of the breakwater. A core and underlayer system 

 that reduces wave energy through turbulence and frictional loss is preferred 

 to a more reflective system. A core that is too permeable can transmit waves 

 as much as 80 percent of the incident wave height (Kogami 1978), and it may 

 pass littoral materials. Some useful experiments with wave transmission 

 through porous rubble-mound breakwaters were performed by Madsen and White 

 (1976) and continued by Seelig (1980a). Their methods are helpful in predict- 

 ing wave transmission characteristics and will be discussed again later in 

 this report. The effects of variations in permeability are discussed further 

 in Bruun ( 1985). 



Head and elbow construction 



39. The inevitable lateral flow across round heads and elbows and the 

 reduced interlocking and compaction in these areas complicate just about every 

 facet of breakwater design. Practical methods to deal with these compli- 

 cations consist primarily of conservative adjustments to analyses as applied 

 to sections of the breakwater trunk. This type of adjustment has limited 

 confidence as evidenced by the frequent need to repair heads and elbows of 

 conventionally designed rubble-mound breakwaters. Model testing in a three- 

 dimensional wave basin is at present the only reliable means of improving this 

 confidence. This is particularly important with slender concrete units (such 

 as dolosse), which may have little or no increased stability over rock or 

 bulky units in lateral flows (Burcharth and Thompson 1982). It is this fact 

 that has caused some investigators to question the reliability of the Hudson 

 formula and the associated K^ factors published in the SPM (1984) for use in 

 head or elbow design (Angerschou et al . 1983). The detail design of heads and 

 elbows will very likely remain a highly subjective and empirical process for 

 some time. 



Toe construction 



40. A number of practical problems related to the toe of rubble-mound 



25 



