Dai and Kamel concluded that their tests gave inconclusive results 

 regarding scale effects in runup. However, when the data are compared 

 for specific wave conditions, some scale effects seem applicable to the 

 rubble-mound structure. Results are given in Table 22 where ratios of 

 runup for dg/H^ = 4.0 and 5.0 are combined and averaged for approximately 

 0.0007 < H^/gT 2 < 0.017. The three high values of runup for the large 

 scale at H^/gT 2 as 0.014 appeared questionable and not included in the 

 derivation of the table. 



Table 22. Scale effects for quarrystone rubble mound with core much 

 below SWL (cot 9 = 1.5) (after Dai and Kamel, 1969). 



V H i 



^ (large scale) 1 

 ^■(medium scale) 2 



R (large scale) 1 

 R (small scale) 3 



4.0 and 5.0 



1.06 



1.10 



1 Large scale: R e - 2.72 x 10 7 , where R g = > / gcQd s /v. 



2 Medium scale: R = 1.33 x 10 6 . 



e 



3 Small scale: R e = 4.69 x 10 5 . 



Dai and Kamel' s (1969) data give runup values considerably higher 

 than Hudson's (1958) data (approximately 30 percent higher at the same 

 scale), even when all of Hudson's data are included, yet the runup data 

 in the two studies appear consistent within each report. Thus, most of 

 the difference is apparently due to differences in experimental pro- 

 cedures rather than scale effect; some of the difference certainly is 

 in the difficulty of measuring runup on rubble slopes. However, Dai 

 and Kamel 's results for the large-scale rough quarrystone are sur- 

 prisingly similar to results of Saville (1962) who tested a large-scale, 

 three-layer, impermeable riprap structure with a 1 on 1.5 slope. Dai 

 and Kamel 's results also seem comparable with trends of Ahrens ' (1975a) 

 data (Figs. 40 and 41), although his H^/k p values were slightly larger 

 (H^/k r = 3.15 compared to Hyk r = 2.5 and 2.7 at dg/H^ = 5.0). 



Because the runup data of Dai and Kamel appear high in relation to 

 other testing, Hudson's runup values are recommended; however, because 

 Dai and Kamel 's runup data appear internally consistent, the scale 

 correction value derived from their data is adopted. Thus, the 

 6-percent correction (i.e., correction factor of 1.06) in Table 13 is 

 recommended for application to the steep structure slope parts of the 

 rubble-mound curves in Figures 25, 26, and 27 derived from Hudson's 

 data. 



Dai and Kamel (1969) also tested runup on quadripods. The rubble- 

 mound cross section was more conventional, with the top of the core 

 located approximately at the SWL. The quadripod tests were also per- 

 formed on rough and smooth armor unit types, as in the quarrystone 



15 



