THE INFLUENCE OF A CROWN WALL ON WAVE OVERTOPPING OVER RUBBLE MOUND STRUCTURES

The crest of a breakwater is often equipped with a crown wall with multiple functions such as a border of the pedestrian walkway, support of the rubble mound or concrete armour units, and reduction of the wave overtopping over the structure in storm conditions. However, the knowledge on how a crown wall influences the overtopping discharge is yet still limited. EurOtop (2016) advises to take the maximum of the crest freeboard Rc and the armour freeboard Ac in case a crown wall is present, and to use the average between Ac and Rc when no crown wall is present. This advice is based on expert judgement rather than on experimental research. The goal of the present work is to carry out experimental research and confirm or update the advice on overtopping over a crown wall for rubble mound structures.


MOTIVATION AND AIM
The crest of a breakwater is often equipped with a crown wall with multiple functions such as a border of the pedestrian walkway, support of the rubble mound or concrete armour units, and reduction of the wave overtopping over the structure in storm conditions.However, the knowledge on how a crown wall influences the overtopping discharge is yet still limited.EurOtop (2016) advises to take the maximum of the crest freeboard Rc and the armour freeboard Ac in case a crown wall is present, and to use the average between Ac and Rc when no crown wall is present.This advice is based on expert judgement rather than on experimental research.The goal of the present work is to carry out experimental research and confirm or update the advice on overtopping over a crown wall for rubble mound structures.
MODEL SET UP Two test campaigns have been carried out in the two wave flumes of Ghent University: 1:20 model tests in a larger wave flume (30m x 1m x 1.20m), and 1:50 model tests in a smaller wave flume (15m x 0.35m x 0.60m).A breakwater according to Figure 1 is built on scale in both flumes.The crown wall height hwall is varied (case 1: Ac = Rc and hwall = 0, case 2: Ac < Rc or hwall > 0 (situations +W1 and +W2 in Figure 1), case 3: Ac > Rc or hwall < 0 (situations -W1 and -W2 in Figure 1)).Also the crest width in front of the crown wall is varied (Gc = 1P, 3P, 5P with P = Dn50).The water level, the wave height and the wave period are also varied in the test matrix.For the most common breakwater geometry, with a crown wall above the armour crest (case 2: hwall > 0 è Rc > Ac), test results show that EurOtop (2016)'s advice (use the maximum of Rc and Ac) underestimates the measured overtopping discharge.
For case 3 with the wall below the rubble mound crest (hwall < 0 è Rc < Ac), using the maximum of both values Ac and Rc is underestimating the actual overtopping discharge.Using the average of Rc and Ac as suggested by EurOtop (2016) gives more promising results but slightly overestimates the results.
Besides the wall height, also a variation of the crest width and the wave period have a rather strong influence on the overtopping discharge: higher periods and/or shorter crests result into more overtopping.However, the advice in EurOtop ( 2016) does not take a variation of Gc or Tp into account for its reduction factor g*.
The research performed (326 tests) has resulted in a more consistent advice for all different variations.This lead to an influence factor to account for the wall height (gv), the crest width (gcrest) and the combined effect (gcrest_v), to replace g* in Eq. (1): 9 = exp 0.3131 (3)  ?@ABC_9 =  9 . ?@ABC (4) Eq. ( 2) to (4) have been derived for a parameter range as given in Table 1. Figure 3 shows the final result after including the different influences from Eq. ( 2) to (4) in Eq. ( 1).results stay well within the +/-5% boundary lines.The analysis of Eq. ( 2) to (4) will be explained in full detail in the presentation and the full paper.

Figure 2 .
Figure 2. Overtopping results on the reference situation in the large wave flume.Dimensionless freeboard on the horizontal axis, with inclusion of the roughness factor gf, and dimensionless overtopping discharge on the vertical axis.

Figure 3 .
Figure 3. Complete dataset corrected by means of the new reduction factor gcrest_v (which is a combination of the influence of the crest width and the crown wall height) found in this research.