In the study, the thickness of wave moved-sediment layers was measured under regular wave conditions as well as the initial slope of sea bed with grain size of medium diameter 0.237mm and 0.128mm. The initial bottom slope ( tan£\) is 1/45, and nineteen wave conditions were studied. For each case, we analyzed the results after about 28800 waves were made. Sands are similar density and grain size but different colors. After wave action, the clear boundary between the two layers (two different colors) of sands will be mixed into gray color by wave-induced vortex.According to unmoved-layer(white) and gray-layer to get the total wave-moved sediment quantity. Finally, we get the relationship between the wave-moved sediment quantity per wave( q) and two parameters( £`b and £Kb ) which were established by Liao (2005, doctorate draft). The grain size does not affect £`b , and we get the q= 6.486*10^-5*£`b . The relationships between and ¡G 0.110mm gets q=5.103*10^-6£Kb , 0.128mm gets q=1.139*10^-5*£Kb, and 0.237mm gets q=1.933*10^-5*£Kb .

In the study, an innovative method is developed in 2-D wave flume tests to explore how much sand is set in motion by waves, and how wave-moved sediment is related to wave properties. Wave conditions on an initial sea bed slopes with grain size of about 0.1mm are varying during the experiments. Three initial bottom slopes of 1/30, 1/45, and 1/60 are analyzed in the study. The total number of waves acting is about 39,600 for each wave condition. The accumulated time of generated waves during the study is more than 1,280 hours; this is equivalent to about 2.45 million waves. The dark sands, along the observing window of the wave tank, of an initial sea bed are replaced by a slice column of white sands. The mixing caused by the waves moved dark and white sands together which generates a layer of grey sands that marks the interface of moved and unmoved white sands on the window. In some cases, three additional white sand columns are merged into the dark sand body perpendicular to the window to verify the uniformity of the moved layer in the wave crest direction. The quantity of the moved sediment is then computed and the wave-moved sediment by each wave is evaluated. Results show that the wave-moved sediment by each wave is linearly correlated to the wave breaking induced turbulent eddy viscosity, based on Prandtls mixing length model. The corresponding proportional coefficient reaches an asymptotic value as the number of acting waves is more than about 10,000. A Similar trend, but more diverse, is found when the wave-moved sediment is related to a movable parameter defined from the Shields number in which the Komars relation of bottom friction and slope is applied. However, the results indicate that the wave-moved sediment does not linearly correlate with the breaking wave power as proposed by most previous studies.