ABSTRACT

Banks and bottoms of inland waterways are subjected to hydraulic actions such as stream flow and ship-induced waves. The contribution focuses on the effects of wave loading on cohesionless sandy soil. In the main part, results of soil column tests under wave loading are presented and the effects on bed stability are analyzed. In this context, the question of incipient sediment motion in the presence of liquefaction is raised. The paper closes with an outlook to further studies with a newly developed testing facility.

A broad research of wave-induced soil mechanical processes is based on analytic solutions with linear elastic soil behavior (e.g. Hsu und Jeng 1994). In the case of wave loading on quasi-saturated sandy soils, relevant seepage forces and hydraulic gradients may occur in the near-surface region within one wave period.

Comparative fine sand column tests with and without drained surface load illustrate different soil mechanical processes. Figure 1 shows the resulting hydraulic gradients close to the top of the soil sample. For the loaded test case, the hydraulic gradient is large whereas the free surface test case does not increase beyond a critical hydraulic gradient i c = 0.92. In this case, the test results give experimental evidence of non-linear and plastic soil behavior as well as of liquefaction during a few seconds within one wave cycle.

In the context of erosion stability analysis in the presence of wave-induced seepage conditions and momentary liquefaction, the transient and cyclic nature of the processes have to be accounted for. Erosion studies with upward seepage (e.g. Cao et al. 2016) analyze steady state conditions and are thus only partially transferable. Therefore a new testing facility, referred to as the river bed simulator, has been developed. The facility permits the analysis of soil water interactions due to combined waves and currents with pressure conditions at natural scale. It is a closed recirculating flume with a rectangular cross section. Pressures can be applied in the flume simulating up to 20 m water depth and modelling 1-dimensional, time dependent pressure fluctuations such as sinusoidal waves. Thus, it is suitable for a wide range of applications in river bed research.