Burial of Short Cylinders Induced by Scour and Bedforms under Waves plus Currents

Experiments were conducted to study burial of cylinders placed on a sandy bed under combined waves and currents. The burial of a finite-length cylinder was determined by local scour around the cylinder and by a more global process associated to the formation and evolution of ripples and sandwaves. Experimental tests covered cylinder Reynolds wave numbers and Keulegan-Carpenter numbers within the ranges 10 ≤ Re ≤ 9.1x10 and 2 ≤ KC ≤ 26, respectively. Observations indicate that existing semiempirical formulae for prediction of equilibrium burial depth for pipelines are not completely suitable for the case of a cylinder having a finite length. In these experiments burial depth shows a strong dependency on both the Shields parameter (θ) and the Keulegan-Carpenter number (KC) combined. Introduction The dynamics of the interaction between objects and movable beds in a wavecurrent combined flow is very relevant having, in nature, an intrinsic strong complexity at the same time. Multiple applications, from military to civil can be associated to this particular phenomenon (Voropayev et al., 1998, 2003; Bennett, 2000). Cokgor (2002) studied the behavior of the lift and drag coefficients around partially buried cylinders on a rigid bottom exposed to combined waves and currents for low Keulegan-Carpenter (KC) numbers (here, D T U KC w m / = , where Um is the maximum value of the undisturbed orbital velocity at the bed, Tw is the wave period, and D is the cylinder diameter). Both coefficients, lift and drag, seem to be governed primarily by KC for both, waves alone and combined flow. Other studies on combined flows, in the laboratory and in the field, have also shown that both the ultimate scour depth and the time scale of the self-burial process of short cylinders are primarily functions of the Keulegan-Carpenter number (KC) and the Shields parameterθ (Bayram and Larson, 1998; Sumer and Fredsoe, 2001). The same parameters also affect the scour around vertical piles in the field subject to combined waves and currents (Sumer and Fredsoe, 2001). Additionally, they showed that the scour depth increases as the current velocity increases and is essentially unaffected by the direction of wave propagation. Earlier, Voropayev et al. (1999) studied the dynamics of disk-shaped objects placed on a sandy Copyright ASCE 2005 EWRI 2005 2 bottom under an oscillatory flow and showed that the burial depth of the cobble does not depend on the specimen’s density. Similar results were obtained for the case of a cylinder horizontally placed over a sandy bed (Sumer and Fredsoe, 2001). More recently, Voropayev et al. (2003) have investigated burial of finite-length cylinders over a sandy floor with a sloped bottom under shoaling waves. Such study showed that the final burial depth depends of both the KC number and the Shields parameter while the length of the scour hole depends mostly on the KC number. Additionally, Voropayev et al. (2003) showed that the cylinder may undergo periodical burial depending on the ratio of cylinder diameter to the height of migrating ripples. Similar observations were reported later by Catano and Garcia (2004) for the case of a zero sloped bottom under combined waves and currents. Systematic experimental research coupled with mathematical modeling that can incorporate the interaction of environmental forces on a cylinder under coexisting flows, considering different properties for both specimen and sediment are still needed (Friedrichs, 2001). Burial and scour processes induced by the hydrodynamic forces acting over the cylinder, coupled with fluidization of muddy bottoms under strong storms, are of particular interest since they convey a more rapid burial process (Admiral and Garcia, 2000; Sumer and Fredsoe, 2002). In this study, laboratory results were analyzed and a new predictive formula relating the burial depth to the KeuleganCarpenter number and the Shields parameter has been developed through basic curve fitting. The main goal of the present work is to contribute to enhance the current understanding of the processes responsible for the burial of finite-length cylinders under the action of waves and currents In this paper, we present results of experiments on the cylinder-burial process for several combinations of wave amplitude, wave length, and period, and constant water depth set to 56 cm. Three dimensional mapping of the evolution of bedforms and scour patterns around cylindrical models under waves alone and combined flow show the influence of both local scour and global bedform evolution on the final relative burial depth and net vertical displacement of the object. Sandwaves are shown to control the vertical displacement of the specimen after they form. It is seen that the vertical displacement of the cylinder does not longer coincide with the burial depth. This is primarily due to the more or less exposition of the specimen to the flow action according to its relative position over the bedforms as they migrate. Experimental setup and test conditions Experiments were conducted in a tilting water flume that is 49 m long, 1.83 m wide, and 1.22 m deep (Figure 1). The longitudinal slope of the flume was set to zero. The floor of the flume is made out of steel while the side walls are made of Plexiglas. A current was achieved by recirculating water in the flume with the help of a constant head tank and a pump. A piston type wavemaker was used to generate regular waves. The wavemaker consists of a flat, vertical steel plate equipped with wooden wipers to prevent leakage around the plate, and is driven by a hydraulic ram that is capable of producing either periodic or random motion. The paddle of the wavemaker moves according to a prescribed sinusoidal motion as given by, ) 2 ( t f Sin S S w o π = (1) Copyright ASCE 2005 EWRI 2005 3 Where So is the paddle stroke that ranges from 0 to 30.48 cm (half the total distance it travels).