Keywords

• Coherent turbulence structures;
• Sediment resuspension;
• Wavelets

1. Introduction

The suspension of sediment in turbulent flows is a complex case of fluid-particle interaction, governed by shear stresses (momentum exchanges) at the bed and within the benthic boundary layer (BBL). Defining the physical processes which dictate the resuspension of sediments in coastal and estuarine settings is fundamental for accurate predictions of bed morphology evolution (van Rijn et al., 2007), and has profound implications for the biogeochemical processes that shape their local ecology (Thompson et al., 2011). It is also a prerequisite to quantifying erosion and deposition trends, and hence guiding engineering applications such as beach nourishment, defence schemes against erosion and flooding, maintenance of marine infrastructure and waterways, and aggregate dredging. There is a genuine need for better, robust models of suspended sediment transport in the coastal zone (Aagaard and Jensen, 2013). In a vision paper on future research needs in coastal dynamics, van Rijn et al., (2013) highlighted the pressing need for research to support such models, focusing in particular on sand transport in the shoreface (non–breaking waves), surf and swash zones; employing field and controlled laboratory experiments.

The mobilisation of sediments in the nearshore and shoreface is dominated by wave-induced bed shear stresses in moderate and stormy conditions (Thompson et al., 2012). The vertical structure of sediment flux components on the shoreface and in the inner surf zone, as well as the dynamics of sediment transport under shoaling waves in the nearshore, are both considered to be insufficiently understood (van Rijn et al., 2013). This requires prioritising research with reference to coherent flow structures and the intermittent stirring of sediments by breaking and shoaling waves, and the time-history effects of suspended sediments under irregular wave conditions [ibid.]. Understanding the spatial, temporal, and frequency characteristics of sediment suspension events in relation to turbulent fluctuations, both in structural form and in temporal distribution, is an important step towards providing a more satisfactory conceptual model for describing suspended sediment transport.

The role played by bed–generated coherent eddy structures in entraining and transporting sediment particles is widely acknowledged, yet the exact mechanism is still unclear (Dey et al., 2012 and Ji et al., 2013). Coherent turbulence structures have been defined, albeit reluctantly, as “connected turbulent fluid masses with instantaneously phase-correlated vorticity over their spatial extent” (Hussain, 1983 and Hussain, 1986). Fiedler (1988) added several criteria to the definition, namely; composite scales, recurrent patterns (lifespan longer than the passage time of the structure), high organisation and quasi-periodic appearance. Besides the “conventional” bursting events which describe the intermittent, energetic process resulting from the passage of near-wall vortices as perceived by passive markers and/or visualisation studies (Schoppa and Hussain, 2002); one may identify vortices induced by wave breaking (Aagaard and Hughes, 2010), or by flow separation from vortex ripples upon reversal in an oscillatory flow, i.e. vortex entrainment/shedding ( Amoudry et al., 2013). Vortex shedding from bedforms in wave dominated flows was first reported by Bagnold and Taylor (1946). As flow reverses over steep two-dimensional bedforms, the benthic boundary layer can separate from the bed, trapping sediments and ejecting them higher into the flow ( O’Hara Murray et al., 2011). This is in essence a repeatable and hence coherent convective entrainment process ( Nielsen, 1992) that is observed in both regular (at half cycle) and irregular wave conditions ( O’Hara Murray et al., 2012, O’Hara Murray et al., 2011 and Thorne et al., 2003). Where vortex pairs may develop, sediments may be violently ejected much higher (several orders of a ripple height) than classically described ( Williams et al., 2007).

The intermittent transfer of momentum by coherent structures of turbulence is manifest by velocity fluctuations, and is linked to short-term variations in near-bed stresses (Heathershaw, 1974 and Laufer, 1975). This is evident in the turbulent “bursting” process (Kline et al., 1967, Offen and Kline, 1974 and Offen and Kline, 1975), which is a critical mechanism for production of turbulent kinetic energy (Dey et al., 2012 and Schoppa and Hussain, 2002). Turbulent bursting may be explained by the advection of spatially distributed vortices and structural features past a fixed point of measurement (Robinson, 1991), although this may not detect how such vortices evolve in time (Schoppa and Hussain, 2002). The largest contributions to stress often occur through ejecting or sweeping motions (Soulsby, 1983). Typically, ejections are associated with entrainment of mass (sediment particles) into suspension, while sweeps are effective at transporting bedload (Cao, 1997, Dyer and Soulsby, 1988, Heathershaw, 1979, Keylock, 2007, Soulsby, 1983 and Yuan et al., 2009). While ejections and sweeps reportedly occur in relatively equal proportions near the bed, the former type dominates higher in the water column (Cellino and Lemmin, 2004). Suspension of sediments is often related to large scale turbulence structures associated with clusters of ejections (Bennett et al., 1998 and Kawanisi and Yokosi, 1993).