INTRAWAVE MODELING OF TURBULENT FLOW AND SEDIMENT TRANSPORT OVER RIPPLED BEDS LAURENT O. AMOUDRY 1 , PAUL S. BELL 1 , ALEJANDRO J. SOUZA 1 , PETER D. THORNE 1 1. National Oceanography Centre, Joseph Proudman Building, 6 Brownlow Street, Liverpool, L3 5DA, UK. laou@pol.ac.uk, psb@pol.ac.uk, ajso@pol.ac.uk, pdt@pol.ac.uk Abstract: We present numerical modeling of sediment transport over ripples in a large flume facility. We investigate flow parameterizations that could be readily implemented at a larger scale, which leads to intra-ripple length scales not being resolved. Different near-bed sediment parameterizations and the turbulence modeling approach are assessed based on model-data comparisons of flow velocity, mean turbulent kinetic energy, and mean suspended sediment concentration. While the flow velocities are found to be well reproduced by the model, wave-averaged turbulence and sediment transport are not, independent of the sediment erosion formulation implemented. We discuss how improved near-bed turbulence parameterizations can lead to reasonable predictions of wave-averaged sediment transport. Introduction Small bed forms such as ripples are ubiquitous in coastal environments and have a significant impact on the ambient flow and associated sediment transport. Vortex formation and ejection can be an important mechanism for sediment suspension over ripples, by which sediment is trapped in lee vortices and entrained in the water column when these vortices are lifted above the ripple at flow reversal (e.g., Thorne et al., 2003). Some models aim to resolve and reproduce the small-scale hydrodynamics due to wave flows over ripples (e.g., Zedler and Street, 2006; van der Werf et al., 2008), but the numerical domain usually remains limited to a few ripple wavelengths in the horizontal direction. Coastal models instead aim to investigate much larger domains and are thus not able to resolve the detailed dynamics associated with small-scale bed forms such as ripples. Instead, their effect needs to be parameterized in some way, which currently is typically done by considering them as roughness elements. Recent advances in measuring techniques allow us to obtain better representations of the physical processes involved (Davies and Thorne, 2008). Part of the difficulty is then how these processes, observed experimentally, should be parameterized in deterministic multi-dimensional coastal models, for which the concentration is the solution of a balance equation. In these models, the suspended sediment