Nearshore sandbar migration predicted by an eddy‐diffusive boundary layer model

We simulated the erosion and accretion of a natural beach using a wave‐resolving eddy‐diffusive model of water and suspended sediment motion in the bottom boundary layer. Nonlinear advection was included in this one‐dimensional (vertical profile) model by assuming that waves propagated almost without change of form. Flows were forced by fluctuating pressure gradients chosen to reproduce the velocity time series measured during the Duck94 field experiment. The cross‐shore flux of suspended sediment beneath each field‐deployed current meter was estimated, and beach erosion (accretion) was calculated from the divergence (convergence) of this flux. Horizontal pressure forces on sediment particles were neglected. The model successfully predicted two bar migration events (one shoreward bar migration and one seaward) but failed to predict a third (seaward migration) event. Simulated seaward sediment transport was due to seaward mean currents. Simulated shoreward sediment transport was due to covariance between wave‐frequency fluctuations in velocity and sediment concentration and was mostly confined to the wave boundary layer. Predicted seaward (shoreward) bar migration was driven by a maximum in the current‐generated (wave‐generated) flux over the sandbar. A wave‐generated downward flux of shoreward momentum into the wave boundary layer contributed to shoreward sediment transport and often had a local maximum over the bar crest. Second‐order nonlinear advection of sediment, mostly representing shoreward advection by the Stokes drift, also often had a local maximum over the bar crest. Together, wave‐generated momentum fluxes and the Stokes drift substantially increased shoreward transport and were essential to predictions of shoreward bar migration.