Incipient motion of coarse particles under regular shoaling waves is studied. Experiments are performed to investigate the effects of bed fluid acceleration on coarse particle stability. By varying wave height, wave period and water depth combinations of similar peak orbital velocities and weak to strong intra-wave accelerations were created. The particles used in these experiments have two different sizes both of a cm order-of-magnitude. The data confirm that acceleration plays a role for the initiation of motion, since combinations of similar orbital velocity and varying acceleration magnitude resulted in no motion, some motion and motion as acceleration increased. Qualitatively we found that initiation of motion occurs at or is very close to the maximum shear stress due to the combined effects of drag/lift and acceleration as introduced by Nielsen and Callaghan (2003). However, quantitatively their formulation does not lead to convincing discrimination between motion and no motion. We expect this to be due to the assumption that the coefficients for drag/lift and acceleration in their formulation are taken equal. From literature and from plotting our data against the Keulegan-Carpenter number we expect that the coefficients strongly vary caused by flow separation effects.To arrive at a more convincing discrimination between motion and no-motion we introduced a new fluid acceleration descriptor for nonlinear shoaling waves. The combination of this descriptor with a Reynolds number Re g clearly delineates the regions with particle motion and without particle motion and has the potential to serve as a descriptor of the incipient motion of coarse particles under nonlinear regular waves.

We present a theoretical model for the wave-induced drift and horizontal deformation of an oil slick. The waves and the mean flow are coupled through the influence of the mean flow on the concentration of slick material, which in turn determines the damping rate of the waves and hence the transfer of momentum from the waves to the mean flow. We also briefly discuss a simplified version of the model that can be used when remote sensing data are available. With this simpler model the wave-induced forcing of the mean flow is obtained directly from observations of the wave field, hence knowledge of any specific slick properties is not required.

Vorticity and enstrophy production and dissipation are studied for both wave-averaged and wave-resolving ͑Boussinesq-type͒ models of wave-induced near shore circulation. Quadratic flow properties of fundamental importance for shallow-water turbulence, i.e., energy and enstrophy, whose sources/sinks are clearly identifiable by positive/negative-definite contributions in the appropriate transport equations, are taken as the most suitable indicators for assessing model performance in describing flows characterized by large-scale vortices. Two state-of-the-art models, SHORECIRC and FUNWAVE2D, have been evaluated in detail. Suitable transport equations for enstrophy are derived and analyzed to get a clear insight into the mechanisms of generation/ dissipation of this quantity in both models. Analytical results show that steep gradients of the total flow depth act as sinks as well as sources for vorticity and entrophy, similar to the results of Brocchini and Colombini ͓M. Brocchini and M. Colombini, Phys. Fluids 16, 2469 ͑2004͔͒. Predictive estimates have been given for the rate of change of circulation for waves breaking over a bar or breakwater and the vorticity source and sink terms have been numerically analyzed. The comparison between numerical results obtained using the two different circulation models reveals that while wave-resolving computations give well-structured rip currents, the wave-averaged model predicts less organized flows, given the different structure of the circulation forcing terms. The analysis of equivalent enstrophy-forcing terms characterizing the two models shows that they are all proportional to depth gradients in the case of wave-resolving models while their intensity is mainly due to the gradients of the wave-induced velocity for wave-averaged models. Energetic considerations are also given in support of the proposed vorticity/enstrophy generation mechanisms. Wave-averaged computations clearly show that, apart from bottom friction, the most intense dissipation mechanism is due to classic viscous effects ͓− T ٌ͑ ͒ 2 ͔ while depth gradients weakly contribute. Rather surprisingly this also occurs for the wave-resolving model.

The hydrodynamic circulation of a nearshore region with complex bathymetry is inves- tigated by means of a point-vortex approach similar, but more complete and suited to practical applications, to that of Kennedy (J. Fluid Mech. vol. 497, 2003, p. 225). The generation and dissipation of each single-point vortex are analysed in detail to obtain a complete description of the vortex dynamics. In particular, we clarify how the mechanism for the generation of breaking-wave-induced macrovortices (large-scale two-dimensional horizontal vortices) can be practically implemented and we discuss in detail the mechanism leading to the dissipation of the circulation assigned to each vortex. Available approximate relations for the rate of generation of bar vortices are placed in context and discussed in detail, and novel approximate relations for the shore vortex generation and for the vortex viscous dissipation are proposed, the latter largely improving the description of the point vortex dynamics. Results have been obtained using three ‘typical’ rip-current bathymetries for which we also test qualitatively and quantitatively the model comparing the vorticity dynamics with the results obtained by means of both wave-resolved and wave-averaged circulation models. A comparison of dynamically equivalent flow configurations shows that the dissipative point-vortex model solutions, neglecting any influence of the wave field, provide rip current velocities in good agreement with both types of numerical solution. A more complete description of the rip current system, not limited to the rip-neck region as given by Kennedy (2003) by mean of an inviscid model, has been achieved by including dissipative effects.