**1. Introduction**

Tidal current energy is a proven type of ocean renewable energy gaining more attention after the successful installation of the first grid-connected tidal turbine in Strangford Loch, Belfast [1]. The vertical-axis tidal current turbine, commonly known as the Darrieus turbine, is able to capture the tidal current from various flow directions without changing turbine facing as horizontal-axis turbine [2]. Scour induced by turbines is an important engineering problem in tidal power projects. Dey et al. [3] reported the damages of scour around offshore structures. The contraction of flow by the turbine rotor changes the flow patterns around the turbine and accelerates the flow between the turbine and seabed due to blockage effects [4]. The dominant feature of scour around turbines is the horseshoe vortex system when monopile foundation is installed. The strength of the horseshoe vortex system is heavily affected by sediment bed velocity [5]. However, the slipstream below turbine can be accelerated due to the compression effect of turbine rotor. The development of scour hole induced by tidal current turbine amplifies the flow speed compared to the bridge pier scour due to flow suppression in the narrow area between rotor tip and seabed. This phenomenon has been proved in paper [6]. Hence the tidal turbine-induced scour is more complex.

Researchers simplified the tidal turbine-induced scour to propose the bridge pier scour equations without consideration of the influences of rotating turbine [7]. Seabed scour around a tidal current turbine can be investigated based on the known knowledge in bridge pier scours. Pier-induced scour focuses on the maximum scour depth, temporal scour evolution, flow field around scour hole, and parametric study on the scour process [5]. These scour works can be found in Sumer and Fredsøe's book "The Mechanics of Scour in the Marine Environment" [8] and Whitehouse's book "Scour at Marine Structures" [9]. The empirical equations for scour around bridge piers can be used for scour prediction around turbine [6]. However, the influences of the turbine on the evolution of scour hole are still unknown.

The horseshoe vortex system and downflow near front side of turbine foundation are the main factors for seabed scour. The horseshoe vortex system greatly strengthens the seabed shear stress inside the vortex's action area [10]. The initial scour hole occurs at both sides of the foundation where the shear stress of the seabed is greater than the critical shear stress. During the scour process, the sediment particles inside the hole are washed away downstream under the exerted action from horseshoe vortex. Rate of sediment scour decreases with the continuous reduction of vortex induced forces in time and finally reached the equilibrium state. On the other hand, the wake–vortex system can expand the scour hole downstream. Large scour holes may develop downstream by the combined action of horse-shoe vortex system and wake–vortex system. The wake–vortex system acts like a shovel to remove the bed material which is then carried downstream by shedding eddies from the supporting piles [11]. Bianchini et al. [12] analyzed the wake structure of vertical axis tidal current turbine using CFD (computational fluid dynamic) method and suggested the main vertical structures were generated by the rotors. Chen and Lam [4] found that tip clearance between the turbine and seabed is the important parameter to determine its impact to the scour by CFD simulation using OpenFOAM. They found the presence of turbine rotor changed the boundary layer profile and results in the altering of horseshoe vortex formation. The axial velocity of flow increased by ~10% of the initial velocity, which can cause deeper scour hole. Axial flow acceleration occurs between the turbine and seabed, but limited discussion was made to unveil the scour mechanism and these works are focused on the horizontal-axis turbine induced scour. Wake patterns of turbines are important foundations to investigate flow near seabed under turbine rotor. Wang et al. [13] measured the slipstream wash in terms of field velocities in turbine wake and surrounding flow, and suggested the turbine should be installed at least one diameter height to decrease its impact on seabed. Myers and Bahaj [14] demonstrated that increased velocity deficits and turbulence exist along the rotor centerline. Pinon et al. [15] used "vortex method" to simulate the flow field of horizontal-axis tidal current turbine. The "vortex method" was a velocity-vorticity numerical implementation of the Navier–Stokes equations to compute an unsteady evolution of the turbine wake by some three-dimensional software. The results show that flow velocity reduced behind rotor and gradually recovered downstream. Lam et al. [16] proposed a wake equation to illustrate the wake distribution of horizontal-axis turbine based on axial momentum theory and the experimental data. Flow acceleration can be clearly observed close to the rotating blade tips. Ma et al. [17] proposed a wake model for vertical-axis tidal current turbine. In their model, the minimum velocity position in lateral wake distribution deviated from centerline at e fflux plane. The turbine's influence on the flow field reduced with the increase of the axial distance in turbine wake.

For the investigation of scour induced by tidal current turbine, Neill et al. [18] found the energy extraction from tidal current turbine reduced the overall bed level change; the asymmetry tidal region had a greater (20%) increase in sediment transport level compared to the tidal symmetry region. Hill et al. [19] carried out the experiments to examine the seabed scour around the foundation of horizontal axis tidal current turbine on an erodible channel. The works stated the scour development was accelerated compared to the scour around piles when rotor being installed upstream. Zhang [6] completed the numerical simulation to predict the horizontal-axis tidal current turbine scour evolution. Giles et al. [20] investigated the e ffect of foundation on the scour protection. Regarding hydrodynamic characteristics, Antheaume et al. [21] suggested that the e fficiency of a single hydraulic Darrieus

turbine could be greatly increased in farm arrangements. The results proved that the hydrodynamic performance of Darrieus-type tidal current turbine could a ffect the flow field significantly. Recently, Sun et al. [22] proposed an empirical model to predict equilibrium scour depth around Darrieus tidal turbines in water. The temporal evolution of scour depth is important for engineering projects of tidal current turbine. Current works provide empirical equations to predict the temporal evolution of scour around foundation of Darrieus-type tidal current turbine.
