Nonlinear interaction of internal solitary-like waves in three-dimensional simulations

Kenji Shimizu, CSIRO Oceans and Atmosphere Flagship, Floreat, Australia and Keisuke Nakayama, Kobe University, Kobe, Japan
Abstract:
Internal solitary-like waves (ISWs) are nonlinear and nonhydrostatic waves with short wavelengths (~1 km) and short periods (~10 min). ISWs provide a pathway of energy from internal tides to shear instability, boundary layer turbulence, and mixing. They are also an essential factor for structural design in offshore engineering. For example, on the Australian Northwest Shelf, ISWs can induce stronger extreme near-bottom currents than tropical cyclones. However, the dynamics of ISWs have not been well understood partly because of the complexity of the physics and partly because of high computational demand for realistic three-dimensional simulations.

Here we investigate the nonlinear interaction of ISWs using three-dimensional MITgcm simulations in the presence of Coriolis effects. Previous studies of barotropic solitons show that solitons intersecting at oblique angles cause nonlinear interaction; however, such effects have not been well investigated for ISWs. To understand the relevance of such a process to ISWs in realistic conditions, we conduct idealized and realistic simulations. In the idealized simulations, we consider the interaction of radially spreading ISWs from two sources for two purposes: 1) to illustrate the nonlinear interaction of ISWs with full nonlinear and nonhydrostatic effects but in simplified settings; and 2) to compare the results against some previous theoretical work. In the realistic simulations, we use realistic bathymetry, stratification, and barotropic tides from the Australian Northwest Shelf. ISWs are generated by topographic generation of internal tides and subsequent nonlinear steepening. We show that the nonlinear interaction of ISWs is essential to understand the complicated interference pattern of internal wave fields in the realistic simulations. This in turn suggests the importance of the process to the interpretation of observations.