4:30 PM - 4:45 PM
[ACG40-10] Development of an air-sea two phase numerical flow model for wind-wave simulation
Keywords:wind wave, gas-liquid two-phase flow
The wind blowing over the sea surface generates small waves, and the waves continue to develop while absorbing wind energy and gradually grow into larger waves. These waves, called wind waves, play a significant role in local and global weather and climate phenomena by affecting the exchange of momentum, heat, and materials (such as CO2) across the surface. Quantitative understanding of the various interactions between the atmosphere and ocean caused by wind waves enable accurate prediction of wave and weather phenomena.
Wind waves have been studied extensively over the years by observational, experimental, or theoretical methods. For example, Miles (1957) and his improved theory (1993) are one of the most common theories of wind wave development by momentum transport from the wind, which agree well with the measured data on development rates compiled by Plant (1982). However, many issues remain, such as the large scatter of observed data in in-situ and laboratory experiments, and the many assumptions made in the theory due to unknown atmospheric turbulence. Another remaining issues is the wave breaking whose effect cannot be precisely considered. Numerical simulations with an air-sea two-phase numerical model capable of representing wave breaking would be one approach to greatly advance these issues.
In this study, we developed a two-phase flow model that numerically solves the weakly compressible Navier-Stokes equation with reference to Matsushita and Aoki (2021), to understand the interaction between the atmosphere and ocean in wind waves with breaking waves. This method solves the incompressible fluid as a compressible fluid with a low Mach number, thus eliminating the need to solve the Poisson equation with poor convergence obtained from the incompressible condition and allowing for a fully explicit calculation. A coupled phase-field and level-set method was used to represent the gas and liquid phases in the calculations.
Some benchmark experiments were conducted to validate the model. In the 2-D dam-breaking experiment, observed time-varying shape of the collapsing water column was in good agreement with those of Martin and Moyce’s laboratory experiment (1952). We also performed the Rayleigh-Taylor instability experiment to test the model and found that the simulated time evolution of interface shapes is similar to those of previous numerical studies. In the presentation, we will report numerical accuracies and costs of our model as well as remaining issues to be solved.
Wind waves have been studied extensively over the years by observational, experimental, or theoretical methods. For example, Miles (1957) and his improved theory (1993) are one of the most common theories of wind wave development by momentum transport from the wind, which agree well with the measured data on development rates compiled by Plant (1982). However, many issues remain, such as the large scatter of observed data in in-situ and laboratory experiments, and the many assumptions made in the theory due to unknown atmospheric turbulence. Another remaining issues is the wave breaking whose effect cannot be precisely considered. Numerical simulations with an air-sea two-phase numerical model capable of representing wave breaking would be one approach to greatly advance these issues.
In this study, we developed a two-phase flow model that numerically solves the weakly compressible Navier-Stokes equation with reference to Matsushita and Aoki (2021), to understand the interaction between the atmosphere and ocean in wind waves with breaking waves. This method solves the incompressible fluid as a compressible fluid with a low Mach number, thus eliminating the need to solve the Poisson equation with poor convergence obtained from the incompressible condition and allowing for a fully explicit calculation. A coupled phase-field and level-set method was used to represent the gas and liquid phases in the calculations.
Some benchmark experiments were conducted to validate the model. In the 2-D dam-breaking experiment, observed time-varying shape of the collapsing water column was in good agreement with those of Martin and Moyce’s laboratory experiment (1952). We also performed the Rayleigh-Taylor instability experiment to test the model and found that the simulated time evolution of interface shapes is similar to those of previous numerical studies. In the presentation, we will report numerical accuracies and costs of our model as well as remaining issues to be solved.