Wave Bottom Boundary Layer Models for Smooth and Rough Beds

Abstract

Abstract : Seven one-dimensional wave bottom boundary layer models have been analyzed based on different methods for estimating the turbulent eddy viscosity (laminar, linear, linear-exponential, parabolic. k-one-equation turbulence closure, k-epsilon-two-equation turbulence closure, and k-omega-two-equation turbulence closure). Two generic test cases displayed similar velocity profiles for all (he models with the exception of the laminar model. Boundary layer and sheer stress estimates, however, did show some differences. The linear and parabolic models predicted bed shear stress twice as large as the k-omega model and 25 percent larger than the k or k-epsilon models. Phase leads between the predicted bed shear stress and the free-stream velocity matched expectations with the laminar case leading by 45 degrees and the other models predicting a phase lead of the shear stress maxima between 12 and 18 degrees with respect to the free-stream velocity maximum. Comparisons to laboratory data on smooth and rough beds showed that overall the linear model was slightly more accurate than the parabolic linear-exponential. k, and k-omega models. The least accurate were the laminar and k-epsilon models. Based on the two laboratory simulations (forced by a 9.72-s sinusoidal wave form with an amplitude of 2 m s(-1)). it is shown that the extra computational effort required for the turbulence closure schemes does not afford an improvement in predictive capability in a one-dimensional boundary layer model. Therefore. it is recommended that the linear or parabolic model be used to rapidly determine flow characteristics for one-dimensional studies of the wave bottom boundary layer.