Yeon S. Chang, Tamay M. Ozgokmen, Hartmut Peters, and Xiaobiao Xu, RSMAS/Univ. of Miami
The Red Sea outflows are numerically simulated and the results are compared with the observational data from the REDSOX experiments. The three-dimensional simulation result shows two distinct plumes as it was observed along the `northern'' and the `southern'' channels. It also shows reasonable agreement with the field data through the direct comparison of the individual T-S profiles. The outflow's propagation speed is found to be 3-fold smaller than the estimated internal wave speed. It is shown that this is because the propagation speed of the computed overflows reduces due to the irregular seabed topography along the channel pathways. Details of topography also impose a hard upper limit in the grid size because the model simulations are only reasonable if the horizontal grid size resolves the channels, which are approximately 5 km wide. When the two parameterization schemes, K-Profile Parameterization (KPP) and one based on Turner's work (TP) are compared, the difference of the two schemes is small and not significant although TP provides better results in terms of entrainment and transport rate. Thus, it is concluded that the simulation of the Red Sea outflow is mainly influenced by the geographical factors such as bottom topography and grid resolution rather than the vertical mixing schemes.