S. Daniel Jacob, NASA/GISS, L. K. Shay and G. R. Halliwell, RSMAS
One of the major uncertainties in a coupled hurricane ocean forecasting system is the choice of mixing scheme in the ocean model as significant oceanic mixed layer cooling and deepening during a storm passage is due to mixing. Using two HYCOM configurations, this issue is investigated in detail to evaluate the performance of five mixing schemes for different forcing characteristics associated with three major hurricanes. Simulated fields are evaluated by comparing to available observations. Results indicate that the performance of the higher order mixing schemes is better than the bulk-schemes in HYCOM. Sensitivity of the schemes to varying vertical resolutions in the upper ocean is also investigated for multiple simulations for the same forcing. While for some schemes the comparison statistics steadily degrade with increasing vertical resolution, this is seen to be improving for the Gaspar scheme. Simulated currents and density structure for these multiple cases are examined to understand and quantify the nature of this variability. While initial conditions with realistic oceanic mesoscale variability are used in these simulations, there is some cold bias in the simulated ocean response due to bias in the initial conditions. The spatial distribution of this bias will also be presented in this paper.