Numerical experiments were conducted using the finite volume community ocean model(FVCOM) to study the impact of the initial density stratification on simulated currents over the Louisiana shelf during Hurricane Katrina. Model results for two simulation scenarios, including an initially stratified shelf and an initially non-stratified shelf, were examined. Comparison of two simulations for two-dimensional(2D) currents,the time series of current speed, and variations of cross-shore currents across different sections showed that the smallest differences between simulated currents for these two scenarios occurred over highly mixed regions within 1 radius of maximum wind(RMW) under the hurricane.For areas farther from the mixed zone, differences increased, reaching the maximum values off Terrebonne Bay. These large discrepancies correspond to significant differences between calculated vertical eddy viscosities for the two scenarios. The differences were addressed based on the contradictory behavior of turbulence in a stratified fluid, as compared to a non-stratified fluid. Incorporation of this behavior in the MellorYamada turbulent closure model established a Richardson number-based stability function that was used for estimation of the vertical eddy viscosity from the turbulent energy and macroscale. The results of this study demonstrate the necessity for inclusion of shelf stratification when circulation modeling is conducted using three-dimensional(3D) baroclinic models. To achieve high-accuracy currents, the parameters associated with the turbulence closures should be calibrated with field measurements of currents at different depths. Numerical experiments were conducted using the finite volume community ocean model (FVCOM) to study the impact of the initial density stratification on simulated currents over the Louisiana shelf during Hurricane Katrina. Model results for two simulation scenarios, including initially stratified shelf and an initially non -stratified shelf, were examined. Comparison of two simulations for two-dimensional (2D) currents, the time series of current speed, and variations of cross-shore currents across different sections showed that the smallest differences between over highly mixed regions within 1 radius of maximum wind (RMW) under the hurricane. For areas farther from the mixed zone, differences increased, reaching the maximum value off Terrebonne Bay. These large discrepancies correspond to significant differences between calculated vertical eddy viscosities for the The differences were addressed based on the contradictory Behavior of turbulence in a stratified fluid, as compared to a non-stratified fluid. Incorporation of this behavior in the Mellor Yamaha turbulent closure model established a Richardson number-based stability function that was used for estimation of the vertical eddy viscosity from the turbulent energy and the results of this study demonstrate the necessity for inclusion of shelf stratification when circulation modeling is conducted using three-dimensional (3D) baroclinic models. To achieve high-accuracy currents, the parameters associated with the turbulence closures should be calibrated with field measurements of currents at different depths.