Summary: | This study tested the utility of a variable mixing efficiency formulation proposed by Mashayek, et al. (2017) for use in oceanographic models other than the modern, pre-industrial ocean. This formulation is used to calculate diapycnal (vertical) mixing due to unresolved subgrid-scale processes. Results from Last Glacial Maximum (LGM) simulations for variables such as sea surface temperature, density, and meridional overturning were compared to those of a model simulation of the pre-industrial ocean using three different parameter conditions for mixing efficiency and one “default” constant value of Γ = 0.2. These three parameter conditions were influenced by Mashayek, et al. (2017), who devised a variable mixing efficiency parameterization and applied it to the pre-industrial ocean. Simulated changes in meridional overturning between the pre-industrial and LGM simulations were very similar for the constant mixing efficiency coefficient models and the variable coefficient models; e.g. the flow of North Atlantic Deep Water out of the Atlantic was ~2 Sv less in the LGM simulations for both variable and constant Γ models. The null hypothesis that a variable mixing efficiency parameterization would not significantly affect differences in physical variables between LGM and pre-industrial simulations was not rejected. A constant Γ is recommended for future modeling of the ocean-climate system. Key Words: mixing efficiency, ocean, turbulence, diapycnal diffusivity, Last Glacial Maximum, pre-industrial, Buoyancy Reynolds number
|