Sensitivity of Ventilation Rates and Radiocarbon Uptake to Subgrid-Scale Mixing in Ocean Models

Abstract

The sensitivity of ventilation timescales and radiocarbon (14C) uptake to subgrid-scale mixing parameterization is studied in a global ocean model. Seven experiments are examined that are identical in every manner except their representation of subgrid-scale mixing of tracers. The cases include (i) two runs with traditional Cartesian mixing (HOR), (ii) a run with enhanced isopycnal mixing (ISO), and (iii) four runs in which the effects of eddies on the mean ocean flow are parameterized following Gent and McWilliams (GM). Horizontal, isopycnal, and isopycnal-thickness diffusion coefficients are varied sequentially in the model runs. Of particular interest is the role of the tracer mixing schemes in influencing longer timescale ventilation processes—centennial and beyond—such as deep water mass renewal and circulation. Simulated ventilation timescales and 14C vary greatly between the three mixing schemes. The isopycnal mixing run exhibits the most rapid water mass renewal due to strong diffusion effects and excessive surface convective overturn, particularly in the Southern Ocean. In contrast, the GM cases show much more gradual renewal of deep and bottom waters, with limited vertical convection of surface waters and slower abyssal currents. Under GM, a background horizontal diffusion or altered isopycnal mixing do not significantly change interior ocean ventilation rates. This means modelers can adjust these background diffusion coefficients under GM (for numerical purposes) without significantly changing model ventilation rates. Reducing the GM isopycnal thickness diffusivity, on the other hand, noticeably increases simulated deep water ventilation rates. In comparison with the HOR runs, deep and bottom water ventilation timescales are reduced by about 30% in ISO, and increased by 30%–40% under GM. Comparison is made between model simulated and observed 14C. The GM runs appear to be the least successful in the North Atlantic Ocean, exhibiting very gradual and only shallow water-mass renewal compared to observations. In the Pacific and Indian Oceans, the HOR and ISO runs are ventilated too rapidly due to strong convection and water-mass contribution from the Southern Ocean. In contrast, the GM runs simulate spuriously old and 14C-depleted bottom and middepth water. The GM cases do, however, capture realistic 14C in the upper 1500 m of the Indian and Pacific Oceans. Overall, none of the model cases reproduce global ocean ventilation rates over centennial timescales (under the chosen set of parameter values). Higher horizontal resolution and a spatially varying GM thickness diffusivity may be required before global models capture long timescale ocean renewal processes with some degree of fidelity.