Estimation of baroclinic tide energy available for deep ocean mixing based on three-dimensional global numerical simulations

The global distributions of the major semidiurnal (M₂ and S₂) and diurnal (K₁ and O₁) baroclinic tide energy are investigated using a hydrostatic sigma-coordinate numerical model. A series of numerical simulations using various horizontal grid spacings of 1/15–1/5° shows that generation of energetic baroclinic tides is restricted over representative prominent topographic features. For example, nearly half of the diurnal (K₁ and O₁) baroclinic tide energy is excited along the western boundary of the North Pacific from the Aleutian Islands down to the Indonesian Archipelago. It is also found that the rate of energy conversion from the barotropic to baroclinic tides is very sensitive to the horizontal grid spacing as well as the resolution of the model bottom topography; the conversion rate integrated over the global ocean increases exponentially as the model grid spacing is reduced. Extrapolating the calculated results in the limit of zero grid spacing yields the estimate of the global conversion rate to be 1105 GW (821, 145, 102, 53 GW for M₂, S₂, K₁, and O₁ tidal constituents, respectively). The amount of baroclinic tide energy dissipated in the open ocean below a depth of 1000 m, in particular, is estimated to be 500–600 GW, which is comparable to the mixing energy estimated by Webb and Suginohara (Nature 409:37, 2001) as needed to sustain the global overturning circulation.