According to Ingram and Prinsenberg (1998):
The open-water circulation of Hudson Bay is primarily driven by wind forcing and buoyancy input. In the summer, northerly winds dominant north of about 60N, with northwesterly winds in the southern half of the bay. The freshwater runoff generates and estuarine-type circulation in the bay as a whole, such that the upper-layer low-salinity flux to Hudson Strait is balanced by an inward flux of higher-salinity ocean water in the lower layer. The incoming higher-salinity and colder water enters at the northern boundary of Hudson Bay and through estuarine-like processes modifies the fresher surface waters. Typical mean surface flow values are 0.04 m s, equivalent to a two-year period to complete a trajectory around Hudson Bay.Ingram and Prinsenberg (1998).
The major regional features of the cycle include rainfall dominating over evaporation in the tropics within the ITCZ, with the subtropics characterized by an excess of precipitation except for the SPCZ. Precipitation again dominates in subpolar latitudes, with data near the poles being too sparse to form any generalizations. The ocean must of course compensate for these latitudinal differences, moving water into evaporation zones and away from regions where precipitation dominates. This transport accounts for a significant percentage of the total poleward heat transport on the planet.
Interbasin differences include the Atlantic being saltier than the Pacific due to the dominance of evaporation in the former and precipitation in the latter. The difference is thought to be maintained by water vapor transport across Central America and the lack thereof into the Atlantic from the east. These differences in surface water fluxes lead to interbasin transports in the ocean, although these are known even less accurately than the precipitation and evaporation patterns. See Schmitt (1995).
If, in standard notation,
is the vertical momentum equation with the height as the vertical coordinate and , then in the hydrostatic approximation the term is omitted. But the condition for the validity of the hydrostatic approximation is much more stringest than , the acceleration due to gravity, because almost all of is balanced by the inert hydrostatic pressure gradient associated with the resting reference state.
More helpfully, we can isolate the hydrostatically balanced and dynamically inactive pressure gradient by writing the Boussinesq form of the previous equation as
where the primes denote a deviation from the hydrostatically balanced reference state and is a standard (constant) value of density. Now it can be clearly seen that the condition for the neglect of is that it should be much smaller than rather than . Let us now try and estimate typical scales for which the hydrostatic approximation is valid, according to this more stringent condition.
Consider a convective event (in an unstratified ocean) which has a characteristic horizontal scale and vertical scale with horizontal and vertical velocity scales and , respectively. The time scale of a particle of fluid moving through the convective system is of order and a typical will be
and hence if
If this last condition is not satisfied, then the full vertical momentum equation must be used.See Jones and Marshall (1993).