Henry Stommel

The Ventilated Thermocline

Abstract A simple theoretical model for the oceanic thermocline and the associated field of current is presented. The model consists of a finite but arbitarily large number of inviscid, homogeneous fluid layers each with a different density. The dynamical balances everywhere are Sverdrupian. IN regions where the Ekman pumping is negative (downward) the surface density is specified, i.e., the position of the outcrop of density interfaces is specified. This outcropping of density layers allows deep motion to be excited by the ventilation provided by Ekman pumping even in latitudes far south of the outcrop where the layer is shielded from direct influence of the wind. Analytical solutions are presented in the case where the density-outcrop lines are coincident with latitude circles. The solutions are not self-similar and important sub-domains of the solution are defined by critical potential vorticity trajectories which separate the ventilated from the unventilated regions in the lower thermocline. These cri...

On the abyssal circulation of the world ocean — II. An idealized model of the circulation pattern and amplitude in oceanic basins

Abstract Stationary planetary flow patterns driven by source sink distributions in an ocean on the rotating earth were developed in Part I. These theoretical results are now used to construct a highly idealized model of the general abyssal circulation of the world ocean. The model is based on the postulation of the existence of two concentrated sources of abyssal waters (one in the North Atlantic and another in the Weddell Sea) and on a uniformly distributed upward flux of water from the abyssal to the upper layers as part of the mechanism of the main oceanic thermocline. Order of magnitude calculations based on this model lead to a variety of estimates of the time in which the deep water is replaced (from every 200 to 1800 years). Comparison of leading terms in the dynamical equations and equation describing the flux divergence of a tracer shows that there is a large range of lateral eddy coefficient which will influence the distribution of the tracer but not affect the dynamically determined planetary flow patterns.

On the abyssal circulation of the world ocean—I. Stationary planetary flow patterns on a sphere☆

Abstract A treatment of stationary planetary flow patterns driven by source-sink distributions in a cylindrical tank (Stommel et al., 1958) is extended to predict flow patterns which might be expected under similar circumstances on a rotating sphere. Flow patterns are sketched for various source-sink distributions and meridional and zonal boundary conditions.

ENTRAINMENT OF AIR INTO A CUMULUS CLOUD

Abstract A theory of convective clouds is presented, the fundamental hypothesis being that the ascending current in a cloud entrains air from its surroundings. A method is developed for computing the amount of entrainment from a knowledge of the temperature and specific humidity inside and outside the cloud. The concentration of water in the form of drops is also determined. Finally the theory is applied to some observations of trade cumulus made near San Juan, Puerto Rico.

On the abyssal circulation of the world ocean—IV: Origin and rate of circulation of deep ocean water as determined with the aid of tracers

Abstract Using box models, and the observed distribution of temperature, salinity and radiocarbon estimates are made of the origins and rate of flow of waters that make up the Common Water in the Pacific and Indian Oceans, and Antarctic Intermediate Water. Difficulties in extending the computation to Antarctic Bottom Water are described. Emphasis is placed upon questions of computational reliability and the occurrence of ‘ill-conditioned’ equations.

The heton, an elementary interaction between discrete baroclinic geostrophic vortices, and its implications concerning eddy heat-flow

Among the interactions of two discrete baroclinic geostrophic vortices in a two-layer system there is one class of interaction that is non-trivial; when the two vortices are of opposite sign and in different layers, and close enough together, they transport heat. Because this particular interaction can transport heat, we propose to call it the heton. It is a tilted baroclinic pair. In the Northern Hemisphere it transports heat to the left of the direction toward which its top tilts. Two warm or two cold hetons repel one another when outside the radius of deformation. A warm and a cold heton attract one another. A simple two-heton engine that exhibits vortex splitting, loss of available potential energy, and meridional heat transport is presented.

Four Views of a Portion of the North Atlantic Subtropical Gyre

Abstract Data from four cruises to the “β-triangle” centered at 27°N, 32°30′W were smoothed by fitting second-degree polynomials at each of about 30 different density surfaces. The density intervals were Δσp=0.3‰, corresponding to about 50 db intervals in pressure. From the polynomials, determination was made of central values, horizontal derivatives and Laplacians of the fields of pressure, salt, oxygen and dynamic height. In addition, maps of the fits and deviations of each nation from the smoothed fits were produced. From the steady advective diffusive equation and the smoothed fits to the age oxygen and dynamic height fields, the lateral isopycnal diffusivity as estimated to be KH ∼ 0.5 × 103 m2 s−1. Although the salt field was reasonably stable from cruise to cruise, the variability of the baroclinic velocity shear was found to be as large as the baroclinic shear itself. The maps suggest a wobble of the gyre. The standard deviation of the fluctuations at each nation from the smoothed fits, when norma...

Is the South Pacific helium-3 plume dynamically active?

It is suggested that the hydrothermal vents of the South Pacific Rise produce a beta-governed circulation at mid-depth, and that perhaps the associated plume of excess3He (Lupton and Craig [1]) points westward because of the dynamics of this circulation rather than as a passive tracer.

Beta spirals and absolute velocities in different oceans

Abstract A simple steady-state theory of the geostrophic velocity field in the main thermocline of large-scale ocean gyres predicts the existence of current spirals associated with either the vertical component of velocity or local density change due to heating or cooling. The sense of rotation of the horizontal velocity vector with increasing depth is anticyclonic wher the vertical component of velocity is downward and it is cyclonic where the vertical component is upward or where there is local cooling. These two cases correspond to what is expected from the Ekman pumping and climatic influences in subtropical and subpolar gyres, respectively. A technique is devised for computing the absolute three-dimensional field of velocity from observed density data alone, assuming further immiscibility of the density stratification and simple linear betaplane vorticity conservation. Hydrographic station data in six mid-ocean areas, smoothed to remove meso-scale eddy noise, reveal spirals in the main thermocline that correspond to predictions. In the subtropics of the North Atlantic, South Atlantic, and North Pacific, a consistent picture of upper-level equatorward and westward flow, with smaller, deeper flows in an opposite direction emerges. Vertical downward flows are found in the top of the main thermocline, presumably as deep extensions of the surface Ekman forcing. A case of a subpolar gyre (North Atlantic) appears to be a strong spiral dominated by cooling rather than the vertical velocity associated with the Ekman layer. Indications in the treatment suggesting departures of the real ocean from the conditions of the model are discussed.

Water masses and patterns of flow in the Somali Basin during the southwest monsoon of 1964

Abstract Closely spaced hydrographic sections made during August–September 1964 in latitudes 3°S–12°N, and between the East African coast and longitude 56°E, define in detail a complex structure of water masses in the Somali Basin under the southwest monsson. Reference to observations made elsewhere in the Indian Ocean permits clear identification of the source waters responsible for this structure. In the near-surface water the distributions of temperature and salinity show the course and lateral extent of the Somali Current, the offshore movement of cold water upwelled near the Somali coast, and two warm saline inflows from the Gulf of Aden and the Arabian Sea. At depths greater than 2000 m, small differences in temperature-salinity characteristics reveal a narrow northward flow adjacent to the continental slope, roughly paralleling the Somali Current. No definite inferences can be drawn concerning flow patterns at intermediate depths, both because of the apparent small scale of horizontal variation there, which is not resolved by the station spacing, and because of inherent ambiguity in core methods when applied to flows which may reverse seasonally.