Ocean circulation
This article introduces the main circulation patterns in the upper ocean, which are driven primarily by winds. The article Thermohaline circulation of the oceans discusses ocean circulation driven mainly by density differences as well as related climate impacts. The influence of Earth’s rotation on ocean circulation is explained in the articles Ekman transport and Geostrophic flow.
By redistributing heat around the globe, ocean currents play an important role in regulating Earth’s climate. In the North Atlantic, ocean heat transport contributes to the relatively mild climate of Western Europe, together with the prevailing westerly winds and coupled atmosphere–ocean circulation patterns. Ocean and atmospheric circulation form a strongly interconnected dynamic system in which instabilities, particularly the El Niño–Southern Oscillation (ENSO), can produce major climate fluctuations. Ocean currents also play a key role in marine ecosystems by storing [math]CO_2[/math] and recycling nutrients.
Although tides are the dominant driver of water motion in shallow coastal seas, their influence on large-scale ocean circulation is smaller. Ocean tides are generated mainly in the open ocean by the gravitational forces of the Moon and Sun and are amplified as they propagate onto continental shelves (see Ocean and shelf tides).
Contents
Currents in the upper ocean
Low atmospheric pressure in the tropics (warm ascending air) and high atmospheric pressure in the subtropics (cooled descending air) drive large-scale atmospheric circulation cells under the influence of the Earth’s rotation. The resulting global wind field consists of dominant westerly winds at latitudes between 30° and 60° in both hemispheres (the Westerlies) and dominant easterly winds in the tropical and subtropical regions (the Trade winds).
Wind stress directly drives currents in the upper ocean mixed layer, whose thickness is typically several tens of meters but varies with season and location. This wind-driven Ekman transport should be distinguished from the large-scale boundary currents described under Geostrophic flow. Ekman transport is a shallow, ageostrophic response to local wind stress: turbulent momentum transfer in the upper ocean is balanced mainly by the Coriolis force, producing a current that rotates and decays with depth and a net transport approximately perpendicular to the wind direction. Observed Ekman-layer velocities are typically of order centimeters per second and are concentrated in the upper tens of meters, although they may reach the upper thermocline at low latitudes (Fig. 1).
Due to the Earth's rotation, the trade winds at low latitudes and the westerlies at mid latitudes drive large basin-scale gyres in the upper ocean: clockwise gyres in the subtropical Northern Hemisphere and anticlockwise gyres in the subtropical Southern Hemisphere.[4]
According to Sverdrup theory, the subtropical wind-stress curl generates downwelling and broad equatorward flow in the ocean interior (see Ekman transport). The return flow is concentrated in narrow, intense western boundary currents such as the Gulf Stream, Agulhas and Kuroshio, a process known as western boundary intensification.[5] (see Western Boundary Currents). Western boundary currents are basin-scale geostrophic currents associated with horizontal pressure gradients, sea-surface slope, density structure, and the closure of the wind-driven Sverdrup circulation. Velocities are commonly of order 1 m/s or more, extending hundreds to more than a thousand meters below the surface (see Fig. 2).
There are five subtropical gyres: the North Atlantic, the South Atlantic, the North Pacific, the South Pacific and the Indian Ocean Gyre, see Fig. 3. They are driven by the anticyclonic wind-stress curl generated by the trade winds at low latitudes and the westerlies at mid-latitudes. Convergence associated with Ekman transport generates large downwelling zones within these gyres.
Besides the subtropical gyres, at high latitudes subpolar gyres are driven by a cyclonic wind-stress curl. Large upwelling zones exist in the subpolar gyres. However, in the North Atlantic, dense, salty surface water cooled by evaporation sinks to depth, fueling the Atlantic Meridional Overturning Current (AMOC), see Thermohaline circulation of the oceans.
The most famous ocean current, the Gulf Stream, is a strong western boundary current that transports large amounts of heat from the Caribbean toward the North Atlantic. Along the US east coast it forms a narrow, fast jet due to western boundary intensification associated with the Earth’s rotation.[7] Farther offshore, it becomes a meandering current that generates numerous mesoscale eddies. The North Atlantic subtropical gyre is completed by the Canary Current, which carries relatively cool water southward along the eastern Atlantic. The Kuroshio is a similar western boundary current in the North Pacific, forming part of the North Pacific subtropical gyre together with the California Current and the North Equatorial Current. The North Equatorial Current and South Equatorial Current are driven by the tropical trade winds. The South Pacific subtropical gyre is bounded in the east by the cool Peru (Humboldt) Current, while the Indian Ocean subtropical gyre includes the warm Agulhas Current and the relatively cool West Australian Current.
Mesoscale and submesoscale eddies
Ubiquitous mesoscale eddies are major agents in the ocean for the transport of heat, salt, nutrients and other tracers across the ocean and therefore play a key role for the ocean ecosystem. They contain a large fraction of ocean kinetic energy and are central to the redistribution and dissipation pathway of energy[8]. Mesoscale eddies mainly emerge from the barotropic and baroclinic instabilities from large-scale wind-driven surface and subsurface circulations. Important source regions are western boundary currents such as the Gulf Stream, Kuroshio and Agulhas Current, where large meanders can pinch off into warm-core and cold-core rings (Fig. 4). Western boundary current rings are among the most energetic and long-lived mesoscale eddies and can transport water masses far from their region of origin. Eddies are also generated by instabilities of eastern boundary upwelling currents such as the California, Canary, Benguela and Humboldt/Peru–Chile systems. In these regions unstable coastal jets, fronts and filaments shed eddies that transport nutrient-rich coastal waters offshore and influence biological productivity, oxygen, carbon and plankton distributions. Other eddies are generated where currents interact with seamounts, ridges, island chains, continental slopes, capes, straits and escarpments.[9]
The horizontal scale of mesoscale eddies is closely related to the first baroclinic Rossby deformation radius, see Geostrophic flow#Geostrophic adjustment and Rossby waves. This radius is the natural scale at which buoyancy forces (characterized by the density difference [math]\Delta\rho[/math]), pressure gradients and Earth’s rotation are all important. In a simple layered ocean it is given by
[math]R_d = \sqrt{g'H}/f,[/math]
where [math]g'=g\Delta\rho/\rho[/math] is the reduced gravity, [math]H[/math] is an active-layer or vertical-mode depth scale, and [math]f[/math] is the Coriolis parameter. Many mesoscale eddies have diameters of order a few deformation radii. Because [math]R_d[/math] is larger in low latitudes and smaller in high latitudes, mesoscale eddies are generally larger in the tropics and subtropics than in polar and subpolar oceans. Typical diameters are tens to hundreds of kilometres, with lifetimes from weeks to months or longer. Surface height anomalies often exceed 10 cm, and maximum azimuthal velocities inside energetic eddies may exceed 20 cm/s. Many eddies are surface-intensified, but some extend several hundred meters below the surface, and energetic rings or mode-water eddies can influence the water column to depths of order 1000 m or more. [10]
Mesoscale eddies are not simply passively advected by the ambient flow. They move as rotating, partly coherent vortices whose propagation is controlled by self-propagation on the planetary vorticity gradient, background currents, eddy–eddy interactions and topographic effects. Most long-lived mesoscale eddies propagate westward (see the Appendix), with a weaker meridional drift: cyclonic eddies tend to drift poleward and anticyclonic eddies equatorward. Their westward translation speed is often close to, or somewhat larger than, the long first-baroclinic Rossby-wave speed,
[math]c_R \approx -\beta R_d^2 \approx -\beta g'H/f^2,[/math]
where the minus sign denotes westward propagation and [math]\beta = \partial f/\partial y[/math] is the meridional gradient of the Coriolis parameter. Typical propagation speeds are of order 1–5 cm/s. Because the rotational velocity inside many eddies is larger than their translation speed, such eddies can trap water in their cores and transport heat, salt, nutrients and biological material over large distances before this water is mixed with the surroundings.[10][11]
Eddies are not isolated structures. They interact, merge, split, form dipoles and exchange energy with surrounding currents and waves[12]. Dipole eddies can propagate more rapidly than single eddies and can enhance vertical exchange, including wind-induced Ekman pumping. At smaller scales, submesoscale fronts, filaments and vortices are especially important in the upper ocean, where they can generate strong vertical velocities, sharpen tracer gradients and contribute to mixing and restratification.
Mixing, restratification and subduction processes in the upper ocean also contribute to the formation and ventilation of mode waters and intermediate waters. Mode waters are relatively thick layers of nearly uniform temperature and salinity, formed mainly by deep winter mixing followed by subduction in subtropical and subpolar regions. Intermediate waters form at somewhat greater depth, especially in high-latitude and frontal regions such as the Southern Ocean and North Atlantic, and spread over large distances between the upper and deep ocean. These water masses form an important link between the upper-ocean wind-driven circulation discussed here and the deeper overturning circulation described in Thermohaline circulation of the oceans.
Upwelling
In coastal regions where alongshore winds drive offshore Ekman transport, surface water is displaced away from the coast and deeper water rises to replace it, see Fig. 5. This phenomenon, which is called 'upwelling', enriches surface waters with inorganic nutrients that stimulate primary production. Upwelling zones are characterized by a very rich marine life with abundant resources for fishery, see Open ocean habitat. Major coastal upwelling systems occur along eastern ocean boundaries, including the California, Canary, Peru/Humboldt and Benguela systems, where prevailing alongshore winds drive offshore Ekman transport.
Equatorial upwelling occurs because trade-wind stress drives surface Ekman transport away from the equator: northward just north of the equator and southward just south of it. This divergence draws cooler, nutrient-rich water upward, see Fig. 6. Downwelling zones exist north and south of the equator, as explained in the article Ekman transport.
The Antarctic Circumpolar Current (ACC) is situated in the Southern Ocean and continuously circles Antarctica because no continental barriers interrupt the flow. It is an eastward-flowing current driven primarily by the strong westerly winds at these latitudes. Northward Ekman transport at the surface induces upwelling of deep water in the Southern Ocean, including Circumpolar Deep Water and other deep water masses. Because no continental barriers exist at Drake Passage latitudes, large-scale zonal pressure gradients cannot easily be maintained, which strongly constrains geostrophic meridional flow in the upper and mid-depth ocean. Part of the compensating return flow therefore occurs through deep pathways steered by bottom topography, where variations in water-column thickness permit meridional motion consistent with potential-vorticity conservation[14] (see Geostrophic flow#Vorticity). In addition, the ACC is structured by several major fronts associated with strong horizontal density gradients. Instabilities of these fronts generate mesoscale eddies that contribute importantly to meridional exchange and partly compensate the wind-driven Ekman transport (“eddy compensation”), thereby influencing the strength of the Southern Ocean overturning circulation.[15]
El Niño–Southern Oscillation
Variability in the coupled ocean–atmosphere system of the tropical Pacific produces the El Niño–Southern Oscillation (ENSO), the dominant mode of interannual climate variability on Earth. ENSO arises through interactions between trade winds, ocean circulation, sea-surface temperature, and the thermocline in the equatorial Pacific. These interactions influence weather and climate patterns far beyond the Pacific region through large-scale atmospheric teleconnections.
Under normal conditions, strong easterly trade winds drive warm surface water westward, causing warm water to accumulate in the western Pacific and promoting upwelling of relatively cold, nutrient-rich water along the equator and the west coast of South America. The associated tilt of the equatorial thermocline maintains relatively cool sea-surface temperatures in the eastern Pacific.
During an El Niño event, the trade winds weaken. Warm surface waters then spread eastward through enhanced Equatorial Counter Currents and Kelvin-wave adjustment of the thermocline. Upwelling in the eastern Pacific is suppressed, causing sea-surface temperatures there to rise. The warmer eastern Pacific lowers atmospheric surface pressure, while relatively cooler conditions in the western Pacific increase surface pressure there. This pressure difference further weakens the trade winds, allowing even more warm water to move eastward and further reducing upwelling. This self-reinforcing ocean–atmosphere interaction is known as the Bjerknes feedback.[16][17]
El Niño events strongly affect marine ecosystems and fisheries because reduced upwelling limits the supply of nutrients to surface waters. ENSO also alters rainfall patterns, storm tracks, drought occurrence and tropical cyclone activity in many parts of the world through atmospheric teleconnections.
The opposite phase, La Niña, is characterized by stronger-than-average trade winds, enhanced upwelling in the eastern Pacific, and increased accumulation of warm water in the western Pacific. ENSO events typically recur every 2–7 years, although their timing and intensity are irregular. The detailed mechanisms controlling the onset, growth and termination of ENSO events remain active topics of research.
Appendix: Westward propagation of a mesoscale eddy
Here follows a heuristic derivation of westward propagation for the idealized case of a cyclonically rotating circular mesoscale eddy in the northern Hemisphere shown in Fig. A1. The longitudinal [math]x-[/math]axis is oriented W-E and the latitudinal [math]y-[/math]axis S-N. The eddy migrates with speed [math]\vec{C}=(c_E,c_N,0)[/math] and the internal eddy motion is represented by the velocity vector [math]\vec{U}=(u,v,0)[/math]. As the eddy dissipates kinetic momentum at the edges, the vorticity [math]\omega = \partial v /\partial x - \partial u / \partial y[/math] is outward decreasing (e.g. [math]\; \partial \omega / \partial x \lt 0[/math] at E and [math]\; \partial \omega / \partial x \gt 0[/math] at W).
Here we consider the simplest barotropic approximation with constant layer thickness and weak friction. Conservation of potential vorticity then reduces to approximate conservation of absolute vorticity [math]f+\omega[/math]. In a frame moving with the eddy, this requirement reads
[math](\vec{U}-\vec{C}) \vec{\nabla} (f+\omega) = 0 \, . \qquad (A1)[/math]
The planetary vorticity (Coriolis parameter) [math]f[/math] depends on the latitudinal coordinate as [math]f=f_0+\beta \, y[/math]. Equation (A1) can therefore be written
[math](u- c_E) \dfrac{\partial \omega}{\partial x} +(v-c_N) (\dfrac{\partial \omega}{\partial y} + \beta) = 0 \, . \qquad (A2)[/math]
At the eddy locations N and S we have [math]v=0, \partial \omega / \partial x =0[/math]. Substitution in Eq. (A2) gives [math]c_N=0[/math].
At the eddy locations W and E we have [math]u=0, \partial \omega / \partial y =0[/math], while at E, [math]\; \partial \omega / \partial x \lt 0, v\gt 0[/math] and at W, [math]\; \partial \omega / \partial x \gt 0, v\lt 0[/math]. Substitution in Eq. (A2) together with [math]c_N=0[/math] implies [math]c_E = \dfrac{\beta \, v}{\partial \omega / \partial x} \lt 0[/math]. This shows that for the situation sketched in Fig. A1, the eddy migrates to the west.
The same qualitative argument applies to anticyclonic eddies and to eddies in the Southern Hemisphere, provided the sign changes of relative vorticity and planetary vorticity are treated consistently. In this idealized β-plane argument, the required compensation of the meridional gradient of planetary vorticity leads to a westward propagation of the eddy. In real oceans, the propagation speed and possible meridional drift are also influenced by stratification, the deformation radius, eddy amplitude, background currents and topography.
Related articles
- Thermohaline circulation of the oceans
- Ekman transport
- Geostrophic flow
- Open ocean habitat
- Shelf sea exchange with the ocean
- Ocean and shelf tides
- Coriolis acceleration
References
- ↑ Price, J.F. and Sundermeyer, M.A. 1999. Stratified Ekman layers. Journal of Geophysical Research: Oceans 104: 20,467–20,494
- ↑ Rossby, T., Flagg, C.N., Donohue, K., Fontana, S., Curry, R., Andres, M. and Forsyth, J. 2019. Oleander is more than a flower: Twenty-five years of oceanography aboard a merchant vessel. Oceanography 32: 126–137
- ↑ Beal, L.M. and Bryden, H.L. 1999. The velocity and vorticity structure of the Agulhas Current at 32°S. Journal of Geophysical Research: Oceans 104: 5151–5176
- ↑ Munk, W.H. 1950. On the wind-driven ocean circulation. J. Met. 7, 79-93.
- ↑ Stommel, H. 1948. The westward intensification of wind-driven ocean currents. Transactions, American Geophysical Union 29: 202-206.
- ↑ https://commons.wikimedia.org/wiki/File:Corrientes-oceanicas-en.svg
- ↑ Stommel, H. 1948. The westward intensification of wind-driven ocean currents. Transactions, American Geophysical Union 29: 202-206.
- ↑ McWilliams, J.C. 2016. Submesoscale currents in the ocean. Proceedings of the Royal Society A 472: 20160117
- ↑ Zhang. Z., Wang, G., Wang, H. and Liu, H. 2024. Three-Dimensional Structure of Oceanic Mesoscale Eddies. Ocean-Land-Atmos. Res. 3, 0051
- ↑ 10.0 10.1 Chelton, D.B., Schlax, M.G. and Samelson, R.M. 2011. Global observations of nonlinear mesoscale eddies. Progress in Oceanography 91: 167–216
- ↑ Barabinot, Y., Speich, S. and Carton, X. 2024. Defining mesoscale eddies boundaries from in‐situ data and a theoretical framework. Journal of Geophysical Research: Oceans 129, e2023JC020422
- ↑ Cite error: Invalid
<ref>tag; no text was provided for refs namedZ24 - ↑ http://www-das.uwyo.edu/~geerts/cwx/notes/chap11/equat_upwel.html
- ↑ Rahmstorf, S. 2006. Thermohaline Ocean Circulation. In: Encyclopedia of Quaternary Sciences, Edited by S. A. Elias. Elsevier, Amsterdam
- ↑ Marshall, J. and Speer, K. 2012. Closure of the meridional overturning circulation through Southern Ocean upwelling. Nature Geoscience 5: 171–180
- ↑ Bjerknes, J. 1969. Atmospheric teleconnections from the equatorial Pacific. Mon. Weather Rev. 97: 163–172.
- ↑ Wyrtki, K. 1973. Teleconnections in the Equatorial Pacific Ocean. Science 180: 66–68.
Please note that others may also have edited the contents of this article.
|