Winds and surface currents
Contents
Winds and surface currents¶
Although the ocean circulation results from the highly nonlinear response of the ocean to the fluxes of momentum and buoyancy through the sea surface, it is the wind that most directly drives the near-surface currents, and indirectly drives the entire ocean circulation. Hence we need to know the general characteristics of the wind field over the ocean. Then we will look at the surface currents, and compare their structure to that of the wind.
Winds¶
Source of data¶
Prior to the satellite era, almost all large-scale wind measurements came from ship reports, the same basic database used for surface temperature and currents. The primary “measurement” was typically a simple visual estimate of the wind direction and speed on the Beaufort scale. This is based on the correspondence between sea state and wind speed for a fully developed sea.
Winds can also be inferred from the atmospheric pressure distribution at sea level. Because of the rotation of the earth, winds tend to blow along lines of constant pressure, with the higher pressure on the right in the northern hemisphere and on the left in the southern. Circular flow around a pressure maximum—clockwise in the northern hemisphere, counterclockwise in the southern–is called anticyclonic, and flow around a pressure minimum is called cyclonic. Such geostrophically balanced flow is equally important in the ocean, and will be explored in detail later in the course. In addition to the tendency to flow along lines of constant pressure, winds in the atmospheric boundary layer—that is, near the earth’s surface—tend to have a weaker flow component from high pressure to low, as would also occur in a nonrotating system. This down-gradient flow component occurs in currents near the ocean bottom also. It is a manifestation of Ekman layer dynamics, another topic to which we will return.
The advent of satellite-based scatterometers has revolutionized the estimation of winds over the ocean. A scatterometer is a type of radar: microwave pulses are directed at the sea surface, and the amplitude of the return depends on surface roughness (windgenerated capillary waves) and on the orientation of the beam relative to the ocean surface and the propagation direction of the waves, which is downwind. The technique is not without problems and uncertainties, but it has revealed dynamically-important smallscale structure in the wind field. An early example is Xie et al. (2001).
Even with wind estimates from scatterometers, the space-time coverage of wind observations is spotty. To fill the gaps and smooth out the glitches, wind estimates used in practice are now now are often made using dynamical numerical models “nudged” by observations. Such wind “products” are particularly convenient for driving ocean numerical models. This technique, assimilating observations into a model, is the basis of numerical weather prediction (for looking ahead), and of reanalysis (the generation of historical wind estimates). See, for example, the European Centre for Medium-Range Weather Forecasts (ECMWF).
Stress¶
Recall that the wind stress is a momentum flux per unit area through the air-sea interface. This momentum is carried down to the sea surface from higher in the atmosphere by turbulent motions that are very hard to measure. So, as a practical matter, the stress is almost always estimated parametrically from quantities that are easier to measure. The usual parameterization looks like this:
where \(\vec{U}_{10}\) is the wind velocity at \(10~\mathrm{m}\) elevation. The drag coefficient, \(C_d\), is typically about \(1.2\times 10^{−3}\). For better accuracy, one uses a variable \(C_d\) that increases with wind speed (causing stronger turbulence and a rougher sea surface) and as the SST exceeds the air temperature (unstable conditions, also causing stronger turbulence). The density of air at sea level, \(\rho_a\), is about \(1.178~\mathrm{kg\,m}^{−3}\).
Important
Examine () and note that wind stress is in the same direction of the wind, but its magnitude goes as the square of the wind speed.
Wind data is therefore sometimes presented in terms of “pseudo-stress”, the magnitude of which is just the square of the speed. Stress and pseudostress estimates depend on the time scale over which the wind speed is averaged before it is squared. The shorter this initial averaging interval, the larger the mean stress estimate.
Features¶
The dominant features in the wind field are very simple:
Tradewinds blow from east to west, and slightly equatorward (or actually towards the ITCZ) over most of the tropical Atlantic and Pacific.
Westerlies blow from west to east in middle latitudes. They are strongest in the respective winter of each hemisphere, but remain strong yearround in the Southern Ocean.
On eastern boundaries (west coasts), winds are typically equatorward in mid latitudes.
Poleward of \(60^\circ\) latitude, mean winds weaken, and in places take on an easterly component.
Monsoons dominate the tropical Indian Ocean and far western Pacific. They blow from the winter hemisphere to the summer hemisphere, generally with an easterly component when blowing equatorward, and a westerly component when blowing poleward.
The strongest winds are in the winter westerlies, especially in the southern ocean and in the northwest corners of the north Atlantic and north Pacific. However, there is also a region of very strong wind in the Arabian Sea during northern hemisphere summer: the southwest monsoon.
Currents¶
Source of data¶
Originally, most knowledge of surface currents came from the historical database of ship reports. A standard byproduct of navigation is the calculation of “set and drift”. This is the difference between the velocity of the ship over the ground as calculated from successive position fixes, and the velocity that would be expected based on the speed of the ship through the water and the heading. Ship drift measurements are inherently noisy, inaccurate, and coarsely resolved, being based on position fixes many hours apart.
During the last few decades, a substantial dataset has been accumulated from satellite-tracked drifters. Typically drogued at \(15~\mathrm{m}\), they are released from research vessels or ships of opportunity. For up to about 2 years after release, they provide current estimates based on position fixes at intervals of 8 hours to a couple of days. Drifter measurements are much more accurate than ship drifts. Motion of the drogue relative to the water is normally under \(2~\mathrm{cm\,s}^{-1}\). Drifters provide much more even and extensive sampling than ships, sampling regions where ships rarely go. Their sampling is still biased, however, not only by their release locations but by the fact that they follow the water, and hence tend to spend more time in regions of convergent surface currents and less in divergent regions such as near the equator in the eastern and central Pacific.
Like winds, near-surface currents can also be estimated from the pressure distribution. But atmospheric pressure on level surfaces can be measured directly with barometers; there is no ocean analog. Instead, pressure gradients on level surfaces are classically estimated using the hydrostatic balance, the density distribution calculated from measurements of salinity and temperature, and an assumption about the pressure distribution at some deep level. (It is becoming possible to make satellite-based estimates of absolute sea-surface height gradients relative to the geoid, but we will defer that topic.) Also unlike the atmospheric case, surface currents have no tendency to flow from high pressure to low; instead, in addition to their flow along lines of constant pressure, they have a component at some angle to the right (northern hemisphere) or left (southern hemisphere) of the wind stress. This will be studied in detail later.
Basic surface circulation pattern¶
Most schematics of surface currents are somewhat misleading in one or more ways: giving an incorrect impression of the relative strengths of western versus eastern boundary currents; giving little indication of the widths and speeds of currents; omitting major currents such as the Mindanao Current; giving no hint of the variability of currents; etc.
Key terms¶
cyclonic: rotating as a cyclone does, counterclockwise in the northern hemisphere, clockwise in the southern; this corresponds to positive relative vorticity
anticyclonic: the opposite of cyclonic
zonal: east-west
meridional: north-south
westward: towards the west (current)
easterly: from the east (wind)
Main features¶
subtropical gyres: each mid-latitude basin is roughly filled by an anticyclonic gyre, loosely corresponding to a subtropical high pressure cell in the atmosphere.
subpolar gyres: poleward of the subtropical gyres are cyclonic circulations, roughly corresponding to atmospheric low pressure cells (the Alaskan Low and the Icelandic Low). Currents in the subpolar gyres tend to run deeper, and be more uniform in the vertical, than those in the subtropical or equatorial gyres. They are correspondingly more strongly steered by topography.
western boundary currents: the poleward current forming the western side of the gyre is swift and deep relative to currents on the eastern sides.
Western boundary currents of the subtropical gyres are:
Gulf Stream system including the Florida Current.
Kuroshio
Agulhas Current
East Madagascar Current flowing south along the southeast coast of Madagascar.
East Australia Current is relatively weak and ill defined by WBC standards; eddies predominate.
Brazil Current is also much weaker than the Gulf stream and the Kuroshio.
Other western boundary currents include:
Mindanao Current flowing southward along Philippine coast. It is fed by the North Equatorial Current, and it feeds into the North Equatorial Countercurrent and into the Indonesian Throughflow.
Somali Current crossing equator in Indian Ocean, flowing north in northern summer, south in northern winter.
North Brazil Current and Guiana Current flowing northwest along the northeast coast of South America.
Labrador Current flowing south along the west side of the Labrador Basin.
East Greenland Current flowing south along the west side of the Greenland Sea, that is, along the east coast of Greenland. It turns the corner at the south tip of Greenland to flow into the Labrador Sea as the West Greenland Current, still generally along the coast of Greenland.
Oyashio flowing south along Kamchatka and the Kuril Islands.
Alaskan Stream flowing mostly west, but with a small southward component, along the southeast side of the Aleutians east of the dateline.
Falklands or Malvinas Current flowing north along the Patagonian coast. It connects the Antarctic Circumpolar Current between the Pacific and the Atlantic sectors of the Southern Ocean.
eastern boundary currents: these are not dynamically or descriptively similar to the WBCs, despite the way they look on schematics; they are very different animals. They are usually weaker, more variable, and shallower than WBCs. Most are cold currents flowing equatorward.
Noteworthy eastern boundary currents include:
California Current
Peru Current, also known as Humboldt Current
Benguela Current
Canary Current
Leeuwin Current, along the west coast of Australia, was little known until a few years ago. It is an anomaly among mid-latitude EBCs—a warm current flowing poleward.
mid-gyre currents: most of the poleward flow in the western boundary currents is returned equatorward not in the EBCs but in very slow drifts across the breadth of the basins.
equatorial currents: within \(15^\circ\) or so of the equator lie strong, nearly zonal currents.
In the Pacific and the Atlantic the equatorial currents are:
North Equatorial Current (NEC) flowing west; feeds the Kuroshio and Mindanao Current in the Pacific, and enters the Caribbean to feed the Florida Current in the Atlantic.
North Equatorial Countercurrent (NECC) flowing east in the range \(4^\circ\mathrm{N}– 10^\circ\mathrm{N}\).
South Equatorial Current (SEC) flowing west on both sides of the equator. It is sometimes split on the equator by the …
Equatorial Undercurrent (EUC), a subsurface eastward current that is strongest in the thermocline at around \(20^\circ\mathrm{C}\), but sometimes extends to the surface in the central and eastern Pacific.
In the South Pacific and the Indian Ocean there is also a
South Equatorial Countercurrent (SECC) flowing eastward. It is well developed in the western Pacific, but weak or intermittent in the central Pacific.
Note the pattern
“Currents” flow westward.
“Countercurrents” and “undercurrents” flow eastward, against the tradewinds.
monsoon currents: In the Indian Ocean and the western Pacific, many of the currents reverse seasonally. In addition, in the Indian ocean, there is an eastward current along the equator that appears twice per year, once in each monsoon transition season (fall and spring). Discovered in historical ship drift data by Klaus Wyrtki, it is sometimes called the “Wyrtki jet”.
Antarctic Circumpolar Current (ACC) flows eastward all the way around Antarctica. The flow is not uniform over a broad band, but is concentrated in narrow fronts. From north to south, these are: the Subantarctic Front, the Polar Front, and the southern ACC front.
See also
There is a nice (but unfinished) web site about currents: https://oceancurrents.rsmas.miami.edu/index.html.
Comparison between currents and winds¶
There is a general tendency for the zonal components of the winds and the currents to agree. Glaring exceptions are the equatorial countercurrents (which is why they are called countercurrents). Except off the Somali and Arabian coasts in the southwest monsoon, the wind field lacks anything like the ocean WBCs. Note also that the swiftest zonal currents are the equatorial currents, which occur under the moderate strength tradewinds; current speeds are generally lower under the strong westerlies.
Source
The original source of this content can be found in PDF form from Eric Firing’s course notes.