Dynamic topography
4.3.8 MODELLING THE CIRCULATION OF THE NORTH ATLANTIC
To simulate the circulation of the North Atlantic well enough to have a chance of re-creating some of the features seen in the real ocean (e.g. Figure 4.20(a)), we would need a three-dimensional model, with sea-floor topography (Figure 4.18) and variations with depth in current flow, temperature and salinity, and hence density (Section 4.2.4). The details of how such a model would be 'constructed' are outside the scope of this book, but modelling is an intrinsic part of modern oceanography and our review of how ideas about the North Atlantic gyre have developed would be incomplete without a brief look at some recent models of the North Atlantic.
One way that modellers can learn about the processes governing ocean circulation at the same time as improving modelling techniques is to compare different types of models with each other and with what is observed in reality. In 1999, as part of a European Community project known as D Y N A M O (Dynamics of North Atlantic Models), three different models of the North Atlantic were compared in detail. One was a 'level' model, one a 'terrain-following" model, and one an isopycnic m o d e l - the coordinate systems for these types of model were shown in Figure 4.18.
All three were forced in the same way, using realistic datasets derived from information about wind patterns, and exchanges of heat and freshwater at the ocean surface, provided by the European Centre for Medium-range Weather Forecasting. The starting conditions were (1) a state of rest, and (2) T, S and density values for each box consistent with the observed mean hydrographic state for the ocean, as detailed in the Climatological Atlas of the World Ocean.* The models were run forward for the equivalent of 20 years, to allow a near-equilibrium state to be reached. The winter sea- surface temperature distributions that were produced by the three models are shown in Figure 4.35.
*This atlas was first produced by NOAA in 1982 and was revised in 1994; it is usually referred to just as "Levitus' (or "Lcvitus climatology'), after its original compiler, S. Levitus (the 1994 edition is by S. Levitus and T.P. Boyer).
(a)
LEVEL MODEL
(b) 'TERRAIN- FOLLOWING' MODEL
Figure 4.35 Winter sea-surface temperatures (~ predicted by the three models tested in the DYNAMO project: (a) the level model, (b) the 'terrain-following' model, and (c) the isopycnic model. All three models produce a Gulf Stream carrying warm water up the western side of the ocean, but the way the Gulf Stream/North Atlantic Current crosses the ocean is slightly different in each case.
(c) ISOPYCNIC MODEL
9
.~_ ~ 9 ~ ~ "~ ~.__. ~'~" ~.-_ ~ .~ .~ ~._ 9
--~ E ~ ~-- = ~'~~
..~ 9 ~ ~ "-~ ~9 --- ~'~
=~-~ ,~ > =~ = ~ ~ 9 o~-='- "~ E E-Z ~
= E ='= g-- ~" = "=~m< o
9 ---~ ~
Ca ~D 1"~ 0 ."d ~D *" ~D
<8~gee~e
"~ r :::::I r~ ~ ~ ~ ~ ~.-= .~
9 ~.~_ ~< E.=_o= ~ ~_~,~
~ *--' ,..., ~ ~ ~, ~1~ ~ ._
o ~-pD~ : :"~'~ ~ ~
~-=~.~.~ {~~~-~-
o ~ ~ ~_ =~-, ~ ~"go =_,~'6~ .-
~.= ~ .~-~ ~-
.,_., . I,.-, = E ~ ~ _>,~ - _
"~ -~z g.
~ ~ o ~ ~ 9 ~.E.- =-~.~ h~ -~ < o 9 :~o.=". ,~ *~ ~ ~ -- 9
-_~ [~ ~~-~ ~
Z ~u .,.- r
"- "-" 9 r
9 ~ "--
.,~
.,..~ pr
r "-- U E Li 9
.> .~ =..2E
~ ~-
I,.., " 8 '-"
~ ~u ~. 9
~~o ~-~ ~'~
r ~ ~ ~ .,--E -E~~ ~
9
~-- .~~~
~ "--
, ,~ "~
~.~ ~ .-
~,~ .~=-~ ~o~
='~" ~" ~-F~4~- oC.
~:l. ~ "..= ..~ ._
~ o4~ -~ ~
o,.,~
~ o ~ ~= ~.~ ~
Figure 4.36 Diagrams (not to scale) to illustrate the essentials of coastal upwelling (here shown for the Northern Hemisphere).
(a) Initial stage: wind stress along the shore causes surface transport 45 ~ to the right of the wind, and Ekman transport (average motion in the wind-driven layer) 90 ~ to the right of the wind (cf. Figure 3.6(b)). (Note: this shows the idealized situation which is never observed in reality.) (b) Cross-section to illustrate the effect of conditions in (a): the divergence of surface waters away from the land leads to their replacement by upwelled subsurface water, and to a lowering of sea-level towards the coast.
(c) As a result of the sloping sea-surface, there is a horizontal pressure gradient directed towards the land (black arrows in (d)) and a geostrophic current develops 90 ~ to the right of this pressure gradient. This 'slope' current flows along the coast and towards the Equator. The resultant surface transport, i.e. the transport caused by the combination of the surface transport at 45 ~ to the wind stress and the slope current, still has an offshore component so upwelling continues.
(d) Cross-section to illustrate the variation with depth of density (the blue lines are isopycnals) and pressure (the dashed black lines are isobars and the horizontal arrows represent the direction and relative strength of the horizontal pressure gradient force). Isopycnals slope up towards the shore as cooler, denser water wells up to replace warmer, less dense surface waters. The shoreward slope of the isobars decreases progressively with depth until they become horizontal; at this depth the horizontal pressure gradient force is zero, and so the velocity of the geostrophic current is also zero. At greater depths, isobars slope up towards the coast indicating the existence of a northerly flow; a deep counter-current is a common feature of upwelling systems.
135 Note that the current arrows on Figure 4.36(a) and (c) are idealized
representations of a steady-state situation, assuming a fully developed Ekman spiral, resulting in Ekman transport at right-angles to the wind (Section 3.1.2).
Even if the wind was steady for some time, a fully developed Ekman spiral would not be possible in shallow coastal waters, and in reality, upwelling occurs in response to particular wind events, which might be quite short- lived. Thus, the actual pattern of isopycnals and along-shore current flow varies from time to time, depending on the direction and strength of the wind, and is also affected by local factors like the topography of the sea-bed and the shape of the coastline. Three examples of upwelling regimes are shown in Figure 4.37 - note that they all include a poleward-flowing counter-current, which is found in most upwelling regions in eastern boundary currents.
The ecological importance of coastal upwelling lies in the fact t h a t - like upwelling in cyclonic gyres (Figure 3 . 2 4 ) - it usually replenishes surface waters with nutrient-rich sub-thermocline water, stimulating greater productivity of phytoplankton (and hence supporting higher trophic levels).
Figure 4.38 illustrates the marked effect that wind has on upwelling, and hence primary productivity, in the surface waters of the Canary Current off north-west Africa. Upwelling off the coast of north-west Africa occurs in response to the North-East Trade Winds. To the north of about 20 ~ N, there is a region where the Trade Winds blow along the coast all year, but to the north and south of this, the wind direction changes seasonally (cf. Figure 2.3).
As a result, although upwelling occurs all year round to the north of Cape Blanc, further to the north and to the south of Cape Blanc it varies seasonally.
Similar seasonal variations are observed in all eastern boundary currents, and there are also marked differences between one year and a n o t h e r - compare Figure 4.38(b) and (c) for the month of November in 1982 and 1983.
Coastal upwelling is most marked in the Trade Wind zones, but it can occur wherever and whenever winds cause offshore movement of water.
Nevertheless, it is a difficult process to investigate directly because it occurs episodically, and because the a v e r a g e speeds of upward motion are very low - generally of the order of 1-2 metres per day though sometimes approaching 10 metres per day. Indirect methods must be used and, as with indirect methods of measuring currents, these may be based on either the causes or the e~'ects of upwelling.
The cause of coastal upwelling is the offshore movement of water in response to wind stress. Since the upwelled water rises to replace that moved offshore by the wind. the rate at which water upwells is the same as that with which it moves offshore. Hence, the rate of upwelling may be calculated using Equation 3.4, which tells us that it must be directly proportional to the wind stress and inversely proportional to the sine of the latitude. This method of calculating the rate of upwelling gives reasonable results only if the
assumption that a steady state has been attained is valid, a n d there is adequate information about local winds. However, as mentioned earlier, the average speed of the wind is not the best indicator of the amount of upwelling it will induce, because the wind stress is proportional to something like the square of the wind speed (Equation 3.1). Thus, occasional strong winds have a
disproportionately large influence on water movement, and upwelling rates fluctuate greatly in response to fairly small changes in wind speed.
Figure 4.37 Three examples of upwelling regimes over different continental margins. In each case, the wind is equatorward and out of the page.
The darker the blue-green tone, the denser the water. The diagrams show the uppermost 200 m or so of the water column; the vertical scale is greatly exaggerated. Note the fronts; these tend to develop wave-like instabilities, eddies and filaments (cf. Figure 4.38).
Figure 4.38 CZCS images showing the concentration of chlorophyll-a pigment in surface waters off the coast of north-west Africa, in each case averaged over about a month: (a) March-April 1983; (b) November 1982; (c) November 1983. In these images, clouds and land are shown as white and turbid coastal waters are black. The colours represent pigment concentration according to a logarithmic scale: dark blue is smallest concentration; dark green largest. Note the wave-like undulations, eddies and filaments, made visible by their higher chlorophyll content.
Figure 4.39 Mean anomaly in the sea-surface temperature off north-west Africa for April. The largearea with a negative anomaly (i.e. region of significant difference between actual surface temperatures and average surface temperatures for these latitudes)is mainly attributable to upwelling (cf. Figure 4.38).
.~ = ~~ ~. ~ ~. "" ~ ~ 0 --. ~_~ ~.= ~~ ~- = ~ ~ ~ ~ ~ ~ ~ c~ m" ~ ~"---" ~ ~ B'~ < ~.o ~ 9 __ ~,< ~ ~ =
-. ~. _~. z~=-~-~~~'~" ~ o_,~ = ~7~o=.~~ ~ ~. ~-~ ~ ~
, ~_~ 9_._~ o~.o~=o~ ~~.~o ~~ ~ ~-
~ ~ '-'" " "~ ~ ~ "* C:~ _ ~-.~ o~ ~o~-~~ ~
~ ~ r.~ ~ ~-~ ~ ~.~ ~.~ -.~ o -~ ~~ =o--4 ~ o o ~.~ =. ~-~~ ~--~~ -~- ~ ~~ ~~176 .~-
~ ~ ~ , ~ ~o = ~ ~ ~ ~~ >~.~ ~~
~-~" (-~o ~~ o ~< ~ -~ =-=..~.~ .... t'~ ~ r ~
~~o~ "~ ~~.o ~ ~F~ ~-.o ~. ~. ~4~ ~ ~ ~. ~ ~ ~.-
_.. ~ .-. r ~ ~---.. ~~_o~~.~ ...
~~~ ~ ~-~
~" ~" ~" ~ ~.~ ~ ~"-~.~~.~ ~ ~ -~ o
~.~& ~ ~ ~ ~ 8 ~ ~'.-,~ ~~ ~-~ ~ ~- ~
- ----o ~'o --" 9 ~ "-,,1The state of the NAO at any one time is described in terms of the NAO Index. This is usually expressed simply as the wintertime pressure
difference between certain meteorological stations in Iceland and the Azores (or, sometimes, Lisbon in Portugal) (Figure 4.40). The Index may also be expressed in terms of the departure of the observed Iceland-Azores pressure difference from the average. In this case, when the pressure difference is higher than usual, the NAO Index is described as positive, and when the pressure difference is lower than usual the Index is described as negative.
Generally, when the pressure difference is large the NAO is described as strong, when it is small the NAO is described as weak.
Figure 4.40 Record of the wintertime difference in atmospheric pressure (reduced to sea-level) between the Azores High and the Iceland Low over the course of the twentieth century. The black line is a 5-year running mean, i.e. a line joining average values of the Index through successive 5-year periods.
The state of the NAO affects not just wind speed and direction - westerlies are stronger when the NAO is strong - but also the paths followed by winter storms, rainfall patterns and transport of heat and moisture between the North Atlantic and the surrounding land masses, and even the strength of the North-East Trades. The importance of the state of the NAO is illustrated by Figure 4.41 which compares conditions during the winter of 1994-95, when the NAO was strong (positive Index), with conditions during the winter of 1995-96, when it was weak (negative Index).
Returning to the apparently mysterious links between the position of the Gulf Stream in the western Atlantic and climate-related ecological
responses in the British Isles and North Sea, it seems likely that the linking factor is heat transfer between the Gulf Stream/North Atlantic Current and the overlying atmosphere far downstream of Cape Hatteras. When the NAO is strong (positive Index) and surface westerlies are strong, the Gulf Stream and North Atlantic Current flow more strongly than usual, and on average their paths are further north; as a result, more heat and moisture is transferred to north-west Europe, which becomes warmer and wetter. The reverse is true when the NAO is weak. However, as Figure 4.41 suggests, the NAO is more complex, and its effects more wide-ranging, than this simple explanation suggests.
139
Figure 4.41 Schematic maps to show the pattern of sea-level atmospheric pressure over the North Atlantic region, and associated extreme conditions, in (a) the winter of 1994-95, when the NAO was positive, and (b) the winter of 1995-96, when the NAO was negative. In 1994-95, the Icelandic Low and Azores High were both more intense than usual. As a result, strong westerly winds (black arrow) steered winter Atlantic storms eastward and brought unusually wet conditions to northern Europe and warmer conditions to a band extending from North Africa to Siberia. In 1995-96, the Icelandic Low and Azores High were both less intense than usual. These weak pressure systems led to a reduction in the influence of the Atlantic over northern Europe, with less rainfall and much lower temperatures.
As far as the atmosphere is concerned, the NAO does not seem to have an obvious periodicity (Figure 4.40) or, rather, it is a m i x of periodicities. Over the second half of the 20th century, it remained in one or other state for five or more years. During most of the 1950s and 1960s, the NAO was weak, and during most of the 1980s and (especially) 1990s the NAO was strong;
sometimes, however, the change from a strong NAO to a weak NAO took place from one winter to the next (cf. Figure 4.41). The ocean (which tends to retain the effects of winter atmospheric conditions) has a slower response time than the atmosphere, and a longer 'memory', so it responds to
relatively short time-scale fluctuations in the atmosphere on a roughly decadal (i.e. ten-year) time-scale. It is generally considered (as will be discussed in Chapter 6) t h a t - except in specific circumstances such as E1 Nifio events - the large-scale pattern of sea-surface temperature is affected by the overlying atmosphere, rather than vice versa. Researchers interested in the NAO have the task of determining whether the ocean might play an a c t i v e role in atmosphere-ocean coupling at higher latitudes, too.
1 The subtropical gyres are characterized by intense western boundary currents and diffuse eastern boundary currents. In the North Atlantic, the western boundary current is the Gulf Stream, and the eastern boundary current is the Canary Current.
2 Exploration of the east coast of America, followed by colonization, trading and whaling, led to the western North Atlantic in general, and the Gulf Stream in particular, being charted earlier than most other areas of ocean. Some of the most notable charts were made by De Brahm and by Franklin and Folger (late eighteenth century) and Maury (mid-nineteenth century): Maury was also the first to encourage systematic collection and recording of oceanographic and meteorological data.
3 The Gulf Stream consists of water that has come from equatorial regions (largely via the Gulf of Mexico) and water that has recirculated within the subtropical gyre. The low-latitude origin of much of the water means that the Gulf Stream has warm surface waters, although the warm core becomes progressively eroded by mixing with adjacent waters as the Stream flows north-east.
4 The prevailing Trade Winds cause sea-levels to be higher in the western part of the Atlantic basin than in the eastern part, and the resulting 'head' of water in the Gulf of Mexico provides a horizontal pressure gradient acting downstream. The flow leaving the Straits of Florida therefore has some of the characteristics of a jet.
5 The Gulf Stream follows the continental slope as far as Cape Hatteras where it moves into deeper water and has an increasing tendency to form eddies and meanders: the flow also becomes more filamentous, with cold counter-currents. Beyond the Grand Banks, the current becomes even more diffuse and is generally known as the North Atlantic Current.
6 Flow in the Gulf Stream is in approximate geostrophic equilibrium, and the strong lateral gradients in temperature and salinity mean that the flow is baroclinic. Confidence in practical application of the geostrophic method was greatly increased when it was successfully used to calculate
geostrophic current velocities in the Straits of Florida.
7 The fast, deep currents in the Gulf Stream are associated with the steep downward slope of the isotherms and isopycnals towards the Sargasso Sea.
The Gulf Stream may be regarded as a ribbon of high velocity water forming a front between the warm Sargasso Sea water and the cool waters over the continental margin. This is a frontal region with large lateral variations in density, and so is subject to wave-like perturbations known as 'baroclinic instabilities'. These lead to the formation of mesoscale eddies, especially downstream of Cape Hatteras. Eddies formed from meanders with an anticyclonic tendency are known as warm-core eddies, and those formed from meanders with a cyclonic tendency are called cold-core eddies.
The importance of mesoscale eddies only began to be appreciated in the 1970s, as a result of MODE and other similar projects.
8 One of the tools for studying fluid flow is the concept of vorticity, or the tendency to rotate. All objects on the surface of the Earth of necessity share the component of the Earth's rotation appropriate to the latitude: this is known as planetary vorticity and, because vorticity is defined as 2 x angular velocity,
~ ~ o~- F~
~" ~.~ ~. ~=---4
~ e e ~~'~ F. _~. o~ ~,~ o~ ,- ~B B ~ ~ ~-
kk
,~. r~ ~, -" ~~k_.= = :.o ~ ~_~ "-" 9 . -~.~ ~ =.~ =:~-= =o ,- =. ~'o~~-.~
~~.~ ~ ~i'D "u =~ o ~,~ ~
~.o~ ~.~
B B =~= o '- ~F' o ~.~ =~'~ = ,-. 9 ~ ~.~< ~ =__. =__. =p= r~== ~.~.> ,,.... ,--. B ~.o~- .~ ~= _.~
~~.
~ ~ r~ "~"~<~ = -,Fo=
='~ ,.--- ~ B r~ u-a~
,_. r/~ ~-~ 9 ~ ,-. r~ u" r~ ~.-,. t~~
0 ~ ,-,= ~ = ~. ~-~. ~~ o~x~ ~ = ~_ o_., ~ B"u = =. =-~8 p~ ~. ~..~ ~'~-~
~.~ s~
-- "-1 ~ 0 r~ -- ~ = ~ ?:i. '-i o ~. 9~.s =_.. d~
9 rD ~ . ,,< --
~-~~
~<~" <16 Satellite radar altimetry may be used to determine variations in the height of the sea-surface (i.e. departures from the geoid), and hence effectively provide a direct measure of the dynamic topography of the sea-surface.
17 The eastern boundary currents of the subtropical gyres are associated with coastal upwelling which occurs in response to equatorward longshore winds. Areas of divergence and upwelling are characterized by cooler than normal surface waters, a raised thermocline and, because nutrients are continually being supplied to the photic zone, high primary productivity.
18 The most important factor affecting wintertime climatic conditions over the northern Atlantic Ocean and the Nordic Seas is the state of the North Atlantic Oscillation (NAO). This is a continual oscillation in the difference in atmospheric pressure between the Iceland Low and the Azores High. A strong NAO (positive NAO Index) leads to strong westerlies, and warm wet winters in north-west Europe. In the ocean, the variability associated with the NAO has a roughly decadal time-scale.
Now t ~ the following questions to consolidate your understanding of this Chapter.
~.~. ~.~ ~ ~ ~ ~ ~ "-" N 9 :~ "I ~ ""I ~+++m+ '-< ,_., m< s r~ "<
+q -
-'- "* 0"0 I:~ = ~'~'~ o~o-~=~%~
m O~ ~ ~ 0 9= ~~' = ~ ~ ,~ ~ -. . ~= 9 ~ ~ ~ ~ .~ ~ ~ ~ _ --- ~ ~-.~~- 9 =~ ~ ~- ~ ~--.~ ~. ~.~.~. ~ -~~. ~-~ ~ ~= ~- -~=. ~. . ~-= -" ~ "~ ~" ~ ~ ~ .0 9 -., ~. 0"~ ~__, ~ m m.. ~ 0,. ~ -- ~ ~. 0 9 ~ ~'~.~ ~~= ~-~ ~ ~ ~. ~= ,~ ~=~ ,.. .. o ;,m. . _. ~ q'~ - ~ %9 ~ -~~
~ +--S~ = - . _ - ~=-' [+-=2 ~ = -p+ J ~.s+m m - ~
=+ +=,-'<=
__ ++ ~.~. +,+ @ ~ ~ ~ ~ ~-~-~ ~ o ~~ ~o_+.+ ~ =,- ~ ~--'+ ~ -,~ 9 ~ ~ ~ ~.+ 9 ~ ~,~. "" ~ ~ ~ ,.-+.. ~ 9 + ~'~ 2r~ '~ -~ """ ~,~ 9 ~r' 9o off ~=-~ ~ "" ~ ~ "
~-=~.~~
~ H~ ~- ~.~ "~ ~ " ~ ~ ~- ~ ~" ~ -'" ~ 0 o.~o _~ ~..~ 0oo ~~~
~~ ~ .~ ~ "1 ~ I~., '-'" ~ =o -o ,--- ~ ~ "-' -,. ~ ~ ,~" ooFigure 5.1 (a) The relationships between the wind direction, the surface current and the Ekman transport (the total transport in the wind-driven layer, shown by short blue arrows), in equatorial latitudes. Note the Doldrum belt between about 5 ~ and 10 ~ N.
(b) North-south diagrammatic section showing the vertical and meridional circulation in equatorial latitudes, and the shape of the sea- surface and thermocline. Regions of eastward and westward flow are indicated by the letters E and W. The darker blue region (in which geostrophic current is assumed to be zero, cf. Figure 3.20(a)) is the deep water below the thermocline. The blue oval at about 100 m depth at the Equator represents the Equatorial Undercurrent (see Section 5.1.1). (Note that the vertical scale is greatly exaggerated.)
Figure 5 l(b) is a dia,,rammatic north-south section across the Equator, showing the vertical and meridional circulation in the mixed surface layer above the thermocline/pycnocline. There is little horizontal variation in density in the upper part of this layer, particularly in the top few tens of metres, and surfaces of constant pressure are more or less parallel to surfaces of constant density, and both are parallel to the sea-surface; in other words, conditions here may be regarded as barotropic. By the depth of the thermocline, however, there are significant lateral variations in density and conditions are baroclinic. The slopes in the thermocline are therefore contrary to those of the sea-surface.
145 Figure 5. l(b) is most easily interpreted in terms of easterly and westerly geostrophic flow if we use the simplified approach illustrated in
Figure 3.20(a), and assume that all of the water above the bottom of the thermocline behaves as one homogeneous layer (shown lighter blue).
This is a justified assumption in low latitudes, because here the density gradient, or pycnocline, between the mixed surface layer and deeper, colder waters is relatively sharp (cf. the dashed profile in Figure 3.19(a)).
Westward flow in surface waters is also a direct result of the Trade Winds. These blow at about 45 ~ to the Equator and surface flow in the South and North Equatorial Currents is 45 ~ cure sole of this, i.e. to the west in both cases.
The easterly E q u a t o r i a l C o u n t e r - C u r r e n t between about 4 ~ and 10~
may be explained as follows. The overall effect of the Trade Winds is to drive water towards the west, but the flow is blocked by the landmasses along the western boundaries. As a result, in equatorial regions the sea- surface slopes up towards the west, causing an eastward horizontal pressure gradient force. Because winds are light in the Doldrums, water is able to flow 'down' the horizontal pressure gradient in a current that is contrary (i.e. 'counter') to the prevailing wind direction. Furthermore, even this close to the Equator, there is some deflection by the Coriolis force. The deflection is to the right and therefore towards the Equator, and so contributes to the convergence at about 4 ~ N (along with the transport across the Equator resulting from the South-East Trades). The sea-surface therefore slopes up from about 10~ to about 4 ~ N, giving rise to a northward horizontal pressure gradient force which drives a geostrophic current towards the east.
This general pattern of westward-flowing North and South Equatorial Currents and an eastward-flowing Counter-Current may be observed in all three oceans. The South Equatorial Current is, on average, the broadest and strongest. Often in the Atlantic and occasionally in the Pacific, it shows two cores of maximum velocity, one at latitude about 2 ~ N and the other on the southern side of the Equator at about 3-5 ~ S (cf. the two parts of the current in Figure 5.1(b)). The Equatorial Counter- Current is best developed in the Pacific, where it reaches its maximum speed between 5 ~ and 10 ~ N, some distance below the surface. In the Atlantic, the Equatorial Counter-Current is present throughout the year only in the eastern part of the ocean, between 5 ~ and 10 ~ N, where it is known as the Guinea Current. In both the Pacific and the Atlantic, there is also a weak South Equatorial Counter-Current which can be
distinguished in the western and central ocean between about 5 ~ and 10 ~ S (not shown in Figure 5.1). For this reason, its counterpart to the north is often referred to as the North Equatorial Counter-Current.
Figure 5.2 Schematic diagram to show the structure of the equatorial current system in the central Pacific, at 170 ~ W, down to a depth of 1000 m (i.e. much deeper than in Figure 5.1(b)).
Westward flow in the North and South Equatorial Currents (NEC and SEC)is shaded pale blue;
strong westward flow (the Equatorial Intermediate Current, EIC) is darker blue.
Eastward flow (including the South Equatorial Counter-Current, SECC)is unshaded; areas of strong eastward flow in the Equatorial Undercurrent (EUC), the North Equatorial Counter-Current (NECC) and the North and South Subsurface Counter-Currents (NSCC and SSCC) are outlined in blue. Note that a banded pattern is also shown for the northern subtropics, where counter-currents have been observed. The numbers are volume transports in sverdrups (106 m3 s-l). Those in red are for 155 ~ W, based on data for April 1979-March 1980, and those in green are for 165 ~ W, based on data for January 1984-June 1986.
As oceanographic instrumentation and techniques have advanced, it has been possible to distinguish finer details in the patterns of ocean circulation.
Figure 5.2 shows a schematic section across the Equator in the central Pacific at 170 ~ W, based on measurements made in the late 1970s and mid-
1980s. This shows not only the equatorial currents mentioned above, plus the Equatorial Undercurrent (to which we will turn shortly), but also, beneath the Equatorial Undercurrent, North and South Subsurface Counter- Currents flowing eastwards either side of a strong westward-flowing Equatorial Intermediate Current. which is effectively an intensification of the South Equatorial Current. The values for volume transport given on this diagram (for different years and different longitudes) illustrate how variable the equatorial current systems can be.
The equatorial current system of the Atlantic does not have the generally simple east-west flows of the Pacific (Figure 3.1 ). This is partly because of the relative narrowness of the ocean basin (it is about one-third the width of the Pacific) combined with the influence of the shape of the African and American coastlines.
5.1.1 THE EQUATORIAL UNDERCURRENT
The Equatorial Undercurrent (Figures 5. l(b) and 5.2) is a major feature
of equatorial circulation. Such an undercurrent occurs in all three oceans, although it is only a seasonal feature in the Indian Ocean.
The effect of the wind is transmitted downwards to deeper layers via
turbulence (eddy viscosity) and is mainly confined to the mixed surface layer above the thermocline/pycnocline (Section 3.1.1 ). At high latitudes, winter cooling of surface waters causes them to become denser, destabilizing the upper part of the water column, so that it may more easily be mixed by wind and waves. At low latitudes, there is no winter cooling and the mixed surface layer is thin - as shown in Figure 5. l(b), in the vicinity of the Equator it may only be 50--100 m thick. In addition, as mentioned earlier, the pycnocline is a