In the course so far, we have explored fundamental processes acting in the atmosphere: the cascade of energy; the relationships between pressure, density and temperature; moist processes; air mass stability; and the forces acting on winds. We have also seen how these processes lead to organised structures and air motion patterns at small to meso-scales in atmosphere: cloud formation, air convection, thunderstorms, and wind vortices such as tornadoes and tropical cyclones. The same processes also allow us to understand the organisation of the atmosphere at larger scales. In this lecture, we examine the general circulation of the global atmosphere.
Views of the Earth from space reveal constantly changing weather patterns. However, there is an overall structure to the worldÕs weather, in which weather systems have repeating spatial arrangements and annual cycles. A surprising amount of this structure can be deduced from the atmospheric principles we have already described. This lecture will examine Tropical circulation patterns, then aspects of circulation at higher latitudes.
The tropics play a very important role in the global climate system. The primary input of energy from the sun is greatest in the tropics, and this energy must be exported to other parts of the atmosphere to maintain a global energy balance. Due to their large area, the tropical oceans are particularly important, although their role in regulating global climate has only recently been recognised. In essence, the Tropical circulation consists of a pair of large convective cells known as the Hadley Cells, named after the 18th C. English meteorologist George Hadley, who first deduced their existence. An introduction to the Hadley Cells can be found here. In this lecture, we will examine aspects of the Hadley circulation with particular emphasis on processes of energy transfer. The Hadley cells consist of an ascending limb - the Intertropical Convergence Zone - associated with the zone of maximum global heating, and a pair of descending limbs - the Subtropical Highs. The ascending and descending limbs are linked by upper-level air flow towards the Subtropical Highs, and low-level flow towards the ITCZ (Trade Winds).
The Intertropical Convergence Zone is a mobile region characterised by deep, moist convection, associated with the zone of maximum heating or thermal equator. Within the ITCZ, areas of convergence and convection grow and decay, and the position and intensity of convergence varies, even on a daily basis. The ITCZ shifts northward and southward on an annual cycle, following the thermal equator, and brings the annual march of rains. The position and timing of movement of the ITCZ therefore has major implications for the onset and duration of rains in Africa (especially the Sahel), South America (the Amazon basin and the Altiplano), Asia (including the monsoon systems of India and Indochina), Indonesia and Australia.
Cloud and rainfall in the ITCZ is structured on several scales. At the smallest scale are individual thunderstorms triggered over warm oceans (>26° C) or warm, moist land surfaces. Such storms commonly penetrate the full thickness of the troposphere (c. 15 km in the Tropics), forming the largest cumulonimbus clouds on Earth (cumulo: heaped; nimbus: cloud). Condensation of moisture from warm air releases large quantities of latent heat, increasing the potential temperature of the rising air. For this reason, tropical thundertorm systems are known as hot towers.
At the next scale, cumulus clouds may be grouped together into Mesoscale Convective Areas (MCAs), up to 100 km across. In turn, several MCAs may comprise a Cloud Cluster, 100s - 1000 km across. In the Atlantic, 10 - 15 cloud clusters form per month near the ITCZ. Cloud clusters may be unstructured weather systems persisting for only 1 or 2 days. Others, however, are embedded within still larger structures known as Wave Disturbances. Some wave disturbances develop a circular, closed structure, and evolve into cyclones.
Wave disturbances are wave forms in atmospheric air pressure, and can be visualised either as wave-like variations in the altitude of pressure surfaces or wave-like variations in surface pressure. Tropical wave disturbances take the form of troughs of low pressure, which move westwards. These can be thought of as incursions of warm, low pressure air spreading polewards into embayments in the subtropical cells. Air behind (to the east of) the trough axis follows converging flowlines, encouraging uplift, so a concentration of convective activity occurs in this area. Conversely, air in front (to the west) of the trough follows diverging flowlines, and is characterised by descending, drying air. Due to this pattern of converging and diverging air, a series of weather zones can be identified around wave disturbances:
(1) A high pressure ridge ahead (west) of the trough, with fine weather, and scattered cumulus;
(2) The trough line, with well developed cumulus, and some showers;
(3) Behind (east of) the trough, with large cumulus and cumulonimbus, and heavy thundery showers.
In the tropical Atlantic, there are c. 90 wave disturbances annually during the June-November hurricane season, over half of which originate over Africa. This amounts to one system every 3-5 days. Wave disturbances change in character as wave propagates westward. The map below is modified from Figure 6.8 in Barry and Chorley (Atmosphere, Weather and Climate), which shows the Pacific between Hawaii and the Philippines on August 17th 1957. This map shows an incipient trough near Hawaii, which filled and dissipated over the next 24 hours. A well developed wave occurs near Wake Island: this was associated with spectacular cumulus extending to 9,100 m altitude some 300 miles behind trough axis. Within 48 hours this wave developed into circular tropical storm with winds up to 20 m per secÊ (but not a full hurricane). A strong closed depression of this type is located to the east of the Philippines on the August 17th map.
The map illustrates how pressure systems associated with the ITCZ are mobile, variable features. Within these waves and depressions, are large numbers of individual cumulus and cumulonimbus systems, each with its own life cycle. This emphasises the point that the rising limb is not a continuous, conveyor belt-style zone of uplift, but as the combined effect of innumerable transient weather events nested within larger regions of convergence and instability.
The Subtropical Highs are a striking feature of global surface pressure patterns. They are most persistent over the subtropical oceans, where they form large east-west elongated cells. In the southern hemisphere, they display higher pressure during winter (DJF), and extend their influence over the continents forming an almost continuous belt of high pressure. In summer, this belt is broken by heat lows over the continents. In the northern hemisphere, the seasonal pattern is different: over the oceans the highs extend over larger areas and have higher pressures during the summer (JJA). In winter, they are often connected to the cold continental highs over North America and Eurasia by high pressure ridges.
The position and form of the subtropical highs reflects both dynamic and thermal factors. The main dynamic factor is the Coriolis effect. Upper-level, poleward flowing air in the Hadley cells is deflected to the right in the northern hemisphere and to the left in the southern hemisphere, forming upper level westerly air flow in the subtropics where geostrophic balance is reached. Further upper-level poleward motion is thus prevented, and air converges at about 30° N and S. Intermittent upper level airflow also converges towards the same areas from the mid-latitudes. This convergence thus increases the atmospheric mass, creating high surface pressure. In turn, the high surface pressure generates the low level outflow required to balance the upper level convergence.
The thermal factors concern the changing energy balance of the air as it flows polewards. Air convected to the top of the troposphere in the ITCZ has a very high potential temperature, due to latent heat release during ascent in hot towers. Air spreading out at higher levels also tends to have low relative humidity, because of moisture losses by precipitation. As this dry upper air drifts polewards, its potential temperature gradually falls due to longwave radiative losses to space (this is a diabatic process, involving exchanges of energy between the air mass and its environment). Decreasing potential temperature leads to an increase in density, upsetting the hydrostatic balance and initiating subsidence. The subsiding air warms (as pressure increases towards lower levels), further lowering the relative humidity and maintaining clear-sky conditions. However, although the subsiding air warms, it does not do so at the dry adiabatic lapse rate. Continuing losses of longwave radiation (radiative cooling) means that the air warms at less than the dry adiabatic lapse rate (i.e. some of the adiabatic warming is offset by diabatic cooling). In short, during descent the potential temperature decreases, creating very stable, cloudless air masses (see diagram below). The relatively high density (i.e. mass per unit thickness of atmosphere) of these cooling air masses produces high pressure at the surface. In the subtropics, the tendency to subsidence and high pressure is greatest over the oceans, where surface temperatures, and consequent upward longwave fluxes, are relatively low. This is particularly marked where ocean circulation patterns advect cool water to low latitudes (e.g. eastern Pacific).
In the subtropical highs, the subsiding air does not extend all the way to the surface. Instead, the surface boundary layer is characterised by lapse rates at or close to the DALR, and potential temperatures somewhat lower than those in the upper subsiding layer. The difference in the potential temperatures of the upper and boundary layers is especially pronounced over the subtopical oceans, reflecting the evaporation of water and consequently weak atmospheric heating of the boundary layer. The relative humidity of the boundary layer is also much higher than that in the upper subsiding layer. The upper subsidence and boundary layer are separated by a zone in which potential temperature increases and relative humidity decreases steeply with height. This is known as the inversion layer or subsidence inversion. Convection commonly occurs in the boundary layer, but vertical development is strictly limited by the inversion layer.
T-phi diagram showing a sounding over the eastern Pacific (30° N, 140° W). Note how the potential temperature declines from 50° at 500 mb (c. 5,800 metres) to c. 30° at the top of the Inversion. This is the result of diabatic cooling (energy losses).
On the equatorward side of the subtropical highs, air within the boundary layer is drawn towards the lower pressures of the ITCZ, forming the Trade Winds. The Trade winds are subject to the (weak) Coriolis effect and are hence generally North-easterlies in the northern hemisphere and South-easterlies in the southern. An exception is in the Indian Ocean in the Northern hemisphere summer, when the ITCZ is drawn particularly far north of the equator by heating of South Asia. Winds originating in the High pressure cell in the southern Indian Ocean (Mascarene High) are initially South-easterly, become Southerly as they cross the equator (no Coriolis effect at the equator), then South-westerly as they are deflected rightwards at higher latitudes.
As the Trade Winds cross the subtropical oceans, they gain in heat and moisture. In consequence, the boundary layer thickens and the inversion (Trade Wind Inversion) occurs at progressively higher altitudes. Initially, convection within the boundary layer is limited, forming only small cumulus or low-level stratus. As the boundary layer thickens, however, vertical development of cumulus increases, and rainstorms become possible. Eventually, as the ITCZ is approached, upper level subsidence breaks down and full depth convection becomes possible (i.e. the full thickness of the troposphere: c. 15 km). The circuit of the Hadley circulation is thus complete.
In the Tropics, air over the land masses heats up more than that over the oceans, because only the land surface is heated by solar radiation, and less evaporation occurs. Consequently, heat lows (warm, low pressure areas) form seasonally over the tropical continents, characterised by rising air. Additionally, the continents create topographic barriers to zonal flow (westerly or easterly winds). This is most pronounced for the Andes. Thus, zonal airflow in the tropics is broken up into a series of cells circling the globe. This circulation pattern is termed the Walker Circulation, after Sir Gilbert Walker, who first described it in 1922-1923 while trying to identify the causes of variability in the Indian monsoon. Over the Atlantic and Pacific oceans, the fundamental Walker circulation consists of easterly Trade winds at the surface, with westerlies aloft. In contrast, mean airflow over the Indian Ocean and Africa is westerly, with easterlies aloft. Easterly airflow over the Atlantic and Pacific Oceans is accentuated by the effects of winds on the eastern boundaries of the oceans: wind stress combined with Eckman flow patterns in the waterÊ result in upwelling, bringing cooler water of southern origin to the surface. This results in an east to west increase in sea-surface temperatures, which in turn creates an atmospheric pressure gradient (high to east, low to west). This effect is particularly marked in the Pacific, where the pressure gradient and wind patterns result in the Pacific warm pool in the western Pacific. This is a ÔlidÕ of warm, low salinity and buoyant surface water >100m thick. The ocean surface is higher in the western Pacific than in east Pacific due to wind stress exerted by the Trades. The warm, moist overlying air overlying the Pacific warm pool is prone to instability and uplift, so, in the standard Walker Circulation pattern, there tends to be a concentration of convective activity in the western Pacific and Indonesia, and relatively stable conditions in the cooler eastern Pacific.Ê The average east to west temperature gradient in the Pacific Ocean is + 4¡ C; with an associated cloudiness gradient of +40%; and and annual precipitation gradient: of +1800mm. Periodic disruption of the Walker Circulation manifests as El Ni–o events.
ENSO refers to a quasi-periodic disruption of the Walker circulation over the Pacific, associated with the suppression of upwelling off Peru and major incursions of warm water from the west. El Ni–o originally referred to these warm waters off Peru. The name literally means The Little One, a phrase that colloquially refers to The Boy Child. The oceanic event was so named due to its usual appearance around Christmas. El Ni–o is part of a larger oscillating system in the ocean and atmosphere. In the basic Walker circulation, atmospheric pressure is high over the eastern Pacific, and low in the west: it is this pressure gradient force that drives the Walker circulation and maintains the Pacific warm pool. However, this may be disrupted and the pressure gradient reversed. The changes in pressure distribution are measured by Southern Oscillation Index (SOI), which is the surface pressure in Tahiti minus the surface pressure in Darwin, Australia, expressed in standard deviation units. Positive values of the SOI indicate higher pressure in the east, and are associated with the Walker circulation. Negative values of the SOI indicate higher pressures in the west, and are commonly associated with El Ni–o conditions. The close association between the Southern Oscillation and El Ni–o events has given rise to composite name for whole phenomenon: ENSO.
The Southern Oscillation is not regular. From the 1950s-mid 1970s the switch occurred every 3-4 years, but this is not typical. Negative values were uncommon between 1932-1939, and in the early 18th C, the return period of El Ni–os was 10 years. Since the mid-1970s, there have been predominantly negative values of the SOI, with severe and prolonged El Ni–os. Ê
The chain of processes underlying ENSO and its quasi periodicity is incompletely known, although understanding has increased enormously in recent years, largely as a result of the TOGA project (Tropical Ocean - Global Atmosphere). The onset of an El Ni–o is associated with the breakdown of atmospheric subsidence in the central Pacific. As a result,Ê lower atmospheric pressure develops, and the Trade winds weaken. This reduces wind stress on the sea surface, and the Pacific warm pool empties eastwards, carrying warm water towards south America and raising sea level in the eastern Pacific. The warm water is guided along the equator because the Coriolis effect resists any tendency for the waters to move into either hemisphere (such guided motions are called Kelvin waves: they are essentially similar to waves guided along a coastline). The warmer water encourages convection and lower atmospheric pressure, accelerating the water movement in a positive feedback. When the event ends, upwelling is reestablished off Peru, and the Walker circulation is resumed. The restablishment of upwelling is due to reflection of waves of warm surface water north and south along the coasts of the Americas: this flow thins the surface layer of warm water and brings deeper, cooler water towards the surface.
It has been argued that the cyclicity of events may reflect the westward return of surface water in westerly Rossby waves - large, low-frequency eddies of surface waters that move only west. These waves and the eastward flowing El Ni–o currents (Kelvin waves) cycle water through the ocean and may drive the atmospheric response. The idea is that an El Nino may be initiated by the reflection from the western boundary of the Pacific of an oceanic Rossby wave. This model had the great appeal that it could allow the prediction of an El Niño, since the Rossby waves can be observed crossing the ocean. The model successfully predicted the 1986-87 and 91-92 events almost a year in advance. However, the more frequent events of the 1990s (1993, 1994-95, 1997-98) were not predicted, indicating that the model is at best incomplete.
Alternatively, the trigger may be due to external disturbances. Tropical convection (organized large-scale thunderstorm activity) tends to occur in bursts that last for about a month, and these bursts propagate out of the Indian Ocean (this is known as the Madden-Julian Oscillation). If the storms are strong enough, or last long enough, then those eastward winds may be enough to start an El Nino. However, specific Madden-Julian Oscillation events are not predictable much in advance (just as specific weather events are not predictable in advance), and if Madden-Julian events are the main trigger, then El Ni–o will not be predictable. Numerical models that did not have the MJO storms were successful in predicting the El Nino events of 1986-87 and 1991-92, suggesting that Rossby waves were a main influence at that time. But those same models have failed to predict the events since then, and the westerlies have appeared to come from nowhere. Other external forcings, such as solar variability and volcanic eruptions have also been suggested, but there is little evidence in support of these ideas. Ê
El Nino events have severe economic consequences. The reduction of upwelling in the eastern Pacific reduces the supply of nutrients to the surface waters, with catastrophic effects on marine life. The fishing industry (especially anchovies) is hard hit. Additionally, warmer SSTs in the eastern and central Pacific bring severe weather, with frequent storms and heavy rain to the central Pacific Islands and Peru. During severe warm events, sufficient heat and moisture can be added to the central Pacific atmosphere to generate hurricanes. For example, in Tahiti hurricanes are rare, but during the1982-3 El Ni–o, five swept the island. Conversely. suppression of convection in the western Pacific leads to drought in Australia, Indonesia, Africa and India. Mid latitudes: disruption of circulation in Pacific during El Ni–o weakens the Trade Winds and suppresses subsidence in the sub-tropical highs. Consequently, the north Pacific westerly storm belt can penetrate further south than usual, forming troughs which move northwest over North America, bringing warmer, wetter weather. The effects can also be felt in Europe (which tends to be cooler and wetter). In the southern Hemisphere, mid-latitude depression tracks tend to be further north, bringing cooler and wetter weather to New Zealand, southern Chile and Argentina.
A site dedicated to all manner of Tropical weather, including cyclones:
An excellent site on ENSO and related phenomena:
An archive of monthly average sea-surface temperatures:
Back to Course Contents