low-latitude tropospheric circulation

low-latitude tropospheric circulation The circulation of air in low latitudes is taken here as referring to circulation in the region of the globe where the Sun is within 10° of zenith at least once a year. Most textbooks lead one to believe that the structure and functioning of the atmosphere in this zone is simple, with little variability in the day-to-day wind flow. Interestingly, there is as yet still no simple theoretical framework similar to that which has been developed for the mid-latitudes (i.e. quasi-geostrophic theory) that can be used to provide an overall understanding of the tropical atmosphere. Instead, largely because of problems of scale, the search for an integrated mathematical framework for the tropics can be regarded as one of great remaining problems of global atmospheric circulation.

One of the simplest criteria for evaluating the large-scale structure of the atmosphere is the distribution of pressure. The surface pressure at 30° N or S is higher than the pressure near the Equator. This drives the surface easterlies (easterly because Coriolis force deflects the flow) or trade winds towards the Equator. Evaporation removes heat and water from the oceans and transports it equatorwards. Consequently, the near-surface air flow is both warm and moist. Convergence and forced ascent near the Equator lead to heavy precipitation at the Intertropical Convergence Zone (ITCZ) or Intertropical Confluence (ITC). The release of latent heat and the convergence of sensible heat flux both help to drive strong upwards motion. Latent and internal energy are here converted into potential energy and exported towards the pole in the upper air. This export exceeds the sum of the flow of latent and internal energy towards the Equator in the lower branch, giving a small net flow of energy towards the pole. As a result of the constraints of the angular momentum balance, the meridional cell is not a particularly efficient means of polewards energy transport.

Ascent in the ITCZ is balanced by descent in the subtropics. This is evidenced by subtropical highs, intense subsidence, clear skies and deserts (e.g. Namib, Karoo, Kalahari, Sahara, Atacama). Diurnal temperatures range by up to 50 °C, because of intense solar radiation by day and loss of terrestrial radiation by night in a dry, cloud-free atmosphere.

Overall, the low latitudes are characterized by a lack of temperature gradients away from land/sea boundaries. This, in turn, means that the atmosphere is driven by energy sources very different from those of the mid-latitudes, whose overlying atmosphere relies on available potential energy derived from the tilting of the atmosphere, itself brought about through temperature contrasts. Instead, the large-scale overturning in the tropical atmosphere is maintained largely by the release of latent heat.

These overturning thermally driven cells in the low latitudes of both hemispheres are known as the Hadley cells (Fig. 1), in honour of George Hadley (1685–1768), who assumed that such a circulation extended from the Equator to the poles. The northern cell dominates from November to March, the southern cell from May to September, with rapid transitions between April and October.

Many of the characteristics of the tropical atmosphere come about because of the small value of the Coriolis parameter, or, put differently, the small effect of the Earth's rotation, which allows air to flow more or less directly from a region of high pressure to one of low pressure. Regions of different pressure and temperature are thus cancelled out efficiently. By contrast, in the mid-latitudes the large effect of the Earth's rotation causes much greater deflection of the wind; in consequence, airflow rotates around regions of low pressure or high pressure, essentially maintaining differences in pressure and temperature.

Unfortunately, while this overview of the low-latitude atmosphere provides the most convenient insight into broad-scale functioning, it also conceals features of the tropical atmosphere which make it both interesting and complicated. For example, the simple model of a large-scale ascent in the ITCZ at the Equator is not consistent with either observation or theory. The large-scale tropical atmosphere is stable above approximately 5 km. Large-scale ascent to the top of the atmosphere would thus have to take place through a stable layer or, in other words, up the energy gradient. This would in effect mean that the upper troposphere would be cooled by the ascent, a situation which would not satisfy the heat imbalance of the tropical zone. Riehl and Malkus showed that ascent needs to occur within clouds, in the so-called hot towers, where the air is saturated. The warmer air could still arrive at the top of the atmosphere by travelling inside the cloud conduits. The ITCZ thus occurs as a collection of cloud clusters rather than as a large ascending branch. At any one time, 1500 to 5000 clouds or hot towers would be necessary to account for the required heat transport of the tropics.

The outcome is that in order to understand and integrate the dynamics of the global tropics, one must really understand and integrate the dynamics at the cloud scale. This introduces one of the most difficult scale problems imaginable in the study of fluids. With this context in mind, we can consider some of the features of the tropical atmosphere, including the ITCZ, easterly waves, tropical cyclones, the monsoons, and the interannual variability of the tropics.

Averaged over time, the ITCZ appears as one of a line of deep convective clouds, for example in the Atlantic and Pacific Oceans between 5 and 10° N. On a day-to-day basis, however, the feature appears as a discrete but transient cloud cluster several hundreds of kilometres in diameter separated by regions of clear skies. Over the continents, the ITCZ clusters are much larger and less zonally constrained, essentially spreading out over the heated land mass. Three equatorial regions of maximum cloudiness are apparent: South America, Africa, and Indonesia. The last-mentioned is an equatorial location, with shallow seas and relatively small land masses, giving rise to the region of greatest rainfall. In fact this region provides a centre for strong thermal forcing of the rest of the tropical atmosphere.

The flat Amazon Basin, oriented towards the moisture-bearing easterlies from the Atlantic, receives more than 2 m of rainfall a year. The moisture is prevented from flowing out over the Pacific by the large barrier of the Andes. Convection leads to release of latent heat and the warming of the lower troposphere, which, in turn, lowers the pressure and increases the influx of surface moisture. Maximum rainfall occurs during the early months of the year and a minimum in August, when precipitation occurs over Central America.

Seasonal shifts in the ITCZ are largest over the land, because of the very small effective heat capacities of land surfaces (Fig. 2). The most dramatic seasonal movement in the mass of convection is centred over Indonesia, with both north–south and east–west components. During the summer of the southern hemisphere, the convection extends southwards over the south Pacific Ocean and is identifiable as far away as 30° S, 140° W. This feature is called the South Pacific Convergence Zone (SPCZ). During the northern summer the SPCZ is retracted back across the dateline, with Indonesian convection extending north-westward into the Bay of Bengal, where it becomes connected with the convection associated with the Asian summer monsoon.

Embedded within the tropical easterly flow are weak equatorial wave disturbances driven by latent heat release, which propagate westward at speeds of 8–10 m s⁻¹ along the ITCZ. The cloud bands are separated by 3000–4000 km with characteristic periods of 4–5 days. The largest disturbances and the highest precipitation tend to occur when mid-tropospheric temperatures are warmer than average (although usually by less than 1 °C). These disturbances involve an interaction in which large-scale convergence at low levels moistens and destabilizes the environment so that small-scale thermals can easily reach the level of free convection and produce deep cumulus clouds. The clouds then act as a large-scale heat source that drives the secondary circulation responsible for low-level convergence. This is true for most parts of the tropics, but in North Africa waves feed from an easterly jet which is driven largely by temperature contrast between the Sahara and equatorial regions.

These systems have the role of converting potential (derived from release of latent heat and ascent) to kinetic energy. This is worth emphasizing, since mid-latitude westerly waves retain potential energy with little conversion to kinetic energy. At times, easterly waves develop into disturbances with closed isobars, which may follow a continuum of intensification from easterly storms to tropical storms (winds of more than 19 m s⁻¹ at 10 m), culminating in tropical cyclones (sea-level pressure of 950 hPa or less, pressure gradients of 1 hPa km⁻¹, winds in excess of 33 m s⁻¹ at 10 m level), which are the most powerful weather disturbance on Earth. Roughly 80 occur per year, forming over oceans where surface temperatures exceed 27 °C over an area of 5° of latitude square (Fig. 3).

Mature tropical cyclones consist of a dense white ring of cumulonimbus and cirrus extending a couple of hundred kilometres out from the centre above a ring of updraft surrounding a central eye. In the eye, pressures drop so low that the stratosphere sinks to the surface; the sinking and warming air is therefore clear. The spirals, arms of cloud stretch-ing hundreds of kilometres outward, often equatorwards, from the central structure, comprise long, curving lines of cumulonimbus.

The warm core of the tropical cyclone sets it apart from other tropical disturbances, which are cool-cored (though penetrated by the hot towers discussed earlier). A warming of the core by 6 °C compared with the surrounding air mass, for example, would lead to a drop in surface pressure of the tropical cyclone of about 50 hPa. Positive feedbacks between frictional convergence in the surface layer supply moisture to the base of the storm. Cumulonimbus convection then strengthens the convergence and hence total growth.

The central pressure of the storm usually rises when it reaches land, which cuts off the moisture supply and therefore energy supply in the form of latent heat. Many storms then travel westwards and then polewards. An interesting scientific advance has been the seasonal forecasting of tropical cyclone frequency in the Atlantic on the basis of sea-surface temperatures, the state of the El Niño–Southern Oscillation (ENSO) event, the quasi-biennial oscillation in the stratosphere, and the rainfall in the Sahel.

While tropical cyclones represent one of the smaller, yet most intense types of tropical disturbance, the monsoons, with a near-global influence, are among the largest. This seasonally reversing circulation system is driven by temperature contrast between land and sea. Land has a different thermal response because the heated layer is thin and its heat capacity is small. Solar radiation thus raises the temperature much more quickly over land. Surface warming leads to cumulus convection and release of latent heat, producing warm temperatures throughout the troposphere. A pressure difference sets in between the land and the sea to drive a direct thermal circulation with low-level convergence and upper-level divergence over the land. The positive correlation between vertical motion and the temperature field points to the role of the monsoons in converting eddy potential energy to eddy kinetic energy, which is frictionally dissipated.

One of the largest meridional seasonal shifts occurs in the Indian–Asian regions and is associated with the monsoon. As a result, India experiences warm, wet summers and cool, dry winters. The monsoon usually becomes established over the Indian subcontinent by late June. Southerly flow into the monsoon is a continuation of the south-easterly trades of the southern hemisphere, except that they are deflected to the right by the Coriolis force in the northern hemisphere to become south-westerlies. These winds evaporate moisture off the Arabian Sea, which reaches temperatures of 29 °C during the boreal summer. Slow-moving monsoon depressions lead to enhanced rainfall. A tropical easterly jet stream is a semi-permanent feature of the region in summer at latitudes of 15° N, developing in response to the heating gradient between the warmer north and the cooler equatorial region (a reversal of the norm). This jet is thus driven by a thermal wind mechanism.

The monsoon begins to withdraw in September and is absent by November. The easterly jet is replaced by a narrow westerly subtropical jet stream on the southern edge of the Himalaya. In winter the circulation reverses to low-level off-land flow as the land cools more quickly than the oceans. Cooling is particularly marked over the Himalaya. High pressure and dry, descending air occurs over the land and rain falls over the ocean. North-westerly cold surges of air occur during winter from the vast pool of air north and east of the Himalaya.

The tropical atmosphere is characterized by large interannual variability in pressure and rainfall. This has been shown to be forced to a large degree by variability in sea-surface temperature; the ENSO mechanism is the best-known example. A unique insight into this behaviour has been derived from the Hadley Centre Climate of the Twentieth Century experiments, in which a general circulation model was forced by observed sea-surface temperatures for the period 1900 to 1993. The model was run several times, but each run had different starting conditions. The degree to which each of the runs produced different interannual variability determined the degree to which the atmosphere was controlled solely by varying boundary conditions. Early results showed that the tropics respond directly to sea-surface temperature forcing, whereas the resulting interannual variability in pressure, temperature, or rainfall in the mid-latitudes is extremely sensitive to starting conditions. The mid-latitudes appear therefore to be more chaotic. An important consequence of this tendency of the tropical atmosphere is that seasonal rainfall may be forecast several months ahead. The forecasts are based on the state of the slowly evolving sea-surface temperatures.

R. Washington

Bibliography

Barry, R. G. and and Chorley, R. (1993) Atmosphere, weather and climate. Routledge, London.
Burroughs, W. J. (1991) Watching the world's weather. Cambridge University Press.
Meteorological Office (1991) Meteorological glossary. HMSO, London.