As well as energy transfers between the earth’s surface and the atmosphere there are also energy transfers at a horizontal level, between low latitudes and high latitudes. Poleward of latitude 38° in both hemispheres there is less solar energy absorbed than terrestrial energy emitted; whereas between 38°N and 38°S there is more energy absorbed than emitted. Energy is transferred from the low latitude surplus areas to the high latitude energy deficient areas. Without this transfer of energy, the low latitudes would get increasingly hotter and high latitudes even colder.

early as 1735, George Hadley, a British scientist, suggested a tropical heat source powers global circulation and effects a transfer of energy from the Tropics to the Poles. Hadley proposed that direct heating of air at the equator causes upward convection and this air then moves poleward to sink at the subtropics. He indicated that similar cells might exist at high latitudes, however, it was left to William Ferrel in 1856 to put forward a three-cell model for each hemisphere.

Atmospheric Circulation Model

The simple three-cell model of atmospheric circulation proposed by Ferrel included the tropical Hadley cell with rising equatorial air and descending air at the subtropics (around 30° latitude). This cell also includes air movement at ground level between the sub-tropics and the equator to replace the rising equatorial air, which has created a zone of low pressure. This area of low pressure is caused by the heated air at the equator which expands, becomes less dense (heavy) and rises. As the heated air rises through the atmosphere the “weight of the air” on the surface is lessened and this results in a low-pressure area.

Differences in air pressure create air movement (wind) and in this case, the wind moves from the high-pressure subtropics to the low-pressure equatorial zone.

The rising equatorial air carries with it considerable amounts of latent heat in the evaporated water from the oceans and land surface. Some of this energy is released as the water vapour condenses in the form of cumulonimbus clouds. The expansion of the rising air in the upper atmosphere creates airflow in both northward and southward directions. More of the water vapour condenses and releases energy as the air moves poleward, warming the atmosphere in the middle latitudes. The poleward moving air subsides (sinks) in the subtropics because the space available in the upper atmosphere decreases poleward. This subsidence causes a subtropical high-pressure zone at about 30° latitude.

A polar cell in this model is due to cold, dense air subsiding in polar regions (polar high-pressure zone) and moving to lower latitudes where it expands as it moves into more space and is warmed by the earth’s surface.

The Ferrel cell between the other two is a response to the two thermally direct cells (circulation cells which owe their origin to temperature differences) and obtains energy from them. This cell feeds warm air to high latitudes and transfers cold air back to the subtropics for warming.

This simple three-cell model is a useful simplification of reality. It does not take into account the seasonal shifts in pressure belts or other forces governing the global circulation.

Forces Governing global circulation

As was mentioned earlier, air moves away from high-pressure areas towards low-pressure areas. This is sometimes called the pressure gradient force. It ought to cause winds to blow directly between high-pressure and low-pressure areas, at right angles to the isobars (lines of equal pressure). However, other forces deflect the winds.

The Coriolis force, named after the nineteenth-century French scientist, Coriolis, is the effect produced by the earth’s rotation. It must be allowed for when studying objects that move with respect to the earth’s surface. From space, winds (or other moving objects) are seen to move in a straight line with the earth rotating beneath. But from the earth’s surface, the effect of the Coriolis force appears to deflect the movement. (for this reason, the force is sometimes called “apparent”). It acts at right angles to the direction of motion, deflecting winds towards the right in the northern hemisphere and to the left in the southern hemisphere. This force affects all objects and fluids and therefore also influences the ocean currents.

The Coriolis force is zero at the equator (because the earth is spinning in a plane almost parallel to the axis of rotation) but increases poleward.

At high latitude, the wind tends to find a balance between the pressure gradient force and the Coriolis force. When the air begins to move from high to low pressure it becomes subject to the Coriolis force which displaces it to the right. As it speeds up, the deflection increases until the two forces are exactly balanced and the wind flows parallel to the isobars. This is known as the geostrophic wind. Near ground level, however, another force comes into play: friction. The friction of the air against the ground partly counteracts the Coriolis force and causes the wind to cross the isobars at an angle of between 20° and 45°. The more rugged the terrain the larger the angle, while for smoother surfaces such as water the angle is smaller. Friction also has the effect of reducing the speed of the wind.

Improvements in atmospheric circulation models

Research early in the 20th century revealed the existence of large-scale belts of fast-moving winds. These wind belts follow a wave-like pattern, first poleward then towards the equator and so on. These wave-like patterns are known as planetary waves or Rossby waves and occur at altitudes of 10,000 to 12,000 metres. Within these wind belts, there is a core of very fast-moving air (the jet stream) where the wind can reach speeds of up to 350-450 km per hour. There are two jet streams in each hemisphere: the polar jet stream which lies between about 30°and 50° latitude and the subtropical jet stream at about 20° to 30° latitude.

In the northern hemisphere, the polar jet stream flows west to east and the subtropical jet stream east to west. The jet streams and their associated Rossby waves are a mechanism for mixing air of different temperatures and are located at key locations in the atmosphere for the transfer of energy. The jet streams are generated by the temperature differences between polar and subtropical air and between subtropical and equatorial air. The polar jet stream is the most vigorous of the two because the temperature difference between polar and subtropical air is more marked than the difference between subtropical and equatorial air.

The Rossby waves themselves are linked to pressure changes at the surface and rotation of the earth but major physical barriers like the Rocky Mountains and the Andes have a great influence on the wave pattern and wind speed. The compression of air as the wind blows over the Rockies causes the wind first to blow south-eastward and then north-eastward This deflection results in approximately 4 or 5 waves that encircle the earth. The wave-like pattern would appear to have a life cycle of about 6 weeks. The wave amplitude and wind speed increase to the point where the polar air transfers southward in cells and the subtropical air moves into higher latitudes. Increased knowledge and understanding of Rossby wave activity and its influence on surface weather has enabled an improvement in forecasting reliability.

The addition of jet streams and a polar front (a boundary between the cold polar air and warm sub-tropical air) improves the simple three-cell model.

Further improvements in our understanding have led to modifications of this Rossby model. The modifications proposed by Palmen in 1951 are still generally accepted today.

The model is still a simplification of reality as the patterns of wind circulation are influenced by the distribution of land and ocean as well as high mountain barriers but, give a good understanding of the mechanics of global circulation.

Planetary wind circulation
The major wind systems largely conform to broad latitudinal zones. There are two main wind belts in each hemisphere. The trade-wind belt covers nearly half the earth's surface and is found between latitudes 30°N and 30°S. The trade winds are fairly constant and predictable because of the relative permanence of the subtropical high-pressure zones, from where they emanate.

The second major wind belt in each hemisphere is the mid­latitude westerlies which develop from the poleward sides of the subtropical high-pressure zone. The westerlies in the southern hemisphere are more persistent than those in the northern hemisphere because of the relative absence of large landmasses. Poleward of the mid-latitude westerlies are smaller polar zones of easterly winds. In the Arctic, these easterlies tend to occur only in winter while in Antarctica they tend to be less seasonal. The trade winds from both hemispheres meet near the equator in what is called the inter-tropical convergence zone (ITCZ).

The ITCZ moves between the tropics with the seasonal shift of the equatorial low pressure. Because this zone is an area of convection (uplift of air) the winds are weak and are called the 'Doldrums'. The other two zones of convergence are the polar fronts in both hemispheres, where the westerlies meet the polar easterlies. There are also two important zones of divergence, the subtropical high-pressure zones, which are areas of relatively light winds usually referred to as the 'horse latitudes'.

These idealised wind belts rarely exist in reality because of seasonal changes in insolation and the distribution of oceans,

continents and major relief features. Average wind patterns for July are shown below for reference.

Prominent circulation features are relatively permanent. The most noticeable features are the subtropical high-pressure cells at about 30°N and 30°S with the equatorial low-pressure zone, with converging winds in the vicinity of the ITCZ.

The larger landmasses in the northern hemisphere tend to break down the latitudinal belts of similar pressure. For example, in summer the warmer landmasses of southern Asia generate a low-pressure cell in subtropical latitudes, while in winter a high-pressure cell develops over northern Asia, in contrast to the Icelandic and Aleutian Lows.

Land and water differences
Land and water have quite different properties in the absorption and radiation of heat. In general, land surfaces heat faster and to a higher level than water surfaces. However, land surfaces also cool faster and reach lower temperatures when solar energy is absent. Oceans cool at a slower rate because the heat is dispersed through some depth, so they act as a store of energy. Temperature ranges over large landmasses are therefore greater than over large areas of water. This difference in properties leads to contrasting amounts of energy being transferred to the atmosphere from land and water.

The difference in properties between land and water can be explained by the fact that water is transparent and allows solar radiation to penetrate to a depth of several metres. The water in the oceans is mixed by rising and sinking motions and therefore the heat is distributed through a great volume of water. More heat is also lost by surface evaporation over water than over land. The land only absorbs heat at the surface and can therefore reach higher temperatures.

Two-thirds of the earth's surface is water and this receives two-thirds of the insolation. Therefore, the redistribution of energy from the equator to the poles by the oceans is significant. Ocean currents are the mechanism for this exchange of energy.

Ocean currents
Atmospheric and oceanic circulation are so closely connected and have such an influence on each other, that it is necessary to have some knowledge of ocean currents in order fully to understand global climate patterns. The ocean, like the atmosphere, acts as a great heat machine transferring energy from the tropics to higher latitudes.

If there were no landmasses, the ocean circulation would be largely controlled by the surface wind systems and would look like three closed loops in each hemisphere. The distribution of the major landmasses breaks down this pattern and only in the Pacific and Atlantic Oceans, where there is sufficient room, do we see elementary development of these loops or gyres which are controlled by the subtropical high-pressure cells.

Ocean currents are largely produced by temperature differences and energy transfer from the wind to water. As with the atmosphere, the temperature contrasts in the oceans create convection currents, in an attempt, to distribute heat more evenly over the globe. Winds over the ocean surface also drag surface water. However, the Coriolis force causes ocean currents to flow at about 45° to the right of the prevailing winds (in the northern hemisphere). Density differences can also produce flow in ocean waters. Density differences may be due to temperature differences e.g. cooled water in polar regions may sink to the ocean floor and be replaced by less dense warmer water. They may also be due to salinity differences, e.g. in the Tropics, equatorial rainfall reduces the salinity of the ocean water while in the drier sub-tropics evaporation is greater than rainfall and hence increased salinity. High salinity water tends to subside while on the surface a current tends to flow from areas of low salinity to areas of high salinity. This flow is however deflected by the Coriolis force by about 90° to the right in the northern hemisphere. Therefore, currents caused by salinity differences tend to move parallel to the salinity gradient.

An interesting feature of ocean currents which results from the circulation pattern is that, below a latitude of about 30°, the west coasts of continents have contact with cold currents (e.g., Canaries Current and California Current) whereas the east coasts are in contact with warm currents (e.g., Gulf Stream, Brazil Current). Above about latitude 45° the positions are reversed with west coasts in contact with warm currents (e.g., North Atlantic Drift) and east coasts in contact with cold currents (e.g., Kamchatka and Labrador Currents).

Ocean currents can have a significant effect on climate. For example, the United Kingdom owes its relatively mild winters and ice-free waters to the North Atlantic Drift which is an extension of the Gulf Stream. This warm current keeps the average January temperature in Valentia (Western Ireland) at about 7°C while on a similar latitude in Tomsk, central Russia, it is -21°C. The influence of ocean currents on climate is examined more fully in the next section which looks at temperature variations across the globe.

Temperature variations across the globe
Global isotherm maps for January and July, show that the mean (average) monthly air temperature decreases away from the equator, reflecting the decrease in solar insolation. The east-west trend and parallelism of the isotherms are best developed in the southern hemisphere where the landmasses are smaller than in the northern hemisphere. In the northern hemisphere, the isotherms are deflected where they pass from land to ocean.

The January isotherms are deflected poleward over the ocean and equatorward over land, in the northern hemisphere. The effects of warm ocean currents such as the North Atlantic Drift are clearly shown by the isotherms. This trend is reversed in the July map because continental interiors rapidly heat up in the summer months. In January the east and west coasts of both South America and Southern Africa illustrate the effect of contrasting warm and cold currents on coastal temperatures.

Throughout the year the isotherms shift with solar insolation changes. Over large water areas, the latitude shift is small whereas over substantial land masses the shift can be considerable. This difference between land and ocean can be clearly seen when the temperature range between January and July is examined. The annual range is much greater in continental than in coastal locations. It is over 55°C in north-east Siberia, while for much of the UK it is about 10°C.

The impact of continentality and the influence of ocean currents can be examined using the concept of temperature anomaly.

Temperature anomalies are calculated by subtracting the mean January or July temperatures for the line of latitude from that of individual stations on that line of latitude. The temperature difference whether positive or negative is the anomalous temperature. Temperature anomalies at sea level are shown on the preceding map. The largest anomalies are in the northern hemisphere where the oceans exhibit positive anomalies and the continental interiors negative anomalies. North-west Europe benefits with a positive anomaly while north-east Siberia is over 24°C colder than the average for its line of latitude.