10.4: Three-Cell Model

If we allow for the effects of a rotating planet, the simple single-cell model above breaks down into multiple cells in each hemisphere as shown in the figure below. It may look more complex and unrelated to the single-cell model, but there are many similarities from above. There is still excess heating in equatorial regions and excess cooling in polar regions. Instead of heat being redistributed by one massive Hadley cell from the equator to the poles, there are now three convective cells. The first of these is still the same thermally direct Hadley cell from before, but now it extends only from the equator to about 30° latitude. The poles still have a large high pressure system, while the equator has a large belt of low pressure along it. Let’s take a closer look at what happens to the rising air just above the equator.

Three-cell model of the rotating Earth and the resulting wind circulations (CC BY-SA 4.0).

At the equator, the air near the surface is warm, winds are light, and the pressure gradient is weak. This region of monotonous weather is known as the doldrums. The warm air here rises, condensing into massive cumulonimbus clouds and thunderstorms, which release large amounts of latent heat as they form. The additional heat makes the air even more likely to rise, and provides the energy that drives the rising branch of the Hadley cell. This rising air reaches the stable tropopause, which blocks it from rising further, causing the air to diverge at upper levels and move poleward. Due to the Coriolis force, this upper level poleward flow is deflected to the right in the Northern Hemisphere and to the left in the Southern Hemisphere, providing westerlies aloft (near the tropopause) in both hemispheres in the Hadley cell.

As air moves poleward from equatorial regions, it is constantly experiencing radiational cooling as it emits infrared radiation. Simultaneously, this air begins to converge and pile up as it approaches the mid-latitudes (around 30° latitude in both hemispheres). This convergence of air far above the surface increases the mass of air aloft, increasing the pressure at the surface. This increase in surface pressure results in a belt of high pressure centers called subtropical highs around 30°N and 30°S. These latitudes are commonly known as the horse latitudes.

As this converging air above the subtropical highs slowly descends, it warms adiabatically by compression. This sinking air, dries the atmosphere creating generally clear skies and little rain. Over the oceans, weak pressure gradients in the high centers produce weak winds. Some of these lighter surface winds begin to move back toward the equator, and are deflected by the Coriolis force. This causes northeasterly winds in the Northern Hemisphere and southeasterly winds in the Southern Hemisphere in tropical regions. These winds are known as the trade winds, and they have a strong influence over the daily wind patterns in Hawai’i. Near the equator, the northeasterly and southeasterly trade winds converge at the surface at what is known as the intertropical convergence zone (ITCZ). Here, convergence further reinforces the rising branch of the Hadley cell.

Back at 30° latitude, while some of the air sinking along the subtropical highs goes equatorward to complete the Hadley cell, some sinking air also moves poleward. This poleward moving surface air travels from from 30° to 60° and is again deflected by the Coriolis force. This results in the prevailing surface westerlies that impact the mid-latitudes in both hemispheres. It is for this reason that weather moves west to east across the continental US. Often, this westerly flow is interrupted by high and low pressure systems that move with the mean surface flow. We’ll learn more about this in the next two chapters. As the surface air travels poleward from 30° to 60°, it collides with cold polar air moving equatorward. These air masses do not mix easily, and are separated by a boundary known as the polar front. At the polar front, surface air converges and rises at the subpolar low, and storms and convection develop here. Some of this rising air goes all the way up to the tropopause where it moves back to 30° latitude and sinks at the subtropical high along with the descending branch of the Hadley cell. This circulation cell from 30° to 60° is known as the Ferrel cell, which is a thermally indirect circulation in which cool air rises and warm air sinks.

Behind the polar front in the Northern hemisphere, cold surface polar air moves from the poles toward 60°. As the air moves equatorward, it is again deflected by the Coriolis force. In the Arctic regions, air typically flows from the northeast while in the Antarctic, air flows from the southeast. These are known as the polar easterlies. Along the polar front where cold polar air collides with warm air from the Ferrel cell, some of the rising air moves back toward the poles, which gets deflected as a westerly wind aloft. Eventually this air reaches the poles, sinks back to the surface, and flows back toward the polar front, which gives us the Polar cell.

To summarize, looking back at the three-cell model picture: there are two major belts of high pressure and two major belts of low pressure in each hemisphere (if you include the equator in both). Areas of high pressure and sinking air exist near 30° latitude and at the poles. Regions of low pressure and rising air exist over the equator and near 60° latitude by the polar front. By knowing that winds travel counterclockwise (clockwise) around low pressure systems in the Northern Hemisphere (Southern Hemisphere), and clockwise (counterclockwise) around high pressure systems in the Northern Hemisphere (Southern Hemisphere), you can get a pretty general idea of how surface winds blow around the world on average. Trade winds blow from the subtropical highs at 30° to the equator, the westerlies blow from the subtropical highs to the polar front, and the polar easterlies blow from the poles to the polar front at the surface. Areas where these winds converge will have rising motion and low pressure at the surface, and regions where these winds diverge will have sinking motion and high pressure at the surface.

How does this three-cell model match with reality? While some minor discrepancies exist, for example in reality much of the upper-level winds in the mid-latitudes are westerly like the surface, while the Ferrel cell suggests there should be easterly winds aloft. However, this model is roughly accurate for surface winds and provides a really good first order pattern for general circulation.

Average near surface winds (CC BY-NC-SA 4.0).

How does the real world’s average surface sea level pressure field compare with the above picture? When we add in the continents, ice masses, oceans, mountains, and forest, we get an average that looks something like the below two figures. The following maps show the mean sea-level pressure field for January and July, averaged from 1981 to 2010.

Looking at the two maps below, you may notice that there are some areas where low and high pressure systems seem to persist throughout the year – these are known as semipermanent highs and semipermanent lows. These include the Bermuda-Azores High, the Pacific High, the Icelandic Low, and the Aleutian Low.

Map of semipermanent highs and lows during the month of January (CC BY-NC-SA 4.0).

Map of semipermanent highs and lows during the month of July (CC BY-NC-SA 4.0).