Science Wiki

The Hadley cell is the large-scale, thermally direct overturning circulation of the tropical atmosphere. In each hemisphere, air rises near the equator, flows poleward at high altitude in the upper troposphere, sinks in the subtropics, and returns equatorward near the surface. It is named after George Hadley, who proposed in 1735 that the trade winds result from a meridional circulation combined with the Earth's rotation. The Hadley cells transport energy and angular momentum poleward, and they set the position of the tropical rain belt, the trade winds, and the subtropical jet. They are also a major influence on the distribution of the world's arid regions. They are one of several components of the general circulation of the atmosphere, together with the mid-latitude Ferrel cell, the polar cell, eddies, and the ocean circulation, and they do not by themselves account for global climate.

The two Hadley cells span roughly the latitudes from the equator to about 30° in each hemisphere. Together they are the dominant mean circulation of the tropics, and, in the zonal mean, they carry moist static energy from the tropics toward the subtropics. The ascending branch is associated with the intertropical convergence zone (ITCZ) and the descending branch with the subtropical high-pressure belts.

Ascent in the ITCZ

Solar heating is strongest in the tropics, so the surface and the near-surface air are warm and moist, and evaporation is large. Near the equator, the trade winds from both hemispheres converge, and the air rises in deep cumulonimbus convection. As moist air rises, it cools, water vapor condenses, and latent heat is released, which warms the troposphere and reinforces the ascent. In the zonal mean the rising air is concentrated in the convective towers of the ITCZ, which occupy a small fraction of the tropical area, and the rest of the tropics has weak, broad subsidence. The ITCZ can be seen as a band of cloud and rain, and its position is not fixed. Over the Pacific and Atlantic oceans it is usually north of the equator in the annual mean, and it moves with the season toward the summer hemisphere.

Poleward flow aloft

Rising air reaches the tropopause, which is a stable layer that limits deep convection, at about 15 to 17 km in the tropics. The air spreads poleward in the upper troposphere. As it moves poleward, its distance from the Earth's axis of rotation decreases, and the conservation of angular momentum makes the air accelerate eastward (a westerly flow). This is the origin of the subtropical jet near the poleward edge of the cell.

Subsidence in the subtropics

The upper-level air cools by emitting infrared radiation to space, and it descends in the subtropics, typically at latitudes of 20° to 35°. The descent is driven by this radiative cooling (together with the effects of eddies at the poleward edge of the cell). Because the air that sinks is dry, having lost most of its water vapor in the ascent, and because it warms by compression as it descends, relative humidity is low, and cloud formation is suppressed. The sinking air creates the subtropical high-pressure belts at the surface. Over large areas of the subtropical ocean, these are the subtropical anticyclones.

In a region of steady subsidence, the vertical velocity is set by a balance between the radiative cooling and the adiabatic warming of descending air. In terms of potential temperature θ:

For example, with a radiative cooling of 1.5 K per day and a static stability ∂_θ_/∂_z_ of 4 K per km, the descent is w ≈ −1.5 / 4 ≈ −0.4 km per day, which is about 4 mm per second. This is much slower than the updrafts in convective towers, which reach several meters per second, which illustrates why a small area of intense ascent balances a large area of slow descent.

For dry adiabatic descent, the air warms at the dry adiabatic lapse rate:

where g is the acceleration of gravity and c_p is the specific heat of air at constant pressure.

Return flow and the trade winds

Near the surface, the air flows from the subtropical highs toward the low-pressure trough of the ITCZ. The Coriolis force deflects this flow to the right in the Northern Hemisphere and to the left in the Southern Hemisphere, which makes the winds blow from the northeast and the southeast, respectively. These are the trade winds. They are easterly, and, over the oceans, they pick up heat and moisture on their way toward the equator, so that the cycle is closed. The trades are an important part of the surface forcing of the ocean, for example in the upwelling at the equator and in the zonal tilt of the thermocline.

Angular momentum conservation

A simple model, due to Isaac Held and Arthur Hou (1980), treats the cell as an axisymmetric circulation that conserves absolute angular momentum in the upper branch, in the absence of eddies and friction. The absolute angular momentum per unit mass of air at latitude φ with zonal wind u is:

where Ω is the rotation rate of the Earth (7.29 × 10⁻⁵ s⁻¹) and a is its radius (6.371 × 10⁶ m). If air starts at rest relative to the Earth at the equator, where M = Ω a², and conserves M as it moves poleward, its zonal wind is:

At φ = 30°, with Ω a ≈ 465 m per second, this gives u ≈ 465 × 0.25 / 0.866 ≈ 134 m per second. Observed zonal-mean jet speeds are lower, about 40 to 60 m per second in the annual mean and higher in winter, because eddies, friction, and mixing of angular momentum reduce the speed. The model nevertheless explains why the Hadley cell has a limited poleward extent and why a jet forms at its edge. In the small-angle approximation, the Held-Hou model gives a poleward extent that scales as:

where Δ_θ_ is the equator-to-pole difference in radiative-equilibrium potential temperature, H is the depth of the circulation, and θ₀ is a reference temperature. For Earth-like parameters, the width is of the order of 20 to 30° of latitude. A faster-rotating planet is predicted to have narrower cells.

The model is idealized. In particular, the Hadley cell is not completely axisymmetric in the real atmosphere. Large-scale eddies in the subtropics and at mid-latitudes mix angular momentum and heat, and they influence the position of the poleward edge and the strength of the cell in winter.

Seasonal cycle

The annual mean circulation consists of two roughly symmetrical cells, but the seasonal cycle is large. In the solstice seasons, the ascending branch shifts into the summer hemisphere, and the cell that crosses the equator from the summer hemisphere into the winter hemisphere becomes much stronger than the other. This winter cell then dominates the cross-equatorial transport of mass and energy, and the opposite cell becomes weak and narrow. The seasonal movement of the ITCZ is one of the main controls of the monsoon systems, for example the African and the South Asian monsoons, which are also influenced strongly by the land-sea contrast, topography, and the surface energy balance.

Strength

The strength of the circulation is measured by the mass streamfunction, which has a peak of the order of 10¹¹ kg per second in the winter cell, and of about half this in the annual mean. The Hadley cells carry a large part of the poleward transport of energy in the tropics, and the total transport of heat by the atmosphere and oceans together reaches about 5 to 6 petawatts in each hemisphere near latitudes of 35 to 40°. At low latitudes, the energy is carried mainly as moist static energy, in which the upper branch carries high energy (a large contribution of the geopotential and latent heat) poleward and the lower branch carries lower energy equatorward.

Rainfall and arid regions

The ascending branch supplies the heavy rainfall of the equatorial rain belt and of the tropical rainforests of the Amazon, the Congo Basin, and Southeast Asia. The descending branch is associated with low rainfall and clear skies in the subtropics, and it is one of the main reasons that many of the world's deserts lie at latitudes of 15 to 35°. Examples include the Sahara, the Arabian Desert, the Kalahari, the Atacama, and the deserts of Australia and the southwestern United States. The location of a particular desert also depends on other factors. They include the distance from the ocean and from moisture sources, topography (rain shadows), cold ocean currents and coastal upwelling (as in the Atacama and the Namib), and the monsoon circulations. The statement that the Hadley cell alone "forms" the deserts is therefore an oversimplification. Subsidence suppresses rainfall, but the details come from the interaction with local geography.

Winds and ocean

The trade winds are among the most persistent surface winds on Earth, and sailing routes have used them for centuries. They drive the equatorial ocean currents, contribute to upwelling along the eastern boundaries of the ocean basins, and carry Saharan dust across the Atlantic. The subtropical jet, at the poleward edge of the cell and near the tropopause, affects the tracks of storms, the formation of clear-air turbulence, and the movement of weather systems in the subtropics.

Relation to the Walker circulation and ENSO

The Hadley circulation is the meridional, zonally averaged overturning. The Walker circulation is the east-west overturning over the tropical Pacific, with rising air over the warm western Pacific and sinking air over the cooler eastern Pacific. The two interact in the El Niño-Southern Oscillation (ENSO). During El Niño, the warm water and the convection move eastward along the equator, and the Walker circulation weakens. The zonal-mean Hadley circulation tends to strengthen in the winter cell at the same time and shifts the subtropical jets, which influences the rainfall and temperature in distant regions. Strong ENSO events therefore affect weather in many regions. The effects depend on the event and the region, and ENSO is only one of several factors in any specific drought or heat wave.

Poleward expansion

Reanalyses and satellite records, among other data, indicate that the edges of the tropics, defined by the poleward limits of the Hadley cells, have moved poleward since about 1979. Estimates of the rate in observations vary by method and data set, with values of the order of a few tenths of a degree of latitude per decade, and with a large uncertainty. Climate models simulate an expansion in response to greenhouse gases, but they usually give a smaller rate than the observations. Natural variability on decadal timescales, notably in the tropical Pacific (the Pacific Decadal Oscillation), contributes to the observed rate and complicates attribution. In the Southern Hemisphere, the depletion of stratospheric ozone has also been a significant driver of the expansion during the summer, with a smaller contribution from greenhouse gases. Other factors that have been proposed include aerosols and changes in sea surface temperature patterns.

The proposed mechanisms of the expansion include an increase in the height of the tropopause and in the static stability of the subtropical troposphere as the troposphere warms, which shifts the latitude at which eddies become unstable to the poleward side, and which therefore moves the descending branch and the jet. The way these processes combine is an area of active research. The statement that a warmer troposphere "delays the cooling" of the poleward-moving air is a simplification that is not accepted as a complete explanation.

Strength

Models usually project a weakening of the overall tropical circulation in a warming climate, because the increase in water vapor changes the efficiency of the energy transport, while the strength of the Hadley cell in observations is more uncertain, and it differs among the cells, seasons, and data sets. Studies of recent trends in reanalyses show a strengthening in some cases, and the differences among data sets are large.

Regional drying

A poleward shift of the subtropical dry zones is one of several factors that are expected to cause drying in some subtropical regions. Models and some observations point to less rainfall in the Mediterranean region, southwestern North America, parts of southern Africa, southwestern South America (Chile), and southwestern and southern Australia, especially in the cool season. The size of the effect differs among models and regions, and there are also changes caused by thermodynamic effects (a warmer atmosphere holds more water vapor and has a higher evaporative demand), by changes in storm tracks, and by decadal variability. The term "permanent desertification" is not supported, because the changes in rainfall are expected to be gradual and uncertain, and they are only a part of the changes in the water balance of a region.

At mid-latitudes, between about 30° and 60°, the mean meridional circulation is in the opposite sense to that of the Hadley cell. This is the indirect Ferrel cell, which is driven by the eddies of the mid-latitude storm tracks and does not represent a thermally driven overturning. At high latitudes, the polar cell has ascent near 60° and descent over the pole. The three-cell picture is a useful teaching device. It describes the zonal-mean circulation, and it does not describe the variability from day to day, which is dominated by eddies, nor the zonal asymmetries, such as the monsoons and the Walker circulation.

The circulation is studied with radiosondes, satellite measurements of winds, temperature, cloud, and precipitation, reanalyses (such as ERA5 and MERRA-2), and models ranging from idealized axisymmetric models to comprehensive general circulation models. The edge of the Hadley cell is defined in several ways, such as the latitude where the mass streamfunction at 500 hPa changes sign, the latitude of the subtropical minimum of precipitation minus evaporation, and the latitude where the tropopause height changes, and the metrics can give different values and trends, so the numbers for the expansion should be compared only when the same definition is used. The interpretation of the measurements is complicated by the limited coverage of observations over the oceans and by differences among reanalyses in the representation of the tropical circulation.

• Atmospheric circulation
• Intertropical convergence zone
• Walker circulation
• Ferrel cell
• Polar cell
• Trade winds
• Subtropical ridge
• El Niño-Southern Oscillation
• Monsoon