Zonally Symmetric and Asymmetric Circulation and Eddy Transports
CSIR NET Part C meteorology: zonal-mean circulation, Hadley cell and angular momentum, Held–Hou, stationary waves, eddy heat and momentum fluxes, EP flux.
- 20 min
On this page · 14 sections
- Prerequisites
- The Hook: Why Hadley’s Picture Was Not Enough
- Why the Atmosphere Must Circulate
- Notation: Zonal Means, Eddies and the Flux Decomposition
- The Zonally Symmetric Circulation
- The Zonally Asymmetric Circulation
- Eddy Transports
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- Hadley, Ferrel and polar cells, the thermal wind and the jets in descriptive form, from Clouds, Precipitation and General Circulation.
- Pressure belts and the planetary winds, from Temperature, Pressure and Wind Belts.
- The radiation budget and the latitudinal surplus and deficit, from Radiation Laws, Scattering, Albedo, Radiative Transfer and the Greenhouse Effect.
- Hydrostatic balance and pressure coordinates, from Composition, Thermal Structure, Hydrostatics and the Standard Atmosphere.
The Hook: Why Hadley’s Picture Was Not Enough
Why the Atmosphere Must Circulate
Annual-mean absorbed solar radiation falls off steeply towards the poles, while outgoing longwave radiation is much flatter in latitude (it depends on temperature and on cloud-top height, which vary less). The result:
- A net radiative surplus equatorward of roughly 35–40° and a deficit poleward of it.
- Without transport the tropics would warm and the poles cool until each latitude was in local radiative equilibrium. The observed equator-to-pole temperature contrast is much smaller than that radiative-equilibrium contrast.
- The atmosphere and ocean together carry energy poleward. The total transport peaks at roughly 5–6 PW (\( 1 \) PW \( = 10^{15} \) W) near 35–40° latitude in each hemisphere. The atmosphere carries most of it in middle latitudes; the ocean’s share is largest in the tropics.
The general circulation is the time-averaged, global-scale motion that does this job, together with the transports of angular momentum and water that come with it.
Notation: Zonal Means, Eddies and the Flux Decomposition
For any variable \( A(\lambda,\phi,p,t) \):
| Symbol | Meaning |
|---|---|
| \( \overline{A} \) | time mean (for example a month or season) |
| \( A' = A - \overline{A} \) | transient eddy part (departure from time mean) |
| \( [A] \) | zonal mean around a latitude circle |
| \( A^* = A - [A] \) | departure from the zonal mean (zonal asymmetry) |
The zonal-mean, time-mean northward flux of a quantity \( T \) (temperature, moisture, momentum) splits exactly into three parts:
$$ [\overline{vT}] \;=\; \underbrace{[\overline{v}]\,[\overline{T}]}_{\text{mean meridional circulation}} \;+\; \underbrace{[\overline{v}^{*}\,\overline{T}^{*}]}_{\text{stationary eddies}} \;+\; \underbrace{[\overline{v'T'}]}_{\text{transient eddies}} $$- Mean meridional circulation (MMC): the Hadley, Ferrel and polar cells.
- Stationary eddies: time-mean troughs and ridges fixed in place by mountains and land–sea heating contrasts (the zonally asymmetric time-mean flow).
- Transient eddies: travelling weather systems, mainly baroclinic cyclones and anticyclones on 2–7-day time scales.
The meridional mass streamfunction describes the zonal-mean overturning:
$$ \Psi(\phi,p) = \frac{2\pi a\cos\phi}{g}\int_0^{p}[\overline{v}]\,dp $$with \( a \) the Earth’s radius. Closed contours of \( \Psi \) are the cells. The Hadley cells have strengths of order \( 10^{11} \) kg s\(^{-1}\), with the winter-hemisphere cell, which straddles the equator, much stronger than the summer one.
The Zonally Symmetric Circulation
The three cells
| Cell | Latitude range (annual mean) | Type | Driven by |
|---|---|---|---|
| Hadley | equator to about 30° | thermally direct: warm air rises, cold air sinks | differential heating and latent heat release in the ITCZ |
| Ferrel | about 30° to 60° | thermally indirect: relatively cold air rises, warm air sinks | the eddy fluxes of heat and momentum (it is an eddy-driven residual) |
| Polar | about 60° to the pole | weak, direct | radiative cooling over the polar cap |
A thermally direct cell converts available potential energy into kinetic energy; an indirect cell does the reverse and must be maintained by something else, which is why the Ferrel cell cannot exist on its own.
Angular momentum and the subtropical jet
The absolute angular momentum per unit mass about the Earth’s axis is
$$ M = (\Omega a\cos\phi + u)\,a\cos\phi $$where \( \Omega = 7.292\times 10^{-5} \) s\(^{-1}\) and \( u \) is the zonal wind. In the free upper troposphere, away from friction, a ring of air moving poleward conserves \( M \). Starting from rest (\( u = 0 \)) at the equator:
$$ u_M(\phi) = \frac{\Omega a\sin^2\phi}{\cos\phi} $$With \( \Omega a = 464.6 \) m s\(^{-1}\) this gives about 14 m s\(^{-1}\) at 10°, 58 m s\(^{-1}\) at 20° and 134 m s\(^{-1}\) at 30°. The observed subtropical jet is much weaker than 134 m s\(^{-1}\) because eddies in the subtropics remove angular momentum from the upper branch; but the rapid growth of \( u_M \) with latitude is why the Hadley cell cannot extend to the pole and why the jet forms at its poleward edge.
The Held–Hou model
Held and Hou (1980) built an idealised, axisymmetric, angular-momentum-conserving Hadley cell. Combining \( u_M \) with thermal-wind balance and requiring that the cell conserve energy (equal-area construction against the radiative-equilibrium temperature) gives the poleward extent
$$ \boxed{\phi_H \approx \left(\frac{5}{3}\,\frac{gH\Delta_h}{\Omega^2a^2}\right)^{1/2}} $$where \( H \) is the tropopause height and \( \Delta_h \) the fractional equator-to-pole difference in radiative-equilibrium potential temperature. With \( H = 15 \) km and \( \Delta_h = 1/3 \), \( \phi_H \approx 0.62 \) rad \( \approx 35° \), close to the observed edge.
| Prediction of Held–Hou | Consequence |
|---|---|
| \( \phi_H \propto \sqrt{H\Delta_h} \) | a deeper troposphere or larger heating contrast widens the cell |
| \( \phi_H \propto 1/\Omega \) | a slower-rotating planet has wider Hadley cells (Venus, Titan) |
| Surface winds | easterly (trades) in the inner cell, westerly towards its edge, so the surface stress integrates to about zero |
| Off-equatorial heating (Lindzen and Hou, 1988) | a strong winter cell crossing the equator and a weak summer cell: the solstitial pattern |
Held–Hou ignores eddies and moisture, so it is a scaling guide, not a model of the real cell. Part C uses it for scaling questions (“if \( \Delta_h \) doubles, by what factor does the width change?”).
Maintaining the surface winds
The surface easterlies (trades) take westerly angular momentum from the Earth; the surface westerlies of middle latitudes give it back. For a steady state the atmosphere must carry angular momentum poleward across about 30° from the easterly belt to the westerly belt. The Hadley cell’s upper branch does part of this in the deep tropics; eddies do most of it in the subtropics and middle latitudes, through a poleward eddy momentum flux \( [\overline{u'v'}] > 0 \) in the Northern Hemisphere upper troposphere.
The convergence of this eddy momentum flux in middle latitudes is what maintains the surface westerlies against friction. The eddy-driven jet sits where the momentum flux converges.
The Zonally Asymmetric Circulation
The zonal mean hides a strongly asymmetric time-mean flow, especially in the Northern Hemisphere winter:
| Feature | Description |
|---|---|
| Semi-permanent centres of action | Aleutian and Icelandic lows, Siberian high (winter); subtropical highs over the oceans (summer); Asian monsoon low (summer) |
| Stationary planetary waves | troughs over eastern Asia and eastern North America, ridges over the eastern oceans and western North America; largest in zonal wavenumbers 1–3 |
| Storm tracks | the North Atlantic and North Pacific, downstream of the main stationary troughs; the Southern Hemisphere storm track is more zonally uniform |
| Monsoon circulations | the Asian summer monsoon and its upper-level Tibetan anticyclone (see the next page) |
Why stationary waves exist
- Orographic forcing. Flow over the Tibetan Plateau and the Rockies stretches and squashes vortex columns and generates Rossby waves (the Charney–Eliassen model of 1949 treated this linearly).
- Diabatic (thermal) forcing. Land–sea heating contrasts and the latent heat of tropical convection create sources and sinks of vorticity in the upper troposphere.
- Transient-eddy feedback. Storm-track eddies feed back on the time-mean flow that steers them.
Southern Hemisphere stationary waves are much weaker because it has less land and less high topography outside Antarctica.
Stationary Rossby waves
For a barotropic Rossby wave in a uniform westerly \( U \), the zonal phase speed is \( c = U - \beta/K^2 \), with \( \beta = 2\Omega\cos\phi/a \) and \( K \) the total wavenumber. A stationary wave (\( c = 0 \)) has
$$ K_s = \sqrt{\beta/U}, \qquad L_s = 2\pi\sqrt{U/\beta} $$At 45° with \( U = 15 \) m s\(^{-1}\), \( L_s \approx 6{,}000 \) km. Waves of this scale can stand still against the westerlies.
Vertical propagation: the Charney–Drazin criterion
Charney and Drazin (1961) showed that stationary planetary waves can propagate upward into the stratosphere only when the mean zonal wind is westerly but weaker than a critical value that decreases with increasing horizontal wavenumber:
$$ 0 < U < U_c $$| Consequence | Why |
|---|---|
| Summer stratosphere has almost no planetary waves | its winds are easterly |
| Only the longest waves (wavenumbers 1–2) reach high in the winter stratosphere | shorter waves have a smaller \( U_c \) and are trapped |
| Northern winter stratosphere is more disturbed than southern | stronger forcing by topography and land–sea contrast; this is where sudden stratospheric warmings are frequent |
Eddy Transports
Signs and roles
| Flux (Northern Hemisphere) | Sign | Role |
|---|---|---|
| Transient eddy heat flux \( [\overline{v'T'}] \) | positive (poleward), peaks in the lower troposphere near the storm tracks | carries heat down the meridional temperature gradient; the signature of baroclinic instability |
| Stationary eddy heat flux \( [\overline{v}^*\overline{T}^*] \) | positive, large in NH winter, small in the SH | warm southerlies ahead of stationary troughs |
| Eddy momentum flux \( [\overline{u'v'}] \) | positive (poleward) in the subtropics and middle latitudes, peaking near the tropopause | carries westerly momentum up the gradient into the jet: maintains the surface westerlies |
| Eddy moisture flux \( [\overline{v'q'}] \) | poleward in middle latitudes | supplies the precipitation excess of high latitudes |
Two points carry most Part C statement questions:
- Eddy heat fluxes are down-gradient: they act like a diffusion, reducing the temperature contrast.
- Eddy momentum fluxes are often up-gradient: they sharpen the jet. This is “negative viscosity” behaviour and cannot be represented as simple diffusion.
Baroclinic instability as the eddy source
Middle-latitude eddies grow by baroclinic instability of the vertically sheared, thermal-wind-balanced westerlies (Charney 1947; Eady 1949). Growing waves tilt westward with height, so their poleward-moving air is warmer than their equatorward-moving air: that is a positive \( [\overline{v'T'}] \). The fastest-growing waves have wavelengths of a few thousand kilometres, of the order of the Rossby radius of deformation.
As these waves reach the upper troposphere they propagate meridionally away from the baroclinic zone, mostly equatorward. Rossby waves carry momentum in the opposite direction to their group propagation, so equatorward wave propagation means poleward momentum flux, converging into the eddy-driven jet.
The Ferrel cell as an eddy-driven residual
Zonal-mean momentum and heat budgets show that in middle latitudes:
- Eddy momentum flux convergence would accelerate the upper westerlies; the Coriolis force on an equatorward \( [\overline{v}] \) aloft balances it.
- Eddy heat flux divergence and convergence would destroy thermal-wind balance; rising motion where eddies deposit heat at high latitudes and sinking where they remove it restore balance.
Both requirements are met by an indirect mean meridional circulation: the Ferrel cell. It is the atmosphere’s response to the eddy forcing.
The transformed Eulerian mean and the Eliassen–Palm flux
Because the Eulerian-mean Ferrel cell largely cancels the eddy heat flux, a “residual” circulation is more physical. In pressure coordinates the transformed Eulerian mean (TEM) residual velocity is
$$ \bar v^{\dagger} = [\overline{v}] - \frac{\partial}{\partial p}\left(\frac{[\overline{v'\theta'}]}{\partial[\overline{\theta}]/\partial p}\right) $$and the eddy forcing of the zonal-mean flow appears as the divergence of the Eliassen–Palm (EP) flux, whose components are proportional to
$$ \mathbf{F} \propto \left( -[\overline{u'v'}],\; f\,\frac{[\overline{v'\theta'}]}{\partial[\overline{\theta}]/\partial p} \right) $$(times \( a\cos\phi \)). The meridional component is the eddy momentum flux; the vertical component is the eddy heat flux.
| TEM fact | What it means |
|---|---|
| Upward EP flux | poleward eddy heat flux (waves propagating up from the baroclinic lower troposphere) |
| Convergence of EP flux | westward (easterly) force on the mean flow, decelerating the westerlies |
| Divergence of EP flux | eastward force on the mean flow |
| Residual circulation | a single thermally direct cell from the tropics to the pole in each hemisphere in the troposphere: the Ferrel cell disappears |
| Non-acceleration theorem | steady, conservative, non-dissipating waves have zero EP flux divergence and do not change the mean flow |
In the stratosphere the residual circulation is the Brewer–Dobson circulation: rising in the tropics and sinking at high latitudes, driven by breaking planetary and gravity waves. It explains why tropospheric air enters the stratosphere mainly in the tropics and why ozone, made mostly in the tropical stratosphere, builds up at high latitudes.
The Lorenz energy cycle
Lorenz (1955) split the atmosphere’s energy into four reservoirs:
| Reservoir | Meaning |
|---|---|
| \( A_Z \) | zonal-mean available potential energy (from the equator-to-pole temperature contrast) |
| \( A_E \) | eddy available potential energy |
| \( K_E \) | eddy kinetic energy |
| \( K_Z \) | zonal-mean kinetic energy |
graph LR
G[Differential heating generates A_Z] --> AZ[A_Z zonal APE]
AZ -->|baroclinic: eddy heat flux| AE[A_E eddy APE]
AE -->|warm air rises, cold sinks in eddies| KE[K_E eddy KE]
KE -->|barotropic: eddy momentum flux into jet| KZ[K_Z zonal KE]
KE --> D[Friction dissipates mostly K_E]
KZ --> D2[Friction]The dominant global path is \( A_Z \to A_E \to K_E \to K_Z \), with dissipation mostly of \( K_E \). Mean-meridional conversions between \( A_Z \) and \( K_Z \) are in opposite directions in the Hadley cell (direct, \( A_Z \to K_Z \)) and the Ferrel cell (indirect, \( K_Z \to A_Z \)), so their net is small. Only a small fraction of the total potential energy is available for conversion to kinetic energy.
Memory Tricks
Don’t Confuse
| These | Difference |
|---|---|
| Stationary vs transient eddies | departures of the time mean from the zonal mean (\( \overline{A}^* \)) vs departures from the time mean (\( A' \)) |
| Thermally direct vs indirect cell | warm rises and cold sinks (APE to KE) vs the reverse (KE to APE) |
| Subtropical vs eddy-driven jet | at the Hadley cell edge, angular-momentum driven, upper-level vs in middle latitudes, maintained by eddy momentum convergence, extends to the surface |
| Eulerian mean vs residual (TEM) circulation | three cells including the indirect Ferrel cell vs one direct cell per hemisphere |
| Baroclinic vs barotropic conversion | \( A_Z \to A_E \) via heat flux vs \( K_E \to K_Z \) via momentum flux |
| Stationary Rossby wave vs stationary planetary wave | the simple barotropic \( L_s = 2\pi\sqrt{U/\beta} \) vs the observed forced waves of wavenumbers 1–3 |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper. Take \( \Omega = 7.292\times 10^{-5} \) s\(^{-1}\), \( a = 6.371\times 10^{6} \) m, \( g = 9.81 \) m s\(^{-2}\), \( c_p = 1004 \) J kg\(^{-1}\) K\(^{-1}\) unless stated.
Level 1: Foundation
- Write the three-term decomposition of the zonal-mean northward heat flux.
- Which of the three cells is thermally indirect?
- What does a convergence of the EP flux do to the zonal-mean westerlies?
Answers
- \( [\overline{vT}] = [\overline{v}][\overline{T}] + [\overline{v}^*\overline{T}^*] + [\overline{v'T'}] \): mean meridional circulation, stationary eddies, transient eddies.
- The Ferrel cell.
- Decelerates them (an easterly force).
Level 2: Part B Style
Practice Q1. A ring of air starts at rest relative to the Earth at the equator and moves poleward in the upper troposphere conserving absolute angular momentum. Its zonal wind on reaching 30° N is closest to:
(a) 58 m s\(^{-1}\) (b) 93 m s\(^{-1}\) (c) 134 m s\(^{-1}\) (d) 232 m s\(^{-1}\)
Answer: (c) 134 m/s
\( u_M = \Omega a\sin^2\phi/\cos\phi = 464.6\times 0.25/0.866 = 134 \) m s\(^{-1}\). Option (a) is the value at 20°; (d) is \( \Omega a/2 \), a wrong shortcut.
Practice Q2. At 45° N in a uniform westerly of 15 m s\(^{-1}\), the wavelength of a stationary barotropic Rossby wave is closest to:
(a) 1,500 km (b) 3,000 km (c) 6,000 km (d) 12,000 km
Answer: (c) about 6,000 km
\( \beta = 2\Omega\cos 45°/a = 1.62\times 10^{-11} \) m\(^{-1}\) s\(^{-1}\). \( L_s = 2\pi\sqrt{U/\beta} = 2\pi\sqrt{15/1.62\times 10^{-11}} \approx 6.05\times 10^{6} \) m. The latitude circle at 45° is about 28,300 km, so this is a wave of zonal wavenumber about 4.7.
Practice Q3. Which conversion in the Lorenz energy cycle is carried out mainly by the poleward eddy heat flux?
(a) \( K_E \to K_Z \) (b) \( A_Z \to A_E \) (c) \( A_E \to K_E \) (d) \( K_Z \to A_Z \)
Answer: (b)
The baroclinic conversion \( A_Z \to A_E \) is proportional to the eddy heat flux down the mean temperature gradient. (a) is the barotropic conversion by eddy momentum flux; (c) is by rising warm and sinking cold air within eddies; (d) is the Ferrel cell’s indirect conversion.
Practice Q4. In the Held–Hou model, if the fractional radiative-equilibrium temperature contrast \( \Delta_h \) is doubled while everything else is unchanged, the predicted Hadley cell width:
(a) is unchanged (b) increases by a factor of about 1.41 (c) doubles (d) halves
Answer: (b)
\( \phi_H \propto \sqrt{\Delta_h} \), so it grows by \( \sqrt 2 \approx 1.41 \). From 35° this would be about 50°, outside the small-angle range where the formula is accurate, but the scaling is what is asked.
Level 3: Part C Style
Practice Q5. At 45° N, suppose the transient-eddy temperature flux \( [\overline{v'T'}] \) is 10 K m s\(^{-1}\), uniform from the surface (1000 hPa) to the top of the atmosphere. The poleward sensible heat transport by transient eddies across this latitude circle is closest to:
(a) 0.3 PW (b) 1.0 PW (c) 2.9 PW (d) 9.0 PW
Answer: (c) about 2.9 PW
Transport \( = 2\pi a\cos\phi \int_0^{p_s} c_p[\overline{v'T'}]\,\frac{dp}{g} \).
- Circumference: \( 2\pi\times 6.371\times 10^{6}\times 0.7071 = 2.83\times 10^{7} \) m.
- Column: \( 1004\times 10\times 10^{5}/9.81 = 1.02\times 10^{8} \) W m\(^{-1}\).
- Product: \( 2.9\times 10^{15} \) W \( = 2.9 \) PW.
The real eddy heat flux is concentrated in the lower troposphere and weaker aloft, so the actual transient-eddy sensible heat transport is smaller; latent heat and the stationary eddies add to it. The exercise shows why the eddy terms dominate middle-latitude energy transport.
Practice Q6. Suppose a reanalysis gives the following illustrative contributions to the vertically integrated northward transport of dry static energy \( c_pT + gz \) (arbitrary but consistent units) in Northern Hemisphere winter:
| Latitude | Mean meridional circulation | Stationary eddies | Transient eddies |
|---|---|---|---|
| 10° N | +6.0 | +0.2 | +0.3 |
| 30° N | +0.4 | +1.5 | +2.0 |
| 50° N | −0.6 | +2.4 | +3.1 |
Which statements are correct?
- At 10° N the transport is dominated by the Hadley circulation.
- At 50° N the mean meridional circulation carries dry static energy equatorward, consistent with a thermally indirect cell.
- At 30° N and 50° N the eddies carry more than 80% of the total.
- The stationary-eddy term would be expected to be much smaller in the Southern Hemisphere at the same latitudes.
(a) 1 and 2 only (b) 1, 2 and 4 only (c) 1, 2, 3 and 4 (d) 2 and 3 only
Answer: (c) 1, 2, 3 and 4
- 1 true: 6.0 of 6.5 units at 10° N comes from the MMC.
- 2 true: dry static energy increases with height in a stable atmosphere. In the Ferrel cell the equatorward branch is aloft, where \( c_pT + gz \) is larger, and the poleward branch is near the surface, so the cell carries dry static energy equatorward, the signature of an indirect cell. (The Hadley cell’s poleward branch is aloft, giving the large positive value at 10° N.)
- 3 true: at 30° N eddies give \( 3.5/3.9 = 90\% \); at 50° N eddies give \( 5.5 \) out of a total of \( 4.9 \), more than 100% because the MMC opposes them.
- 4 true: the Southern Hemisphere has much weaker stationary waves because it has little land and topography outside Antarctica.
Practice Q7. Which statements about the transformed Eulerian mean (TEM) view are correct?
- The residual circulation in the troposphere is a single thermally direct cell in each hemisphere.
- An upward-pointing EP flux indicates a poleward eddy heat flux.
- Steady, conservative, linear waves with no dissipation can still decelerate the mean flow through a non-zero EP flux divergence.
- The stratospheric residual circulation (Brewer–Dobson) rises in the tropics and sinks at high latitudes.
(a) 1, 2 and 4 only (b) 1 and 3 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 4 only
- 1 true: the indirect Ferrel cell cancels against the eddy heat flux term and disappears from the residual circulation.
- 2 true: the vertical component is \( f[\overline{v'\theta'}]/(\partial\theta/\partial p) \). In a stable atmosphere \( \partial\theta/\partial p < 0 \), so a poleward heat flux (with \( f > 0 \)) gives a negative \( p \)-component, which points towards lower pressure: upward.
- 3 false: this is exactly what the non-acceleration theorem rules out. Wave transience, dissipation or breaking is needed.
- 4 true: that is the Brewer–Dobson circulation, driven by breaking planetary and gravity waves.
Quick Revision Table
| Idea | Fact or formula |
|---|---|
| Flux decomposition | \( [\overline{vT}] = [\overline{v}][\overline{T}] + [\overline{v}^*\overline{T}^*] + [\overline{v'T'}] \) |
| Mass streamfunction | \( \Psi = (2\pi a\cos\phi/g)\int_0^p[\overline v]\,dp \) |
| Angular momentum | \( M = (\Omega a\cos\phi + u)a\cos\phi \); \( u_M = \Omega a\sin^2\phi/\cos\phi \) (134 m s\(^{-1}\) at 30°) |
| Held–Hou width | \( \phi_H \approx (5gH\Delta_h/3\Omega^2a^2)^{1/2} \approx 35° \) |
| Cells | Hadley direct, Ferrel indirect and eddy-driven, polar weak direct |
| Total energy transport | peaks about 5–6 PW near 35–40° |
| Stationary Rossby wave | \( L_s = 2\pi\sqrt{U/\beta} \) |
| Charney–Drazin | upward propagation only for \( 0 < U < U_c \) |
| Eddy fluxes (NH) | \( [\overline{v'T'}] > 0 \), down-gradient; \( [\overline{u'v'}] > 0 \), into the jet |
| EP flux | \( (-[u'v'],\ f[v'\theta']/\theta_p) \); convergence decelerates westerlies |
| Lorenz cycle | \( A_Z \to A_E \to K_E \to K_Z \) |
Key Takeaways
What’s Next
- Tropical East–West Circulations: the Walker circulation, the Gill response to tropical heating and the monsoon’s divergent outflow.
- Grid-Point and Spectral GCMs: how models reproduce this circulation from the primitive equations.
- Back to the chapter overview.