Atmospheric Energetics
CSIR NET Part C meteorology: kinetic energy equation, total and available potential energy, the Lorenz energy cycle and the atmospheric heat engine.
- 17 min
On this page · 14 sections
Prerequisites
- Isobaric equations and \( \omega \), from Forces, Equations of Motion, Scale Analysis and Dimensionless Numbers.
- Barotropic and baroclinic instability, from Gravity, Inertia and Rossby Waves; Barotropic and Baroclinic Instability.
- Friction as a sink, from Boundary Layer, Turbulence and the Ekman Layer.
- The global radiation budget and poleward heat transport, from Radiation Balance and Insolation.
- The first law and heat engines, from Laws of Thermodynamics.
The Hook: A Very Inefficient Engine
The Kinetic Energy Equation
Take the dot product of the horizontal momentum equation in isobaric coordinates with \( \mathbf V \). The Coriolis term drops out (it is perpendicular to \( \mathbf V \)):
$$\boxed{\frac{D}{Dt}\left(\frac{\mathbf V\cdot\mathbf V}{2}\right) = -\mathbf V\cdot\nabla_p\Phi + \mathbf V\cdot\mathbf F}$$Horizontal kinetic energy changes only through:
- Work by the pressure gradient force, \( -\mathbf V\cdot\nabla_p\Phi \): positive when air flows across the contours towards lower geopotential (lower pressure). Purely geostrophic flow, parallel to the contours, does no work. KE generation therefore depends on the ageostrophic wind.
- Friction, \( \mathbf V\cdot\mathbf F \), always a sink.
From pressure work to \( \omega\alpha \)
Integrated over the whole atmosphere (mass element \( dm = dx\,dy\,dp/g \)), and using continuity and the hydrostatic relation \( \partial\Phi/\partial p = -\alpha \):
$$-\int\mathbf V\cdot\nabla_p\Phi\,dm = -\int\omega\alpha\,dm$$Globally, kinetic energy is produced where warm air (large \( \alpha \)) rises (\( \omega < 0 \)) and cold air sinks. This is the same term with opposite sign in the thermodynamic energy equation: it represents conversion between potential (plus internal) energy and kinetic energy.
Potential and Internal Energy
Their fixed ratio
For a hydrostatic column of unit cross-section extending to the top of the atmosphere:
$$P = \int_0^\infty\rho gz\,dz,\qquad I = \int_0^\infty\rho c_vT\,dz$$Using \( \rho g\,dz = -dp \) and integrating by parts (with \( pz\to 0 \) at both ends):
$$P = \int_0^{p_s}z\,dp = \int_0^\infty p\,dz = \int_0^\infty\rho RT\,dz = \frac{R}{c_v}\,I$$$$\boxed{\frac{P}{I} = \frac{R}{c_v}\approx 0.40,\qquad P + I = \frac{c_p}{c_v}I = \int_0^\infty\rho c_pT\,dz = \frac{c_p}{g}\int_0^{p_s}T\,dp}$$In a hydrostatic atmosphere, potential and internal energy are proportional and always change together; it makes sense to treat their sum, the total potential energy (TPE), as one reservoir. Heating a column raises its temperature (\( I \)) and expands it upward (\( P \)) in the fixed ratio 5:2 (for a diatomic ideal gas, \( c_v : R = 5 : 2 \)).
Available Potential Energy
Definition
Not all TPE can be converted to KE. Imagine rearranging the atmosphere adiabatically (each parcel keeps its \( \theta \)) to the state of minimum TPE: isentropic surfaces flattened to horizontal, stably stratified, with the same mass between each pair of isentropes. That reference state has no horizontal temperature gradients and cannot produce any more KE.
$$\boxed{\text{APE} = \text{TPE} - \text{TPE}_{\text{reference}}}$$The idea goes back to Margules; Lorenz gave it its modern form.
Properties:
- APE is zero when isentropes are horizontal and the stratification is stable, whatever the temperature.
- APE is positive otherwise, and depends on the variance of temperature (or pressure) on isentropic or isobaric surfaces, not on the mean temperature.
- APE is a small fraction of TPE (often quoted as of order 0.5%, i.e. a fraction of a percent), and the kinetic energy is smaller still.
Approximate form
For small departures \( T' \) from the horizontal mean \( \bar T(p) \) in a stably stratified atmosphere (lapse rate \( \Gamma < \Gamma_d \)), the APE per unit mass is approximately
$$\boxed{a\approx\frac{1}{2}\,\frac{g}{\bar T}\,\frac{\overline{T'^2}}{\Gamma_d - \Gamma}=\frac{1}{2}\frac{g^2}{N^2}\overline{\left(\frac{\theta'}{\bar\theta}\right)^2}}$$The second form follows from \( N^2 = (g/\bar T)(\Gamma_d - \Gamma) \): it is the buoyancy analogue of the elastic energy of a spring, \( \tfrac12 b'^2/N^2 \) with \( b' = g\theta'/\bar\theta \). Strong static stability (large \( N \)) makes a given temperature anomaly hold less APE, because a smaller vertical displacement suffices to flatten the isentropes.
The Lorenz Energy Cycle
Split each field into a zonal mean (denoted Z) and eddies (departures from it, denoted E). This gives four reservoirs:
| Reservoir | Meaning |
|---|---|
| \( A_Z \) | zonal APE: from the mean equator-to-pole temperature gradient |
| \( A_E \) | eddy APE: from temperature variance along latitude circles |
| \( K_Z \) | zonal KE: mean zonal winds (jets) and mean meridional circulations |
| \( K_E \) | eddy KE: cyclones, anticyclones, waves |
and the processes linking them:
| Process | Symbol | Physical mechanism |
|---|---|---|
| Generation of \( A_Z \) | \( G_Z \) | heating at low latitudes (warm) and cooling at high latitudes (cold) |
| \( A_Z\to A_E \) | \( C_A \) | eddy heat transport down the mean temperature gradient (poleward \( \overline{v'T'} \)) |
| \( A_E\to K_E \) | \( C_E \) | baroclinic conversion: warm eddy air rises, cold sinks (\( -\overline{\omega'\alpha'} > 0 \)) |
| \( K_E\to K_Z \) | \( C_K \) | eddy momentum flux convergence into the jets (eddies tilted with the shear) |
| \( A_Z\leftrightarrow K_Z \) | \( C_Z \) | mean meridional circulations: thermally direct Hadley cell converts \( A_Z\to K_Z \); thermally indirect Ferrel cell converts \( K_Z\to A_Z \) |
| Dissipation | \( D_Z, D_E \) | friction, mostly in the boundary layer |
graph LR
G[Differential heating G_Z] --> AZ[A_Z zonal APE]
AZ -->|C_A: poleward eddy heat flux| AE[A_E eddy APE]
AE -->|C_E: warm rises, cold sinks| KE[K_E eddy KE]
KE -->|C_K: eddy momentum flux into jets| KZ[K_Z zonal KE]
KZ -.->|C_Z: Ferrel indirect, Hadley direct| AZ
KE --> DE[Friction D_E]
KZ --> DZ[Friction D_Z]Reading the cycle
- Radiation creates zonal APE. Heating where it is warm and cooling where it is cold generates \( A_Z \).
- Baroclinic eddies take over. The mean state is baroclinically unstable. Eddies transport heat poleward, converting \( A_Z \) to \( A_E \), then release \( A_E \) as \( K_E \) by warm ascent and cold descent. This is the energetics of baroclinic instability.
- Eddies feed the jets. On average mid-latitude eddies give kinetic energy to the zonal flow (\( K_E\to K_Z \)), maintaining the eddy-driven jet against friction. This is the reverse of barotropic instability.
- The Ferrel cell runs backwards. It is thermally indirect (air rises at its colder, poleward edge and sinks at its warmer edge), converting \( K_Z \) back to \( A_Z \). It is driven by the eddies, not by heating. The Hadley cell is thermally direct. Their contributions to \( C_Z \) partly cancel in the global mean.
- Friction closes the cycle. Dissipation, mostly in the boundary layer, balances the conversions.
The atmosphere as a heat engine
- Heat is absorbed at higher temperature (tropical surface and latent heat release in deep convection) and rejected at lower temperature (infrared emission to space from the cold upper troposphere and high latitudes), so the atmosphere does mechanical work.
- Its efficiency (KE generation divided by heat input) is about 1%, far below the Carnot limit for the temperatures involved, because most heat is transported by large-scale motions and moisture without passing through a work-producing cycle, and because of irreversible processes such as moist mixing and frictional dissipation of falling precipitation.
- Tropical cyclones are a clear example of a heat engine: heat taken from the warm sea surface, given up at the cold outflow level near the tropopause. This Carnot-cycle picture underlies the potential-intensity theory associated with Emanuel.
Energy Budget of the Whole Column
Combining the thermodynamic, kinetic-energy and moisture equations gives the total energy per unit mass
$$E = c_pT + \Phi + Lq + \frac{V^2}{2}$$(the moist static energy plus KE). Its vertical integral is changed only by radiation and surface fluxes and redistributed by transports. The poleward transport of \( c_pT + \Phi + Lq \) by the circulation (Hadley cell, eddies) and by the ocean balances the radiative surplus in the tropics and deficit at high latitudes. The KE term is negligible in this budget; the dynamics is a small “leak” through which the circulation that does the transporting is maintained. See Vertical Stability and Tropical Convection for moist static energy itself.
Memory Tricks
Don’t Confuse
| Pair | Distinction |
|---|---|
| TPE vs APE | \( P + I \) of the actual state vs its excess over the minimum reachable adiabatically |
| \( P \) vs \( I \) | gravitational vs internal; in a hydrostatic column \( P = (R/c_v)I \) |
| Baroclinic vs barotropic conversion | \( A_E\to K_E \) via \( \omega'\alpha' \) vs exchange \( K_E\leftrightarrow K_Z \) via \( u'v' \) and shear |
| Direct vs indirect circulation | warm rising, cold sinking (Hadley, monsoon) vs the reverse (Ferrel), driven by eddies |
| Geostrophic vs ageostrophic wind in KE | does no pressure work vs generates KE by cross-contour flow |
| Generation vs conversion | diabatic creation of APE by heating vs adiabatic transfer between reservoirs |
| Efficiency vs Carnot efficiency | about 1% vs much higher theoretical limit |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper.
Level 1: Foundation
- Write the ratio of potential to internal energy in a hydrostatic column.
- Define available potential energy.
- State the main path of the Lorenz energy cycle.
Answers
- \( P/I = R/c_v\approx 0.40 \).
- The difference between the total potential energy of the atmosphere and the minimum total potential energy obtainable by an adiabatic redistribution of mass (isentropes made horizontal).
- \( A_Z\to A_E\to K_E\to K_Z \), with generation of \( A_Z \) by differential heating and dissipation of KE by friction.
Level 2: Part B Style
Practice Q1. With \( R = 287 \) and \( c_p = 1004 \) J kg\(^{-1}\) K\(^{-1}\), the ratio of total potential energy to internal energy of a hydrostatic atmosphere is:
(a) 0.29 (b) 0.40 (c) 1.40 (d) 3.50
Answer: (c) 1.40
\( (P + I)/I = c_p/c_v = 1004/717 = 1.40 \). Option (b) is \( P/I \); (a) is \( R/c_p \).
Practice Q2. Kinetic energy of the horizontal wind is generated in the free atmosphere when:
(a) the geostrophic wind blows parallel to the contours (b) air flows across the contours towards lower geopotential (c) the Coriolis force accelerates the flow (d) air flows across the contours towards higher geopotential
Answer: (b)
\( -\mathbf V\cdot\nabla\Phi > 0 \) when \( \mathbf V \) has a component down the geopotential gradient. Geostrophic flow does no work; Coriolis does no work; flow towards higher geopotential destroys KE (as in the exit region of a jet streak).
Practice Q3. In a region of the troposphere, the r.m.s. temperature departure on isobaric surfaces is 12 K, mean temperature 260 K and lapse rate 6.5 K km\(^{-1}\) (\( \Gamma_d = 9.8 \) K km\(^{-1}\)). The APE per unit mass is closest to:
(a) 80 J kg\(^{-1}\) (b) 820 J kg\(^{-1}\) (c) 8,200 J kg\(^{-1}\) (d) 82,000 J kg\(^{-1}\)
Answer: (b)
\( a = \tfrac12\times 9.81\times 144/(260\times 0.0033) = 706.3/0.858 = 823 \) J kg\(^{-1}\), about 0.3% of \( c_p\bar T \).
Practice Q4. Which mean meridional circulation converts zonal kinetic energy into zonal available potential energy?
(a) Hadley cell (b) Ferrel cell (c) Polar cell (d) Walker circulation
Answer: (b) Ferrel cell
The Ferrel cell is thermally indirect: cold air rises at its poleward edge and warm air sinks at its equatorward edge, so \( -\overline{\omega\alpha} < 0 \) and energy flows from \( K_Z \) to \( A_Z \).
Level 3: Part C Style
Practice Q5. Eddy statistics at 45° N in winter (mid-troposphere) show: \( \overline{v'T'} = +15 \) K m s\(^{-1}\); \( \overline{\omega'T'} = -0.05 \) K Pa s\(^{-1}\); \( \overline{u'v'} = +20 \) m\(^2\) s\(^{-2}\), with the jet core to the south (\( \partial\bar u/\partial y < 0 \) at this latitude). The mean temperature decreases poleward. Which statements are correct?
- Eddies convert zonal APE to eddy APE, since they carry heat down the mean temperature gradient.
- Eddies convert eddy APE to eddy KE, since warm air rises (negative \( \omega' \) correlates with positive \( T' \)).
- The term \( -\overline{u'v'}\,\partial\bar u/\partial y \) is positive, so eddies are taking KE from the zonal flow here.
- The eddy momentum flux is northward, away from the jet core, consistent with eddies tilted against the local shear.
(a) 1 and 2 only (b) 1, 2 and 3 only (c) 1, 2, 3 and 4 (d) 2 and 4 only
Answer: (c) all four
- 1 true: poleward heat flux down a poleward-decreasing temperature gradient is the \( C_A \) conversion.
- 2 true: \( \overline{\omega'T'} < 0 \) means warm anomalies have negative \( \omega \) (rising), so \( -\overline{\omega'\alpha'} > 0 \).
- 3 true: \( -\overline{u'v'}\,\partial\bar u/\partial y = -(+20)\times(\text{negative}) > 0 \): eddy KE grows at the expense of \( K_Z \) at this location (barotropic conversion \( K_Z\to K_E \)).
- 4 true: positive \( \overline{u'v'} \) is a northward flux of westerly momentum, away from the jet to the south; momentum is carried down the gradient of \( \bar u \), which is what drains the jet locally. This is the local, flank picture; integrated over the jet as a whole, eddies usually converge momentum into it.
Practice Q6. Which statements about atmospheric energetics are correct?
- Uniformly warming the whole atmosphere by 2 K increases total potential energy but leaves available potential energy unchanged (to first order).
- A more stable stratification makes a given r.m.s. temperature anomaly contain less APE.
- The total rate of KE dissipation is a few tens of W m\(^{-2}\), comparable with the latent heat flux.
- In a hydrostatic atmosphere, potential energy and internal energy cannot vary independently.
(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: APE depends on horizontal contrasts, not the mean.
- 2 true: \( a\propto\overline{T'^2}/(\Gamma_d - \Gamma)\propto 1/N^2 \).
- 3 false: dissipation is about 2 W m\(^{-2}\), roughly 1% of absorbed solar radiation, far below the surface latent heat flux (tens of W m\(^{-2}\)).
- 4 true: \( P = (R/c_v)I \).
Practice Q7. The globally averaged KE of the atmosphere is about \( 1.1\times 10^6 \) J m\(^{-2}\) and the dissipation rate is 2 W m\(^{-2}\). Which statements are correct?
- If generation stopped, the winds would decay on a time scale of about a week.
- With absorbed solar radiation of 240 W m\(^{-2}\), the atmosphere’s efficiency as a heat engine is about 0.8%.
- The kinetic energy corresponds to a mean wind speed of roughly 15 m s\(^{-1}\) (taking the column mass as \( 1.02\times 10^4 \) kg m\(^{-2}\)).
- The Coriolis force converts some of the dissipated energy back into kinetic energy.
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2 and 3 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
- 1 true: \( 1.1\times 10^6/2 = 5.5\times 10^5 \) s, about 6.4 days.
- 2 true: \( 2/240 = 0.83\% \).
- 3 true: KE per unit mass \( = 1.1\times 10^6/1.02\times 10^4 = 108 \) J kg\(^{-1}\); \( \sqrt{2\times 108} = 14.7 \) m s\(^{-1}\).
- 4 false: Coriolis does no work.
Quick Revision Table
| Idea | Fact or formula |
|---|---|
| KE equation | \( D(V^2/2)/Dt = -\mathbf V\cdot\nabla_p\Phi + \mathbf V\cdot\mathbf F \); Coriolis absent |
| Global conversion | \( -\int\omega\alpha\,dm \): warm rising, cold sinking makes KE |
| \( P \) and \( I \) | \( P/I = R/c_v\approx 0.4 \); TPE \( = (c_p/g)\int T\,dp \) |
| APE | TPE minus minimum after adiabatic levelling of isentropes; a fraction of a percent of TPE (often quoted about 0.5%) |
| APE approximation | \( a\approx\tfrac12(g/\bar T)\overline{T'^2}/(\Gamma_d - \Gamma) = \tfrac12(g/N)^2\overline{(\theta'/\bar\theta)^2} \) |
| Lorenz cycle | \( G_Z\to A_Z\to A_E\to K_E\to K_Z \); friction \( D_E \), \( D_Z \) |
| \( C_A \), \( C_E \), \( C_K \) | \( \overline{v'T'} \), \( \overline{\omega'\alpha'} \), \( \overline{u'v'} \) |
| Mean cells | Hadley direct (\( A_Z\to K_Z \)), Ferrel indirect (\( K_Z\to A_Z \)) |
| Rates | about 2 W m\(^{-2}\); efficiency about 1%; KE replenished in about a week |
Key Takeaways
What’s Next
- Dynamic Meteorology: Quick Revision: the whole chapter on one page.
- Clouds, Precipitation and General Circulation: the Hadley, Ferrel and polar cells in descriptive form.
- Climate Change: how changes in the energy budget alter the circulation.