The Venus Test

Venus's surface is at 735 K, hot enough to melt lead. Is that the 92 bar of pressure, or the greenhouse effect? Same planet, same sunlight, same pressure, same lapse rate. Two runs, and an energy ledger that has to balance.

In 60 seconds

16,576 W/m²infrared emitted by the ground, σT⁴ at 735 K
150 W/m²leave Venus: the same as the sunlight it absorbs
16,426 W/m²never reach space: the air traps them. That is the greenhouse effect
227 Kthe ground with the same 92 bar but no infrared absorbers (327 K even reflecting nothing)

Run

Sunlight reflected

With nothing reflected Venus absorbs all 650 W/m² that reach it, the most it could ever get.

Nothing in the air can radiate: air that doesn't absorb infrared doesn't emit it either, so the ground is the only thing radiating to space. Raising the emission height means adding absorbers, and that is the second run.

More absorbing gas, higher emission level. Venus's measured profile reaches the emission temperature at 71 km, near the cloud tops.

Grey model: one opacity for the whole infrared, absorber mixed with the air, so τ(z) = τ₀ p(z)/p₀. Venus radiates from where τ = 1: . Doubling τ₀ lifts that level by ; the effect grows with the logarithm and never saturates.

Hover or touch the measured curve to read altitude, temperature and pressure; click or tap to pin the reading.

Energy ledger, per square metre, planet average

Venus as a whole
Sunlight absorbed
Infrared sent to space
Sent − absorbed
The ground
Infrared emitted by the ground, σT⁴
Sunlight absorbed by the ground
Must come back from the air
Greenhouse effect
Ground's infrared that never reaches space (emitted − sent to space)
Earth, for comparison (396 − 240)156 W/m²

The published model: switch absorbers off

Bullock and Grinspoon's radiative-convective model of Venus, as reported by Taylor and Grinspoon (2009, Table 2): how much the surface cools when one species is removed from the model's greenhouse calculation, with the planet's albedo unchanged, even when the clouds are removed. Switch off one or more.

What leaves the ground, what leaves Venus

Planck's law, Mλ = 2πhc² / λ⁵ / (ehc/λkT − 1), per square metre; the area under each curve is σT⁴. Both curves follow the run and the emission height above. Grey picture: real spectra have bands and windows, but the totals are set by the temperatures.

The argument, step by step

  1. Venus is in energy balance. It absorbs about 150 W/m² of sunlight, averaged over the planet, and sends the same back to space as infrared. Measurements by the Pioneer Venus Orbiter confirmed that the planet is in overall energy balance to within the accuracy of the measurements. Taylor & Grinspoon 2009, §2.4
  2. Without infrared absorbers, the ground can't be warmer than 227 K. If the air doesn't absorb infrared, it doesn't emit any either: the ground is the only thing that radiates, and all of its radiation escapes. So σT⁴, averaged over the ground, equals the 150 W/m² absorbed, and the ground sits at 227 K whatever the pressure, the winds, the convection or the lapse rate. A ground with hot and cold patches averages lower, not higher. Even if Venus reflected nothing, the limit would be 327 K. An airless body with the same albedo and heliocentric distance as Venus would reach radiative equilibrium at a mean surface temperature of only about 230 K. §2.4
  3. The real ground is at 735 K, and its infrared doesn't get out. At 735 K the ground emits about 16,600 W/m² of infrared, 110 times what Venus absorbs and 25 times all the sunlight that reaches Venus. Only 150 W/m² leave Venus. The other ~16,400 W/m² of the ground's infrared is absorbed by the air. That is the greenhouse effect: the infrared radiation energy trapped by atmospheric gases and clouds. Raval & Ramanathan 1989 On Earth the same difference is about 156 W/m² (396 − 240); on Venus it is about 100 times larger. This step doesn't depend on how energy moves inside the atmosphere, only on how much leaves the ground and how much leaves Venus.
  4. The ground lives on what the air sends back. Enough sunlight diffuses through the cloud layers on Venus to provide about 17 W per m2 of average surface insolation. §2.4 A ground that emits 16,600 W/m² and gets 17 W/m² from the Sun receives at least ~16,560 W/m² from the air above. Without the infrared trapped in step 3, the air would only have the ~130 W/m² of sunlight it absorbs itself to pass on, a hundred times too little.
  5. The lapse rate fills in the shape. The absorbed energy must instead be raised by convection along an approximately adiabatic temperature-pressure profile to a level near the cloud tops where it can radiate to space. §2.4 On Venus that level is about 71 km, where the air is at 225 K; below it the temperature rises by about 7.1 K per km: 225.30 + 7.145 × 71.38 = 735.3 K. The absorbers set the height, the lapse rate sets the slope. This step explains the profile, it doesn't prove anything: the 71 km and the slope are read from the measured profile, so 735 K comes out by construction. The proof is in steps 1 to 4.

The First Law, in time

Add absorbers and the emission level moves up. For a while Venus radiates from colder air: less power leaves than arrives, and the difference is stored as heat, dU/dt = Pin − Pout. The column warms at the same lapse rate until the new level radiates the absorbed power again. Nothing is created: the warming is paid for by the imbalance.

Outgoing − absorbed, right after the step
Ground warming at the end
Energy stored in the column, C × ΔT
Imbalance added up over time, ∫ (Pin − Pout) dt

Rough time scale: the whole atmosphere warms together at a fixed lapse rate, with heat capacity C = (p₀ / g) × cp = (9.21 × 10⁶ Pa / 8.87 m/s²) × 1.1 kJ/(kg·K) ≈ 1.1 × 10⁹ J/(m²·K), cp of CO₂ at 600–700 K (Çengel and Cimbala, Table A-10). The end state does not depend on C; only the timing does.

If compression paid the bill

Suppose there were no greenhouse effect and the ground were kept at 735 K by compression. The ground loses more than the sunlight that reaches it, every second. Compression can supply that only while the air is actually being compressed: gravity does work as the column sinks, at a rate p₀ × v per square metre.

The air would have to sink at
That is, every year
Gravitational energy of the whole atmosphere, ∫ p dz
Its centre of mass sits at
All of it would pay the ground's bill for

A static atmosphere does no work: compression heats only while it lasts. Energy from the VIRA profile (log-linear between levels, the column above 80 km added with a 4.7 km scale height); the bill is σT⁴ at 735.3 K minus the 17 W/m² of sunlight at the ground.

The Mercury test: closer to the Sun, and two months of night

Mercury is the closer, darker planet, with no atmosphere. If distance from the Sun explained Venus, Mercury would be hotter.

MercuryVenus
Sunlight arriving9,082.7 W/m²2,601.3 W/m²
Bond albedo (reflected)0.0680.77
Sunlight absorbed, planet averageS (1 − A) / 42,116 W/m²150 W/m² 14 times less
Black-body temperature439.6 K226.6 K
Average surface temperature440 K about its black-body temperature737 K 510 K above its own
Length of the solar day176 days117 days nights of about two months
Day–night temperature rangenot listed sunward side alone: 590–725 K~0

Mercury absorbs 14 times more sunlight per square metre and sits at its black-body temperature. Venus absorbs 14 times less and is 300 K hotter. And through nights that last about two months, with no sunlight at all, Venus's surface doesn't cool. What keeps it warm in the dark is the air above it: the infrared it absorbs and sends back down.

NASA NSSDC Planetary Fact Sheets, Mercury and Venus: solar irradiance, Bond albedo, black-body temperature, average temperature, length of day, diurnal temperature range.

The explicit equations

The same chain for Earth and Venus. It reproduces both measured surfaces, and it shows what the middle-of-the-mass rule leaves unexplained.

EarthVenus
1. Energy balanceσTe4 = S (1 − A) / 4 1361 × 0.706 / 4 = 240.2 W/m²
Te = 255.12 K
2601.3 × 0.23 / 4 = 149.6 W/m²
Te = 226.63 K
2. Emission sphere, radius R + zeTem = Te (R / (R + ze))½ 255.02 K R = 6371.0 km 225.30 K R = 6051.8 km
3. Emission height: the absorbersτ(ze) = 1 → ze = H ln τ0
measured where T(z) = Tem
5.097 km 71.38 km
4. Lapse rate: thermodynamicsdT/dz = −g / cp, measured mean 6.5 K/km 7.145 K/km
5. GroundTs = Tem + Γ ze 255.02 + 6.5 × 5.097 = 288.15 Kmeasured 288.15 K 225.30 + 7.145 × 71.38 = 735.3 Kmeasured 735.3 K
The middle-of-the-mass ruleTs = Te + Γ zhalf, half of the air's mass above zhalf 255.12 + 6.5 × 5.477 = 290.72 Koff by −2.57 K 226.63 + 7.145 × 10.40 = 300.94 Koff by +434.4 K
6. Without absorbersτ0 < 1 → ze = 0 → Ts = Te 255.1 K 226.6 K the whole residual belongs to the absorbers

First Law: if ze rises, less power leaves than arrives, and dU/dt = Pin − Pout > 0 until the ground has warmed by Γ Δze. Earth: U.S. Standard Atmosphere 1976, 288.15 K at the surface and 6.5 K/km in the troposphere, pressure from the hydrostatic equation (half the mass above 5.48 km); NASA Earth Fact Sheet, solar irradiance 1361 W/m², Bond albedo 0.294. Venus: VIRA, half the mass above 10.40 km. The emission heights are read from the measured profiles; computing them from the spectra is what the published model in the panel above does.

Objections

"Taylor and Grinspoon themselves say it's the mass of the atmosphere."

They do: the high surface temperature on Venus, is largely a consequence of the large mass of the atmosphere, rather than any mysterious thermal process (§2.2). The same paper says how the mass works: Most of the energy deposited at depth cannot escape as long-wave radiation and has to be carried up to near the cloud tops before it can radiate to space (§2.4). A massive CO₂ atmosphere is a thick layer of infrared absorber, and that is what lifts the radiating level to 71 km. In the same paper, switching off CO₂'s absorption in the model cools the surface by 423 K (Table 2).

"They also say Earth with 92 bar would be hot even without more greenhouse gas."

The sentence is: If the Earth had such a high surface pressure, it too would be extremely hot, even without the increased proportion of greenhouse gases that is found on Venus (§2.2). With the same proportions, 92 times the air means 92 times the CO₂ above every square metre, and the higher pressure broadens the absorption lines: the absorber grows with the mass. What doesn't work is mass without absorbers, which is the first run: 227 K.

"It's just the adiabatic lapse rate."

The measured profile, in the paper's words, roughly follows the hydrostatic and adiabatic formulae (§2.2). The lapse rate says how much warmer the ground is than the level that radiates to space; it doesn't say where that level is. With no absorbers that level is the ground itself, and the ground sits at 227 K.

"Compression heats the gas."

Compressing a gas warms it while it is being compressed: the work done is pressure times the change in volume. Venus's atmosphere isn't being compressed. A pressure that doesn't change does no work, while the ground radiates away 16,600 joules per square metre every second. To pay that by compression the air would have to sink by about 57 km a year, and the whole gravitational energy of the atmosphere would last about three months: see "If compression paid the bill".

"Venus is closer to the Sun."

Venus reflects about 76% of the sunlight it receives and absorbs less than Earth: the paper gives equilibrium temperatures of about 230 K for Venus and 250 K for Earth (§2.4). Even if Venus reflected nothing, a ground that radiates freely to space couldn't average more than 327 K. Try it in the first run. And Mercury, closer and darker, absorbs 14 times more sunlight and is 300 K colder: see "The Mercury test".

"The heat comes from the interior, from volcanoes."

Interior heat release cannot, in any case, be the dominant influence in surface temperature, because of the constraints placed by the energy balance measurements (§2.4). What leaves Venus matches the sunlight it absorbs.

"The first run is unrealistic: no gas at 92 bar is transparent in the infrared."

Right, and that absorption is the greenhouse effect. The first run is the control: same mass, same sunlight, no infrared absorption. Whatever the real air adds on top of it, it adds by absorbing infrared.

"The air could give the heat back by convection, not by radiation."

It doesn't matter how the energy gets back to the ground. The ground sends out 16,600 W/m² as infrared and only 150 W/m² leave Venus, so the rest is absorbed by the air. Air that absorbs the ground's infrared is the greenhouse effect, whatever happens next. And on Venus convection carries energy up, not down: it is raised by convection towards the cloud tops (§2.4).

"You tuned the second run to give 735 K."

Yes: its height and slope are read from the measured profile, so it matches Venus by construction. It is there to show the shape, not to prove the greenhouse effect. The proof doesn't use it: it uses the ground's temperature, the energy that leaves Venus and the sunlight that reaches the ground, all measured.

"You need the adiabatic lapse rate to compute the emission height."

No. The emission height comes from radiation and hydrostatics alone: with the absorber mixed in the air, τ(z) = τ₀ p(z)/p₀, and Venus radiates from where τ ≈ 1, so ze = H ln τ₀, with H = RT/(Mg). The lapse rate is needed only to go from that level down to the ground. And with radiation alone, no convection, the ground would be far hotter: press "Show radiation only" under the chart. Convection, at the adiabatic lapse rate, is what brings it down to 735 K.

"At 1 bar Venus is at 15 °C, like the Earth."

Not in the measurements. In the Venus International Reference Atmosphere the pressure is 1 bar at 50.5 km, where the air is at 346 K, 73 °C. The air is at 15 °C higher up, at 56.8 km, where the pressure is 0.4 bar. Press "Show the 1-bar level" under the chart. Anchoring Earth's surface temperature at Venus's 1-bar level is a choice, not a measurement, and the anchor is the whole question.

"It's the Sun: the long Venus day bakes the surface."

The solar day on Venus lasts about 117 Earth days, so the nights last about two months. NASA lists Venus's day–night temperature range as ~0: the surface doesn't cool in the dark. Mercury, with a longer day and 14 times more absorbed sunlight but no atmosphere, averages 440 K. See "The Mercury test".

"Only 17 W/m² of sunlight reaches the ground. How can it be the greenhouse effect?"

That's the strongest evidence for it. 17 W/m² of sunlight can't keep up a ground that emits 16,600 W/m². Almost all of the energy the ground receives arrives as infrared from the air above.

Found an error?

Every number on this page comes from the equations and data listed below, and the code is public. If one is wrong, show which one, the value you get instead and how you get it. The report opens on GitHub, already filled in with a link to the view you are looking at.

See the reports

Corrections so far