Tidal Heating

A moon kneaded like dough

Io should be a dead rock. It is almost exactly the size of Earth's Moon, made of similar silicate material, and four and a half billion years old. By every expectation it should have cooled and solidified eons ago, its interior heated only by the slow trickle of radioactive decay. Instead, Io is the most volcanically active body in the Solar System: more than 400 active volcanoes, lava lakes the size of countries, sulphur plumes that rise hundreds of kilometres above the surface, and a global heat flow of order 10¹⁴ watts — roughly thirty times Earth's per square metre. Something is pouring an enormous amount of energy into Io's interior, far more than radioactivity could ever supply.

That something is tidal heating: the conversion of orbital and rotational energy into internal heat when a body is repeatedly flexed by a time-varying tidal field. Jupiter raises a tidal bulge on Io, just as the Moon raises tides on Earth. On a circular orbit that bulge would sit still and do nothing. But Io's orbit is slightly eccentric, so over each 1.77-day orbit the bulge grows, shrinks, and rocks back and forth. Io's rocky interior cannot flex without internal friction, and that friction becomes heat — deposited continuously, day after day, for billions of years. The same mechanism, gentler, keeps a global liquid-water ocean from freezing beneath the ice of Europa.

The deep elegance of the story is that the eccentricity which drives the heating should not survive — tides themselves try to erase it. It persists only because Io, Europa, and Ganymede are locked in a gravitational resonance that pumps the eccentricity back in as fast as the heating drains it. Strip away the resonance and Io goes dark within tens of millions of years.

How the flexing becomes heat

Start with the tide itself. A nearby planet of mass M_p raises a bulge on a satellite of radius R at distance a; the height of the equilibrium tide scales as M_p R⁴ / (M_s a³). For a satellite in synchronous rotation — spinning once per orbit, like our Moon — that bulge would, on a circular orbit, point permanently at the planet and never move relative to the satellite's body. A static bulge stores elastic energy but dissipates none.

Eccentricity breaks the stasis in two distinct ways, and both matter:

• Radial tide (amplitude variation). On an eccentric orbit the satellite is closer at perihelion than at aphelion. Tidal force falls as 1/a³, so the bulge is taller at perihelion and shorter at aphelion. The body breathes radially once per orbit — actually with a dominant term at twice the orbital frequency once you carry the expansion to first order in e.
• Librational tide (orientation variation). By Kepler's second law the satellite moves faster along its orbit at perihelion than at aphelion, but its spin rate is constant (synchronous). So the planet appears to rock east and west in the satellite's sky by roughly ±2e radians over an orbit — an "optical libration." The tidal bulge tries to follow the planet, sweeping back and forth across the surface.

Together these produce a periodic strain field throughout the satellite's interior, oscillating at the dominant frequency 2n (n being the mean motion). Real rock and ice are anelastic: the strain lags the stress, and over each cycle a hysteresis loop encloses an area equal to the energy dissipated as heat. Integrate that loss over the whole body and you get the total tidal power.

The scaling law: e² n⁵ R⁵ / Q

For a synchronously rotating satellite on a low-eccentricity orbit, the tidal power dissipated in the satellite is, to leading order,

Ė = (21/2) · (k₂/Q) · (G M_p² / a) · (R/a)⁵ · n · e²

where the symbols are: M_p the planet mass, a the orbital semi-major axis, R the satellite radius, n the mean motion (orbital angular frequency), e the eccentricity, k₂ the degree-2 tidal Love number (how much the body deforms), and Q the dimensionless quality factor (how lossy that deformation is — a low Q means a lot of energy lost per cycle). Substituting Kepler's third law n² = G M_p / a³ to trade a for n, this collapses to the proportionality used everywhere in the field:

Ė  ∝  (k₂ / Q) · e² · n⁵ · R⁵

Three sensitivities dominate everything that follows:

• e² — quadratic in eccentricity. No eccentricity, no heat. Halving e quarters the dissipation. This is why eccentricity pumping by resonance is not a footnote but the central plot device.
• n⁵ — fifth power of orbital frequency (equivalently a⁻¹⁵ᐟ²). Tidal heating is a savagely close-in phenomenon. Move a moon outward by a factor of two and, at fixed e, the heating falls by 2¹⁵ᐟ² ≈ 180. This is why Io (innermost) cooks while Callisto (outermost Galilean, and not in the resonance) is inert.
• k₂ / Q — the interior response. A more deformable, lossier body dissipates more. For Io k₂/Q is of order 0.01–0.04; for a rigid icy shell it can be far smaller, but a body with a subsurface ocean decouples its shell and can have a surprisingly large effective k₂.

Worked example: putting numbers into Io

Let us estimate Io's dissipation from the formula and see if it lands near the observed 10¹⁴ W. Use the following values:

M_p (Jupiter)   = 1.90 × 10²⁷ kg
a (Io orbit)    = 4.217 × 10⁸ m
R (Io radius)   = 1.822 × 10⁶ m
n (mean motion) = 4.11 × 10⁻⁵ s⁻¹   (period 1.769 d)
e (eccentricity)= 0.0041
k₂/Q            ≈ 0.015            (Io estimate)
G               = 6.674 × 10⁻¹¹ m³ kg⁻¹ s⁻²

Plug into Ė = (21/2)(k₂/Q)(G M_p²/a)(R/a)⁵ n e²:

G M_p² / a   = (6.674e-11)(1.90e27)² / (4.217e8)
             = (6.674e-11)(3.61e54) / (4.217e8)
             ≈ 5.71 × 10³⁵  J

(R/a)⁵       = (1.822e6 / 4.217e8)⁵
             = (4.321e-3)⁵
             ≈ 1.51 × 10⁻¹²

Ė ≈ 10.5 × 0.015 × 5.71e35 × 1.51e-12 × 4.11e-5 × (0.0041)²
  ≈ 10.5 × 0.015 × 5.71e35 × 1.51e-12 × 4.11e-5 × 1.68e-5
  ≈ 0.9 × 10¹⁴  W

So a back-of-envelope estimate gives roughly 10¹⁴ W — right in the observed band of (0.6–1.6)×10¹⁴ W from Voyager, Galileo, and decades of ground-based infrared monitoring. Spread over Io's surface area (4πR² ≈ 4.2×10¹³ m²) that is about 2 W/m², dwarfing Earth's ~0.09 W/m² geothermal flux. The estimate is crude — the result swings by a factor of a few with k₂/Q — but the order of magnitude is robust, and it was on the strength of exactly this calculation that Peale, Cassen, and Reynolds famously predicted Io's volcanism in a 1979 paper that appeared just weeks before Voyager 1 photographed the plumes.

Why the eccentricity doesn't die: the Laplace resonance

Here is the paradox. Tidal dissipation removes orbital energy and, for a synchronous satellite, damps the eccentricity on a timescale of order Q/(k₂) · (1/n) · (M_s/M_p)(a/R)⁵. For Io this is tens of millions of years — a hundred times shorter than the age of the Solar System. Io's orbit should have circularised long ago, the heating should have switched off, and Io should be cold. It is not. What is feeding the eccentricity?

The answer is the Laplace resonance, a triple mean-motion resonance among the three inner Galilean moons. Their orbital periods are in a near-perfect 1:2:4 ratio:

Io       P = 1.769 d
Europa   P = 3.551 d   (≈ 2 × Io)
Ganymede P = 7.155 d   (≈ 2 × Europa, 4 × Io)

More precisely, the mean motions obey n_Io − 3 n_Europa + 2 n_Ganymede ≈ 0, and the resonant argument Φ = λ_Io − 3λ_Europa + 2λ_Ganymede librates tightly about 180°. Because the moons line up in the same geometry over and over, their mutual gravitational tugs add coherently instead of averaging away. These repeated kicks force the orbits to stay slightly eccentric, pumping eccentricity back in as fast as tides drain it. The energy is sourced ultimately from Jupiter's spin and the moons' orbital migration, relayed through the resonant chain. Io's heating, in other words, is paid for by Jupiter's rotation, laundered through a resonance. The identical principle, scaled down, runs at Saturn: Enceladus is held eccentric by a 2:1 resonance with Dione, and that maintained eccentricity powers its south-polar geysers.

Where it shows up across the Solar System

Body	Host	Eccentricity e	Resonance source	Estimated heating	Surface effect
Io	Jupiter	0.0041	Laplace 1:2:4	~1×10¹⁴ W	Hundreds of volcanoes, lava lakes
Europa	Jupiter	0.0094	Laplace 1:2:4	~10¹¹–10¹² W	Subsurface ocean, chaos terrain
Ganymede	Jupiter	0.0013	Laplace 1:2:4	~10¹⁰ W (low now)	Possible deep ocean; past activity
Enceladus	Saturn	0.0047	2:1 with Dione	~5×10¹⁰ W	South-polar geysers, ocean
Triton	Neptune	~0 (now)	Capture & circularisation (past)	Low now; huge during capture
The Moon	Earth	0.0549	None (no pumping)	Negligible today	Geologically dead
Titan	Saturn	0.0288	Weak / damped	Modest, debated	Possible ocean, cryovolcanism?

The contrast between Io and the Moon is the cleanest control experiment nature provides. Both are ~1,800 km rocky bodies. The Moon's eccentricity (0.055) is actually larger than Io's, but its mean motion is far smaller (a 27-day orbit versus 1.77 days), the n⁵ factor crushes the heating, and crucially nothing pumps its eccentricity. One world is paved in fresh lava; the other has not erupted in a billion years.

Variants and regimes

• Constant-Q vs. viscoelastic (Maxwell) models. The simplest treatment assumes Q is frequency-independent. More realistic Maxwell or Andrade rheologies make Q depend on temperature and forcing frequency, which couples the heating to the thermal state and can produce feedback — heat lowers viscosity, which changes k₂/Q, which changes the heat.
• Thermal runaway and self-regulation. Because dissipation depends on temperature through the rheology, tidally heated bodies can sit in a self-regulating equilibrium or oscillate. Io may be in a non-steady episode: its inferred surface heat flow exceeds what the orbits supply in equilibrium, hinting at a ~100-Myr heating/cooling cycle.
• Shell vs. ocean vs. core deposition. Where the heat goes matters as much as how much. In Europa most dissipation occurs in the warm basal ice and at the silicate sea floor; in Io it may concentrate in a partially molten asthenosphere or even a global magma ocean.
• Obliquity and spin–orbit tides. Beyond eccentricity tides, a non-zero obliquity or non-synchronous rotation adds further dissipation terms. For most major moons these are sub-dominant, but they can matter for captured or recently perturbed bodies.
• Tides raised on the planet. The flip side — tides the moon raises on Jupiter — drive the moons' outward migration and ultimately power the resonance. The energy budget is a coupled planet–satellite system, not a moon in isolation.

Common misconceptions

• "Tidal heating comes from a circular orbit's tides." No. A synchronous moon on a perfectly circular orbit has a static bulge and zero eccentricity-tide dissipation. The time-variation — supplied by eccentricity (or obliquity, or non-synchronous spin) — is the whole point.
• "It's the gravity that heats it directly." Gravity supplies the forcing, but the heat comes from internal friction — the anelastic lag in the rock or ice as it flexes. A perfectly elastic body would store and return all the strain energy and never warm up.
• "Bigger eccentricity always wins." The Moon has a larger eccentricity than Io yet is dead, because the n⁵ factor and the absence of a pumping resonance dominate. You cannot judge heating from e alone.
• "The eccentricity is left over from formation." Tides would have circularised these orbits long ago. The eccentricity is actively maintained, in real time, by mean-motion resonances. Remove the resonance and the heating fades.
• "Io's heat is mostly radioactive." Radioactivity contributes well under 1% of Io's heat flow; tidal dissipation outpaces it by more than two orders of magnitude. For Earth the reverse is true — internal heat is mostly radiogenic plus primordial.
• "Tidal locking and tidal heating are the same thing." Locking is the end state where spin equals orbit; heating is the ongoing dissipation. A locked moon on a circular orbit has no eccentricity tide; a locked moon on an eccentric, pumped orbit is a furnace.

Observational status and missions

The prediction–confirmation arc here is one of the cleanest in planetary science. In March 1979 Peale, Cassen, and Reynolds published "Melting of Io by Tidal Dissipation," forecasting that Io's interior should be widely molten. Days later, Voyager 1 imaged the first active extraterrestrial volcanism — Linda Morabito spotted a 300-km plume on the limb. Galileo (1995–2003) then mapped Io's hot spots and confirmed the global heat flow at the 10¹⁴ W level, while detecting the induced magnetic signatures that revealed conductive (salty, liquid) oceans inside Europa, Ganymede, and Callisto.

Two flagship missions are now sharpening the picture. NASA's Europa Clipper (launched 2024, arriving 2030) will measure Europa's tidal Love number k₂ by tracking how its shape flexes over an orbit, directly constraining ice-shell thickness and ocean depth. ESA's JUICE (launched 2023, arriving 2031, settling into Ganymede orbit) will do the same for Ganymede and characterise the Laplace resonance's present state. Together they aim to resolve the long-standing tension between Io's observed and equilibrium heat output, and to test whether these oceans are habitable. Ground-based and JWST infrared monitoring continues to track Io's eruptions, which vary on timescales of days to years.

Quantitative derivation sketch

To see where the e² and n⁵ come from, model the satellite as a body whose response to the tidal potential is characterised by the complex Love number k₂. The tide-raising potential at the satellite's surface from the planet is, to degree 2, W ∝ (G M_p R² / a³) P₂(cosψ). On an eccentric orbit, expand the inverse-cube distance and the longitude of the sub-planet point to first order in e; the time-dependent part of W oscillates with amplitude proportional to e at frequency ~2n.

The body responds with a deformation whose imaginary (out-of-phase) component is governed by Im(k₂) = k₂/Q. The time-averaged dissipation is the product of the forcing and the out-of-phase response, integrated over the body. Schematically,

⟨Ė⟩ ∝ Im(k₂) · (forcing amplitude)² · (forcing frequency)
     ∝ (k₂/Q) · [ G M_p R² / a³ · e ]² · n

Collecting the a-dependence: (1/a³)² · (1/a from the energy normalisation) gives a⁻⁷ before accounting for the (R/a)⁵ geometric factor and the n from the cycle frequency. Folding in Kepler's law a = (G M_p)^{1/3} n^{-2/3} converts every factor of a into n, and the powers tally to the canonical result:

Ė = (21/2)(k₂/Q)(G M_p²/a)(R/a)⁵ n e²   ∝   (k₂/Q) e² n⁵ R⁵

The numerical prefactor 21/2 comes from the degree-2 spherical-harmonic bookkeeping and the specific combination of radial and librational tides for a synchronous rotator; different conventions (e.g. tracking the secular eccentricity damping separately) shuffle the constant but never the e², n⁵, R⁵ scaling. That scaling is the load-bearing physics: it explains why a moon's fate is decided by its orbit, and why a resonance that holds e away from zero is the difference between a frozen rock and a world with oceans and volcanoes.