Grand Tack Hypothesis

Tacking a planet like a sailboat

A sailboat cannot sail straight into the wind. To make ground upwind it sails at an angle, then sharply reverses across the wind — a manoeuvre called a tack. The Grand Tack hypothesis borrows that image for Jupiter. In the first few million years of the Solar System, while the Sun was still wrapped in a disk of gas and dust, Jupiter did not sit quietly at 5.2 AU where we find it today. Instead it drifted inward, plunging toward the Sun as far as roughly 1.5 AU — about the present orbit of Mars — and then abruptly reversed and sailed back out. That turnaround is the "tack."

The reason a planet drifts at all is that a young giant planet is not orbiting in vacuum. It is embedded in a dense, viscous gas disk, and it exchanges angular momentum with that gas through gravitational torques. Left alone, those torques pull a Jupiter-mass planet steadily inward. The hypothesis's central claim is that a second planet — Saturn — grew up behind Jupiter, caught it, and flipped the sign of the net torque so that both planets reversed and migrated outward together. The single event reshaped the inner Solar System: it explains why Mars is so surprisingly small and why the asteroid belt is both nearly empty and chemically split into a dry inner zone and an icy outer zone.

The physics of disk migration

A planet embedded in a gas disk raises spiral density waves at its Lindblad resonances. The waves interior to the planet's orbit carry angular momentum the planet's way (pushing it out); the waves exterior carry it away (pushing it in). For a low-mass planet that does not perturb the disk much, this is Type I migration, and the outer torque usually wins slightly, so the planet drifts inward fast. Once a planet is massive enough to clear an annular gap in the gas — the gap-opening criterion is roughly

M_p / M_star  ≳  ( H / r )³        (thermal mass / viscous criterion)

where H/r is the disk aspect ratio (≈ 0.03–0.05 at a few AU) — the planet enters Type II migration. Now it is locked to the gap and drifts inward on the disk's viscous timescale,

t_visc ~ r² / ν        with  ν = α c_s H        (α ≈ 10⁻³–10⁻²)

which for the giant-planet region is of order 10⁵–10⁶ years — fast compared with the few-million-year lifetime of the gas disk. A lone Jupiter in Type II migration would therefore spiral all the way in toward the Sun unless something stops it. The Grand Tack supplies that something: Saturn.

How Saturn flips the direction

The reversal mechanism was identified by Frédéric Masset and Mark Snellgrove in 2001 and built into the full Solar System model by Walsh and collaborators in 2011. The sequence is:

1. Jupiter forms first, near 3.0–3.5 AU, reaches its gap-opening mass, and begins Type II migration inward.
2. Saturn forms later and lighter, around 4.5 AU. Because it is below Jupiter's mass, it migrates inward in a faster, runaway-like regime (sometimes called Type III) and rapidly closes the gap to Jupiter.
3. Capture into 2:3 resonance. Saturn is caught in a mean-motion resonance with Jupiter — for every two of Saturn's orbits, Jupiter completes three. The two planets are now dynamically locked.
4. A shared, asymmetric gap. Saturn is not massive enough to fully clear gas on its own, so the pair carves one common gap. Gas leaks across the gap differently on the two sides, and the inner disk's torque now exceeds the outer disk's. The net torque on the locked pair points outward.
5. The tack. Both planets reverse and migrate outward together until the gas disk dissipates, parking Jupiter near 5.2 AU and Saturn near 7 AU — close to their pre-Nice-model configuration.

The 2:3 commensurability is the load-bearing detail. The reversal only works in a narrow window of mass ratio; hydrodynamic simulations show the torque flips outward when Saturn reaches roughly 0.3–0.6 of Jupiter's mass and the two are locked in 2:3 (some studies favour 1:2). Outside that window the pair keeps drifting inward, and there is no tack.

The numbers that have to line up

Quantity	Value in the Grand Tack	Comment
Jupiter formation radius	~3.0–3.5 AU	Just beyond the water snow line
Inner turnaround radius	~1.5 AU	Near present-day Mars orbit
Saturn formation radius	~4.5 AU	Forms after Jupiter
Jupiter–Saturn resonance	2:3 (some studies 1:2)	Locks the pair before reversal
Final Jupiter radius	~5.2 AU	Set when gas disperses
Disk truncation radius	~1.0 AU	Sets the small-Mars outcome
Event timescale	~10⁵ yr per leg, within first 3–5 Myr	Must finish before gas is gone
Resulting belt mass	~10⁻³ M⊕	Depleted by ~3 orders of magnitude

For scale: 1 AU = 1.496 × 10⁸ km, Jupiter's mass is 318 M⊕ = 9.5 × 10⁻⁴ M☉, Saturn's is 95 M⊕, and Mars's is just 0.107 M⊕. The asteroid belt today contains barely 4–5 × 10⁻⁴ M⊕ of material spread between 2.1 and 3.3 AU — Ceres alone holds about a third of it.

Solving the small-Mars problem

The original motivation for the Grand Tack was an embarrassment in terrestrial-planet formation. When you seed a smooth disk of planetesimals from ~0.5 AU out to ~4 AU and let it accrete, the simulations reliably build a Mars analogue that is far too heavy — typically 0.5 to 1.0 Earth masses, when the real Mars is only 0.107 M⊕. Earth and Venus come out about right; Mars does not. The Earth-to-Mars mass ratio in nature is roughly 9:1, but classical accretion gives something closer to 1:1 or 2:1.

The Grand Tack fixes this by cutting off Mars's food supply. As Jupiter sweeps inward to 1.5 AU it gravitationally herds solid material ahead of it, piling planetesimals into a dense annulus and clearing the region beyond ~1 AU. When the gas is gone and terrestrial accretion proceeds, the planets grow from a narrow ring between roughly 0.7 and 1.0 AU. Earth and Venus form near the centre of that ring where the mass is concentrated; Mars and Mercury are built from the starved fringes scattered to the edges. Simulations starting from such a truncated annulus reproduce not only a small Mars but also the relatively low eccentricities and the radial mass distribution of the four terrestrial planets — the so-called "concentric annulus" success.

Re-sculpting and re-seeding the asteroid belt

The asteroid belt is the Grand Tack's second great success — and arguably its most distinctive prediction. The belt has two puzzles. First, it is almost empty: its total mass is about a thousandth of an Earth, far too little to have ever assembled a planet there, yet the planetesimals plainly formed across that region. Second, it is compositionally stratified. The inner belt (inside about 2.5 AU) is dominated by dry, silicate-rich S-type asteroids; the outer belt by dark, volatile- and water-rich C-type asteroids. Why should two chemically distinct populations sit side by side in the same narrow zone?

The double pass answers both. On the way in, Jupiter ploughs through the primordial belt, scattering most of it — flinging some bodies inward to be swept up by the growing terrestrial planets and ejecting others outward — and depleting the region by a factor of about 1,000. On the way back out, Jupiter scatters volatile-rich planetesimals from beyond its orbit (the cold, icy reservoir from which C-types formed) back inward, capturing a small fraction of them onto belt-like orbits in the outer belt. The dry S-types that originated inside 2.5 AU are partly preserved or re-implanted in the inner belt. The result is exactly the observed layering: rocky in, icy out, with the whole belt severely under-massed. As a bonus, those scattered C-type bodies are a candidate source for delivering Earth's water, since their isotopic (D/H) signature resembles terrestrial ocean water better than comets do.

Grand Tack vs the Nice model

The two flagship migration models of Solar System history are easy to confuse because both move giant planets around. They operate in different epochs by different physics and are best understood as complementary chapters.

Feature	Grand Tack	Nice model
Epoch	First ~3–5 Myr	~50–700 Myr after formation
Gas disk present?	Yes — gas drives it	No — gas long gone
Driving force	Planet–gas torques (Type II)	Planetesimal scattering / resonance crossing
Planets involved	Mainly Jupiter + Saturn	All four giants + a 5th (possibly ejected)
Net Jupiter motion	In to 1.5 AU, then out to 5.2 AU	Small inward jump (~0.2–0.4 AU)
Main thing explained	Small Mars, split asteroid belt, Earth's water	Outer-planet orbits, Kuiper belt, Trojans, Late Heavy Bombardment
Trigger	Saturn catching Jupiter in resonance	Crossing a mutual resonance after disk dispersal

A self-consistent history can run both in sequence: the Grand Tack first parks the giants in a compact resonant chain while the gas disk is still around, and then — hundreds of millions of years later — the Nice-model instability breaks that chain and disperses the giants and the Kuiper belt to their present orbits. The Grand Tack handles the inner Solar System; the Nice model handles the outer.

Evidence and what would falsify it

• The mass of Mars. The single strongest motivation. The Grand Tack reproduces a 0.1-Earth-mass Mars where smooth-disk models cannot.
• The compositional gradient of the belt. The interleaving of S-types inside 2.5 AU and implanted C-types outside it is a direct prediction of the two-pass scattering.
• Belt mass and orbital excitation. The depletion to ~10⁻³ M⊕ and the spread of inclinations and eccentricities match observations of the modern belt.
• Hot-Jupiter analogues. Disk migration of giants is independently required to explain the hundreds of hot and warm Jupiters found around other stars — the Solar System would simply be a case where the migration reversed in time.
• Falsifiers. If the Jupiter–Saturn mass ratio or formation timing could be shown never to fall in the resonance-capture window, or if isotopic dating of meteorites placed the belt's compositional split too early or too late for a few-Myr event, the scenario would fail. Competing "low-mass asteroid belt" and "early instability" models reproduce a small Mars without migration, so the Grand Tack is not the only way to the same outcome.

Common misconceptions and edge cases

• "Jupiter became a hot Jupiter." No — the inner turnaround is around 1.5 AU, not 0.05 AU. The Grand Tack is a modest excursion into the inner system, not a plunge to the stellar surface. It explains why the Solar System lacks a hot Jupiter: the migration reversed before Jupiter got close.
• "The tack happened recently." It is an event of the first few million years, while gas was still present. It must complete before disk dispersal; once the gas is gone there is no torque to drive Type II migration.
• "Grand Tack and Nice model are rival theories of the same thing." They are sequential and complementary, acting in different epochs on different regions by different physics.
• "It is settled science." It is a leading but actively contested hypothesis. The required fine-tuning of timing and the existence of migration-free alternatives keep it under debate.
• "The asteroids we see are the original belt." In the Grand Tack the present belt is mostly re-implanted material — both surviving locals and bodies scattered in from elsewhere — not the pristine planetesimals that first condensed at 2–3 AU.
• "Mars formed where it did from local material." The model implies Mars was largely built from planetesimals originating closer to 1 AU, then left stranded on the truncated disk's edge — a clue some seek in Martian meteorite chemistry.