Planetary Migration

Planetary Migration is the astrophysical process by which a growing planetary embryo or formed planet exchanges orbital energy and angular momentum with a surrounding gaseous protoplanetary disk or residual disk of solid planetesimals, causing its semi-major axis (a), eccentricity (e), or inclination (i) to evolve away from its initial birth radius. First analyzed theoretically by Peter Goldreich and Scott Tremaine in 1979–1980 to explain resonant spiral density waves in Saturn's rings and generalized to gas disks by Douglas N. C. Lin, John Papaloizou, and William R. Ward, planetary migration moved to the forefront of astrophysics in 1995 when Michel Mayor and Didier Queloz discovered 51 Pegasi b—a 0.47-Jupiter-mass gas giant orbiting a solar-type star with an orbital period of 4.23 days at a semi-major axis of 0.052 astronomical units (AU).

Because gas-giant cores require the high solid surface density beyond the water frost line (r > 2–3 AU) to reach the ~10-Earth-mass (M_⊕) threshold for runaway gas accretion—and because inner-disk temperatures exceeding 1,500 K at 0.05 AU preclude in situ condensation of icy cores—the existence of hot Jupiters like 51 Pegasi b and resonant chains of compact super-Earths (such as the seven planets of TRAPPIST-1) proved that planetary orbits are dynamically mobile. Within our Solar System, both gas-driven migration during the first 5 million years (the Grand Tack Hypothesis) and later planetesimal-driven migration after gas dispersal (the Nice Model) sculpted the masses of the terrestrial planets, mixed the Asteroid Belt, and captured Pluto into mean-motion orbital resonances with Neptune.

When a low-mass planetary embryo or sub-Neptune core (0.5 M_⊕ to ~10 M_⊕) is embedded within a gaseous protoplanetary disk, its gravitational perturbation is weak enough that pressure forces and turbulent viscosity prevent the planet from clearing an annular gap in the gas. In this linear regime—designated Type I migration (Ward, 1997; Tanaka, Takeuchi, & Ward, 2002)—the planet interacts with the surrounding gas through two classes of gravitational torques:

1. Differential Lindblad Torques (Spiral Density Waves): A planet orbiting at angular velocity Ω_p(r_p) launches two-armed spiral density waves at radial locations where the Doppler-shifted forcing frequency matches the local epicyclic frequency κ(r) of the gas (Lindblad resonances, r_L = r_p [1 ± 1/m]^2/3). Because gas interior to the planet (r < r_p) orbits faster than the planet (Ω > Ω_p), gravitational attraction between the planet and the inner spiral wake pulls the planet forward, transferring angular momentum from the inner disk to the planet (positive inner Lindblad torque, Γ_ILR > 0, pushing the planet outward). Conversely, gas exterior to the planet (r > r_p) orbits slower (Ω < Ω_p), so the outer spiral wake pulls the planet backward, draining its angular momentum into the outer disk (negative outer Lindblad torque, Γ_OLR < 0, driving the planet inward). Because the radial pressure gradient in a standard disk shifts the outer Lindblad resonances slightly closer to the planet than the inner resonances, the negative outer torque outweighs the positive inner torque, yielding a net inward differential Lindblad torque proportional to M_p² and (H/r)⁻².
2. Corotation (Horseshoe) Torques and Planet Traps: Gas parcels orbiting near the planet's semi-major axis (r ≈ r_p) execute U-turn horseshoe orbits in the corotating frame, librating around the tadpole Lagrange points. As Sijme-Jan Paardekooper and Garrelt Mellema (2006) discovered, in a non-isothermal, radiative disk with a radial entropy gradient, warm gas executing an inward U-turn ahead of the planet cools and contracts while cool gas executing an outward U-turn behind the planet expands. This azimuthal density asymmetry generates a positive entropy-related corotation torque that can cancel or reverse the inward Lindblad torque, driving outward Type I migration toward zero-torque convergence zones (planet traps) at opacity transitions and the water frost line.

As a growing protoplanet surpasses the gap-opening threshold mass (30 to 100 M_⊕), its gravitational Lindblad torques locally exceed the viscous shear torque that refills the co-orbital zone, and its Hill radius exceeds the disk scale height (R_H ≥ H, the Crida–Morbidelli–Masset criterion). The giant planet carves a deep annular density gap, positioning itself as a movable dam in the gas disk. In this Type II migration regime (Lin & Papaloizou, 1986), the giant planet drifts inward on the viscous diffusion timescale (τ_ν ≈ r_p² / [3 ν] ≈ 10⁵ to 10⁶ years), carrying isolated gas giants inward to ~0.05 AU as hot Jupiters.

However, as Masset and Snellgrove (2001) demonstrated, Type II migration reverses direction when a second, lower-mass outer giant planet (Saturn, 95 M_⊕) forms outside a larger inner giant planet (Jupiter, 318 M_⊕). Because the lighter outer planet clears only a shallower gap, it undergoes rapid inward migration and catches up to the slower inner giant, becoming trapped in the mutual 2:3 mean-motion orbital resonance. Once their gaps overlap into a shared cavity, Jupiter experiences a strong positive inner Lindblad torque from the inner disk while lower-mass Saturn experiences a weaker negative outer Lindblad torque from the outer disk, driving coupled outward migration of both gas giants in the Grand Tack Hypothesis.

After photoevaporation disperses the gaseous disk at 3–10 Myr, giant planets migrate by gravitationally scattering a residual disk of solid icy planetesimals (Julio A. Fernández and Wing-Huen Ip, 1984). When Neptune scatters trans-Neptunian planetesimals inward toward Uranus and Saturn, which pass them to high-mass Jupiter, Jupiter ejects the planetesimals out of the Solar System (or into the Oort Cloud). To conserve angular momentum, Jupiter migrates slightly inward by ~0.2 AU, whereas Saturn, Uranus, and Neptune migrate outward (Neptune moving from ~17 AU to 30.1 AU).

In 1993 and 1995, Renu Malhotra showed that as Neptune migrated outward, its exterior 3:2, 5:3, and 2:1 mean-motion resonances swept radially across the primordial Kuiper Belt, adiabatically capturing Pluto (a = 39.48 AU, e = 0.249) and hundreds of plutinos into the 3:2 resonance while pumping up their eccentricities according to Δ(e²) ≈ (1/3) ln(a_final / a_initial).

How does the Rossiter–McLaughlin effect distinguish disk-migrated Hot Jupiters from high-eccentricity tidally migrated Hot Jupiters?

During a planetary transit across a rotating star, the planet blocks the blueshifted approaching limb and then the redshifted receding limb of the stellar photosphere, creating a radial-velocity anomaly (Rossiter–McLaughlin effect) that measures the sky-projected angle (λ) between the stellar spin axis and the planet's orbital normal. Hot Jupiters migrated via Type II gas-disk migration maintain low obliquities (λ ≈ 0°), whereas roughly one-third of Hot Jupiters exhibit tilted or retrograde orbits (λ = 60° to 180°) produced by planet–planet scattering or Kozai–Lidov secular cycles followed by stellar tidal circularization.

Why didn't our Solar System form a close-in chain of Super-Earths like most exoplanetary systems?

Kepler and TESS surveys show that 30% to 50% of sun-like stars host compact systems of super-Earths and sub-Neptunes orbiting inside Mercury's orbit (P = 3 to 100 days), formed when an unimpeded inward pebble flux feeds embryos that undergo Type I migration to the inner disk edge. In our Solar System, the early formation of Jupiter blocked the inward pebble flux and, via its Grand Tack excursion with Saturn, prevented giant-core migration inside 1.5 AU.

What is resonant chain breaking in compact exoplanet systems?

Type I migration parks inward-migrating super-Earths at the inner disk edge in resonant chains (4:3, 3:2, 2:1, as preserved in TRAPPIST-1 and HD 110067). Once the damping gas disk photoevaporates at ~5 Myr, mutual gravitational perturbations destabilize ~90–95% of resonant super-Earth chains within 10–100 Myr ("breaking the chains"), leaving ~5–10% as pristine resonant chains.

1. Goldreich, P., & Tremaine, S. (1980). "Disk-satellite interactions." The Astrophysical Journal, 241, 425–441. https://doi.org/10.1086/158356
2. Fernández, J. A., & Ip, W.-H. (1984). "Some dynamical aspects of the accretion of Uranus and Neptune: The exchange of orbital angular momentum with planetesimals." Icarus, 58(1), 109–120. https://doi.org/10.1016/0019-1035(84)90101-590101-5)
3. Malhotra, R. (1995). "The origin of Pluto's orbit: Implications for the Solar System beyond Neptune." The Astronomical Journal, 110, 420–429. https://doi.org/10.1086/117532
4. Ward, W. R. (1997). "Protoplanet migration by nebula tides." Icarus, 126(2), 261–281. https://doi.org/10.1006/icar.1996.5647
5. Masset, F., & Snellgrove, M. (2001). "Reversing type II migration: Resonance trapping of a lighter giant protoplanet." Monthly Notices of the Royal Astronomical Society, 320(4), L55–L59. https://doi.org/10.1046/j.1365-8711.2001.04159.x