Terrestrial planet formation in low eccentricity warm-Jupiter systems

We examine the effect of giant planet migration on the formation of inner terrestrial planet systems. We consider situations in which the giant planet halts migration at semi-major axes in the range 0.13 - 1.7 AU due to gas disk dispersal. An N-body code is employed that is linked to a viscous gas disk algorithm capable of simulating: gas loss via accretion onto the central star and photoevaporation; gap formation by the giant planet; type II migration of the giant; optional type I migration of protoplanets; gas drag on planetesimals. We find that most of the inner system planetary building blocks survive the passage of the giant planet, either by being shepherded inward or scattered into exterior orbits. Systems of one or more hot-Earths are predicted to form and remain interior to the giant planet, especially if type II migration has been limited, or where type I migration has affected protoplanetary dynamics. Habitable planets in low eccentricity warm-Jupiter systems appear possible if the giant planet makes a limited incursion into the outer regions of the habitable zone (HZ), or traverses its entire width and ceases migrating at a radial distance of less than half that of the HZ’s inner edge. We conclude that Type II migration does not prevent terrestrial planet formation.

💡 Research Summary

This paper investigates how the migration of a giant planet influences the formation of terrestrial planets interior to its final orbit, focusing on systems where the giant halts its Type II migration at semi‑major axes between 0.13 AU and 1.7 AU because of gas‑disk dispersal. The authors employ a hybrid numerical framework that couples a three‑dimensional N‑body integrator with a viscous protoplanetary‑disk model. The disk model accounts for gas accretion onto the star, photo‑evaporative loss, gap opening by the giant, Type II migration of the giant, optional Type I migration of protoplanets, and aerodynamic drag on planetesimals. By varying the giant’s stopping distance, the timing of gas‑disk dispersal, the initial mass distribution of planetary embryos, and the inclusion or exclusion of Type I migration, the study explores a wide parameter space relevant to observed warm‑Jupiter and hot‑Jupiter systems.

The simulations reveal two dominant dynamical pathways for the inner solid material. In the “shepherding” channel, embryos that lie ahead of the migrating giant are pushed inward by a combination of resonant torques and gas drag. Their orbits shrink, leading to a cascade of collisions that builds one or more compact, rocky planets interior to the giant. These “hot‑Earths” typically attain masses of 0.1–0.5 M⊕ and remain on low‑eccentricity orbits. Shepherding is most efficient when the giant’s migration is curtailed early—either because the gas disk dissipates rapidly or because the planet reaches its final orbit before traversing the entire inner disk.

In the “scattering” channel, embryos and planetesimals are gravitationally scattered outward as the giant passes. The scattered material can re‑accumulate beyond the giant’s orbit, potentially forming additional terrestrial planets in the outer disk. When the giant’s final orbit lies near or just inside the habitable zone (HZ), the outward‑scattered debris can populate the HZ with sufficient mass to assemble water‑rich, potentially habitable worlds. The authors define the HZ for a solar‑type star as roughly 0.95–1.67 AU; they find that if the giant stops at a distance less than half the inner edge of this zone (≈0.5 AU), the inner system still retains enough solid material to build Earth‑mass planets inside the HZ. Conversely, if the giant penetrates deep into the HZ but halts before the zone’s outer edge, a fraction of the original solid mass remains in the outer part of the HZ, allowing for the formation of habitable planets with modest water inventories.

Including Type I migration for the embryos accelerates inward drift, enhancing shepherding but simultaneously depleting the outer disk of water‑rich material. This trade‑off leads to more massive, drier hot‑Earths and fewer water‑bearing planets in the HZ. When Type I migration is switched off, the solid mass distribution remains more uniform, favoring the formation of both interior hot‑Earths and exterior, water‑rich terrestrial planets.

The paper’s key conclusions are: (1) Type II migration of a giant planet does not categorically prevent terrestrial planet formation; the outcome depends sensitively on the migration distance, the timing of gas‑disk dispersal, and the dynamical response of the embryo population. (2) Warm‑Jupiter systems with low eccentricities can host multiple inner rocky planets, especially when the giant’s incursion into the HZ is limited or when it stops well inside the HZ’s inner edge. (3) Habitable planets are plausible in such systems if the giant either grazes the outer HZ or halts at a radius smaller than roughly half the HZ’s inner boundary, leaving enough material to assemble Earth‑mass bodies with adequate volatile content. (4) The presence or absence of Type I migration critically shapes the mass and composition of the resulting planets, offering a potential diagnostic for interpreting observed system architectures.

Observationally, the results suggest that low‑eccentricity warm‑Jupiter systems—such as WASP‑47, Kepler‑9, and similar multi‑planet configurations—should be prime targets for searches of interior hot‑Earths and exterior habitable‑zone planets using transit timing variations, radial‑velocity follow‑up, and future direct‑imaging missions. The study thus bridges theoretical planet‑formation modeling with emerging exoplanet surveys, providing a framework to assess the habitability potential of systems dominated by a single, migrated giant planet.