Introduction
The Nice model is a dynamical scenario for giant-planet migration in the early Solar System, originally developed by Tsiganis, Gomes, Morbidelli, and Levison at the Cote d'Azur Observatory (Nice) in 2005. The model proposes that the giant planets formed in a more compact configuration than their present-day positions and underwent a chaotic dynamical instability driven by mean-motion-resonance crossings. The instability triggered the Late Heavy Bombardment (LHB), shaped the Kuiper Belt and Oort Cloud, and rearranged the giant-planet system into its present configuration. This report reviews the model and the observational tests.
Pre-instability configuration
The Nice model proposes that the four giant planets initially had semi-major axes between approximately 5.5 and 14 AU, substantially more compact than their present positions (Jupiter at 5.2 AU, Saturn at 9.5 AU, Uranus at 19 AU, Neptune at 30 AU). A dense planetesimal disc extended from approximately 14 to 35 AU, containing approximately 35 Earth masses of material. The system was stable for approximately 700 million years before the instability triggered.
The Jupiter-Saturn 2:1 resonance crossing
The instability trigger is the Jupiter-Saturn 2:1 mean-motion resonance crossing. Slow gravitational interaction between the planets and the planetesimal disc caused the planets to migrate inward (Jupiter) and outward (Saturn, Uranus, Neptune) over the 700 Myr pre-instability phase. As the orbits diverged, Jupiter and Saturn approached the 2:1 resonance, and the resonance crossing produced a rapid increase in eccentricity for both planets. The eccentric orbits then perturbed Uranus and Neptune onto chaotic, near-crossing orbits.
Chaotic phase
During the chaotic phase, lasting approximately 100 Myr, the giant planets exchanged angular momentum and the planetesimal disc was substantially disrupted. Uranus and Neptune migrated outward through the disc, with Neptune moving from approximately 12 AU to its present 30 AU. The migration scattered planetesimals throughout the inner Solar System, producing the LHB. Some planetesimals were captured into stable Trojan orbits at the L4 and L5 Lagrange points of Jupiter and Neptune; others were deposited into the present-day Kuiper Belt with characteristic resonance-trapped and scattered-disc populations.
Trojan asteroid populations
The Jupiter Trojan asteroids (the L4 and L5 swarms at Jupiter's orbit) are predicted by the Nice model to be a captured planetesimal-disc population rather than primordial Jupiter-orbital material. The capture mechanism requires the chaotic phase: under the standard pre-instability conditions, the Jupiter Trojan capture rate is too small to produce the observed population. The composition of Jupiter Trojans (similar to dark, organic-rich Kuiper Belt objects rather than to inner-Solar-System asteroids) is consistent with a Kuiper-Belt-region origin.
NASA's Lucy mission (launched 2021) is conducting a tour of the Jupiter Trojan asteroids through 2027-2033, providing the first direct observations of selected Trojans. Early results from the 2023 main-belt asteroid Dinkinesh flyby and the 2025 first Trojan encounter at Eurybates have been consistent with the predicted population properties.
Kuiper Belt structure
The Nice model predicts specific structural features in the Kuiper Belt:
The hot classical population is implanted from a more distant primordial reservoir during the chaotic phase, accounting for the inclination distribution that primordial in-situ formation cannot reproduce.
The cold classical population is in-situ-formed at the present-day distance, surviving the chaotic phase because of its dynamically-cold initial orbits.
Resonance populations at the 3:2, 2:1, and higher-order resonances are captured during Neptune's outward migration. The relative occupation of different resonances depends on the smoothness of the migration; the observed pattern favours a chaotic-then-smooth migration history rather than purely smooth migration.
Observational tests
The Nice model has passed several specific observational tests:
Lunar LHB ages. The radiometric ages of Apollo lunar basins cluster around 3.85-4.0 Ga, consistent with the Nice-model timing of the chaotic phase.
Trojan asteroid composition and binary fraction. Both are consistent with Kuiper-Belt-region origin rather than Jupiter-orbital primordial origin.
Kuiper Belt cold-vs-hot dichotomy. Reproduced naturally in Nice-model simulations.
Several tensions remain. The pre-instability planetesimal disc mass implied by the model is somewhat higher than independent estimates. The 5:2 resonance population in the Kuiper Belt is anomalously large for the standard Nice model and requires a more complex migration history (sometimes called the "chaotic Nice" model). The lunar LHB age clustering has been called into question by recent re-analyses suggesting smoother age distributions, which would favour a smoother migration model.
Outlook
The Nice model framework remains the primary working hypothesis for early-Solar-System dynamical evolution. Continued refinement comes from improved observational constraints (Lucy mission Trojan results through 2033, Vera C. Rubin Observatory Kuiper Belt survey beginning 2026) and from progressively-detailed numerical simulations. Alternative migration scenarios (smooth-only migration, the Grand Tack model in which Jupiter migrated inward and outward) are still in active development for specific use-cases but the Nice model framework is the most-cited.