Circum-Jovian Disc (CJD): Formation & Dynamics
- Circum-Jovian Disc is the gas disc surrounding a growing proto-Jupiter, formed by inflowing gas with angular momentum and regulating satellite formation.
- Recent models detail its dynamic interaction with the circumstellar disc, showing phenomena such as tidal truncation, vertical inflow, and pressure-induced substructures.
- Observations and simulations reveal how thermal structure, photoevaporation, and angular momentum transport control gas accretion rates and subsequent satellite development.
The circum-Jovian disc (CJD) is the small gas disc around the gas-accreting proto-Jupiter, formed as inflowing gas with angular momentum settles around the growing planet. In contemporary usage, the term refers primarily to the circumplanetary disc that mediated Jupiter’s late gas accretion and supplied the environment in which the Galilean satellites formed, although analogous Jovian or super-Jovian discs are also discussed observationally. Across current models, the CJD is not treated as an isolated reservoir but as a dynamically coupled component of the circumstellar disc (CSD), with its structure determined by gas inflow, stellar tides, viscosity, irradiation, photoevaporation, and solid delivery from the surrounding nebula [(Rivier et al., 2012); (Shibaike et al., 8 Aug 2025)].
1. Formation, scope, and conceptual models
Classical CJD treatments distinguish between a minimum mass model, which assumes a static, isolated disc with just enough solids to build the Galilean moons, and a gas-starved model, which assumes continuous inflow of low-density gas from the CSD. A more recent viewpoint describes the CJD as an open system with continuous gas and dust or solid inflow from the CSD, shaped by the evolving gap opened by Jupiter and by the dynamics of the surrounding nebula (Shibaike et al., 8 Aug 2025).
Three-dimensional hydrodynamical simulations summarized in recent work indicate an envelope to disc transition during proto-Jupiter’s growth. At early, low mass, proto-Jupiter is surrounded by an envelope; as its mass increases to about , the structure flattens into a disc. The effective outer edge of the CJD is then typically about $0.03$–$0.1$ Hill radii, or about $30$–, depending on the specific angular momentum of the gas and the details of the accretion flow. Gas enters the Hill sphere at high altitude via the region, falls toward the disc surface, and then spirals inward; inside about , the gas is nearly Keplerian in some simulations, while other calculations find outward flows at some radii (Shibaike et al., 8 Aug 2025).
An older but distinct modeling tradition reconstructs a primordial Jovian nebula by fitting an isothermal oscillatory density model to the present-day Galilean and nearby satellites. In that framework, the density maxima of an isothermal Lane-Emden solution with rotation are aligned with the current semimajor axes of the moons, yielding very small radial scale lengths, low rotational support, and long-term stability against self-gravity induced instabilities for millions of years. This is a reconstruction of a possible primordial structure rather than a hydrodynamic accretion-flow model, and it addresses a different level of description from open-system CJD calculations (Christodoulou et al., 2019).
2. Global structure, tidal truncation, and flow geometry
A central structural result is that circumplanetary discs are tidally truncated by the star near a substantial fraction of the Hill radius. In ballistic calculations, stable non-intersecting orbits exist only inside about , and the disc is correspondingly truncated near . The Hill radius is written as
Beyond the orbit-crossing radius, shocks and efficient angular-momentum removal by the stellar tidal field impose a hard outer edge in the cold, pressureless limit (Martin et al., 2010).
In real discs, pressure and viscosity smooth the truncation region. During the T Tauri stage, the disc is expected to be fairly thick, with $0.03$0, and the edge tapering occurs over a radial scale of order $0.03$1. For a circular or slightly eccentric orbit planet, no significant resonances lie within the main body of the disc, but tidally driven waves involving resonances still play an important role in truncating the disc, especially when the disc is fairly thick. One-dimensional time-dependent and steady-state models, together with two-dimensional SPH simulations, show that the internal structure depends on the radial variation of turbulent viscosity and is relatively insensitive to the angular distribution of accreting gas (Martin et al., 2010).
The flow geometry inferred from three-dimensional studies is correspondingly nontrivial. The gas supply is predominantly vertical, not purely radial, and the CJD is replenished from the overlying CSD. This makes the disc neither a simple miniature protoplanetary disc nor a closed subnebula. A plausible implication is that the outer truncation radius, the altitude-dependent inflow geometry, and the internal thermodynamic state cannot be treated independently, because each constrains where gas arrives, where it circularizes, and where solids can be retained or lost [(Rivier et al., 2012); (Shibaike et al., 8 Aug 2025)].
3. Angular-momentum transport and the bottleneck for gas accretion
A major line of work treats the CJD as a possible bottleneck for gas accretion onto giant planets. In the limiting case of a totally inviscid circumplanetary disc, angular momentum is removed only by the gravitational perturbation from the star. The steady-state model introduced for this regime assumes vertical gas inflow at rate $0.03$2 per unit area and an external stellar torque $0.03$3. The angular momentum inside radius $0.03$4 is
$0.03$5
and the radius of the steady-state depleted region is
$0.03$6
The corresponding accretion rate is
$0.03$7
In this formulation, the accretion rate depends almost linearly on the stellar torque and only weakly on the inflow rate (Rivier et al., 2012).
Two-dimensional hydrodynamical simulations in a planetocentric frame show that the disc develops a two-armed spiral wave caused by stellar tides. The spiral propagates from the outer edge toward the planet and acts as the medium by which torque is transmitted through the disc. The torque is negative, is applied mostly at the outer edge, and in the inviscid limit is փոխանցed through pressure from ring to ring until it exerts an effective negative torque at the inner edge, causing depletion from the inside out. The measured torque is small, relatively insensitive to disc aspect ratio $0.03$8, and minimally affected by low but non-zero numerical viscosity (Rivier et al., 2012).
For a Jupiter-mass planet, the inflow estimate drawn from earlier simulations is
$0.03$9
and the resulting accretion rate is about
$0.1$0
with a steady-state radius $0.1$1. This corresponds to a mass-doubling time for Jupiter of order $0.1$2 Myr, far slower than the $0.1$3 yr associated with classic runaway accretion models. Under the paper’s limit assumptions, this is explicitly a lower bound on the real accretion rate. The central implication is that gas accretion onto a giant planet can be regulated by its circumplanetary disc, and that differences in disc viscosity may contribute to the diversity of exoplanet giant masses (Rivier et al., 2012).
4. Thermal structure, irradiation, self-shadowing, and photoevaporation
The thermodynamic state of the CJD is highly sensitive to accretion history and radiative environment. Recent synthesis of three-dimensional hydrodynamic results emphasizes that the disc can be very hot near Jupiter, with midplane temperatures exceeding the water-ice sublimation point during early, high-accretion phases. In one quantitative estimate, a gas accretion rate of $0.1$4 yields midplane temperatures above $0.1$5 K, preventing icy moon formation; lower accretion rates are therefore required for ice survival (Shibaike et al., 8 Aug 2025).
A dedicated two-dimensional quasi-stationary model of Jupiter’s irradiated circumplanetary disc adds a further geometric effect: self-shadowing. In that model, Jupiter’s intense radiative heating warms the optically thick inner disc, but the inner disc can cast shadows onto more distant regions. The shadowed zone is associated with a temperature drop of about $0.1$6 K relative to surrounding regions and, in the reference calculation, lies around $0.1$7, specifically between $0.1$8 and $0.1$9. These cold regions can act as traps for volatiles such as $30$0, $30$1, and $30$2. Depending on metallicity and viscosity, the shadowed structures may persist for $30$3 kyr in low-metallicity cases and up to $30$4 kyr in dustier cases, with radial coverage as large as $30$5–$30$6. This suggests that the CJD may have had a non-monotonic thermal structure during the epoch of Galilean moon formation (Schneeberger et al., 2024).
An additional external control is photoevaporation. Once Jupiter opens an annular gap in the protoplanetary disc and the gap becomes sufficiently optically thin, the Jovian CPD is exposed to far-ultraviolet radiation from the young Sun and from the Solar birth cluster. In this regime, the outer radius is set by competition between viscous replenishment and photoevaporative loss. A key scaling derived for the steady-state truncation radius is
$30$7
For $30$8 and $30$9, the modal truncation radius is 0, within Callisto’s present orbit at 1. For CPD accretion rates 2, photoevaporative truncation explains the lack of additional satellites outside the orbit of Callisto; for 3, dispersal can occur before Callisto migrates into the Laplace resonance (Oberg et al., 2020).
5. Transport of solids and satellite formation within the CJD
The transport of solids from the circumstellar disc into the CJD remains a central unresolved problem. Once Jupiter opens a deep gap in the CSD, pebbles of about centimeter-to-meter size are trapped at the gas-pressure maximum at the outer gap edge and cannot be directly supplied to the CJD. The leading alternatives are therefore: small dust particles accreting onto the CJD together with the gas, or planetesimals, less affected by the gas, being captured by the CJD. Recent 3D simulations indicate that turbulent diffusion can loft small grains to high altitudes where they can join meridional inflows, while planetesimal capture may be assisted by gravitational scattering and gas drag. None of the proposed solid-delivery scenarios has been fully accepted (Shibaike et al., 8 Aug 2025).
Satellite formation models within the CJD are correspondingly divided into satellitesimal accretion and pebble accretion scenarios, and in both cases the satellite growth timescale of about 4–5 Myr depends on continuous supply of material to the disc. The observed compositional gradient and degree of differentiation of the Galilean moons place strong constraints on these models, because the radial temperature structure determines where water ice survives, and inward migration may transport bodies across the snow line. A plausible implication is that solid transport, thermochemical evolution, and migration stopping mechanisms must be solved simultaneously rather than sequentially (Shibaike et al., 8 Aug 2025).
A separate line of N-body work explored a massive, static, low viscosity circum-planetary disc motivated by minimum mass sub-nebula prescriptions. In that model, the gas mass is about 6, the solid mass about 7, the turbulent viscosity is low with 8, and the surface density follows 9 with either 0 or 1. Hydrodynamic simulations found no sign of gap opening by satellites, so type II migration was not expected; type I migration, eccentricity damping, and inclination damping were therefore used. Within that framework, a flatter profile (2) and a hotter disc with 3 place the ice line near 4 and are more favorable for forming massive icy satellites such as Ganymede and Callisto, whereas a colder disc with 5 places the ice line near 6, too close to Jupiter (Moraes et al., 2017).
6. Substructure, resonances, alternative architectures, and observational extensions
One longstanding issue is why Io, Europa, and Ganymede participate in the 7 Laplace resonance while Callisto does not. The prevailing explanation has been late or slow Callisto formation, so that migration stalled before resonant capture. An alternative now proposed is that disk substructure—specifically a pressure bump—can act as a migration trap. In the corresponding N-body simulations, the bump is positioned interior to the birthplaces of all four moons. Io, Europa, and Ganymede can be sequentially trapped at the bump and then ushered across it through resonant lockstep migration with an exterior neighbor, while Callisto remains isolated at the bump edge. The outcome depends sensitively on bump structure: if the bump is too sharp, the system becomes unstable; if too flat, all four moons cross and form a four-body 8 chain. The successful “Goldilocks” range corresponds to a bump aspect ratio 9–0 (Yap et al., 2 Jan 2026).
This substructure-based interpretation does not eliminate earlier explanations, but it reframes them. The controversy is not whether migration mattered—it clearly did—but whether Callisto’s present orbit primarily records timing of accretion, disk dispersal, or disk substructure. The current literature supports all three as viable ingredients: rapid photoevaporative clearing can strand Callisto (Oberg et al., 2020), pressure bumps can isolate it dynamically (Yap et al., 2 Jan 2026), and the uncertainty in Callisto’s degree of differentiation remains relevant to slow-formation arguments (Shibaike et al., 8 Aug 2025).
Observationally, direct evidence for active circumplanetary accretion now extends the CJD concept beyond the Solar System. High-resolution near-UV spectroscopy of the 1-Myr super-Jovian companion Delorme 1 (AB)b shows resolved hydrogen emission lines and supports a planetary accretion shock with 2. The favored models imply pre-shock velocity 3, planetary mass 4, and either 5 or 6, with the higher density requiring a line-emitting area of about 7 or less of the planetary surface. The small filling factor and asymmetric line profiles favor magnetospheric accretion and show that gas-rich circumplanetary discs can persist to ages far beyond canonical few-Myr dispersal times, in so-called “Peter Pan disc” systems (Ringqvist et al., 2022).
A distinct, post-gaseous extension of the subject concerns circumplanetary debris discs produced by satellite-system disruption during planet-planet scattering. Simulations with Galilean-like satellite systems around 8 planets find that 9 percent of planets undergoing scattering will possess debris from satellite destruction, yielding tilted, optically thick debris discs around eccentric planets. These are not proto-satellite CJDs in the usual gas-rich sense, but they show that circumplanetary disc phenomena persist across very different dynamical regimes and observational manifestations (Mustill et al., 2024).
