Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Planetary Science. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 19 January 2021

The Interiors of Jupiter and Saturnfree

  • Ravit HelledRavit HelledUniversity of Zurich, Center for Theoretical Astrophysics and Cosmology


Probing the interiors of the gaseous giant planets in our solar system is not an easy task. It requires a set of accurate measurements combined with theoretical models that are used to infer the planetary composition and its depth dependence. The masses of Jupiter and Saturn are 317.83 and 95.16 Earth masses (M), respectively, and since a few decades, it has been known that they mostly consist of hydrogen and helium. The mass of heavy elements (all elements heavier than helium) is not well determined, nor are their distribution within the planets. While the heavy elements are not the dominating materials inside Jupiter and Saturn, they are the key to understanding the planets’ formation and evolutionary histories.

The planetary internal structure is inferred from theoretical models that fit the available observational constraints by using theoretical equations of states (EOSs) for hydrogen, helium, their mixtures, and heavier elements (typically rocks and/or ices). However, there is no unique solution for determining the planetary structure and the results depend on the used EOSs as well as the model assumptions imposed by the modeler.

Major model assumptions that can affect the derived internal structure include the number of layers, the heat transport mechanism within the planet (and its entropy), the nature of the core (compact vs. diluted), and the location (pressure) of separation between the two envelopes. Alternative structure models assume a less distinct division between the layers and /or a non-homogenous distribution of the heavy elements. The fact that the behavior of hydrogen at high pressures and temperatures is not perfectly known and that helium may separate from hydrogen at the deep interior add sources of uncertainty to structure models. In the 21st century, with accurate measurements of the gravitational fields of Jupiter and Saturn from the Juno and Cassini missions, structure models can be further constrained. At the same time, these measurements introduce new challenges for planetary modelers.


Investigating the interiors of the giant planets in the solar system goes back several decades. Jupiter and Saturn are located at radial distances of about 5.2 and 9.6 AU from the Sun and their composition is dominated by light elements, in particular, hydrogen and helium (hereafter H–He).1 Jupiter and Saturn are massive fast rotators, and their atmospheres are characterized by impressive signatures of dynamics. These colorful atmospheres, however, represent only the “skin” of the planets and cannot reveal the secrets of their internal structures. Therefore, despite the significant progress on both the observational and theoretical fronts, Jupiter and Saturn remain mysterious planets.

Due to their large distances from the Earth and their gaseous nature, revealing information on the deep interiors of Jupiter and Saturn must be done by using indirect measurements. As more information about the planets is collected, more comprehensive theoretical structure models must be developed. However, it becomes increasingly challenging to find a self-consistent theoretical framework that meets all the observational constraints. The accurate measurements of the physical and chemical planetary properties provide new and important constraints, but they also lead to new open questions. The current situation in modeling planetary interiors can be summarized by Albert Einstein quote: “The more I learn, the more I realize how much I don’t know.” Fitting the new data requires more complex structure models and the inclusion of various physical processes and assumptions which are not well justified or completely understood.

Currently, planetary scientists still don’t have unique and self-consistent views of the interiors of Jupiter and Saturn, but there is a better understanding of the relevant physical and chemical processes that should be considered as well as the limitations of the theoretical approaches. Making progress in that direction does not only help in studying planetary interiors but also in better understanding the behavior of simple elements at high pressures and temperatures, and in putting important constraints on giant planet formation and evolution models.

In this article, the current knowledge of the internal structures of Jupiter and Saturn is summarized. Recent reviews of this topic include Militzer, Soubiran, Wahl, and Hubbard (2016), Guillot and Gautier (2014), Helled and Guillot (2018), Baraffe, Chabrier, Fortney, and Sotin (2014), Fortney and Nettelmann (2010), Fortney et al. (2016), and the references therein.

Making an Interior Model

Observational Constraints

Internal structure models are designed to fit the observed physical data of the planets, such as their masses, radii, gravitational and magnetic fields, 1-bar temperatures, atmospheric composition, and internal rotations. Key physical properties of Jupiter and Saturn are summarized in Table 1. Interestingly, the atmospheres of both Jupiter and Saturn show a depletion in helium in comparison to the helium mass fraction of the protosolar value of Yproto ∼ 0.275 as inferred from stellar evolution models for the Sun (e.g., Bahcall, Pinsonneault, & Wasserburg, 1995). The measured helium mass fractions in Jupiter and Saturn are found to be ∼ 0.238 (von Zahn, Hunten, & Lehmacher, 1998) and 0.18–0.25 (Conrath & Gautier, 2000), respectively. As discussed below, this does not imply that Jupiter and Saturn are depleted in helium globally but rather that the distribution of helium is inhomogeneous within their interiors due to the phenomenon of helium settling (see the section “Helium and Hydrogen” for details). In addition, for Jupiter, the Galileo entry probe provided abundances measurements of other components, suggesting that Jupiter’s outer envelope is enriched with heavy elements by a factor of ∼ 24 compared to protosolar abundance (e.g., Atreya, Mahaffy, Niemann, Wong, & Owen, 2003; Guillot & Gautier, 2014). Two exceptions were neon and oxygen, both found to be depleted; however, neon is expected to be affected by the process of helium rain (Roustlon & Stevenson, 1995; Wilson & Militzer, 2010) and the low abundance of water is probably linked to the special location of the probe’s entry, which is known as a “dry spot” where the atmosphere is dry and does not represent the bulk of the atmosphere. Therefore, at the moment the oxygen abundance in Jupiter is still unknown.

Since giant planets consist of mostly fluid hydrogen and helium, they do not have a solid surface below the clouds like terrestrial planets. Therefore, the “surface” of the planet is defined as the location where the pressure is 1 bar, comparable to the pressure at the Earth’s surface. The temperature at this location is measured (with a small uncertainty). Then, with information of the temperature at 1 bar, the entropy of the outer envelope is determined and adiabatic models can be constructed.2

The gravitational field of the planet, or more precisely, the total potential, which also includes the rotational term, and is given by


where (r, θ‎, φ) are spherical polar coordinates, a is the equatorial radius, and M is the total planetary mass. The potential U is represented as an expansion in Legendre polynomials (e.g., Zharkov & Trubitsyn, 1978), where typically only the even indices (i.e., P2n) are taken into account due to the primarily north–south symmetry of the two hemispheres.3 The gravitational harmonic coefficients J2n are typically inferred from Doppler tracking data of a spacecraft orbiting or flying by the planet and are used to constrain the density profile as discussed below.

Governing Equations

The planetary interior is modeled by solving the standard structure equations, which include the mass conservation, hydrostatic balance, and thermodynamic equations as follows:


where P is the pressure, ρ is the density, m is the mass within a sphere of a radius r, and ω is the rotation rate. In order to account for rotation, the hydrostatic equation (Equation 2) includes an additional term that depends on ω, which is assumed to be constant (i.e., uniform rotation), for a non-spinning planet: ω = 0. For a rapidly rotating planet, this equation is valid in the limit of a barotropic fluid and a solid-body rotation. The radius r is then considered as a mean volumetric radius. Equation 3 is the first-order expansion of the total potential U . The temperature gradient ∇T ≡ d ln T/d ln P depends on the heat transport mechanism (convection vs. conduction/radiation). Typically, the temperature gradient is taken to be the smallest among the adiabatic ∇ad, radiative/conductive ∇rad/cond gradients, since the heat transport mechanism that leads to the smallest temperature gradient is the most efficient one. In other words, the temperature gradient is taken to be ∇T = min[∇ad, ∇rad/cond].4 Finally, in order to solve this set of equations, the density dependence on the temperature and pressure needs to be known, that is, ρ(P, T ), which is determined by the equation of state.

Table 1. Basic Observed Properties of Jupiter and Saturn

Physical Property



Distance to Sun (AU)



Mass (1024 kg)

1,898.13 ± 0.19

568.319 ± 0.057

Mean radius (km)

69,911 ± 6

58,232 ± 6

Equatorial radius (km)

71,492 ± 4

60,268 ± 4

Mean density (g/cm3)

1.3262 ± 0.0004

0.6871 ± 0.0002

J2 × 106

14,696.572 ± 0.014

16,290.557 ± 0.028

J4 × 106

−586.609 ± 0.004

−935.318 ± 0.044

J6 × 106

34.24 ± 0.24

86.334 ± 0.112




Rotation period

9 hr 55 min 29.56 s

10 hr 39 min ± ~10 min

Effective temperature (K)

124.4 ± 0.3

95.0 ± 0.4

1-bar temperature (K)

165 ± 4

135 ± 5

Note: Adapted from Helled and Guillot (2018), NASA(2019a), and references therein. Jupiter’s gravitational field is taken from Iess et al. (2018). The gravitational coefficients correspond to the reference equatorial radii of 71,492 km and 60,330 km for Jupiter and Saturn, respectively (see NASA [2019b]). These are theoretical values based on interior model calculations. See “Rotation Rate and Depth of Winds” and Helled et al. (2015) for discussion on Saturn’s rotation rate uncertainty.

The density profile of the planets is set to reproduce the measured gravitational moments J2n. The relation between the gravitational moments and the density profile is given by (e.g., Zharkov & Trubitsyn, 1978):


where the integration is carried out over the volume τ. Traditionally, the theoretical density profile and gravitational moments were calculated by using the theory of figures (TOF) in which the harmonics are computed from a series approximation in the smallness parameter m = ω2R3/GM , where R is the mean radius of the planet, typically up to an order of three or four. This was sufficient as long as no information was known about the gravitational coefficients beyond J6 and the measurements have relatively large uncertainties. An alternative method to TOF that is designed to be compatible with accurate data and provide estimates for the higher-order harmonics was developed by W. Hubbard (e.g., Hubbard, 2012, 2013). In this approach, called concentric Maclaurin spheroid (CMS), the density profile is represented by a large number of Maclaurin spheroids where a continuous density can be achieved if the number of spheroids is large enough. While the computational resources needed are large, the gravitational coefficients can be calculated to any order with an excellent precision (∼109). Discussion and comparison between the TOF and CMS methods can be found in Hubbard, Schubert, Kong, and Zhang (2014), Wisdom and Hubbard (2016), Nettelmann (2017), Debras and Chabrier (2018), and references therein.

Equation of State

In thermodynamics, the equation of state (EOS) relates the state variables such as the temperature, pressure, density, internal energy, and entropy. Since Jupiter and Saturn are mostly composed of H–He, modeling their structures relies on information of the EOS of hydrogen, helium, and their mixture. Giant planet interiors serve as natural laboratories for studying different elements at exotic conditions that do not exist on Earth. At the same time, calculating the EOS of materials in Jupiter and Saturn interior conditions is a challenging task because molecules, atoms, ions, and electrons coexist and interact, and the pressure and temperature range varies by several orders of magnitude, going up to several tens of megabars (Mbar) (i.e., 100 GPa and several 104 Kelvins). Therefore, information on the EOS at such conditions requires performing high-pressure experiments and/or solving the many-body quantum mechanical problem to produce theoretical EOS tables that cover such a large range of pressures and temperatures. Despite the challenges, there have been significant advances in high-pressure experiments and ab initio EOS calculations. The EOS of hydrogen, helium, and heavier elements is briefly described below. More information on that topic can be found in Fortney and Nettelmann (2010), Baraffe et al. (2014), Militzer et al. (2016), Guillot and Gautier (2014), Helled and Guillot (2018), and references therein.


Hydrogen is the most abundant element in the universe, and yet its phase diagram is still a topic of intensive research. The behavior of hydrogen at high pressures can be investigated experimentally and theoretically. There are several types of laboratory experiments such as gas guns, convergent shock waves, and laser-induced shock compression that can probe hydrogen (or its isotope, deuterium) at Mbar pressures. Unfortunately, the available experimental data have a large range, and each experiment suffers from different limitations and systematics. Nevertheless, some progress has been made toward convergence when comparing the shock Hugoniot curve of hydrogen from different studies. Although empirical, the information from experiments is limited, and theoretical calculations are required to provide a wide-range EOS for hydrogen. On the other hand, laboratory experiment results are of primary importance since they are used to calibrate the theoretical EOS used for planetary modeling.

The most popular ab initio technique in materials science, and the most common approach to probe the EOS of hydrogen and helium at planetary conditions, is density functional theory (DFT). Although the theory is exact, all existing practical implementations rely on approximations. While DFT calculations provide a relatively accurate determination of the EOS of hydrogen in a large range of temperatures and pressures, using moderate computational resources, it shows a poor performance in assessing phase transitions (e.g., Azadi & Foulkes, 2013). An alternative approach is quantum Monte Carlo (QMC), which is a wave function-based method and can accurately solve the electronic problem (e.g., Mazzola, Helled, & Sorella, 2018 and references therein). This approach is much more computationally expensive, but is potentially one order of magnitude more accurate than DFT (e.g., Foulkes, Mitas, Needs, & Rajagopal, 2001) and therefore can closely simulate the phase transitions, where the (free) energy difference between the competing phases (at given thermodynamic conditions) can be small. At the moment, QMC calculations can be used to calibrate other existing wide-range EOS tables and are expected to play an important role in EOS calculations in the 2020s.

Figure 1 shows the phase diagram of dense hydrogen and H–He mixture. Shown is the transition between the insulating molecular and the metallic-atomic hydrogen fluid (shaded area). The location of this transition is not only important for understanding the generation of strong magnetic fields in these planets but also for determining the division of the planets to layers with different structure models (brown and mustard color lines, respectively).

There are several interesting conclusions about Jupiter and Saturn that can be made simply by looking at the hydrogen phase diagram. First, both planets lie in the regime above solid hydrogen, suggesting that they are fluid planets, as already suggested by Hubbard (1968). Second, both planets cross the critical point of hydrogen, which indicates that in the outer parts of the planets hydrogen is in the molecular form (H2) and in the metallic form in the deep interiors. Metallic hydrogen is a phase of hydrogen at high pressures/temperatures where electrons are free and hydrogen becomes an excellent conductor, like a metal. The main uncertainty concerning the hydrogen EOS is in the region of 0.5–10 Mbars (50−103 GPa), where the transition from the molecular phase to the metallic phase occurs. In fact, the metallization of hydrogen is an active area of research and the exact metallization pressure/temperature is still being debated, but it is expected to be at ∼1 Mbar for Jupiter’s conditions.5 Finally, it is clear that Saturn’s adiabat covers lower temperatures and pressures (due to its lower mass). As a result, Saturn’s interior consists of a smaller fraction of metallic hydrogen in comparison to Jupiter, and since this regime of the EOS is less understood, there is less uncertainty in Saturn’s structure due to the hydrogen EOS (Saumon & Guillot, 2004). On the other hand, as discussed below, Saturn is more likely to be affected by the phase separation of helium.

Figure 1. Phase diagram of hydrogen. Gray area: solid phase (experimental); shaded blue: insulating molecular liquid; solid blue: first-order liquid–liquid phase boundary (QMC); dashed blue: continuous liquid–liquid phase boundary (QMC); black points: possible location of the critical point (endpoint of the first-order line); solid red: first-order for the H–He (QMC); dashed red: continues liquid–liquid boundary H–He (QMC); empty triangles: static compression experiments H (Zaghoo, Salamat, & Silvera, 2016; Ohta et al., 2015); solid triangles: dynamic compression H (Knudson et al., 2015).

Adapted from Mazzola et al. (2018).
Helium and Hydrogen

The behavior of pure helium at the extreme conditions inside giant planet interiors is more constrained than that of hydrogen simply because helium ionization requires larger pressures and a phase transition is not expected to take place. In Figure 1, the blue and red curves correspond to the phase diagram of pure hydrogen and a H–HE mixture (red) with a protosolar value, respectively (see Mazzola et al. [2018] for details). It is clear that the presence of helium delays the dissociation (metallization) pressure compared to pure hydrogen and therefore the presence of helium cannot be neglected when estimating the metallization pressures in Jupiter’s and Saturn’s interiors.

Figure 2. Phase diagram for a H–He mixture. The orange region shows the region of the H- He separation as derived by Lorenzen et al. (2011). The red curve shows the critical temperature for the separation, according to Morales et al. (2013). Numerical and experimental results by Schouten, de Kuijper, and Michels (1991) and Loubeyre, Letoullec, and Pinceaux (1991) are also presented. The back curves show the isoentropes of Jupiter (plain) and Saturn (dashed), respectively.

Adapted from Guillot and Gautier (2014).

In addition, the interaction between hydrogen and helium under the interior conditions of Jupiter and Saturn leads to challenges in determining the EOS. This is because helium is expected to become immiscible in hydrogen, leading to helium settling (known as “helium rain”) that results in a non-homogenous distribution of helium within the planet, where helium settles (and is therefore enriched) toward the deep interior. This phenomenon of helium rain was already predicted in the 1970s (e.g., Salpeter, 1973; Stevenson, 1975; Stevenson & Salpeter, 1977a, 1977b) and received observational support when the helium in the atmospheres of Jupiter and Saturn was found to be depleted in comparison to the protosolar value. Recently, ab initio calculations of the phase diagram have confirmed the immiscibility of helium in hydrogen (Salpeter, 1973; Lorenzen, Holst, & Redmer, 2011; Morales, Hamel, Caspersen, & Schwegler, 2013; Sch¨ottler & Redmer, 2018, and references therein). Figure 2 shows the phase diagram for a H–He mixture with a helium mole concentration slightly lower than protosolar (see Guillot & Gautier [2014] for details). The exact location in the phase diagram in which helium rain occurs is still being investigated and is of great importance for understanding the structure and evolution of both Jupiter and Saturn. What seems to be robust is that since Saturn has a smaller mass than Jupiter and therefore its internal temperatures and pressures are lower, it is located “deeper” within the phase diagram in comparison to Jupiter (see Figure 2). This means that the process of helium rain is more significant inside Saturn and has begun earlier. Jupiter’s interior has to cool for a longer time to reach the temperatures corresponding to this phase separation. This is consistent with the measurements of helium in the giant planet atmospheres where Saturn’s atmosphere is found to be more depleted in helium.

Heavy Elements

In astrophysics, heavy elements represent all the elements that are heavier than helium. Ideally, structure models should include all the possible elements when modeling the planetary interior. However, this introduces an additional complexity to the models because the ratios between the different elements within the planets have to be assumed. In addition, the details of the EOS of the heavy elements are less crucial, since the temperature dependence on the density of these elements is rather weak at giant planet interior conditions, and their contribution to the planetary density is of a second-order effect in comparison to H–He (e.g., Baraffe et al., 2008; Saumon & Guillot, 2004; Fortney et al., 2016). Often, the heavy elements in Jupiter and Saturn are represented by water and/or rock, where rocks lead to about 50% less massive cores compared to water (e.g., Fortney & Nettelmann, 2010). While the measured gravitational field is essentially blind to the innermost regions of the planet, these regions can affect the density profile indirectly via the constraints on the outer envelope (e.g., Helled, Anderson, Schubert, & Stevenson., 2011; Guillot & Gautier, 2014). Currently there is ongoing progress in ab initio calculations of the EOSs for water, ammonia, silicates, and iron as well as their miscibility in metallic hydrogen (e.g., French et al., 2009; Knudson et al., 2012; Wilson & Militzer, 2010, 2012). As such ab initio EOS calculations become available, it is desirable to include them in structure models and further investigate their effect on the inferred composition and internal structure.

Internal Structure Models

Since several decades, studies aimed to better constrain the interiors of Jupiter and Saturn (e.g., Saumon & Guillot, 2004; Militzer, Hubbard, Vorberger, Tamblyn, & Bonev, 2008, Militzer et al., 2016; Nettelmann et al., 2008; Nettelmann, Becker, Holst, & Redmer, 2012; Nettelmann, Fortney, Moore, & Mankovich, 2015; Helled & Guillot, 2013; Hubbard & Militzer, 2016; Miguel et al., 2016). Unfortunately, there is no unique solution for the internal structure of a planet. The non-unique nature of the problem is inherent because the available data are (and will remain) insufficient to uniquely infer the planetary internal structure. In addition, the inferred structure depends on the model assumptions and the EOSs used by the modeler. The main uncertainties in structure models are linked to the following assumptions and setups: (a) the number of layers, (b) the composition and distribution of heavy elements, (c) the heat transport mechanism, (d) the transition pressure of hydrogen metallization, and (e) the rotation period and the dynamical contribution of winds (e.g., differential rotation).

Typically, the interiors of Jupiter and Saturn are modeled assuming the existence of a distinct heavy-element core which is surrounded by an inner envelope of metallic hydrogen and an outer envelope of molecular hydrogen. Due to the indication of helium rain in the planets, the inner and outer envelopes are set to be helium-rich and helium-poor, respectively. For the heavy elements distribution there are two common assumptions. In the first, they are assumed to be homogeneously mixed within the two envelopes. Then, if Zin and Zout represent the heavy element mass fraction in the inner and outer envelopes, respectively, for this case Zin = Zout. In the second case, the heavy element enrichment is assumed to be higher at the metallic region (inner envelope) (i.e., Zin > Zout).

Figure 3. Sketches of the internal structures of Jupiter and Saturn as inferred from structure models. For each planet, two possible structures are shown: one consisting of distinct layers and one with a gradual distribution of heavy elements. Schematic representation of the interiors of Jupiter and Saturn. The core masses of Jupiter and Saturn are not well constrained; for Saturn, the inhomogeneous region could extend down all the way to the center, resulting in a “helium core.”

In both of these cases, the heavy elements are taken to be homogeneously distributed, suggesting a homogenous composition, at least within one part of the envelope. The location where the envelope is divided into a helium-poor–helium-rich region corresponds to the pressure in which helium becomes immiscible in hydrogen. For simplicity, for the models with ZinZout the location of the heavy element discontinuity is assumed to occur at the same location.

The division of the planetary structure into three layers is not written in stone; it only represents the simplest model that can be considered. It may indeed be that the discontinuity in helium and the heavy elements occurs at different pressures, and especially in the case of Saturn, that a nearly pure helium layer also exists (e.g., Fortney & Hubbard, 2003). In addition, the core itself may not be a distinct region (as well as non-existing), and the heavy elements may have a gradual distribution along the planetary interior. Even within this simple three-layer model framework, the inferred composition and core mass depend on the model assumptions. More complex models increase the range of possible solutions even further. Figure 3 shows a simple sketch of the interiors of Jupiter and Saturn. Figure 4 shows representative density and pressure profiles within the planets for standard three-layer models.


For Jupiter, structure models typically differ by the assumption regarding the heavy element distribution, the assumed number of layers, and the calculated entropy from EOS calculations. Saumon and Guillot (2004) explored the possible range of solutions for Jupiter using different EOSs and inferred total heavy element mass between 10 and 40 M, and core masses between 0 and 10 M. Later, when DFT EOS calculations became available, new Jupiter models were presented. The first set of

Figure 4. Representative density (left) and pressure (right) profiles of Jupiter and Saturn as a function of the planetary mean radius. The data are taken from Miguel et al. (2016) and Helled and Guillot (2013), respectively.

models was based on the entropy calculation of the Rostock H–He EOS, which was calculated using ab initio DFT (e.g., Nettelmann et al., 2008, 2012; Becker et al., 2014). In Nettelmann et al. (2008), Nettelmann, Helled, Fortney, and Redmer (2012), Becker et al. (2014), and Nettelmann (2017), the models relied on the three-layer assumption, where the interior is separated into a distinct core and two homogeneous envelopes. The helium mass fraction in the outer envelope, Yout, was set to match the Galileo entry probe value of Y = 0.238. The inner envelope helium abundance Yin is chosen to yield a bulk helium mass fraction that reproduces the protosolar value. The heavy element mass fractions Zout and Zin were chosen to match the then measured values of the low-order gravitational harmonics J2 and J4, with J4 being slightly more sensitive to Zout than J2 is, and vice versa for Zin. In the transition between the inner and outer envelopes, the pressure and temperature are assumed to change continuously, while the density and entropy have discontinuities. In these models, the transition pressure Ptrans is taken to be a free parameter, between 1 and 5 Mbar, although H–He phase diagram calculations suggest that Ptrans ∼ 1 Mbar. Higher assumed Ptrans values lead to higher envelope metallicity and smaller inferred values for the core mass. Overall, the studies of this group have confirmed the ranges derived by Saumon and Guillot (2004), where the core mass was smaller than ∼10 M, with a global enrichment of tens M of heavy elements.

A second type of Jupiter models based on Militzer’s H–He EOS calculation, also using ab initio DFT-MB (e.g., Militzer et al., 2008; Hubbard & Militzer, 2016). A comprehensive equation of state of H–He mixtures and their inferred internal energies as well as a Jupiter adiabat have been presented by Militzer and Hubbard (2013). These models lead to a significant inferred core mass for Jupiter of the order of 15–20 M with a low envelope metallicity, sometimes even less than solar.

Note that in these Jupiter models, the transition pressure is not taken as a free parameter but is set by identifying the location in which the pressure of Jupiter’s adiabat intersects with the H–He immiscibility region as derived by Morales et al. (2013; see Figure 2). It should be noted that the EOSs used by Militzer and collaborators and Nettelmann and collaborators are not very different in terms of the raw data, but in the entropy calculation, and therefore in the constructed adiabat. A hotter adiabat, as inferred by Nettelmann and collaborators, leads to a larger inferred heavy element mass (see Militzer et al. [2016] and Miguel et al. [2016] for further discussion).

The recent gravity measurements from the Juno spacecraft (Bolton et al., 2017; Iess et al., 2018) introduced new constraints on Jupiter structure models. Jupiter models that fit the Juno data have been presented by Wahl et al. (2017), Nettelmann (2017), and Guillot et al. (2018). Overall, it seems that preferable solutions are ones with cores (∼10 M) and a discontinuity of the heavy element enrichment in the envelope, with the inner helium-rich envelope consisting of a larger fraction of heavy elements than the outer helium-poor envelope (i.e., Zin > Zout). In addition, interior models of Jupiter that fit Juno data suggest that another feasible solution for Jupiter’s internal structure is the existence of a diluted/fuzzy core (e.g., Wahl et al., 2017). In this case, Jupiter’s core is no longer viewed as a pure heavy element central region with a density discontinuity at the core-envelope-boundary, but as a central region whose composition is dominated by heavy elements, which could be gradually distributed or homogeneously mixed. Such a diluted core could extend to a few 10s of percentages of the planet’s total radius and can also consist of lighter elements (H–He). While the total amount of heavy elements in the central region of diluted core models in the central region does not change much (e.g., Wahl et al., 2017, Nettelmann, 2017), the size of the core increases significantly due to the inclusion of H–He.

The existence of a diluted core or a steep heavy element gradient inside Jupiter is actually consistent with formation models of Jupiter (see “Constraints on Internal Structure and Origin” for details). Giant planet formation models in the core accretion scenario (e.g., Pollack et al., 1996) suggest that once the core mass reaches ∼ 12M, the accreted solid material (heavy elements) vaporizes and remains in the planetary envelope (e.g., Stevenson, 1982). This leads to a structure in which the deep interior is highly enriched with heavy elements, with no sharp transition between the core and the inner envelope (e.g., Helled & Stevenson, 2017, and references therein). Another explanation for a diluted core is core erosion. If the heavy elements within a compact core are miscible in metallic hydrogen (e.g., Wilson & Militzer, 2010, 2012), the presence of vast convection could mix some of the core elements in the deep interior (e.g., Guillot, Stevenson, Hubbard, & Saumon, 2004). Long-term evolution models of Jupiter with composition gradients suggest that steep composition gradients can persist up to present-day (see the section “Evolution Models”).

Finally, it is important to note that inferred core mass and total planetary enrichment do not only depend on the assumed EOS but also on the model assumptions. Jupiter’s structure models as presented by Wahl et al. (2017) show that both fuzzy and compact cores are consistent with the Juno data, with the core mass being between ∼ 1.5 and 20 M, depending on the model. A Jupiter structure model with a diluted core resembles the primordial structure derived by formation models (Stevenson, 1985; Helled & Stevenson, 2017), providing a potential link between giant planet formation models and the current-state structure of the planets. In all of these models, Jupiter’s normalized moment of inertia was found to be ∼0.264. This value is relatively well constrained, at least from the modeling point of view, due to the accurate determination of Jupiter’s gravitational field by Juno (e.g., Wahl et al., 2017). Indeed, it has been shown that there is a very strong correlation between J2 and the moment of inertia, but this is not a perfect one-to-one correspondence (e.g., Helled, Anderson, Podolak, & Schubert [2011], and references therein).


Saturn is often considered to be a small version of Jupiter but, in fact, the two planets have significant differences. First, the relative enrichment in heavy elements is rather different, as well as the geometry, the magnetic field, the axis tilt, and the long-term evolution. Just from a simple comparison of their normalized moment of inertia values, one can conclude that Saturn is more centrally condensed compared to Jupiter. Naively, one would expect that it is easier to model Saturn’s interior, since a smaller portion of its mass sits in the region of the high uncertainty in the hydrogen EOS, but this is not the case due to the possibility of helium rain. Additional complication arises from the uncertainty in Saturn’s rotation period and shape (see Fortney et al., 2016, and references therein).

Saturn models calculated by Helled and Guillot (2013) also used the three-layer model approach. The range of the helium mass fraction in the outer envelope Yout was taken to be between 0.11 and 0.25, with a global Y = 0.265–0.275 consistent with the protosolar value. Here the EOS of H–He was set to the Saumon, Chabrier, van Horn value (SCVH), which was calculated for a large range of pressures and temperatures (Saumon et al., 1995) and has been broadly used in the astrophysics community. A range of temperatures at 1 bar was considered (130–145 K) and Ptrans was allowed to range between 1 and 4 Mbars. For the heavy element distribution, they assumed Zin = Zout. These Saturn models also accounted for the uncertainty in Saturn’s shape and rotation rate and rotation profile (see “Rotation Rate and Depth of Winds”). Saturn models were constructed for two different assumed rotation periods for both the Voyager and Cassini gravity data. For the range of different model assumptions, the derived core mass was found to range between ∼ 5 and 20 M, while the heavy element mass in the envelope was found to be between ∼ 0 and 7 M. Like for Jupiter, increasing Ptrans leads to smaller core masses and more enriched envelopes.

Finally, the Cassini gravity data reduces the inferred core mass by about 5 M. It should be noted, however, based on the recent Juno data for Jupiter as well as new studies on giant planet formation and evolution, that the assumption of Zin = Zout may be inappropriate, and a more realistic assumption is Zin > Zout. The total heavy element mass in Saturn is estimated to be ∼16–30 M, with a core mass between zero and 20 M (e.g., Saumon & Guillot, 2004; Nettelmann et al., 2012; Helled & Guillot, 2013). However (see “Internal Structure Models”), this conclusion is based on relatively simple interiors models. Figure 5 provides a schematic presentation of the two possible internal structures of Jupiter and Saturn.

Figure 5. Sketches of the internal structures of Jupiter and Saturn.

Non-Adiabatic Interiors

Standard structure models of Jupiter and Saturn assume that the dominating energy transport mechanism is convection; that is, that the temperature gradient is given by the adiabatic one, apart from the (thin) outer radiative atmosphere. This assumption simplifies the calculation, since the temperature profile is then well constrained, and in addition, one can assume that the composition within the envelope(s) is homogenous. However, it is now realized that in some cases (and perhaps in most cases), a fully adiabatic model for the giant planets is too simplistic. Non-adiabatic giant planet interiors are in fact a natural outcome of their formation process where the accreted heavy elements result in a non-homogenous interior (e.g., Stevenson, 1985; Helled & Stevenson, 2017; Lozovsky, Helled, Rosenberg, & Bodenheimer, 2017). Non-adiabatic interiors can also be a result of core erosion (e.g., Guillot et al., 2004) and immiscibility of materials in metallic hydrogen (e.g., Wilson & Militzer, 2012; Soubrian & Militzer, 2016).

The existence of composition gradients can inhibit convection due to their stabilizing effect. Moderate composition gradients can be erased by overturning convection, especially at early evolution stages where convection is strong, which leads to a rapid mixing and homogenization of the planet. Otherwise, they can either lead to layered convection, a less efficient type of convection (e.g., Wood, Garaud, & Stellmach, 2013), or inhibit convection and lead to heat transport by conduction and/or radiation.

Leconte and Chabrier (2012, 2013) accounted for the possibility of double-diffusive convection in both Jupiter and Saturn interiors caused by heavy element gradients. It was shown that both Jupiter and Saturn can satisfy all the observational constraints also when assuming non-adiabatic structures with compositional gradients throughout the entire planetary interiors. Since in this scenario heat loss (cooling) is less efficient, the planetary interiors can be much hotter, and the planets can accommodate larger amounts of heavy elements. The core masses derived by these models were found to be 0–0.5 M for Jupiter and ~10–21 M for Saturn. The heavy element mass in the envelope was found to be 41–63.5 M and 10–36 M for Jupiter and Saturn, respectively (see Leconte & Chabrier, 2012 for further details). Although these models could be viewed as extreme cases, since the composition gradients are assumed to persist across the entire planetary interiors, they clearly demonstrate the importance of the model assumptions and the limitation of the simple three-layer models. It is also interesting to note that although the semi-convective structure models for Jupiter and Saturn are more rich in heavy elements, the solution for Jupiter indicates the absence of a core.

Evidence of a non-adiabatic interior for Saturn is also indicated from the observed frequency spectrum of its ring oscillations. Some of Saturn’s ring modes observed by the Cassini spacecraft can be attributed to oscillations within the planetary interior (Hedman & Nicholson, 2013). An analysis of the splitting of these oscillation modes suggests the existence of a thick stably stratified region above the core where gravity modes can penetrate (Fuller, 2014). Currently, this is the only proposed explanation to the unexpected splittings via interactions between f-modes propagating in the convective envelope and g-modes propagating in the stable region of the deep interior. While further investigations on this topic are required, these important observation and analysis further suggest that a fully convective structure is too simplistic for describing Saturn’s (and possibly Jupiter’s) interior.

Evolution Models

Another piece of information that can be used to constrain structure models is the planetary evolution. The idea is that the current-state structure of the planets must be consistent with the age of the solar system (~4.56×109 years) (i.e., with the planetary evolution). In fact, the simple assumption of an adiabatic structure was originated by an evolution model where it was shown that the high thermal emission of Jupiter is somewhat consistent with an interior that is convective (e.g., Hubbard, 1968; Guillot, Gautier, Chabrier, & Mosser, 1994; Fortney, Ikoma, Nettelmann, Guillot, & Marley, 2011). Evolution models with layered convection in the helium rain region of Jupiter and Saturn have recently been calculated (Nettelmann et al., 2015; Mankovich, Fortney, & Moore, 2016). In these models, the molecular envelope cools over time, but the deep interior can even heat up if the super-adiabaticity in the inhomogeneous He rain zone is strong. While it is not yet clear whether layer convection occurs in the helium mixing region, these models show that non-adiabaticity is an important aspect that should be considered when calculating the long-term evolution of gaseous planets such as Jupiter and Saturn.6

Evolution models with primordial composition gradients for Jupiter and Saturn have also been presented (Vazan, Helled, Podolak, & Kovetz, 2016; Vazan, Helled, & Guillot, 2018). It was found that a moderate primordial heavy element gradient becomes homogenous via convective mixing after several million years and that this mixing leads to an enrichment of the planetary envelope with heavy elements. Of the other hand, if the primordial composition gradient is steep, convection in the deep interior was found to be inhibited. This affects the thermal evolution and leads to hotter interiors in comparison to the standard adiabatic case. As in the structure models with layered convection, also here the total heavy element mass in the planets is higher than in the adiabatic models and was found to be up to 40 M and 36 M for Jupiter and Saturn, respectively.

New Insights from the Juno and Cassini Missions

In July 2016, the Juno mission began to orbit Jupiter and, among other things, has provided an accurate measurement of Jupiter’s gravitational and magnetic fields (e.g., Bolton et al., 2017; Folkner et al., 2017; Iess et al., 2018). At the same time, the Cassini spacecraft performed its last orbits having geometries similar to that of Juno, known as Cassini Grand Finale, providing similar information about Saturn’s fields (e.g., Spilker, 2012), allowing a comparative study of the solar system giant planets. Studies and investigations are still ongoing and more results are likely to appear at the time of (or after) this article, but some key conclusions and new insights on the structure of Jupiter and Saturn from these recent measurements have already been achieved.

Rotation Rate and Depth of Winds

The atmospheres of both Jupiter and Saturn have strong zonal winds, with equatorial speeds of ~100 m s1 and 400 m s1, respectively. These zonal wind velocities are relative to the assumed rotation period of the planet’s deep interior (Table 1). In fact, it is not necessarily intuitive to think that giant planets rotate as solid bodies and are therefore represented by a single rotation period due to the fact that they are fluid objects and are characterized by zonal winds that hint the possibility of differential rotation (on cylinders).

Jupiter’s rotation period is assumed to be represented by the rotation period of its magnetic field which is tilted ~10 ° from its spin pole and has not changed in many decades (e.g., Riddle & Warwick, 1976; Higgins, Carr, & Reyes, 1996). On the other hand, Saturn’s magnetic pole is aligned with its rotation axis. This spin-aligned configuration and the fact that the magnetic field is dipolar prevent a direct determination of the rotation rate of Saturn’s deep interior because there is no variable component of the magnetic field that is associated with the planetary rotation (e.g., Cao, Russell, Christensen, Dougherty, & Burton, 2011; Cao, Russell, Wicht, Christensen, & Dougherty, 2012). The 10 hr 39 min 22 s rotation period of Saturn, which leads to an equatorial speed of 400 m s1 , was derived from the Voyager spacecraft measurement of the periodicity in Saturn’s kilometric radiation (e.g., Ingersoll & Pollard, 1982; Dessler, 1983). In fact, measurements from the Cassini spacecraft did not only measure a different periodicity by several minutes but also showed that the period is changing with time (e.g., Gurnett et al., 2007), suggesting that the periodicity in Saturn’s kilometric radiation does not represent the rotation of the deep interior. Therefore, at the moment Saturn’s rotation period is not well constrained. Accordingly, the atmospheric zonal wind velocities with respect to the underlying rotating planet are also unknown for Saturn. Several theoretical approaches have been presented to constrain Saturn’s rotation period, and the estimated values range between ~10 hr 32 min and 10 hr 45 min (e.g., Anderson & Schubert, 2007; Read, Dowling, & Schubert, 2009; Helled, Galanti, & Kaspi, 2015; Mankovich et al., 2018). While an uncertainty of about 10 minutes sounds small, it can affect the inferred internal structure of the planet and also has implications on its atmosphere dynamics.

The relation between the rotation period of Jupiter and Saturn to their zonal wind, physical shapes, and gravitational and magnetic fields has been studied for decades and is still being investigated. Nevertheless, recently substantial progress in this direction has been made thanks to the Cassini and Juno missions. Deep winds can change the planetary density profile and therefore contribute to the measured gravitational harmonics, and as a result this contribution has to be accounted for as an uncertainty in structure models, since they are hydrostatic and do not include dynamical effects. The depth of the winds can be constrained by accurate measurements of the high-order gravitational and/or the odd harmonics (e.g., Hubbard, 1999; Kaspi, Hubbard, Showman, & Flierl, 2010; Kaspi et al., 2018). A determination of the depth of the winds in Jupiter was recently possible thanks to the Juno data (Iess et al., 2018; Kaspi et al., 2018; Guillot et al., 2018). The winds were found to penetrate to depths of 2000–3000 km, suggesting that 1% of the outer planetary mass rotates differentially in patterns similar to that of the observed atmospheric winds. This depth is consistent with the one expected from ohmic dissipation constraints linked to the metallization of hydrogen (e.g., Liu, Goldreich, & Stevenson, 2008). Since the metallization in Saturn occurs at deeper regions (due to its smaller mass and resulting pressures), then by following the same argument the depth the winds in Saturn is predicted to reach deeper, down to ~9000 km. These estimates correspond to depths of around 95% and 80% of the total planetary radius for Jupiter and Saturn, respectively (e.g., Cao & Stevenson, 2017).

Magnetic Fields

Both Jupiter and Saturn possess dipolar intrinsic magnetic fields. The existence and nature of the magnetic fields provide important observational constraints on their current interior structure and dynamics. The existence of an intrinsic magnetic field requires large-scale motions in a medium that is electrically conducting (e.g., Roberts & King, 2013). For Jupiter and Saturn, large-scale radial motions are caused by convective motions which also transport heat from the deep interior toward the outer regions, with the conducting material being metallic hydrogen. Indeed, a significant electrical conductivity is expected inside Jupiter and Saturn before the full metallization of hydrogen at Mbar (100 GPa) pressures (e.g., French et al., 2012). The magnitude of the electrical conductivity inside Jupiter and Saturn combined with the planetary measured magnetic field strength and surface luminosity can be used to estimate the internal ohmic dissipation and introduce additional constraints for structure models (e.g., Liu et al., 2008; Cao & Stevenson, 2017). Thus, our ability to decode the interiors of Jupiter and Saturn from the measured properties of the magnetic field is limited by our current understanding of the dynamo process.

Jupiter’s intrinsic magnetic field is strongest among all solar system planets, with surface field strength ranging from 4 Gauss to 20 Gauss (Connerney et al., 2018; Moore et al., 2018). Recent Juno observations revealed several surprising factors in the morphology of Jupiter’s magnetic field. When viewed at the dynamo surface, Jupiter’s magnetic field is characterized by an intense isolated magnetic spot near the equator with negative flux, an intense and relatively narrow band of positive flux near 45 ° latitude in the northern hemisphere, and a relatively smooth magnetic field in the southern hemisphere. The north–south dichotomy in Jupiter’s magnetic field morphology has been speculated to be due to the existence of a diluted core inside Jupiter, which either limits the dynamo action to the upper layer of Jupiter or creates two spatially separate dynamo actions inside Jupiter (Moore et al., 2018). This provides a nice link between internal structure models that are based solely on the gravity data and magnetic field measurements.

Saturn’s intrinsic magnetic field is unusually weak, with surface field strength ranging from 0.2 Gauss to 0.5 Gauss (Dougherty et al., 2005; Cao et al., 2011, 2012). Surprisingly, Saturn’s magnetic field seems to be perfectly symmetric with respect to the spin axis (Cao et al., 2011, 2012). Both the weak strength and the extreme spin axis-symmetry of the magnetic field of Saturn were attributed to helium rain, which could create a stably stratified layer atop the deep dynamo. However, whether helium rain or composition gradients inside Saturn create substantial stable stratification and whether this stratification layer is above rather than below the deep dynamo are still being investigated.

Constraints on Internal Structure and Origin

In the standard view of giant planet formation, known as core accretion (e.g., Pollack et al., 1996), a giant planet forms in three stages:

Phase 1: Primary core/heavy element accretion. During this early phase, the core accretes solids (planetesimals and/or pebbles) until it empties its gravitational dominating region (feeding zone). The mass associated with the end of this stage is known as “isolation mass” and its exact value depends on the local formation conditions. At this point, the planet is primarily composed of heavy elements with a negligible fraction of an H–He envelope.

Phase 2: Slow envelope accretion. During this phase, the solid accretion rate decreases, and the H–He accretion rate increases until the envelope accretion rate exceeds the heavy element accretion rate. The growth of the envelope enlarges the planet’s feeding zone and thus allows heavy elements to be accreted but at a slow rate.

Phase 3: Rapid gas accretion. Once the H–He mass is comparable to the heavy element mass, the gas accretion rate continuously increases and exceeds the heavy element accretion rate until the disk can no longer supply gas fast enough to maintain equilibrium and keep up with the planetary contraction, and a rapid hydrodynamic accretion of H–He initiates.

In the early core accretion simulations, for the sake of numerical simplicity, it was assumed that all the heavy elements reach the core while the envelope is composed of H–He. However, formation models that follow the heavy element distribution during the planetary formation show that once the core mass reaches a small value of ~1–2 M and is surrounded by a small envelope, the solids composed of heavy elements tend to dissolve in the envelope instead of reaching the core (e.g., Lozovsky et al., 2017; Helled & Stevenson, 2017). In this case, the resulting giant planet has a small core mass and an inner envelope which is enriched with heavy elements. Interestingly, this view is consistent with the possibility of Jupiter having a diluted/fuzzy core. Although the prediction on the distribution of heavy elements corresponds to Jupiter right after its formation, evolution models confirm that in several cases such a structure can persist until present time (e.g., Vazan et al., 2016, 2018). A point that still needs to be investigated is whether the composition discontinuity in the heavy elements is caused by the formation process or a result of phase separations and core erosion that occur at later stages during the planetary long-term evolution.

Another missing piece of the puzzle in our understanding Jupiter is linked to its water abundance. The low abundance of water in Jupiter’s atmosphere measured by the Galileo probe is likely to be a result of the special entry spot, which keeps the enrichment of water in Jupiter’s atmosphere unknown. Jupiter’s water abundance is now being measured by Juno using a microwave radiometer (MWR), which probes down to pressure levels of ~100 bar at radio wavelengths ranging from 1.3 to 50 cm using six separate radiometers to measure thermal emissions. The water measurement is not only important for constraining Jupiter’s origin (e.g., Helled & Lunine, 2014 and references therein) but also for further constraining structure models. First, since updated Jupiter interior models suggest that Zin > Zout, the water measurement will provide a lower bound for the total (water) enrichment within Jupiter. Second, since several of the new interior models infer a low metallicity for the outer envelope, they could be excluded. Finally, the variation of water with depth can provides information about Jupiter’s atmosphere dynamics and put constraints on the convective behavior in its upper atmosphere and also indicates the presence of a non-convective region within Jupiter’s interior. It should be kept in mind that the MWR measurement still reveals the information about a very small fraction of the planet. Nevertheless, when combined with other measurements, it will provide new insights about the most massive planet in the solar system. Future studies should explore the relations between various formation and evolution model assumptions and the inferred planetary composition and internal structure.


There are still many unsolved questions regarding the origins and internal structures of Jupiter and Saturn. As open questions are being solved, new questions arise and our understanding is still incomplete. Nevertheless, the early decades of the 21st century is a golden era for giant planet exploration given the ongoing Juno mission and the recent measurements from Cassini Grand Finale that are still being processed. The possibility of having similar information about Jupiter and Saturn simultaneously opens opportunities to improve our understanding of giant planets and to explore the physical and chemical processes that lead to the differences. It is now known that even within our planetary system, there are significant differences between the two giant planets, suggesting that there is no one simple way to model giant planet interiors.

The continuous theoretical efforts and the new measurements from Juno and Cassini provide data that will keep planetary modelers busy for a while. In the meantime, the knowledge of the EOS of different elements and their interaction needs to keep improving, and all the available information (gravity field, magnetic field, atmospheric composition, etc.) needs to be combined to further constrain the planetary interior. In addition, it is desirable to develop a united theoretical framework for giant planet formation, evolution, and current-state structure.

Future missions will also play an important role in better constraining the interiors of Jupiter and Saturn. The upcoming JUICE mission can reveal further information on Jupiter, and a potential Saturn probe mission will provide constraints on Saturn’s atmospheric composition and the process of helium rain. Finally, the detection and characterization of giant planets around other stars, combined with the knowledge of solar system giants, can lead to a more comprehensive understanding of gaseous planets.


The author thanks Nadine Nettelmann, Hao Cao, and Guglielmo Mazzola for their important contributions. The author also acknowledges valuable comments and support from David Stevenson, Tristan Guillot, and Allona Vazan, as well as the two anonymous referees. Finally, the author acknowledges all the Juno science team members for inspiring discussions.


  • Anderson, J. D., & Schubert, G. (2007). Saturn’s gravitational field, internal rotation, and interior structure. Science, 317, 1384.
  • Atreya, S. K., Mahaffy, P. R., Niemann, H. B., Wong, M. H., & Owen, T. C. (2003). Composition and origin of the atmosphere of Jupiter: An update, and implications for the extrasolar giant planets. Planetary and Space Science, 51, 105–112.
  • Azadi, S., & Foulkes, W. M. C. (2013). Fate of density functional theory in the study of high-pressure solid hydrogen. Physical Review B, 88, 014115.
  • Bahcall, J. N., Pinsonneault, M. H., & Wasserburg G. J., (1995). Solar models with helium and heavy- element diffusion. Reviews of Modern Physics, 67, 781808.
  • Baraffe, I., Chabrier, G., & Barman, T. (2008). Structure and evolution of super-Earth to super-Jupiter exoplanets. I. Heavy element enrichment in the interior. Astronomy and Astrophysics, 482(1), 315–332.
  • Baraffe, I., Chabrier, G., Fortney, J., & Sotin, C. (2014). Planetary internal structures. In H. Beuther, R. S. Klessen, C. P. Dullemond, & T. Henning (Eds.), Protostars and planets VI (p. 763). Tucson: University of Arizona Press.
  • Becker, A., Lorenzen, W., Fortney, Jonathan J., Nettelmann, N., Schöttler, M., & Redmer, R. (2014). Ab Initio Equations of State for Hydrogen (H-REOS.3) and Helium (He-REOS.3) and their Implications for the Interior of Brown Dwarfs. The Astrophysical Journal Supplement, 215(2), 14.
  • Bolton, S. J., Adriani, A., Adumitroaie, V., Allison, M., Anderson, J., Atreya, S., . . . Wilson, R. (2017). Jupiter’s interior and deep atmosphere: The initial pole-to-pole passes with the Juno spacecraft. Science, 356, 821–825.
  • Cao, H., Russell, C. T., Christensen, U. R., Dougherty, M. K., & Burton, M. E. (2011). Saturn’s very axisymmetric magnetic field: No detectable secular variation or tilt. Earth and Planetary Science Letters, 304, 22–28.
  • Cao, H., Russell, C. T., Wicht, J., Christensen, U. R., & Dougherty, M. K. (2012). Saturn’s high degree magnetic moments: Evidence for a unique planetary dynamo. Icarus, 221, 388–394.
  • Cao, H., & Stevenson, D. (2017). Zonal flow magnetic field interaction in the semi-conducting region of giant planets. Icarus, 296, 59–72.
  • Connerney J. E. P., Kotsiaros, S., Oliversen, R. J., Espley, J. R., Joergensen, J. L., Joergensen, P. S., . . . Levin, S. M. (2018). A new model of Jupiter’s magnetic field from Juno’s first nine orbits. Geophysical Research Letters, 45(6), 2590–2596.
  • Conrath, D., & Gautier, D. (2000). Saturn helium abundance: A reanalysis of Voyager measurements. Icarus, 144, 124–134.
  • Debras, F., & Chabrier, G. (2018). A complete study of the precision of the concentric MacLaurin spheroid method to calculate Jupiter’s gravitational moments. Astronomy and Astrophysics, 609, Article 97.
  • Dessler, A. J. (1983). Physics of the Jovian magnetosphere. New York, NY: Cambridge University Press.
  • Dougherty, M. K., Achilleos, N., Andre, N., Arridge, C. S., Balogh, A., Bertucci, C., . . . Tsurutani, B. T. (2005). Cassini magnetometer observations during Saturn orbit insertion. Science, 307, 1266–1270.
  • Folkner, W. M. et al. (2017). Jupiter gravity field estimated from the first two Juno orbits. Geophysical Research Letters. Manuscript submitted for publication.
  • Fortney, J. J., Helled, R., Nettelmann, N., Stevenson, D. J., Marley, M. S., Hubbard, W. B., & Iess, L. (2016). Invited review for the forthcoming volume Saturn in the 21st century. eprint arXiv:1609.06324.
  • Fortney, J. J., & Hubbard, W. B. (2003). Phase separation in giant planets: Inhomogeneous evolution of Saturn. Icarus, 164, 228–243.
  • Fortney, J. J., Ikoma, M., Nettelmann, N., Guillot, T., & Marley, M. S. (2011). Self-consistent model atmospheres and the cooling of the solar system’s giant planets. Astrophysical Journal, 729, 32.
  • Fortney, J. J. & Nettelmann, N. (2010). The interior structure, composition, and evolution of giant planets. Space Science Reviews, 152, 423–447.
  • Foulkes, W. M. C., Mitas, L., Needs, R. J., & Rajagopal, G. (2001). Quantum Monte Carlo simulations of solids. Reviews of Modern Physics, 73, 33–83.
  • French, M., Becker, A., Lorenzen, W., Nettelmann, N., Bethkenhagen, M., Wicht, J., & Redmer, R. (2012). Ab initio simulations for material properties along the Jupiter adiabat. Astrophysics Journal Supplement Series, 202(1), Article 5.
  • French, M., Mattsson, T. R., Nettelmann, N., & Redmer, R. (2009). Equation of state and phase diagram of water at ultrahigh pressures as in planetary interiors. Physical Review B, 79(5), 054107
  • Fuller, J. (2014). Saturn ring seismology: Evidence for stable stratification in the deep interior of Saturn. Icarus, 242, 283–296.
  • Guillot, T., & Gautier, D. (2014). Giant planets. In T. Spohn & G. Schubert (Eds.), Treatise on geophysics (2nd ed.). Amsterdam, The Netherlands: Elsevier.
  • Guillot, T., Gautier, D., Chabrier, G., & Mosser, B. (1994). Are the giant planets fully convective? Icarus, 112, 337–353.
  • Guillot, T., Miguel, Y., Militzer, B., Hubbard, W. B., Kaspi, Y., Galanti, E., . . . Bolton, S. J. (2018). A suppression of differential rotation in Jupiters deep interior. Nature, 555, 227–230.
  • Guillot, T., Stevenson, D. J., Hubbard, W. B., & Saumon, D. (2004). The interior of Jupiter. In F. Bagenal, T. E. Dowling, & W. B. McKinnon (Eds.), Jupiter: The planet, satellites and magnetosphere. Cambridge Planetary Science. Cambridge, U.K.: Cambridge University Press.
  • Gurnett, D. A., Persoon, A. M., Kurth, W. S., Groene, J. B., Averkamp, T. F., Dougherty, M. K., & Southwood, D. J. (2007). The variable rotation period of the inner region of Saturn’s plasma disk. Science, 316, 442–445.
  • Hedman, M. M., & Nicholson, P. D. (2013). Kronoseismology: Using density waves in Saturn’s C ring to probe the planet’s interior. Astronomy Journal, 146(12), 1–16.
  • Helled, R., Anderson, J. D., Podolak, M., & Schubert, G. (2011). Interior models of Uranus and Neptune. Astrophysical Journal, 726(1), 1–7.
  • Helled, R., Anderson, J. D., Schubert, G., & Stevenson, D. J. (2011). Jupiter’s moment of inertia: A possible determination by JUNO. Icarus, 216, 440.
  • Helled, R., Bodenheimer, P., Podolak, M., Boley, A., Meru, F., Nayakshin, S., . . . Boss, A. P. (2014). Giant planet formation, evolution, and internal structure. In H. Beuther, R. Klessen, C. Dullemond, & T. Henning (Eds.), Protostars and planets VI (pp. 643–665). Tuscon: University of Arizona Press.
  • Helled, R., Galanti, E., & Kaspi, Y. (2015). Saturn’s fast spin determined from its gravitational field and oblateness. Nature, 520, 202–204.
  • Helled, R., & Guillot, T. (2013). Interior models of Saturn: Including the uncertainties in shape and rotation. Astrophysical Journal, 767, 113.
  • Helled, R., & Guillot, T. (2018). Internal structure of giant and icy planets: Importance of heavy elements and mixing. In H. J. Deeg & J. A. Belmonte (Eds.), Handbook of exoplanets (pp. 167–185). New York, NY: Springer.
  • Helled, R., & Lunine, J. (2014). Measuring Jupiter’s water abundance by Juno: The link between interior and formation models. Monthly Notices of the Royal Astronomical Society, 441(3), 2273.
  • Helled, R., & Stevenson, D. (2017). The fuzziness of giant planets cores. Astrophysical Journal Letters, 840(1), 1–4.
  • Higgins, C. A., Carr, T. D., & Reyes, F. (1996). A new determination of Jupiter’s radio rotation period. Geophysical Research Letters, 23, 2653–2656.
  • Hubbard, W. B. (1968). Thermal structure of Jupiter. Astrophysical Journal, 152, 745.
  • Hubbard, W. B. (1999). Gravitational signature of Jupiter’s deep zonal flows. Icarus, 137, 357–359.
  • Hubbard, W. B. (2012). High-precision Maclaurin-based models of rotating liquid planets. Astrophysical Journal Letters, 756(1), L15.
  • Hubbard, W. B. (2013). Concentric Maclaurin spheroid models of rotating liquid planets. Astrophysical Journal, 768, 43.
  • Hubbard, W. B., & Militzer, B. (2016). A preliminary Jupiter model. Astrophysical Journal, 820, 80–93.
  • Hubbard, W. B, Schubert, G., Kong, D., & Zhang, K. (2014). On the convergence of the theory of figures. Icarus, 242, 138–141.
  • Iaroslavitz, E., & Podolak, M. (2007). Atmospheric mass deposition by captured planetesimals. Icarus, 187, 600.
  • Iess, L., Folkner, W. M., Durante, D., Parisi, M., Kaspi, Y., Galanti, E., . . . Bolton, S. J. (2018). Measurement of Jupiter’s asymmetric gravity field. Nature, 555, 220–222.
  • Ingersoll, A. P., & Pollard, D. (1982). Motion in the interiors and atmospheres of Jupiter and Saturn: Scale analysis, anelastic equations, barotropic stability criterion. Icarus, 52, 62–80.
  • Kaspi, Y. (2013). Inferring the depth of the zonal jets on Jupiter and Saturn from odd gravity harmonics. Geophysical Research Letters, 40, 676.
  • Kaspi, Y., Flierl, G. R., & Showman, A. P. (2009). The deep wind structure of the giant planets: Results from an anelastic general circulation model. Icarus, 202(2), 525–542.
  • Kaspi, Y, Galanti, E., Hubbard, W. B., Stevenson, D. J., Bolton, S. J., Iess, L., . . .Wahl, S. M. (2018). Jupiter’s atmospheric jet streams extend thousands of kilometers deep. Nature, 555, 223–226.
  • Kaspi, Y., Hubbard, W. B., Showman, A. P., & Flierl, G. R. (2010). Gravitational signature of Jupiters internal dynamics. Geophysical Research Letters, 37, L01204.
  • Knudson, M. D., Desjarlais, M. P., Lemke, R. W., Mattsson, T. R., French, M., Nettelmann, N., & Redmer, R. (2012). Probing the Interiors of the Ice Giants: Shock Compression of Water to 700 GPa and 3.8g/cm3. Physical Review Letters, 108(9), 091102
  • Knudson, M., Desjarlais, M. P., Becker, A., Lemke, R. W., Cochrane, K. R., Savage, M. E., . . . Redmer, R. (2015) Direct observation of an abrupt insulator-to-metal transition in dense liquid deuterium. Science, 348, 1455–1460.
  • Leconte, J., & Chabrier, G. (2012). A new vision on giant planet interiors: The impact of double diffusive convection. Astronomy and Astrophysics, 540, A20.
  • Leconte, J., & Chabrier, G. (2013). Layered convection as the origin of Saturn’s luminosity anomaly. Nature Geoscience, 6, 347.
  • Liu, J., Goldreich, P. M., & Stevenson, D. J. (2008). Constraints on deep-seated zonal winds inside Jupiter and Saturn. Icarus, 196(2), 653–664.
  • Lorenzen, W., Holst, B., & Redmer, R. (2009). Demixing of hydrogen and helium at megabar pressures. Physical Review Letters, 102(11), 115701.
  • Lorenzen, W., Holst, B., & Redmer, R. (2011). Metallization in hydrogen-helium mixtures. Physical Review B, 84(23), 235109.
  • Loubeyre, P., Letoullec, R. & Pinceaux, J. P. (1991). A new determination of the binary phase diagram of H2-He mixtures at 296 K. Journal of Physics: Condensed Matter, 3, 3183.
  • Lozovsky, M., Helled, R., Rosenberg, E. D., & Bodenheimer, P. (2017). Jupiter’s formation and its primordial internal structure. Astrophysical Journal, 836, Article 227.
  • Mankovich, C., Fortney, J. J., & Moore, K. L. (2016). Bayesian evolution models for Jupiter with helium rain and double-diffusive convection. Astrophysical Journal, 832, Article 113.
  • Mankovich, C., Marley, M. S., Fortney, J. J., & Movshovitz, N. (2018). Cassini ring seismology as a probe of Saturn’s interior I: Rigid rotation. Astrophysical Journal. Manuscript submitted for publication.
  • Mazzola, G., Helled, R., & Sorella, S. (2018). Phase diagram of hydrogen and a hydrogen-helium mixture at planetary conditions by quantum Monte Carlo simulations. Physical Review Letters, 120, 025701.
  • Miguel, Y., Guillot, T., & Fayon, L. (2016). Jupiter internal structure: The effect of different equations of state. Astronomy and Astrophysics, 596, A114.
  • Militzer, B., & Hubbard, W. B. (2013). Ab initio equation of state for hydrogen-helium mixtures with recalibration of the giant-planet mass-radius relation. Astrophysical Journal, 774, 148.
  • Militzer, B., Hubbard, W. B., Vorberger, J., Tamblyn, I., & Bonev, S. A. (2008). A massive core in Jupiter predicted from first-principles simulations. Astrophysical Journal Letters, 688, L45.
  • Militzer, B., Soubiran, F., Wahl, S. M., & Hubbard, W. (2016). Understanding Jupiter’s interior. Journal of Geophysical Research: Planets, 121, 1552–1572.
  • Mirouh, G. M., Garaud, P., Stellmach, S., Traxler, A. L., & Wood, T. S. (2012). A new model for mixing by double-diffusive convection (semi-convection): I. The conditions for layer formation. Astrophysical Journal, 750, 61.
  • Moore, K. M., Yadav, R. K., Kulowski, L., Cao, H., Bloxham, J., Connerney, J. E. P., . . . Levin, S. M. (2018). A complex dynamo inferred from the hemispheric dichotomy of Jupiter’s magnetic field. Nature, 561, 76–78.
  • Morales, M. A., Hamel, S., Caspersen, K., & Schwegler, E. (2013). Hydrogen-helium demixing from first principles: From diamond anvil cells to planetary interiors. Physical Review B, 87, 174105.
  • Morales, M. A., Schwegler, E., Ceperley, D., Pierleoni, C., Hamel, S., & Caspersen, K. (2009). Phase separation in hydrogen–helium mixtures at Mbar pressures. Proceedings of the National Academy of Sciences USA, 106, 1324–1329.
  • NASA. (2019a). NASA Jet Propulsion Laboratory. California Institute of Technology.
  • Nettelmann, N. (2017). Low- and high-order gravitational harmonics of rigidly rotating Jupiter. Astronomy & Astrophysics, 606, 9.
  • Nettelmann, N., Becker, A., Holst, B., & Redmer, R. (2012). Jupiter models with improved ab initio hydrogen equation of state (H-REOS.2). Astrophysical Journal, 750, 52.
  • Nettelmann, N., Fortney, J. J., Moore, K., & Mankovich, C. (2015). An exploration of double diffusive convection in Jupiter as a result of hydrogen–helium phase separation. Monthly Notices of the Royal Astronomical Society, 447, 3422.
  • Nettelmann, N., Helled, R., Fortney, J. J., & Redmer, R. (2012). New indication for a dichotomy in the interior structure of Uranus and Neptune from the application of modified shape and rotation data. Planetary and Space Science, 77, 143–151.
  • Nettelmann, N., Holst, B., Kietzmann, A., French, M., Redmer, R., & Blaschke, D. (2008). Ab initio equation of state data for hydrogen, helium, and water and the internal structure of Jupiter. Astrophysical Journal, 683, 1217.
  • Nettelmann, N., Puttelm, R., & Redmer, R. (2013). Saturn layered structure and homogeneous evolution models with different EOSs. Icarus, 225, 548.
  • Ohta, K., Ichimaru, K., Einaga, M., Kawaguchi, S., Shimizu, K., Matsuoka, T., Hirao, N., & Ohishi, Y. (2015). Phase boundary of hot dense fluid hydrogen. Scientific reports, 5, 16560
  • Pollack, J. B., Hubickyj, O., Bodenheimer, P., Lissauer, J. J., Podolak, M., & Greenzweig, Y. (1996). Formation of the giant planets by concurrent accretion of solids and gas. Icarus, 124, 62.
  • Püstow, R., Nettelmann, N., Lorenzen, W., & Redmer, R. (2016). H/He demixing and the cooling behavior of Saturn. Icarus, 267, 323.
  • Read, P. L., Dowling, T. E., & Schubert, G. (2009). Saturn’s rotation period from its atmospheric planetary-wave configuration. Nature, 460, 608–610.
  • Riddle, A. C., & Warwick, J. W. (1976). Redefinition of system III longitude. Icarus, 27, 457–459.
  • Roberts, P. H., & King, E. (2013). On the genesis of the Earth’s magnetism. Reports on Progress in Physics, 76, 096801.
  • Rosenblum, E., Garaud, P., Traxler, A., & Stellmach, S. (2011). Erratum: “Turbulent mixing and layer formation in double-diffusive convection: Three-dimensional numerical simulations and theory.” Astrophysical Journal, 742, 132.
  • Roulston, M. S., & Stevenson, D. J. (1995). Prediction of neon depletion in Jupiter’s atmosphere. EOS Transactions, 76, 343 (abstract).
  • Salpeter, E. E. (1973). On convection and gravitational layering in Jupiter and in stars of low mass. Astrophysical Journal Letters, 181, L83.
  • Saumon, D., Chabrier, G., & van Horn, H. M., (1995). An equation of state for low-mass stars and giant planets. Astrophysical Journal Supplement, 99, 713.
  • Saumon, D., & Guillot, T. (2004). Shock compression of deuterium and the interiors of Jupiter and Saturn. Astrophysical Journal, 609, 1170.
  • Schöttler, M., & Redmer, R. (2018). Ab initio calculation of the miscibility diagram for hydrogen–helium mixtures. Physical Review Letters, 120(11), 115703.
  • Schouten, J. A., de Kuijper, A., & Michels, J. P. J. (1991). Critical line of He-H2 up to 2500 K and the influence of attraction on fluid-fluid separation. Physical Review B, 44, 6630.
  • Soubiran, F., & Militzer, B. (2016). The Properties of Heavy Elements in Giant Planet Envelopes. The Astrophysical Journal, 829(1), 12.
  • Spilker, L. J. (2012). Cassini: Science highlights from the equinox and solstice missions. Lunar and Planetary Institute Science Conference Abstracts, 43, 1358.
  • Stevenson, J. (1975). Thermodynamics and phase separation of dense fully ionized hydrogen-helium fluid mixtures. Physical Review B, 12(10), 3999–4007.
  • Stevenson, D. J. (1982). Formation of the giant planets. Planetary and Space Science, 30, 755–764.
  • Stevenson, D. J., & Salpeter E. E. (1977a). The dynamics and helium distribution in hydrogen–helium fluid planets. Astrophysical Journal Supplement, 35, 239.
  • Stevenson, D. J., & Salpeter, E. E. (1977b). The phase diagram and transport properties for hydrogen– helium fluid planets. Astrophysical Journal Supplement, 35, 221.
  • Stevenson, D. J. (1985). Cosmochemistry and structure of the giant planets and their satellites. Icarus, 62, 4–15.
  • Vazan, A., Helled, R., & Guillot, T. (2018). Jupiter’s evolution with primordial composition gradients. Astronomy and Astrophysics, 610, L14.
  • Vazan, A., Helled, R., Kovetz, A., & Podolak, M. (2015). Convection and mixing in giant planet evolution. Astrophysical Journal, 803, 32.
  • Vazan, A., Helled, R., Podolak, M., & Kovetz, A. (2016). The evolution and internal structure of Jupiter and Saturn with compositional gradients. Astrophysical Journal, 829, 118.
  • Venturini, J., Alibert, Y., & Benz, W. (2016). Planet formation with envelope enrichment: New insights on planetary diversity. Astronomy and Astrophysics, 596, Article 90.
  • von Zahn, U., Hunten, D. M., & Lehmacher, G. (1998). Helium in Jupiter’s atmosphere: Results from the Galileo probe helium interferometer experiment. Journal of Geophysical Research: Planets, 103, 22815.
  • Wahl, S. M., Hubbard, W. B., Militzer, B., Guillot, T., Miguel, Y., Kaspi, Y., . . . Bolton, S. J. (2017). Comparing Jupiter interior structure models to Juno gravity measurements and the role of a dilute core. Geophysical Research Letters, 44, 4649–4659.
  • Wilson, H. F., & Militzer, B. (2010). Sequestration of noble gases in giant planet interiors. Physical Review Letters, 104, 121101.
  • Wilson, H. F., & Militzer, B. (2012). Solubility of water ice in metallic hydrogen: Consequences for core erosion in gas giant planets. Astrophysical Journal, 745, 54.
  • Wisdom, J., & Hubbard, W. B. (2016). Differential rotation in Jupiter: A comparison of methods. Icarus, 267, 315.
  • Wood, T. S., Garaud, P., & Stellmach, S. (2013). A new model for mixing by double-diffusive convection (semi-convection). II. The transport of heat and composition through layers. Astrophysical Journal, 768, 157.
  • Zaghoo, M., Salamat, A., &. Silvera, I. F. (2016). Evidence of a first-order phase transition to metallic hydrogen, Physical Review B, 93, 155128.
  • Zharkov, V. N., Hubbard, W. B., & Trubitsyn, V. P. (1978). Physics of planetary interiors. Tucson, AZ: Pachart.


  • 1. 1 AU is an astronomical unit, the average distance between the Earth and the Sun.

  • 2. In adiabatic models, the temperature profile is set by the adiabatic gradient, and the entropy is (nearly) constant within the planet (see Militzer et al., 2016 and references therein for details).

  • 3. However, this is a simplification as odd harmonics have been measured for Jupiter with the Juno spacecraft (Iess et al., 2018). The measurement of Jupiter’s gravitational field being north–south asymmetric, has been used to reveal the planets atmospheric and interior flows (Kaspi et al., 2018).

  • 4. The adiabatic gradient ad=lnTlnP|s, where S is the entropy, corresponds to a case in which the material is homogenous and convective. The radiative/conductive gradient is given by rad/cond=3kLP64πσT4Gm, where κ is the Rosseland opacity which accounts for contributions from both radiation and conduction, and σ is the Stephan–Boltzmann constant. See Guillot et al. (2004), Militzer et al. (2016), and references therein for further details.

  • 5. It should also be noted that the nature of the transition of hydrogen from “molecular” to “metallic” along Jupiter’s and Saturn’s adiabats is still debated. The transition could be a “first-order” one or a smooth one, although most studies imply that in Jupiter’s and Saturn’s interiors the transition is smooth and is a first-order transition at lower (intermediate) temperatures.

  • 6. This would depend on the thermodynamic behavior of the H–HE mixture in the presence of a phase separation.