C365 : Physics of Planetary Interiors Essay

The Internal Structure of Ganymede and Callisto
 

By: Peter Grindrod
       MSci Planetary Science III

For: Dr. L.Vocadlo

Go to:
 Introduction
 Internal Structure of Ganymede
 Internal Structure of Callisto
 Conclusion
 References

Introduction

Since the surfaces of these moons were seen by the Voyager probes in 1979, much speculation has occurred as to the origin of the apparent dichotomy between the features observed on the surfaces on these two large moons. To a certain extent, this contrast has been explained by different internal structures and evolution, however, in the course of this essay, the focus will be on the internal structures of these Galilean satellites, and the differing surfaces only mentioned as a consequence of (and evidence for) this.

Ganymede and Callisto are the third and fourth of Jupiter's four large moons, respectively, known as the Galilean satellites (after their discoverer, Galileo Galilei). Despite their similar sizes and densities [Schenk, 1995], they have contrasting surfaces and therefore possibly differing interiors and/or evolutions. They have been observed by the Voyager spacecraft in 1979 and Galileo in 1996 to present, and much of the information in this essay originates from data received from these probes.

The apparent dichotomy between the surfaces arises from the fact that Callisto has a very old, heavily cratered surface, whereas the surface of Ganymede varies between old and cratered to younger grooved terrain [Schubert et al. 1981]. Table 1 compares the properties of Ganymede and Callisto, and illustrates the fact that the main differences can only be seen when studying the surface features.

The Internal Structure of Ganymede

The density of Ganymede shown above is consistent with a body that has a high proportion of ice present; the moon is nearly twice as dense, which is thought to consist of 95% silicates [Schenk, 1995]. Therefore, with the definite identification of water ice on the surface of Ganymede [Pollack et al. 1978, Clark et al. 1986, Calvin et al. 1995], the main question asked is to what extent is the interior of Ganymede differentiated? The answers range from uniform ice-rock mixtures (completely undifferentiated) to rock cores surrounded by ice mantles (fully differentiated) to states somewhere in between. Although it is widely held that all the Galilean satellites have undergone differentiation to some degree, there is no conclusive evidence at present that allows the precise extent of differentiation to be established [Schubert et al. 1994].

Ganymede would have originally been formed from planetesimals that were composed of an ice/rock mixture [Stevenson et al. 1986]; differentiation then occurs when the body is (radioactive) heated, thus melting the ice and forcing it to rise as it is more buoyant than the remaining rock, which sinks towards the centre of the early moon. Therefore a "three layer model" can be expected, which consists of a silicate core, an inner mantle of mixed ice and rock and an outer mantle/crust of refrozen ice [Mueller & McKinnon, 1988]. Thus the final internal structure definition depends on the composition and proportion of the rock and ice constituents.

Mueller & McKinnon (1988) investigated the validity of three different rock types for the interior of Ganymede: CI carbonaceous chondrite, "Prinn-Fegley" assemblage (P/F) and Pretemolite Condensate (PTC). The P/F rock is an assemblage predicted by Prinn & Fegley (1981) to have condensed from the proto-Jovian nebula, and the PTC rock is an anhydrous assemblage. These rocks cover a range of water content from high to low, and, in order of decreasing water content, are CI, P/F then PTC. Their mineral content is shown in table 3. Mueller & McKinnon then calculated the likelihood of each rock being in a certain layer by using the whole-rock equation of state determined from the component mineral properties:

	(g cm-3)	(10-5 K-1)	(10-11 Pa-1)
CI rock	2.766	6.953	1.878
P/F rock	3.262	5.997	1.479
PTC rock	3.756	3.507	0.862

They concluded that CI rock is plausible as a lower mantle constituent, but not as a core component, and therefore the core should be made of either P/F or PTC rock. CI chondrite material is assumed to be the rocky constituent in the rock/ice mixture that made up the planetesimals and subsequently formed Ganymede; this therefore would be the obvious choice for any rocky layer within the interior. However, Mueller & McKinnon (1988) state that CI rock is too "wet" to be a good analogue for primordial planetary rock in the Jupiter region, and anhydrous chondrite is only useful as an end-member mineral assemblage. Therefore the above rock types were chosen for a three layer model, leaving only the top layer as consisting of nearly pure water ice (and no rock constituent), which has been proved by spectroscopic methods (Calvin & Clark, 1993).

Mueller & McKinnon narrowed down the range for the value of the important parameter that is the state of differentiation: they calculated a value for the fraction of rock as between 0.54 – 0.59, rather than the previous (and more ambiguous) value of 0.49 – 0.64 [Schubert et al. 1986].

After limiting (but, unfortunately, not fixing) a value for the degree of differentiation, and predicting a number of possible rock types for the core and lower mantle, the final step in describing the interior of Ganymede is to study the phase changes of ice according to their depth, and therefore temperature and pressure. Figure 1 shows the phase diagram for ice. Kargel (1991) pointed out that pressures at the core/mantle boundary would be sufficient for the phase ice VI to exist (~ 12-13 kbars [i.e. ~ 1.2 GPa]) and also that the mantle and crust would subsequently be layered with different polymorphs of ice. Durham et al. (1983) give a possible temperature profile for Ganymede and Callisto and this has been plotted on figure 1. This profile, if correct, shows that the general trend of phase change with increasing depth would be ice I_H, ice II, ice IV and finally ice VI. These different ice phases define how ice at different pressures and temperatures behave, according to the crystalline structure of the H₂O molecules (for example, ice I_H is hexagonal as seen by six-cornered snow flakes on Earth).

 Durham et al. (1983) found that the ice at the bottom of the I_H layer is slightly stronger than expected, but the ice near the surface is considerably softer than previously predicted. It is this soft layer therefore that water (or ice via cryovolcanism) must have broken through to form the light terrain areas that are widely accepted to be younger than the dark areas.
 
 

The Internal Structure of Callisto

The lack of resurfacing on Callisto is thought by some to represent a body that has undergone no differentiation. Instead of a surface complicated by young, grooved terrain, Callisto has a very old surface that has been dominated by heavy cratering. Schubert et al. (1981) claimed that the only conclusion they could reach for the vastly differing surfaces between Ganymede and Callisto was that the latter was completely undifferentiated (see figure 2). This was inferred by the lack of a bright terrain intruding the dark, heavily-cratered material (i.e. Callisto had undergone no resurfacing due to internal differentiation, as thought to have happened on Ganymede).
 

However Malhotra (1991) proposed a theory that if true, is quite intriguing: that is that Ganymede's light terrain is not the result of differentiation from accretional heat, but represents a later resurfacing period caused by the heat created from the chaotic evolution of Ganymede's eccentricity. Schenk (1995) links this fact with an analogy between crater morphologies and dimensions on Callisto and Ganymede to propose a now popular theory that Callisto infact represents a baseline for what Ganymede once looked like, and would still resemble if it wasn't for its (now considered) unusual history.

Therefore the interior of Callisto could now be looked upon as almost identical to that of Ganymede, although it could be slightly less differentiated as a result of Ganymede's extra heating period late in its history. Thus Callisto could also be assumed to be made of a three layer system: the top layer (crust) being layered ice phases, the middle layer (lower mantle) being made up of polymorphs of ice with silicate rock (possibly CI rock) with a silicate mantle (possibly P/F or PTC rock).

Schenk (1995) states the outermost 10 km or so of Callisto appear to be as differentiated as Ganymede, from crater and palimpset studies. Another possibility, given by Osrto et al. (1992), is that Callisto has a crust that is actually more differentiated than Ganymede. This was calculated as a possibility to explain the lower radar reflectivity (by a factor of 2) that Callisto has compared to Ganymede. Ice is still the main crustal constituent despite this possible rockier regolith.
 
 

Conclusion

After studying various possibilities for the interiors of Ganymede and Callisto, contemporary models for their structures can determined.

Ganymede can be assumed to be a differentiated body that has evolved into three layers: the icy crust, ice/rock mantle and (silicate) rocky core. It is thought to have undergone differentiation, which formed the three layers, after a period of heavy bombardment. It then underwent a second period of differentiation due to its eccentric orbit causing internal heating; this heating caused some ice to melt and rise to the surface to form the young, grooved terrain that, due to its light appearance, reflects its icy nature. The outer crust is probably made of nearly completely water ice that is in the phase ice I_H at the surface, and gradually changes phase according to the (presently unknown) pressure/temperature profile with depth. This crust could be as thick as ~ 300 km [Kargel, 1991]. The rock/ice mantle is ~ 500-600 km thick [Kargel, 1991] and is made of pressure/temperature dependant ice phases and silicate rock. This rock could resemble a CI chondrite and therefore would be rich in MgSO₄ and Na₂SO₄ (~ 15-20% by mass [Kargel, 1991]). This leaves a core ~ 1600 km thick which could also be quite hydrous (P/F rock) or contain higher levels of troilite ([FeS] which is the ideal pyrrhotite structure and is found in meteorites) and enstatite (16[MgSiO₃]). The iron in the troilite could contribute towards the magnetic field that has just been discovered by the Galileo spacecraft, thus strengthening the case for the presence of this mineral, but in a wider context, the presence of a core and the past process of differentiation on Ganymede.

The structure of Callisto could reflect that of Ganymede for its initial formation and differentiation, but appears not to have undergone the second (shorter) period of differentiation (as inferred from the lack of recent/if any resurfacing features). Therefore it could also be assumed to have a three-layer model composed of similar ice/ rock constituents as Ganymede; it will however have a slightly less differentiated interior due to the lack of a second heating phase. Therefore the ice crust could be smaller than that of Ganymede (as differentiation normally leads to an increase in radius), and therefore could have slightly different ice phases present due to the differing pressures and temperatures. The internal structure could however, according to some scientists, consist of one layer of completely undifferentiated material (an ice and rock mixture) and therefore resemble figure 2 and not figure 3.
 

 References