Voyager 1 and 2 performed the first unambiguous low-energy (E ≥ 30 keV) ion measurements in and around the Jovian magnetosphere in 1979. The magnetosphere contains a hot (kT ~ 30 keV), multicomponent (H, He, O, S) ion population dominated by convective flows in the corotation direction out to the dayside magnetopause and on the nightside to ~ 130–150 Rj beyond which the ion flow direction changes to predominantly antisolar, but with a strong component radially outward from Jupiter. This tailward flow of hot plasma, the magnetospheric wind, accounts for the loss of ~ 2 × 1027 ions/s and ~ 2 × 1013 W from the magnetosphere. Comparison of energetic (≥ 30 keV) ion to magnetic field pressure reveals that particle and magnetic pressures are comparable from the magnetopause inward to at least ~ 10 Rj, that is, magnetosphere dynamics is determined by pressure variations in a high-β plasma. This particle pressure is responsible for inflation of the magnetosphere and it (rather than the planetary magnetic field) determines the standoff distance with the solar wind. The ion spectrum can be described by a convected Maxwellian component at E ≤ 200 keV, and a nonthermal tail at higher energies described by a power law of the form E−γ. New theoretical techniques were developed in order to interpret the low-energy solid-state detector measurements of temperature, number densities, pressures, and flow velocities in this novel hot-plasma environment.

The magnetosphere of Jupiter is unique in the solar system because of its large extent and rapid rotation, and because of the prodigious source of plasma provided by the satellite Io. Io and its associated neutral clouds inject > 1029 amu/s of freshly ionized material into the magnetosphere, producing a plasma torus with a density maximum near the L-shell of Io and a total mass of ~ 1036 amu. The innermost region of this torus contains a cool plasma corotating with the planet and dominated by S+ ions with temperatures of a few eV. At greater distances, beyond 5.6 Rj, the plasma ions are warmer, consisting primarily of sulfur and oxygen ions with temperatures of ~ 40 eV. The plasma electrons here have mean energies of 10 to 40 eV, and exhibit distribution functions which are non-Maxwellian, with both a thermal and suprathermal component. Near Io itself, the Alfvén wave generated by Io gives rise to observed perturbations in the magnetospheric velocities as the ambient plasma flows around the Io flux tube. In the middle magnetosphere, between ~ 8 and ~ 40 Rj, the ions and electrons tend to be concentrated in a plasma disc or sheet that is routinely cooler than its higher latitude surroundings. This plasma tends to move azimuthally but does not rigidly corotate with the planet. The electron density enhancements at the plasma sheet are due primarily to an increase in the electron thermal population with little change in the suprathermal population.

The radio spectrum of Jupiter spanning the frequency range from below 10 kHz to above 3 GHz is dominated by strong nonthermal radiation generated in the planet's inner magnetosphere and probably upper ionosphere. At frequencies above about 100 MHz, a continuous component of emission is generated by synchrotron radiation from trapped electrons between equatorial distances of about 1.3 and 3 Rj. This component exhibits a broad spectral peak at decimetric (DIM) wavelengths, distinct longitudinal asymmetries arising from asymmetries in Jupiter's magnetic field, and slow intensity variations that are presumably related to temporal changes in the energy, pitch angle, or spatial distributions of the radiating electrons. High resolution mapping of this component will probably continue to provide detailed information on the inner magnetosphere structure that is presently unobtainable by other means. Jupiter's most intense radio emissions occur in the frequency range between a few tenths of a MHz and 39.5 MHz. This decameter-wavelength (DAM) component is characterized by complex, highly organized structure in the frequency-time domain and by a strong dependence on the longitude of the observer and in some cases, of Io. The DAM component is thought to be generated near the electron cyclotron frequency in and above the ionosphere on magnetic field lines that thread the Io plasma torus, but neither the specific location(s) of the radio source(s) nor the specific plasma emission process are firmly established. At frequencies below about 1 MHz there exist two independent components of emission that have spectral peaks at kilometer (KOM) wavelengths. One is bursty, relatively broadbanded (typically covering 10 to 1000 kHz), and strongly modulated by planetary rotation.

Introduction

The dynamical properties of a planet depend on the way in which the density varies with radius, and seismological properties depend also on the way in which the elastic moduli and elastic dissipation vary with radius. The data we have for the Earth are sufficiently complete that the variations of density and elastic moduli with radius can be derived from them and we are then presented with the problem of inferring the mineralogical and chemical composition consistent with them. When, as for the other planets, seismic data are lacking, we must proceed in a different way and derive the variation of density from a postulated composition, asking if it leads to the observed mass and moment of inertia. In either case, we must know how the density depends on pressure, temperature and composition, for all these vary with radius, and when we discuss the Earth and the Moon, for which we have seismic data, we must also examine the dependence of the elastic moduli upon the three variables. Some idea of the problems that arise, of the theoretical principles, of possible experimental methods, and of the systematics of equations of state of minerals has already been given in Chapter 1, and it is the aim of this chapter to give a more extensive and systematic account.

The wonders of the heavens

The planets have been a subject of wonder to man from earliest recorded times. Their very name, the Wandering Ones, recalls the fact that their apparent positions in the sky change continually, in contrast to the fixed stars. Greek astronomers, Ptolemy particularly, had shown how the motions of the planets, the Sun and the Moon could be accounted for if they were all supposed to move around a stationary Earth, and in mediaeval times an elaborate cosmology was created, at its most allegorical, evocative and poetic in the Paradiso of Dante. The men of the Renaissance overthrew these ideas but provided fresh cause for wonder in their place. Placed in motion around the Sun by Copernicus, their paths observed with care by Kepler, the planets led Newton to his ideas of universal gravitation. Galileo, his telescope to his eye, showed that they had discs of definite size and that Jupiter had moons, the Medicean satellites, which formed a system like the planets themselves.

The discoveries of the seventeenth century settled notions of the planets for three centuries, but within that framework a most extra-ordinary flowering of the intellect attended the working out of the ideas of Newton. Closer and closer observation showed ever more intricate departures of the paths of the planets from the simple ellipses of Kepler, and each was accounted for by ever subtler applications of mechanics as the consequence of the gravitational pull of each planet upon its fellows.

Introduction

The models of the planets which have been adopted so far depend explicitly on the assumption that the planet is in the hydrostatic state, so that the density is a function only of radial distance (in a generalized sense when the planet is flattened by spin). That may be an appropriate first assumption to provide a starting point for further developments, but it is clearly not adequate: the gravity fields of the Earth, the Moon and Mars contain harmonic components that would be absent if the internal state were hydrostatic; the irregular surface features of the terrestrial planets are inconsistent with strict hydrostatic equilibrium; and the structure seen in the atmosphere of Jupiter reveals internal motions, if only superficial. A density distribution not in hydrostatic equilibrium requires a stress system to support it that departs from the simple normal pressure to which hydrostatic equilibrium corresponds. Such a stress system may be developed in two ways: statically, through strains of the planet, or dynamically, through movements of the material. According to which mode is effective, so the planet may be considered to be cold or hot (though, as has already been argued, no planet is hot in relation to the effect on the equation of state). If the planet is cold, the materials within it have high strengths, can support large stresses and so maintain statically non-hydrostatic distributions of density. If the planet is hot, then parts will be molten, as is the core of the Earth, or will be sufficiently hot to creep steadily under applied stress.

Introduction

Mars, Venus and Mercury form with the Earth and the Moon a group of rather similar bodies. By comparison with the giant planets on the one hand and the small satellites on the other, the sizes lie in a relatively restricted range, while the mean densities are higher than those of most other bodies in the solar system. It is natural to think that their compositions are similar and that the structures of Mars, Venus and Mercury might be inferred from what is known of the Earth and the Moon.

Seismological data are, of course, not available for any of the planets other than the Earth, so that the structures of the terrestrial planets must be derived from the dynamical data, together with such inferences as may be drawn from the magnetic and electrical properties, together with analogies with the Earth and the Moon.

Unfortunately, the dynamical data themselves are less informative for Mars, Venus and Mercury than they are for the Earth and the Moon or for the major planets. The solar precession of Mars has not so far been observed and, in consequence, the moment of inertia cannot be derived from the value of J2 without making the assumption of hydrostatic equilibrium. Yet it is clear that Mars is not in hydrostatic equilibrium. The theory of the errors likely to be committed by making the assumption of hydrostatic equilibrium was given in Chapter 3 and subsequently in this chapter (section 6.6) it will be applied to Mars.

Introduction

The possibility that, at the pressures encountered in the planets, materials ordinarily non-metallic at low pressures might transform into metals has been discussed for more than forty years. Two main ideas have been considered: one, that metal silicates, such as olivine, might become metallic at pressures developed in the core of the Earth, and the other, that hydrogen, helium and other light elements might transform to metals at pressures encountered in the major planets. Sufficient is now known about changes of density in metallic transformations under high pressure to be sure that the jump of density between the mantle and the core of the Earth is too great to be explained by such a transformation and in the preceding chapters on the terrestrial planets it has been assumed that the difference between the core and mantle of the Earth is one of composition (see also Anderson, 1977). It is otherwise with Jupiter and Saturn. The mean densities of those planets are too low for them to be composed of anything but hydrogen, helium and other materials of low atomic number, and the likelihood of a metallic transformation of hydrogen in particular is crucial to a discussion of their internal structures. One of the first studies of the metallic transformation in hydrogen (Kronig, de Boer and Korringa, 1946) was prompted by the idea of Kuhn and Rittman (1941) that the inability of the core of the Earth to support shear waves might be because it was of solar composition, that is, mainly of hydrogen, and by the subsequent suggestion of van der Waals that at core pressures the hydrogen might be metallic.

The lesser objects of the solar system

So far in our studies, no notice has been taken of the smaller bodies in the solar system, i.e. Pluto, the asteroids and the satellites of Mars and the major planets, for their properties are but poorly known on the whole and it is not very rewarding to apply to them the type of analysis that was applied in the foregoing parts of this book to the greater objects. Yet, in considering the solar system as a whole, their existence and such information as we have of them cannot be ignored.

Pluto, the outermost known planet, is in a highly eccentric orbit highly inclined to the ecliptic, and, in consequence, although it comes on occasion within the orbit of Neptune, it never approaches Neptune closely, and detailed studies have shown the outer solar system to be stable. Pluto is a very small object as seen from the Earth. Its diameter is estimated from the brightness and supposed reflectivity. The latter has recently been redetermined from infra-red spectroscopy and the diameter of Pluto is consequently now estimated to lie between 2800 and 3300 km (Cruickshank, Pilcher and Morrison, 1976). The mass of Pluto was originally estimated from the perturbations of the orbits of Uranus and Neptune, but a satellite has now been detected (Christy and Harrington, 1978) with a period of 6.4 d, from which the mass of Pluto is estimated to be about 0.002 times that of the Earth (Meadows, 1980).

Introduction

The only observed mechanical data we have for any planet are the mass and moment of inertia, and infinite sets of models can be constructed consistent with such pairs of data. The sets of models are not, however, unbounded, and, further, certain models are in some sense more probable than others. It is the purpose of this appendix to set out the bounds on two particular models and to give some most probable models. The models considered are: that of two zones, each of constant density, and that in which the density is determined by hydrostatic compression alone. The terrestrial planets may be modelled by the former, and the major planets by the latter. Neither model can represent the complexities of actual planets but, given only two data, no more elaborate model is justified. Guided by the constitution of the Earth, and by such seismic data as are available for the Moon, it is natural to choose the two-zone model as an approximation to the structures of the terrestrial planets. In this model, the maximum pressure is such that changes of density under self-compression are less than differences of density arising from differences of chemical composition or crystal structure in different parts of the planet. Thus, a model comprising two zones of different density is chosen as a basis for study of the terrestrial planets.