Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 Interiors and Magnetic Fields INTRODUCTION There are many areas of interest to planetary astronomers for which our knowl- edge is fragmentary, either because extensive studies have not yet been carried out, or because we are limited in our ability to investigate certain problems from the surface of the Earth. Two such areas are planetary interiors and planetary magnetic fields. A knowledge of the structure and composition of planetary interiors is essential to a better understanding of the present state of the planetary system, how it was formed, and the manner in which it evolved. The only method that has so far been applied to observe planetary magnetic fields from the Earth's surface is to detect radiation from charged particles trapped in the field. Thus far, only Jupiter has displayed such nonthermal emission. The study of this emission permits some understanding of the nature of the Jovian magnetosphere. This is of interest both for possible clues to the activities occurring within the planet and for comparison with the Earth's magnetosphere, which may aid in understanding the processes underlying both. PLANETARY INTERIORS The goal of the study of planetary interiors, ideally, would be to produce a table (or analytic expression) including such parameters as density, pressure, 39
40 PLANETARY ASTRONOMY temperature, composition, and magnetic field at every point within each planet. Such a wealth of detail, however, is not possible nor even desirable on any reasonable scale of scientific priorities. Nonetheless, certain averages which could in principle be extracted from such ideal tables are of great interest, for from them could be obtained the answer to such questions as: Did a given planet originate by accretion of cold, solid objects or by contracting clouds of gas and dust? What are the gross mole fractions of the elements in extrasolar material in the present solar system? How will the radiative temperature of the planets vary with time ? Would formation of a system of planets be a common or rare phenomenon in the Universe ? Since the data in such hypothetical tables cannot be obtained directly by measurements, they must be obtained from calculations using measurable parameters and theory. In principle, this is not a difficult process. A system of equations can be written, usually assuming hydrostatic equilibrium, that de- fines the equation of state, guarantees the conservation of mass and energy, and defines the mode of energy transport and the distribution of composition. Then, using the mass, radius, and measurable moments of inertia as boundary conditions, the set of differential equations can be integrated to produce a model of the planet. There are, however, a number of problems in formulating the detailed equations, problems that stem from our very incomplete under- standing of the behavior of matter under high pressure. One obstacle to model building is our limited knowledge of the equation of state. Experimental investigations using static compression and shock compres- sion have, in recent years, yielded much information on the behavior of matter under high pressure. Experiments alone, however, cannot cover the range of situations that are of interest to planetary physicists, and advances in theory have not been nearly so impressive as experimental advances. Some theoretical studies of certain elements and compounds in cold, solid phases should now be made, despite their complexity. In addition, whereas it is possible to estimate for a given compound or mixture the variation of density with pressure and temperature, one cannot at the present time make any a priori prediction of phase boundaries. In the case of the Earth, some boundaries can be predicted when experimental data on chemically similar isomorphic structures are avail- able, but, for other planetary interiors, the questions of phase boundaries are likely to remain enigmatic for a long period. A second area of difficulty involves thermal transport. Energy in planetary interiors may be transported by ordinary thermal conduction, convection, or radiation. How much each of these transport mechanisms contributes to the energy budgets of planetary interiors is not precisely known. The thermal con- ductivity of planetary material is very difficult to estimate, if only because such conductivities are extremely sensitive to composition, lattice structure (or lack
INTERIORS AND MAGNETIC FIELDS 41 of it), temperature, density. Convection is well understood in gases and certain liquids when nonturbulent; but even here any slight complexity in geometry raises severe mathematical difficulties. Convection of very imperfect gases, or convection in condensed systems exhibiting phase changes coupled with the presence of gyroscopic forces (magnetic or Coriolis), presents formidable difficulties indeed. Finally, radiative transport is presumably of importance in planetary interiors. With planets, however, even if the hypothesis of local thermodynamic equilibrium suffices, one cannot use the simplified version of Kirchhoff's law or even its more general expression which includes the square of the refractive index. This formula holds only for weakly absorbing media, because matter cannot have refractive indices which differ significantly from unity in some parts of the spectrum without having nontrivial absorption also present. Plasma physi- cists are faced with similar problems, and it seem likely that fundamental progress can be made in the next decade. Because so little of the fundamental physics is well understood, and even when understood, is so sensitive to the value of the parameters involved, many workers have used the theory of the Earth as an aid in developing models of the terrestrial planets. The density of the Earth is fairly well determined as a function of depth, probably to within 5 percent over most of its volume. The distribution of its density, however, already includes assumptions about its composition. The general picture of a predominantly iron core and a silicate mantle seems established, but to attribute this composition to other planets is questionable. Some models of Venus have been based on the planet's similarity to the Earth; one should bear in mind, however, that although Venus has almost the same mass and radius as the Earth, its atmospheric composition, for example, is very different. Until the reasons for the major differences are clearer, it may be premature to assume that Venus and the Earth have similar interior compositions. A like attitude toward those inner planets which resemble the Earth even less, Mercury and Mars, is obviously prudent. Objections are also raised against planetary models that assume abundances and structure based upon meteoritic samples, since evidence is rapidly accumu- lating that iron meteorites do not originate from the core of a sizable (>200 km) planet. For example, the irons were not surrounded by pallasites, since pallasite meteorites cooled ten times more slowly than most iron meteorites; and the assumption of chondritic composition of the Earth's mantle can be criticized. Even if meteorites originated on a parent planet, most of them (the chondrites and 80 percent of the observed falls) were formed near the surface. Although available information is normally not sufficient to determine a planet's composition, Jupiter and Saturn are exceptions. In these cases, simple
42 PLANETARY ASTRONOMY models based on the planetary radius, mass, moments of inertia, and the as- sumption of hydrostatic equilibrium yield such low densities that either the planets are composed largely of hydrogen or their internal temperatures are so high that thermal pressure is a substantial fraction of the total pressure. If such high thermal pressures were to exist, the planets would be extremely unstable to convection and would radiate tremendous quantities of heat. Such radiation is contrary to observation. Since simple models indicate that 80 percent of the mass of Jupiter and 75 percent of the mass of Saturn are hydrogen, it follows that at least 70 per- cent of the extrasolar material in the solar system is hydrogen. These numbers are quite insensitive to anything but the assumed equation of state of hydrogen and the assumption that these planets cannot be stable against convection if the thermal pressure is high. Unfortunately, the remaining constituents of Jupiter and Saturn cannot be identified so simply. Here one is forced to resort to such arguments as: Any object containing so much hydrogen must have essentially cosmic abundances of the elements, and consequently most of the remaining matter is helium. Whereas that is a reasonable assumption, it must for the present remain merely an assumption. The foregoing may perhaps best be summarized by stressing that the theory of planetary interiors is semiempirical. The models of planetary structure are far less convincing from the standpoint of physical theory than are the models of stellar structure. For the present we lack a detailed knowledge of the compo- sition of any planet; and our theoretical understanding of such factors as the energy transport mechanisms and phase changes is extremely limited. Until fundamental theory is significantly advanced, there is no alternative to working under these unsatisfying conditions; and these conditions must be kept in mind before one places too much credence on planetary models. PLANETARY MAGNETIC FIELDS AND MAGNETOSPHERES Planetary magnetic fields are of interest because of the plasma phenomena produced by the interaction of the fields with the solar wind, and because, being produced by processes occurring deep within the planet, they provide a clue to the nature and extent of those processes. In principle there are four kinds of observational evidence from which the existence of a planetary mag- netic field may be inferred: space-probe magnetometer measurements; non- thermal electromagnetic radiation from the planet of appropriate frequency, polarization, and intensity; evidence that in modulating the solar wind the effective planetary cross section greatly exceeds the cross section of the visible planet; and Faraday rotation of the plane of linear polarization either of
INTERIORS AND MAGNETIC FIELDS 43 visible light as it traverses the residual gas above the scattering atmosphere or of radar echoes from the surface of a planet or one of its satellites as the radio waves traverse the planet's magnetoionosphere. Space-probe magnetometer measurements have so far been made for only two planets, Venus and Mars. Measurements made by Mariner V as it flew past Venus at a distance of 1.7 radii from the center of the planet gave no evidence of a planetary magnetic field; a bow-shock interaction between the solar wind and the Venus ionosphere was observed. These results place an upper limit on the intrinsic magnetic field of 10"* that of Earth. There is no evidence of any electromagnetic radiation of nonthermal origin from observations of Venus made to wavelengths as long as 11 m. Magnetometer data obtained as Mariner IV flew past Mars at a distance of about two planetary radii from its center gave no evidence at any point on the trajectory of a Martian magnetic field. If a fluctuation in field strength ob- served by Mariner IV was a shock front associated with the Martian magneto- sphere, the ratio of the magnetic moment of Mars to that of the Earth is <2xlO~*. There is no evidence of nonthermal radiation from Mars to a wavelength of 17 m. Nonthermal electromagnetic radiation from Jupiter at decimeter and deca- meter wavelengths is so strong that this planet is one of the brightest radio sources in the sky. At wavelengths of 10 cm to about 1 m the spectrum is nearly flat and is produced by synchrotron radiation from the magnetosphere. This radiation is 20 percent linearly polarized, and a small, circularly polarized com- ponent has recently been reported from equatorial regions at about ±2.5 radii. From this, an order-of-magnitude estimate of 2 G for the strength of the mag- netic field was obtained, consistent with the other characteristics of the synchro- tron radiation and with the field estimated from measurements of the low- frequency decameter radiation. The orientation of the electric vector of the linearly polarized radiation defines the magnetic axis of Jupiter as tilted by 10° from the rotation axis. The resultant rocking of the plane of polarization as Jupiter rotates defines a rotation period essentially the same as the System III period derived from the low-frequency observations at about 15-m wavelength. A small variation of intensity with rotation is observed, which is apparently due to beaming of the synchrotron radiation. Asymmetries in the rocking effects and analyses of the low-frequency radiation suggest asymmetry of the dipolar magnetic field. High-resolution observations of the synchrotron radiation show the extent of the source to be about three times the disk diameter in the equatorial plane and roughly equal to the disk diameter in the polar direction, consistent with a radiation-belt structure generally similar to that of the Earth. High-resolution observations of the polarized brightness distribution are needed at other wave-
44 PLANETARY ASTRONOMY lengths to define the radiation-belt structure and to allow a separation of the thermal and nonthermal components of radiation so that the planetary proper- ties relevant to the two components may be recognized and studied. Apparent changes of the level of the synchrotron radiation over periods of days or longer have been reported; in some cases a connection with solar activity was indicated. It is important to observe Jupiter over long enough periods to confirm these results. The low-frequency decameter radiation from Jupiter is intermittent, fluc- tuating, and of very high intensity. It comes in periods of short noise-like bursts, some of the pulse structure being due to propagation effects. The inten- sity of the radiation increases over a wavelength range from approximately 7 m to the ionospheric cutoff at about 100 m. The radiation is observed less frequently at the short wavelength end. The probability of occurrence of the radiation is greater (up to about 0.8) during three distinct portions of Jupiter's rotation period. This indicates beaming from one or more sources, and a new rotation period and longitude system for Jupiter System III have been defined on this basis. This period, 9h55m29.37', based on observations made between 1950 and 1960, differs from both System I and System II but agrees well with the period found from the wobble of the magnetic axis shown by the synchro- tron radiation. There is evidence from more recent observations of slightly different periodicities in the apparent source regions. A gradual change in the rotation period of the radio source was first suspected, but the observed ap- parent period may be influenced by changes with time in the source-observer geometry or in the radio-noise storm duration. The polarization of the radia- tion, except at the longer wavelengths, is measured as predominantly right- hand elliptical or partially right-hand circular, implying a magnetic field of at least several gauss. The angular extent of a low-frequency source on Jupiter has been determined by interferometer techniques as less than 3 sec of arc. Recently, it has been recognized that the probability of occurrence of the low-frequency radiation depends not only on the System III longitude of Jupiter at the central meridian but also on the position of lo in its orbit. lo and the Jovian magnetosphere apparently interact in a way that is still unexplained, influencing the particle streams which may produce the low-frequency radiation. Because of its sporadic nature and many interrelated variables, this radiation is very difficult to study, and it is important to continue and to improve the observations. It is also important to extend the ability to make observations at the low-frequency end of the spectrum by means of satellites in high Earth orbit or by space probes; to test coherence at very widely spaced stations to clarify the millisecond burst structure, which is apparently related to Jupiter; and to measure the absolute positions of the decameter sources.
INTERIORS AND MAGNETIC FIELDS 45 If definite changes in Jupiter's magnetic field or changes in period are estab- lished, the reconciliation of these changes with the motion of other parts of the planet, as evidenced by the excursions in longitude of the Great Red Spot, will pose fascinating problems. The theoretical exploitation of these observations will bear both on the internal structure of the planet and on the origin of its magnetic field. The possibility that a dynamo mechanism is operating in the lower reaches of Jupiter's atmosphere is not inconsistent with the (admittedly limited) theoretical and observational evidence. If this is indeed the case, variations in the motions of the Red Spot and of the radio sources could be manifestations of hydromagnetic torsional oscillations of the planet. Moreover, the "topographical feature" with which the Red Spot may be associated might be magnetic in nature.
