We can measure the masses of the planets using the periods and distances of their natural satellites (Kepler’s laws, described in Chapter 3) or (for moonless Mercury and Venus) via observing their small gravitational effects on the slightly elliptical orbits of neighboring planets or on flyby spacecraft. Knowing their distances from us, we can measure the size of each of the planets using powerful telescopes and resolving the disk of the planet.
Given the mass and the size, it is possible to determine the density of each planet, that is, the mass divided by the volume.
The density for the terrestrial planets is given in Figure 11.1.
The bulk density of a planet is determined both by the mate- rial out of which it is made and the amount of compression of that material caused by the planet’s own gravitational field.
Because different materials compress to differing extents (for example, a soft pillow versus a slab of rock), the composition and compression are coupled. The more massive the planet, fur- thermore, the more compression. Also listed in Figure 11.1 are the uncompressed densities, that is, the densities of the plan- etary materials in the absence of self-compression. We can display these as single numbers only for the solid planets;
113
8
7
6
5
Density
, grams per cubic centimeter 4
3
2
0.0 0.5
crust of the Earth
ordinary chondritic meteorites metallic iron
(Moon)
Mercury Venus Earth Mars
1.0
Distance from sun in astronomical units
1.5 2.0
Figure 11.1Densities of terrestrial planets and candidate mineral components. The planets are plotted as a function of their distance from the Sun.
The symbol×indicates the measured density; the symbol◦refers to the uncompressed density of the planet when the effect of gravitational compression of the material is removed. The compressed and uncompressed densities for the Moon are shown in parentheses, to distinguish them from Earth’s. The density of Earth’s crust also is shown; for comparison, the density of liquid water under low pressure is 1 gram per cubic centimeter.
for the giant planets, most or much of the uncompressed mate- rial would be an ideal gas for which the density depends on the pressure and temperature under which the gas is contained.
Using the uncompressed densities, the abundances of the elements shown in Figure 3.8 of Chapter 3, and some chem- ical knowledge of how these elements tend to combine in the interiors of planets, we can infer the most abundant con- stituents in each of the planets. Obviously, this is not a simple deductive exercise because the uncompressed densities given in Figure 11.1 were computed on the basis of some assumed com- position. Instead, it is aniterativeexercise, wherein one elim- inates certain materials for certain planets because they do not produce the right compressed density. We will not go through the details of the exercise, but instead summarize the results for the solid planets, and then for the giant planets.
11.1.1 Solid planets
The designation solid planets includes the terrestrial planets (Mercury, Venus, Earth, Mars, and the Moon), Pluto, and the larger moons of the outer solar system. The terrestrial planets all have uncompressed densities in the range of 3 to 6 grams per cubic centimeter (g/cm3); by comparison, the density of liquid water at normal conditions is 1 g/cm3. Chondritic meteorites, composed to a large extent of minerals containing silicon, mag- nesium, and oxygen, have a density in the 3- to 4-g/cm3range.
The meteorites provide us with clues as to the nature of planet- building materials in the right density range. In terms of cosmic abundance, the rock-forming elements silicon and magnesium
are less abundant than elemental oxygen. The silicon and mag- nesium combine with abundant oxygen to form minerals such as enstatite (MgSiO3), forsterite (Mg2SiO4), and other “rocky”
compounds.
Densities of the silicate minerals are too low to account fully for the uncompressed density of all but Earth’s Moon and per- haps Mars. A clue to the identity of a denser material lies in the high abundance of iron in chondrites, as well as in the exis- tence of theiron meteorites, which have densities of 7.5 g/cm3, approaching that of metallic iron. Nearly as abundant as silicon and magnesium, iron is an excellent candidate for the material that raises the terrestrial planet densities beyond those of sili- cates. Mercury has by far the largest abundance of iron (most of its mass); the Moon has the least (close to zero). This is inter- esting in view of the fact that these two heavily cratered planets have similar sizes, the smallest of the terrestrial planets; clearly, their histories and probably their origins were quite different, given the distinct compositions. Nickel also is present in mete- orites at roughly 6% of the abundance of iron, and in planets it is expected to be present in similar amounts.
About one-third the mass of Earth is in the form of iron. Much of this may be chemically combined with sulfur or oxygen, and is known to be mostly segregated in a core, for reasons we discuss in section 11.3. Hence Earth is stratified with the densest material toward the center; this is likely to be the case for all of the planets and most of the moons. The outermost chemical layer, or crust, of Earth is composed of minerals containing largely silicon and magnesium with an admixture of lower-density minerals.
Aluminum, for example, underabundant relative to silicon and
magnesium but with similar mineral-forming properties, is more abundant in the crust of Earth than throughout the rest of its interior. Venus’ strong resemblance to Earth in density and size leads us to conclude that, in bulk composition, it is similar to Earth. Limited chemical measurements of the surface from landed, Russian space probes suggest this to be the case, even though the geologic processes shaping the face of Venus appear to be different from those on Earth (see Chapter 15).
Most of the major moons of the outer solar system, and Pluto, have densities around 2 g/cm3. This is too low to be accounted for by common silicate minerals, but too high for pure ices. A sensible inference is that these bodies are roughly one-half ice and one-half silicate by mass. Because oxygen is significantly more abundant than carbon, nitrogen, or other ice- forming elements, it is a logical assumption that water dom- inates the ice component of these moons, with admixtures of ammonia, methane, carbon dioxide, and other ices. Spectro- scopic identification of water ice, and of other ices in the cases of Triton and Pluto, seem to confirm these general ideas. Two major moons of Jupiter – Io and Europa – exceed 3 g/cm3in density. Io is almost entirely silicate; Europa is lower in density and appears to have a water-ice veneer. They both may have lost water early on, or never acquired significant quantities in the first place.
Interpreting planetary compositions in terms of the internal arrangement, or structure, of the various major components is a challenge that requires additional observational tools. We describe the monitoring of earthquakes to infer Earth’s internal structure in section 11.3.
11.1.2 The giant planets
Determining the detailed composition of the giant planets, pic- tured in Figure 11.2, is difficult because of their distance from Earth, and the inaccessibility of their vast interiors. Density can be measured from size and mass, and the values are 1.33 for Jupiter, 0.69 for Saturn, 1.27 for Uranus, and 1.64 for Neptune, all in units of grams per cubic centimeter. These are much lower than the densities of the terrestrial planets plotted in Figure 11.1.
Equally important to understanding composition is the determi- nation of the shape of the giant planet and hence its gravitational field. Such information provides constraints on whether espe- cially dense layers are located near the planet’s center, and to what extent the outer gaseous layers are pure hydrogen and helium. Measuring the gravitational field requires precise track- ing of the orbits of a planet’s moons, particularly its closest ones, and this must be done using flyby robotic spacecraft such as the Cassini Saturn Orbiter. Tracking the paths of the space- craft themselves also yields gravity data. The spin rate of the planet also must be measured, because a faster spin tends to flatten gaseous planets significantly, affecting the distribution of mass in their interiors and hence their gravitational fields.
Both Jupiter and Saturn have such low densities that they must be made up mostly of the light, primordial elements hydrogen and helium; spacecraft spectroscopic measurements show that helium is 10 to 15% the abundance of hydrogen in the outer lay- ers of Jupiter, Uranus, and Neptune, but much lower in Saturn.
The giant planets much more closely resemble the Sun in com- position than do the terrestrial planets. However, there must be
much more of the elements heavier than hydrogen and helium in the giant planets than in the Sun, based on their densities and observed gravitational fields. The composition of the deep interiors cannot be sampled directly but is likely to be largely the abundant rock- and ice-forming elements such as oxygen, carbon, nitrogen, silicon, magnesium, and iron. The tremen- dous pressures, 70 million times Earth’s sea level pressure at the center of Jupiter, force these materials to exist in chemical con- figurations different from those we are used to seeing on Earth.
A schematic slice of Jupiter’s interior is shown in Figure 11.3.
Spacecraft and Earth-based identification of hydrogen and helium by spectroscopy and other techniques confirm their pres- ence, at least in the outermost layers. However, other molecules are found to be present, such as methane and ammonia, which likely contain most of the carbon and nitrogen atoms in the outer layers; water is probably present but temperatures in the atmospheres are low enough that it is condensed out below the measurable outer layer of these planets. If Jupiter and Saturn had an overall composition equal to that of the Sun, we would expect that no more than 1% of the mass of each planet would be in the form of heavy elements. However, 10% or more by mass of each planet must be elements heavier than hydrogen or helium, based again on gravity tracking, and this important fact drives scientists to the model described in Chapter 10 in which solid cores form first and then gravitationally attract nebular gas to form the giant planets. In effect, the genesis of Jupiter and Saturn begins with the formation of terrestrial- or ice-moon-type bodies, a process that does not stop until these protoplanets are drowned in hundreds of Earth masses (in the case of Jupiter) of hydrogen-helium gas. Continued infall of icy planetesimals during and after this process “salts” the gas envelopes of these planets with more ice- and rock-forming material.
Uranus and Neptune hold far less hydrogen-helium gas than do Jupiter and Saturn but vastly more than the terrestrial plan- ets. Though not as rich in hydrogen and helium as are Jupiter and Saturn they may contain large amounts of water. Because of their great distances from the Sun, their atmospheres are extremely cold; the water is frozen out of the upper, observable, atmospheres, where instead clouds of methane are seen to form.
Careful measurement of the absorption of sunlight and emis- sion of heat from each of the giant planets reveals that, with the exception of Uranus, each releases more energy in the form of heat than it receives as sunlight. There must be an internal source of energy in each planet, but it cannot be hydrogen fusion, because temperatures and pressures computed for the center of each body are too small to overcome the repulsive forces that prevent protons from fusing together. Even deuterium fusion, which is easier to initiate, cannot be achieved; computations show that a body must be 13 times the mass of Jupiter for such reactions to take place.
The most plausible source of heat comes from the forma- tion of the giant planets themselves. In compressing gas into a self-gravitating, bound sphere, from a state in which the gas originally is spread over a large region of space, potential energy is lost and converted into random energy of motion of the atoms and molecules, that is, into heat. It is the same process that heats the air that you pump into your bicycle tire, but the source of compression is gravitational energy rather than the stored chemical energy in your muscles that you use to move the pump
(c)
(b) (d)
(a)
Figure 11.2The giant planets of our solar system: (a) Jupiter from Hubble Space Telescope; (b) Saturn from Hubble, with contrast exaggerated to show atmospheric patterns; (c) Uranus from Voyager 2, also with enhanced contrast to show very faint banding; (d) Neptune from Voyager 2.
Photos (a) and (b) courtesy of NASA and the Space Telescope Science Institute; (c) and (d) courtesy NASA and the Jet Propulsion Laboratory. See color versions of (a), (c), and (d) in plates section.
piston. The initial energy of formation cannot be lost all at once;
processes such as conduction, radiation, and convection, which transport heat from the inside of a large body such as a planet, do so in a finite amount of time. Hence heat is still being lost today. University of Arizona scientist W. B. Hubbard and col- leagues showed, back in the 1970s, that the excess heat emitted by Jupiter today is consistent with residual heat of formation if Jupiter formed some 4 billion to 5 billion years ago – consistent with the age of the solar system. Neptune’s heat yields a similar result.
Saturn and Uranus, however, are a mystery. Saturn emits almost twice as much heat as it should if the energy is that of
its initial collapse some 4.5 billion years ago. Is Saturn younger than the other giant planets? A more elegant and sensible expla- nation comes from the helium abundance, which is depleted in the upper layers as measured from spacecraft. David Stevenson of the California Institute of Technology and Edwin Salpeter of Cornell showed, over two decades ago, that Saturn’s interior is cool enough (in contrast to Jupiter’s) that helium is chemically insoluble in hydrogen. The helium has been separating out as droplets that fall toward Saturn’s center because their density is larger than that of the surrounding hydrogen. This slow rainout of helium contributes additional gravitational energy, which is detected as excess heat emission.
0 km
~17,000 K
~70 Mbar
metallic hydrogen molecular
hydrogen core
10,000 km
20,000 km
30,000 km
40,000 km
50,000 km
60,000 km
70,000 km
165 K 1 bar
Figure 11.3A slice through the interior of Jupiter, with distances to the center marked, as well as the pressure and temperature at the top and in the center.
The process ofplanetary differentiation, in which the interior materials are sorted out according to density, is important in Saturn because the ringed planet is less massive than Jupiter, and hence its interior cooled quickly to the point where separation could begin. Subsequent calculations by Stevenson suggested that helium rainout also is occurring in Jupiter, but began more recently than in Saturn.VoyagerandGalileodata on the helium abundances of these giant planets, showing a strong helium depletion in Saturn’s atmosphere, but not Jupiter’s, appear to support the model.
Measurements showing that Uranus emits essentially no heat, other than what it derives from sunlight, do not yet yield a tidy explanation. Nearly Neptune’s twin in terms of size, mass, and density, we expect it to have a similar source of internal heat.
However, Uranus spins on an axis that lies parallel to the plane of its orbit around the Sun, rather than close to perpendicular as with Earth and most of the other planets. Over the past decade, Uranus has had one pole tipped toward the Sun, and hence is receiving solar energy in a very different distribution of latitudes than is Neptune. It is possible that this has bottled up or redi- rected the internal heat of Uranus so that it is not observable.
As Uranus moves in its orbit so that the equator, rather than the poles, points toward the Sun, interior energy may be released.
That this is happening is suggested by Hubble Space Telescope images of Uranus in the late 1990s showing clouds becoming more abundant on its surface, a surface thatVoyager 2 found to be bland in 1989 (Figure 11.4). Perhaps the “cork” has been popped from the planetary bottle as Uranus moves from a sol- stice orientation to one of equinox, that is, from summer/winter to spring/fall.