(Background, above) This odd looking photo, taken by the Hubble telescope orbiting Earth, shows the planet Uranus with its many rings and five of its inner moons. The picture seems odd because each of the moons appears as a string of three dots. It is actually a composite of three images, taken 6 minutes apart, thereby showing the moons' rather rapid revolution just beyond the planet's rings.

(Inset A) The largest of Uranus's inner moons is Miranda, seen here from the perspective of the oncoming Voyager spacecraft. Also visible at lower right in the Hubble photo, Miranda has less than one percent of the mass of Earth's Moon.

(Insets B, C, and D) Progressively closer views of Miranda, radioed back as Voyager skirted to within 30,000 kilometers of this peculiar moon, show several of its prominent features. So many major terrain types are evident in these photos that some scientists have likened Miranda to a cosmic geology museum. Currently, astronomers have no clear consensus about how Miranda's unusual surface features originated, but it seems clear that some very violent events must have been involved, probably about 4 billion years ago. Consult Figure 13.14 for further details about Miranda's cliffs, grooves, and greatly puzzling terrain.

LEARNING GOALS

Studying this chapter will enable you to:

Describe how both calculation and chance played a major role in the discoveries of the outer planets.

Summarize the similarities and differences between Uranus and Neptune, and compare these planets with the other two jovian worlds.

Explain what the moons of the outer planets tell us about their past.

Contrast the rings of Uranus and Neptune with those of Jupiter and Saturn.

Summarize the orbital and physical properties of Pluto, and explain how the Pluto-Charon system differs fundamentally from all the other planets.

The three outermost planets were unknown to the ancients. All were discovered by telescopic observations--Uranus in 1781, Neptune in 1846, and Pluto in 1930. Uranus and Neptune have very similar masses and radii, so it is natural to consider them together. They are part of the jovian family of planets. Pluto, by contrast, is not a jovian world. It is very much smaller than even the terrestrial planets and generally seems much more moonlike than planetlike in character. Indeed, it may well be a one-time moon that has escaped from one of the outer planets, most likely Neptune. Because of Pluto's similarity to the jovian moons and its possible (although unproven) connection with Neptune, we study it here along with its larger jovian neighbors.

Back The planet Uranus was discovered by British astronomer William Herschel in 1781. Herschel was engaged in charting the faint stars in the sky when he came across an odd-looking object that he described as "a curious either nebulous star or perhaps a comet." Repeated observations showed that it was neither. The object appeared as a disk in Herschel's 6-inch telescope and moved relative to the stars, but it traveled too slowly to be a comet. Herschel soon realized that he had found the seventh planet in the solar system. Since this was the first new planet discovered in well over 2000 years, the event caused quite a stir at the time. The story goes that Herschel's first instinct was to name the new planet "Georgium Sidus" (Latin for "George's star") after his king, George III of England. The solar system was saved from a planet named George by the wise advice of another astronomer, Johann Bode. He suggested instead that the tradition of using names from Greco-Roman mythology be continued and that the planet be named after Uranus, the father of Saturn.

Careful observations since its discovery have allowed astronomers to determine the orbital properties of Uranus. Its orbit has a semi-major axis of 19.2 A.U., an eccentricity of 0.05, and a sidereal orbital period of 84.0 years. Since its discovery in 1781, Uranus has completed only two and a half revolutions about the Sun. The planet is in fact just barely visible to the naked eye, if you know exactly where to look. At opposition, it has a maximum angular diameter of 4.1´´ and shines just above the unaided eye's threshold of visibility. It looks like a faint, undistinguished star. No wonder it went unnoticed by the ancients. Even today, few astronomers have seen it without a telescope.

Through a large Earth-based optical telescope (see Figure 13.1), Uranus appears hardly more than a tiny pale greenish disk. With the flyby of Voyager 2 in 1986, our detailed knowledge of Uranus increased dramatically. Figure 13.2 is a combination of close-up visible-light images of the planet and one of its moons. The apparently featureless atmosphere of Uranus contrasts sharply with the bands and spots visible on all the other jovian worlds.

Figure 13.1 Details are barely visible on photographs of Uranus made with large Earth-based telescopes. (Arrows point to three of its moons.)

Figure 13.2 A montage of Uranus and one of its satellites, composed of photos sent back to Earth by the Voyager 2 spacecraft as it whizzed past this giant planet at 10 times the speed of a rifle bullet. The image of Uranus, taken from a distance of about 100,000 km, shows the planet's blue yet featureless upper atmosphere. The image in the foreground is Miranda, one of Uranus's moons. (Uranus's rings, not visible in the photograph, have been added by an artist.)

Once Uranus was discovered, astronomers set about charting its orbit. The figures we have just listed do indeed describe Uranus's orbital motion, but eighteenth-century astronomers quickly discovered a small discrepancy between the planet's predicted position and where they actually observed it. Try as they might, astronomers could not find an elliptical orbit that fit the planet's trajectory to within the accuracy of their measurements. Half a century after Uranus's discovery, the discrepancy had grown to a quarter of an arc minute, far too big to be explained away as observational error.

The logical conclusion was that an unknown body must be exerting a gravitational force on Uranus--much weaker than that of the Sun, but still measurable. But what body could this be? Astronomers realized that there had to be another planet in the solar system perturbing Uranus's motion.

In the 1840s, two mathematicians independently solved the difficult problem of determining this new planet's mass and orbit. A British astronomer, John Adams, reached the solution in September 1845; in June of the following year, the French mathematician Urbain Leverrier came up with essentially the same answer. British astronomers seeking the new planet found nothing during the summer of 1846. In September, a German astronomer named Johann Galle began his own search from the Berlin Observatory, using a newly completed set of more accurate sky charts. He found the new planet within one or two degrees of the predicted position--on his first attempt. After some wrangling over names and credits, the new planet was named Neptune, and Adams and Leverrier (but not Galle!) are now jointly credited with its discovery.

Neptune orbits the Sun with a semi-major axis of 30.1 A.U. (4.5 billion km) and an eccentricity of just 0.01. Since its sidereal orbital period is 164.8 years, it has not yet completed one revolution since its discovery. Unlike Uranus, Neptune cannot be seen with the naked eye, although it can be seen with a small telescope--in fact, Galileo may actually have seen Neptune, although he had no idea what it really was at the time. Through a large telescope, Neptune appears as a bluish disk, with a maximum angular diameter of 2.4´´ at opposition.

Figure 13.3 shows a long Earth-based exposure of Neptune and its largest moon, Triton. Neptune is so distant that surface features are virtually impossible to discern. Even under the best observing conditions, only a few markings can be seen. These are suggestive of multicolored cloud bands--light bluish hues seem to dominate. With Voyager 2's arrival, much more detail emerged, as shown in Figure 13.4. Superficially, at least, Neptune resembles a blue-tinted Jupiter, with atmospheric bands and spots clearly evident.

Figure 13.3 Neptune and two of its moons, Triton (large arrow) and Nereid (small arrow), imaged with a large Earth-based telescope.

Figure 13.4 (a) Neptune as seen by Voyager 2, from a distance of roughly 1 million km. A closer view (b), resolved to about 10 km, shows cloud streaks ranging in width from 50 to 200 km.

MASSES AND RADII

Back Figure 13.5 shows Uranus and Neptune to scale, along with the Earth for comparison. The two giant planets are quite similar in their bulk properties, too. The diameter of Uranus is 51,100 km (4.0 times that of the Earth); that of Neptune, 49,500 km (3.9 Earth diameters). Their masses (first determined from terrestrial observations of their larger moons and later refined by Voyager 2) are 8.7 × 1025 kg (14.5 Earth masses) for Uranus and 1.0 × 1026 kg (17.1 Earth masses) for Neptune. Thus, Uranus's average density is 1200 kg/m3 (1.2 g/cm3), and Neptune's is 1700 kg/m3. These densities imply that large rocky cores constitute a greater fraction of the planets' masses than do the cores of either Jupiter or Saturn. The cores themselves are probably comparable in size, mass, and composition to those of the two larger giants.

Jovian Planet Part I

Figure 13.5 Jupiter, Saturn, Uranus, Neptune, and Earth, drawn to scale. Uranus and Neptune are quite similar in their bulk properties. Each probably contains a core about 10 times more massive than the Earth. Jupiter and Saturn are each substantially larger, but their rocky cores are probably comparable in mass to those of Uranus and Neptune.

ROTATION RATES

Like the other jovian planets, Uranus has a short rotation period. Earth-based observations of the Doppler shifts in spectral lines first indicated that Uranus's "day" was between 10 and 20 hours long. The precise value of the planet's rotation period--accurately determined when Voyager 2 timed radio signals associated with its magnetosphere--is now known to be 17.2 hours. Again as with Jupiter and Saturn, the planet's atmosphere rotates differentially, but Uranus's atmosphere actually rotates faster at the poles (where the period is 14.2 hours) than near the equator (where the period is 16.5 hours).

Each planet in our solar system seems to have some outstanding peculiarity, and Uranus is no exception. Unlike all the other planets, which have their spin axes roughly perpendicular to the ecliptic plane, Uranus's rotation axis lies almost within that plane--98° from the perpendicular, to be precise. Relative to the other planets, we might say that Uranus lies tipped over on its side. As a result, the "north" (spin) pole* of Uranus, at some time in its orbit, points almost directly toward the Sun. Half a "year" later, its "south" pole faces the Sun, as illustrated in Figure 13.6. When Voyager 2 encountered the planet in 1986, the north pole happened to be pointing nearly at the Sun, so it was midsummer in the northern hemisphere.

*As in Chapter 9, we adopt the convention that a planet's rotation is always counterclockwise as seen from above the north pole (that is, planets always rotate from west to east).

Figure 13.6 Uranus's 98 axial tilt places its equator almost perpendicular to the ecliptic. As a result, the planet experiences the most extreme seasons known in the solar system. The equatorial regions experience two warm seasons (around the two equinoxes) and two cold seasons (at the solstices) each year, while the poles are alternately plunged into darkness for 42 years at a time.

The strange orientation of Uranus's rotation axis produces some extreme seasonal effects. Starting at the height of northern summer, when the north pole points closest to the Sun, an observer near that pole would see the Sun move in gradually increasing circles in the sky, completing one circuit (counterclockwise) every 17 hours and dipping slightly lower in the sky each day. Eventually the Sun would begin to set and rise again in a daily cycle, and the nights would grow progressively longer with each passing day. Twenty-one years after the summer solstice, the autumnal equinox would occur, with day and night each 8.5 hours long.

The days would continue to shorten until one day the Sun would fail to rise at all. The period of total darkness that followed would be equal in length to the earlier period of constant daylight, plunging the northern hemisphere into the depths of winter. Eventually, the Sun would rise again; the days would lengthen through the vernal equinox and beyond, and in time the observer would again experience a summer of uninterrupted (though dim) sunshine.

From the point of view of an observer on the equator, by contrast, summer and winter would be almost equally cold seasons, with the Sun never rising far above the horizon. Spring and fall would be the warmest times of year, with the Sun passing almost overhead each day.

No one knows why Uranus is tilted in this way. Some scientists have speculated that a catastrophic event, such as a grazing collision between the planet and another planet-sized body, might have altered the planet's spin axis. There is no direct evidence for such an occurrence, however, and no theory to tell us how we should seek to confirm it.

Neptune's clouds show more variety and contrast than do those of Uranus, and Earth-based astronomers studying them determined a rotation rate for Neptune even before Voyager 2's flyby in 1989. The average rotation period of Neptune's atmosphere is 17.3 hours (virtually identical to that of Uranus). Measurements of Neptune's radio emission by Voyager 2 showed that the magnetic field of the planet, and presumably also its interior, rotates once every 16.1 hours. Thus, Neptune is unique among the jovian worlds in that its atmosphere rotates more slowly than its interior. Neptune's rotation axis is inclined 29.6° to a line perpendicular to its orbital plane, quite similar to the 27° tilt of Saturn.

Detailed Imaging of Uranus' Atomsphere

COMPOSITION

Spectroscopic studies of sunlight reflected from Uranus's and Neptune's dense clouds indicate that the two planets' outer atmospheres (the parts we actually measure spectroscopically) are similar to those of Jupiter and Saturn. The most abundant element is molecular hydrogen (84 percent), followed by helium (about 14 percent) and methane, which is more abundant on Neptune (about 3 percent) than on Uranus (2 percent). Ammonia, which plays such an important role in the Jupiter and Saturn systems, is not present in any significant quantity in the outermost jovian worlds.

Comparing this information with that given previously for Jupiter and Saturn, we can see that the abundances of gaseous ammonia and methane vary systematically among the jovian planets. Jupiter has much more gaseous ammonia than methane, but as we proceed outward, we find that the more distant planets have steadily decreasing amounts of ammonia and relatively greater amounts of methane. The reason for this variation is temperature. Ammonia gas freezes into ammonia ice crystals at about 70 K. This is cooler than the cloud-top temperatures of Jupiter and Saturn but warmer than those of Uranus (58 K) and Neptune (59 K). Thus, the outermost jovian planets have little or no gaseous ammonia in their atmospheres, so their spectra (which record atmospheric gases only) show just traces of ammonia.

The increasing amounts of methane are largely responsible for the outer jovian planets' blue coloration. Methane absorbs long-wavelength red light quite efficiently, so that sunlight reflected from the planet's atmospheres is deficient in red and yellow photons and appears blue-green or blue. As the concentration of methane increases, the reflected light should appear bluer. This is just the trend observed: Uranus, with less methane, looks blue-green, while Neptune, with more methane, looks distinctly blue.

WEATHER

Voyager 2 detected only a few features in Uranus's atmosphere, and even these became visible only after extensive computer enhancement (see Figure 13.7). Uranus apparently lacks any significant internal heat source, and because of the planet's low surface temperature, its clouds are found only at lower-lying, warmer levels in the atmosphere. The absence of high-level clouds means that we must look deep into the planet's atmosphere to see any structure, and the bands and spots that characterize flow patterns on the other jovian worlds are largely "washed out" on Uranus by intervening stratospheric haze.

Figure 13.7 (a) This Voyager view of Uranus approximates the planet's true color but shows little else. Parts (b), (c), and (d) are Hubble Space Telescope photographs made at roughly four-hour intervals, showing the motion of a pair of bright clouds (labeled A and B) in the planet's southern hemisphere. (The numbers at the top give the time of each photo.)

With computer-processed images, astronomers have learned that Uranus has atmospheric clouds and flow patterns that move around the planet in the same sense as the planet's rotation, with wind speeds in the range 200-500 km/h. Tracking these clouds allowed the measurement of the differential rotation mentioned earlier. Despite the odd angle at which sunlight is currently striking the surface (recall that it is just after midsummer in the northern hemisphere), the planet's rapid rotation still channels the wind flow into bands reminiscent of those found on Jupiter and Saturn. Even though the predominant wind flow is in the east­west direction, Uranus's atmosphere seems to be quite efficient at transporting energy from the heated north to the unheated southern hemisphere. Although the south is currently in total darkness, the temperature there is only a few kelvins less than in the north.

Neptune's cloud and band structure is much more easily seen. Although it lies at a greater distance from the Sun, Neptune's upper atmosphere is actually slightly warmer than Uranus's. Like Jupiter and Saturn, but unlike Uranus, Neptune has an internal energy source--in fact, it radiates 2.7 times more heat than it receives from the Sun. The cause of this heating is still uncertain. Some scientists have suggested that Neptune's excess methane has helped "insulate" the planet, tending to maintain its initially high internal temperature. If that is so, then Neptune's internal heat has the same basic explanation as Jupiter's--it is energy left over from the planet's formation. The combination of extra heat and less haze may be responsible for the greater visibility of Neptune's atmospheric features (see Figure 13.8), as its cloud layers lie at higher levels in the atmosphere than do Uranus's.

Figure 13.8 (a) Close-up views, taken by Voyager 2 in 1989, of the Great Dark Spot of Neptune, a large storm system in the planet's atmosphere, possibly similar in structure to Jupiter's Great Red Spot. Resolution in the photo on the right is about 50 km; the entire dark spot is roughly the size of planet Earth. (b) These three Hubble Space Telescope views of Neptune were taken about 10 days apart in 1994, when the planet was some 4.5 billion km from the Earth. The aqua color is caused by absorption of red light by methane; the cloud features (mostly methane ice crystals) are tinted pink here because they were imaged in the infrared, but are really white in visible light. Neptune apparently has a remarkably dynamic atmosphere that changes over just a few days. Notice that the Great Dark Spot has now disappeared.

Neptune sports several storm systems similar in appearance to those seen on Jupiter (and assumed to be produced and sustained by the same basic processes). The largest such storm, known simply as the Great Dark Spot, is shown in Figure 13.8(a). Discovered by Voyager 2 in 1989, the Spot was about the size of the Earth, was located near the planet's equator, and exhibited many of the same general characteristics as the Great Red Spot on Jupiter. The flow around it was counterclockwise, as with the Red Spot, and there appeared to be turbulence where the winds associated with the Great Dark Spot interacted with the zonal flow to its north and south. The flow around this and other dark spots may drive updrafts to high altitudes, where methane crystallizes out of the atmosphere to form the high-lying cirrus clouds. Astronomers did not have long to study the Dark Spot's properties, however. As shown in Figure 13.8(b), when the Hubble Space Telescope viewed Neptune in 1994, the Spot had disappeared.

Voyager 2 found that both Uranus and Neptune have fairly strong internal magnetic fields--about 100 times stronger than Earth's field and 1/10 as strong as Saturn's. However, because the radii of Uranus and Neptune are so much larger than the radius of Earth, the magnetic fields at the cloud tops, spread out over a far larger volume than the field on Earth, are actually comparable in strength to Earth's field. Uranus and Neptune each have a substantial magnetosphere, populated largely by electrons and protons either captured from the solar wind or created from ionized hydrogen gas escaping from the planets themselves.

When Voyager 2 arrived at Uranus, it discovered that the planet's magnetic field is tilted at about 60° to the axis of rotation. On Earth, such a tilt would put the North (magnetic) Pole somewhere in the Caribbean. Furthermore, the magnetic field lines are not centered on the planet. It is as though Uranus's field were due to a bar magnet that is tilted with respect to the planet's rotation axis and displaced from the center by about one-third the radius of the planet. Figure 13.9 compares the magnetic field structures of the four jovian planets. The locations and orientations of the bar magnets represent the observed planetary fields, and the sizes of the bars indicate magnetic field strength.

Figure 13.9 A comparison of the magnetic field strengths, orientations, and offsets in the four jovian planets: (a) Jupiter, (b) Saturn, (c) Uranus, (d) Neptune. The planets are drawn to scale, and in each case the magnetic field is represented as though it came from a simple bar magnet. The size and location of each magnet represent the strength and orientation of the planetary field. Notice that the fields of Uranus and Neptune are significantly offset from the center of the planet and are significantly inclined to the planet's rotation axis. The Earth's magnetic field is shown for comparison.

Because dynamo theories generally predict that the magnetic axis should be roughly aligned with the rotation axis--as on Earth, Jupiter, Saturn, and the Sun--the misalignment on Uranus suggested to some researchers that perhaps the planet's field had been caught in the act of reversing. Another possibility was that the oddly tilted field was in some way related to the planet's axial tilt--perhaps one catastrophic collision skewed both axes at the same time. Those ideas evaporated in 1989 when Voyager 2 found that Neptune's field is also inclined to the planet's rotation axis, at an angle of 46° (see Figure 13.9d), and also substantially offset from the center. It now appears that the internal structures of Uranus and Neptune are different from those of Jupiter and Saturn, and this difference somehow changes the nature of the field-generation process.

Theoretical models indicate that Uranus and Neptune have rocky cores similar to those found in Jupiter and Saturn--about the size of Earth and perhaps 10 times more massive. However, the pressure outside the cores of Uranus and Neptune (unlike the pressure within Jupiter and Saturn) is too low to force hydrogen into the metallic state, so hydrogen stays in its molecular form all the way in to the planets' cores. Astronomers theorize that, deep below the cloud layers, Uranus and Neptune may have high-density, "slushy" interiors containing thick layers of water clouds. It is also possible that much of the planets' ammonia could be dissolved in the water, accounting for the absence of ammonia at higher cloud levels. Such an ammonia solution would provide a thick, electrically conducting ionic layer that could conceivably explain the planets' magnetic fields. At the present time, however, we don't know enough about the interiors to assess the correctness of this picture. Our current state of knowledge is summarized in Figure 13.10, which compares the internal structures of the four jovian worlds.

Jovian Planet Part II

Figure 13.10 A comparison of the interior structures of the four jovian planets. (a) The planets drawn to scale. (b) The relative proportions of the various internal zones.

Back Like Jupiter and Saturn, both Uranus and Neptune have extensive moon systems, each consisting of a few large moons, long known from the Earth, and many smaller moonlets, discovered by Voyager 2.

URANUS'S MOONS

William Herschel discovered and named the two largest of Uranus's five major moons, Titania and Oberon, in 1789. British astronomer William Lassell found the next largest, Ariel and Umbriel, in 1851. Gerard Kuiper found the smallest, Miranda, in 1948. In order of increasing distance from the planet, they are Miranda (at 5.1 planetary radii), Ariel (at 7.5), Umbriel (at 10.4), Titania (at 17.1), and Oberon (at 22.8). The 10 smaller moons discovered by Voyager 2 all lie inside the orbit of Miranda. Many of them are intimately related to Uranus's ring system. All the moons revolve in Uranus's skewed equatorial plane, almost perpendicular to the ecliptic, in circular, tidally locked orbits. Their properties are listed in Table 13-1. Because the satellites share Uranus's odd orientation, they experience the same extreme seasons as their parent planet.

The five largest moons are similar in many respects to the six midsized moons of Saturn. Their densities lie in the range 1300-1600 kg/m3, suggesting composition of ice and rock, like Saturn's moons, and their diameters range from 1600 km for Titania and Oberon, to 1200 km for Umbriel and Ariel, to 490 km for Miranda. Uranus has no moons comparable to the Galilean satellites of Jupiter, nor to Saturn's single large moon, Titan. Figure 13.11 shows Uranus's five large moons to scale, along with the Earth's Moon for comparison.

Figure 13.11 The five largest moons of Uranus, to scale. In order of increasing distance from the planet, they are Miranda, Ariel, Umbriel, Titania, and Oberon. Earth's moon is shown for comparison.

The two outermost moons, Titania and Oberon (shown in Figure 13.12), are heavily cratered and show little indication of any geological activity. Their overall appearance (and quite possibly their history) is comparable to that of Saturn's moon Rhea, except that they lack Rhea's wispy streaks. Also, like all Uranus's moons, they are considerably less reflective than Saturn's satellites, suggesting that their icy surfaces are quite dirty. One possible reason for this might simply be that the planetary environment in the vicinity of Uranus and Neptune contains more small "sooty" particles than does the solar system closer to the Sun. An alternative explanation, now considered more likely by many planetary scientists, cites the effects of radiation and high-energy particles striking the surfaces of these moons. These impacts tend to break up the molecules on the moons' surfaces, eventually leading to chemical reactions that slowly build up a layer of dark, organic material. This radiation darkening is thought to contribute to the generally darker coloration of many of the moons and rings in the outer solar system. In either case, the longer a moon has been inactive and untouched by meteoritic impact, the darker its surface should be.

Figure 13.12 Close-up comparison of Uranus's two largest moons, Titania (a) and Oberon (b). Their appearance, structure, and history may be quite similar to those of Saturn's moon Rhea. Smallest details visible on both moons are about 15 km across.

The darkest of the moons of Uranus is Umbriel (Figure 13.13a). It displays little evidence for any past surface activity; its only mark of distinction is a bright spot about 30 km across, of unknown origin, on its sunward side. By contrast, Ariel (Figure 13.13b), similar in size to Umbriel but closer to Uranus, does appear to have experienced some activity in the past. It shows signs of resurfacing in places and exhibits surface cracks a little like those seen on another of Saturn's moons, Tethys. However, unlike Tethys, whose cracks are probably due to meteoritic impact, Ariel's activity probably occurred as internal forces and external tidal stresses (due to the gravitational pull of Uranus) distorted the moon and cracked its surface.

Figure 13.13 (a) The moon Umbriel is one of the darkest bodies in the solar system. Its most noteworthy feature is a bright white spot on its sunward side. (b) Ariel is similar in size but has a brighter surface. Unlike Umbriel, its surface shows signs of past geological activity. Resolution is approximately 10 km.

Strangest of all Uranus's icy moons is Miranda, shown in Figure 13.14. Before the Voyager 2 encounter, astronomers thought that Miranda would most resemble Mimas, the moon of Saturn whose size and location it most closely approximates. However, instead of being a relatively uninteresting cratered, geologically inactive world, Miranda displays a wide range of surface terrains, including ridges, valleys, large oval faults, and many other tortuous geological features. In order to explain why Miranda seems to combine so many different types of surface features, some researchers have hypothesized that this baffling object has been catastrophically disrupted several times (from within or without), with the pieces falling back together in a chaotic, jumbled way. Certainly, the frequency of large craters on the outer moons suggests that destructive impacts may once have been quite common in the Uranus system. It will be a long time, though, before we can obtain more detailed information to test this theory.

Figure 13.14 Miranda, an asteroid-sized moon of Uranus photographed by Voyager 2, has a strange, fractured surface suggestive of a violent past, but the cause of the grooves and cracks is presently unknown. The resolution here is about 1 km.

NEPTUNE'S MOONS

From Earth, we can see only two moons orbiting Neptune. William Lassell discovered the inner moon, Triton, in 1846. The outer moon, Nereid, was located by Gerard Kuiper in 1949. Voyager 2 discovered six additional moons, all less than a few hundred kilometers across and all lying within Nereid's orbit. Neptune's known moons are listed in Table 13-2.

In its moons we find Neptune's contribution to our list of solar system peculiarities. Unlike the other jovian worlds, Neptune has no regular moon system. The larger moon, Triton, is 2800 km in diameter and occupies a circular retrograde orbit 354,000 km (14.2 planetary radii) from the planet, inclined at about 20° to Neptune's equatorial plane. It is the only large moon in our solar system to have a retrograde orbit. The other moon visible from Earth, Nereid, is only 200 km across. It orbits Neptune in the prograde sense, but on an elongated trajectory that brings it as close as 1.4 million km to the planet and as far away as 9.7 million km. Nereid is probably similar in both size and composition to Neptune's small inner moons.

Voyager 2 approached to within 24,000 km of Triton's surface, providing us with essentially all that we now know about that distant, icy world. Astronomers redetermined the moon's radius (which was corrected downward by about 20 percent) and measured its mass for the first time. Along with Saturn's Titan and the four Galilean moons of Jupiter, Triton is one of the six large moons in the outer solar system. Triton is the smallest of them, with about half the mass of the next smallest, Jupiter's Europa.

Lying 4.5 billion km from the Sun, and with a fairly reflective surface, Triton has a surface temperature of just 37 K. It has a tenuous nitrogen atmosphere, perhaps a hundred thousand times thinner than Earth's, and a surface that most likely consists primarily of water ice. A Voyager 2 mosaic of Triton's south polar region is shown in Figure 13.15. The moon's low temperatures produce a layer of nitrogen frost that forms and evaporates over the polar caps, a little like the carbon dioxide frost responsible for the seasonal caps on Mars. The frost is visible as the pinkish region on the left of Figure 13.15. Overall, there is a marked lack of cratering on Triton, presumably indicating that surface activity has obliterated the evidence of most impacts. There are many other signs of an active past. Triton's face is scarred by large fissures similar to those seen on Ganymede, and the moon's odd cantaloupe-like terrain may indicate repeated faulting and deformation over the moon's lifetime. In addition, Triton has numerous frozen "lakes" of water ice (Figure 13.16), which are believed to be volcanic in origin.

Figure 13.15 The south polar region of Triton, showing a variety of terrains, ranging from deep ridges and gashes to what appear to be frozen water lakes, all indicative of past surface activity. The pinkish region at the left is nitrogen frost, forming the moon's polar cap. Resolution is about 4 km.

Figure 13.16 Scientists believe that this lakelike feature on Triton may have been caused by the eruption of an ice volcano. The water "lava" has since solidified, leaving a smooth surface. The absence of craters indicates that this eruption was a relatively recent event in Triton's past. The nearly circular feature at the center of this image spans some 200 km in diameter; its details are resolved to a remarkable 1 km. The insert is a computer-generated view illustrating the topographic relief of the same area.

Triton's surface activity is not just a thing of the past. As Voyager 2 passed the moon, its cameras detected two great jets of nitrogen gas erupting from below the surface, rising several kilometers into the sky. It is thought that these "geysers" result when liquid nitrogen below Triton's surface is heated and vaporized by some internal energy source, or perhaps even by the Sun's feeble light. Vaporization produces high pressures, which force the gas through cracks and fissures in the crust, creating the displays Voyager 2 saw. Scientists conjecture that nitrogen geysers may be very common on Triton and are perhaps responsible for much of the moon's thin atmosphere.

The event or events that placed Triton on a retrograde orbit and Nereid on such an eccentric path are unknown, but they are the subject of considerable speculation. Triton's peculiar orbit and surface features suggest to some astronomers that the moon did not form as part of the Neptune system but instead was captured, perhaps not too long ago. Other astronomers, basing their views on Triton's chemical composition, maintain that it formed as a "normal" moon but was later kicked into its abnormal orbit by some catastrophic event, such as an interaction with another similar-sized body. It has even been suggested that the planet Pluto may have played a role in this process, although no really convincing demonstration of such an encounter has ever been presented. The surface deformations on Triton certainly suggest fairly violent and relatively recent events in the moon's past. However, they were most likely caused by the tidal stresses produced in Triton as Neptune's gravity circularized its orbit and synchronized its spin, and they give little indication of the processes leading to the orbit in the first place.

Whatever its past, Triton's future is fairly clear. Because of its retrograde orbit, the tidal bulge Triton raises on Neptune tends to make the moon spiral toward the planet rather than away from it (as our Moon moves away from Earth). Thus, Triton is doomed to be torn apart by Neptune's tidal gravitational field, probably in no more than 100 million years or so, the time required for the moon's inward spiral to bring it inside Neptune's Roche limit (see Chapter 12). By that time, it is conceivable that Saturn's ring system may have disappeared, so that Neptune will then be the planet in the solar system with spectacular rings!

THE RINGS OF URANUS

Back All of the jovian planets have rings. The ring system surrounding Uranus was discovered in 1977, when astronomers observed it passing in front of a bright star, momentarily dimming its light. Such a stellar occultation (see Figure 13.17) happens a few times per decade and allows astronomers to measure planetary structures that are too small and faint to be detected directly. The 1977 observation was actually aimed at studying the planet's atmosphere by watching how it absorbed starlight. However, 40 minutes before and after Uranus itself occulted the star, the flickering starlight revealed the presence of a set of rings. The discovery was particularly exciting because, at the time, only Saturn was known to have rings. Jupiter's rings went unseen until Voyager 1 arrived there in 1979, and those of Neptune were unambiguously detected only in 1989, by Voyager 2.

Figure 13.17 Occultation of starlight allows astronomers to detect fine detail on a distant planet. The rings of Uranus were discovered using this technique.

The ground-based observations revealed the presence of a total of nine thin rings. The main rings, in order of increasing radius, are named Alpha, Beta, Gamma, Delta, and Epsilon, and they range from 44,000 to 51,000 km from the planet's center. All of these lie within the Roche limit of Uranus, which is about 62,000 km. A fainter ring, known as the Eta ring, lies between the Beta and Gamma rings, and three other faint rings, known as 4, 5, and 6, lie between the Alpha ring and the planet itself. In 1986, Voyager 2 discovered two more even fainter rings, one between Delta and Epsilon and one between ring 6 and Uranus. The main rings are shown in Figure 13.18. More details on the rings are provided in Table 13-3.

Figure 13.18 The main rings of Uranus, as imaged by Voyager 2. All of the rings known before Voyager 2's arrival can be seen in this photo. From the inside out, they are 6, 5, 4, Alpha, Beta, Eta, Gamma, Delta, and Epsilon. Resolution is about 10 km, which is just about the width of most of these rings. The two rings discovered by Voyager 2 are too faint to be seen here.

Uranus's rings are quite different from those of Saturn. While Saturn's rings are bright and wide with relatively narrow gaps between them, the rings of Uranus are dark, narrow, and widely spaced. With the exception of the Epsilon ring, which is about 100 km wide, the rings of Uranus are all less than 10 km wide, and the spacing between them ranges from a few hundred to about a thousand kilometers. However, like Saturn's rings, all of Uranus's rings are less than a few tens of meters thick (that is, measured in the direction perpendicular to the ring plane).

The density of particles within the rings themselves is comparable to that found in Saturn's A and B rings. The particles that make up Saturn's rings range in size from dust grains to boulders, but in the case of Uranus, the particles show a much smaller spread--few if any are smaller than a centimeter or so in diameter. The ring particles are also considerably less reflective than Saturn's ring particles, possibly because they are covered with the same dark material as Uranus's moons. The Epsilon ring (shown in detail in Figure 13.19), the widest of the 11, exhibits properties a little like those of Saturn's F ring. It is slightly eccentric (its eccentricity is 0.008) and of variable width, although no braids are found. It also appears to be composed of ringlets.

Figure 13.19 A close-up of Uranus's Epsilon ring, showing some of its internal structure. The width of the rings averages 30 km; special image processing has magnified the resolution to about 100 meters.

Like the F ring of Saturn, Uranus's narrow rings require shepherding satellites to keep them from diffusing away. In fact, the theory of shepherd satellites was first worked out to explain the rings of Uranus, which had been detected by stellar occultation even before Voyager 2's Saturn encounter. Thus, the existence of the F ring did not come as quite such a surprise as it might otherwise have done! Presumably many of the small inner satellites of Uranus play some role in governing the appearance of the rings. Voyager 2 detected the shepherds of the Epsilon ring, Cordelia and Ophelia (see Figure 13.20). Many other, undetected shepherd satellites must also exist.

Figure 13.20 These two small moons, discovered by Voyager 2 in 1986 and named Cordelia and Ophelia, tend to "shepherd" Uranus's Epsilon ring, thus keeping it from diffusing away into space.

THE RINGS OF NEPTUNE

As shown in Figure 13.21 and presented in more detail in Table 13-4, Neptune is surrounded by four dark rings. Three are quite narrow, like the rings of Uranus, and one is quite broad and diffuse, more like Jupiter's ring. The dark coloration probably results from radiation darkening, as discussed earlier in the context of the moons of Uranus. All the rings lie within Neptune's Roche limit. The outermost (Adams) ring is noticeably clumped in places. From Earth we see not a complete ring, but only partial arcs--the unseen parts of the ring are simply too thin (unclumped) to be detected. The connection between the rings and the planet's small inner satellites has not yet been firmly established, although many astronomers believe that the clumping is caused by shepherd satellites.

<

Figure 13.21 Neptune's faint rings. In this long-exposure image, the planet (center) is heavily overexposed and has been artificially blotted out to make the rings easier to see. One of the two fainter rings lies between the inner bright ring and the planet. The other lies between the two bright rings.

While all the jovian worlds have ring systems, the rings themselves differ widely from planet to planet. This leads us to ask: Is there some "standard" way in which rings form around a planet? And is there a standard manner in which ring systems evolve? Or do the processes of ring formation and evolution depend entirely on the particular planet in question? If, as now appears to be the case, ring systems are relatively short-lived, their formation must be a fairly common event. Otherwise, we would not expect to find rings around all four jovian planets at once. There are also many indications that the individual planetary environment plays an important role in determining a ring system's appearance and longevity. Although many aspects of ring formation and evolution are now understood, it must be admitted that no comprehensive theory yet exists.

Surface Detail on Planet Pluto

Historical Observations of Pluto

Map of Pluto

By the end of the nineteenth century, observations of the orbits of Uranus and Neptune suggested that Neptune's influence was not sufficient to account for all of the irregularities in Uranus's motion. Further, it seemed that Neptune itself might be affected by some other unknown body. Following their success in the discovery of Neptune, astronomers hoped to pinpoint the location of this new planet using similar techniques. One of the most ardent searchers was Percival Lowell, a capable, persistent observer and one of the best-known astronomers of his day. (Recall that he was the leading proponent of the theory that the "canals" on Mars were constructed by an intelligent race of Martians--see Interlude 10-1.)

Basing his investigation primarily on the motion of Uranus (Neptune's orbit was still relatively poorly determined at the time), Lowell set about calculating where the supposed ninth planet should be. He searched for it, without success, during the decade preceding his death in 1916. Not until 14 years later did American astronomer Clyde Tombaugh, working with improved equipment and photographic techniques at the Lowell Observatory, finally succeed in finding Lowell's ninth planet, only 6° away from Lowell's predicted position. The new planet was named Pluto for the Roman god of the dead who presided over eternal darkness (and also because its first two letters and its astrological symbol are Lowell's initials). Its discovery was announced on March 13, 1930, Percival Lowell's birthday.

On the face of it, the discovery of Pluto looks like another spectacular success for celestial mechanics. Unfortunately, it now appears that the supposed irregularities in the motions of Uranus and Neptune did not exist and that the mass of Pluto, not measured accurately until the 1980s, is far too small to have caused them anyway. The discovery of Pluto owed much more to simple luck than to elegant mathematics!

Pluto's orbital semi-major axis is 39.5 A.U. (5.9 billion km). But unlike the paths of the other outer planets, Pluto's orbit is quite elongated, with an eccentricity of 0.25. It is also inclined at 17.2° to the plane of the ecliptic. Already we have some indications that Pluto is unlike its jovian neighbors. Because of its substantial orbital eccentricity, Pluto's distance from the Sun varies considerably. At perihelion, it lies 29.7 A.U. (4.4 billion km) from the Sun, inside the orbit of Neptune. At aphelion, the distance is 49.3 A.U. (7.4 billion km), well outside Neptune's orbit. Pluto last passed perihelion in 1989, and it will remain inside Neptune's orbit until 1999. Its sidereal period is 248.6 years, so the next perihelion passage will not occur until the middle of the twenty-third century. Pluto's orbital period is apparently exactly 1.5 times that of Neptune--the two planets are locked into a 3:2 resonance (two orbits of Pluto for every three of Neptune) as they orbit the Sun. The orbits of these two outer planets are sketched in Figure 13.22.

Figure 13.22 The orbits of Neptune and Pluto cross, although Pluto's orbital inclination and a 3:2 resonance prevent the planets from actually coming close to one another. Between 1979 and 1999, Pluto is inside Neptune's orbit, making Neptune the most distant planet from the Sun.

At nearly 40 A.U. from the Sun, Pluto is often hard to distinguish from the background stars. As the two photographs of Figure 13.23 indicate, the planet is actually considerably fainter than many stars in the sky. Like Neptune, it is never visible to the naked eye.

Figure 13.23 These two photographs, taken one night apart, show motion of the planet Pluto (arrow) against a field of much more distant stars. Most of Pluto's apparent motion in these two frames is actually due to the orbital motion of the Earth rather than that of Pluto.

Back Pluto is so far away that little is known of its physical nature. Until the late 1970s, studies of its reflected light variations suggested a rotation period of nearly a week, but measurements of its mass and diameter were very uncertain. All this changed in July 1978, when astronomers at the U.S. Naval Observatory discovered that Pluto has a satellite. It is now named Charon, after the mythical boatman who ferried the dead across the river Styx into Hades, Pluto's domain. The discovery photograph of Charon is shown in Figure 13.24(a). Charon is the small bump near the top of the image. Knowing the moon's orbital period of 6.4 days, astronomers could determine the mass of Pluto to much greater accuracy. It is 0.0025 Earth masses (1.5 × 1022 kg), far smaller than any earlier estimate--more like the mass of a moon than of a planet. In 1990, the Hubble Space Telescope imaged the Pluto-Charon system (Figure 13.24b). The improved resolution of that instrument clearly separates the two bodies and allowed even more accurate measurements of their properties.

Figure 13.24 (a) The discovery photograph of Pluto's moon, Charon. The moon is the small bump on the top right portion of the image. (b) The Pluto-Charon system, to the same scale, as seen by the Hubble Space Telescope. The angular separation of the planet and its moon is about 0.9 arc second.

Before Charon was discovered, Pluto's radius was also poorly known. Pluto's angular size is much less than 1´´, so its true diameter is blurred by the effects of Earth's turbulent atmosphere. But Charon's orbital orientation has given astronomers new insight into the system. By pure chance, Charon's orbit over the 6-year period from 1985 to 1991 (less than 10 years after the moon was discovered) has produced for Earth viewers a series of eclipses. Pluto and Charon repeatedly passed in front of one other, as seen from our vantage point. Figure 13.25 sketches this orbital configuration. With more good fortune, these eclipses took place while Pluto was closest to the Sun, making for the best possible Earth-based observations.

Figure 13.25 The orbital orientation of Charon produced a series of eclipses between 1985 and 1991. Observations of eclipses of Charon by Pluto and of Pluto by Charon have provided detailed information about both bodies' sizes and orbits.

Basing their calculations on the variations in light as Pluto and Charon periodically hid each other, astronomers have computed their masses and radii and have determined their orbit plane. Additional studies of sunlight reflected from Pluto's surface indicate that the two objects are tidally locked as they orbit each other. Pluto's diameter is 2250 km, about one-fifth the size of the Earth. Charon is about 1300 km across and orbits at a distance of 19,700 km from Pluto. If planet and moon have the same composition (probably a reasonable assumption), Charon's mass must be about one-sixth that of Pluto, giving the Pluto-Charon system by far the largest satellite-to-planet mass ratio in the solar system. As shown in Figure 13.26, Charon's orbit is inclined at an angle of 118° to the plane of Pluto's orbit around the Sun. Since the spins of both planet and moon are perpendicular to the plane of Charon's orbit around Pluto, the geographic "north" poles of both bodies lie below the plane of Pluto's orbit. Thus, Pluto is the third planet in the solar system found to have retrograde rotation.

Figure 13.26 Charon's path around Pluto is circular, synchronous, and inclined at 118° to the orbit plane of the Pluto-Charon system about the Sun. The Pluto-Charon orbit plane is itself inclined at 17° to the plane of the ecliptic.

The known mass and radius of Pluto allow us to determine its average density, which is 2300 kg/m3--too low for a terrestrial planet, but far too high for a mixture of hydrogen and helium of that mass. Instead, the mass, radius, and density of Pluto are just what we would expect for one of the icy moons of a jovian planet. In fact, Pluto is quite similar in mass and radius to Neptune's large moon, Triton. The planet is almost certainly made up mostly of water ice. In addition, spectroscopy reveals the presence of frozen methane as a major surface constituent. Pluto is the only planet in the solar system on which methane exists in the solid state, implying that the surface temperature on Pluto is no more than 50 K. Pluto may also have a thin methane atmosphere, associated with the methane ice on its surface. Recent computer studies indicate that Charon may have bright polar caps, but their composition and nature are as yet unknown.

Because Pluto is neither terrestrial nor jovian in its makeup, and because of its similarity to the ice moons of the outer planets, some researchers suspect that Pluto is not a "true" planet at all. Pluto may be an escaped planetary moon or a large icy chunk of debris left over from the formation of the solar system. This idea is bolstered by Pluto's eccentric, inclined orbit, which is quite unlike the orbits of the other known planets. Since 1978, the explanation of Pluto's origin has been greatly complicated by the presence of Charon. It was much easier to suppose that Pluto was an escaped moon before we learned that it had a moon of its own. There is still no clear or easy answer to the puzzle of Pluto's origin.

Pluto may be just what it seems--a planet that formed in its current orbit, possibly even with its own moon right from the outset. Because we know so little about the environment in the outer solar system, we cannot rule out the possibility that planets beyond Neptune should simply look like Pluto. There is evidence for large chunks of ice circulating in interplanetary space beyond the orbits of Jupiter or Saturn (see Chapter 14), and some researchers have even suggested that there might have been thousands of Pluto-sized objects initially present in the outer solar system. The capture of a few of these objects by the giant planets would explain the strange moons of the outer worlds, especially Triton. And if there were enough moon-sized chunks originally orbiting beyond Neptune, it is quite plausible that Pluto could have captured Charon following a collision (or near-miss) between the two. At present, our scant knowledge of the compositions of the two bodies does not allow us to confirm or disprove either the coformation or the capture theory of the Pluto-Charon system. Unfortunately, this uncertainty may persist for some time--there is no present or proposed space mission that might suddenly and radically improve our understanding of these distant worlds.