14-7 Uranus’s larger and smaller satellites

Before the Voyager 2 flyby of Uranus, five moderate-sized satellites—Titania, Oberon, Ariel, Umbriel, and Miranda—were known to orbit the planet (Figure 14-13). These are Uranus’s largest moons. Most are named after sprites and spirits in Shakespeare’s plays. They range in diameter from about 1600 km (1000 mi) for Titania and Oberon to less than 500 km (300 mi) for Miranda. These moons have average densities around 1500 kg/m³, which is consistent with a mixture of half water-ice and half rock (although Miranda is mostly ice).

Figure 14-13: R I V U X G
Uranus’s Principal Satellites This “family portrait” (a montage of five Voyager 2 images) shows Uranus’s five largest moons to the same scale and correctly displays their respective reflectivities. (The darkest satellite, Umbriel, is actually more reflective than Earth’s Moon.) All five satellites have grayish surfaces, with only slight variations in color.

(NASA/JPL-Caltech/R. Hurt [SSC])

Voyager 2 discovered 11 other small Uranian satellites, most of which have diameters of less than 100 km (60 mi); 11 more were found using ground-based telescopes and the Hubble Space Telescope between 1997 and 2005. Only Jupiter and Saturn have more known satellites than Uranus. In the infrared, Uranus, its rings, and some of its moons are easily visible (Figure 14-14).

Figure 14-14: R I V U X G
Uranus’s Rings and Moons (a) This false-color infrared image from the Hubble Space Telescope shows eight of Uranus’s inner moons, all of which were discovered by Voyager 2 when it flew past Uranus in 1986. They all lie within 86,000 km of the planet’s center (only about one-fifth of the distance from Earth to our Moon). The arcs show how far each moon moves around its orbit in 90 minutes. Note the pole and equator; unlike the other planets, Uranus nearly lies on its side with its rotational axis close to its orbital plane around the Sun. (b) This infrared image from the Very Large Telescope shows five of Uranus’s larger moons. All the moons visible in this image are in prograde orbit, orbiting in the same direction as Uranus rotates. The rings in this image appear even brighter than Uranus: Gaseous methane in Uranus’s atmosphere absorbs infrared light, while ice in the rings reflects this wavelength.

(a: Erich Karkoschka, University of Arizona; and NASA, b: European Southern Observatory)

The moons of Uranus are generally divided into three groups: the larger moons, the smaller inner moons orbiting near the rings, and the smaller outer moons that are probably captured asteroids.

Uranus’s Larger Satellites

Umbriel and Oberon both appear to be geologically dead worlds, with surfaces dominated by impact craters. By contrast, Ariel’s surface appears to have been cracked at some time in the past, allowing some sort of ice lava to flood low-lying areas. A similar process appears to have taken place on Titania. This geologic activity may be due to a combination of the satellites’ internal heat and tidal heating like that which powers volcanism on Jupiter’s satellite Io (see Section 13-4). Io’s tidal heating is only possible because of the 1:2:4 ratio of the orbital periods of Io, Europa, and Ganymede. While there are no such simple ratios between the present-day orbital periods of Uranus’s satellites, there may have been in the past. If so, tidal heating could have helped reshape the surfaces of some of the satellites.

Unique among Uranus’s satellites is Miranda, which has a landscape unlike that of any other world in the solar system. Much of the surface is heavily cratered, as we would expect for a satellite only 470 km in diameter, but several regions have unusual and dramatic topography (Figure 14-15). Detailed analysis of Miranda’s geology suggests that this satellite’s orbital period was once in a whole-number ratio with that of more massive Umbriel or Ariel. The resulting tidal heating melted Miranda’s interior, causing dense rocks in some locations on the surface to settle toward the satellite’s center as blocks of less dense ice were forced upward toward the surface, thus creating Miranda’s resurfaced terrain. Tidal heating must have ceased before this process could run its full course, which explains why some ancient, heavily cratered regions remain on Miranda’s surface.

Figure 14-15: R I V U X G
Miranda This composite of Voyager 2 images shows that part of Miranda’s surface is ancient and heavily cratered, while other parts are dominated by parallel networks of valleys and ridges. At the very bottom of the image—where a “bite” seems to have been taken out of Miranda—is a range of enormous cliffs that jut upward to an elevation of 20 km, twice as high as Mount Everest.

(NASA/JPL)

Uranus’s Larger Satellites Constrain Models of an Axis-Tilting Impact

The five large moons of Uranus help us understand the collision that is proposed to have tilted the rotation axis of Uranus. A collision large enough to turn a planet on its side would be quite a violent event. Such an impactor is estimated to be several times the mass of Earth—quite large!—though there might instead have been several smaller impacts. While several impacts might seem less likely to occur, modeling indicates that a single large impact would have resulted in retrograde orbits for Uranus’s large moons. On the other hand, two or more smaller impactors, each the size of Earth, would better explain the observed prograde orbits of the large Uranian moons.

Another constraint on the model of impacts with Uranus is that if these impacts occurred after the large moons formed, the moons would have remained in their original orbital plane, which is assumed to be in Uranus’s original equatorial plane. Instead, what we see today is both Uranus and the orbital plane of its moons tilted together, so that the larger moons are still in the equatorial plane (see Figure 14-14b). In order to tilt Uranus and its moons together, models suggest that impacts might have rotated a disk of protoplanetary material from which both Uranus and its moons formed. While frequent impacts are expected during the early solar system, impacts of this magnitude are not expected, and the origin of Uranus’s tilt is still an open question.

Uranus’s Perplexing Small Satellites

While Uranus’s small satellites are unlikely to show the kind of geology found in Miranda, some of them move in curious orbits. Eight of the nine smaller outer satellites in large orbits beyond Oberon are in retrograde orbits, moving around Uranus opposite to the direction in which Uranus rotates. Many of the small outer satellites of Jupiter and Saturn have retrograde orbits and are probably captured asteroids (see Section 13-9 and Section 13-10); the same is probably true of the outer satellites of Uranus.

The 13 inner satellites that orbit closest to Uranus (inside the orbit of Miranda) are all in prograde orbits, so they travel around Uranus in the same direction as the planet rotates (Figure 14-14a). Just like the rings of Uranus, these inner moons are very dark; this is probably due to the same process of radiation darkening. In fact, it is thought that the Uranian rings were probably created by the fragmentation of one or more inner moons.

Will the inner moons orbit Uranus indefinitely? Probably not. When Mark Showalter and Jack Lissauer compared 11 of these satellites’ orbits in 2005 with their orbits in 1994, they found surprisingly large differences. (The other two inner satellites were only discovered in 2003, so it is not known whether their orbits underwent similar changes.) These satellites can pass within a few thousand to a few hundred kilometers from each other, so they can exert strong gravitational forces on each other. Over time these forces can modify their orbits. Computer simulations of these gravitational interactions indicate that the inner satellites might begin colliding with each other within a hundred million years. If these inner satellites are in such unstable orbits, have they really been in orbit around Uranus since it formed more than 4.56 billion years ago? Or did Uranus somehow acquire these satellites in the relatively recent past?