LCROSS moon impactFAJARDO, Puerto Rico—"We could have just stayed in bed" was one comment I overheard this morning from planetary scientists who had woken up early to see NASA's Lunar Crater Observation and Sensing Satellite (LCROSS) crash into the lunar surface. At 7:31 A.M. (Eastern Daylight Time) the spacecraft's Centaur-class rocket booster slammed (deliberately) into the moon, the idea being to set off a dramatic explosion and send up a plume of material that telescopes could analyze for signs of water in the lunar soil. The spacecraft flew through the plume and, four minutes later, itself crash-landed. To watch the event, planetary scientists at the annual Division for Planetary Sciences meeting here had crowded into an over-air-conditioned ballroom and poured themselves insipid hotel coffee. As the designated time approached, the moon's surface loomed larger and larger as the spacecraft closed in [photograph at left].

At 7:30 the room went silent as everyone prepared to cheer wildly as the camera showed a bright flash. But the time came and went, and there was no flash, and no cheers. "It's hard to know what we saw there," admitted Mike Bicay, science director at NASA Ames Research Center in Moffett Field, Calif. Shortly before the spacecraft itself hit, word came through that the infrared camera had indeed seen a thermal signature of the booster's crater. This comment was barely audible, though, over the bemused laughter as images of the mission control center showed one controller conspicuously failing to respond to another's high five.

Although the event was something of a letdown, all this will surely be forgotten when the data comes in and LCROSS adds to the string of recent discoveries showing that the moon is not quite the bone-dry world that scientists once assumed.

People often ask me what I think the biggest news of the meeting has been. For me, these meetings—and science generally—aren't about the "biggest news." Instead, I come to keep tabs on the broader trends, watching the slow and patient accumulation of knowledge. Yesterday, for example, I attended an engrossing session on the formation of planets and their satellites. The field has inched forward since I started to attend these meetings 20 years ago, yet many of the same questions remain. If planet formation is a mystery, satellite formation is a mystery upon a mystery. Starting from a disk of material orbiting a planet, a satellite can coalesce extremely rapidly—in just a few thousand years. That is faster than the disk itself forms, which seems contradictory. Moreover, why are Jupiter's and Saturn's satellite systems so different? Jupiter has four large satellites fairly close to the planet (Io, Europa, Ganymede, and Callisto), whereas Saturn has only one giant satellite (the distant Titan). And why does Saturn have such a huge retinue of runty moons?

Bill Ward of the Southwest Research Institute in Boulder, Colo., has been working with his colleague Robin Canup on reconciling satellite and disk formation times. Their idea is that disk and satellite formation are not sequential but concurrent: they occur together as material dribbles in from the broader disk of material orbiting the sun. The inward dribble slows down satellite formation so that it never gets out of whack with the disk. In his talk yesterday, Ward argued that the requisite conditions naturally occur toward the end of planet-building.

Takanori Sasaki of the Tokyo Institute of Technology followed this up with an elegant explanation for the differences between the Jovian and Saturnian systems. Perhaps Saturn did once have multiple large satellites but rudely cannibalized them. Sasaki and his colleagues proposed that Jupiter's strong magnetic field opened up a hole in the surrounding disk, whereas Saturn's disk extended all the way down to the planet. Jupiter's disk thus looked like a doughnut and Saturn's like a saucer. One by one, satellites coalesced in the outer reaches of the disk and spiraled inward because of drag forces. In Jupiter's case, satellites stopped migrating when they reached the inner edge of the disk, and they began to queue up there. In Saturn's case, there was nothing to stop the hapless satellites from falling all the way in. Titan avoided this fate only because migration ceased as the disk dissipated.

As for Saturn's family of little moons, Sébastien Charnoz of Paris Diderot University gave a talk entitled "Are Saturn's small moons the children of Saturn's rings?" Scientists often think of the rings as a moon that broke up or at least failed to form—the rings reside inside the so-called Roche limit, the distance within which a planet's tidal forces rip apart any moon. Charnoz and his colleagues inverted this thinking. In their model, as the rings spread out, material exits the Roche limit and coalesces into moons. Once a moon forms, its gravitational interaction with the rings causes it to spiral outward, opening up room for another moon, and another. The outer edge of the rings becomes an assembly line of moons. Eventually the moons' interaction with the rings becomes so strong that it shut downs the assembly line. This idea not only explains the moons, but also Saturn's funky F ring: it may be a product of material coming off the innermost satellites.

With work like this, scientists are gradually piecing together a comprehensive view of solar system formation. Now if they can only prevail on their conference hotels to serve better coffee.


From carbon planets to the lakes of Titan: Dispatch from the annual planets meeting

What caused Saturn to lurch? Second dispatch from the annual planets meeting

Planetary bombardments, past and future: Third dispatch from the annual planets meeting

View of the moon from LCROSS as the probe neared impact: NASA