A NASA spacecraft that in 2005 discovered watery plumes spewing from the surface of Saturn's icy moon Enceladus has now found compelling evidence that the plumes stem from an underground reservoir of saltwater.

The Cassini probe in 2008 and 2009 flew through a towering plume emanating from the moon's southern polar region and sampled its contents. In an analysis published online June 22 in Nature, a team of researchers reports that the composition of the plume is most easily explained by a sizable subterranean body of water. (Scientific American is part of Nature Publishing Group.) Cassini's instruments were not designed to make such measurements—and in fact the mission was supposed to have ended before the flyby took place—but with a postponed retirement and a few on-the-fly software tweaks the versatile spacecraft was able to get a whiff of the geyserlike ejecta.

The plumes have since their discovery been known to be rich in water vapor, but their origin has remained unsettled. Even in the absence of a liquid reservoir belowground, water vapor could stem from some of Enceladus's abundant ice sublimating directly to vapor in the vacuum of space or from the breakdown of hydrated solids called clathrates.

But whereas a liquid reservoir in contact with the moon's rocky core should contain dissolved salts that would be injected into an upwelling geyser, sublimating ice or decomposing clathrates would be much less efficient at producing a salty plume. Planetary scientist Frank Postberg of Heidelberg University and the Max Planck Institute for Nuclear Physics in Heidelberg, Germany, and his colleagues gathered some support for the saltwater hypothesis in 2009 when they showed that some particles in Saturn's diffuse E ring were salt-rich. The E ring is fed by Enceladus's plumes, so the implication was that the salty grains originated in the icy moon's hypothesized subterranean ocean and were ejected into the ring as a kind of frozen sea spray.

But with salty grains constituting only a small percentage of the E ring particles, the ocean hypothesis was hardly a lock. In the new analysis of Cassini's dives through the plumes Postberg and his co-authors found a much greater salt concentration—almost all the particles near the source of the plumes are salty ices. It now becomes much more difficult to explain Enceladus's eruption without invoking a large underground reservoir. "Over 99 percent of the emitted ice being salt-rich, that makes a much stronger case [for an ocean], and it's not in agreement with ice sublimation," Postberg says. "Now, with 99 percent, we know that it's just not plausible to be coming from a solid." The salt-rich grains are heavy and tend to fall back to the surface, explaining their relative paucity in Saturn's E ring compared with lighter, salt-free particles.

"They got a sniff of these salty ice grains when they flew through the E ring," says planetary scientist Francis Nimmo of the University of California, Santa Cruz, who did not contribute to the new study. "Now that sniff has become—practically everything is salty. It makes the case that these grains are coming from some liquid reservoir kind of inescapable."

The icy moon, just 500 kilometers in diameter, could be one of several moons in the solar system to harbor underground stores of liquid water. Some evidence has hinted at a subterranean ocean for Titan, a much larger Saturnian moon, as well as for Ganymede, Callisto and Europa, three of Jupiter's largest satellites, and for Neptune's moon Triton.

But just what Enceladus's reservoir would look like is somewhat uncertain. The salt content implies a body of water in contact with the moon's rocky core, which Enceladus's density indicates is dozens of kilometers below the surface, but the escaping vapor at the surface points to evaporation at much shallower subterranean depths. One possibility is a series of near-surface misty caverns fed by a saltwater ocean at Enceladus's core. "You have an ocean at depth at the interface of the ice and the rocky core," Postberg says. "But it must be connected with reservoirs that are only a few hundred meters below the surface."

The scale and complexity of that hypothesized plumbing raises some questions. "These 'deep misty caverns' must be truly immense, and connected in complicated ways with the ocean and with the surface," says Nicholas Schneider, a planetary scientist at the University of Colorado at Boulder. The detection of salt in the plumes is indeed consistent with a liquid source, Schneider says, but geophysicists now need to come up with a viable description of a watery internal structure for the satellite. "After all, we're really using the plumes to tell us what's going on inside, and nobody's taken up that challenge," he says. "We're watching what little Enceladus spits up, but that hardly tells us much about the baby's insides!"

Another question is how a tiny, icy satellite like Enceladus could maintain a large body of liquid water. The tidal energy generated by Enceladus's orbit around Saturn provides some heating, but not enough to keep a large amount of water from freezing over billions of years. "The big question that we still don't have answered is: How can an ocean survive for geological time?" Nimmo says. "Most likely the answer is it's not a global ocean at all but more of a regional sea." In other words, Enceladus's tidal heat could be concentrated on the south polar region, allowing for a localized reservoir of liquid there on an otherwise frozen moon.

Perhaps Cassini, which has been exploring Saturn since 2004, will deliver more answers about the mysterious ice world in the coming years. The spacecraft's mission, originally set to end in 2008, has been extended through 2017.