The bottom of the North Atlantic Ocean midway between Bermuda and the Canary Islands must rank highly on any list of unlikely places to find a bustling city. Yet there, in the darkness that reigns nearly a kilometer below the sunlit surface, nature has built an undersea metropolis, a complex of limestone towers as tall as skyscrapers that is home to masses of snails, crabs and mussels. The towers form as minerals precipitate out of warm, alkaline water jetting from hydrothermal vents along the ocean floor. Biologists using submersibles and remote cameras found this exotic “Lost City” in the early 2000s and have been studying it ever since to learn how hydrothermal vents can sustain thriving ecosystems so far from the life-giving light of the sun. In the meantime, planetary scientists using the Cassini space probe have made a revolutionary series of related discoveries in the outer solar system, finding strong evidence that hydrothermal vents much like those of Lost City exist not only on Earth but also in the mysterious subsurface ocean of a small, icy Saturnian moon called Enceladus. Could life exist there, too?

Naturally, the possibility of extraterrestrial life tantalizes scientists, but they would have been excited about otherworldly hydrothermal vents even without aliens dangling in front of their eyes. The same evidence that suggests hydrothermal activity on this faraway moon is also providing critical information about the composition and longevity of Enceladus's ocean. Those secrets might otherwise be forever hidden underneath a frozen crust, as they might be for other ocean-bearing moons currently lacking strong evidence for hydrothermal activity, such as Jupiter's Europa.

More fundamentally, the very existence of Enceladus's hydrothermal vents poses an irresistible puzzle. Other than water, the most important ingredient for hydrothermal activity is obviously heat, but this icy moon's sizzling innards are not easily explained. Enceladus is roughly the diameter of England—relatively runty for a moon and far too small to hold onto the primordial heat left over from its formation. Some other source of warmth must be at work deep within. Learning how Enceladus generates and sustains its toasty interior could revolutionize our understanding of icy moons—and their prospects for life.

First Hints

Scientists began to suspect an ocean within Enceladus in 2005, about a year after Cassini's arrival in the Saturnian system, when the spacecraft observed a huge plume of water vapor and ice grains rising hundreds of kilometers into space from tectonically active terrain around the moon's south pole. In a series of subsequent flybys, Cassini traced the plume to multiple jets emanating from four linear fissures so much warmer than their frigid surroundings that they glowed in infrared. Mission scientists called the fissures “tiger stripes” and pinpointed their jets as the source of the tenuous, sprawling ring of icy particles—known as the E ring—that encloses Saturn's classical ring system. Most of the jets' ice grains, however, travel too slowly to reach this ring and instead fall back to Enceladus as a fine-powdered snow. Based on the 100-meter-high snowdrifts that cover parts of its southern hemisphere, researchers estimate that Enceladus has been venting water into space for 10 million years or more.

Although the “ocean hypothesis” for Enceladus's jets was initially controversial, an extended series of studies by Cassini has now irrefutably confirmed that a global deep sea hides inside the moon. Most recently, an analysis of Enceladus's gravitational field, surface topography and slight rotational wobble by Ondrej Cadek of Charles University in Prague and his collaborators, including one of us (Tobie), placed the best limits yet on the ocean's size and extent. Their work suggests that the crust must be about 35 kilometers thick near Enceladus's equator but less than five kilometers thick in and around the south pole. Indeed, Alice Le Gall of the University of Versailles Saint-Quentin in France and her collaborators recently detected unusually large thermal microwave emissions at the south polar terrain, indicating an ice shell thickness of only a few kilometers. The ocean floor would lie about 60 to 65 kilometers under the surface, which would mean Enceladus's sea holds about a tenth as much water as the Indian Ocean. And based on data gathered by Cassini in 2009 and 2011, one of us (Postberg) has shown that the water is alkaline and salty, containing sodium chloride—standard table salt—in its ejected plume. This means the ocean most likely lies atop (and leaches minerals from) the moon's rocky core.

The crucial evidence for the presence of hydrothermal vents was gathered beginning in 2004 by Cassini's Cosmic Dust Analyzer (CDA), before the spacecraft even arrived at Saturn and discovered Enceladus's plume. As Cassini approached Saturn from interplanetary space, unexpected showers of microscopic, fast-moving nanoparticles struck the CDA like buckshot. Years later, after the discovery of the plume, Postberg examined the distribution of the nanoparticle sizes and frequency in the CDA data, finding none of them to be larger than 20 nanometers and all of them to have a composition most consistent with that of essentially pure silicon dioxide—silica, the main constituent of quartz rock and beach sand. Using numerical simulations to trace the most probable trajectories for the silica nanoparticles, Hsiang-Wen Hsu of the University of Colorado Boulder surmised that they had originated in the outer reaches of the E ring. Because we knew that Enceladus produces the E ring, this finding strongly suggested that the nanoparticles had come from the icy moon. Their composition proved to be the smoking gun for discovering Enceladus's hydrothermal activity.

Enceladus (center), seen from two million kilometers away, is embedded in Saturn's E ring, which is formed from the moon's icy plume. Credit: NASA/JPL and Space Science Institute

Pure silica jetting from Enceladus was a surprise because its only plausible source would be deep underneath the ice and the ocean, in the moon's rocky core, where silicon mostly exists in minerals chemically bound with other elements such as iron and magnesium. Collisional grinding of those minerals—the rough-and-tumble shattering of rock to make ever smaller pieces—might conceivably create silica nanoparticles. Yet such particles would come in a wide range of sizes, not the very narrow range Cassini observed. Only one other natural explanation remained: the nanoparticles could have crystallized from a supersaturated, silica-rich solution of hot, alkaline water flowing through rock—that is, from hydrothermal vents of exactly the kind found at Lost City on Earth.

A Habitable Ocean?

At Lost City, and presumably on the seafloor of Enceladus, hot water absorbs silica as it flows up through silicate rocks. As the water vents out into the surrounding sea and cools, its capacity to carry absorbed minerals diminishes, and silica nanoparticles form. At this stage, other molecules could glom onto the nanoparticles, making them larger and heavier and thus causing them to eventually precipitate to the bottom—unless, that is, the water were alkaline and not too salty. This relation between the size and longevity of nanoparticles and the temperature and chemistry of their aqueous birthplaces offers researchers an unprecedented window into the environmental conditions of Enceladus's ocean.

Following on Cassini's initial detection of the nanoparticles, a team led by Yasuhito Sekine of the University of Tokyo conducted laboratory experiments to confirm how the nanoparticles formed and to reveal the conditions deep within Enceladus. The scientists found that water at or above 90 degrees Celsius with alkalinity above and salinity slightly below the value of Earth's sea is ideal for creating small, long-lived silica nanoparticles. According to their experiments, the alkalinity of Enceladus's ocean must be between that of terrestrial seawater and ammonia-based household cleaning products. If it were more alkaline than aqueous ammonia, the water's high silica solubility would not allow nanoparticles to form. If the water were less alkaline than seawater on Earth, it would have to be inconceivably hot to dissolve sufficient amounts of silicon dioxide to form silica nanoparticles. Altogether Hsu's, Postberg's and Sekine's respective work raises the possibility that the rich ecosystems of Lost City and other terrestrial hydrothermal vents could conceivably survive and thrive if relocated to the depths of Enceladus. In other words, the ocean of this distant icy moon looks like it could be habitable.

Of course, it could be that Enceladus is currently inhospitable for life and that the silica nanoparticles detected by Cassini are simply relics of ancient hydrothermal activity that ceased long ago. But the work of Sekine and other collaborators suggests this is not the case. In lab experiments and numerical models, freshly formed silica particles average about four nanometers wide, only growing larger over time spans of at least a few months and, at most, a few years. The CDA data show that the typical nanoparticle from Enceladus is between four and 16 nanometers wide, with none wider than 20. Hence, the nanoparticles collected by Cassini must have been created only a short time before being measured. Otherwise, they would have been larger than observed.

From Seafloor to Deep Space

We can now trace a typical nanoparticle's journey from the bottom of Enceladus's buried sea out to the wider solar system. After forming at the cooling edges of hot, silica-rich fluids gushing into the cold surrounding ocean, a nanoparticle will spend as much as a few years drifting up through about 60 kilometers of open water.

When it reaches the top of the ocean, the nanoparticle ascends into water-filled fractures that crisscross through the few-kilometer-thick overlying frozen crust of the south polar terrain. Because the seawater is denser than the surrounding ice, its progress upward should halt less than a kilometer underneath Enceladus's surface. But here the so-called champagne effect provides a further boost: as the water, which contains dissolved carbon dioxide and other gases, rises, and the pressure on it decreases, it becomes fizzy with bubbles. The bubbles help to lift the seawater within perhaps 100 meters of Enceladus's surface. Tidal forces may also help the process by repeatedly opening and closing the cracks.


There, we suspect, temporary pools of gas-saturated liquids loaded with suspended particles throw off clouds of ocean mist from bursting bubbles. In such close proximity to the harsh vacuum of space, water evaporates efficiently, although it is near its freezing point. The vapor is dragged into the vacuum and, as if through a chimney, rises up through cracks in the near-surface ice into space. Along with the vapor, the ocean spray from the effervescing pools is carried upward and quickly freezes into micron-sized ice grains, which incorporate the silica nanoparticles like raisins in a bun. Some of the vapor freezes on the walls of ice, releasing latent heat that we see as the infrared glow of Enceladus's surface tiger stripes. The vapor that does not freeze carries the nanoparticle-laden grains up to the surface, hurling them into space as icy fountains.

Most of the grains in the plume fall back to the surface as snow, but those with the highest speeds escape Enceladus to accumulate in the E ring. In the E ring, ionized gas erodes the ice grains and frees the embedded nanoparticles. The liberated nanoparticles then accumulate an electric charge from the ionized gas and free electrons and become the playthings of Saturn's immense electromagnetic fields. Finally, boosted by solar winds, some of the nanoparticles reach velocities up to a million kilometers per hour—about 1 percent of light speed—and zip off into the solar system. A small fraction of the escapees may even reach interstellar space, to surf the voids between the stars.

Hot Topics

As elaborate, beautiful and, we believe, true as this narrative is, it does not address what has become the central conundrum of Enceladus: What is the source of the internal heat required to maintain its dynamic ocean? That heat, which is essential for liquid water and life, obviously cannot come from sunlight. The sun's rays are about 99 percent weaker at Enceladus than in Earth's vicinity, giving the icy moon a surface temperature about that of liquid nitrogen.

About half of Earth's internal warmth comes from the slow decay of radioactive isotopes of uranium, thorium and potassium. This radiogenic heating has sustained temperatures exceeding several thousands of degrees C in Earth's interior for billions of years. Although Enceladus probably contains comparable concentrations of radiogenic elements, at only 500 kilometers wide, the diminutive moon loses its internal heat much more efficiently than Earth does. In the absence of an additional heat source, Enceladus's interior should be frozen solid. The moon's small size and weak gravity also make its internal dynamics very different from bulky planets like Earth: lower pressures and more moderate temperatures within Enceladus limit the compaction and consolidation of material in its core, allowing water to circulate down through porous rock to create hydrothermal processes in the moon's very center. Contrast this with Earth, where the rapid subterranean increases in pressure and temperature limit water circulation to the top few kilometers of crust.

One might expect the flushing of Enceladus's core to hasten its cooling, sweeping away any radiogenic heat and precluding the high temperatures required to make silica nanoparticles. Yet there is another possible energy source beyond standard radiogenic heating that could explain the moon's present-day hydrothermal activity: tidal heating.

Similar to how Earth's ocean tides arise from our moon and the sun pulling on our planet, tidal heating occurs when a planet's or moon's interior periodically flexes as it moves through a noncircular, eccentric orbit. The flexure from shifting gravitational forces creates friction in a planet's or moon's inner layers, which in turn creates heat. Tidal heating would be particularly powerful within the porous, water-suffused core of a world such as Enceladus. And indeed, data from Cassini clearly show that Saturn's tidal pull profoundly influences the tiny moon—the brightness of its erupted jets and therefore the amount of material ejected vary periodically as the moon whirls around the ringed planet. Evidently, the chimneylike cracks that serve as conduits for mist and water vapor through the ice are at turns squeezed together and pulled apart by tidal tugs that also generate significant amounts of heat.

The Turning of the Tide

What we do not know is whether the ocean we observe today is a transient phenomenon persisting for only tens of millions of years or a sustained feature of the moon that has endured for hundreds of millions or even billions of years. The answer depends on how long the action of tides has heated up Enceladus's interior, which in turn depends on how the moon influences Saturn, as well as its lunar neighbor Dione.

To understand these tidal interactions, we can consider the familiar system of our Earth and moon, which bears some similarities to that of Saturn and Enceladus. Our moon raises tides on Earth, and Enceladus does the same to Saturn. In Earth's ocean, these tidal flows gradually dissipate because of friction against coastlines and the seafloor—an effect that measurably slows Earth's rotation. One hundred years from now, the day will be two milliseconds longer than it currently is, and Earth will have sapped enough of lunar tidal energy to push the moon's orbit out by nearly four meters. Similarly, tidal friction in Saturn's interior infinitesimally affects the giant planet's rotation while increasing Enceladus's distance from Saturn and orbital eccentricity. Higher eccentricities translate into larger tidal effects—and thus more heating—within Enceladus. Early theoretical estimates suggested that Enceladus would generate only weak tidal friction within Saturn, leading to the moon's orbit losing its eccentricity and thus limiting the lifetime of any tidal-heated ocean to no more than one million years.

Hydrothermal fluids percolating through Enceladus's hot interior could potentially form seafloor mineral deposits, as in this artist's conception. Credit: Ron Miller

Recently Valéry Lainey of the Paris Observatory and his collaborators (including Tobie) performed detailed analyses of the motions of Saturn's large moons to place more accurate constraints on the magnitude of tidal friction in the giant planet's interior. They found that the tidal friction within Saturn is at least 10 times larger than predicted by previous models. If true, this larger figure would mean that Enceladus's orbital eccentricity is stable and long-lived, allowing strong tides that can sustain an ocean for at least tens of millions of years and potentially much, much longer. The longer Enceladus's ocean can persist, it would seem, the greater the likelihood of life's emergence and flourishing there.

Brave New Underworlds

In the meantime, there is a second possible heat source to consider beyond tidal heating. When water percolates through silicate rock, it can hydrate and change the crystalline structure of certain minerals, releasing substantial amounts of heat in a process called serpentinization. Bolstered by the ready circulation of water through the moon's porous, silicate-rich rocky core, serpentinization could generate several gigawatts of power and be a crucial part of Enceladus's internal heat budget. As long as fresh, unaltered minerals are in contact with circulating water, this heat supply will persist. But as the rock becomes fully serpentinized over millions of years, it will cease producing heat and should cool down in the absence of other influences, such as tidal friction. So it seems that serpentinization alone can hardly sustain a global ocean long enough to allow prebiotic chemistry to evolve.

Even so, serpentinization could still contribute to possible biospheres within Enceladus's depths. On Earth, scientists have observed serpentinization processes powering the hydrothermal vents at Lost City and other undersea sites. Besides making heat, these reactions also produce hydrogen, methane and other organic compounds that sustain the microbes that form the base of the food chain for their isolated, sun-starved ecosystems. Studying such organisms, some researchers have wondered whether life really needs the sun at all.

The idea that similar scenarios are occurring within Enceladus's water-filled core recently got a substantial boost: during the last close flyby on October 28, 2015, Cassini dove deeper into the plume than ever before to allow its ion and neutral mass spectrometer (INMS) to look for hydrogen. In April 2017 the INMS team reported that it indeed detected hydrogen, which makes up about 1 percent of the plume vapor, supporting the idea that serpentinization happens deep in the moon's interior. Serpentinization likely does not contribute substantially to the total heat budget, but the measured ratio of hydrogen and methane in the plume indicated that the so-called methanogenesis reaction could provide a source of chemical energy at hydrothermal sites.

In the late 1980s Michael Russell, then at the University of Strathclyde in Scotland, and his collaborators hypothesized that alkaline hydrothermal vents might have been the birthplace of the first living organisms on the early Earth. Although none were yet known on Earth, Russell argued that such sites would offer a relatively benign yet energy-rich environment in which prebiotic chemistry could brew to form the precursors of modern membranes, metabolisms and self-replicating molecules. Few people took the idea seriously enough to discuss or debate it outside of rarefied academic circles.

The discovery of Lost City catalyzed new interest in Russell's hypothesis, catapulting it to the forefront of contemporary discussions of life's origins. Now the discovery of similar environments within Enceladus—and the potential for their existence in other icy moons such as Jupiter's Europa—is catalyzing another shift in how we think about the possibilities for life elsewhere in the solar system. Biology need not be confined to the warm, wet surfaces of sunlit rocky planets but could perhaps proliferate in a much wider range of environments, sustained in whole or in part by heat from radioisotopes, serpentinization or tidal forces. Enceladus and Europa may be proverbial tips of the iceberg—telltale hints that subsurface oceans also exist in Jupiter's moons Ganymede and Callisto, as well as Saturn's moons Titan and Mimas and even the dwarf planet Pluto. Researchers who, like us, are interested in life beyond Earth are only beginning to grapple with these speculative possibilities and their implications, but it appears increasingly likely that until now, we have drastically underestimated the universe's biological fecundity.

For the time being, we must remain in the dark about whether the interiors of icy moons really do supply all the necessary ingredients for extraterrestrial habitability. The duration and intensity of hydrothermal activity within Enceladus remain an open issue, and discussion of possible hydrothermal activity inside Europa is scarcely more than speculation. Both NASA and its counterpart the European Space Agency are eagerly pursuing answers to these questions and are planning missions to Jupiter's icy moons that could seek Enceladus-style plumes in the late 2020s or early 2030s. In September 2017 Cassini completed its mission—crashing into Saturn to preclude any possibility of contaminating Enceladus or another icy moon with earthly biology. Eventually a new generation of spacecraft could be sent there to undertake in situ investigations, landing on the moon and even gathering samples for return to Earth. At present, such missions exist only in the hopes and dreams of astrobiologists—but perhaps not for long.