If you're still having trouble coming to terms with the idea of subsurface oceans, possibly harboring life, on Enceledus or Pluto get ready to be truly bamboozled. Evidence has emerged Ganymede may have not one but several oceans trapped between ice sheets many kilometers thick. In at least one of these, it may snow upwards.

 

The idea of interleaved ice and oceans was first proposed last year by NASA's Jet Propulsion Laboratory, but now researchers at the same institution have published evidence in Planetary and Space Science. “Ganymede's ocean might be organized like a Dagwood sandwich," says lead author Dr Steve Vance, referring to the cartoon character's fondness for multilayered luncheon foods.

 

Confidence that Ganymede even has liquid water is relatively recent. Whereas it's nearest neighbor Europa has a surface that hints at an ocean whose tidal forces constantly wipe the icy surface features clean, Ganymede is cratered. Moreover, Ganymede's greater distance from Jupiter and near circular orbit means it has nothing like the tidal flexing that keeps Io volcanic and Europa warm inside. Nevertheless support for the idea of a subsurface ocean has grown. It is possible that Ganymede previously had an orbit that induced more tidal flexing, not only keeping its innards liquid but releasing geysers that explain how its lowlands were resurfaced.

 

Ganymede is the heaviest moon in the solar system, but its density is well below that of Europa and Io,  indicating it has almost as much ice or water as rock. With most of the rock at its core, that leaves hundreds of kilometers of ice or water at the upper layers. The presence of magnesium sulfate has also been detected, and this would act as a salt within the ocean.

 

However, modeling of Ganymede's oceans previously ignored the interaction between salt and the exceptional pressures that would occur at such mighty depths. Salt ions attract water molecules and, combined with their own higher density, cause salty water to be denser than fresh. Vance experimented with water under extreme pressure and demonstrated that the increased density of salty over fresh becomes important under the conditions similar to Ganymede's depths.

 

It's not new that dense salty water sinks, but under intense pressures, so does ice. While the ordinary ice we are familiar with is less dense than water and floats, ice placed under enough pressure contracts. "It's like finding a better arrangement of shoes in your luggage—the ice molecules become packed together more tightly," says Vance.

 

This ice sinks beneath the water. However, Vance's careful modeling suggested there could be three layers of ice. Surprisingly, even the densest layer of ice is still lighter than the densest water, so it is likely the Ganymede's rocky core would have water, not ice, immediately above it. “This is good news for Ganymede,” says Vance, apparently assuming that what a moon really wants is to support life. “It's ocean is huge, with enormous pressures, so it was thought that dense ice had to form at the bottom of the ocean. When we added salts to our models, we came up with liquids dense enough to sink to the sea floor."

 

In the uppermost ocean cold patches could see an intermediary dense ice, known as Ice III, form above the mezzanine level. As Ice III forms salts precipitate out and fall to the bottom of this layer. Without the salts the ice would become like the material we are familiar with on Earth, Ice I, lighter than any ocean. It would thus start to rise, as snow-like flakes rather than bobbing icebergs, potentially melting on the way to leave a slushy layer beneath the surface.

 

"We don't know how long the Dagwood-sandwich structure would exist," says co-author Christophe Sotin of JPL. "This structure represents a stable state, but various factors could mean the moon doesn't reach this stable state.”

 

The same structure could arise on exoplanets with suitable temperatures. The theory will build support for the European Space Agency's proposed Jupiter Icy moons Explorer (JUICE) mission, as well as encouraging such a mission to not spend most of its time focused on Europa as some have envisaged. Just a few says ago NASA sought suggestions for ways to conduct a mission to Europa for less than $1 billion. Twinned missions to two of Jupiter's moons could share costs and reduce the price for each individually.