Simulated Moon Crashes Reshape Icy Oceans but Never Create Them

Within 48 hours of impact, the wreckage of a shattered moon had pulled itself back into a single lump, at least inside a computer. In simulations of head-on crashes between icy moons, a collision mostly altered how thick a buried ocean grew and how long it lasted, and it settled whether the moon had an ocean at all for only a stretch of its history. None of four collision simulations, run at speeds from one to three times the mutual escape velocity (the speed at which two bodies could just pull free of each other’s gravity), conjured an ocean in a moon that would otherwise have stayed frozen. Which real moons that describes is a narrower question than it sounds.

Out past Jupiter the solar system is a rough neighborhood, and some of today’s moons around Saturn and Uranus are thought to be reassembled remnants of earlier generations, smashed apart and pulled back together from debris. Some of those same moons, Mimas and Ariel among them, are candidates for a recent or present-day ocean. Would a crash wipe that ocean out, or stoke it?

In Nature Astronomy, Marc Neveu, an astronomy associate research scientist at the University of Maryland, and colleagues at the Southwest Research Institute in Colorado and the Weizmann Institute of Science in Israel describe stitching together two very different kinds of computer model: a smoothed-particle hydrodynamics code, which tracks roughly 500,000 particles of rock and ice as they shatter, heat up and clump back together, and a thermal evolution model that follows a moon’s interior for billions of years, asking where heat builds, where it leaks away and whether ice melts into an ocean. They could not simply run one model start to finish. The two do not share assumptions, so the output of the first had to be rebinned by hand into the input of the second. Collisions are expensive to compute and slow burns are cheap, which is why a handful of crashes fed many more thermal runs.

The Same Crash Cut Two Ways

Size turned out to matter more than the team expected, and the same kind of impact that helped one moon hurt another. For a moon roughly 1,000 kilometers in radius, struck head-on by a 500-kilometer body at about 2.1 kilometers per second, the simulated ocean did more than survive. It thickened after an impact about 500 million years into the solar system’s history and stayed thick for about two billion years, whereas an untouched twin grew an ocean roughly 200 kilometers deep that dwindled to almost nothing by the present day. Neveu puts the mechanism plainly: “In larger moons, the energy of the crash converts into extra heat that can actually thicken up an existing ocean for a couple billion years.”

The result surprised Neveu and his co-authors, and they had stacked the deck against it by choosing head-on impacts, which maximize both heating and disruption. “These simulations were pretty much the biggest collisions we could come up with,” he says. “If those didn’t make a difference, it’s unlikely smaller ones would either.”

Small Moons Lose an Early Ocean

So why would a crash that heated a small moon cost it an ocean? Moons around 500 kilometers in radius tell a different story, and the difference is insulation. Left alone, the interior of such a moon warms slowly, and in the two larger of the three simulated cases an ocean formed roughly 0.25 to 0.3 billion years after the moon assembled and lasted about a billion years before it froze again. Add a collision, though, and the picture changes. Impacts at two and three times escape velocity heated the outer layers enough to let rock sink and ice rise, which put water closer to the cold surface and stripped away the rocky cover that had been insulating it; even the gentlest impact left no lasting ocean, and the hotter ones left a moon warmer than its untouched twin but fully frozen. Hardly a reset button. In the model, a crash that heated a small moon cost it the ocean it would otherwise have had.

The university’s release says the findings apply to a long list of real worlds, Mimas, Enceladus, Miranda and Triton among them, but the paper is more careful. Its simulated targets were sized to bracket moons such as Ariel, Umbriel, Titania, Oberon, Tethys, Dione and Rhea; Mimas, Enceladus and Miranda are smaller than any moon the team simulated, the authors flag them for future work, and Triton is not treated at all.

Rhea is the tempting test case, with ancient craters that have slumped as if warmed from below. “It’s like building a snowman in the winter, and then a week later, there’s been warming and sunlight so the snowman is melting away,” Neveu explains. The heat, in his telling, came from below, and a long-ago collision might account for it; but at about 763 kilometers in radius Rhea falls between the two sizes simulated, and the paper stops at proposing the test.

The authors are open about what the models leave out. Impacts were head-on and material strength was ignored, which would probably mean less disruption in reality; the moons sat at Uranus’s distance from the Sun, and when a large moon’s surface was made 45 percent warmer, a rough proxy for a moon nearer the Sun, its simulated ocean was less likely to last to the present day.

Microbes or Whales

NASA and the wider planetary science community are weighing missions to these worlds, and a moon’s likely ocean history bears on where to go and what to look for: the gravitational fingerprint of a hidden ocean, salty deposits, icy volcanoes on the surface. It could even set how sensitive a life-detecting instrument needs to be. “If there’s only a handful of microbes, you won’t be designing the same kind of search mission as you would if the ocean were full of whales,” Neveu says.

Collision history is only one factor, Neveu cautions, and heat from tides matters at least as much. He wants a longer film: “Down the line, I would love to press play on a moon system and watch how the moons move, how they smash into each other and how their interiors change as a result,” he says. “Maybe then we can reconstruct what happened around Saturn, Uranus and Neptune.”

FAQ