≡ Menu

Giant Jovian Impact Could Explain Juno Data

Impacts seem to have run rampant in the early Solar System, to judge from what we keep uncovering as we survey today’s evidence. The Moon is widely considered to be the result of Earth’s impact with a Mars-class object, while Mercury’s big iron core may show what happens when a larger world is stripped of much of its mantle in another ‘big whack.’ Then there’s Uranus, spinning lopsidedly in the outer system.

We also know that impacts continue to make their mark. They’re shown up on Jupiter at a fairly brisk pace, with Shoemaker-Levy striking the gas giant in 1994, and another evident impact from an asteroid earlier this month, creating a definitive flash.

For that matter, we have a Hubble image from 2009 showing an impact, an expanding spot twice the length of the United States. That one was discovered by Australian amateur astronomer Anthony Wesley. Later observations allowed scientists to estimate the impactor’s diameter at 200 to 500 meters, with an explosion thousands of times more powerful than the Tunguska event in 1908. Juno mission scientist Ravit Helled (University of Zurich) jokes that when planetary scientists lack a solution, they tend to invoke a giant impact. If so, it seems to be an understandable assumption.

Image: Hubble’s view of the 2009 impact event on Jupiter. Credit: NASA, ESA, and H. Hammel (Space Science Institute, Boulder, Colo.), and the Jupiter Impact Team.

But Helled and colleagues in the National Centre of Competence in Research PlanetS (Bern, Switzerland) aren’t joking when it comes to results from Juno that are forcing them to change their views of Jupiter’s core. The gravity data demand explanation, which may emerge in a massive impact early in the history of Jupiter’s formation:

“Instead of a small compact core as we previously assumed, Jupiter’s core is ‘fuzzy,’” Helled explains. “This means that the core is likely not made of only rocks and ices but is also mixed with hydrogen and helium and there is a gradual transition as opposed to a sharp boundary between the core and the envelope.”

In a paper just published in Nature, Helled and an international team led by Shang-Fei Liu (Sun Yat-sen University in Zhuhai, China) present the results of simulated collisions between an early Jupiter and planetary embryos. They worked with software code developed by PhD student Simon Müller that probed planetary evolution. Particularly puzzling is the thermal evolution of the planet after the impact. Could the diluted core Juno found really persist for billions of years until today?

Giant impacts, the authors argue, are most likely to occur not long after runaway gas accretion. This is when the gravitational effects of the growing planet increase 30-fold in the space of a few million years, destabilizing the orbits of nearby embryos. The team’s simulations show that Jupiter’s gravitational effect on nearby planetary embryos would have been profound, with at least a 40 percent chance that a large embryo would hit Jupiter within the first few million years.

The scientists worked through tens of thousands of simulations to model this effect, and went on to use separate computer code to investigate what these impacts would do to Jupiter’s internal structure. To produce the diluted core we see today, heavy elements in the core and the embryo need to mix with the surrounding gas envelope.

Analyzing heat transport and heavy element mixing, the team finds that it would take an impactor with about 10 times Earth’s mass to stir Jupiter’s core, mixing denser layers with less dense layers above. The team’s 3D models show the effects of a major hit below.

Image: This is Figure 3 from the letter in Nature. Caption: Three-dimensional cutaway snapshots of density distributions during a merger event between a proto-Jupiter with a 10M rock/ice core and a 10M impactor. a, Just before the contact. b, The moment of core–impactor contact. c, 10 h after the merger. Owing to impact-induced turbulent mixing, the density of Jupiter’s core decreases by a factor of three after the merger, resulting in an extended diluted core… Credit: Shang-Fei Liu/Sun Yat-sen University.

The authors go on to compute the thermal evolution following the impact of a 10 Earth mass impactor until the present day, a span of 4.56 billion years. There is only one solution that produces a diluted core like that found by Juno. From the paper:

We conclude that Jupiter’s diluted-core structure could be explained by a giant impact event, but only under specific conditions including a head-on collision with a massive planetary embryo, a post-impact central temperature of about 30,000 K or an initial thermal structure created by the accretion shock during the runaway phase. Indeed, the hydrodynamic simulation suggests that most of the impact energy is not deposited in the deep interior, and therefore the central temperature is unlikely to increase substantially, supporting the diluted core solution.

Interestingly, such an impact would demand the collision be head-on, for grazing impacts would not produce the core-density profile that Juno has now measured. Even a grazing embryo of 10 Earth masses, in this scenario, would be disrupted while sinking to the center of the planet. Meanwhile, smaller impactors (1 Earth mass or less) disintegrate in the envelope of the gas giant before they ever reach its center.

How definitive is this impact solution? The paper points out in its conclusion that a gradual accretion of planetesimals along with runaway gas accretion could produce a disrupted core, but the authors question whether this would allow a diluted core to be preserved to the present day. They also note that giant impacts like the one they model here may be producing an observational signature in extrasolar gas giants, in the form of the high metallicity found in some of these worlds. And this is interesting:

Since impacts of planetary embryos are expected to be frequent after a gas giant’s runaway gas accretion phase, such an event with different impact conditions (such as a small impactor or an oblique collision) may have also happened to Saturn, and could in principle explain the differences between the internal structures of Jupiter and Saturn.

Shang-Fei Liu et al., “The formation of Jupiter’s diluted core by a giant impact,” Nature 572 (15 August 2019), pp. 355–357 (abstract).

{ 22 comments… add one }
  • Antonio August 22, 2019, 15:02

    “and could in principle explain the differences between the internal structures of Jupiter and Saturn”

    Do we know the internal structure of Saturn with that level of detail?

    • Ashley Baldwin August 24, 2019, 9:39

      Yes. Militao et al conducted a provisional analysis of the Cassini Grand Finale data and published in May. The results for Saturn were essentially what was expected for Jupiter. Consistent with a dense rocky core ( 15-18 Me) clearly demarcated from a surrounding 1.5-5 Me area of heavier elements . Underlying a thick gaseous H/He atmospheric envelope . No core dilution or elemental mixing.

  • John walker August 22, 2019, 15:20

    It would be amazing to see a high res time lapse of the planetary pinball during the early solar system. What gigantic(on a planetary scale) events these impacts were.
    I remember the end sequence of Lars von Trier’s Melancholia. Emotionally devastating. But, despite my tears a gruesomely fascinating collision.

  • Laintal August 22, 2019, 16:52

    Thanks Paul

    Another great article and it was one interesting paper to read too


  • Harry R Ray August 22, 2019, 18:56

    From today’s ESSIV presentations: TIC 260128333(TESS’s first CIRCOMBINARY planet). Orbital period – 95.2 days. Eccentricity – 0.086. Mass – ~72 Earth mass. Radius – 3.87 Earth radii(ALSO: EXACTLY Neptune’s radius). Teff – 450 Kelvins. Density – ~ Earth’s density. THEREFORE: This planey OBVIOUSLY has a huge gas envelope, BUT: ~50 Earth mass consists of a rocky core! My take: This planet was once a gas giant like Jupiter, and had a collision almost exactly like the one proposed in this posting, BUT; because it orbits a binary star, the great majority of the gas expelled had its orbit ALTERED by the binary stars’ gravitational field and was unavailable for the transformed planet to re-integrate into itself. To see the data for yourself: ONE-log onto https://solar-flux.forumotion.com TWO-Scroll down to and click on “Extrasolar News and Discoveries. THREE-Clock on “TIC260128333: TESS’ first transiting circumbinary planet” by clicking the box next to “Led Zep”. FOUR-Click EACH of the four charts to ENLARGE them on “image Barn”.

  • wdk August 23, 2019, 10:47

    Just one interesting article after another!
    I’ve heard other versions of “just one thing after another”, but I like this type. May you live in interesting times does not sound so bad.

    Aware from reports that there does seem to be some odd homogeneity about Jupiter’s interior that makes one wonder how it got started.
    And then we had reports earlier of some ancient suns of low metals nature that seemed to have a gas giant or two of their own… But what puzzles me right at the moment about collating all these data and findings is that there is also an understanding or case for Jupiter’s migration back and forth in the solar system due to accretion disk interactions, planetary resonances and the ejection of a Neptune or two. The Nice model.

    Now here or today we have a head on collision. That would tend to cause migration inward, right?

    And if Jupiter hit a ten Earth mass going 15 km/sec in one direction, say, and Jupiter was moving 15 km/sec in the other… Well, well that would be like negative specific impulse of about 2800 seconds with a change in mass of 1/30th, judging from my rocket exercises. A very short finite burn at that.

    At the very least that would put that fireball in an elliptical path. And it was so head on, apparently that the rotational axis of Jupiter either decided to become perpendicular to the ecliptic, or else was that way already. Never mind that another gas giant out there (Uranus) skewed way off, perhaps due to another encounter. How’s its interior? And one out of ten thousand simulations does it about right? .. Do you suppose the celestial mechanics as described above were taken into account too, or just the eventual steady state of the interior?

    Well, the investigation solved the immediate problem.

  • David Evans August 23, 2019, 11:51


    But “several football fields”? Come on, we’re all grown-ups here, we can handle sizes in meters (even approximate ones). And some of us may be used to different football fields than others.

    • Alex Tolley August 23, 2019, 13:30

      Exactly. American football (L=109.1m) or English football (soccer) ( Lmin = 90m, Lmax = 120m) fields? ;) At least they are roughly similar in size. Meters would be so much simpler.

      • Paul Gilster August 23, 2019, 13:33

        Yes, this seems to have set off a number of people. Let me see if I can dig out the figure in meters.

        • Paul Gilster August 23, 2019, 14:57

          OK, about 200 to 500 meters in diameter. I’ve inserted that into the text.

  • Neil Stahl August 23, 2019, 14:57

    Maybe you use “football field” as a unit when you just don’t have an accurate measurement.

    • Paul Gilster August 23, 2019, 14:59

      Could be, but in this case I dug up 200 to 500 meters.

      • ljk August 26, 2019, 9:43

        Paul, next time you should use Colosseum lengths, just to mix things up a bit. :^)

        The dimensions (from Wikipedia):

        It is elliptical in plan and is 189 meters (615 ft / 640 Roman feet) long, and 156 meters (510 ft / 528 Roman feet) wide, with a base area of 24,000 square metres (6 acres). The height of the outer wall is 48 meters (157 ft / 165 Roman feet). The perimeter originally measured 545 meters (1,788 ft / 1,835 Roman feet). The central arena is an oval 87 m (287 ft) long and 55 m (180 ft) wide, surrounded by a wall 5 m (15 ft) high, above which rose tiers of seating.

  • Ron S. August 23, 2019, 15:16

    Football field units would put us in the ball park?

    Who knew science could be so much fun.

    To Paul: sorry, used the wrong name in the first version. please delete.

  • Mike Serfas August 23, 2019, 15:37

    This may be a silly question, but with the outer atmosphere of Jupiter there are layers of ammonia, ammonium hydrosulfide, and water clouds at the temperature levels where these chemicals precipitate out, more or less. If the core is so dilute, does that imply that more or less every component of Jupiter has a cloud layer somewhere in the supercritical atmosphere of the planet – clouds of condensing iron, tin, gold, etc., or compounds thereof? Has anyone been able to model the planet’s chemistry given the new core data and propose more of what lies beneath at least the inferred water layer?

  • Geoffrey Hillend August 24, 2019, 22:59

    I don’t see how an examination using gravitation from comet impacts to Jupiter could lead to the assumption that the core is mixed. The heavier elements and rock will stay in the center. It seems rather far fetched. I would like to see how can we differentiate a gas giant with a solid core from a mixed one using gravity. Also a direct collision of two 10 Earth mass objects blast each other into bits and do not leave much left of a core..

    A grazing collision would knock Jupiter on it’s side like Uranus. Consequently, I don’t think that the theory matches observations that is the heat from inside Jupiter and Jupiter’s present moons which are high in volatiles, which are assumed to have formed from an accretion disk with Jupiter in the center like a mini solar system. Jupiter would also have many more moons and some would have heavy elements being parts of a core of a giant impact collision.

    • wdk August 27, 2019, 18:59

      Regarding the issue of the internal mass distribution, I think there might be a couple of ways to get some information, based on Juno’s trajectory.
      For one, it has high enough inclination to get a sampling of what is going on with respect to Jupiter’s oblateness. The Earth’s J2 term in the polynomial series that describe its gravitational potential, is derived from the Earth’s flattening or oblateness. Or we could say that the mass distribution is not uniformly spherical. In both worlds we can see it visually. Jupiter is much more flattened, but it also a fluid world. A lot of what we see is simply reflective, not massive. The earth as a solid or a slowly flowing one, means it. A ten mile or kilometer (?) between equatorial and polar diameters is backed by dense materials.

      Now suppose the center of Jupiter is the dense sphere? The J2 due to
      oblateness would not be as large as suspected by visual images.

      I submit that as an indicator that what trajectories detect is a J2 reparesentative of an intermediate value, larger than suspected with the premise of a solid spherical core , but smaller than what the visual gas bag would indicate.

  • Geoffrey Hillend August 26, 2019, 20:42

    Intuitively, it seems to me that a ten Earth mass planet would do more than just mix Jupiter’s core. I think we would at least see some stripping of the outer atmosphere of Jupiter over the collision area and some debris reaching orbital velocity considering the extreme kinetic energy and momentum from the gravitational potential to two very massive bodies. Another idea is that Jupiter got it’s rotational angular momentum from the gas in the protoplanetary disk around our Sun. https://astrobites.org/2018/04/19/why-jupiter-spins-fast-but-not-really-fast/

  • Geoffrey Hillend August 27, 2019, 15:10

    There could also have been some collisions of planetesimals and protoplanets or even planets which collided with Jupiter to give it angular momentum and a fast rotation as well as gas and dust.

  • Geoffrey Hillend August 28, 2019, 16:05

    I like the disk instability theory which argues gas giants like Jupiter were formed from pebbles, gas and dust instead of planetary bombardment periods like the inner, rocky planets.

  • ljk September 3, 2019, 10:44

    Looks like Saturn’s interior isn’t exactly rock-solid, either:


Leave a Comment