The story of the birth of our solar system has been worn smooth through years of retelling. It starts billions of years ago with a black, slowly spinning cloud of gas and dust. The cloud collapses, forming our sun at its heart. In time, the eight planets, along with lesser worlds such as Pluto, emerge from leftover gas and debris swirling about our star. This system of sun and planets has been whirling through space ever since, its motions as accurate and predictable as clockwork.

In recent years astronomers have glimpsed subtle clues that belie this familiar tale. In comparison with the architectures of thousands of newfound exoplanetary systems, our solar system's most salient features—its inner rocky worlds, its outer gas giants and its lack of planets interior to Mercury—are actually quite anomalous. Turning back the clock in computer simulations, we are learning that these quirks are the products of a troubled youth. The emerging rewrite of the solar system's history includes far more drama and chaos than most anyone had expected.

The new history is a tale of wandering planets evicted from their birthplaces, of lost worlds driven to fiery destruction in the sun eons ago and of lonely giants hurled into the frigid depths of near-interstellar space. By studying these ancient events and the scars they may have left—such as the recently postulated Planet Nine that could be lurking unseen beyond Pluto—astronomers are gaining both a cohesive picture of the solar system's crucial formative epochs and a new appreciation for its cosmic context.

The Classical Solar System

Planets are a by-product of star formation, which occurs in the hearts of giant molecular clouds 10,000 times the mass of our sun. Dense core regions within a cloud can collapse in on themselves, forming a central glowing protostar encircled by a sprawling, opaque ring of gas and dust called a protoplanetary disk.

For decades theorists have looked to our sun's primordial protoplanetary disk to explain one of the solar system's most distinctive features: its bifurcated brood of rocky and gassy planets. Four terrestrial worlds are confined between Mercury's 88-day and Mars's 687-day orbital periods. In contrast, the known gas-rich giant planets reside on much more distant orbits, have orbital periods ranging from 12 to 165 years and contain more than 150 times the mass of the terrestrial bodies.

Both varieties of planet are thought to come from a universal formation process, in which motes of dust swirling within the gassy, turbulent disk collided and stuck together to make kilometer-scale objects called planetesimals, akin to the dust balls formed by air currents and electrostatic forces on an unswept kitchen floor. The largest planetesimals also had the greatest gravitational pull and rapidly grew even larger as they swept up lingering debris in their orbits. Within perhaps a million years of its collapse from a cloud, our solar system's protoplanetary disk—just like any other in the universe—teemed with moon-sized planetary embryos.

The largest embryo resided past the present-day asteroid belt, far enough from the newborn sun's light and heat for ice to exist in the protoplanetary disk. Beyond this “ice line,” embryos could feast on plentiful planet-building ices to grow to enormous sizes. In a familiar example of the rich getting richer, the largest embryo was also the fastest-growing, as its greater gravitational field rapidly carved most of the available ice, gas and dust from the surrounding disk. Within only a million years or so, the greedy embryo had grown to become the planet Jupiter. This, theorists believed, was the crucial moment where our solar system's bifurcated architecture emerged. Outpaced by Jupiter, our sun's other giant planets formed into smaller bodies because they grew slower, ramping up their gas-attracting gravitational pulls only after Jupiter had diminished the amount available. The inner worlds were far smaller still because they were born inward of the ice line where the disk was relatively devoid of gas and ice.

Save for a few bothersome details, such as the exceedingly small masses of Mars and Mercury, this “Jupiter-first” narrative appeared satisfactory as an explanation for our solar system's architecture. The expectations were clear for systems orbiting other stars: giant planets would eventually be found in long-period orbits beyond the ice line, whereas rocky worlds would abound with orbital periods on the order of a few years or less. These preconceptions, however, proved to be deceptive.

The Exoplanet Revolution

When astronomers began discovering exoplanets more than 20 years ago, they also put the theory of the solar system's formation to the test on a galactic scale. Many of the first known exoplanets were “hot Jupiters,” gas giant planets whizzing around their stars with orbital periods of just a few days. The existence of giant planets in such scorching proximity to a stellar surface, where ice is utterly absent, is entirely contradictory to the classical picture of planet formation. To reconcile this discrepancy, theorists concluded that these planets formed farther out before somehow migrating inward.

Furthermore, based on thousands of exoplanets found by surveys such as NASA's Kepler mission, astronomers are now arriving at the uneasy conclusion that solar system look-alikes are relatively rare. The average planetary system contains one or more super Earths (planets a few times bigger than Earth), with orbital periods shorter than about 100 days. Conversely, giant planets—Jupiter and Saturn analogues—are found around only about 10 percent of stars, with even lower fractions occupying sedate, nearly circular orbits.

With their expectations in tatters, theorists realized that the “few bothersome details” of the classical theory of our solar system's formation demanded better explanations. Why is the solar system's inner region so depleted in mass compared with its exoplanetary counterparts, with relatively runty rocky worlds instead of super Earths and no worlds at all inside Mercury's 88-day orbit? And why are the orbits of the sun's giant planets so calm and spread out?

As it stands, answers to these questions can be drawn from the failure of classical planet formation theory to account for the fluid mutability of protoplanetary disks. It turns out that a newborn planet, like a life raft in an ocean, can drift far from its point of origin. Once a planet grows large enough, its gravitational influence propagates through the surrounding disk, raising spiraling waves that themselves exert gravitational forces of their own, generating powerful positive and negative feedbacks among planets and disks. Correspondingly, time-irreversible exchanges of momentum and energy can occur, allowing young planets to set off on epic journeys through their natal disks.

When the process of planetary migration is accounted for, ice lines within disks no longer play a singular role in shaping the architectures of planetary systems. For instance, giant planets born beyond an ice line can become hot Jupiters by drifting inward, traveling along with gas and dust spiraling down toward a star. The trouble is that this process works almost too well and seems to be a ubiquitous property of all protoplanetary disks. So how could one account for Jupiter's and Saturn's distant orbits from the sun?

The Grand Tack

The first hint of a compelling explanation arrived in 2001 from computer simulations by Frederic Masset and Mark Snellgrove, both then at Queen Mary University of London. Masset and Snellgrove modeled the simultaneous evolution of Saturn's and Jupiter's orbits within the sun's protoplanetary disk. Because of Saturn's lower mass, its inward migration rate is more rapid than Jupiter's, and as their migrations proceed, the two planets draw closer. Eventually the orbits reach a specific configuration known as a mean motion resonance, in which Jupiter makes three revolutions around the sun for every two orbital periods of Saturn.

Two planets linked by a mean motion resonance can exchange momentum and energy back and forth between each other like an interplanetary game of hot potato. Because of the coherent nature of resonant perturbations, both worlds essentially exert an amplified common gravitational influence on each other and their surroundings. In the case of Jupiter and Saturn, this seesawing allowed the planets to collectively throw their weight against the protoplanetary disk, carving a great gap within it, with Jupiter on the inner side and Saturn on the outer side. At this point, because of its larger mass, Jupiter exerted a greater gravitational pull on the inner disk than Saturn did on the outer disk. Counterintuitively, this caused both planets to reverse course and begin drifting away from the sun. This inward-then-outward swoop is often referred to as the Grand Tack, after its similarity to the motions of a sailboat tacking to change directions against a steady wind.

In 2011, a decade after the Grand Tack's initial conception, computer simulations by Kevin J. Walsh, then at the Côte d'Azur Observatory in Nice, France, and his colleagues showed that it can neatly explain not only the dynamical history of Jupiter and Saturn but also the distribution of rocky and icy asteroids, as well as the diminutive mass of Mars. As Jupiter migrated inward, its gravitational influence captured and shepherded planetesimals in its path through the disk, scooping them up and pushing them ahead of it like a snowplow. If we suppose that Jupiter migrated as close to the sun as the present orbit of Mars before turning back around, it could have ferried icy building blocks totaling approximately 10 times the mass of Earth into the terrestrial region of the solar system, seeding it with water and other volatiles. This process would have also created a clear outer edge to the inner nebula's planet-forming material, truncating the growth of a nearby planetary embryo that went on to become the world we know as Mars.

Jupiter’s Grand Attack

As compelling as the Grand Tack scenario appeared to be in 2011, its relation to the other great remaining mystery of our solar system, namely, the utter lack of planets inward of Mercury, remained elusive. In comparison with other systems packed with close-in super Earths, ours seems almost hollowed out. Why? It seems strange that our solar system did not participate in the dominant mode of planet formation we see elsewhere in the cosmos. In 2015 two of us (Batygin and Laughlin) considered what the consequences of the Grand Tack would be on a hypothetical retinue of close-in super Earths around the sun. Our startling conclusion is that they would not have survived the Grand Tack. Remarkably, Jupiter's inward-outward migration can account for many properties of the planets that we do have, as well as for the ones we do not.

As Jupiter plunged into the inner solar system, its snowplowlike influence on the planetesimals in its way should have stirred their neat, circular orbits into a disordered swarm of spiraling, intersecting trajectories. Some of the planetesimals would collide with great force, shattering into fragments that inevitably generated further fragmenting collisions. Jupiter's inward migration thus most likely triggered a collisional cascade that eroded the planetesimal population, essentially grinding them back down to boulders, pebbles and sand.

Assaulted by collisional grinding and aerodynamic drag within the gassy confines of the inner protoplanetary disk, the fragmenting, eroding planetesimals bled off their energy and rapidly spiraled down closer to the sun in an avalanche of orbital decay. As they fell, they would have been easily captured in further resonances, ominously stacking up on the horizons of any primordial close-in super Earths.

This would have been very bad news for those planets, which would suddenly be hectored by parasitic swarms of debris feeding off their orbital energy. Continuously hindered by gas streaming through the disk, the swarms should have spiraled straight into the sun. But thanks to their resonances with the super Earths, the swarms were held in place, siphoning off orbital energy from the planets and bleeding it off as heat from aerodynamic drag. The net effect was that the swarms of eroded planetesimals pushed the planets into death spirals with ruthless efficiency, progressively lowering each world's orbit so that one by one they all fell into the sun. Our simulations suggest none of these hypothetical planets would have survived longer than hundreds of thousands of years after the collisional cascade began.

Thus, the Grand Tack of Jupiter and Saturn may have unleashed a bona fide Grand Attack on a population of primordial close-in planets in our solar system. As these erstwhile super Earths decayed onto the sun, they would have left behind a desolate unpopulated cavity in the solar nebula, extending out to an orbital period of perhaps 100 days. As a result, Jupiter's glancing swoop through the early solar system produced a relatively narrow ring of rocky debris, from which the terrestrial planets neatly coalesced hundreds of millions of years later. The concatenation of chance events required for this delicate choreography suggests that small, Earth-like rocky planets—and perhaps life itself—could be rare throughout the cosmos.

A Nice Model

By the time Jupiter and Saturn plowed back outward from their foray into the inner system, the sun's surrounding disk of gas and dust was on the wane. The resonant pair of Jupiter and Saturn eventually encountered newly formed Uranus and Neptune, along with, perhaps, an additional, similarly sized body. Aided by the gravitational effects of the dissipating gas, the dynamic duo locked these smaller giants into resonances as well. Thus, just as most of the disk's gas disappeared, the solar system's inner architecture probably consisted of a ring of rocky debris in the neighborhood of Earth's current orbit. In its outer reaches, a compact and resonant chain of at least four giant planets resided in nearly circular orbits between Jupiter's current orbit and roughly the halfway point to the present orbit of Neptune. Beyond the outermost giant planet's orbit, the frozen, icy planetesimals of the outer disk stretched to the far edge of the solar system. Over hundreds of millions of years the terrestrial planets formed, and the once wild outer worlds settled down into what could have been enduring stability. But as chance would have it, this was not the final phase of our solar system's evolution.

The Grand Tack and coeval Grand Attack had arranged one last gasp of interplanetary violence in the solar system's history, a finishing touch that brings our sun's retinue of worlds close to the configurations we witness today. The last gasp is known as the Late Heavy Bombardment, a time between 4.1 billion and 3.8 billion years ago when the solar system temporarily transformed into a shooting gallery filled with barrages of impacting planetesimals. We see its scars today in huge craters pockmarking Earth's moon.

Working with several colleagues at the Côte d'Azur Observatory in Nice in 2005, one of us (Morbidelli) produced the so-called Nice model to explain how interactions between the giant planets could produce the Late Heavy Bombardment. Where the Grand Tack ends, the Nice model begins.

The closely packed giant planets were still resonant with one another and still felt the slight gravitational tugs of the outlying icy planetesimals. They were in fact poised on the knife-edge of instability. Accumulating over millions of orbits across hundreds of millions of years, each individually insignificant tug from the outer planetesimals subtly shifted the motions of the giants, slowly chipping away at the delicate balance of resonances that bound them together. The tipping point came when one of the giants fell out of resonance with another, unraveling the balance and kicking off a chaotic series of planet-planet perturbations that jolted Jupiter slightly inward while scattering the other giants outward. In a cosmically brief span of a few million years the outer solar system experienced a jarring transition from a closely packed, nearly circular state to an expansive, disordered configuration characterized by planets with wide, eccentric orbits. The interactions among the giant planets were so violent that one or more may have been scattered away, ejected beyond the boundary of interstellar space.

Had dynamical evolution stopped here, the outer solar system's architecture would have fit nicely into the trends we witness in giant exoplanets, many of which occupy eccentric orbits around their stars. Thankfully, however, the disk of icy planetesimals that ignited the disorder also helped to eradicate it through subsequent interactions with the eccentric orbits of the giant planets. One by one, most close-passing planetesimals were flung out by Jupiter and the other giant planets, gradually drawing orbital energy from the planets and circularizing their orbits once again. Whereas most planetesimals were ejected beyond the sun's gravitational reach, a small fraction remained in bound orbits, forming a disk of icy debris we now call the Kuiper belt.

A Ninth Planet, a Final Theory

Patient observational work with the largest telescopes is gradually revealing the full expanse of the Kuiper belt, slowly unveiling unexpected structure. In particular, astronomers have spied a peculiar pattern among the most far-flung objects of the Kuiper belt that exist at the outer limits of detectability. Despite having a range of distances from the sun, the orbits of these objects are highly clustered, as if they are all subject to a common, very large perturbation. Computer simulations performed by Batygin and Michael E. Brown of the California Institute of Technology have shown that this state of affairs is naturally produced by an as yet unobserved ninth planet, having a mass roughly 10 times that of Earth and in a highly eccentric orbit around the sun of approximately 20,000 years. Such a planet is unlikely to have formed so far out, but it can be quite readily understood as an exile ejected from closer in during the solar system's infancy [See “Planet Nine from Outer Space” below].

If confirmed, the existence of a ninth planet around the sun would dramatically tighten the constraints on our understanding of our weird, hollowed-out solar system, placing new limits on the theories we could weave to explain all its anomalies. Even now astronomers are marshaling some of Earth's largest telescopes to ardently seek this putative world. Its discovery could mark the penultimate chapter in the long, complex tale of how we discovered our place in the universe, surmounted only by the yet to be written conclusion, when we at last find living worlds orbiting other stars.

Like strands of DNA, that on sequencing, reveal the story of humankind's ancient migrations across the surface of our small planet, astronomical clues have permitted our computer simulations to reconstruct the planets' majestic wanderlust during the solar system's multibillion-year lifetime. From its birth in roiling molecular clouds, to the formation of its first planets, to the world-shattering growing pains of the Grand (At)Tack and the Nice model, to the emergence of life and sentience around at least one sun in the vast Milky Way, the complete biography of our solar system will be one of the most significant accomplishments in modern science—and undoubtedly one of the greatest stories that ever can be told.

Planet Nine from Outer Space

Does the newly postulated “Planet Nine” fit in with the latest thinking about the origin of the solar system?

By Michael D. Lemonick

The idea that the solar system was violently reshuffled in the distant past may explain the existence of the Kuiper belt and the Oort cloud of icy bodies that surround us, the ancient bombardment of the inner planets by asteroids billions of years ago, and the seeming absence of so-called super Earths, which other solar systems have in abundance. But now planetary scientists have something new to wrestle with: a putative planet, with perhaps 10 times the mass of Earth, orbiting in the dark regions beyond Pluto. If it exists, the gravity of the world provisionally known as Planet Nine might be the reason why a handful of known Kuiper belt objects are following suspiciously similar paths around the sun.

But it might also be yet another clue to the wrenching changes the solar system went through early in its history. With an estimated minimum distance from the sun of 30.5 billion kilometers—five times farther than Pluto’s average distance—it is unlikely that this massive world could have formed where it is now. There simply would not have been enough material to build it with. “If it’s there,” says Harold F. Levison, a planetary formation theorist at the Southwest Research Institute, “it most likely formed in the region of between about five and 20 [Earth-sun distances] and was scattered outward by [a gravitational interaction with] Jupiter or Saturn.”

This point is uncontroversial. Jupiter, in particular, is so massive, says Scott Tremaine of the Institute for Advanced Study in Princeton, N.J., that “it doesn’t care whether it’s scattering a comet or a 10-Earth-mass planet.” Once it got the boot, however, a planet would tend to keep going, eventually escaping into interstellar space. The odds that it would instead settle into orbit around the sun are extremely low. Statistically, Levison says, you would need to start with 50 or 100 to end up with one—which he considers unlikely.

If astronomers actually spot Planet Nine through a telescope, the question of likeliness becomes moot, of course. Still, the question of how something so improbable happened is something theorists will have to wrestle with. “My guess,” Tremaine says, “is that the scattering process is more efficient than the standard model would lead us to believe”—that is, a higher percentage of outward-flung objects manages to stay within the solar system than everyone thinks.

One way this might happen, according to Ben Bromley of the University of Utah, is if the scattering of a super Earth took place very early in the life of the solar system, before the gas in the protoplanetary disk that formed into planets dissipated. If the scattering of a super Earth took place within that period, Bromley notes, “The planet could interact with the gas and settle out in the boondocks.”

Or perhaps, says Nathan Kaib, a theorist at the Carnegie Institution for Science in Washington, D.C., Planet Nine, should it exist, did not come from our solar system. The sun formed not alone but in a cluster of perhaps thousands of stars, each (most likely) with its own planetary system. At least some of those systems would have undergone their own violent reshuffling, ejecting objects just as the sun presumably did. “These,” Kaib says, “can be captured by our own sun.”

The best explanation will depend on what Planet Nine’s orbit turns out to be; its proponents have calculated only a range of possibilities. If it does exist, scientists should be able to figure out how it got to where it is. The answer to whether Planet Nine fits with current thinking about the early solar system, Tremaine says, “is a definite ‘maybe.’”

Michael D. Lemonick is opinion editor at Scientific American.