One of the major breakthroughs in astronomy in the last decade has been the discovery that extra-solar planets are ubiquitous. We now have a variety of ways of detecting exoplanets, and our telescopes have now found thousands of planetary systems. Perhaps the most surprising discovery was that most known extra-solar planetary systems look nothing like the Solar System: we see giant planets orbiting their stars in just few days; planets on highly eccentric orbits; and even systems where multiple rocky or icy planets are packed into very closely-spaced orbits. The latter type of system, discovered by the Kepler mission, are particularly interesting, as these planets orbit too close to their host stars to have formed in their current locations. The conventional picture of planet formation suggests that such planets formed much further from their stars, and then “migrated” inwards due to gravitational interactions with their parent protoplanetary gas disc. Planets of different mass migrate at different rates, however, and in multi-planet systems this process invariably leads to neighbouring planets becoming trapped in so-called mean-motion resonances. These resonances are special locations in planetary systems, where adjacent planets’ orbital periods are integer multiples of one another (i.e., a 2:1 resonance occurs when the outer planet has exactly twice the period of the inner planet). When in such a resonance the planets’ feel one another’s gravity much more strongly than they would do otherwise. Planets which become “trapped” in these locations are therefore expected to remain in resonance indefinitely, but this is at odds with what we see in real systems: most exoplanets in multiple systems are not in resonance with their neighbours. This stark discrepancy between theory and reality has led astronomers to question the very basics of our understanding of how planetary systems form.

In the last few years astronomers have suggested a wide variety of ways around this problem, but as yet no solution has gained acceptance. The Theoretical Astrophysics Group in Leicester has recently used high-resolution numerical simulations to study how planets in multiple systems migrate through their parent protoplanetary discs. The computational power provided by DiRAC allowed us to follow the evolution of multi-planet systems for many thousands of orbits, capturing the planet-disc and planet-planet interactions in exquisite detail (see Figure 1). These simulations show that under certain conditions (conditions which are prevalent in many planet-forming discs) the gravity between the disc and the planet can overcome the gravity between the planets, breaking the system out of the resonance. This mechanism appears to operate efficiently, with planets escaping resonance after as little as a thousand years. This potentially allows multi-planet systems to migrate into their observed tightly-packed orbits without becoming “trapped” in resonances, and offers an elegant answer to the question of how to assemble the tightly-packed planetary systems observed by Kepler.