Since 2010, radio astronomers have been contending with a mystery: a handful of very intense, very short lived signals have been detected, and disappeared just as quickly. Only recently have we even had any indication as to their origin, and even then, it's just the faintest hint: that they seem to somehow originate from pulsars or other dead stars (and probably not aliens).
In August, the Astrophysical Journal published the research of two astrophysicists at Nanjing University, which was just released on pre-print server Arxiv this week. In the paper, researchers Y.F. Huang and J.J. Geng detail their own take: that fast radio bursts are the collision of two objects, but not necessarily two pulsars or a pulsar and a black hole. Instead, it's something much simpler: asteroids colliding into pulsars.
The astronomers used data collected up until July 2015, detailing 10 such events in all. Unlike the radio pulses that give pulsars their name, fast radio bursts don't appear with any regularity. Without a way to pinpoint their origin, radio astronomers have to hope they pick them up in other kinds of sky surveys and that it helps understand the context of the event.
Huang and Geng believe that, based on the examples of radio bursts at their disposal, any object that might collide with a pulsar would have to be highly elongated, not unlike a mid-sized asteroid that never had enough gravity to form into a more spherical shape. Because of a limited size and mass, especially compared to the dense, almost-all-neutron material inside a pulsar, the asteroids would be destroyed quickly, which is why fast radio bursts would only last a few milliseconds.
The burst itself is likely the energy given off by an explosion in the outer layers of the pulsar. "Just before the final collision, the material is compressed to a dense thin sheet by the magnetic field of the neutron star," Huang and Geng write. "The collision will launch a rapidly expanding plasmoid fireball from the surface, leading to a fan of field lines filling with relativistic electrons."
Such an event, they estimate, is likely to happen every 1-10 million years per pulsar, which is why they're so elusive: it's not a periodic event, but rather an occasional cataclysm. Given that pulsars (or neutron stars) are the dense remnants of certain kinds of supernovae, under this model, there's a finite (though still massive) sampling of stars to work with.
The results aren't conclusive. Rather, it just adds on to the number of possible scenarios that might create a fast radio burst. So of course, it could always still be aliens. But probably not.