Neutron stars – the dead stellar remnants of old, burned-out stars – are some of the most extreme objects in the universe. They weigh as much as the entire Sun, but are small enough to fit into downtown Sydney, and they rotate up to 700 times every second. Imagine that: a whole star rotating faster than the fastest kitchen blender.
Astronomers know of a few thousand neutron stars, but one in particular is a stand-out. As part of the Parkes Pulsar Timing Array, we have been observing pulsar J1909-3744 with the CSIRO's Parkes Radio Telescope for 11 years.
During this time, we have accounted for every single one of the neutron star's 116 billion rotations (115,836,854,515, to be precise). We know the rotational period of this star to 15 decimal places, making it truly one of the most accurate clocks in the universe.
But, as we show in a paper published today in the journal Science, it was not supposed to be this way. Gravitational waves from all of the black holes in the universe were supposed to ruin the timing precision of this pulsar. But they have not.
Gravitational waves stretch and squeeze space, causing the distance between us and the neutron star to change. The gravitational waves we were looking for should have altered that distance by about ten metres, a tiny fraction given that this neutron star is about 3.6 x 1019 metres from Earth (that's 3.6 with 19 zeros following)! But this should have been enough to show up in our measurements.
Yet the fact that our measurements are so accurate tells us that something is wrong with the theory. This doesn't mean that gravitational waves don't exist. There are other facets of our understanding of the universe that might be off track.
Whatever the resolution to this quandary, it is sure to change the way we understand the most massive black holes in the universe.
The centre of our galaxy harbours a black hole that weighs more than four million times the mass of our sun. But this is a lightweight; other galaxies contain black holes weighing more than 17 billion times the mass of our Sun.
And we have good reason to believe that most, if not all, galaxies contain supermassive black holes in their cores. We also know that galaxies throughout the universe grow by merging with one another.
Following the merger of any two galaxies, the two black holes from the parent galaxies sink to the centre of the daughter galaxy, forming a supermassive black hole binary pair. At some point, the subsequent evolution of the binary pair becomes dominated by the emission of gravitational waves.
Ripples in Spacetime
When any two black holes are spiralling around one another, they ought to emit gravitational waves. These carry energy away from the system, causing the two black holes to move closer together.
The sum of all the binary supermassive black holes in the universe should produce a background of gravitational waves (similar to the cosmic microwave background). It is this background that was expected to ruin our precision timing of PSR J1909-3744.
Astrophysicists have made a number of predictions about the strength of the background. These predictions incorporate state-of-the-art measurements of galaxy formation and evolution, and the most sophisticated theoretical models of how the universe evolves following the Big Bang.
Why No Gravitational Waves?
But we want to be very clear that our lack of a detection does not imply that Einstein's theory of relativity is wrong, nor does it imply that gravitational waves don't exist. While we don't know the real solution, we have a number of ideas.
Perhaps not every galaxy in the universe contains a supermassive black hole. Reducing the fraction of galaxies that host supermassive black holes in the models reduces the predicted amplitude of the gravitational wave background, potentially making it undetectable by our observations.
Perhaps we do not understand the relationship between the mass of the host galaxy and the mass of the black hole. We use empirical relationships between galaxy and black hole masses to determine the latter. While we believe these are robust in the local universe, the black hole mergers we are most sensitive to occur billions of light years from us, where our understanding of these empirical relations is far from complete.
Perhaps one of our assumptions about the process that drives the mergers is too simplistic. For example, if the centres of galaxies contain significant amounts of gas, it can act like an extra friction force, causing black holes to merge with one another quicker than expected. This would also cause a smaller-than-expected amplitude of the gravitational wave background.
At the moment, each of these scenarios is equally plausible. Continued observations of pulsars, as well as observations of the distant universe with large optical telescopes, may soon allow us to distinguish between these ideas. And, one day, we may finally find the direct evidence for the existence of gravitational waves that we're looking for.