Galaxy Evolution and Gravitational Waves, Part II

This post is a continuation from Part I last week.

It turns out that there is another way to detect gravitational waves (GWs) without laser interferometers such as LIGO: once again, through pulsars, except this time in a way that directly demonstrates the presence of gravitational waves nearby the earth. (And it will be particularly suited for supermassive black hole GW detection, as I’ll explain later.) This method is entirely different from the indirect method I spoke of earlier that won Russell Hulse and Joseph Taylor the Nobel Prize in Physics in 1993, but it too relies on the fact that pulsars emit very regular pulses. Imagine a distant (single) pulsar rotating like a lighthouse and sending trains of regular pulses propagating through space and eventually reaching us. Because the timing, and thus the distance, between lightspeed-propagating pulses is very regular, these interstellar radio waves themselves are like a very long ruler! Should there be a stretching of space at or around the earth, a careful pulsar astronomer will find that the number of pulses received per second (typically on the order of a thousand per second) decreases slightly, then increases slightly, as a gravitational wave stretches the fabric of spacetime back and forth. (Note that the gravitational wave I’m referring to otherwise has nothing to do whatsoever with the pulsars being used to detect it.) This method, when it uses multiple pulsars, is referred to as a pulsar timing array (PTA), where the word “array” refers to the pulsars timed regularly in order to look for this slight change in arrival time. NANOGrav (North American Nanohertz Observatory for Gravitational waves) is such an effort underway using the Arecibo Observatory along with the Green Bank Observatory in West Virginia. NANOGrav is also part of a larger international effort known as the International Pulsar Timing Array, using eight radio telescopes. (Just as a miles-long laser, in the case of LIGO can be referred to as an “observatory”, so can forty pulsars spread around the Milky Way galaxy!)

Diagram of a pulsar timing array such as NANOGrav.
Each line of sight to a particular pulsar (yellow) functions
as a "lever arm" with which to measure waves in space-time
(i.e. the hills and valleys in the green grid).

As said before, individual pairs of merging supermassive black holes are unlikely to produce gravitational waves we can detect here on earth. But it turns out that the sum total of all merging black holes throughout a giant volume of space stretching back to about one half of the universe’s elapsed history, produces a combined GW background that can be detected. Thus, there would be gravitational waves coming equally from all directions in the sky. (This is the gravitational wave equivalent to the extragalactic background light or EBL, which is the uniform “glow” in the night sky due to numerous unresolved individual galaxies. CANDELS has also contributed to studying the EBL, as described here. See also the recent posting on the EBL.) 

What does this have to do with galaxies? A lot, because it turns out that there is a very close correlation between the rates at which smaller galaxies merge to form larger galaxies, and the (relative) rates at which small supermassive black holes (SMBHs) merge to become more massive (single) SMBHs. Because most galaxies have a SMBH at the center, if two galaxies merge, then eventually their central black holes will also undergo a merger. If one wants to predict the expected gravitational wave background, then, one needs to look carefully at what galaxy evolution simulations have to say. Carefully adding up the number of galaxy mergers over cosmic time leads to a prediction of how intense the gravitational wave background due to SMBHs should be.

As it turns out, doing this calculation reveals that the strongest part of the gravitational wave background should be at long periods – where a single oscillation takes greater than five years! Fortunately, pulsar timing arrays such as NANOGrav will be able to observe these very slowly vibrating (i.e. “nanohertz”) waves. This is because laser interferometers cannot remain stably phase locked for decades at a time. Millisecond pulsars, however, often do remain stable clocks (and thus rulers) for this length of time. Should the SMBH gravitational wave background be detected, it will be a direct measurement of the galaxy merger rate across cosmic time. This information could then be used to go back and make an even better computer simulation of galaxy evolution.

The Green Bank Telescope in Green Bank, WV.
Along with the Arecibo Observatory, astronomers
hope to detect gravitational waves emitted by
supermassive black holes in the centers of merging galaxies

In other words, all roads lead back to galaxies. At the time of this writing, a new study has been released from the Parkes Pulsar Timing Array in Australia, which uses its pulsar timing measurements to rule out models of galaxy evolution in which galaxy (and thus SMBH) mergers happened very late in cosmic history. (See here: http://arxiv.org/abs/1310.4569) It will be interesting to see what new limits on gravitational waves will come about in the next few years. Or even better, a detection!