There are many enduring mysteries in astrophysics that challenge our theoretical understanding of different phenomena, are understood through careful observation or simulation, then become deep mysteries again when new data conspires to reveal some heretofore ignored or misunderstood aspect of their character. One of the most prominent examples of this in the modern era are black holes in the early Universe, recently reviewed by Volonteri and Billovary (Rep. Prog. Phys. 75, 124901 (2012), arxiv:1209.2243).
The Report poses several questions which are still much debated and the focus of ongoing research: What physical mechanisms lead to the formation of the first massive black holes? How massive were the initial massive black hole seeds? When and where did they form? How is the growth of black holes linked to that of their host galaxy? These questions, and many others, can be considered with gravitational wave observations from future space-based detectors like LISA through the study of massive black hole binaries.
There is ample observational evidence to infer the existence of massive black hole binaries. Consider the images to the right, showing the famed double source in 3C 75 (suggesting two, active, accreting black holes; see Chandra’s page, or 3C75’s SIMBAD record), and the “kinked jet” source in NGC 326 (which might be due to a jet orientation change after the merger of two spinning black holes; NGC 326’s SIMBAD record). These kinds of “x-shaped sources” were first cataloged by Leahy and Parma [1992].The figure below shows the last two hours of a simulated gravitational waveform (courtesy, John Baker [GSFC])from a 10^5 + 10^5 Msun black hole binary at z = 15, including the expected instrumental and astrophysical noise for an observatory like LISA. The waveform is easily detectable! Heuristically, gravitational wave astronomers think about binary black holes in three distinct phases of their evolution.
1. Inspiral. This phase lasts much longer than the short bit shown in this two-hour segment of the waveform. The long slow inspiral of two massive black holes will be visible to a spaceborn gravitational wave observatory for a year or more. During this time, the localization by a detector in a LISA-like orbit will slowly pare down the source’s location on the sky. Information gleaned from waveform characterization in this time will also help determine the initial spins of the black holes, the total mass of the system, and the luminosity distance (one of the most important, classic results that will emerge from gravitational wave astronomy: Schutz [1986]).2. Merger. Long regarded as one of the holy grails of numerical simulation, the last decade has seen tremendous strides in our understanding of the late inspiral and merger of black hole systems. Current gravitational wave research in this area is focused on producing accurate gravitational waveforms, and predicting the astrophysical outcome of the mergers, particularly as they depend on spin and mass ratio, and whether they result in significant “kicks” to the final remnant (see a recent review by Komossa, arxiv: 1202.1977).
3. Ringdown. The remnant that forms from the merger of two black holes will itself be a black hole. How big is it? How is it spinning? How do the distortions from merger settle down into a well behaved black hole described by general relativity? The answers to these questions are encoded in the ringdown phase, when the black hole is excited much like a struck bell, radiating the distortions and energy away in gravitational waves. This is the phase where searches for electromagnetic counterparts are expected to begin in earnest, depending on whether there is prompt or delayed emission associated with the merger event and the interaction of the remnant with the surrounding astrophysical environment.
The biggest unknown with regard to these sources is the number of them that will merge as a function of time. That uncertainty, in and of itself, is one of the outstanding questions in this field, deeply connected to formation and growth scenarios for massive black holes. Recent provocative ideas from McWilliams et al (arxiv: 1211.5377) and Sesana (arxiv: 1211.5375) have examined models for massive black hole merger scenarios (particularly with an eye toward another gravitational wave detection technique, known as puslar timing arrays). Their results suggest the number of detectable systems may be much higher than previously estimated, underscoring how poorly constrained our knowledge of the massive black hole distribution functions are using current observational techniques. Gravitational wave detections of binary black hole systems will be a direct measurement of the merger rate that will allow us to significantly constrain the models currently in vogue.