Missing Link: Bayesian detection and measurement of intermediate-mass black-hole binaries

We perform Bayesian analysis of gravitational-wave signals from nonspinning, intermediate-mass black-hole binaries (IMBHBs) with observed total mass, Mobs, from 50M to 500M and mass ratio 1-4 using advanced LIGO and Virgo detectors. We employ inspiral-merger-ringdown waveform models based on the effective-one-body formalism and include subleading modes of radiation beyond the leading (2,2) mode. The presence of subleading modes increases signal power for inclined binaries and allows for improved accuracy and precision in measurements of the masses as well as breaking of degeneracies in distance, orientation and polarization. For low total masses, Mobs50M, for which the inspiral signal dominates, the observed chirp mass Mobs=Mobsη3/5 (η being the symmetric mass ratio) is better measured. In contrast, as increasing power comes from merger and ringdown, we find that the total mass Mobs has better relative precision than Mobs. Indeed, at high Mobs (≥300M), the signal resembles a burst and the measurement thus extracts the dominant frequency of the signal that depends on Mobs. Depending on the binary's inclination, at signal-to-noise ratio (SNR) of 12, uncertainties in Mobs can be as large as ∼20-25% while uncertainties in Mobs are ∼50-60% in binaries with unequal masses (those numbers become ∼17% vs. ∼22% in more symmetric mass-ratio binaries). Although large, those uncertainties in Mobs will establish the existence of IMBHs. We find that effective-one-body waveforms with subleading modes are essential to confirm a signal's presence in the data, with calculated Bayesian evidences yielding a false alarm probability below 10-5 for SNR9 in Gaussian noise. Our results show that gravitational-wave observations can offer a unique tool to observe and understand the formation, evolution and demographics of IMBHs, which are difficult to observe in the electromagnetic window.