Intrinsic selection biases of ground-based gravitational wave searches for high-mass BH-BH mergers

The next generation of ground-based gravitational wave detectors may detect a few mergers of comparable-mass M\simeq 100-1000 Msun ("intermediate-mass'', or IMBH) spinning black holes. Black hole spin is known to have a significant impact on the orbit, merger signal, and post-merger ringdown of any binary with non-negligible spin. In particular, the detection volume for spinning binaries depends significantly on the component black hole spins. We provide a fit to the single-detector and isotropic-network detection volume versus (total) mass and arbitrary spin for equal-mass binaries. Our analysis assumes matched filtering to all significant available waveform power (up to l=6 available for fitting, but only l<= 4 significant) estimated by an array of 64 numerical simulations with component spins as large as S_{1,2}/M^2 <= 0.8. We provide a spin-dependent estimate of our uncertainty, up to S_{1,2}/M^2 <= 1. For the initial (advanced) LIGO detector, our fits are reliable for $M\in[100,500]M_\odot$ ($M\in[100,1600]M_\odot$). In the online version of this article, we also provide fits assuming incomplete information, such as the neglect of higher-order harmonics. We briefly discuss how a strong selection bias towards aligned spins influences the interpretation of future gravitational wave detections of IMBH-IMBH mergers.