Distinguishing short duration noise transients in LIGO data to improve the PyCBC search for gravitational waves from high mass binary black hole mergers

Abstract: "Blip glitches" are a type of short duration transient noise in LIGO data. The cause for the majority of these is currently unknown. Short duration transient noise creates challenges for searches of the highest mass binary black hole systems, as standard methods of applying signal consistency, which look for consistency in the accumulated signal-to-noise of the candidate event, are unable to distinguish many blip glitches from short duration gravitational-wave signals due to similarities in their time and frequency evolution. We demonstrate a straightforward method, employed during Advanced LIGO's second observing run, including the period of joint observation with the Virgo observatory, to separate the majority of this transient noise from potential gravitational-wave sources. This yields a $\sim 20\%$ improvement in the detection rate of high mass binary black hole mergers ($> 60 M_{\odot}$) for the PyCBC analysis.