Mitigating the effects of glitches on gravitational-wave event sky localizations

Electromagnetic follow up of gravitational-wave signals relies on fast, accurate sky localizations based on gravitational-wave data. Transient instrumental artifacts, known as glitches, are present in the data and can coincidently overlap a transient gravitational-wave signal. If this occurs, then the estimated source sky location of the signal may be biased and prevent identification of an electromagnetic counterpart. Numerous methods have been developed to subtract the glitch noise from data, but are either not quick enough to be used in low latency or introduce their own biases. In this talk, we demonstrate how inpainting techniques can be used to remove glitches that overlap gravitational-wave signals from compact binary mergers. We then present a method to account for the impact of the data that was removed and calculate new bias-free sky localizations in only a few seconds. Only knowledge of the time and duration of a glitch is required to calculate a new sky localization, allowing this method to be used for a wide array of different artifacts. We further show how this method can also be used to test if the estimated source sky location was biased before this method was applied.