Automating Veto Algorithms for Application to the Search for Burst Gravitational Waves with LIGO

This research has applied automated veto (glitch rejection) tools to evaluate their potential to improve the search for burst (unmodelled) gravitational waves with LIGO (Laser Interferometer Gravitational-Wave Observatory). During the process of data acquisition, transient noise from environmental and instrumental sources creates glitches in the data. The impact of these glitches on the search for gravitational waves can be mitigated by removing (vetoing) these glitches from the data. Auxiliary channels, data streams that are unable to detect gravitational waves, are used to generate glitches and minimize the chance that a glitch may be a true gravitational wave. Omicron is an algorithm that is currently used to automatically identify data segments likely to be glitches. Two different statistical algorithms evaluate data quality when Omicron glitches originating from different auxiliary channels are vetoed: hierarchical veto (Hveto) and Used Percentage Veto (UPV). We have developed software to collect the daily results of Hveto and UPV and apply those candidate vetoes to the burst search algorithm, Coherent WaveBurst (cWB). The results are evaluated by the Veto Evaluation Tool (VET) to measure a veto’s efficiency (fraction of the background events that were vetoed) in removing cWB triggers, deadtime (amount of time removed by the veto), and the ratio of efficiency to deadtime. The higher this ratio, the more effective the veto. Results of this research show that Hveto and UPV provide vetoes that identify unique glitch features, and both have the potential to improve data quality for burst gravitational wave search. New features have been added to assist data scientists in identifying problematic channel sources. This material is based upon work supported by the National Science Foundation under Grant PHY-2110157.