Getting the Best Multi-Messenger Science out of Gravitational-Wave Data with Better Modeling of its Noise and Automation

This award will support the pursuit for understanding the nature of short-duration gamma-ray bursts (SGRBs) and their progenitors. An SGRB is one of the most energetic explosions in the cosmos and is triggered by the collision of two neutron stars or a neutron star and a black hole. It is accompanied by the emission of electromagnetic radiation across a wide range of wavelengths, as well as gravitational waves. While the prototypical binary neutron star collision, GW170817, expanded our understanding of such bursts in unprecedented ways, it also left certain questions unresolved. One such question is the fate of these collisions, which have been conjectured to affect the afterglows of SGRBs. This work will support multi-messenger astronomical observations of SGRB counterparts of gravitational wave (GW) signals in LIGO and Virgo to understand the nature of the remnants produced by them. The project will involve citizen scientists in this investigation via the development of Machine Learning tools for distinguishing noise artifacts in GW detectors that can masquerade as GW signals from these collisions. It will also contribute to training students at the Tri-Cities campus of the Washington State University and education and public outreach activities in neighboring areas. This campus is close to the LIGO-Hanford site and caters to a large percentage of Hispanic and Native-American populations. This project will enhance the science exploitation of gravitational-wave observations with LIGO, Virgo, and KAGRA during their fourth and fifth observation runs. It will reduce the number of false alarms among the low-latency alerts issued by those detectors. For this purpose, it will specifically target GW signals from binaries involving neutron stars. This aids astronomical observatories that follow-up these GW alerts in their quest for detecting accompanying electromagnetic and particle emissions. The project will pursue this objective by (a) contributing to the more complete accounting of sources of non-Gaussian transient noise, (b) identifying different types of non-linear couplings of noise sources, and (c) improving the significance measurement of online GW alerts by utilizing Machine Learning solutions and automated data quality inputs. It will also develop signal-based noise discriminators, or “chi-square” tests, for targeting especially those noise artifacts that are responsible for a majority of false alerts. Observations of these binaries will be used to improve our understanding of the neutron star equation of state and gravitational-wave aided measurements of the Hubble constant. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.