A cross-correlation search for GWs from long-lived magnetars with Advanced LIGO

With the detection of gravitational waves (GWs) from the binary neutron star (BNS) in-spiral GW170817, associated with the gamma-ray burst (GRB) 170817A and its broad-band electromagetic afterglow, multi-messenger astronomy is now at the forefront of astrophysical research. Through the use of multi-messenger astronomy we now have an unprecedented opportunity to study the mechanisms that drive GRBs. While several theoretical models predict an accreting black hole (BH) acting as the central engine for a GRB, some GRBs have so-called ‘plateaus’ in their X-ray afterglow light curves that may signal the presence of a long-lived (100-1000 seconds) highly magnetized NS (magnetar). If a magnetar is the GRB central engine and powers the X-ray plateau, GWs may be emitted from it and could be detectable with ground-based GW detectors such as the Laser Interferometer Gravitational wave Observatory (LIGO). Recent searches that focus on GWs of duration similar to those of the X-ray plateau ( ~ 100-1000 seconds) have highlighted the need for improved data analyses techniques that could push the horizon distances of current (or near-future) GW detectors to distances at least comparable to that of GW170817. In this work, I present a new search algorithm dubbed the Cross Correlation Algorithm (CoCoA), which has the ability to achieve greater sensitivity than other current methods for the detection of long-lived GWs. The improved sensitivity of CoCoA comes with substantial computational cost, and thus this technique does not substitute, but rather complements, other less sensitive but more robust and computationally cheaper search methods. Hereafter, I describe the full implementation and testing of CoCoA, as well as first results of its application to a search for long-lived GWs from GW170817/GRB,170817A. I show how the sensitivity achieved in this last search for a specific set of magnetar models brings a 10-fold improvement on that of a similar search performed by the LIGO Scientific Collaboration (LSC) in 2017. While the algorithm and analyses presented in this Thesis have been fully designed, developed, and tested at Texas Tech University, I acknowledge invaluable inputs and feedback from several LSC colleagues.