Improving the Performance and Sensitivity of Gravitational Wave Detectors

Abstract

The field of observational gravitational wave astronomy has begun in earnest, starting with the detection of the strain signal from the binary black hole merger GW150914 by the Laser Interferometer Gravitational-wave Observatory (LIGO) in 2015. The current incarnation of the LIGO observatories, known as Advanced LIGO, has achieved strain sensitivities on the order of 10−23/√Hz in the hundreds of Hz region, which has enabled unambiguous detection of astrophysical gravitational wave signals. Nevertheless, the scientific output from the LIGO observatories is constrained by the instrumental performance and sensitivity, as there remain many more distant and exotic sources to be observed. This thesis describes a few topics in experimental gravitational physics, broadly unified by the desire to improve the performance and sensitivity of gravitational wave interferometers. First, it describes an experimental effort to search for a novel form of nonlinear mechanical noise that may be relevant for the ultimate performance of the mirror sus- pension systems used throughout the instrument. Next, it summarizes work done at the CalTech 40m LIGO controls prototype to realize its fully operational state, and a novel automated controls algorithm developed and tested there that may be useful in simplifying the control of current and future interferometers. Finally, it describes work done on a system to identify and subtract unwanted noise couplings out of recorded aLIGO strain data in an automated fashion. The noise subtraction system applied to GW150914 is demonstrated to reduce the uncertainties of the black hole mass parameters by about 10%.