Doctor of Philosophy (PhD)


Physics and Astronomy

Document Type



This dissertation describes a novel method of analyzing fluctuations in the time-dependent calibration models of the LIGO interferometers to estimate their effect on strain reconstruction for gravitational-wave detections. The time-dependence of the calibration model of each detector is tracked with a set of parameters which are continuously measured while the interferometers are operating. These parameters track slow variations in the sensing function of the detectors as well as the actuators that hold the detectors in an operational state. The time-dependent parameter data during the second observation run (O2 [November 30, 2016 16:00 UTC to August 25, 2017 22:00:00 UTC]) and the first epoch of the third observation run (O3a [April 1, 2019 15:00 UTC to October 1, 2019 15:00 UTC]) were combed for significant fluctuations. A modified z-score was used as a standardized metric to identify and sort time segments which correspond to noise transients in the time-dependent calibration parameters for both detectors. The results of our search through O2 data from the LIGO Hanford Observatory (LHO) identified a set of relatively few noise transients all with negligible statistical significance, demonstrating that the behavior of the time-dependent calibration parameters at LHO were largely consistent with Gaussian noise on 30-minute or less time scales. Our search through O2 data from the LIGO Livingston Observatory (LLO) showed similar results for the parameters that track variations in actuation. Likewise, the sensing function parameters were mostly consistent with Gaussian noise as well; however, our search identified a small set of statistically significant noise transients. None of which were coincident with gravitational-wave signals, and the most significant of which were estimated to momentarily increase the strain reconstruction 1-sigma uncertainty from 1% to 3% in the regime of the interferometers' most sensitive frequency band (150Hz). Like the O2 results, analysis of O3a data revealed that all the time-dependent calibration parameters at each detector were mostly consistent with Gaussian noise on 30-minute or less time scales. Our results demonstrate that overall, the time-dependent calibration parameters experienced very few significant fluctuations throughout O2 and O3a, and even largest transient we found would have minimal impact on gravitational-wave measurements.



Committee Chair

Gonzalez, Gabriela