Use of Bayesian Inference to Diagnose Issues in Experimental Measurements of Mechanical Disk Resonators

Abstract

Gravitational wave detectors, such as LIGO, are predominantly limited by coating Brownian thermal noise (CTN), arising from mechanical losses in the Bragg mirror coatings used on test-mass optics. Accurately characterizing and minimizing these losses is crucial for enhancing detector sensitivity. This paper introduces a general mathematical and statistical framework leveraging Bayesian inference to precisely analyse mechanical ring-down measurements of disk resonators, a standard method for quantifying mechanical loss in coating materials. Our approach presents a refined model that fully captures the non-linear behaviour of beam spot motion on split photodiode sensors, significantly improving upon traditional simplified exponential-decay methods. We achieve superior estimation accuracy for decay constants ($τ_1$ and $τ_2$), especially for measurements exhibiting larger oscillation amplitudes. Specifically, we observe improvements in estimation accuracy by up to 25$\%$ over traditional methods, with strong Bayesian evidence favouring our framework. Our simulations and experimental validations reveal that previously discarded measurements due to fitting inaccuracies can now be reliably analysed, maximizing the use of available data. This enhanced analytical capability not only provides more precise mechanical loss estimations but also offers deeper insights into systematic issues affecting disk resonator measurements, paving the way toward improved coating materials and ultimately, more sensitive gravitational wave detectors.