Identification and monitoring of scattered light noise sources in laser interferometers with adaptive algorithms

Gravitational waves (GWs) are ripples in space-time produced by accelerating compact objects such as a binary system made of two black holes or neutron stars and about to merge. To detect GWs, km-scale instruments such as LIGO and Virgo, which are Michelson interferometers, are needed. Such detectors need to be extremely sensitive, and all noise sources, i.e., fundamental, environmental, or technical, that could pollute the detector’s output need to be identified, controlled, and mitigated. Scattered light is a non-stationary noise that can couple with the differential arm motion (DARM) degree of freedom and decrease detectors’ sensitivity. Scattering is often caused by an excess of microseismicity at the detector’s location. While localization of scattering culprits by experimental means can be difficult and time-consuming, adaptive algorithms can be used. Adaptively decomposing DARM provides the instantaneous amplitude, well correlated with the culprit’s predictor, which helps detector characterization efforts. Adaptive algorithms used are empirical mode decomposition and time-varying filter EMD . A daily monitoring system of scattering sources is developed and successfully applied to both Virgo and LIGO data during days affected by scattering. A large scale analysis carried out on Virgo data allowed us to introduce automation and improve overall performances of the monitoring system. The high values of correlation obtained suggest that adaptive decomposition can be used to precisely subtract scattering, improving parameter estimation and sky localization of the astrophysical source. The topic of scattering sources identification and monitoring in GW detectors by means of adaptive algorithms is reviewed.