Impact of correlated magnetic noise on directional stochastic gravitational-wave background searches

TL;DR

This paper assesses how correlated magnetic noise, notably Schumann resonances, could bias directional stochastic gravitational-wave background searches with Earth-based detectors. It introduces a cross-correlation formalism that includes a magnetic-noise term and measures the magnetic correlations using external magnetometers at LVK sites, then applies this framework to ASAF, BBR, and isotropic SGWB analyses on O3 data. The key finding is that magnetic noise levels during O3 are below detector sensitivities for all targeted searches, with magnetic budgets remaining negligible for ASAF, BBR, and isotropic analyses, though A+ projections indicate possible contamination at a few narrow lines. This work provides a quantitative bound on environmental magnetic budgets and informs future SGWB analyses as detectors approach design sensitivity, highlighting the need to model phase couplings and to monitor magnetic noise in future runs.

Abstract

One potential factor that could impede searches for the stochastic gravitational-wave background (SGWB), arising from the incoherent superposition of a multitude of weak and unresolvable gravitational-wave signals in the Universe, is correlated magnetic noise at Earth-scale distances, such as the Schumann resonances. As the sensitivity of terrestrial detectors to SGWBs increases, these effects are expected to become even more pronounced, emphasizing the importance of considering correlated noise sources in SGWB searches. In this study, we explore for the first time the impact of this class of noise on anisotropic SGWB searches. We focus on the low-frequency range of 20-200 Hz with a resolution of 1/32 Hz, using data from the LEMI-120 and Metronix MFS-06e magnetometers, located at the sites of the Advanced LIGO and Advanced Virgo interferometers, respectively, during the third observing run. These measurements were then compared with the directional upper limit derived from gravitational-wave data collected during the third observing run of the LIGO-Virgo-KAGRA collaboration. The results indicate that the magnetic noise correlations are below the interferometers' sensitivity level during the O3 observing run. The potential impact of magnetic correlations on the Advanced LIGO detectors has also been investigated at the A+ design sensitivity level.

Figures (7)

• Figure 1: The effective GW strain $h_{\rm mag}(f,\boldsymbol{\hat{n}})$ related to correlated magnetic noise (left panel) and the 95% confidence Bayesian upper limits from KAGRA:2021rmt on the GW strain $h_{\rm gw}(f,\boldsymbol{\hat{n}})$ amplitude for each sky direction in the 20-200 Hz band (right panel). Both plots are derived with data from the O3 run, combining the HL, HV, and LV baselines.
• Figure 2: The ratio of the effective GW strain, $h_{\rm mag}(f,\boldsymbol{\hat{n}})$, associated with correlated magnetic noise, to the 95% confidence Bayesian upper limits from KAGRA:2021rmt on the GW strain $h_{\rm gw}(f,\boldsymbol{\hat{n}})$ amplitude, for each sky direction in the 20-200 Hz frequency range.
• Figure 3: The intensity of the effective GW energy density is depicted as a Mollweide projection of the sky in equatorial coordinates at the fourth Schumann resonance peak of 27.3 Hz.
• Figure 4: Mollweide projection in the equatorial coordinates of the ratio of the effective GW flux of magnetic origin $\mathcal{F}_{\mathrm{mag}}(f_{\rm ref},\boldsymbol{\hat{n}})$ to the BBR upper limits on the GW flux $\mathcal{F}^{95\%, \rm UL}_{\rm gw}(f_{\rm ref},\boldsymbol{\hat{n}})$ using the O3 data sets of the LVK detectors' network.
• Figure 5: Comparison of the IMB from Janssens:2022tdj expressed with the blue line and the IMB derived from the effective GW anisotropic estimators (orange line). In both analyses, the IMBs have been constructed for the HLV detector network. The black and red curves indicate the narrow-band and broad-band sensitivities for the O3 run, respectively KAGRA:2021kbb. Lastly, the green curve represents the Design A+ broad-band sensitivity A+.