Detecting a stochastic gravitational-wave background in the presence of correlated magnetic noise

A detection of the stochastic gravitational-wave background (SGWB) from unresolved compact binary coalescences could be made by Advanced LIGO and Advanced Virgo at their design sensitivities. However, it is possible for magnetic noise that is correlated between spatially separated ground-based detectors to mimic a SGWB signal. In this paper we propose a new method for detecting correlated magnetic noise and separating it from a true SGWB signal. A commonly discussed method for addressing correlated magnetic noise is coherent subtraction in the raw data using Wiener filtering. The method proposed here uses a parametrized model of the magnetometer-to-strain coupling functions, along with measurements from local magnetometers, to estimate the contribution of correlated noise to the traditional SGWB detection statistic. We then use Bayesian model selection to distinguish between models that include correlated magnetic noise and those with a SGWB. Realistic simulations are used to show that this method prevents a false SGWB detection due to correlated magnetic noise. We also demonstrate that it can be used for a detection of a SGWB in the presence of strong correlated magnetic noise, albeit with reduced significance compared to the case with no correlated noise. Finally, we discuss the advantages of using a global three-detector network for both identifying and characterizing correlated magnetic noise.