The impact of compact binary confusion noise on tests of fundamental physics with next-generation gravitational-wave detectors

Next-generation ground-based gravitational-wave observatories such as the Einstein Telescope and Cosmic Explorer will detect $O(10^{5}-10^{6})$ signals from compact binary coalescences every year, the exact number depending on uncertainties in the binary merger rate. Several overlapping signals will be present in band at any given time, generating a confusion noise background. We study how this confusion noise affects constraints on possible deviations from general relativity induced by modified gravity and environmental effects. Confusion noise impacts only the signals that last longer in band. Even for a "golden" GW170817-like signal, the constraints broaden by a factor in the range $[10\%,40\%]$ $([70\%,110\%])$ for the fiducial (highest) value of the local binary neutron star merger rate. Our ability to test general relativity or constrain environmental effects will be limited by systematic errors, and not by confusion noise.