A lower limit for Newtonian-noise models of the Einstein Telescope

Abstract

The Einstein Telescope (ET) is a proposed third-generation gravitationalwave (GW) underground observatory. It will have greatly increased sensitivity compared to current GW detectors, and it is designed to extend the observation band down to a few Hz. At these frequencies, a major limitation of the ET sensitivity is predicted to be due to gravitational fluctuations produced by the environment, most importantly by the seismic field, which give rise to the so-called Newtonian noise (NN). Accurate models of ET NN are crucial to assess the compatibility of an ET candidate site with the ET sensitivity target also considering a possible reduction of NN by noise cancellation. With NN models becoming increasingly complex as they include details of geology and topography, it is crucial to have tools to make robust assessments of their accuracy. For this purpose, we derive a lower bound on seismic NN spectra, which is weakly dependent on geology and properties of the seismic field. As a first application, we use the lower limit to compare it with NN estimates recently calculated for the Sardinia and Euregio Meuse-Rhine (EMR) candidate sites. We find the utility of the method, which shows an inconsistency with the predictions for the EMR site, which indicates that ET NN models require further improvement.

Impact of correlated seismic and correlated Newtonian noise on the Einstein Telescope

“…with S ξz the PSD, or in our case the CSD, of the vertical displacement of the Rayleigh wave and ρ 0,Surface = 2 800kg m −3 the surface density [46]. γ is a factor given by the elastic properties of the half space that takes into account the partial cancellation of the NN given by the compression/decompression of the rocks caused by the Rayleigh waves below the surface: we assume γ = 0.8 (see Fig 10 of [39]).…”

“…Earlier work has studied the effect from seismic and Newtonian noise on the instantaneous sensitivity P ET (f ) [18,29,46]. It was found that seismic noise could be dominant up to ∼ 2 Hz-3 Hz rather than just below 2 Hz [18].…”