Early cosmology and the stochastic gravitational wave background

TL;DR

This paper explores how non-standard early universe models, like thermal inflation and early matter domination, affect the stochastic gravitational wave background, impacting detectability by LIGO and LISA.

Contribution

It analyzes the impact of alternative early cosmological epochs on gravitational wave spectra and their detectability, highlighting features like oscillations and spectrum suppression.

Findings

Second inflation reduces gravitational wave spectrum at certain scales.

Pre big bang scenario can still produce observable signals in LISA.

Early matter domination causes moderate spectrum reduction.

Abstract

The epoch when the Universe had a temperature higher than a GeV is long before any time at which we have reliable observations constraining the cosmological evolution. For example, the occurrence of a second burst of inflation (sometimes called thermal inflation) at a lower energy scale than standard inflation, or a short epoch of early matter domination, cannot be ruled out by present cosmological data. The cosmological stochastic gravitational wave background, on scales accessible to interferometer detection, is sensitive to non-standard cosmologies of this type. We consider the implications of such alternative models both for ground-based experiments such as LIGO and space-based proposals such as LISA. We show that a second burst of inflation leads to a scale-dependent reduction in the spectrum. Applied to conventional inflation, this further reduces an already disappointingly low signal. In the pre big bang scenario, where a much more potent signal is possible, the amplitude is reduced but the background remains observable by LISA in certain parameter space regions. In each case, a second epoch of inflation induces oscillatory features into the spectrum in a manner analogous to the acoustic peaks in the density perturbation spectrum. On LIGO scales, perturbations can only survive through thermal inflation with detectable amplitudes if their amplitudes were at one time so large that linear perturbation theory is inadequate. Although for an epoch of early matter domination the reduction in the expected signal is not as large as the one caused by a second burst of inflation, the detection in the context of the pre big bang scenario may not be possible since the spectrum peaks around the LIGO frequency window and for lower frequencies behaves as $f^3$.