Robust and improved constraints on higher-curvature gravitational effective-field-theory with the GW170608 event

TL;DR

This paper develops a waveform model within higher-curvature effective field theory extensions of general relativity, analyzes GW170608 data, and places improved constraints on the theory's coupling constants, disfavoring non-Einsteinian effects.

Contribution

It introduces an approximate waveform model for higher-curvature gravity effects in binary black hole mergers and applies it to GW170608 to derive new constraints.

Findings

Bayesian analysis disfavors non-Einsteinian theory with log Bayes factor -2.81.

Constraints on coupling parameters are 3.5 times stronger than previous limits.

Results are independent of prior choices on the coupling parameters.

Abstract

Effective field theory methods allow us to modify general relativity through higher-curvature corrections to the Einstein-Hilbert action, while preserving Lorentz invariance and the number of gravitational degrees of freedom. We here construct an approximate inspiral-merger-ringdown waveform model within the cubic, parity-preserving class of effective-field-theory extensions to Einstein's theory for the gravitational waves emitted by quasi-circular binary black holes with aligned/anti-aligned spins. Using this waveform model, we first explore the detectability of non-Einsteinian effective-field-theory effects through an extended version of effective cycles to illustrate the need to include non-Einsteinian amplitude corrections. We then use this model to analyze the GW170608 event in a full Bayesian framework, and we place new improved and more robust constraints on the coupling constants of the effective field theory. Our Bayesian model selection study disfavors the non-Einsteinian theory with a (log) Bayes factor of $\log \mathcal{B}^{\text{EFT}}_{\text{GR}} = -2.81$. Our Bayesian parameter estimation study places the constraints $\bar{\alpha}_1=0.87^{+1.95}_{-1.03}$ and $ \bar{\alpha}_2=-0.35^{+4.12}_{-2.92}$ at $90\%$ confidence on the coupling parameters of the effective-field theory. These constraints are $3.5$ stronger than previous constraints, informative relative to the prior, and independent of the choice of prior on the coupling parameters of the modified theory.