BEBOP II: Sensitivity to sub-Saturn circumbinary planets using radial-velocities

TL;DR

BEBOP II advances the detection of sub-Saturn circumbinary planets using radial velocities, revealing sensitivities down to Neptune and Saturn masses and highlighting the importance of orbit assumptions in detection limits.

Contribution

This paper introduces improved methods for identifying circumbinary planets in radial velocity data and demonstrates the impact of orbit assumptions on detection sensitivity.

Findings

Assuming circular orbits overestimates detection limits by up to 40%.

Achieved residual RMS scatter of 3 m/s after binary signal removal.

Sensitive to planets with masses down to Neptune and Saturn for periods up to 1000 days.

Abstract

BEBOP is a radial-velocity survey that monitors a sample of single-lined eclipsing binaries, in search of circumbinary planets by using high-resolution spectrographs. Here, we describe and test the methods we use to identify planetary signals within the BEBOP data, and establish how we quantify our sensitivity to circumbinary planets by producing detection limits. This process is made easier and more robust by using a diffusive nested sampler. In the process of testing our methods, we notice that contrary to popular wisdom, assuming circular orbits in calculating detection limits for a radial velocity survey provides over-optimistic detection limits by up to $40\%$ in semi-amplitude with implications for all radial-velocity surveys. We perform example analyses using three BEBOP targets from our Southern HARPS survey. We demonstrate for the first time a repeated ability to reach a residual root mean squared scatter of $3~\rm m.s^{-1}$ (after removing the binary signal), and find we are sensitive to circumbinary planets with masses down to that of Neptune and Saturn, for orbital periods up to $1000~\rm days$.