The CARMENES search for exoplanets around M dwarfs

Abstract

Context. The nearby ultra-compact multiplanetary system YZ Ceti consists of at least three planets, and a fourth tentative signal. The orbital period of each planet is the subject of discussion in the literature due to strong aliasing in the radial velocity data. The stellar activity of this M dwarf also hampers significantly the derivation of the planetary parameters. Aims. With an additional 229 radial velocity measurements obtained since the discovery publication, we reanalyze the YZ Ceti system and resolve the alias issues. Methods. We use model comparison in the framework of Bayesian statistics and periodogram simulations based on a method by Dawson and Fabrycky to resolve the aliases. We discuss additional signals in the RV data, and derive the planetary parameters by simultaneously modeling the stellar activity with a Gaussian process regression model. To constrain the planetary parameters further we apply a stability analysis on our ensemble of Keplerian fits. Results. We find no evidence for a fourth possible companion. We resolve the aliases: the three planets orbit the star with periods of 2.02 d, 3.06 d, and 4.66 d. We also investigate an effect of the stellar rotational signal on the derivation of the planetary parameters, in particular the eccentricity of the innermost planet. Using photometry we determine the stellar rotational period to be close to 68 d and we also detect this signal in the residuals of a three-planet fit to the RV data and the spectral activity indicators. From our stability analysis we derive a lower limit on the inclination of the system with the assumption of coplanar orbits which is imin = 0.9 deg. From the absence of a transit event with TESS, we derive an upper limit of the inclination of imax = 87.43 deg. Conclusions. YZ Ceti is a prime example of a system where strong aliasing hindered the determination of the orbital periods of exoplanets. Additionally, stellar activity influences the derivation of planetary parameters and modeling them correctly is important for the reliable estimation of the orbital parameters in this specific compact system. Stability considerations then allow additional constraints to be placed on the planetary parameters.

The GAPS programme at TNG

“…It has also often been found to be in agreement with Hill stability criterion and dynamical simulations (e.g. Zinzi & Turrini 2017;Agnew et al 2018;Petit et al 2018;Stock et al 2020).…”

“…Ultimately, AMD-stability appears to be possible for more than 90% of the simulated systems only in the case of Neptunemass (M inj ≥ 10) planets orbiting on relatively low-eccentric (e inj ≤ 0.30) orbits in the outer portion (a inj > 0.5 au) of the habitable zone. Barring resonances and secular interaction that can stabilise an AMD-unstable system or vice versa (Zinzi & Turrini 2017;Agnew et al 2018;Petit et al 2018;Stock et al 2020) that are not yet considered by the AMD stability test, we therefore expect yet-undetected temperate Neptune-mass planetary bodies, if any, to be present in the BD-11 4672 system only on low-to moderate-eccentricity orbits on the outer region of the circumstellar habitable zone.…”

The CARMENES search for exoplanets around M dwarfs

“…Moreover, every peak at a frequency f has a corresponding peak at a frequency − f , and both peaks in the positive and negative frequency domain are subject to aliasing. In general, aliases can appear at frequencies f alias = |m f + n f W | (Dawson & Fabrycky 2010;Van-derPlas 2018;Stock et al 2020).…”

CARMENES input catalog of M dwarfs

“…Being designed for high-precision RV measurements of cool stars, it comprises two spectrograph channels with wavelength ranges of 0.52 to 0.96 µm and 0.96 to 1.71 µm, respectively, with spectral resolutions of R ≈ 80 000-96 000 (Quirrenbach et al 2014) capable of providing an RV accuracy of ∼ 1.2 m s −1 (Bauer et al 2020). More than 350 stars from the CARMENES guaranteed time observations (GTO) sample have been monitored since January 2016, resulting in 26 planet detections to date (e.g., Bluhm et al 2020;Nowak et al 2020, see Sabotta et al 2021 for an exhaustive review), as well as several planet confirmations (Trifonov et al 2018(Trifonov et al , 2020Sarkis et al 2018;Lalitha et al 2019;Stock et al 2020a).…”