[Paper Review] The GAPS Programme with HARPS-N@TNG XIV. Investigating giant planet migration history via improved eccentricity and mass determination for 231 transiting planets

A. S. Bonomo, S. Desidera+35 more|Max Planck Institute for Plasma Physics|Apr 2, 2017

Stellar, planetary, and galactic studiesPhysics and Astronomy146 references64 citations

TL;DR

A Bayesian homogeneous analysis of RV data (literature + 782 HARPS-N measurements) for 231 transiting giant planets to refine orbital, planetary, and stellar parameters, with emphasis on eccentricity and migration history.

ABSTRACT

We carried out a Bayesian homogeneous determination of the orbital parameters of 231 transiting giant planets (TGPs) that are alone or have distant companions; we employed DE-MCMC methods to analyse radial-velocity (RV) data from the literature and 782 new high-accuracy RVs obtained with the HARPS-N spectrograph for 45 systems over 3 years. Our work yields the largest sample of systems with a transiting giant exoplanet and coherently determined orbital, planetary, and stellar parameters. We found that the orbital parameters of TGPs in non-compact planetary systems are clearly shaped by tides raised by their host stars. Indeed, the most eccentric planets have relatively large orbital separations and/or high mass ratios, as expected from the equilibrium tide theory. This feature would be the outcome of high-eccentricity migration (HEM). The distribution of $α=a/a_R$, where $a$ and $a_R$ are the semi-major axis and the Roche limit, for well-determined circular orbits peaks at 2.5; this also agrees with expectations from the HEM. The few planets of our sample with circular orbits and $α>5$ values may have migrated through disc-planet interactions instead of HEM. By comparing circularisation times with stellar ages, we found that hot Jupiters with $a < 0.05$ au have modified tidal quality factors $10^{5} < Q'_p < 10^{9}$, and that stellar $Q'_s > 10^{6}-10^{7}$ are required to explain the presence of eccentric planets at the same orbital distance. As a by-product of our analysis, we detected a non-zero eccentricity for HAT-P-29; we determined that five planets that were previously regarded to have hints of non-zero eccentricity have circular orbits or undetermined eccentricities; we unveiled curvatures caused by distant companions in the RV time series of HAT-P-2, HAT-P-22, and HAT-P-29; and we revised the planetary parameters of CoRoT-1b.

Motivation & Objective

• Determine a homogeneous set of orbital and physical parameters for 231 known transiting giant planets.
• Quantify and refine orbital eccentricities and their uncertainties using a Bayesian DE-MCMC approach.
• Investigate tidal evolution and migration scenarios shaping the observed eccentricities and separations.
• Identify long-term RV trends due to distant companions and update planetary/system parameters.
• Provide constraints on tidal quality factors and assess implications for planet migration histories.

Proposed method

• Compile literature RV datasets with at least four measurements per system and add 782 HARPS-N RVs for 45 systems.
• Fit RV data with Keplerian models, including options for long-term drifts or curved trends to account for outer companions or activity cycles.
• Use differential evolution Markov chain Monte Carlo (DE-MCMC) to sample posterior distributions of orbital parameters.
• Incorporate priors on transit epoch Tc, period P, and secondary eclipse times Te to constrain e cos ω and e sin ω.
• Adopt a Jacobian-aware parameterization to ensure proper priors on eccentricity, and perform Bayesian model selection via Bayes factors to decide circular vs eccentric orbits.
• Combine RV results with literature transit and stellar parameters to derive planetary masses, densities, and gravities.

Experimental results

Research questions

• RQ1 What are the precise orbital eccentricities of 231 transiting giant planets when combining heterogeneous RV data with new HARPS-N observations?
• RQ2 How do tidal interactions and migration mechanisms (high-eccentricity vs disc migration) shape the observed eccentricity and semi-major axis distributions for hot/warm Jupiters?
• RQ3 What are the updated planetary masses, densities, and surface gravities when analyzed uniformly across the sample?
• RQ4 Do RV time series show long-term trends or curvature indicative of distant companions, and how do these affect derived orbital solutions?
• RQ5 What constraints can be placed on tidal quality factors (Q'p, Q's) and stellar ages to explain circularisation timescales and the presence of eccentric planets?

Key findings

• Eccentricities of non-compact TGPs are consistent with tides shaping their orbits, suggesting migration from highly eccentric paths.
• The distribution of α = a / aR for well-determined circular orbits peaks around 2.5, compatible with high-eccentricity migration scenarios.
• Hot Jupiters with a < 0.05 au imply planetary modified tidal quality factors 10^5 ≲ Q′p ≲ 10^9 and stellar Q′s ≳ 10^6–10^7 are needed to explain observed eccentricities.
• A non-zero eccentricity e = 0.104−0.018+0.021 is detected for HAT-P-29; several planets once thought eccentric are now circular or have undetermined eccentricities (CoRoT-2b, CoRoT-23b, TrES-3b, HAT-P-23b, WASP-54b).
• Evidence of curvatures and long-term trends in RV data points to distant companions in systems like HAT-P-2, HAT-P-22, and HAT-P-29; the orbit of HAT-P-17c is significantly improved; CoRoT-1b is more inflated than previously reported.