Radial velocity detection of the TRAPPIST-1 planetary system with SPIRou and NIRPS

Le résumé fourni par la source

Results of the radial velocity (RV) modeling of the TRAPPIST-1 system using the near-infrared spectrographs SPIRou and NIRPS. The system is modeled with three components: Systematics model: fitting for residual telluric signal in the SPIRou time series. Two models are considered a Gaussian process (GP) model with a stochastically-driven harmonic oscillator kernel (SHO, Foreman-Mackey et al. 2017; Foreman-Mackey 2018), parametrized by the standard deviation of the process $\sigma_\text{syst}$, the damping timescale $\tau_\text{syst}$ and the undamped period $\rho_\text{syst}$ the sum of two sinusoids, with respective amplitudes ($A_1$, $A_2$), phases ($\Phi_1$ and $\Phi_2$) and periods ($P_\text{syst}$ and $P_\text{syst}/2$) Stellar activity model: characterizing the stellar activity RV signal. Two models are considered a GP with an SHO kernel, parametrized by $\sigma_\text{act}$, $\tau_\text{act}$ and $\rho_\text{act}$ a GP model with a quasi-periodic (QP) kernel, parametrized by $B$, the amplitude of the GP, $C$, an additive factor impacting the amplitude, $L$, the timescale and $P$, the rotation period (Foreman-Mackey et al. 2017) Planetary model: characterizing the planetary RV signal of TRAPPIST-1. Two parametrizations are considered Seven Keplerian (individual) model, where all the planets are classically modeled by their own independant circular Keplerian with fixed ephemeris. This model is described by seven parameters: the RV semi-amplitudes $K_\text{b}$, $K_\text{c}$, $K_\text{d}$, $K_\text{e}$, $K_\text{f}$, $K_\text{g}$, $K_\text{h}$ Combined model, where we inform the relative masses of the TRAPPIST-1 planets on the TTV model. We thus only fit for the absolute scale of the system with $K_\text{b}$ by reparametrizing the RV semi-amplitudes as $K_j\longrightarrow K_\text{b}\cdot(K_{j,\,\text{TTV}}/K_{\text{b},\,\text{TTV}})$, with $j=\{\text{b,}\,\text{c,}\,\text{d,}\,\text{e,}\,\text{f,}\,\text{g,}\,\text{h}\}$. The best-fit baseline model (as shown in evidence_and_Kb_all_models.pdf) uses GP SHO for the systematics and stellar activity, and the combined scheme for the planets. The posterior distribution is shown in cornerplot_main_parameters_baseline_model_spirou_nirps_nightbin.pdf with the fit over the data in rv_fit_baseline_model_spirou_nirps_nightbin.pdf. The model comparison presented in evidence_and_Kb_all_models.pdf shows that perfectly consistent $K_\text{b}$ are recovered in all the different combinations of the three model components. The planetary RV detection is thus not an artifact of the chosen systematics and activity models. We present here three additional posterior distributions: for the alternative treatment of systematics, using a sinusoidal model instead of the baseline GP SHO (cornerplot_main_parameters_alternative_syst_model_spirou_nirps_nightbin.pdf) for the alternative treatment of stellar activity, using a GP QP instead of the baseline GP SHO (cornerplot_main_parameters_alternative_act_model_spirou_nirps_nightbin.pdf) for the alternative treatments of systematics and stellar activity, using a sinusoidal model and a GP QP, respectively (cornerplot_main_parameters_alternative_syst+act_model_spirou_nirps_nightbin.pdf)

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.