Probing the Atmospheres of Young Long-Period Sub-Neptune Progenitors with ELT/ANDES

Abstract

High-resolution cross-correlation spectroscopy (HRCCS) has emerged as a powerful technique for detecting and characterizing exoplanet atmospheres from the ground. While it has been highly successful for ultra-hot Jupiters and hot Jupiters, next-generation facilities such as ELT/ANDES will extend its reach to smaller and longer-period planets, including young sub-Neptunes and their progenitors. We investigate whether HRCCS with ELT/ANDES can robustly recover orbital parameters and atmospheric signals of the long-period sub-Neptunes V1298Taub and TOI-451c. For such systems, the slow Doppler drift of the planetary signal over a single night reduces its separation from stationary telluric contamination, increasing the risk of signal removal while detrending. We therefore quantify the impact of including out-of-transit exposures on signal recovery and parameter estimation. We simulate transmission observations in the YJH bands using the Ratri pipeline and analyze them with the HRCCS framework Upamana. For V1298Taub, we inject atmospheric signals consistent with HST, Spitzer, and JWST constraints. For TOI-451c, we explore models spanning sub-solar to super-solar C/O ratios to assess compositional sensitivity. We find that incorporating out-of-transit exposures significantly enhances detectability, provided that the detrending effects are consistently propagated to the template models prior to cross-correlation. Without this reprocessing step, recovered orbital parameters can deviate substantially from injected values or yield reduced detection significance. For V1298Tau b, >5$σ$ detections of H2O, H2S, and CO are achievable within $\lesssim$10 hours (~2 nights). For TOI-451c, distinguishing sub-solar from super-solar C/O ratios requires $\geq$18 hours. HRCCS with ELT/ANDES will therefore be a key tool for probing the atmospheric diversity of young, long-period sub-Neptunes in the ELT era.

Figures (14)

• Figure 1: Overview of the modeling and analysis workflow used in this study. The Chemistry module (top) combines elemental abundances, opacities, and star–planet parameters to compute 1D pressure–temperature profiles using 1D radiative-convective-thermochemical equilibrium model petitCODE, which are then coupled to the 1D chemical kinetics code, VULCAN, to produce volume mixing ratio profiles under stellar UV irradiation. The Observation module (middle) uses petitRADTRANS to generate line-by-line high-resolution transmission spectra, which are propagated through the Ratri ELT/ANDES simulator together with PHOENIX stellar models, telluric transmission from ESO SkyCalc, and observatory and airmass information obtained via astroplan. The resulting wavelength-dependent signal-to-noise ratios feed into two detectability pathways: (1) a 1D "optimistic” case assuming perfect detrending, and (2) a full 3D detectability test incorporating time-variable systematics and applying the SVD + MLR detrending pipeline. This end-to-end framework is used to assess HRCCS detectability for young sub-Neptunes in this study.
• Figure 2: Comparison of the ultraviolet stellar flux spectra used to model photochemistry in the atmospheres of V1298 Tau b and TOI-451 c. The V1298 Tau spectrum from duvvuri2023high is rescaled to the stellar surface, whereas the "Gueymard-solar" spectrum provided with VULCAN is adopted as a proxy for TOI-451, whose effective temperature is comparable to that of the Sun.
• Figure 3: Vertical chemical abundance profiles for the dominant molecular species in the atmospheres of V1298 Tau b (top panel) and TOI-451 c (bottom panels). The top panel shows the disequilibrium chemistry results for V1298 Tau b computed using a self-consistent radiative–convective pressure–temperature profile and the star-planet parameters listed in Table \ref{['tab:systems']} The three bottom panels correspond to TOI-451 c models assuming sub-solar (C/O = 0.22), solar (C/O = 0.55), and super-solar (C/O = 0.80) elemental carbon-to-oxygen ratios at solar metallicity. All models include vertical mixing and photochemistry computed with VULCAN. The profiles illustrate the sensitivity of key species to elemental composition and disequilibrium processes across the observable atmosphere. Molecules shown are selected based on having average volume mixing ratios exceeding 10$^{-8}$ between the 1 bar and 0.1 mbar pressure levels, corresponding to the observable photospheric region.
• Figure 4: High-resolution transmission spectra simulated for V1298 Tau b and TOI-451 c at the resolving power of ELT/ANDES: (a) spectrum of V1298 Tau b computed using disequilibrium chemistry and a self-consistent pressure–temperature profile, while panels; (b)-(d) correspond to TOI-451 c assuming sub-solar (C/O = 0.22), solar (C/O = 0.55), and super-solar (C/O = 0.80) elemental carbon-to-oxygen ratios, respectively. For V1298 Tau b, cloud-free and cloudy atmospheric configurations are investigated, where the latter includes a gray cloud deck at 0.01 bar and an enhanced Rayleigh-scattering haze. All spectra include collision-induced absorption and Rayleigh scattering scaled by a factor of 10 and are convolved with the instrumental line-spread function of ANDES. The contributions of individual molecular species to the total transmission spectrum are shown as a function of wavelength.
• Figure 5: The variation of airmass versus exoplanet orbital phase in simulated nights using Ratri for the case of V1298 Tau b (top panel) and TOI-451 c (bottom panel). Phases falling inside the shaded region represent in-transit phases. For more discussions about the synthetic nights, see Sections \ref{['v1298taubobs']} and \ref{['toi451cobs']}.