Interior–atmosphere modelling to assess the observability of rocky planets with JWST

¹ Max-Planck-Institut für Astronomie, Königstuhl 17, 69117 Heidelberg, Germany
e-mail: acuna@mpia.de
² Aix-Marseille Université, CNRS, CNES, Institut Origines, LAM, Marseille, France
³ Institut universitaire de France (IUF), 75231 Paris Cedex 05, France

Received: 20 December 2022
Accepted: 24 April 2023

Abstract

Context. Super-Earths present compositions dominated by refractory materials. However, there is a degeneracy in their interior structure between a planet that has no atmosphere and a small Fe content, and a planet that has a thin atmosphere and a higher core mass fraction. To break this degeneracy, atmospheric characterisation observations are required.

Aims. We present a self-consistent interior–atmosphere model to constrain the volatile mass fraction, surface pressure, and temperature of rocky planets with water and CO₂ atmospheres. The parameters obtained in our analysis can be used to predict observations in emission spectroscopy and photometry with JWST, which can determine the presence of an atmosphere and, if present, its composition.

Methods. We coupled a 1D interior model with a supercritical water layer to an atmospheric model. In order to obtain the bolometric emission and Bond albedo for an atmosphere in radiative-convective equilibrium, we used a low-resolution k-correlated atmospheric model. We generated emission spectra with the same atmospheric model at a higher resolution (R = 200–300). An adaptive Markov chain Monte Carlo was employed for an efficient sampling of the parameter space at low volatile mass fractions.

Results. From our interior structure retrieval, we conclude that TRAPPIST-1 c most likely has a bare surface, although the presence of an atmosphere cannot be ruled out. We estimate a 1σ confidence interval of the surface pressure for a water-dominated atmosphere of P_surf = 40 ± 40 bar. We generated spectra for these two scenarios to compare with the emission flux of TRAPPIST-1 c recently observed in the MIRI F1500W filter. This is compatible with bare rock surfaces or a thin atmosphere with little or no CO₂. In the case of 55 Cancri e, a combined spectrum with NIRCam and MIRI LRS may present high uncertainties at wavelengths between 3 and 3.7 µm. However, this does not affect the identification of H₂O because it does not present spectral features in this wavelength range.