Exoplanet Atmosphere Retrievals in 3D Using Phase Curve Data with ARCiS: Application to WASP-43b^★

¹ SRON Netherlands Institute for Space Research, Sorbonnelaan 2, 3584 CA Utrecht, The Netherlands
e-mail: m.min@sron.nl
² Centre for Exoplanet Science, University of St Andrews, North Haugh, St Andrews, UK
e-mail: klc20@st-andrews.ac.uk

Received: 1 December 2021
Accepted: 26 May 2022

Abstract

Aims. Our goal is to create a retrieval framework which encapsulates the three-dimensional (3D) nature of exoplanet atmospheres, and to apply it to observed emission phase curve and transmission spectra of the ‘hot Jupiter’ exoplanet WASP-43b.

Methods. We present our 3D framework, which is freely available as a stand-alone module from GitHub. We use the atmospheric modelling and Bayesian retrieval package ARCiS (ARtful modelling Code for exoplanet Science) to perform a series of eight 3D retrievals on simultaneous transmission (HST/WFC3) and phase-dependent emission (HST/WFC3 and Spitzer/IRAC) observations of WASP-43b as a case study. Via these retrieval setups, we assess how input assumptions affect our retrieval outcomes. In particular we look at constraining equilibrium chemistry vs. a free molecular retrieval, the case of no clouds vs. parametrised clouds, and using Spitzer phase data that have been reduced from two different literature sources. For the free chemistry retrievals, we retrieve abundances of H₂O, CH₄, CO, CO₂, AlO, and NH₃ as a function of phase, with many more species considered for the equilibrium chemistry retrievals.

Results. We find consistent super-solar C/O (0.6–0.9) and super-solar metallicities (1.7–2.9 dex) for all retrieval setups that assume equilibrium chemistry. We find that atmospheric heat distribution, hotspot shift (≈15.6° vs. 4.5° for the different Spitzer datasets), and temperature structure are very influenced by the choice of Spitzer emission phase data. We see some trends in molecular abundances as a function of phase, in particular for CH₄ and H₂O. Comparisons are made with other studies of WASP-43b, including global climate model (GCM) simulations, available in the literature.

Conclusions. The parametrised 3D setup we have developed provides a valuable tool to analyse extensive observational datasets such as spectroscopic phase curves. We conclude that further near-future observations with missions such as the James Webb Space Telescope and Ariel will greatly improve our understanding of the atmospheres of exoplanets such as WASP-43b. This is particularly evident from the effect that the current phase-dependent Spitzer emission data has on retrieved atmospheres.