[Paper Review] WASP-39b: exo-Saturn with patchy cloud composition, moderate metallicity, and underdepleted S/O

L. Carone, David Lewis+3 more|arXiv (Cornell University)|Jan 20, 2023

Astro and Planetary Science4 citations

TL;DR

This study uses a 3D general circulation model (GCM) coupled with a non-equilibrium kinetic cloud model to simulate heterogeneous cloud composition in WASP-39b’s atmosphere, revealing patchy silicate and metal oxide clouds with variable particle sizes and opacities. The results show that a moderate (10× solar) metallicity with reduced vertical mixing better explains JWST observations than high-metallicity clear-sky models, and that sulfur is under-depleted due to condensation, making it a more reliable tracer of primordial composition than iron, magnesium, or silicon.

ABSTRACT

WASP-39b is one of the first extrasolar giant gas planets that has been observed within the JWST ERS program. Fundamental properties that may enable the link to exoplanet formation differ amongst retrieval methods, for example metallicity and mineral ratios. In this work, the formation of clouds in the atmosphere of WASP-39b is explored to investigate how inhomogeneous cloud properties (particle sizes, material composition, opacity) may be for this intermediately warm gaseous exoplanet. WASP-39b's atmosphere has a comparable day-night temperature median with sufficiently low temperatures that clouds may form globally. The presence of clouds on WASP-39b can explain observations without resorting to a high (> 100x solar) metallicity atmosphere for a reduced vertical mixing efficiency. The assessment of mineral ratios shows an under-depletion of S/O due to condensation compared to C/O, Mg/O, Si/O, Fe/O ratios. Vertical patchiness due to heterogeneous cloud composition challenges simple cloud models. An equal mixture of silicates and metal oxides is expected to characterise the cloud top. Further, optical properties of Fe and Mg silicates in the mid-infrared differ significantly which will impact the interpretation of JWST observations. We conclude that WASP-39b's atmosphere contains clouds and the underdepletion of S/O by atmospheric condensation processes suggest the use of sulphur gas species as a possible link to primordial element abundances. Over-simplified cloud models do not capture the complex nature of mixed-condensate clouds in exoplanet atmospheres. The clouds in the observable upper atmosphere of WASP-39b are a mixture of different silicates and metal oxides. The use of constant particles sizes and/or one-material cloud particles alone to interpret spectra may not be sufficient to capture the full complexity available through JWST observations.

Motivation & Objective

• To investigate the impact of inhomogeneous cloud properties—such as variable particle size, composition, and opacity—on atmospheric retrieval of WASP-39b.
• To assess whether a moderate metallicity (10× solar) with reduced vertical mixing can reproduce JWST observations without requiring high-metallicity (100× solar) clear-sky models.
• To evaluate how condensation processes affect elemental ratios, particularly S/O, and assess sulfur’s potential as a formation tracer.
• To challenge the validity of simplified cloud models (e.g., grey, mono-material, constant-size particles) in interpreting JWST transmission spectra.
• To develop a physically motivated cloud microphysics framework that improves atmospheric retrieval accuracy for exoplanets.

Proposed method

• 3D GCM simulations (expeRT/MITgcm) generate atmospheric profiles used as input for a kinetic, non-equilibrium cloud formation model.
• The cloud model computes particle size, number density, and volume fraction of condensates (silicates, metal oxides, high-temperature species) across pressure levels.
• Opacity is calculated from the resulting cloud microphysical properties, accounting for wavelength-dependent optical properties of Fe and Mg silicates.
• The model incorporates pressure-dependent particle size and number density, avoiding the assumption of constant cloud parameters.
• Retrieval consistency is tested by comparing simulated transmission spectra with JWST ERS data under varying metallicity and vertical mixing assumptions.
• The model is validated against WASP-96b data, showing consistent results with reduced vertical mixing (~100× lower than standard) for both planets.

Experimental results

Research questions

• RQ1 Can a moderate-metallicity (10× solar) atmosphere with reduced vertical mixing explain the observed transmission spectrum of WASP-39b without requiring high-metallicity (100× solar) clear-sky models?
• RQ2 How does the heterogeneous composition of mixed-condensate clouds affect the interpretation of JWST mid-infrared observations?
• RQ3 To what extent is sulfur under-depleted relative to oxygen due to condensation, and can it serve as a more reliable tracer of primordial composition than iron, magnesium, or silicon?
• RQ4 How do variable particle sizes and multi-material cloud layers bias atmospheric retrievals that assume grey, mono-disperse, or single-composition clouds?
• RQ5 Can a non-equilibrium cloud model with pressure-dependent microphysics improve agreement with JWST observations compared to simplified cloud assumptions?

Key findings

• WASP-39b’s atmosphere is globally covered in clouds due to efficient day-night heat redistribution and sufficiently low temperatures, despite its intermediate equilibrium temperature (~1100 K).
• A 10× solar metallicity atmosphere with reduced vertical mixing (by ~100×) provides a better fit to JWST data than a 100× solar metallicity clear-sky model.
• The upper cloud deck consists of an almost equal mixture of silicates and metal oxides, with vertical patchiness in composition due to varying thermodynamic conditions.
• Sulphur is significantly under-depleted in the atmosphere due to condensation, making it a more robust tracer of primordial element abundances than Fe, Mg, Si, or O.
• The optical properties of Fe and Mg silicates differ substantially in the mid-infrared, meaning that multi-material cloud models are essential for accurate spectral interpretation.
• Simplifying cloud models by assuming constant particle size, single composition, or grey opacity leads to systematic biases in retrieval results, especially in near- and mid-infrared observations.