Mean opacity tables for probing the interior and atmosphere of giant planets

TL;DR

This work delivers new Rosseland and Planck mean opacity tables for giant planets, spanning $P$–$T$ conditions far beyond previous datasets and incorporating both cloud-free and cloud-inclusive (cloudy) opacities. Using the latest molecular/atomic line lists, refined pressure-broadening (including Na D and K I), and rainout condensation chemistry, the authors produce nine metallicities across $100-6000\ \mathrm{K}$ and $10^{-6}-10^{5}\ \mathrm{bar}$, with a flexible cloud framework over multiple grain sizes. They demonstrate that cloud opacities can markedly boost $\kappa_{\rm R}$ at $T \lesssim 2800$ K and modestly affect $\kappa_{\rm P}$, while high-temperature Planck opacities are highly sensitive to atomic species and resonance lines, leading to substantial deviations from older tables like F14. The publicly available cloudy tables and enhanced cloud-free data enable more realistic modeling of giant planet atmospheres and interiors, though caution is needed at pressures above ~1000 bar where non-ideal effects become important.

Abstract

We present new Rosseland and Planck mean opacity tables relevant to the shallow interiors and atmospheres of giant planets. The tables span metallicities from 0.31 to 50 times solar, temperatures from 100 - 6000 K, and pressures from 1e-6 - 1e5 bar, thereby covering a wider parameter space than previous data sets. Our calculations employ the latest molecular and atomic line lists and pressure-broadening treatments, and include contributions from collision-induced absorption, free electrons, and scattering processes. We further provide cloudy mean opacity tables that account for cloud particle extinction across a range of particle sizes and capture the sequential removal of condensates as the gas cools. We benchmark our cloud-free tables against widely used opacity tables and find significant relative differences, exceeding 100% in Rosseland mean opacities at T \gtrsim 3000 K due to the inclusion of additional short-wavelength absorbers. Differences in Planck mean opacities at high temperatures are even larger, in some cases exceeding two orders of magnitude, which is most likely driven by the inclusion of Ca, Mg, and Fe cross-sections and updated Na D and K I resonance line treatments. Cloud opacities substantially increase Rosseland mean opacities for T \lesssim 2800 K, while their effect on Planck mean opacities is weaker. We also discuss limitations of our mean opacities at high pressures, where non-ideal effects become important. This work provides improved cloud-free mean opacity tables for giant planets, as well as the first publicly available cloudy mean opacity tables, which will enable more realistic modeling of their atmospheres and interiors.

Figures (13)

• Figure 1: Number mixing ratio of species at a fixed pressure of $1 \ \rm bar$ assuming a solar composition from Asplund_2021. The left panel shows the chemistry of condensates using the rainout approach (solid curves) and equilibrium condensation (dotted curves). The right panel is the same as the left panel but showing the gas chemistry.
• Figure 2: Monochromatic opacities at 1 bar for a solar composition. Each panel applies to a different local gas temperature $T_{\rm g}$. The opacities of all neutral molecules and atoms are shown in black, CIA is brown, free-free (ff), and bound-free (bf) absorption is turquoise, Rayleigh and Thomson scattering is given in cyan, and the opacity due to clouds is yellow. To compute the cloud opacity a mean particle radius $r_{\rm g} = 1 \ \rm \mu m$ was used. The solid and dotted grey lines mark the wavelengths where the Planck function $B_{\lambda}$ and its temperature derivative $dB_\lambda/dT$, respectively reach their maxima for the local gas temperature $T_{\rm g}$.
• Figure 3: Local Rosseland mean opacity $\kappa_{\rm R}$ computed for a solar composition over the grid of temperature and pressure considered in this work. In black we show example thermal profiles which fall within our opacity grid. The Jupiter model is taken from Siebenaler_2025, the brown dwarf model is from Marley_2021, and the hot Jupiter model is from Goyal_2020.
• Figure 4: Upper panels: Local Rosseland mean opacity $\kappa_{\rm R}$ as a function of temperature for various fixed pressures. The solid curves correspond to the data from this study, while the dashed curves come from Freedman_2014. Lower panels: The relative difference in $\kappa_{\rm R}$ between our data and that of Freedman_2014 at fixed pressures. Right panels: Apply to a metallicity of $[\rm M/H] = +0.5$. Left panels: Apply to a metallicity of $[\rm M/H] = +1.7$.
• Figure 5: Upper panel: Local Planck-mean opacity $\kappa_{\rm P}$ as a function of temperature for various fixed pressures at a metallicity of $[\rm M/H] = +0.5$. The solid curves correspond to the data from this study, while the dashed curves come from Freedman_2014. Lower panel: The relative difference in $\kappa_{\rm P}$ between our data and that of Freedman_2014 at fixed pressures.