Helium Atmospheres May Hide in Current Exoplanet Analysis Frameworks

TL;DR

This work tackles a key bias in exoplanet atmosphere retrievals: the common fixing of the helium-to-hydrogen ratio to a solar-system value. By treating He/H$_2$ as a free parameter and applying a proof-of-concept to HD 209458 b with JWST data, the authors show that He-rich solutions can reproduce the transmission spectrum, often reducing the inferred water abundance and increasing the mean molecular weight $μ$, thereby mimicking volatile-rich or cloudy scenarios. The study highlights a degeneracy between $μ$, $H$, and absorber abundances and argues that this can bias metallicity inferences if He/H$_2$ is held fixed. It proposes multiple complementary diagnostics—pressure-broadening differences, outflow fractionation, and atmospheric chemistry—to disentangle He-rich from volatile-rich atmospheres and calls for higher-fidelity opacity, outflow, and chemistry models to enable robust, unbiased inferences about exoplanet atmospheric composition and formation histories.

Abstract

The increasing number of detailed exoplanet observations offers an opportunity to refine our analyses and interpretations. Here, we show that atmospheres that appear volatile-rich and/or cloudy may instead be helium-rich. As transmission spectra constrain the atmospheric scale height ($H$), a He-enriched atmosphere can be misinterpreted as H$_2$-dominated water-rich to bring the mean molecular weight ($μ$) to intermediate values ($\sim$4$-$10) when He/H$_2$ is fixed. Similarly, a cloud deck can reduce the spectral features, and thus the apparent (i.e., cloud-free equivalent) $H$. We present a proof-of-concept reanalysis of HD~209458~b's JWST transmission spectrum treating He/H$_2$ as a free parameter, resulting in sets of He-rich solutions. We argue that He enhancement must be considered to reliably constrain atmospheric composition, be sensitive to a more diverse planetary population, and ultimately yield robust trends to inform formation and evolution pathways. Looking ahead, we suggest leveraging insights from differences in pressure-broadening effects, outflow measurements, and atmospheric chemistry to disentangle reliably between He-, volatile-rich, and cloudy atmospheres -- while recognizing that associated models need targeted upgrades to reach the fidelity level required to this end.

Figures (6)

• Figure 1: The two regimes of effects of water abundance on a transmission spectrum. At low water abundances (MR$_{\rm H_2O} \lesssim 1\%$), increasing the amount of water primarily increases $\rm H_2O$ absorption features while leaving the rest of the spectrum mostly unchanged. At high water abundances (MR$_{\rm H_2O} \gtrsim 5\%$), increasing the water content primarily raises the atmospheric mean molecular weight, shrinking the atmospheric scale height and compressing the entire spectrum. This effect is expected for other volatiles and can be generalized to "volatile-rich" atmospheres.
• Figure 2: JWST's spectrum of HD 209458 b is consistent with water-rich or He-rich cloud-free atmosphere. Left: Best fits to HD 209458 b's transmission spectrum from the Log$_{10}$[He/H$_2$]$<-0.5$ (blue) and $>0.5$ (red) subsample of solutions. Data from Xue2024. Center: Posterior probability distributions (PPDs) for water abundance under low (red) and high (blue) He/H$_2$ ratios. Higher helium enrichment leads to lower inferred H$_2$O, since the increased mean molecular weight alone can match the scale height of the atmosphere. Right: The same comparison for the trace gas H$_2$S shows negligible differences between the two subsamples, as non-dominant absorbers are less sensitive to assumptions about bulk atmospheric composition.
• Figure 3: Normalized posterior distribution of the mean molecular weight $\mu$ for cloud-free HD 209458 b peaking above the canonical $\mu \simeq 2.3$ expected for a He/H$_2$ ratio of 0.157.
• Figure 4: Forming a He-dominated atmosphere for a giant exoplanet. Schematic representation of the "before" and "after" states of atmospheric evolution for a stratified giant atmosphere -- e.g., via helium rain. He fractionation is enhanced via the formation of a H$_2$-dominated He-poor layer at the top of the atmosphere, allowing for the preferential loss of H$_2$ and preservation of He as seen for HD 209458 b, for example Xing2023.
• Figure 5: Spectroscopic differences between a H$_2$- and He-dominated atmosphere. Left: Best fits to HD 209458 b's transmission spectrum from Fig.\ref{['figure:bestfit']} (Log$_{10}$[He/H$_2$]$<-0.5$ in blue and $>0.5$ in red) propagated on a wider wavelength range to present where the H$_2$- and He-dominated scenarios diverge, esp. due to scattering ($\lesssim1\mu$m) and collision-induced absorption (CIA, $\sim2.4\mu$m) -- see dotted blue line for the H$_2$-only signal. While differences in scattering can yield $\sim$100-ppm signals, hazes will often overpower the scattering of H$_2$/He Sing2016. Similarly, for volatile-rich atmospheres, H$_2$'s CIA signal will not shine between features of, e.g., H$_2$O. (In comparison, H$_2$'s CIA signal is clearly detectable for $\lesssim$100 ppm water abundance, dashed blue line.) Right: Synthetic transmission spectra from the best-estimate parameters for HD 209458 b using perturbed cross-sections with doubled and halves pressure-broadening coefficients from Niraula2022 to simulate the detectable effects of H$_2$- vs He-broadened absorption features of other gases.