ExoJAX Retrievals of VLT/CRIRES Spectra of Luhman 16AB: C/O Ratios and Systematic Uncertainties

TL;DR

This study uses ExoJAX to perform high-resolution spectral retrievals of Luhman 16AB from VLT/CRIRES in the $K$-band, aiming to measure C/O ratios and assess systematic uncertainties. By comparing multiple CO line lists within a power-law $T$–$P$ framework and then applying a flexible Gaussian-process (GP) $T$–$P$ profile, the authors quantify how line-list choices and thermal-structure parameterizations impact inferred abundances and cloud properties. They find C/O values around $0.67$ for both components, with line-list systematics contributing the dominant $\sim7\%$ uncertainty, while the $T$–$P$ parameterization and photometric variability have smaller effects. The GP approach yields broader, more conservative temperature uncertainties and reveals localized deviations from a simple power law, illustrating the method's utility for robust atmospheric characterization and for guiding future joint analyses with JWST data. Overall, Luhman 16AB emerges as a crucial anchor for substellar C/O measurements, and the work highlights the importance of systematic tests and flexible thermal modeling in the interpretation of substellar atmospheres.

Abstract

We present atmospheric retrievals of the benchmark brown dwarf binary Luhman 16AB using high-resolution VLT/CRIRES spectra and the differentiable framework ExoJAX. We derive elemental abundances and temperature-pressure ($T$-$P$) profiles while explicitly testing the robustness of the results against major sources of systematic uncertainty. We first perform retrievals with a power-law $T$-$P$ profile and assess the sensitivity of inferred molecular abundances and C/O ratios to different CO line lists (ExoMol, HITEMP with air- and H2-broadening). We then introduce a flexible Gaussian process-based $T$-$P$ profile, allowing a non-parametric characterization of the thermal structure and a more conservative treatment of uncertainties. For both components, we infer C/O ratios of about 0.67, slightly above solar, with line list systematics at the 7 percent level emerging as the dominant source of uncertainty, whereas assumptions about $T$-$P$ parameterization or photometric variability play a lesser role. The retrieved $T$-$P$ profiles and molecular abundances are broadly consistent with atmospheric models and equilibrium chemistry. Our results establish Luhman 16AB as a key anchor for substellar C/O measurements, demonstrate the utility of flexible $T$-$P$ modeling in high-resolution retrievals, and highlight the importance of systematic tests -- particularly line list uncertainties -- for robust comparisons between brown dwarfs and giant exoplanets.

Figures (15)

• Figure 1: Comparison of C/O ratios versus mass for brown dwarfs and directly-imaged super Jupiters based on high-resolution spectra. Literature values and masses are summarized in Table \ref{['tab:co_lit']}; the measurements for Luhman 16A and 16B are from this work. Blue symbols denote objects that are wide-orbit companions, and orange symbols denote isolated objects. For Luhman 16AB, the vertical error bars include the systematic uncertainty from the choice of CO line list quantified in this study.
• Figure 2: Retrieval results for the VLT/CRIRES high-resolution spectra of Luhman 16A (left) and Luhman 16B (right), modeled with the $\mathrm{HITEMP_{H_2}\,\,CO}$ line list and a power-law $T$--$P$ profile. The wavelength coverage is split into four detector segments (chips 1–4). For each chip, the top panel shows the normalized flux data (black points) and the best-fit physical model (blue line). The bottom panel shows the residuals (data minus the best-fit model); orange lines indicate the correlated noise predicted by the GP. Each chip is normalized independently.
• Figure 3: Corner plots for the posterior samples obtained by fitting the spectra with models that use CO line lists from $\mathrm{ExoMol\,\,CO}$ (orange), $\mathrm{HITEMP_{Air}\,\,CO}$ (green), and $\mathrm{HITEMP_{H_2}\,\,CO}$ (blue). (Left) Luhman 16A. (Right) Luhman 16B.
• Figure 4: Retrieved $T$--$P$ profiles and normalized contribution fuctions for Luhman 16A (left) and Luhman 16B (right) under the $\mathrm{HITEMP_{H_2}\,\,CO}$ configuration. For each object, the left panel shows the contribution function, and the right panel shows the posterior $T$--$P$ profile. The horizontal shaded band marks the inferred cloud deck pressure, $\log_{10}P\left(\tau_c=1\right)$. Note that $\tau_c$ denotes the cumulative optical depth due only to cloud opacity.
• Figure 5: Retrieved VMRs of $\mathrm{H_2O}$ (purple), $\mathrm{CO}$ (brown), $\mathrm{CH_4}$ (pink), and $\mathrm{HF}$ (yellow) for Luhman 16A (left) and Luhman 16B (right). Points are plotted at the pressure of the contribution function peak; the vertical extent indicates the pressure range that contributes appreciably for each molecule, with the upper bound truncated at the cloud-deck pressure. For comparison, thermochemical-equilibrium VMRs computed with FastChem, including equilibrium condensation, are shown as 5–95% credible bands.