11:45 AM - 12:00 PM
[PAE18-11] Validation of High-Resolution Spectral Analysis for Brown Dwarfs Using Luhman16 AB
Keywords:atmospheric retrieval, High-resolution spectroscopy, brown dwarf, ExoJAX
Brown dwarfs are cold and faint objects with atmospheric characteristics similar to those of gas giants. Since it is difficult to observe gas giants directly with current technology, studying the atmospheres of brown dwarfs is crucial for understanding the atmospheres of these cold and faint objects.
High-resolution spectral analysis is highly effective for elucidating the atmospheric characteristics of brown dwarfs. Recent advancements in near-infrared observation technology (e.g., IRD/Subaru) have made high-resolution spectroscopic observations and analyses of brown dwarfs possible (e.g., Kawashima et al., 2024). High-resolution spectra contain crucial information on physical parameters such as molecular abundances, temperature-pressure (T-P) profiles, and surface gravity. Low- to medium-resolution spectroscopy can easily obtain spectra of faint objects but cannot resolve individual absorption lines. In contrast, high-resolution spectroscopy enables the precise detection of continuum and individual absorption line profile shapes. This allows for the estimation of molecular abundances and the T-P profile of brown dwarf atmospheres through high-resolution spectral analysis.
However, uncertainties remain in the high-resolution spectral analysis due to differences in the T-P profile models, molecular line lists, and the presence or absence of clouds in the spectral models used for analysis, and the analytical methods have not yet been fully established. For example, Picos et al. (2025) constructed spectral models using two different T-P profile models and analyzed high-resolution spectra, finding that the estimated surface gravity varied significantly depending on the T-P profile model used. Furthermore, de Regt et al. (2023) analyzed high-resolution spectra using spectral models with different H2O line lists and found that the estimated 12CO/13CO ratio varied depending on the line list used.
In this study, we analyzed the high-resolution spectra of the brown dwarf binary Luhman16 A and B (CRIRES/VLT, K-band, R~100,000; Crossfield et al., 2014) to investigate the uncertainties in high-resolution spectral analysis. Luhman16 A and B, as the closest and brightest brown dwarf binary to the Solar System, have been observed using various methods. Their masses have been dynamically determined (Lazorenko & Sahlmann, 2018), and the presence of clouds has been inferred from their photometric variability (Biller et al., 2024). Since these known observations can be compared with the results of high-resolution spectral analysis to verify their consistency, Luhman16 A and B serve as an excellent testbed for investigating uncertainties in high-resolution spectral analysis. We conducted high-resolution spectral analysis with the auto-differentiable spectral modeling tool ExoJAX (Kawahara et al., 2024) to investigate how uncertainties arise due to differences in CO line lists, the flexibility of T-P profile models, and the presence or absence of clouds.
High-resolution spectral analysis is highly effective for elucidating the atmospheric characteristics of brown dwarfs. Recent advancements in near-infrared observation technology (e.g., IRD/Subaru) have made high-resolution spectroscopic observations and analyses of brown dwarfs possible (e.g., Kawashima et al., 2024). High-resolution spectra contain crucial information on physical parameters such as molecular abundances, temperature-pressure (T-P) profiles, and surface gravity. Low- to medium-resolution spectroscopy can easily obtain spectra of faint objects but cannot resolve individual absorption lines. In contrast, high-resolution spectroscopy enables the precise detection of continuum and individual absorption line profile shapes. This allows for the estimation of molecular abundances and the T-P profile of brown dwarf atmospheres through high-resolution spectral analysis.
However, uncertainties remain in the high-resolution spectral analysis due to differences in the T-P profile models, molecular line lists, and the presence or absence of clouds in the spectral models used for analysis, and the analytical methods have not yet been fully established. For example, Picos et al. (2025) constructed spectral models using two different T-P profile models and analyzed high-resolution spectra, finding that the estimated surface gravity varied significantly depending on the T-P profile model used. Furthermore, de Regt et al. (2023) analyzed high-resolution spectra using spectral models with different H2O line lists and found that the estimated 12CO/13CO ratio varied depending on the line list used.
In this study, we analyzed the high-resolution spectra of the brown dwarf binary Luhman16 A and B (CRIRES/VLT, K-band, R~100,000; Crossfield et al., 2014) to investigate the uncertainties in high-resolution spectral analysis. Luhman16 A and B, as the closest and brightest brown dwarf binary to the Solar System, have been observed using various methods. Their masses have been dynamically determined (Lazorenko & Sahlmann, 2018), and the presence of clouds has been inferred from their photometric variability (Biller et al., 2024). Since these known observations can be compared with the results of high-resolution spectral analysis to verify their consistency, Luhman16 A and B serve as an excellent testbed for investigating uncertainties in high-resolution spectral analysis. We conducted high-resolution spectral analysis with the auto-differentiable spectral modeling tool ExoJAX (Kawahara et al., 2024) to investigate how uncertainties arise due to differences in CO line lists, the flexibility of T-P profile models, and the presence or absence of clouds.