The JWST weather report from the nearest brown dwarfs III: Heterogeneous clouds and Thermochemical instabilities as possible drivers of WISE 1049AB's spectroscopic variability

TL;DR

This study uses JWST/NIRSpec time-resolved spectra (0.6–5.3 μm) of the nearby brown-dwarf binary WISE 1049AB to probe the drivers of spectroscopic variability across H2O, CH4, and CO bands. By analyzing molecular-band light curves, their correlations, and depth-dependent contribution functions, the authors show that clouds alone cannot account for the enhanced CH4 and CO variability and that thermochemical instabilities likely play a role, supported by period analyses near the known rotation periods and by 3D-like atmospheric mapping. The results imply a weather system in the L/T transition brown dwarfs where chemical disequilibrium and vertical structure modulate band-specific variability, and they demonstrate the feasibility of constructing atmospheric maps from molecular-band contributions. The work advances our understanding of brown-dwarf atmospheres and demonstrates JWST's capability to disentangle cloud and chemistry-driven variability, with implications for interpreting exoplanetary atmospheres in similar regimes.

Abstract

We present a new analysis of the spectroscopic variability of WISE~J104915.57$-$531906.1AB (WISE~1049AB, L7.5+T0.5), observed using the NIRSpec instrument onboard the James Webb Space Telescope (GO 2965 - PI: Biller). We explored the variability of the dominant molecular bands present in their 0.6--5.3~$μ$m spectra (H$_2$O, CH$_4$, CO), finding that the B component exhibits a higher maximum deviation than the A component in all the wavelength ranges tested. The light curves reveal wavelength-(atmospheric depth) and possibly chemistry-dependent variability. In particular, for the A component, the variability in the light curves at the wavelengths traced by the CH$_4$ and CO molecular absorption features is higher than that of H$_2$O, even when both trace similar pressure levels. We concluded that clouds alone are unlikely to explain the increased variability of CO and CH$_4$ with respect to H$_2$O, suggesting that an additional physical mechanism is needed to explain the observed variability. This mechanism is probably due to thermochemical instabilities. Finally, we provide a visual representation of the 3D atmospheric map reconstructed for both components using the molecular band contributions at different pressure levels and the fit of planetary-scale waves.

Figures (7)

• Figure 1: Upper panel: Median of the 21 maximum spectra (purple line) and median of the 21 minimum spectra (black line) of WISE 1049A in left panel and WISE 1049B in right panel, taken from "white light curve" (0.6 -- 5.3 $\mu$m) 7 hr NIRSpec/PRISM data. Color segments represent the range of the molecular bands selected in the spectra (H$_2$O-blue, CH$_4$-orange, CO-pink). Lower panel: Max/min flux ratio of the two spectra shown in the upper panel. The horizontal dotted line represents the wavelengths where there is no variability.
• Figure 2: WISE 1049A (left) and WISE 1049B (right) light curves for each molecular band selected of H$_2$O - blue, CH$_4$ - orange, CO - pink, sorted by wavelength (top -- 0.6 $\mu$m and bottom 5.3 $\mu$m).
• Figure 3: Left panel: WISE 1049A light curves for different molecular bands of H$_2$O - blue, CH$_4$ - orange, CO - pink, each with the best fit (black), next to each is the number of sinusoidal functions found as seen in the best fit. Right panel: the residual fit (data minus fit).
• Figure 4: Left panel: WISE 1049B light curves for different molecular bands of H$_2$O - blue, CH$_4$ - orange, CO - pink, each with the best fit (black), next to each is the number of sinusoidal functions found as seen in the best fit. Right panel: the residual fit (data minus fit).
• Figure 5: Lomb-Scargle Periodogram of WISE 1049A. The vertical lines represent the half-period at $3.46$ hr (dashed) and the period at $6.94$ hr (solid) reported by Apai2021. Each color and line style represents a different range of molecular bands selected. In addition, we add the periodogram (grey) calculated with the 'only NIRSpec light curve', also presented in biller2024jwst.