Atmospheric circulation of brown dwarfs and directly imaged exoplanets driven by cloud radiative feedback: global and equatorial dynamics

Abstract

Brown dwarfs, planetary-mass objects and directly imaged giant planets exhibit significant observational evidence for active atmospheric circulation, raising critical questions about mechanisms driving the circulation, its fundamental nature and time variability. Our previous work has demonstrated the crucial role of cloud radiative feedback on driving a vigorous atmospheric circulation using local models that assume a Cartesian geometry and constant Coriolis parameters. In this study, we extend the models to a global geometry and explore properties of the global dynamics. We show that, under relatively strong dissipation in the bottom layers of the model, horizontally isotropic vortices are prevalent at mid-to-high latitudes while large-scale zonally propagating waves are dominant at low latitudes near the observable layers. The equatorial waves have both eastward and westward phase speeds, and the eastward components with typical velocities of a few hundred m s−1 usually dominate the equatorial time variability. Lightcurves of the global simulations show variability with amplitudes from 0.5 percent to a few percent depending on the rotation period and viewing angle. The time evolution of simulated lightcurves is critically affected by the equatorial waves, showing wave beating effects and differences in the lightcurve periodicity to the intrinsic rotation period. The vertical extent of clouds is the largest at the equator and decreases poleward due to the increasing influence of rotation with increasing latitude. Under weaker dissipation in the bottom layers, strong and broad zonal jets develop and modify wave propagation and lightcurve variability. Our modeling results help to qualitatively explain several features of observations of brown dwarfs and directly imaged giant planets, including puzzling time evolution of lightcurves, a slightly shorter period of variability in IR than in radio wavelengths, and the viewing angle dependence of variability amplitude and IR colors.

Aerosols in Exoplanet Atmospheres

“…Tan and Showman (2017) included a parameterization of silicate condensation and latent heating in their GCM study and found that isolated silicate storms can occur when the condensation level sinks below the radiative‐convective boundary due to the onset of moist convection, which could explain the inferred patchiness of clouds and temporal variability of objects at the L‐T transition. Variability is also likely impacted by the rotation rate and cloud radiative feedback (Tan & Showman, 2019, 2020, 2021), such that changes in the Coriolis force with latitude could lead to corresponding changes in cloud opacity and patchiness inline with observations of brown dwarfs at different inclinations, where objects viewed equator‐on are redder and more variable than objects viewed pole‐on (Vos et al., 2017).…”

Mapping stellar surfaces III: An Efficient, Scalable, and Open-Source Doppler Imaging Model

“…This may be an acceptable prior if truly nothing is known about the typical angular scale of features on the surface, but this is seldom the case. Both observations of the Sun and magnetohydrodynamic simulations of stellar magnetic fields give us prior constraints on the typical scales of starspots (e.g., Berdyugina 2005;Solanki et al 2006), and climate models of brown dwarfs tell us about typical sizes of clouds and other atmospheric features (e.g., Tan & Showman 2019, 2021a. These can be encoded into our Gaussian prior by choosing a prior variance that scales with spherical harmonic degree, as in Equation ( 31), where the variance at each degree is parametrized as, e.g.,…”

Radiative-dynamical Simulation of Jupiter’s Stratosphere and Upper Troposphere

“…Radiative transfer equations including multiple scattering are solved with a two-stream approximation (Kylling et al 1995). A similar numerical implementation has been applied to GCMs of exoplanets and brown dwarfs (Komacek et al 2017;Tan & Showman 2021). An internal heat flux of 7.48 W m −2 (Li et al 2018b) is also added as a radiative flux at the bottom of our model.…”