The cloudy shape of hot Jupiter thermal phase curves

Abstract: Hot Jupiters have been predicted to have a strong day/night temperature contrast and a hot spot shifted eastward of the substellar point. This was confirmed by numerous phase curve observations probing the longitudinal brightness variation of the atmosphere. Global circulation models, however, systematically underestimate the phase curve amplitude and overestimate the shift of its maximum. We use a global circulation model including non-grey radiative transfer and realistic gas and cloud opacities to systematically investigate how the atmospheric circulation of hot Jupiters varies with equilibrium temperature from 1000 to 2200K. We show that the heat transport is very efficient for cloudless planets cooler than 1600K and becomes less efficient at higher temperatures. When nightside clouds are present, the day-to-night heat transport becomes extremely inefficient, leading to a good match to the observed low nightside temperatures. The constancy of this low temperature is, however, due to the strong dependence of the radiative timescale with temperature. We further show that nightside clouds increase the phase curve amplitude and decreases the phase curve offset at the same time. This change is very sensitive to the cloud chemical composition and particle size, meaning that the diversity in observed phase curves can be explained by a diversity of nightside cloud properties. Finally, we show that phase curve parameters do not necessarily track the day/night contrast nor the shift of the hot spot on isobars, and propose solutions to to recover the true hot-spot shift and day/night contrast.