THE STATISTICS OF ALBEDO AND HEAT RECIRCULATION ON HOT EXOPLANETS

Abstract

[Abridged] If both the day-side and night-side effective temperatures of a planet can be measured, it is possible to estimate its Bond albedo, 0<A_B<1, as well as its day-night heat redistribution efficiency, 0<epsilon<1. We attempt a statistical analysis of the albedo and redistribution efficiency for 24 transiting exoplanets that have at least one published secondary eclipse. For each planet, we show how to calculate a sub-stellar equilibrium temperature, T_0, and associated uncertainty. We then use a simple model-independent technique to estimate a planet's effective temperature from planet/star flux ratios. We use thermal secondary eclipse measurements -those obtained at lambda>0.8 micron- to estimate day-side effective temperatures, T_d, and thermal phase variations -when available- to estimate night-side effective temperature. We strongly rule out the "null hypothesis" of a single A_B and epsilon for all 24 planets. If we allow each planet to have different parameters, we find that low Bond albedos are favored (A_B<0.35 at 1 sigma confidence), which is an independent confirmation of the low albedos inferred from non-detection of reflected light. Our sample exhibits a wide variety of redistribution efficiencies. When normalized by T_0, the day-side effective temperatures of the 24 planets describe a uni-modal distribution. The dimensionless quantity T_d/T_0 exhibits no trend with the presence or absence of stratospheric inversions. There is also no clear trend between T_d/T_0 and T_0. That said, the 6 planets with the greatest sub-stellar equilibrium temperatures (T>2400 K) have low epsilon, as opposed to the 18 cooler planets, which show a variety of recirculation efficiencies. This hints that the very hottest transiting giant planets are qualitatively different from the merely hot Jupiters.