Kepler planets: a tale of evaporation

(Abridged) Inspired by the Kepler planet discoveries, we consider the thermal contraction of planets close to their parent star, under the influence of evaporation. The mass-loss rates are based on hydrodynamic models of evaporation that include both X-ray and EUV irradiation. We find that only low-mass planets with hydrogen envelopes are significantly affected by evaporation, with evaporation being able to remove massive hydrogen envelopes inward of 0.1 AU for Neptune-mass objects. We construct a theoretical population of planets with varying core masses, envelope masses, orbital separations, and stellar spectral types, and compare these against the sizes and densities measured for low-mass planets, both in the Kepler mission and from radial velocity surveys. This exercise leads us to conclude that evaporation is the driving force of evolution for close-in Kepler planets. In fact, some 50% of the Kepler planet candidates may have been significantly eroded. Evaporation explains two striking correlations observed in these objects: a lack of large radius/low density planets close to the stars, and a bimodal distribution in planet sizes with a deficit of planets around 2R_E. Planets that have experienced high X-ray exposures are generally smaller than this size, and those with lower X-ray exposures are typically larger. A bimodal distribution is naturally explained by the evaporation model, where, depending on their X-ray exposure, close-in planets can either hold on to hydrogen envelopes 1% in mass, or be stripped entirely. To quantitatively reproduce the observed features, we argue that not only do low-mass Kepler planets need to be made of rocky cores overlaid with hydrogen envelopes, but few of them should have initial masses above 20 M_E, and the majority of them should have core masses of a few Earth masses.