2003ApJ...598L.121L

Past studies addressing the thermal atmospheric escape of hydrogen from ``hot Jupiters'' have been based on the planet's effective temperature, which, as we show here, is not physically relevant for loss processes. In consequence, these studies led to significant underestimations of the atmospheric escape rate (≤10³ g/s) and to the conclusion of long-term atmospheric stability. From more realistic exospheric temperatures, determined from X-ray and extreme-ultraviolet (XUV) irradiation and thermal conduction in the thermosphere, we find that energy-limited escape and atmospheric expansion arise, leading to much higher estimations for the loss rates (~10¹² g/s). These fluxes are in good agreement with recent determinations for HD 209458b based on observations of its extended exosphere. We also show that for young solar-type stars, which emit stronger XUV fluxes, the inferred loss rates are significantly higher. Thus, hydrogen-rich giant exoplanets under such strong XUV irradiances may evaporate down to their core sizes or shrink to levels at which heavier atmospheric constituents may prevent hydrodynamic escape. These results could explain the apparent paucity of exoplanets so far detected at orbital distances less than 0.04 AU.