Distorted, non-spherical transiting planets: impact on the transit depth and on the radius determination

Abstract: We quantify the systematic impact of the non-spherical shape of transiting planets and brown dwarfs, due to tidal forces and rotation, on the observed transit depth. Such a departure from sphericity leads to a bias in the derivation of the transit radius from the light curve and affects the comparison with planet structure and evolution models which assume spherical symmetry. As the tidally deformed planet projects its smallest cross section area during the transit, the measured effective radius is smaller than the one of the unperturbed spherical planet. This effect can be corrected by calculating the theoretical shape of the observed planet. We derive simple analytical expressions for the ellipsoidal shape of a fluid object (star or planet) accounting for both tidal and rotational deformations and calibratre it with fully numerical evolution models in the 0.3Mjup-75Mjup mass range. Our calculations yield a 20% effect on the transit depth, i.e. a 10% decrease of the measured radius, for the extreme case of a 1Mjup planet orbiting a Sun-like star at 0.01AU. For the closest planets detected so far (< 0.05 AU), the effect on the radius is of the order of 1 to 10%, by no means a negligible effect, enhancing the puzzling problem of the anomalously large bloated planets. These corrections must thus be taken into account for a correct determination of the radius from the transit light curve. Our analytical expressions can be easily used to calculate these corrections, due to the non-spherical shape of the planet, on the observed transit depth and thus to derive the planet's real equilibrium radius. They can also be used to model ellipsoidal variations of the stellar flux now detected in the CoRoT and Kepler light curves. We also derive directly usable analytical expressions for the moment of inertia, oblateness and Love number (k_2) of a fluid planet as a function of its mass.