Dusty phenomena in the vicinity of giant exoplanets

Abstract

Context. Hitherto, searches for exoplanetary dust have focused on the tails of decaying rocky or approaching icy bodies only at short circumstellar distances. At the same time, dust has been detected in the upper atmospheric layers of hot jupiters, which are subject to intensive mass loss. The erosion and/or tidal decay of hypothetic moonlets might be another possible source of dust around giant gaseous exoplanets. Moreover, volcanic activity and exozodiacal dust background may additionally contribute to exoplanetary dusty environments. Aims. In the present study, we look for photometric manifestations of dust around different kinds of exoplanets (mainly giants). Methods. We used linear approximation of pre- and post-transit parts of the long-cadence transit light curves (TLCs) of 118 Kepler objects of interest after their preliminary whitening and phase-folding. We then determined the corresponding flux gradients G1 and G2, respectively. These gradients were defined before and after the transit border for two different time intervals: (a) from 0.03 to 0.16 days and (b) from 0.01 to 0.05 days, which correspond to the distant and adjoining regions near the transiting object, respectively. Statistical analysis of gradients G1 and G2 was used for detection of possible dust manifestation. Results. It was found that gradients G1 and G2 in the distant region are clustered around zero, demonstrating the absence of artifacts generated during the light curve processing. However, in the adjoining region, 17 cases of hot jupiters show significantly negative gradients, G1, whereas the corresponding values of G2 remain around zero. The analysis of individual TLCs reveals the localized pre-transit decrease of flux, which systematically decreases G1. This effect was reproduced with the models using a stochastic obscuring precursor ahead of the planet. Conclusions. Since only a few TLCs show the presence of such pre-transit anomalies with no analogous systematic effect in the post-transit phase, we conclude that the detected pre-transit obscuration is a real planet-related phenomenon. Such phenomena may be caused by dusty atmospheric outflows or background circumstellar dust compressed in front of the mass-losing exoplanet, the study of which requires dedicated physical modeling and numeric simulations. Of certain importance may be the retarding of exozodiacal dust relative to the planet by the Poynting-Robertson effect leading to dust accumulation in electrostatic or magnetic traps in front of the planet.