NEAR-INFRARED THERMAL EMISSION FROM TrES-3b: A Ks-BAN DETECTION AND AN H-BAND UPPER LIMIT ON THE DEPTH OF THE SECONDARY ECLIPSE

We present H- and Ks-band photometry bracketing the secondary eclipse of the hot Jupiter TrES-3b using the Wide-field Infrared Camera on the Canada-France-Hawaii Telescope. We detect the secondary eclipse of TrES-3b with a depth of 0.133(-0.016)(+0.018)% in the Ks band (8 sigma)-a result that is in sharp contrast to the eclipse depth reported by de Mooij & Snellen. We do not detect its thermal emission in the H band, but place a 3 sigma limit of 0.051% on the depth of the secondary eclipse in this band. A secondary eclipse of this depth in Ks requires very efficient day-to-nightside redistribution of heat and nearly isotropic reradiation, a conclusion that is in agreement with longer wavelength, mid-infrared Spitzer observations. Our 3 sigma upper limit on the depth of our H- band secondary eclipse also argues for very efficient redistribution of heat and suggests that the atmospheric layer probed by these observations may be well homogenized. However, our H- band upper limit is so constraining that it suggests the possibility of a temperature inversion at depth, or an absorbing molecule, such as methane, that further depresses the emitted flux at this wavelength. The combination of our near-infrared measurements and those obtained with Spitzer suggests that TrES-3b displays a near-isothermal dayside atmospheric temperature structure, whose spectrum is well approximated by a blackbody. We emphasize that our strict H- band limit is in stark disagreement with the best-fit atmospheric model that results from longer wavelength observations only, thus highlighting the importance of near-infrared observations at multiple wavelengths, in addition to those returned by Spitzer in the mid-infrared, to facilitate a comprehensive understanding of the energy budgets of transiting exoplanets.