Evidence for Atmospheric Cold-trap Processes in the Non-inverted Emission Spectrum of Kepler-13Ab Using HST/WFC3

We observed two eclipses of the Kepler-13A planetary system, on UT 2014 April 28 and UT 2014 October 13, in the near-infrared using Wide Field Camera 3 on the Hubble Space Telescope. By using the nearby binary stars Kepler-13BC as a reference, we were able to create a differential light curve for Kepler-13A that had little of the systematics typically present in HST/WFC3 spectrophotometry. We measure a broadband (1.1$\mu$m to 1.65$\mu$m) eclipse depth of $734\pm28$ ppm, and are able to measure the emission spectrum of the planet at $R\approx50$ with an average precision of 70 ppm. Our observations do not well sample either the eclipse ingress or egress, and so we are not able to provide meaningful constraints on the eclipse timing offset observed by Shporer et al. (2014). We do find that our observations, combined with those of Shporer et al. (2014) give an average dayside brightness temperature of 3000 K, and are consistent with a non-inverted, monotonically decreasing vertical temperature profile at 2.4$\sigma$. We exclude an isothermal profile and an inverted profile. We also find that the dayside emission of Kepler-13Ab appears generally similar to an isolated M7 brown dwarf at a similar effective temperature. Due to the relatively high mass and surface gravity of Kepler-13Ab, we suggest that the apparent lack of an inversion is due to cold-trap processes in the planet's atmosphere. Using a toy-model for where cold-traps should inhibit inversions, and observations of other planets in this temperature range with measured emission spectra, we argue that with more detailed modeling and more observations we may be able to place useful constraints on the size of condensates on the daysides of hot Jupiters.