No Arabic abstract
There has been increasing progress toward detailed characterization of exoplanetary atmospheres, in both observations and theoretical methods. Improvements in observational facilities and data reduction and analysis techniques are enabling increasingly higher quality spectra, especially from ground-based facilities. The high data quality also necessitates concomitant improvements in models required to interpret such data. In particular, the detection of trace species such as metal oxides has been challenging. Extremely irradiated exoplanets (~3000 K) are expected to show oxides with strong absorption signals in the optical. However, there are only a few hot Jupiters where such signatures have been reported. Here we aim to characterize the atmosphere of the ultra-hot Jupiter WASP-33b using two primary transits taken 18 orbits apart. Our atmospheric retrieval, performed on the combined data sets, provides initial constraints on the atmospheric composition of WASP-33b. We report a possible indication of aluminum oxide (AlO) at 3.3-sigma significance. The data were obtained with the long slit OSIRIS spectrograph mounted at the 10-meter Gran Telescopio Canarias. We cleaned the brightness variations from the light curves produced by stellar pulsations, and we determined the wavelength-dependent variability of the planetary radius caused by the atmospheric absorption of stellar light. A simultaneous fit to the two transit light curves allowed us to refine the transit parameters, and the common wavelength coverage between the two transits served to contrast our results. Future observations with HST as well as other large ground-based facilities will be able to further constrain the atmospheric chemical composition of the planet.
We present an atmospheric transmission spectrum for the ultra-hot Jupiter WASP-121b, measured using the Space Telescope Imaging Spectrograph (STIS) onboard the Hubble Space Telescope (HST). Across the 0.47-1 micron wavelength range, the data imply an atmospheric opacity comparable to - and in some spectroscopic channels exceeding - that previously measured at near-infrared wavelengths (1.15-1.65 micron). Wavelength-dependent variations in the opacity rule out a gray cloud deck at a confidence level of 3.8-sigma and may instead be explained by VO spectral bands. We find a cloud-free model assuming chemical equilibrium for a temperature of 1500K and metal enrichment of 10-30x solar matches these data well. Using a free-chemistry retrieval analysis, we estimate a VO abundance of -6.6(-0.3,+0.2) dex. We find no evidence for TiO and place a 3-sigma upper limit of -7.9 dex on its abundance, suggesting TiO may have condensed from the gas phase at the day-night limb. The opacity rises steeply at the shortest wavelengths, increasing by approximately five pressure scale heights from 0.47 to 0.3 micron in wavelength. If this feature is caused by Rayleigh scattering due to uniformly-distributed aerosols, it would imply an unphysically high temperature of 6810+/-1530K. One alternative explanation for the short-wavelength rise is absorption due to SH (mercapto radical), which has been predicted as an important product of non-equilibrium chemistry in hot Jupiter atmospheres. Irrespective of the identity of the NUV absorber, it likely captures a significant amount of incident stellar radiation at low pressures, thus playing a significant role in the overall energy budget, thermal structure, and circulation of the atmosphere.
We present the optical transmission spectrum of the hot Jupiter WASP-104b based on one transit observed by the blue and red channels of the DBSP spectrograph at the Palomar 200-inch telescope and 14 transits observed by the MuSCAT2 four-channel imager at the 1.52 m Telescopio Carlos Sanchez. We also analyse 45 additional K2 transits, after correcting for the flux contamination from a companion star. Together with the transit light curves acquired by DBSP and MuSCAT2, we are able to revise the system parameters and orbital ephemeris, confirming that no transit timing variations exist. Our DBSP and MuSCAT2 combined transmission spectrum reveals an enhanced slope at wavelengths shorter than 630 nm and suggests the presence of a cloud deck at longer wavelengths. While the Bayesian spectral retrieval analyses favour a hazy atmosphere, stellar spot contamination cannot be completely ruled out. Further evidence, from transmission spectroscopy and detailed characterisation of the host stars activity, is required to distinguish the physical origin of the enhanced slope.
We present ground-based, spectroscopic observations of two transits of the ultra-hot Jupiter WASP-121b covering the wavelength range $approx$500 - 950 nm using Gemini/GMOS. We use a Gaussian process framework to model instrumental systematics in the light curves, and also demonstrate the use of the more generalised Students-T process to verify our results. We find that our measured transmission spectrum, whilst showing overall agreement, is slightly discrepant with results obtained using HST/STIS, particularly for wavelengths shortward of $approx$650 nm. In contrast to the STIS results, we find evidence for an increasing blueward slope and little evidence for absorption from either TiO or VO in our retrieval, in agreement with a number of recent studies performed at high-resolution. We suggest that this might point to some other absorbers, particularly some combination of recently detected atomic metals, in addition to scattering by hazes, being responsible for the excess optical absorption and observed vertical thermal inversion. Our results are also broadly consistent with previous ground-based photometry and 3D GCM predictions, however, these assumed different chemistry to our retrievals. In addition, we show that the GMOS observations are repeatable over short periods (days), similarly to the HST/STIS observations. Their difference over longer periods (months) could well be the result of temporal variability in the atmospheric properties (i.e. weather) as predicted by theoretical models of ultra-hot Jupiters; however, more mundane explanations such as instrumental systematics and stellar activity cannot be fully ruled out, and we encourage future observations to explore this possibility.
We obtained long-slit optical spectroscopy of one transit of WASP-48b with the Optical System for Imaging and low-Intermediate-Resolution Integrated Spectroscopy (OSIRIS) spectrograph at the 10.4 m Gran Telescopio Canarias (GTC). We integrated the spectrum of WASP-48 and one reference star in several channels with different wavelength ranges, creating numerous color light curves of the transit. We fit analytic transit curves to the data taking into account the systematic effects present in the time series in an effort to measure the change of the planet-to-star radius ratio ($R_p/R_s$) across wavelength. After removing the transit model and systematic trends to the curves we reached precisions between 261 ppm and 455-755 ppm for the white and spectroscopic light curves, respectively. We obtained $R_p/R_s$ uncertainty values between $0.8 times 10^{-3}$ and $1.5times 10^{-3}$ for all the curves analyzed in this work. The measured transit depth for the curves made by integrating the wavelength range between 530 nm and 905 nm is in agreement with previous studies. We report a relatively flat transmission spectrum for WASP-48b with no statistical significant detection of atmospheric species, although the theoretical models that fit the data more closely include of TiO and VO.
We present a new ground-based visible transmission spectrum of the high-gravity, hot Jupiter HAT-P-23b, obtained as part of the ACCESS project. We derive the spectrum from five transits observed between 2016 and 2018, with combined wavelength coverage between 5200 {AA} - 9269 {AA} in 200 {AA} bins, and with a median precision of 247 ppm per bin. HAT-P-23bs relatively high surface gravity (g ~ 30 m/s^2), combined with updated stellar and planetary parameters from Gaia DR2, gives a 5-scale-height signal of 384 ppm for a hydrogen-dominated atmosphere. Bayesian models favor a clear atmosphere for the planet with the tentative presence of TiO, after simultaneously modeling stellar contamination, using spots parameter constraints from photometry. If confirmed, HAT-P-23b would be the first example of a high-gravity gas giant with a clear atmosphere observed in transmission at optical/NIR wavelengths; therefore, we recommend expanding observations to the UV and IR to confirm our results and further characterize this planet. This result demonstrates how combining transmission spectroscopy of exoplanet atmospheres with long-term photometric monitoring of the host stars can help disentangle the exoplanet and stellar activity signals.