Transmission spectra of highly irradiated extrasolar planet atmospheres
by Lisa Nortmann
Date of Examination:2015-11-19
Date of issue:2016-11-11
Advisor:Prof. Dr. Stefan Dreizler
Referee:Prof. Dr. Stefan Dreizler
Referee:Prof. Dr. Artie Hatzes
Files in this item
Name:nortmann_phd_thesis.pdf
Size:7.59Mb
Format:PDF
Abstract
English
The ground-based facilities ESO/VLT+FORS2 (Very Large Telescope + FOcal Reducer and low dispersion Spectrograph) and GTC+OSIRIS (Gran Telescopio CANARIAS + Optical System for Imaging and low Resolution Integrated Spectroscopy) were used to probe the atmospheres of the two hot Jupiter planets WASP-17b and HAT-P-32Ab using multi-object-spectrophotometry. For WASP-17b, observed with FORS2, 14 transit light curves were extracted from the observed spectra and for HAT-P-32Ab, observed with OSIRIS, 20 light curves. Each light curve represents a 20 nm wide wavelength interval and was fitted with an analytical transit model to determine the wavelength dependent transit depth. For both planets strong non-astrophysical systematic noise, caused by the instrument, was found to affect the light curves. This thesis presents new findings regarding the source, nature and proper correction of these instrument specific and observing set-up dependent systematics, as well as the results for the planet transmission spectra. By including terms for the correction of the systematic noise signals into the light curve modeling, results for the transit depths were reached with uncertainties between 108 and 297 ppm for HAT-P-32Ab and between 127 and 481 ppm for WASP-17b. The results for WASP-17b show a large-scale trend towards larger transit depth with shorter wavelength indicating strong absorption in the blue. Furthermore, these results act as a possible connection between two spectrophotometric studies in neighboring wavelength regions which are separated by a $>0.2\%$ gap in transit depth. Neither a hot atmosphere containing titanium and vanadium oxide nor a cool atmosphere free of these molecules is fully consistent with the measurements. In contrast, a flat transmission spectrum was measured for HAT-P-32Ab, which could be indicative of a cloud layer at high altitude masking any atmospheric features below. Other possible explanations are an atmospheric depletion in alkali metals or a smaller than expected atmospheric scale height caused by either a lower planet equilibrium temperature or a heavier atmospheric mean molecular weight. The result is in excellent agreement with another ground-based study probing HAT-P-32Ab's optical transmission spectrum. This independent reproduction of compatible results demonstrates that ground-based measurements of exoplanet atmospheres can achieve robust and dependable measurements if the sources of systematic noise are well studied and corrected for appropriately. Furthermore, an optical spectrum of the M-dwarf HAT-P-32B was obtained to correct its flux contribution to the HAT-P-32Ab transit light curves. By comparing the spectrum to PHOENIX stellar atmosphere models an effective temperature of $T_\mathrm{eff} = 3187^{+60}_{-71}$\,K was derived. This value is slightly lower than results obtained by other groups in studies using broadband infrared data only.
Keywords: stellar astrophysics; extrasolar planet atmospheres; HAT-P-32b; WASP-17b; spectroscopy