Transit spectra with the Hubble Space Telescope: the dream


So, after the transit of Venus, back to other planetary transits far, far away. Tomorrow, the Hubble Space Telescope will turn to WASP-17, to measure the transit spectrum of its planet in the visible with the STIS instrument. This Sunday, it will measure the transit of WASP-6.

Last year we were awarded 200 hours of Hubble Space Telescope time to attempt to measure the transit spectrum of a collection of hot Jupiters. It’s not a typo: two hundred hours, or 124 orbits with the Space Telescope! As far as these things go, it is a massive amount of observations.

The idea is to collect the transmission spectra of eight hot Jupiters encompassing visible wavelengths down to 300 nanometers. We already have decent atmospheric transmission spectra in the visible for two planets, HD 189733b and HD 209458b. The eight new targets are chosen to cover a wider range in planet temperature and mass.

The Space Telescope was collecting similar measurements last year, but in the near infrared. This is a programme lead by Drake Deming from NASA. The near infrared is sampling the signatures of molecules such as water and methane, while in the visible we expect to see things like sodium and potassium, titanium oxide, and dust grains. The main advantage of working in the visible though, is that the accuracy of the measurement can be much higher  (simply because visible-light detectors are more precise).

The eight targets were chosen primarily because of their high temperature, large size and low mass. This implies that their atmospheres are very extended, which makes transmission spectroscopy easier. The atmospheric scale heights range from 500 km to 1700 km (Jupiter has a scale height of 30 km, it is 200 km for HD189733b). The target planets also cover a good fraction of the temperature and gravity span of hot Jupiter atmospheres.

Target planets for the visible transit HST programme in mass vs temprature. The bands show my guess for three interesting possible "transition regions". The two green dots are HD 189733b and HD 209458b, for which we already have decent spectra (I'll let you guess which is which).

So what do we expect to see? Silicate and iron dust, titanium oxide vapour, carbon or sulphur soot. The temperatures of the targets cross three interesting – and still speculative – regimes: the formation of silicate and iron dust, the inversion of the temperature profile and formation of a hot stratosphere due to Titanium oxide absorption, and the breakdown of day-night heat recirculation at high temperatures.

Jonathan Fortney ran a set of models for WASP-17 b for instance. The planet sits close to the temperature where titanium oxide is expected to condensate. TiO is an extremely efficient absorber of visible light. We could either be seeing most of the visible spectrum of WASP-17 b blocked out by TiO lines, or not at all, leaving the sodium and potassium lines to dominate. The figure below shows two possibilities, with either TiO and hydrogen sulphide are as culprit for the temperature inversion observed in the day-side infrared spectrum. TiO would block a large swath of the visible light in the transmission spectrum, while HS would produce a sharp rise of opacity at blue wavelengths.

Two possible transmission spectra for WASP-17 b, according to different assumptions about the nature of the stratospheric absorber. (Figure from the Space Telesope proposal by David Sing).

So, that’s the dream anyway. Will the observations really be that good? Will the features be that clear? We’ll know soon. About half the data has already been collected, most of the rest will be in by the end of this summer.

Feature Image: [collage from various sources]


About Author

I am a professor of planetary science at the University of Exeter. My specialty is the study of exoplanets, in particular the observation and modelling of exoplanet atmospheres. I have done my PhD a the University of Geneva and worked in Chile, France and Switzerland.