The diversity of hot Jupiter atmospheres


Last week, our team published in Nature the global results of our 124-orbit Hubble Space Telescope programme to study the atmospheres of hot Jupiters with transmission spectroscopy. One orbit is 90 minutes, so the observations represent more than a week’s worth of Hubble.

Here is an earlier post about the goals of the programme:  “transit spectra with the Hubble Space Telescope, the dream”, and another post about early results.

According to the press coverage, and to some extend to the Nature paper itself, the key result of the programme is that we solved the mystery of the absence of water in hot Jupiter atmospheres.

Screen Shot 2016-01-18 at 10.26.29

Example of news coverage (source:

An earlier series of observations with the space telescope in the infrared had found that the spectroscopic signature of water was weak in several hot Jupiters – see this post:  the unexpected underabundance of water in three hot jupiters . The weak water signature could indicate that water was much scarcer than expected in those atmospheres. But it could also be explained by the effect of hazes or clouds obscuring all features in the spectra. The team publishing those results had put much emphasis on the first possibility, elaborating on the consequences of the lack of water on planet formation scenarios.

Our HST programme provides, for about ten exoplanets, a combined view of the atmospheric spectrum in the visible and infrared. The addition of data in the visible was the key factor: signatures of molecules like water tend to stand out in the infrared, while hazes and clouds are easier to detect in the visible. What we saw when confronting the two types of spectra is clear evidence for hazes and clouds on most of the planets. Globally, as shown in the figure below, the shape of the spectra was as expected with haze, and highly incompatible with the water-depletion scenario.

Plot representing the spectra of eight hot Jupiters in indices sensitive to clouds, hazes and water depletion ("sub-solar" water abundance, red line). The data clearly exclude water depletion, and favor hazes over clouds.

Plot representing the spectra of eight hot Jupiters in indices sensitive to clouds, hazes and water depletion. The data clearly exclude water depletion (red line), and favour hazes over clouds. (Source: Figure 3 in the paper)

To me, the most interesting result of the HST programme is not the issue of dryness (I confess always having believed hazes were responsible for the reduced molecular signatures), but the fact that the ten hot Jupiter atmospheres we measured are all different from each other. Try as we may – and we tried hard – these atmospheres do not fall into any easy categorising. Some of the hot ones have clouds, others don’t. Same for the cooler ones. Some have high-altitude sodium or potassium, others not. We would have loved to see them line up in a nice sequence, indicating some profound order in hot Jupiters’s atmospheres, but in fact they might be as diverse as the planets in our own Solar System. And you don’t hear anybody complain that Jupiter and Saturn look too different from each other.

Sketchy spectra for ten gas giant exoplanets. Caution: the detailed colored lines are models, the space telescope observations are the much sparser dots with error bars.

Spectra for ten gas giant exoplanets. Caution: the detailed coloured lines are models, the Space Telescope observations are the dots with error bars. (Source: Figure 1 in the paper).

When we chose “the diversity of planetary atmospheres” as a subtitle for Exoclimes, I was thinking more of the differences between hot Jupiters, water planets, hot Neptunes, and so on. I hadn’t imagined that within the same class, the diversity would be so high as well.

Feature Image: From the NASA/ESA press release.


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.