3.6 and 4.5 micron phase curves of HD 189733b



There is good news and bad news in this new paper.

The good news is that there is an extraordinary new series of measurements from the Spitzer Space Telescope on the best-measured exoplanet, HD 189733b. Knutson et al. present two phase curves of the planet over its entire 2.2-day orbit, in the 3.6 micron and 4.5 micron bands of the IRAC instrument. The time series clearly show the signal of the changing brightness of the planet in the infrared as the day and night sides alternate.  One transit and two secondary eclipse are also clearly measured.

Knutson et al. Fig. 6 and 7

Phase curve of HD 189733b at 3.6 microns (left) and 4.5 microns (right) measured with the Spitzer Space Telescope. Figures 6 and 7 of Knutson et al. (2012).

The same team had already obseved the phase curve of ‘189 at two longer wavelengths, 8 microns and 24 microns. The 8-micron data is at the source of the famous “temperature map” of  the atmosphere of the planet. The eastward shift of the hottest point in this map has confirmed the model prediction that hot Jupiter atmospheres are dominated by powerful eastwards jets.

The new 3.6 and 4.5 micron phase curves show the same eastward shifts.  Since the peak of the planet’s thermal flux is near 3 microns, this shows that the temperature maps derived from the 8 micron data correctly reflect the heat distribution in the atmosphere of the planet (as opposed to tracing some anomalous absorber that would affect the spectrum at 8 microns).

Going into more details, Knutson et al. show that the observed position of the hottest spot is about twice closer to the centre of the day side than predicted by the models. This can mean two things: either the winds are slower than we think, or the heat is lost faster than we think. Slower winds could be due to the planet rotating more slowly than it orbits (thus being out of tidal synchronisation, a very unlikely behaviour on such a close orbit), or to winds being dragged down by turbulence more than the model predicts. A larger heat loss could be due to the presence of more heavy elements in the atmosphere (heavier atoms have more free electrons available to radiate heat away).

And the bad news? The new data, and re-analysis of previous data by the same team, show that previous measurements were much more strongly affected by the difficulty of dealing with Spitzer instrument corrections than we previously thought. It turns out that a large part of the “classic” 8 micron phase curve is actually rendered unuseable by instrumental effects.  Moreover, the eclipse depths at 3.6 and 4.5 microns turn out to need corrections much larger than their formal error bars. This is a stark warning for theorist trying to match models too closely to the Spitzer observations.

Knutson et al. 2012 Fig. 10

Brightness of the day side and night side of HD 189733b in the Spitzer passbands (in blue) compared to predictions from standard atmosphere models. The models struggle to explain all observations simultaneously. Figure 10 from Knutson et al.

There is one sign, though, that we are on the right track: the two main observations that are overturned are the position of the coolest spot on the atmosphere and the high brightness temperature at 3.6 microns. These were actually the two features that model-makers like Adam Showman had the most difficulty reproducing.

Feature Image: The IRAC instrument. NASA. Found here.


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.