Detection of carbon monoxide in two very different planets


An outline of the two papers: Detection of CO in the high-resolution day-side spectrum of HD189733b by de Kok et al., and Detection of CO and water absorption lines in an exoplanet atmosphere by Konopacky et al.

Significant detection of molecular features are still rare enough to constitute big pieces of news in the field of exoplanet atmospheres. Recently, absorption by carbon monoxide was detected in two very different planets and using two very different methods.

de Kok et al. detected CO in the hot Jupiter HD 189733b, the most intensely scrutinized exoplanet at present, using the very tricky technique developped by the group of Ignas Snellen at Leiden University. With the CRIRES infrared spectrograph on the VLT, they obtained K-band (2.0-2.3 microns) spectra during three nights, and looked for a set of narrow spectral lines from the CO molecules. The tricky part is to separate the lines from the atmosphere of the planet from the dominant contribution of the star by using the different Doppler shifts of the two bodies as they orbit each other. The stellar lines move in one direction at low velocity, whereas the planetary lines whizz across at high velocity.

Water, methane and CO2 also have lines in this spectral range, but were not detected.

In terms of interpretation, the detection of CO in absoprtion implies that the temperature decreases with altitude in the regions of the atmosphere were these lines are formed. In our last paper, we hypothesised that the temperature could be increasing because of the effect of silicate dust, but this does not seem to be the case.

Konopacki and et al. detected CO in the young gas giant HR 8799c, the second companion in the spectacular system, HD 8799, identified by direct imaging. They used the integral-field OSIRIS spectrograph on Keck II. In this case the planet can be separated from the star in the images, and a direct spectrum of the planet, also around 2 microns, was obtained. The “band-head” of CO was detected in the spectrum (a band-head is the side of a series of closely packed molecular lines in the spectrum). Water was also detected, and methane not seen.


FIGURE: the observed spectrum of HR 8799b (black) and a model spectrum (green). The CO band head is visible as a drop near 2.29 microns [Figure 2 of Konopacki et al.]

FIGURE: the observed spectrum of HR 8799c (black) and a model spectrum (green). The CO band head is visible as a drop near 2.29 microns [Figure 2 of Konopacki et al.]

Apart from the simple feat of the detection of a molecular feature in an exoplanet atmosphere, the detection of CO rather than methane implies the presence of vigorous mixing in the atmosphere of the planet HR 8799c. At the temperatures in the part of the atmosphere seen in those spectrum (~1100 K), CO turns into methane (CH4). Only if the methane is rapidly mixed into lower, hotter regions inside the atmosphere can it be “cooked” back into CO before it accumulates. According to the authors, the observation of CO implies a mixing time smaller than 2 days in the atmosphere of this planet.

Interestingly, in spite of the fundamental differences in technique and type of planet, the two results are similar in terms of telescope size, wavelengths, and detection significance.

Both HD 189733b and HR 8977c are exceptionally favourable systems to observe, and extending these methods to other objects will present new challenges.


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.