Artist’s illustration of a cloudy exoplanet. A new study suggests that the Jupiter-like exoplanet 51 Eri b may have partly cloudy skies.
Mark Garlick/University of Warwick

Direct imaging of exoplanets was once only possible for the brightest of planets orbiting the dimmest of stars — but improving technology is turning this into an increasingly powerful technique. In a new study, direct-imaging observations of the Jupiter-like exoplanet 51 Eridani b provide tantalizing clues about its atmosphere.

Direct Imaging of 51 Eri b

Images of 51 Eri b from GPI (top) and Keck (bottom).
Rajan et al. 2017

While transit detections remain the best way to discover large quantities of new exoplanets, direct imaging provides a unique advantage: you can measure the light from the exoplanet itself. With proper constraints on the host star, it therefore becomes possible to measure the spectrum of the planet’s atmosphere.

One target for this technique is 51 Eri b, a Jupiter-like exoplanet located roughly 100 light-years away. This object was the first exoplanet directly imaged by the Gemini Planet Imager Exoplanet Survey, a project that used the Gemini Planet Imager (GPI) instrument in Chile to search for exoplanets around 600 young nearby stars.

A team of scientists led by Abhijith Rajan (Arizona State University) has now made new near-infrared observations of 51 Eri b: spectroscopy in the K band using GPI, and photometry in the Ms band with a camera on the Keck I telescope in Hawaii. Rajan and collaborators combined this new data with past observations and modeling to better characterize the 51 Eri b’s properties.

Cloudy Transition

Color–magnitude diagrams for brown dwarfs and imaged exoplanets (click for a closer look). The color of 51 Eri b (marked with red star) places it among late T dwarfs, but it is redder than most comparable-temperature brown dwarfs. The tracks for the L–T transition for two different planet masses are shown.
Rajan et al. 2017

One intriguing aspect of 51 Eri b is the challenge of determining its spectral type. Though its spectrum is consistent with that of a T dwarf, photometry shows that it’s unusually red for this spectral type. There may be a reason for this, however: clouds.

Rajan and collaborators find that the best fitting models for 51 Eri b’s spectra all have an atmosphere consisting of patchy clouds. This result holds true both for models with the salt and sulfide clouds expected to condense in the atmospheres of mid-to-late T dwarfs, and for models with the iron and silicate clouds common in atmospheres of redder L-dwarfs.

The authors hypothesize that 51 Eri b may be in the process of transitioning from a warmer L-type body to a cooler T-type body. As an L-type planet cools, holes and low-opacity patches appearing in an initially uniform cloud deck could cause the transition of the planet’s spectrum to T-type.

An Unusual Start?

The ten best-fitting cloudy (red) and cloudless (blue) atmospheres over the wavelength range of JWST, illustrating that JWST will likely be able to differentiate between atmospheric models for 51 Eri b.
Rajan et al. 2017

In addition to examining 51 Eri b’s atmosphere, Rajan and collaborators use its luminosity to explore how it may have formed. They demonstrate that 51 Eri b is one of the only directly imaged planets that’s consistent with what’s known as the cold-start scenario, in which planets slowly grow via accretion onto a solid core.

While much remains to be learned about 51 Eri b, these new results provide an excellent step in the right direction. The authors also show that future observations — such as with the James Webb Telescope — will allow us to further differentiate between models describing this planet. 51 Eri b’s intriguing atmosphere makes it a prime target to revisit as our observational capabilities continue to improve.


Abhijith Rajan et al 2017 AJ 154 10. doi:10.3847/1538-3881/aa74db

Related Journal Articles

This post originally appeared on AAS Nova, which features research highlights from the journals of the American Astronomical Society.


Image of Anthony Barreiro

Anthony Barreiro

June 22, 2017 at 4:13 pm

I'm confused by the reference to L- and T- dwarfs. L, T, and Y are spectral classes of brown dwarfs, right? So is 51 Eri b a planet or a brown dwarf, or are these terms used interchangeably?

Also, some background on infrared spectroscopy would be helpful. The wikipedia article on infrared astronomy is helpful in this regard.

You must be logged in to post a comment.

Image of Margarita


June 25, 2017 at 12:37 pm

It seems that the definitions brown dwarf, sub-brown dwarf, and giant planet are under debate.

Regarding 51 Eri b
"51 Eri b has an effective temperature ranging between 605 - 737 K", Rajan et al tell us * whilst Wikipedia† tells is that "In 2009, the coolest known brown dwarfs had estimated effective temperatures between 500 and 600 K, and have been assigned the spectral class T9. "
So 51 Eri b is hotter than some of the coolest brown dwarfs.

[*Characterizing 51 Eri b from 1-5 μm: a partly-cloudy exoplanet
Abhijith Rajan, Julien Rameau et al

As background to this topic, I found this paper to be useful.
"Giant planet and brown dwarf formation
G. Chabrier, A. Johansen, M. Janson, R. Rafikov
(Submitted on 29 Jan 2014)
Understanding the dominant brown dwarf and giant planet formation processes, and finding out whether these processes rely on completely different mechanisms or share common channels represents one of the major challenges of astronomy and remains the subject of heated debates. It is the aim of this review to summarize the latest developments in this field and to address the issue of origin by confronting different brown dwarf and giant planet formation scenarios to presently available observational constraints. As examined in the review, if objects are classified as "Brown Dwarfs" or "Giant Planets" on the basis of their formation mechanism, it has now become clear that their mass domains overlap and that there is no mass limit between these two distinct populations. 
Furthermore, while there is increasing observational evidence for the existence of non-deuterium burning brown dwarfs, some giant planets, characterized by a significantly metal enriched composition, might be massive enough to ignite deuterium burning in their core. Deuterium burning (or lack of) thus plays no role in either brown dwarf or giant planet formation. Consequently, we argue that the IAU definition to distinguish these two populations has no physical justification and brings scientific confusion. In contrast, brown dwarfs and giant planets might bear some imprints of their formation mechanism, notably in their mean density and in the physical properties of their atmosphere. Future direct imaging surveys will undoubtedly provide crucial information and perhaps provide some clear observational diagnostics to unambiguously distinguish these different astrophysical objects."

You must be logged in to post a comment.

Image of Anthony Barreiro

Anthony Barreiro

June 26, 2017 at 4:02 pm

Thank you very much, Margarita. It's reassuring to know that I'm not the only one who is confused!

You must be logged in to post a comment.

You must be logged in to post a comment.