HR 8799b: Low Temperatures, Surprising Spectrum

by Paul Gilster on August 31, 2010

Photos of the Earth from a significant distance are always fascinating, dating back to the spectacular shot of the rising Earth over lunar mountains taken by Apollo 8 in December of 1968. The image below, showing Earth and its Moon, comes from the Messenger spacecraft, taken at a distance of some 183 million kilometers. I see things like this and think about our future imaging of exoplanets, and the possibilities of space-based missions that can study their atmospheres. Learning how we look helps us understand what to look for around other stars, and also offers a bit of the ‘wow’ factor.

We’re nowhere near this kind of imaging with exoplanets but we’re getting better all the time, and that’s providing some curious results. HR 8799 is the interesting young system some 130 light years from Earth (in Pegasus) that has yielded direct images of three planets. Some eighteen months after the discovery of the system here, we’ve now managed to get a spectrum of HR 8799b, useful for what it can tell us about temperatures, clouds, and chemical composition.

Image: The Earth/Moon system as seen from the Messenger spacecraft. Credit: NASA/JHU-APL.

HR 8799b shows little or no methane in its atmosphere, a fact that used in conjunction with models of low-temperature atmospheres yields an estimate of 1200 K (926 degrees Celsius or 1700 Fahrenheit) as the coolest possible temperature for this young object. Oddly, the planet ought to be some 400 Kelvin cooler than what the new measurements show, extrapolating from the amount of energy the planet emits and its assumed age.

The culprit? Scientists at the University of Hawaii, who made these measurements at the Keck Observatory, think dust in the planet’s atmosphere must be to blame. If you change the computer models of gas giant planets to incorporate thick dust clouds, you wind up with essentially the same result. Thus direct spectroscopy of exoplanets may be telling us that young gas giants are cloudier than we had realized. The results are a reminder, as if we needed one, that exoplanets will continue to surprise us, especially given the fact that direct imaging of these worlds has just begun. Only six planets — three in this system — have been directly imaged. Says Michael Liu, co-author of the paper on this work:

“Direct studies of extrasolar planets are just in their infancy. But even at this early stage, we are learning they are a different beast than objects we have known about previously.”

True enough, and we should consider the advantages that direct imaging puts on the table. For one thing, it gives us the opportunity to study planets in wide orbits — the planets around HR 8799, discovered by Christian Marois (Herzberg Institute of Astrophysics) and team in 2008, orbit the star with semimajor axes of 24 AU, 38 AU and 68 AU respectively, with the interesting possibility of a 1:2:4 orbital resonance (not yet confirmed). At 68 AU, HR 8799b is too far from its star for radial velocity detection, but direct imaging gives us both planet and spectrum.

Image: Keck II image of the young extrasolar planet HR 8799b, seen as the point source in center of image. The bright light from the parent star HR 8799 is seen in the background in yellow/red and has been removed in an annular region centered on the planet. Credit: Brendan Bowler and Michael Liu, IfA/Hawaii.

But are the HR 8799 objects planets or brown dwarfs? At least one recent paper has suggested the latter, but this work advocates a younger age for the system and sees the objects as hot, young planets. Even so, explicit comparisons with brown dwarfs are useful, and the authors note that these objects are massive analogs of giant planets, objects that offer up similar physics, radii, effective temperatures and cooling histories. The comparison is intriguing, and allows the researchers to peg HR 8799b as a unique object:

Altogether, our spectral and photometric comparisons to field brown dwarfs suggest a spectral type between L5 and T2 for HR 8799 b. Although peculiar compared to most L and T dwarfs in the field, the planet’s photometry is consistent with the reddest field L dwarfs. These results imply that HR 8799 b is the lowest-mass L/T transition object currently known.

And later, the paper refers to the planet this way:

These observations, combined with spectroscopy of HR 8799 c and d, will elucidate the physical properties of this emerging class of low-mass objects which are characterized by low surface gravities, low luminosities, and exceptionally cloudy atmospheres.

‘An emerging class of low-mass objects’ is a phrase that reminds us how much we’ve learned in the past decade about small, cool stars and planets that skirt the boundary between planet and star. The paper is Bowler et al., “Near-Infrared Spectroscopy of the Extrasolar Planet HR 8799b,” accepted by The Astrophysical Journal (preprint).

tzf_img_post

Carl Keller August 31, 2010 at 9:51

“Thus direct spectroscopy of exoplanets may be telling us that young gas giants are cloudier than we had realized. ”

The list of Classes on http://en.wikipedia.org/wiki/Category:Extrasolar_planets may need a ‘young jovian’ or ‘young Jupiter’, young ‘Neptune’. It’s inspiring to see witness this new science as it develops. Exoplanetary science will always yield new discoveries.

Thomas August 31, 2010 at 13:02

“Photos of the Earth from a significant distance are always fascinating, dating back to the spectacular shot of the rising Earth over lunar mountains taken by Apollo 8 in December of 1968.”

Comrade?
http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19660084822_1966084822.pdf
http://www.donaldedavis.com/2004%20new/5_30_66E.jpg

And if these aren’t far enough, at least consider the Zond missions.

Thomas August 31, 2010 at 13:05

“Only six planets — three in this system — have been directly imaged.”

http://exoplanet.eu/catalog-imaging.php — Looks like 13.

Mike August 31, 2010 at 13:35

Yeah Carl you couldn’t have said it better. It’s been surprise after surprise. After all, who predicted hot Jupiters? Noooooo body!
Is it a star? Is it a planet? Depends on mass and how it formed, in a stellar formation process or in a proto-planetary disk. Not an entirely clear distinction or very well understood processes. And the debate continues.

Paul Gilster August 31, 2010 at 14:24

Thomas, unless I’m mistaken, those 13 listed at exoplanet.eu are candidates — not all have been confirmed.

Thomas August 31, 2010 at 15:52

Those are the confirmed planets. The unconfirmed ones are here, and not organised by detection method:
http://exoplanet.eu/catalog-contro.php

andy August 31, 2010 at 16:40

After all, who predicted hot Jupiters? Noooooo body!

Wrong.

Mike August 31, 2010 at 17:52

Nooooo body but Otto Struve!
I would like to point out that Gordon Walker,the Canadian astronomer who developed the radial velocity method for detecting exoplanets back in the early 1980s conducted a multi-year observing program that actually acquired data detecting hot Jupiters but he and his team didn’t recognize what they had discovered. Exoplanets around main sequence stars weren’t identified until 1995.
Otto Struve, another neglected genius like Fritz Zwicky.

Martin September 1, 2010 at 11:54

“Those are the confirmed planets. The unconfirmed ones are here, and not organised by detection method:”

4 of the items on that list are actually Brown Dwarfs.

Exciting to see spectra of these objects, more to come in the next few years I think.

andy September 1, 2010 at 14:09

4 of the items on that list are actually Brown Dwarfs.

I’d actually go further and state that the majority of these objects are probably brown dwarfs. I generally prefer the formation-based criterion to decide whether an object is a brown dwarf or a planet, rather than the criterion of deuterium fusion (which incidentally is not necessarily 13 Jupiter masses). Most of these objects have highly suspicious properties that indicate that they may have formed via star-type formation processes.

As regards the planets of HR 8799, it is interesting to note that the star is a Lambda Boötis-type chemically-peculiar star. This does not however seem to be related to HR 8799b’s peculiarities, as they note that 2M1207b (which is one of the objects that is probably a brown dwarf despite its mass below the deuterium-fusion threshold) shows similar effects.

ljk September 2, 2010 at 7:08

Exoplanet Atmospheres

Physical Processes

Sara Seager

To read the entire book description or the introduction, please visit:

http://press.princeton.edu/titles/9225.html

Over the past twenty years, astronomers have identified hundreds of extrasolar planets–planets orbiting stars other than the sun. Recent research in this burgeoning field has made it possible to observe and measure the atmospheres of these exoplanets.

This is the first textbook to describe the basic physical processes–including radiative transfer, molecular absorption, and chemical processes–common to all planetary atmospheres, as well as the transit, eclipse, and thermal phase variation observations that are unique to exoplanets.

Tobias Holbrook September 7, 2010 at 9:03

“I generally prefer the formation-based criterion to decide whether an object is a brown dwarf or a planet.”

Ah, I prefer the heat based method – is it producing a significant amount of energy through gravitational compression? I seem to recall Jupiter emits ~8 watts above what it recieves? Oh, as well as composition, of course – if it’s made of the same material, and radiates heat like, a star, but doesn’t use P-P fusion, it’s a brown dwarf.

andy September 7, 2010 at 15:27

Tobias Holbrook: surely that would result in all gas giants being classed as brown dwarfs when they are formed, and all brown dwarfs ending up as gas giants as they age?

Tobias Holbrook September 8, 2010 at 17:08

Erm… that *is* a potential flaw. But when a star leaves the main sequence, it’s no longer a star.

I don’t agree with a hard line seperating the two objects; I’d class them on composition and energy emitted. We already class brown dwarfs this way; all it takes is extending the line downwards.

Comments on this entry are closed.