First images of a planet in the act of forming

Hydrogen-alpha light is a dead giveaway that the planet is still growing.

Read the whole story

 

[greatn]

Maybe someone with greater astrophsyics credentials could explain, but this sounds similar to the phenomenon we're seeing on that one star where it was postulated that there could be an artificial structure(like a Dyson sphere) because over a regular period the light is decreasing at certain intervals. Could that one also be the formation of a planet we are seeing?

 

[Statistical]

Maybe someone with greater astrophsyics credentials could explain, but this sounds similar to the phenomenon we're seeing on that one star where it was postulated that there could be an artificial structure(like a Dyson sphere) because over a regular period the light is decreasing at certain intervals. Could that one also be the formation of a planet we are seeing?

The "stange" KIC star was strange because there was no interval. No periodic pattern just seemingly random reductions in observed light.

 

[Joriarty]

DOI link is broken – there's a %20 that needs to be deleted.

 

[SirBedwyr]

I'd like to ask a super dumb question. (Sorry in advance.)


If you're making an observation that gathers about 20-30 degrees of rotation about the center, how do you gather the orientation of the system's disc? That is, the artist's rendition has us peaking "underneath" with the planets forward and overhead of the star. How do we determine that this is the correct orientation and that the planets are not overhead and behind the star as would be the case in an optical illusion? Is the star's light in that upper right region different than when the planets are on the back-side of their orbit? I'm also assuming we haven't observed a full orbit yet.

 

[seaborgium]

SirBedwyr[/url]":1iadhw1p]I'd like to ask a super dumb question. (Sorry in advance.)


If you're making an observation that gathers about 20-30 degrees of rotation about the center, how do you gather the orientation of the system's disc? That is, the artist's rendition has us peaking "underneath" with the planets forward and overhead of the star. How do we determine that this is the correct orientation and that the planets are not overhead and behind the star as would be the case in an optical illusion? Is the star's light in that upper right region different than when the planets are on the back-side of their orbit? I'm also assuming we haven't observed a full orbit yet.

I imagine it has to do with the curvature of the observed orbits. The concave part of the curve would point toward the star and the convex part would point away.

 

[Joriarty]

greatn[/url]":2fs2mztf]Maybe someone with greater astrophsyics credentials could explain, but this sounds similar to the phenomenon we're seeing on that one star where it was postulated that there could be an artificial structure(like a Dyson sphere) because over a regular period the light is decreasing at certain intervals. Could that one also be the formation of a planet we are seeing?

Not likely. KIC 8462852 is a main-sequence star. To form a planet you need a protoplanetary disk: these are easily inferable because of their strong infra-red and sub-mm emission.

Even a planetary collision – which could create a debris field weird enough to create the irregular dimming we saw – seems an unlikely explanation. Such a collision would also create an excess of infrared light, and we have not seen an IR excess from KIC 8462852.

 

[genphp]

SirBedwyr[/url]":20qpk5pt]I'd like to ask a super dumb question. (Sorry in advance.)

As my old professor once told me "there are no dumb questions when seeking to enlighten one self".


I think the artist took a little "artistic liberties ", but im sure someone who knows more than me will give you a better explanation

Really excited about this !

 

[Joriarty]

SirBedwyr[/url]":2glaqewn]I'd like to ask a super dumb question. (Sorry in advance.)


If you're making an observation that gathers about 20-30 degrees of rotation about the center, how do you gather the orientation of the system's disc? That is, the artist's rendition has us peaking "underneath" with the planets forward and overhead of the star. How do we determine that this is the correct orientation and that the planets are not overhead and behind the star as would be the case in an optical illusion? Is the star's light in that upper right region different than when the planets are on the back-side of their orbit? I'm also assuming we haven't observed a full orbit yet.

Good question! Really good question. It's not always easy to tell.

With an inclined disk, you're gonna see asymmetries in the brightness of each side. But is the asymmetry due to forward-scattering of starlight from optically thin material at the outer rim of the near side of the disk, or from backscattered light from optically thick material on the other side?

Recent results reckon that we're looking at the forward-scattering case, where the northwestern edge is closer to us (section 6.1 of the paper is especially relevant). Basically, we have to model both scenarios and see which computer model best fits the observations. It's not immediately obvious.

 

[SirBedwyr]

Joriarty[/url]":1au6n06x]

SirBedwyr[/url]":1au6n06x]I'd like to ask a super dumb question. (Sorry in advance.)


If you're making an observation that gathers about 20-30 degrees of rotation about the center, how do you gather the orientation of the system's disc? That is, the artist's rendition has us peaking "underneath" with the planets forward and overhead of the star. How do we determine that this is the correct orientation and that the planets are not overhead and behind the star as would be the case in an optical illusion? Is the star's light in that upper right region different than when the planets are on the back-side of their orbit? I'm also assuming we haven't observed a full orbit yet.

Good question! Really good question. It's not always easy to tell.

With an inclined disk, you're gonna see asymmetries in the brightness of each side. But is the asymmetry due to forward-scattering of starlight from optically thin material at the outer rim of the near side of the disk, or from backscattered light from optically thick material on the other side?

Recent results reckon that we're looking at the forward-scattering case, where the northwestern edge is closer to us (section 6.1 of the paper is especially relevant). Basically, we have to model both scenarios and see which computer model best fits the observations. It's not immediately obvious.

Ok, so it's not totally obvious or trivial then. Could you take a stab at it based purely on geometry (ignoring scattering) given a full 360 degree orbit's worth of information?

 

[AliceWonder]

So it really doesn't take 6 days to create a planet and seed it with life?

 

[Z1ggy]

AliceWonder[/url]":2mzcaavw]So it really doesn't take 6 days to create a planet and seed it with life?

sure it does.

for sufficiently large values of 'day'

 

[greatn]

For further inquiries about time length and days, please consult with the world's foremost experts, Morris Day and the Time.

 

[Chuckstar]

Statistical[/url]":2xvy4izr]

Maybe someone with greater astrophsyics credentials could explain, but this sounds similar to the phenomenon we're seeing on that one star where it was postulated that there could be an artificial structure(like a Dyson sphere) because over a regular period the light is decreasing at certain intervals. Could that one also be the formation of a planet we are seeing?

The "stange" KIC star was strange because there was no interval. No periodic pattern just seemingly random reductions in observed light.

That's not why it was strange. We didn't observe it long enough to know if there was an interval or not.

What was strange was that the star was too old to just now have a planet forming and there also weren't the infrared emissions we would expect from a protoplanetary disk or other such dust formation.

 

[Billy_ca]

How do we know that these celestial bodies have cleared their neighbouring regions of planetesimals yet? Shouldn't we hold off on calling them planets until we've established that?

 

[dermott]

seaborgium[/url]":fmr9rbxo]

SirBedwyr[/url]":fmr9rbxo]I'd like to ask a super dumb question. (Sorry in advance.)


If you're making an observation that gathers about 20-30 degrees of rotation about the center, how do you gather the orientation of the system's disc? That is, the artist's rendition has us peaking "underneath" with the planets forward and overhead of the star. How do we determine that this is the correct orientation and that the planets are not overhead and behind the star as would be the case in an optical illusion? Is the star's light in that upper right region different than when the planets are on the back-side of their orbit? I'm also assuming we haven't observed a full orbit yet.

I imagine it has to do with the curvature of the observed orbits. The concave part of the curve would point toward the star and the convex part would point away.

Figuring out which side is "concave" vs "convex" isn't as easy as you might think...
Draw an oval; figure out whether the top or bottom is the closer side of the "circle" you just drew at an oblique angle. Could be either one, unless you're prepared to measure red-shift!

This becomes surprisingly obvious when you, for example, shift point of view below the object and suddenly the orbital lines go the reverse of where you expect. Any Elite Dangerous pilot here can confirm it's a hell of an optical illusion.

Now, if you could measure red-shift of the right- and left-most sides of the orbit, you should be able to figure out whether the "top" of the path is nearer or further from your point of view.


ETA:
Later point on light scattering was excellent; didn't think of that (or read the paper, obviously!).

To the other later question, about whether this could be figured out purely geometrically... I don't think so. You need some way of inferring "depth" to figure out which side of the orbit is further from you. At our distance, purely tracing the orbital path (what I *think* you mean) probably isn't sufficient.

 

[jmai86]

This is really awesome stuff.
On a related note, would it be possible to directly image a planet in true visible light? What tech would we need, or how big does a telescope need to be?

EDIT for clarity: Photograph an extra solar planet in detail, enough to discern surface features

 

[Chuckstar]

jmai86[/url]":1ylqa3sx]This is really awesome stuff.
On a related note, would it be possible to directly image a planet in true visible light? What tech would we need, or how big does a telescope need to be?

Hubble already did:

https://en.wikipedia.org/wiki/Fomalhaut_b

EDIT: Changed the link to be the wikipedia page. Had originally linked to a pdf.

 

[jmai86]

Chuckstar[/url]":1k24oyu8]

jmai86[/url]":1k24oyu8]This is really awesome stuff.
On a related note, would it be possible to directly image a planet in true visible light? What tech would we need, or how big does a telescope need to be?

Hubble already did:

https://en.wikipedia.org/wiki/Fomalhaut_b

EDIT: Changed the link to be the wikipedia page. Had originally linked to a pdf.

Ah right, I should clarify my original question: Photograph an extra solar planet in detail, enough to discern surface features. I'm thinking along the lines of the detail of some of the first photographs of Pluto by New Horizons.

 

[wyrmhole]

jmai86[/url]":3j6rr4p8]

Chuckstar[/url]":3j6rr4p8]

jmai86[/url]":3j6rr4p8]This is really awesome stuff.
On a related note, would it be possible to directly image a planet in true visible light? What tech would we need, or how big does a telescope need to be?

Hubble already did:

https://en.wikipedia.org/wiki/Fomalhaut_b

EDIT: Changed the link to be the wikipedia page. Had originally linked to a pdf.

Ah right, I should clarify my original question: Photograph an extra solar planet in detail, enough to discern surface features. I'm thinking along the lines of the detail of some of the first photographs of Pluto by New Horizons.

Oh no. Not even close, and not for a very long time. It's going to be a very long time until images of exoplanets match the best pre-New Horizons image of Pluto.

The only star (other than Sol of course) that we've ever been able to resolve as a disk instead of a point, is Betelgeuse, and it looked like this. And that star is pretty close (~650 ly) and huge -- it's about as wide as the orbit of Jupiter.

This star system in question is around 400 ly away. If I had a bit more time I could try my hand at the math to figure out what kind of improvement in angular resolution we'd need to get from that image of Betelgeuse, to having Jupiter occupy two pixels in an image... but given the distances aren't wildly different, you could get a basic idea by comparing the diameter of Jupiter to the diameter of Jupiter's orbit. Biiiig.

 

[funkioto]

When the article says 'images' but there's no actual images

 

[Z1ggy]

funkioto[/url]":bsn4xhuh]When the article says 'images' but there's no actual images

ok im not the only one.

 

[Joriarty]

jmai86[/url]":269a4ice]

Chuckstar[/url]":269a4ice]

jmai86[/url]":269a4ice]This is really awesome stuff.
On a related note, would it be possible to directly image a planet in true visible light? What tech would we need, or how big does a telescope need to be?

Hubble already did:

https://en.wikipedia.org/wiki/Fomalhaut_b

EDIT: Changed the link to be the wikipedia page. Had originally linked to a pdf.

Ah right, I should clarify my original question: Photograph an extra solar planet in detail, enough to discern surface features. I'm thinking along the lines of the detail of some of the first photographs of Pluto by New Horizons.

Gliese 674b seems like a good bet... only 15 light years away. We don't know the planet's radius, but 10% the radius of Jupiter might be a good guess. This makes the angular diameter of the planet 1e-10 radians.

The Rayleigh criterion says that if we were to resolve the planet with 50 resolution elements in each direction (i.e. a 50*50 pixel image, though it would be rather blurry), at ~500nm (that's green), we need a 313km diameter mirror.

Three hundred and thirteen kilometers.

One hell of an interferometer could do the trick, yeah. But I don't even want to think about how you'd build such a thing.

 

[spartanical]

Regarding these, "artist's conceptions" images, how reliable are these? We obviously have no actual or usable photos or they would be posted.

So who is this artist and what makes him/her qualified to represent anything astrophysical in the universe? Are they some creative scientist; directly guided by a scientist; or given liberty to embellish a conceptual understanding? Are we looking at imagination in graphical form or are these images meant to be a precise representation of the reality of these systems as they would be seen with direct observation?

 

[Chuckstar]

Joriarty[/url]":m7a9w5ah]

jmai86[/url]":m7a9w5ah]

Chuckstar[/url]":m7a9w5ah]

jmai86[/url]":m7a9w5ah]This is really awesome stuff.
On a related note, would it be possible to directly image a planet in true visible light? What tech would we need, or how big does a telescope need to be?

Hubble already did:

https://en.wikipedia.org/wiki/Fomalhaut_b

EDIT: Changed the link to be the wikipedia page. Had originally linked to a pdf.

Ah right, I should clarify my original question: Photograph an extra solar planet in detail, enough to discern surface features. I'm thinking along the lines of the detail of some of the first photographs of Pluto by New Horizons.

Gliese 674b seems like a good bet... only 15 light years away. We don't know the planet's radius, but 10% the radius of Jupiter might be a good guess. This makes the angular diameter of the planet 1e-10 radians.

The Rayleigh criterion says that if we were to resolve the planet with 50 resolution elements in each direction (i.e. a 50*50 pixel image, though it would be rather blurry), at ~500nm (that's green), we need a 313km diameter mirror.

Three hundred and thirteen kilometers.

One hell of an interferometer could do the trick, yeah. But I don't even want to think about how you'd build such a thing.

Someone has already figured that out. At least theoretically. They admit the cost would be prohibitive:

https://lise.oca.eu/spip.php?rubrique78


EDIT: And thanks for calculating the size requirement. I'm actually shocked by it. Not because it's so big, but because my wild guess when first reading wyrmhole's post was way bigger than that. I was thinking like "orbit of Neptune".

 

[Chuckstar]

spartanical[/url]":3tbul0wf]Regarding these, "artist's conceptions" images, how reliable are these? We obviously have no actual or usable photos or they would be posted.

So who is this artist and what makes him/her qualified to represent anything astrophysical in the universe? Are they some creative scientist; directly guided by a scientist; or given liberty to embellish a conceptual understanding? Are we looking at imagination in graphical form or are these images meant to be a precise representation of the reality of these systems as they would be seen with direct observation?

These images are not meant to be particularly precise. But they are, generally, done with the input of the astrophysicists. For example, in the image in the article they did not include the planetary rings just to be artistic. These are planets during formation, so would have a disk of dust spiraling in.

 

[jmai86]

Joriarty[/url]":vijgf8ou]

jmai86[/url]":vijgf8ou]

Chuckstar[/url]":vijgf8ou]

jmai86[/url]":vijgf8ou]This is really awesome stuff.
On a related note, would it be possible to directly image a planet in true visible light? What tech would we need, or how big does a telescope need to be?

Hubble already did:

https://en.wikipedia.org/wiki/Fomalhaut_b

EDIT: Changed the link to be the wikipedia page. Had originally linked to a pdf.

Ah right, I should clarify my original question: Photograph an extra solar planet in detail, enough to discern surface features. I'm thinking along the lines of the detail of some of the first photographs of Pluto by New Horizons.

Gliese 674b seems like a good bet... only 15 light years away. We don't know the planet's radius, but 10% the radius of Jupiter might be a good guess. This makes the angular diameter of the planet 1e-10 radians.

The Rayleigh criterion says that if we were to resolve the planet with 50 resolution elements in each direction (i.e. a 50*50 pixel image, though it would be rather blurry), at ~500nm (that's green), we need a 313km diameter mirror.

Three hundred and thirteen kilometers.

One hell of an interferometer could do the trick, yeah. But I don't even want to think about how you'd build such a thing.

Seriously though, thanks for the insight. That is a really big mirror. Sounds like we'll need to come up with new tech either to directly image, or new tech to get closer to these bodies to photograph with current imaging tech.

 

[lewax00]

funkioto[/url]":cgfizhrx]When the article says 'images' but there's no actual images

ok im not the only one.

I was very disappointed as well...I wanted to see said images...then again, I assume they're way to low resolution so actually discern anything meaningful out of them unless you already know what you're looking for.

 

[bhspencer]

The href for the DOI should be

http://dx.doi.org/10.1038/nature15761

Not

http://dx.doi.org/doi:%2010.1038/nature15761

 

[JohnDeL]

A study released today describes a new technique that has allowed scientists to observe a planet forming for the first time.[/url]

It may be the first time that they have used this technique, but it is far from the first time that we've seen a planet forming. Back in 1984 (i.e., the Stone Age), a protoplanetary disc was imaged around Vega using IRAS.

As for imaging planets around other stars, NASA has had that in the planning stages for about thirty years now. They wanted to start with a binocular telescope in a heliocentric orbit then move to twinned scopes at L1 and L3, and then move to scopes orbiting at Jupiter's L1 and L3 points. They would have been able to image planets out to about 50 light years with the big set-up. Unfortunately, budget cuts and NSA paranoia kept them from ever launching the program.

 

[wyrmhole]

JohnDeL[/url]":1jkhb21n]

A study released today describes a new technique that has allowed scientists to observe a planet forming for the first time.[/url]

It may be the first time that they have used this technique, but it is far from the first time that we've seen a planet forming. Back in 1984 (i.e., the Stone Age), a protoplanetary disc was imaged around Vega using IRAS.

Seeing a protoplanetary disk is not the same as seeing a still-forming planet.

But it's still a cool observation.

Edit: I'm pretty sure we've also observed planets that are in a protoplanetary disk, but did not have observations that indicated those planets were still growing, as we do here. Well here's an ALMA image showing clearly that something has swept up material in various portions of a disk.

 

[The Opinionaire]

So, where are these images?

 

[wyrmhole]

The Opinionaire[/url]":rjdr8hyc]So, where are these images?

You can see a tiny version of them in the corrected DOI link bhspencer posted:

http://dx.doi.org/10.1038/nature15761

They are... not particularly impressive visually.

 

[StratThinker]

Fortunately, a forming planetary system has its own observational advantages: the planets are putting out a lot of energy in the form of light. They're especially bright in hydrogen-alpha emissions, a particular shade of red that is very often useful in astronomy because it’s strongly emitted where hydrogen gas is ionized. It’s thought that the region around a still-forming planet can become hot enough to cause the hydrogen atoms to emit hydrogen-alpha light.

According to the theory, if a planet is forming, then it will emit hydrogen-alpha light. However is the inverse (which is equivalent to the converse) true? i.e. if a planet is not forming, then it will not emit hydrogen-alpha light? In other words is it possible for a non-forming planet to emit hydrogen-alpha light?

Lets analyse this more closely, according to the theory: If a planet is forming, then it will be rather hot. If a planet is rather hot, then hydrogen gas will be ionized. If hydrogen gas is ionized, then emit hydrogen-alpha light. Are the inverses of all these statements true? I.e. is it possible for non-forming planet to be rather hot (maybe due a collision)? Is it possible for a non-hot planet to have ionized hydrogen gas (maybe a strong enough magic field together with ice moon with active water geysers is sufficient to provide enough ionized hydrogen gas)? Is it possible for anything other ionized hydrogen gas to emit hydrogen-alpha light? The answer to the last question is probably a strong no.

Now a better question is not whether it is possible, but whether these are probable, and what relative probabilities between this hydrogen-alpha light is coming from a forming planet or whether it is coming from another source.

 

[JustUsul]

JohnDeL[/url]":3th3sqtt]

A study released today describes a new technique that has allowed scientists to observe a planet forming for the first time.

It may be the first time that they have used this technique, but it is far from the first time that we've seen a planet forming. Back in 1984 (i.e., the Stone Age), a protoplanetary disc was imaged around Vega using IRAS.

Actually, Vega has what is known as a debris disk, which is the older remnants of a so-called protoplanetary disk. But it looks like your link points to an article on Beta Pictoris (which, as it happens, does have a planet discovered 24 year later in 2008). These observations are the first time we have observed H-alpha emission at the location of a purported planetary companion, indicating that the proto-planet is accreting material from the disk and in the process of forming.

As for imaging planets around other stars, NASA has had that in the planning stages for about thirty years now. They wanted to start with a binocular telescope in a heliocentric orbit then move to twinned scopes at L1 and L3, and then move to scopes orbiting at Jupiter's L1 and L3 points. They would have been able to image planets out to about 50 light years with the big set-up. Unfortunately, budget cuts and NSA paranoia kept them from ever launching the program.

While SIM (which you link) would have been great for finding planets, it wasn't actually used for direct imaging. Instead, it would have been extremely sensitive to any astrometric movement of a star. Given that planets gravitationally pull the host star as they orbit, SIM would have been able to determine the masses and orbital positions of some planets around certain stars based on the stellar motion. Gaia is the European version that launched in 2013 and is currently surveying the sky.

 

[JustUsul]

wyrmhole[/url]":2zmsqaed]

The Opinionaire[/url]":2zmsqaed]So, where are these images?

You can see a tiny version of them in the corrected DOI link bhspencer posted:

http://dx.doi.org/10.1038/nature15761

They are... not particularly impressive visually.

Well, as someone that does stuff like this for a living, I find it very impressive (but I may be slightly biased in that assessment).

But yes, I can see how they might be disappointing given all the jaw-dropping images of Pluto coming down off of New Horizons these days.

 

[wyrmhole]

Jarron[/url]":3344euo7]

wyrmhole[/url]":3344euo7]

The Opinionaire[/url]":3344euo7]So, where are these images?

You can see a tiny version of them in the corrected DOI link bhspencer posted:

http://dx.doi.org/10.1038/nature15761

They are... not particularly impressive visually.

Well, as someone that does stuff like this for a living, I find it very impressive (but I may be slightly biased in that assessment).

But yes, I can see how they might be disappointing given all the jaw-dropping images of Pluto coming down off of New Horizons these days.

Only in terms of visual aesthetics. It's no Pillars of Creation, is all I'm saying. In terms of what it represents, both the accomplishment of taking it and what it means scientifically, it's awesome.

 

[Joriarty]

SirBedwyr[/url]":1yw32x80]

Joriarty[/url]":1yw32x80]

SirBedwyr[/url]":1yw32x80]I'd like to ask a super dumb question. (Sorry in advance.)


If you're making an observation that gathers about 20-30 degrees of rotation about the center, how do you gather the orientation of the system's disc? That is, the artist's rendition has us peaking "underneath" with the planets forward and overhead of the star. How do we determine that this is the correct orientation and that the planets are not overhead and behind the star as would be the case in an optical illusion? Is the star's light in that upper right region different than when the planets are on the back-side of their orbit? I'm also assuming we haven't observed a full orbit yet.

Good question! Really good question. It's not always easy to tell.

With an inclined disk, you're gonna see asymmetries in the brightness of each side. But is the asymmetry due to forward-scattering of starlight from optically thin material at the outer rim of the near side of the disk, or from backscattered light from optically thick material on the other side?

Recent results reckon that we're looking at the forward-scattering case, where the northwestern edge is closer to us (section 6.1 of the paper is especially relevant). Basically, we have to model both scenarios and see which computer model best fits the observations. It's not immediately obvious.

Ok, so it's not totally obvious or trivial then. Could you take a stab at it based purely on geometry (ignoring scattering) given a full 360 degree orbit's worth of information?

Nope – geometrically, there's no way distinguish whether the planets are moving towards us, or away from us, at any particular point in the orbit.

You can figure it out with spatially-resolved spectroscopy, because the rotation of the disk gives you doppler shifts. The ALMA telescope is very good this – it's by far the best instrument we have to spatially resolve protoplanetary disks.

 

[JustUsul]

StratThinker[/url]":4s3z92hq]

Fortunately, a forming planetary system has its own observational advantages: the planets are putting out a lot of energy in the form of light. They're especially bright in hydrogen-alpha emissions, a particular shade of red that is very often useful in astronomy because it’s strongly emitted where hydrogen gas is ionized. It’s thought that the region around a still-forming planet can become hot enough to cause the hydrogen atoms to emit hydrogen-alpha light.

According to the theory, if a planet is forming, then it will emit hydrogen-alpha light. However is the inverse (which is equivalent to the converse) true? i.e. if a planet is not forming, then it will not emit hydrogen-alpha light? In other words is it possible for a non-forming planet to emit hydrogen-alpha light?

Lets analyse this more closely, according to the theory: If a planet is forming, then it will be rather hot. If a planet is rather hot, then hydrogen gas will be ionized. If hydrogen gas is ionized, then emit hydrogen-alpha light. Are the inverses of all these statements true? I.e. is it possible for non-forming planet to be rather hot (maybe due a collision)? Is it possible for a non-hot planet to have ionized hydrogen gas (maybe a strong enough magic field together with ice moon with active water geysers is sufficient to provide enough ionized hydrogen gas)? Is it possible for anything other ionized hydrogen gas to emit hydrogen-alpha light? The answer to the last question is probably a strong no.

Now a better question is not whether it is possible, but whether these are probable, and what relative probabilities between this hydrogen-alpha light is coming from a forming planet or whether it is coming from another source.

I imagine that H-Alpha would not always be detectable in a forming proto-planet. However, observing H-Alpha (that is clearly not associated with the star) is probably a dead giveaway of the planet formation process.

In order to get the H-Alpha emission, the gas would need to be pretty hot (~10,000K). We commonly see H-Alpha in young stars with circumstellar disks, where this emission is associated with accretion shocks as material from the disk is magnetospherically funneled onto the surface of the star. In addition, disk winds also contribute to H-Alpha emission, but the majority of the emission is from the stellar accretion. Basically, you need a continuous highly-energetic event in order to explain this observed strong emission in young stars. The shape of the H-Alpha spectral line profile tells us a lot about the geometry of the emission region, which is why we can infer with great confidence that the majority of this emission seen in young star comes from relatively small spots of accretion at the stellar surface.

For the particular case of the LkCa 15 exoplanet companion(s), the use of high-resolution adaptive optics allows us to spatially locate exactly where the H-Alpha emission is coming from. A good chunk of emission appears to be concentrated on a point-like source seen in the infrared. The bright source in the IR suggests a giant planet or very warm proto-planet, while the H-Alpha likely indicates accretion; therefore, proto-planet caught in formation. So far, it's proven difficult to come up with other scenarios where both these things hold true.

People have raised the possibility that this is simply scattered light off of a clump of material that originated from the star, but that poses problems such as the relative brightnesses at the various wavelengths don't correspond to what you would expect for scattering. Also, I'm not sure if we would expect the clump to stay intact over the full observed orbital motion.

 

[JustUsul]

Joriarty[/url]":10rwfkyk]
Nope – geometrically, there's no way distinguish whether the planets are moving towards us, or away from us, at any particular point in the orbit.

Unless you're dealing with something like Beta Pic b, which I believe will be going behind the parent star pretty soon.