PDA

View Full Version : Alien Atmosphere Detected



Chip
2001-Nov-28, 05:51 AM
I like to post topics to discuss, but once in a while something comes up that's pretty cool as news, and I have to just offer a link to it:

For the first time, Hubble scientists detect an alien planet's atmosphere! Interesting article:
http://science.nasa.gov/headlines/y2001/ast27nov_1.htm?friend

Sodium in a hot Jupiter - Wow. And the star (not the planet) can be seen from Earth with a small telescope. /phpBB/images/smiles/icon_smile.gif

Argos
2001-Nov-28, 09:45 AM
I'm not sure if we should call it an "atmospheric" planet like Earth, Venus and Mars. Judging by what I read so far it should be deemed only as another giant gaseous planet. The only difference is that this time we have a direct glimpse of one of the chemical components of the huge gas giant, instead of solely infering its mass and orbit.

A thing that intrigues me is that most of the planets discovered so far orbit tightly the stars (this one 4 million miles). Why do we find this pattern in every star system we discover? When will we discover a planet orbiting at the "right" place?

Ok. You will say that is precisely because of it that we get to discover those planets, with the means available. But to me it is getting harder and harder to explain statistically. So many Jupiters so close to the stars? One or two is an acceptable oddity. But several ones? How do they keep their "atmosphere"? Are they losing it? Strong magnectic fields? Why don't we discover four Jupiters at the double distance, so that they can be observed by the means we have? Why always a lonely giant planet some million miles from the stars?

Questions, questions, questions...

<font size=-1>[ This Message was edited by: Argos on 2001-11-28 07:55 ]</font>

NottyImp
2001-Nov-28, 12:51 PM
"Ok. You will say that is precisely because of it that we get to discover those planets, with the means available. But to me it is getting harder and harder to explain statistically. So many Jupiters so close to the stars? One or two is an acceptable oddity. But several ones?"

I think you're right. We discover Jupiter-sized planets close to stars because these are the easiest to detect with our current methods. It's kind of self-selecting. It doesn't mean that these planets are at all common, and indeed they may be very rare. It's really a similar situation to the stars we see in the night sky. Most of these are in the top 10% of stellar masses, and that's why we can see them. But they're not typical of the broad mass of stars out there, many of which are tiny and dim in comparison.


<font size=-1>[ This Message was edited by: NottyImp on 2001-11-28 07:52 ]</font>

ToSeek
2001-Nov-28, 01:42 PM
On 2001-11-28 04:45, Argos wrote:

A thing that intrigues me is that most of the planets discovered so far orbit tightly the stars (this one 4 million miles). Why do we find this pattern in every star system we discover? When will we discover a planet orbiting at the "right" place?



The primary current method of detecting planets around other stars is to look for Doppler shifts in the light from the star. The closer the planet is to the star, the greater the Doppler shift and the quicker the full cycle. So there's definitely a selection effect toward close-in planets.

Torsten
2001-Nov-29, 02:13 AM
I was just reading the December issue of Sky and Telescope and in a short article they discuss this very phenomenon:

..."However, there's a sampling bias at work here: the tighter a planet's orbit, the faster the reflexive wobble it induces in the motion of its host star -- and the quicker the wobble can be detected. But now astronomers have been at the planet-hunting game long enough to find wobbles that take several years to complete, and to no one's big surprise, such long-period worlds are beginning to show up."

The article continued with a discussion of a second planet discovered orbiting 47 Ursae Majoris. See National Science Foundation News (http://www.nsf.gov/od/lpa/news/press/01/pr0164.htm)

Chip
2001-Nov-29, 06:41 AM
On 2001-11-28 21:13, Torsten wrote:
I was just reading the December issue of Sky and Telescope and in a short article they discuss this very phenomenon:

..."However, there's a sampling bias at work here: the tighter a planet's orbit, the faster the reflexive wobble it induces in the motion of its host star -- and the quicker the wobble can be detected. But now astronomers have been at the planet-hunting game long enough to find wobbles that take several years to complete, and to no one's big surprise, such long-period worlds are beginning to show up."


This leads to another question, (which has been asked before I'm sure.)

Given the current techniques for detecting big "Jupiter-like" extra solar planets in close orbits -- Is the discovery and/or detection of smaller extra solar terrestrial planets "merely" a technological problem?

In other words, given the current development of space telescope designs and other detection technologies, will we be able to detect or even actually see "Earthlike" planets in the future?


<font size=-1>[ This Message was edited by: Chip on 2001-11-29 01:43 ]</font>

ToSeek
2001-Nov-29, 01:56 PM
On 2001-11-29 01:41, Chip wrote:
given the current development of space telescope designs and other detection technologies, will we be able to detect or even actually see "Earthlike" planets in the future?


Yes (http://tpf.jpl.nasa.gov/)

David Simmons
2001-Nov-29, 04:20 PM
On 2001-11-29 01:41, Chip wrote:
given the current development of space telescope designs and other detection technologies, will we be able to detect or even actually see "Earthlike" planets in the future?




On 2001-11-29 08:56, ToSeek wrote:

Yes (http://tpf.jpl.nasa.gov/)


The answer might be closer to "maybe."

Just my take on it.

Planets only relect a small part of the light that falls on them. This makes the number of photons from a planet falling on an earth (or solar system) based detector exceedingly small. If it falls below the random noise of the background and the internal noise of the detection system then some sort of integration needs to be used to improve the signal to noise ratio.

For example, we assume that photons from a planet are falling on a pixel detector at a regular rate but that noise photons are intermittent. So if we sum detector photons for an extended period the pixel with the planet's photons will finally emerge from the background noise. The improvement is the sqrt of the number of samples.

But this assumes that our mechanical system is so steady that the planetary photons always fall on the same one or two pixel detectors, so the mechanical noise (vibrations etc.) also enters into the picture when integrating over long time periods.

Never say "never," but this looks to me like it's a long way down the road.

ToSeek
2001-Nov-29, 05:42 PM
On 2001-11-29 11:20, David Simmons wrote:

Never say "never," but this looks to me like it's a long way down the road.


The real challenge, as I understand it, is to do interferometry in space. (Mechanical pointing is less of an issue because you can coordinate on the star.) This requires being able to set up multiple spacecraft with distances between them controllable to nanometers and measurable to picometers. (For interferometry to work, the distance between the telescopes must be known to significantly less than the wavelength being used.)

In short, the technology isn't there yet, but it is being worked on.

frenchy
2001-Nov-29, 10:50 PM
Other space missions are on the way.
They are mainly based n photometric detection of transits, it's fair to say that detecting an Earth-like planet would need some good fortune.

The following sites are pretty technical
http://www.astrsp-mrs.fr/projets/corot/planets.html (the french COROT)
http://astro4.ast.villanova.edu/MONS/planets.htm (the danish MONS)
http://www.kepler.arc.nasa.gov/ (the american KEPLER)

David Simmons
2001-Nov-29, 11:07 PM
On 2001-11-29 18:02, David Simmons wrote:
[quote]
On 2001-11-29 12:42, ToSeek wrote:

The real challenge, as I understand it, is to do interferometry in space. (Mechanical pointing is less of an issue because you can coordinate on the star.) This requires being able to set up multiple spacecraft with distances between them controllable to nanometers and measurable to picometers. (For interferometry to work, the distance between the telescopes must be known to significantly less than the wavelength being used.)

In short, the technology isn't there yet, but it is being worked on.


It's been over 20 years since I retired so I haven't kept up very well. But the original question was about imaging, in addition to just detecting, an Earth-sized planet.

Hmmm. I don't see how you can image using passive interferometry. A hologram produces and image using interference, but the light reflected from the object interferes with a sample from the same, illuminating source to produce the interference pattern. That isn't the same as the passive case.

I guess I do agree that detecting Earth-sized planets is probably within our capabilities, and not too long from now either.

Donnie B.
2001-Nov-30, 02:11 PM
The interferometry in question here is not the sort where you pass the light through a slit and look at the fringes. It's the sort where you combine the light received by several instruments and generate an image that's equivalent to a single instrument whose aperture is as large as the distance between the instruments (though, of course, its light-gathering power is only the sum of the two instruments').

This is what the VLBI (Very Long Baseline Interferometer) does at RF wavelengths. There are other examples, and IIRC, there's an optical interferometer in the works using the two Keck telescopes (somebody correct me if I have this wrong).

To image a planet at interstellar distances, you'd need an interferometer that's perhaps 2AU wide. The idea is to launch two space telescopes into solar orbit and keep them in very accurate positions, combine their outputs, and resolve a very small object at very large distances.

David Simmons
2001-Nov-30, 02:50 PM
On 2001-11-30 09:11, Donnie B. wrote:
The interferometry in question here is not the sort where you pass the light through a slit and look at the fringes.


I understand that. But I'm having trouble seeing the physics of image formation by the process of interference.

Donnie B.
2001-Nov-30, 07:18 PM
OK, I think I misread your post. I can't answer your question in detail, but I know it involves combining the wavefronts. In some of the arrays, they do this with the RF signals by piping them to a central location using waveguides. However, I understand that it's also possible to record data at distant locations and do the interferometry with computer processing offline. A very accurate timebase signal must be recorded with the data.

Here's a fairly non-technical description:

http://www.drao.nrc.ca/science/vlbi/principles/principles.shtml

This site may be of interest too:

http://sgra.jpl.nasa.gov/mosaic_v0.0/svlbi_old.html

David Simmons
2001-Dec-01, 01:58 AM
On 2001-11-30 14:18, Donnie B. wrote:
OK, I think I misread your post. I can't answer your question in detail, but I know it involves combining the wavefronts. In some of the arrays, they do this with the RF signals by piping them to a central location using waveguides.


OK, if the term "image" is extended to include power distribution at various frequencies as seen by the interference between two widely separated receivers I get it.

I was thinking of the image as in a picture of an earth-like planet in the visible.

Manchurian Taikonaut
2004-Mar-08, 07:36 AM
yes, most of these planets are too big, large gravity and too hot to host life.

http://www.exn.ca/news/images/1999/11/29/19991129-extrasolar.jpg


Yet, sooner or later they will find better planets. Plus they have many new projects for better detection systems.[/img]

Maksutov
2004-Mar-08, 12:19 PM
On 2001-11-29 01:41, Chip wrote:
given the current development of space telescope designs and other detection technologies, will we be able to detect or even actually &lt;i>see&lt;/i> "Earthlike" planets in the future?


Yes (http://tpf.jpl.nasa.gov/)

Agreed. Despite David Simmons' response, it's a definite "yes". It's just a matter of resolution. All that is needed is a larger baseline, which is already being achieved in Chile, and will soon be surpassed by computer-linked space or Moon based telescopes.

eburacum45
2004-Mar-08, 02:35 PM
I think a space-based baseline thousands of kilometers long will be needed to resolve earth-like planets; but such an instrument should also be sensitive enough to see galaxies a few million years old, at the edge of the observable universe; and red dwarf stars thousands of parsecs away.

Just looking for planets would use a fraction of the potential of such an array.

Tinymidget
2004-Mar-09, 05:54 AM
Also, remember, there are many other stars out there. The reason we are discovering so many gaseous giants that are close to the sun is not because they are common, but because we can see them easily and are identified as planets. Take this for example: everybody can tell where Venus is in the night sky. It's the brightest one, it reflects alot of light and is an okay size. Now take a look at mars, in the other direction, a casual eye wouldn't pick it up without being told its location or having a general knowledge of our solar system. Apply this to bigger terms. When we scan the skys for planets, the most noticble ones will be 1) Things that reflect light and 2) are BIG. We probably miss alot of solar systems that look like ours. If you were on another solar system looking our our home system, you would probably see jupiter, but not earth. So you would say, "Oh...another solar system with a lone gas giant in the middle of nowhere." and move on.

Kaptain K
2004-Mar-09, 08:10 AM
We have yet to see any exoplanets directly. Most of the exoplanets we have detected have been by the Doppler shift of the parent star's spectrum. This method is heavily biased toward big planets in tight orbits (hot Jupiters). It only gives a lower limit to the mass of the planet (the mass if the inclination of the orbit is zero to our line of sight).

A couple of exoplanets have been discovered from the light curve of the parent star caused by transits of the planet across the face of the star. This method has the advantage of giving the mass directly, since we know the inclination (zero). This method also favors large planets, since the variation in the light curve is so small.

Direct imaging of planets has been compared to looking at a searchlight and trying to see a candle next to it.

Manchurian Taikonaut
2005-Sep-22, 10:06 PM
The planet around 70 Vir orbits the star in an eccentric, elongated orbit every 116 days and has a mass about nine times that of Jupiter. Using standard formulas that balance the sunlight absorbed and the heat radiated, Marcy and Butler calculated the temperature of the planet at about 85 degrees Celsius (185 degrees Fahrenheit), cool enough to permit water and complex organic molecules to exist. The star 70 Vir is nearly identical to the Sun, though several hundred degrees cooler and perhaps three billion years older. US scientists have found evidence that more than one planet is orbiting a star called 47 Ursae Majoris a mere 50 light years away from Earth.
The planet spotters said they are likely to find signs of even more planets as instruments become more sensitive and sky surveys become more comprehensive.
Now the scientists are pushing for funds to develop new telescopes that do nothing but look for planets orbiting nearby stars.

some future missions:

COROT will contribute to the search for habitable, Earth-like planets around other stars. COROT will do this by detecting planets as they pass in front of their parent stars, blocking some of the light.
http://sci.esa.int/science-e/www/object/index.cfm?fobjectid=31709

Terrestrial Planet Finder is a suite of two complementary space observatories: a visible-light coronagraph and a mid-infrared formation-flying interferometer. They will detect and characterize Earth-like planets around as many as 150 stars up to 45 light-years away.
http://planetquest.jpl.nasa.gov/TPF/tpf_facts.cfm

Darwin is a flotilla of four spacecraft that will search for Earth-like planets and analyse their atmospheres for the chemical signature of life. One spacecraft will be a central communications hub. The other three will function as 'light collectors', redirecting light beams to the hub spacecraft.
http://www.esa.int/esaSC/SEMZ0E1A6BD_index_0.html

Kepler... NASA's first mission capable of finding Earth-size and smaller planets around other stars.
http://kepler.nasa.gov/about/