I spy in the night sky, don't I...
May 12, 2004 1:07 PM Subscribe
Best evidence yet (maybe) for keeping the Hubble around: Hubble snaps first photo (maybe) of extrasolar planet (maybe).
Wait a minute. We've never had photographic evidence of a planet circling another star before? I had no idea that ... well, that we hadn't, or that such a photo would be novel.
Huh.
posted by blueshammer at 1:53 PM on May 12, 2004
Huh.
posted by blueshammer at 1:53 PM on May 12, 2004
I'll say... I've been waiting for something like this for some time, and now that it's here it looks like pixelated bird poo. Nasa says they're thinking of sending up a robotic mission to repair the Hubble due to the public outcry against the d/c order in the wake of Columbia. Score that as a major victory for space exploration, folks.
posted by moonbird at 1:55 PM on May 12, 2004
posted by moonbird at 1:55 PM on May 12, 2004
that's not very convincing. as far as i can tell, all they've done is take pictures of a bunch of faint stars and then subtracted what they think the light from the star would be (perhaps by averaging all their observations together to get an "average star"). what they're left with, in a few cases, is a bright dot. that's it.
incidentally, the strange "noise" pattern where the star was is normal - it's what you always see when you try to subtract away a bright object, and is simply the noise in the original observation and model (the noise where the star was is noisier than the noise elsewhere because the star was brighter - it's just how "photon statistics" works). the idea is that by doing the subtraction you can see what's left in the data, in this case the circled dot (without the subtraction, they would just have seen a bright blob that was the star - that little dot to the side would have been swallowed up in the stars emission and not at all noticeable).
the dot might be noise (you get bright dots in some observations if a cosmic ray strikes the detector - although i don't know if that particular detector has that problem), but is probably real because i doubt they'd have announced anything without finding the same dot in a couple of images (you tend to make more than one observation of the same thing just to rule out randome events like this).
however, there's absolutely nothing to suggest that it's a planet. it could just as easily be another, fainter, more distant star that just happens to be aligned behind the star they observed.
you could perhaps make some arguments about how common faint stars are and work out whether it's extremely unlikely to have a faint star so close, but they don't appear to have done so (and such arguments are famously unreliable - they were used to support cosmological theories that are generally regarded as "crackpot" in the past).
to verify that the "planet" is real, they need to do more observations. they probably can't do spectroscopy, because the star is too close (and i'm not sure that would help anyway, as you'd just see the stellar spectrum reflected), so they'll need to wait for 6 months (say) and take another picture to see if it's shifted as you might expect for a planet in orbit.
maybe they announced this so that they would get time to do those observations (getting time on telescopes is very competitive). more likely, this was rushed out as pro-hubble propoganda (with everyone fighting over grants and support, there's a lot of pressure in astronomy to get "cool" science out into the popular press; that's just normal astronomy, so the pressure must really be on people working with hubble data).
and for those hoping this means alien life, the news is even worse - that star was previously a giant and shedded a planetary nebula (see here), so anything that was alive on the planet is almost certainly toast. in fact, i'm wondering whether planets are expected to survive around white dwarves - i don't know enough stellar astronomy to say (i'll ask my partner when she comes in).
oops. got carried away. if this all looks very authoratitive, remember i've not worked as an astronomer for years and was never a very good one. but even so, this looks pretty dodgy...
posted by andrew cooke at 2:01 PM on May 12, 2004
incidentally, the strange "noise" pattern where the star was is normal - it's what you always see when you try to subtract away a bright object, and is simply the noise in the original observation and model (the noise where the star was is noisier than the noise elsewhere because the star was brighter - it's just how "photon statistics" works). the idea is that by doing the subtraction you can see what's left in the data, in this case the circled dot (without the subtraction, they would just have seen a bright blob that was the star - that little dot to the side would have been swallowed up in the stars emission and not at all noticeable).
the dot might be noise (you get bright dots in some observations if a cosmic ray strikes the detector - although i don't know if that particular detector has that problem), but is probably real because i doubt they'd have announced anything without finding the same dot in a couple of images (you tend to make more than one observation of the same thing just to rule out randome events like this).
however, there's absolutely nothing to suggest that it's a planet. it could just as easily be another, fainter, more distant star that just happens to be aligned behind the star they observed.
you could perhaps make some arguments about how common faint stars are and work out whether it's extremely unlikely to have a faint star so close, but they don't appear to have done so (and such arguments are famously unreliable - they were used to support cosmological theories that are generally regarded as "crackpot" in the past).
to verify that the "planet" is real, they need to do more observations. they probably can't do spectroscopy, because the star is too close (and i'm not sure that would help anyway, as you'd just see the stellar spectrum reflected), so they'll need to wait for 6 months (say) and take another picture to see if it's shifted as you might expect for a planet in orbit.
maybe they announced this so that they would get time to do those observations (getting time on telescopes is very competitive). more likely, this was rushed out as pro-hubble propoganda (with everyone fighting over grants and support, there's a lot of pressure in astronomy to get "cool" science out into the popular press; that's just normal astronomy, so the pressure must really be on people working with hubble data).
and for those hoping this means alien life, the news is even worse - that star was previously a giant and shedded a planetary nebula (see here), so anything that was alive on the planet is almost certainly toast. in fact, i'm wondering whether planets are expected to survive around white dwarves - i don't know enough stellar astronomy to say (i'll ask my partner when she comes in).
oops. got carried away. if this all looks very authoratitive, remember i've not worked as an astronomer for years and was never a very good one. but even so, this looks pretty dodgy...
posted by andrew cooke at 2:01 PM on May 12, 2004
duh. should have spell checked. authoritative and god knows what else.
posted by andrew cooke at 2:02 PM on May 12, 2004
posted by andrew cooke at 2:02 PM on May 12, 2004
sorry, wasn't criticising it as a post. if they're right you're certainly going to hear about it again, and maybe there's more info than appeared in the article. perhaps the stars were already known to be likely candidates for having planets, for example.
posted by andrew cooke at 2:11 PM on May 12, 2004
posted by andrew cooke at 2:11 PM on May 12, 2004
Wow. First-ever picture of an extrasolar planet, and it turns out to be a square one.
Maybe it be one of these.
posted by Wet Spot at 3:23 PM on May 12, 2004
Maybe it be one of these.
posted by Wet Spot at 3:23 PM on May 12, 2004
One of the articles I've seen added the following tidbits (which might be off a bit, as I'm reciting from memory):
1. They, essentially, took several shots, and between each rotated the telescope. Oversimplified, the spiky diffraction patten from the star rotates with the telecope (as it is caused by the telescope's optics), but the position of the possible planet on the image will change as the telescope rotates. This information can be used to mostly erase the star, but keep the thingie, whatever it is.
2. In 6 months, followups are planned. If the possible planet is a background object, it won't display the same proper motion as the star. Proper motion is the movement of objects across the sky, in this case, due to the movement of the earth over six months. If the thingie is actually close to the star, they'll move together.
posted by doorsnake at 3:25 PM on May 12, 2004
1. They, essentially, took several shots, and between each rotated the telescope. Oversimplified, the spiky diffraction patten from the star rotates with the telecope (as it is caused by the telescope's optics), but the position of the possible planet on the image will change as the telescope rotates. This information can be used to mostly erase the star, but keep the thingie, whatever it is.
2. In 6 months, followups are planned. If the possible planet is a background object, it won't display the same proper motion as the star. Proper motion is the movement of objects across the sky, in this case, due to the movement of the earth over six months. If the thingie is actually close to the star, they'll move together.
posted by doorsnake at 3:25 PM on May 12, 2004
This is a nice discovery, but can someone explain exactly why, when the Hubble finally dies, we can't wait a few years for the Hubble's replacement and save some $$$ in the process. Why the fixation on repair?
The Universe isn't going anywhere.
posted by moonbiter at 3:41 PM on May 12, 2004
The Universe isn't going anywhere.
posted by moonbiter at 3:41 PM on May 12, 2004
pauli came home and says some planets are expected to survive around these kinds of stars; also apparently several different groups have been trying this which helps explain the early press release, but you could argue that makes the odds worse (presumably a lot of stars have been measured by other groups without finding anything, so odds are eventually a background star will be in the right place by accident).
(moonbiter - i'm not arguing for hubble. in fact in the past i've argued that astronomy generally isn't particularly useful, though it keeps us clothed and fed, thank-you very much)
posted by andrew cooke at 4:14 PM on May 12, 2004
(moonbiter - i'm not arguing for hubble. in fact in the past i've argued that astronomy generally isn't particularly useful, though it keeps us clothed and fed, thank-you very much)
posted by andrew cooke at 4:14 PM on May 12, 2004
Moonbiter, the Hubble telescope has significant advantages to astronomers over Earth-based telescopes. There is no atmosphere to distort the image in space, and without an atmosphere to absorb certain parts of the electromagnetic spectrum, more detail can be seen with Hubble. Taking Hubble away from astronomers is like telling the New York Times to switch to typewriters for a couple of years.... (well sorta). And anyway, the universe may not be going anywhere soon, but it IS constantly changing.
posted by banished at 4:40 PM on May 12, 2004
posted by banished at 4:40 PM on May 12, 2004
moonbiter, the replacement for Hubble (the James Webb Space Telescope) is actually a very different piece of kit. It's primary function is to examine the remnant radiation from the big bang ("first light") and as a result it is optimised to look at infra-red rather than optical frequencies.
It's also going to be put at the L2 Lagrangian point, well beyond reach of the shuttle. So, there will be no maintenance missions and when it fails it's gone.
The ideal would be to run both Hubble and the JWST at the same time, but beyond the problem of cost there is also the requirement of sending a repair mission to Hubble, which because of it's orbit would mean a shuttle being sent without any option of diverting to the ISS - unacceptable now of course.
More info here.
posted by thatwhichfalls at 11:55 PM on May 12, 2004
It's also going to be put at the L2 Lagrangian point, well beyond reach of the shuttle. So, there will be no maintenance missions and when it fails it's gone.
The ideal would be to run both Hubble and the JWST at the same time, but beyond the problem of cost there is also the requirement of sending a repair mission to Hubble, which because of it's orbit would mean a shuttle being sent without any option of diverting to the ISS - unacceptable now of course.
More info here.
posted by thatwhichfalls at 11:55 PM on May 12, 2004
« Older Modern Ruins | International ME/CFS/Fibromyalgia Awareness Day Newer »
This thread has been archived and is closed to new comments
posted by muppetboy at 1:51 PM on May 12, 2004