Astronomers detect smallest extrasolar planets yet 16
Bob Kopp writes "A team of astronomers using Keck Observatory has discovered the smallest extrasolar planets yet: one with 80% of Saturn's mass, orbiting HD46375, and one with 70% of Saturn's mass, orbiting 79 Ceti. These are the first extrasolar planets discovered with masses less than that of Jupiter. Both are, however, quite close to their parent stars; the HD46375 planet has a period of 3 days, while the 79 Ceti planet has a period of 75 days. CNN has coverage. "
Close orbit gas planets (Score:1)
All may not be as it seems... (Score:1)
I wonder if we could detect anything but Jupiter from our ouwn sun's wobble? How big a planet does old Sol's wobble say is in orbit around it - 2 Jupiters or 1 jupiter and 8 to 9 others? How does an Oort cloud or Kuiper belt of material affect these calculations?
Until we have a spaced based interferometer array (which I beleive NASA is trying to get funding for)which can do DIRECT imaging of these planets, we will not know of any reall numbers and sizes.
BTW, NASA's space-based interferometer design is capable of imaging earth sized planets and doing spectral analysis on them.
Any professional asttronomers or planet finders out there care to comment?
Re:All may not be as it seems... (Score:1)
Build and interferometer!
Thanks agian..great answer...you should be moderated up.
Re:This may actually be a bad thing... (Score:1)
1. Saturn is a Jovian planet. Jovian refers to the composition of the planted, not its size or mass.
2. There have been on the order of 100 planets found. There are billions of stars. When we find billions of solar systems and show that none possess terrestrial planets, then we can conclude ours is not typical.
Re:Close orbit gas planets (Score:1)
The reports only tell the mass of the planets. It doesn't say anything about there composition, so it is not necessarily a "gas giant", just a planet with a large mass. The densities of palntes vaires quite a bit. Saturn has density of less than 1. It's plausable that there could be dense planets also.
provolt
Orbit of 3 days!? (Score:1)
It seems to me for a gas giant to orbit a sun in 3 days, they would have to be really close, right? My gut reaction is that an orbit that close would be unstable. Has anyone studied the dynamics of a situation like this?
-y
Re:Implications (Score:1)
Re:We don't have capacity...Computing in space... (Score:1)
How pathetic is that.
You'd think that might be something we want to find out about so that when our sun blows up in a few million years (if not sooner) we will have some idea of where we can go (let alone how to get there).
But I guess that's a little ways off.
BTW for those who care, is there any realizable time pay off to computing in 0 gravity (I suppose Moore's law keeps relativity (which tells us time goes faster where there is less gravity) in perspective.)
Atmosphere Retention (Score:1)
Oops... (Score:2)
But while I'm on the topic: the sun is something like 1050 Jovian masses, dwarfing Jupiter even more than Jupiter does Earth. This ratio is why detecting extrasolar planets smaller than Jupiter is so hard -- it's common to say that the planets orbit the sun, but actually a planet and its sun both orbit their common center of gravity. As it turns out, Jupiter orbits a bit more than 1100 times the sun's radius away from the sun's center, so the center of mass is just about at the sun's surface.
This means, to detect Jupiter at interstellar distances, we'd be looking for a Doppler shift based on a "wobble" about equal to the sun's radius (not quite 700 thousand kilometers) in half of Jupiter's orbital period (six years, since Jupiter's orbit takes 4433 days)... not much variation, over a very long time -- meaning the velocity is small, and therefore difficult to separate from measurement errors. The extrasolar planets we've been finding have mostly been larger than Jupiter, and have all orbited much closer to their stars -- barely a quarter of them have been as far out as Earth, while Jupiter is over five times that distance from Sol.
As I said, it's a difficult task.
---
Re:Orbit of 3 days!? (Score:3)
Yes.
Yes. See, for example:
"Orbital Evolution and Migration of Giant Planets: Modeling Extrasolar Planets", Trilling, D.E., Benz, W., Guillot, T., Lunine, J.I., Hubbard, W.B., Burrows, A., 1998, Astrophysical Journal, 500, 428. [harvard.edu]
[TMB]
Re:We don't have capacity...Computing in space... (Score:3)
Don't be so quick to judge. You can't just point the Hubble at a star and see if there are any planets going around it. First of all, since they are so faint and close to the disk of the star, no existing instrument can resolve planets optically.
Secondly, the major method used to find planets (doppler shift in stellar spectra to detect the star's wobble) works as well from the ground as from space, and there are more telescopes on the ground.
Thirdly, you can't find earth-like planets this way, so this won't help any when it comes time to look for other planets to move to. But the sun won't explode, either, so we're fine for quite some time. The gradual heating of the sun will, if anything, probably mean that, in 100 Myr, when the Earth is uninhabitably hot, Mars should be plenty warm enough to support liquid water (just add air).
Fourthly, to see earth-like planets, you need huge interferometers. Look at Deep Space 3 [nasa.gov] for an example of what NASA is trying to do in this direction. But interferometers are hard, so it'll be a while before the planet finder [nasa.gov] goes on line, or is even technically feasible, for that matter.
Not really. From the perspective of a zero-g environment, calculations would seem to take longer on the surface of the earth than in space, but by such a miniscule amount that it doesn't matter. For the money spent to develop and orbit the thing, you can probably double many times over the number of processor nodes in your computing system.
Implications (Score:3)
Yes, finding more planets reinforces the notion that planetary systems are relatively common. This theory didn't need a whole lot of shoring up, though ... it's patently obvious by now.
On the other hand, I don't at all see what this discovery has to say about the relative abundance of terrestrial-class planets. In fact, since the current models of planetary formation indicate that these bodies must have migrated in from the outer reaches of their solar systems (via either migration or interaction with other large bodies) we can actually surmise that, in all of the systems thus far examined, any inner, terrestrial planets that once existed have been ejected by interacting with these gas giants.
Remember, in our own solar system, the major reason there are (practically) no small sun-orbiting bodies between the asteroid belt and the Kupier belt is that the gas giants swept up most of the mass and ejected the rest in that zone.
Re:All may not be as it seems... (Score:3)
I'm not a professional in the field (I've used my physics degree to bootstrap myself into engineering consulting), but my field of interest was astrophysics, so I'll take a swing at it. Lotsa questions, though...
That information comes from the Doppler data: compare the present discovery [berkeley.edu] with a system which has multiple planets. [berkeley.edu] In the first case, you'll see a simple periodic variation in the Doppler shift -- a distorted sine curve, if you will, but one which has a single periodic structure which corresponds to the period of the planet. In the second case, there are three periods -- one for each of the three planets -- so they stack up on top of each other, to form a complex periodic structure. The second plot on that page shows the second and third planets' Doppler curve, with the very short-period inner planet removed; you see a long-period sinusoidal curve (the outer planet's), with a shorter-period curve (the second planet's) making about 5 "ripples" in the long-period curve (meaning that the second planet orbits about 5 times for each orbit of the outer one -- 241 days to 1309 days, in fact).
I won't go into the details -- there's plenty of that on the second link -- but it's just a matter of analysis of the data, fitting a model to it, and making a few wise choices if you find that the simplest models won't do. And BTW, it would take a lot more than a handful of terrestrial planets to equal the mass of a super-jovian -- although Jupiter is about 11 Earth diameters wide, it is much more massive (around 80 terrestrial masses, IIRC -- but don't quote me on that, I don't have a text or a link handy).
That may very well be the case, but we won't find that out with the present equipment. More reason to keep searching, and to get better instruments!
Right now, we'd might be able to detect Jupiter and possibly Saturn -- but both of those planets are in more-distant orbits than what we've found so far [sfsu.edu], so detection would be more difficult. The Doppler method would completely miss everything else in the Solar system. (Again, the mass of the rest of the system, after Jupiter and a bit from Saturn, is trivial...)
The Oort cloud and Kuiper belt don't have much effect at all -- orbits of stuff out there are quite long (hundreds of years and way up), compared to the few days to a few years for all the extrasolar planets detected so far. And as far as we know, the masses of bodies in both aren't very large -- in fact, they don't even detectably perturb the orbits of the planets in the Solar system, and we've been looking for that for a long time. Even if something quite large was out there, it would take hundreds to thousands of years of watching to detect it -- either here or in the extrasolar planetary systems.
Actually, we do know one planet's [berkeley.edu] mass and size, because it passed between us and the disk of its star; that allows us to remove the uncertainty in mass (from Doppler data, we only know a lower limit on the planet's mass) because we now know the plane of the orbit, and from the measured dimming of the star's light, we know the size of the planet.
But that's a rare case, and I won't argue that we don't need a space-based interferometer array.
Good questions, all of them -- and I hope I've answered them, at least partially.
---
This may actually be a bad thing... (Score:3)
I still intend to look up the characteristics of the stars involved... my guess it they are not sollike... but the fact that gas giants have been detected close enough to their primaries to have 3- and 75- day orbits means that the models that predict gas giants at larger distances and rock/ice bodies in close, with scattered debris interspersed, may be incorrect.
On the positive side, these are only two data points, and the most obvious sort (it's easier to find a planet close to a star, and large, because of reflection), so they may be statistical anomolies yet.
We don't have capacity to detect sol-like systems (Score:4)
There are two boundary conditions on what we can detect: the size of the planets and the distance of their orbit from the star. This is because we can detect planets by noting how much their presence causes their parent star to move. At present, we can detect Saturn-class planets with periods of a couple of years or less. With ground based technology, we may eventually be able to detect Neptune-class planets and planets in Jupiter-like orbits. To detect terrestrial planets, we'll need orbital or lunar telescopes designed for planet searching.
Bob Kopp