"All these worlds are (like) yours except . . . "

July 23, 2010 1:30 AM

More than 100 Earth-like planets found . . .

. . . well, no, not really.

Reports that stated so were erroneous. NASA's Kepler mission has confirmed 5 extrasolar planets so far, none of them Earth-like. Among the over 700 candidates a large percentage of them appear to be small, rocky worlds instead of large gaseous planets, like Jupiter. In June NASA released data from 43 days of the Kepler mission. The data can be found online. The June data release caused some division among astronomers because the Kepler team released information on only about 350 planet candidates of the more than 700 candidates that they had data on. Those candidates need ground based observations to figure out whether they're planets or not. Until then, current extrasolar planet count: 453, or 466, or something. (Pre-viously)

posted by IvoShandor (48 comments total)

What always concerns me about extra-solar planet discoveries, Earth-like or Jovian, is that their orbits always seem to weird compared to our own solar system. Everytime I read one of these reports, it's like "The orbit is much closer to the star than Mercury is to the Sun" or "The planet orbits the star once every 8 days". Doesn't sound like our solar system. Doesn't sound promising.
posted by Jimbob at 1:44 AM on July 23, 2010

Hopefully one day my grandchildren will colonize these planets and strip them like a stolen Lexus.
posted by Mayor Curley at 1:48 AM on July 23, 2010 [28 favorites]

I was surprised it was a link to the Daily Mail, until I saw the "Not really" part hidden behind the More Inside. LAME.
posted by JHarris at 1:48 AM on July 23, 2010 [3 favorites]

Jimbob: That's because radial velocity measurements to detect extrasolar planets are most sensitive to massive planets and to planets close to the star.
It's not truly representative of the true distribution of extrasolar planetary systems, and everyone is aware of what kind of sensitivity these searches have and why they keep popping up odd systems like that.
posted by edd at 2:00 AM on July 23, 2010 [7 favorites]

What always concerns me about extra-solar planet discoveries, Earth-like or Jovian, is that their orbits always seem to weird compared to our own solar system.

It's probably down to having pretty blunt tools for discovery, which only uncover planets large enough to perturb their stars' brightness enough for us to measure. We're looking for slight perturbations that are very, very far away.
posted by Blazecock Pileon at 2:20 AM on July 23, 2010

Plus, they probably used to be in normal orbits but they got put in wacky ones as a last-ditch attempt by the inhabitants to save their ecosystems, once filling the air with sulfur and detonating the moon(s) had ceased to be effective.
posted by No-sword at 2:33 AM on July 23, 2010 [4 favorites]

"It's probably down to having pretty blunt tools for discovery, which only uncover planets large enough to perturb their stars' brightness enough for us to measure."
Well it's not just the brightness that can be used (it's one method for sure though) - the star orbits a little bit around the planet just as the planet orbits around the star, and the motion of the star back and forth as it does so is measurable. This velocity change is greater when the planet is in a tight and short orbit, which is why we're sensitive to planets very close in as well as very big.
posted by edd at 2:45 AM on July 23, 2010

Wikipedia's List of extra solar planets is always helpful. You can sort by mass. One planet is about twice as massive as earth (0.003 times as heavy as jupiter) and one is about 1/50th. All the other ones are larger then that. And. The twice as massive one has an orbital period of 1.2 days

The fastest orbital period is 0.7888399 days and the longest is 1699 years.
posted by delmoi at 2:53 AM on July 23, 2010 [2 favorites]

I was surprised it was a link to the Daily Mail, until I saw the "Not really" part hidden behind the More Inside. LAME.

But who else will tell me what this foreign real estate boom will do to MY property value?!
posted by protorp at 2:54 AM on July 23, 2010

They keep finding those 100 new planets.
posted by Phanx at 2:56 AM on July 23, 2010 [1 favorite]

If I won the lottery, the first thing I'd spend my money on was a chance to sit down with the extra-solar planet wobble detector scientists and quiz them.

"We're looking for slight perturbations that are very, very far away."

Understatement? Apparently we're detecting moths passing in front of headlights a million miles down the road?

I don't doubt there are solar systems around most/many stars, but how are we observing anything conclusive from our vantage point?
posted by vectr at 2:56 AM on July 23, 2010

Daily Mail - "Hundreds of planets full of immigrants may be planning to seek asylum in Britain according to NASA"
posted by quarsan at 3:34 AM on July 23, 2010 [2 favorites]

I don't doubt there are solar systems around most/many stars, but how are we observing anything conclusive from our vantage point?

I think these people know what they're doing. What else do you think could be causing the periodic oscillations?
posted by delmoi at 3:36 AM on July 23, 2010

vectr: Well you have to remember these are headlights that don't move around and stay on for a very long time - quite unlike headlights really, with things moving past them on the timescale of days to years - quite unlike moths really, so you can have very sensitive instruments look at them with great precision.
posted by edd at 4:00 AM on July 23, 2010 [2 favorites]

They keep finding those 100 new planets.

Well, we now know of several hundred planets now, so there may have been multiple 100+ planet hauls.
posted by delmoi at 4:01 AM on July 23, 2010

Vectr, wikipedia is a pretty good starting point if you want to read up on your extrasolar planetary astronomy. It's pretty awesome stuff, I do recommend it. There are many different methods of detection currently in use, and yes we have directly observed quite a few of these guys already. But you're right in that most of these suspected exoplanets are not confirmed, and many could be non-planetary bodies in orbit around these stars, like brown dwarfs. But what is certain is there's something there, and if that something is a planet we can make a lot of pretty good guesses about what kind of planet it will be. Of course, we can also be very wrong.
posted by mek at 4:34 AM on July 23, 2010

We'd have a hard time finding planets like our own solar system around other stars. Which may mean that all those stars where we can't find planets? Might be the ones that actually have earth like planets.
posted by empath at 4:49 AM on July 23, 2010

It worries me that everyone who finds their keys outside at night does so under a streetlamp. What mysterious force is causing people to lose their keys under street lamps??
posted by DU at 4:52 AM on July 23, 2010 [8 favorites]

What I don't understand is this: In order to make these transit calculations, we would have to be looking at these systems edge-on, right? Are a large percentage of solar systems in the galaxy oriented exactly north-south? I would think this would severely limit the number of observations of this type we could make...
posted by joecacti at 5:50 AM on July 23, 2010

They keep finding those 100 new planets.

Wow. . . one for each "eskimo word for snow".
posted by Herodios at 6:04 AM on July 23, 2010

joecacti: Not exactly edge on, and radial velocity measurements work better with more inclined orbits than optical transits I assume, but one can calculate the probability assuming random distributions of inclination to figure out statistics etc. It's limiting, but there's plenty of other limiting factors to consider too.
posted by edd at 6:04 AM on July 23, 2010

In order to make these transit calculations, we would have to be looking at these systems edge-on, right?

For the transit ones, more or less.

Are a large percentage of solar systems in the galaxy oriented exactly north-south?

I think you are misvisualizing the geometry. Even if we say it has to be *exactly* edge on, there's a whole plane the extrasolar system can be oriented in. If we allow some obliquity, then there's some solid angle of exclusion where you can't observe a transit (which depends on the parameters of the system, such as planet and sun size and closeness of orbit).

But you don't actually have to see a transit either, if you are instead observing wobbles. In that case, you'd like your view to be exactly those views that are bad for transits.
posted by DU at 6:10 AM on July 23, 2010

"But you don't actually have to see a transit either, if you are instead observing wobbles. In that case, you'd like your view to be exactly those views that are bad for transits."
Eh? I don't understand that. They're both best when the orbit is edge-on.
posted by edd at 6:26 AM on July 23, 2010

Bah. Stupid tabloids.

Still, they are finding rather a lot of them these days, and they are getting smaller and more rocky.
posted by Artw at 6:27 AM on July 23, 2010

A wobble is just fine edge-on, but I would think it would be best end-on. The first is a projection of circular motion, the second is actual circular motion. Extra degree of information.
posted by DU at 7:00 AM on July 23, 2010 [1 favorite]

Dear Douglas Adams, how big is the universe?
posted by blue_beetle at 7:02 AM on July 23, 2010

Fine. I think one earth is plenty enough.
posted by notmydesk at 7:21 AM on July 23, 2010

700 planets? That's a lot of scanning just to find more Element Zero.
posted by yeti at 7:43 AM on July 23, 2010 [1 favorite]

So do the each have their own God, one per "Earth", or do they all fall within the previous jurisdiction?

This is important. I need to know just where to move the goalposts.
posted by LordSludge at 7:46 AM on July 23, 2010

The problem with the end-on ones is that you typically have information only about the line-of-sight velocity of the star (to the precision we are talking about here, i.e. tens of meters per second). In the case of a planet orbiting in the plane of the sky around a star, the line-of-sight reflex motion of the star is zero. In other words, you really do want mostly edge-on systems for the radial velocity detections, too.

If we could routinely measure motion in the plane of the sky to the same precision as radial (line-of-sight) velocities, this wouldn't matter. But we can't.
posted by chalkbored at 7:58 AM on July 23, 2010

What always concerns me about extra-solar planet discoveries, Earth-like or Jovian, is that their orbits always seem to weird compared to our own solar system.

The discovery of so many of these sorts of planets would seem to me to add a bit of support to the Rare Earth hypothesis. Lots of planets, lots of possibilities for extremophiles or other microbial life, but not necessarily very many Earths.
posted by CosmicRayCharles at 8:20 AM on July 23, 2010

DU: you may be missing that it's a lot easier to measure velocity along the line of sight than position transverse to the line of sight.
posted by edd at 8:28 AM on July 23, 2010

Or what chalkbored said. Curse mobile internet disappearing.
posted by edd at 8:30 AM on July 23, 2010

There is almost zero chance that any of our current methods (other than Kepler) would detect an earth-like planet right now. We know this. The fact that we don't detect earth-like planets, with instruments that can't detect earth-like planets, tells you nothing about the prevalence of earth-like planets.
posted by kiltedtaco at 8:30 AM on July 23, 2010 [3 favorites]

joecacti : What I don't understand is this: In order to make these transit calculations, we would have to be looking at these systems edge-on, right?...I would think this would severely limit the number of observations of this type we could make...

My understanding is that they look for the wobble first, which might be indicative of an extra solar planet, and then they start investigating to see if there is an observable transit. Using the wobble allows them to significantly reduce the number of stars that they are looking at, IIRC Kepler looks at something like 150K stars and measures variations in brightness once every 30 minutes or so.
posted by quin at 8:42 AM on July 23, 2010

What I don't understand is this: In order to make these transit calculations, we would have to be looking at these systems edge-on, right? Are a large percentage of solar systems in the galaxy oriented exactly north-south? I would think this would severely limit the number of observations of this type we could make...

Mostly just guessing here, but I'd expect most planets' orbital axes would be fairly similar to the galactic axis around which they are rotating, yeah. For the exact same reason that the axis a planet rotates on is similar to its orbital axis. Obviously there's some variation, but there's a lot of force being exerted through gravity pulling things together along the galactic plane, and a lot of momentum that's going to make the two ellipses tend towards being coplanar, and to become more so over time.
posted by atbash at 8:45 AM on July 23, 2010

Our solar system is a paradise in comparison to some of the other hellholes out there.
posted by pashdown at 9:14 AM on July 23, 2010 [3 favorites]

Atbash: our own solar system is fairly inclined. I don't think the correlation is as strong as you think.
posted by edd at 9:26 AM on July 23, 2010

My understanding is that they look for the wobble first, which might be indicative of an extra solar planet, and then they start investigating to see if there is an observable transit.

If we're talking about Kepler specifically, it's more like the other way around. There are many, many more stars in the Kepler field than have ever been targeted for radial-velocity-based searches, so if you had to wait for the RV "wobble" first and only then do follow-up with Kepler, you'd be waiting a very long time for results. So they're mostly identifying promising candidates from the Kepler data, then doing RV follow-up on some of them (though, as everyone has noted, some of the systems detectable by transits will not be detectable in radial velocity measurements).

So, just to summarize, the basic program is something like this: Kepler is looking for the periodic brightening and dimming of stars that you'd expect as a planet passes in front of a host star¹. Because there are lots of things other than planets that can cause a star to look brighter or dimmer, the Kepler team will probably not announce a planet as "confirmed" until they have seen it transit, say, three times.² What this means is that planets that are genuinely "Earth-like" -- i.e., planets of roughly Earth's size, orbiting with periods of a year or so -- probably aren't going to be trumpeted as solid detections for two to three years.³ But there are lots of "candidate" systems, where they've presumably seen, say, one transit. In the meantime, you can bet your ass that people affiliated with Kepler are doing lots of follow-up radial velocity measurements on those candidates, hoping to confirm them that way.

It is hard to convey just how insanely difficult this whole thing is. Kepler is simply a beautiful mission, and I think the results coming out of it are going to be transformative -- both for our understanding of planets around other stars, and for all kinds of other stuff (like the way we think about stellar activity and rotation, for instance) too. I can't wait.

¹: At the level of precision needed to detect a transiting Earth, all kinds of other effects come into play. Starspots, for instance, like the spots we see on the surface of the Sun, can make a star look a little dimmer (or, weirdly, brighter -- e.g., the Sun is actually a little brighter at sunspot maximum); this variation can actually be bigger than what you'd expect from a transiting Earth.

²: Seeing a planet transit twice gives you the period of its orbit around the host star, so you can then make a prediction for when you ought to see the third transit. If you see it, great -- detection confirmed.

³: It could be less. Many of the stars Kepler's looking at are dimmer than the Sun, so the "habitable zone" (which is basically the region where a planet could have liquid water at its surface) is closer to the star than we are to the Sun. That means the orbits take less time, so you don't have to wait quite as long for the planet to transit a few times. I have no idea how many of the "candidates" the Kepler team has mentioned are around these types of stars, though.
posted by chalkbored at 9:34 AM on July 23, 2010 [3 favorites]

On the other hand: Claims of 100 Earth-Like Planets Not True
posted by saulgoodman at 9:45 AM on July 23, 2010

So do the each have their own God, one per "Earth", or do they all fall within the previous jurisdiction?

Every God has a sector with many planets and stars. Sort of like the Green Lantern Corps.
posted by octobersurprise at 12:45 PM on July 23, 2010 [2 favorites]

Morbid as it sounds, a lot has been made about the "when I die shoot my ashes into space" concept. Me, I volunteer to man a one-way mission to die on another planet.
posted by swimming naked when the tide goes out at 1:06 PM on July 23, 2010

Way of classifying digital numbering schemes: Big Endian.
Way of classifying extra solar planets: Big Assian.
posted by storybored at 1:19 PM on July 23, 2010

So do the each have their own God, one per "Earth", or do they all fall within the previous jurisdiction?

This is important. I need to know just where to move the goalposts.

This is absolutely not a problem for the Mormons. They may even be vindicated :-P
posted by dibblda at 3:41 PM on July 23, 2010 [1 favorite]

Does this mean more episodes of Star Trek (TOS) are possible?
posted by vapidave at 7:33 PM on July 23, 2010

There are a lot of low-grade 'science news' sites out there these days. Not just anyone can do science reporting. They (not we) should be checking the old reliable sources (for example, what SciAm had to say on the subject back in January) before they fly-off-the-handle and confuse the innocents.

Spacecraft are not magic crystal balls; they don't find anything ... they ship data back to the ground where it has to be interpreted by intelligent beings. Too, the Kepler team ought to keep a lid on its joy, joy around reporters until it's got -science- to report.
posted by Twang at 10:26 PM on July 23, 2010

Damn, I'd already packed! Now where am I supposed to go?
posted by tiamat at 5:40 AM on July 24, 2010

I don't work directly with this crowd anymore, but I know several of these people. What they're doing is being cautious and making DARN sure that it's a planet before officially announcing it.

Not that I know anything, even if I dropped them an email they'd be pretty tight-lipped but I wouldn't be surprised to see the hint of a Cheshire-cat grin.

If you had the chance to announce that you've found a planet the size of Earth right in the middle of the habitable zone of a star, wouldn't you be a little careful about making sure, too?
posted by chimaera at 3:19 PM on July 24, 2010