Looking for Planets in all the wrong places

Factual Questions
1

OK, maybe not the wrong places, but I like the catchy title…
When we look for planets outside our solar system, we detect them by measuring the decrease in radiation from a star as a planet passes between it and the earth (or wherever our telescopes are). So, we only see planets who orbits are a plane defined by the star and earth.

Is there any reason that planets in our galaxy will tend to more often be in such a plane as opposed to some other plane in which the orbit never takes the planet between it’s star and earth? The only thing I can think of is that the galaxy has a preferred direction (defined by the spiral), but I have no idea if that influences planetary orbits.

IOW, we can only “see” a tiny, tiny fraction of the planets out there, right?

2

Yep, we’ll only see a very small fraction, but even then it’s enough to get a good sample. Astronomers can then estimate how many undetected planets remain, based on a number of assumptions (i.e. totally random distribution of orbital planes, or some amount of bias due to the overall rotation of everything in the galaxy). If you have a method to reliably detect the remaining majority of planets, there’s a Nobel prize waiting for you…

ETA: There’s damn few situations where science can observe and catalog every single type of thing in existence. That’s why we work hard to get good samples and apply good statistics to infer things about the entire population.

3

Do we also not detect the more massive planets by the gravitational tugs they give their stars? That method is independent of orientation.

4

Not entirely, since astronomers measure spectral doppler shift caused by stars wiggling back and forth towards us, since a lateral change in position is too small to measure (usually… wiki says there have been a few controversial attempts with that method). Thus there is still a significant bias towards planets with orbital planes that line up with us, though the alignment doesn’t have to be perfect with the doppler method.

5

There are other methods of detecting planets.

It’s a simplification to say that planets orbit their suns. Rather, they both orbit the centre of gravity of the system. That means that a star with planets wobbles a bit. The wobble can be detected by astronomers, and they can deduce the presence of planets.

Edit - beaten to it.

6

You all may already know this and hence the thread, but the 219th meeting of the American Astronomical Society concludes today. Among other things, planetary searchers announced their latest findings. 160 Billion Alien Planets May Exist in Our Milky Way Galaxy | Space

A few highlights:

Stars in our galaxy have on average 1.6 planets per star, so that’s 160 billion planets in our galaxy alone. That results in a lot of moons, too.

Small planets like Mercury, Earth, as well as super-Earths (10x the size of Earth) etc. are more common than gas giants (just like our system).

There are huge numbers of rogue planets flying around the galaxy unbound to a parent star.

Tatooine-type planets with two suns in the sky are not only possible but common.

Cool stuff.

7

How can they know that?

8

Yep, at least right now. There are plans for larger orbital telescopes that should be able to see with better resolution, or detect more subtle changes, but in the end we will probably never be able to see more than a very tiny fraction of planets, since they are so hard to see.

As others noted, there are other ways to detect probable planets. I think the wobble method discussed earlier was actually the first one used to find an extra-solar planet. What amazes me is they are able to get so much data from such a small variation, even to roughly the composition, temperature and size of the planet in question.

Probably because our galaxy formed from a large spinning disk of gas and dust, so things tend to stay in that plane unless there are collisions or perturbations. I don’t think our galaxy has (yet) to collide with another galaxy, and that’s when things tend to get really weird. It’s going to happen though…I seem to recall that there is another galaxy that is on a collision course with ours some time in the next several billion years. At that point you will probably have a lot of junk flying about outside of the general plane.

-XT

9

You might find this interesting.

10

**tdn **has got it. :slight_smile:

11

Fascinating article, but I’m somewhat taken aback by the startling spelling of “fiercesome.” That didn’t look odd to the author or any editors Discover still might be using?

12

Aside from the transit method (used by the Kepler telescope) and the Doppler-shift method (the only successful method until a few years ago), there has also been proposed a third method known as [. Basically, you take a bunch of telescopes and combine the light waves coming into them in such a way that the light from the star cancels out, but the light from stuff orbiting around the star doesn’t. This greatly mitigates the “firefly-next-to-a-searchlight” problem that has been the bugbear of exoplanet research for years; in principle, this method could detect a planet whose orbital plane was in any orientation at all. There have been a couple of proposed [url=Darwin (spacecraft) - Wikipedia]dedicated](]nulling interferometry[/url) projects to use this technique, but it would be a pretty tricky thing to pull off to find Earth-like planets; you basically need to either fly satellites in a formation that never deviates by more than a few millimeters, or park the thing on the dark side of the Moon. Maybe in 50 years.

I think that the prevailing thinking is that the orientations of planetary systems will be pretty much random relative to the galactic plane. (I recall hearing that this is the case for binary star systems, at least.) Also note that the Solar System’s ecliptic plane is nowhere near the galactic plane — it’s about 60° off.

That would be the closest comparable galaxy, the Andromeda Galaxy. However, it’s still not clear whether it will actually collide with us. We probably need another few thousand years of data before we know for sure.

13

From that article:

Imagine what it would be like for life on a moon heated by tides orbiting an orphaned planet. Fascinating!

14

The big advantage of the Kepler method over the wobble method is that it can detect Earthlike (in size and orbital distance) planets. The wobble method could, in principle, do this too, but our instruments aren’t anywhere near good enough to pull it off-- Mostly all we can see is “hot Jupiters”, gas giants that are very close to their parent star. Yeah, only about 1% of planets will be lined up well enough to use the Kepler method, but you can deal with that by just observing millions of stars at once (which is what Kepler does).

Well, there are two fundamental properties of the wobble: How big it is, and how long its period is. And we generally have a pretty good idea of the mass of the parent star, based on its spectral type. Given the mass of the star and the period of the orbit, you can use Kepler’s laws (actually, Newton’s form for them) to find the mean distance of the planet from the star. Given how large and hot the star is (again, known from its spectral type) and how far away the planet is, you can estimate the temperature of the planet. And based on how big the wobble is, you can estimate the mass of the planet compared to the mass of the star, though that runs into the problem that the inclination angle will also change the size of the wobble we see: What they actually measure isn’t m, but m*sin(i), where m is the mass and i is the inclination angle.

If you detect the planet through an occultation, like Kepler does, then you can learn a different set of information. You can still get the period and all that follows from that (distance and temperature). Based on how long it takes to transition from full brightness to reduced brightness, you can get the radius of the planet (you can also in principle get a lot of other information about the orbit from the shape of the dropoff, but I don’t think this has been done with planets, only with stars). You know that the inclination is very close to 90 degrees, because if it weren’t, you wouldn’t have detected it in the first place. And if the planet has an atmosphere, you can sometimes even learn something about its composition, from the light that just skims the planet and passes through that atmosphere.

In the ideal case, of course, you can detect both the wobble and an occultation. In that case, Bob’s your uncle, and you can learn pretty much everything there is to know: All of the orbital parameters, the size and mass of the planet (which pretty much tells you the composition of the planet itself), and so on.

15

Impressive. I had no clue they could detect such tiny variations.

16

Another thing they can discover with the transit method is if the planet has any large satellites. They haven’t found any yet, but they’re looking. I expect it takes the observation of a goodly number of transits to find a moon. This page here gives a fairly accessible idea of how it works.

BTW, someone above asked what method the first exoplanet was discovered with: Doppler-shift or transit. It turns out it wasn’t either one. The first exoplanets were discovered around a pulsar. Irregularities in the timing of the pulses, when they should have been very regular, revealed the existence of the planets.

17

I don’t know enough to imagine it. Can you expound on this? I’d love to know what that would look like.

It’s going to be tough to memorize all the planet names now.

18

A bit of a nitpick, because I think it’s important, and because that figure seemed low to me: The article initially says at least 1.6 planets per star, and later clarifies that the 1.6 planet figure is for planets in the range 0.5 to 10 AU, about to Saturn. So no Mercury, Neptune or Uranus included in that figure.

The wobble detection method requires at least one orbit, and better several, to determine there’s a planet. I also believe the transit method requires at least two or three measurements of the blip before they call it confirmed. We haven’t been looking long enough to detect planets farther out yet, so that 1.6 planet per star figure will rise.

(Not sure where the 0.5 AU lower bound comes from.)

19

I agree, it’s a conservative number and will grow.

20

Also note that the 1.6 planets per star includes all stars, including those with no planets at all. It’d be consistent with most stars having none and a few being like our Solar System, for instance. And I don’t think there are any red dwarfs in the Kepler sample (they’re just too dim).