My understanding of the Kepler and TESS missions is that they detect exoplanets by looking for changes in a star’s intensity as a planet transits in front of the star.
Does this mean that we are only able to detect planetary systems when the orbital plane is oriented just right so that we can detect the transits?
The other method is looking at a star’s wobble, which would be indicative of an orbit around a common barycenter outside of the star’s core. In other words, if a Jupiter sized planet is orbiting a star, it will tug the star a small amount in its direction.
Aka Doppler spectroscopy.
No, there’s also the radial velocity/doppler shift technique. There are other techniques, but I don’t think any have actually resulted in the discovery more than one or two exoplanets.
Ah, ok cool. That makes sense. Of course then the question becomes how accurate those measurements of planet size are since the spectroscopy change from the perspective of earth would depend on the inclination of the orbital plane.
Also, how does this method account for multiple planets?
I see some of these limitations are discussed in the Wikipedia article.
The Doppler shift method only gives minimum masses for the planets and never the size. If we know the angle of the planet’s orbit (which we virtually never do except for transiting planets), we could get the true mass by multiplying by the cosine of the angle it makes with the line between the star and the Earth.
For multiple planets, I believe you could take a long stream of data (measured over several times the period of the outermost planet) and run it through Fourier analysis to get info (how many, periods, perhaps masses) on the various planets.
The Kepler satellite only uses the transit method, and yes, that does mean that it misses a lot (about 99%). But that’s not so bad, for two reasons: First, 1% of stars is still a heck of a lot of them. And second, the percentage missed is really easy to calculate precisely, meaning that we can extrapolate very well from the number we see to say how many there are total. Meanwhile, the transit method has the huge advantage over the Doppler method that it can detect Earthlike planets and solar systems like our own: The Doppler method is strongly biased towards finding high-mass planets which are very close to their star (“hot Jupiters”), and can’t say anything about how common planets are which it can’t detect (like, say, any of the ones in our system).
Oops, mistake in the above. Divide, not multipy, by the cosine.
To get an idea how limited the transit method is, you should think on the fact that we see transits of Venus only 4 times in every 243 years.
An article from 4 days ago showing the use of filters to detect objects.
Still not the best, as the object is 214 AU from the binary pair and is suspected of being at least Jupiter mass.
For reference, Eris ranges from 37 to 97 AU from the sun and likely wouldn’t be detected with this process.
2012 VP113 is a rock anywhere from 300 to 1,000 km in diameter with an orbit that varies from 80 to 438 AU, the farthest known object orbiting the Sun, but I don’t think they could possibly detect something so small and so dim around another star.
Cool! Off-topic — the article claims the binary pair (as a system?) is only 2 or 3 million years old — can that be right? Sounds awfully young to me.
Age (or at least lifespan) of stars correlates strongly with star type, and that article doesn’t say what they are, but the mentions of accretion disks does seem to indicate a very young system. And of course, even old stars were young once (insert Harrison Ford joke here).
There are actually some exoplanets detected by direct imaging. This boggles my mind. Check out Lists of exoplanets - Wikipedia and click on the sort button for the Discovery Method column. Look at all the ones discovered by “imaging”. For some of them, if you click on the Name you get to see images, including images of them traveling around their star.
Thanks. Great post/username combo!