Our solar system is pretty neat and orderly. Yeah, it has some issues, but in general we can make some broad statements about it: the planets all orbit the Sun in the same direction, for one thing, and they also orbit pretty much in the same plane. If you look at the system from the side, the orbits would all look flat, like a DVD seen from the side.
That's left over from the formation of the solar system itself, which happened when a cloud of dust and gas collapsed into a disk. The planets formed from that disk, so they all orbit in roughly the same plane. We see other systems forming in the same way, so we assume that when we look at those planets, they'll also have all their planets in a plane.
Oops. Maybe not so much. Astronomers have just announced that they've confirmed a system where the planets are not all aligned this way, and in fact the planets are titled relative to each other by as much as 30°!
Ironically, the parent star is Upsilon Andromedae -- that made me chuckle, because it was one of the very first stars found to have planets orbiting it, back in 1996. It's actually a binary star, two stars orbiting each other; one is a star slightly more massive and hotter than the Sun, and the other a dinky red dwarf orbiting pretty far out (well outside the frame of that illustration of the system above). Three planets (called Upsilon Andromedae b, c, and d) at least are known to orbit the primary star. The planets were initially detected by their gravitational pull on the star; as they orbit they move the star in a mutual tug-of-war. We can't (usually) see that motion directly, but it can be detected as a Doppler shift in the star's light.
Due to the physics of the situation, that method only gives us a minimum mass for a planet. The actual mass might be much higher. It also doesn't tell us the tilt of the orbit of the planet, or of any of the other planets in the system.
What's new here is that astronomers used telescopes on Hubble called the Fine Guidance Sensors, which are incredibly accurate and highly precise. The FGSs are so accurate that they could see the physical motion of the star on the sky, the wobbling as the planets tugged on it this way and that. Think of it like a harried parent at a mall with two little kids holding her hand. As the kids see one store or another they want to visit, they pull on her in different directions as she walks with them, so her path down the mall corridor shifts left and right.
Combining the new Hubble data with the older Doppler data has revealed a wealth of information about the planets in that system. For one thing, it nailed the masses. Instead of lower limits, we now have accurate masses for planets: Ups And c is 14 times the mass of Jupiter, and Ups And d is 10 times Jupiter's mass*. Mind you, Jupiter is a bit of a bruiser, so these are hefty planets. These masses are far larger than thought before, so the new observations really changed our thinking here.
But the amazing thing is that it looks like Ups And c and d are in wildly different orbits: instead of being almost exactly in the same plane as expected, they are tilted relative to one another by 30°! The illustration on the right compared those orbits with those of planets in our own solar system, and you can see how weird this is.
But does this mean astronomers are wrong about how planets form?
Probably not. We're pretty sure we understand how planets form, at least in general terms. What this does mean is that something happened to the planets after they formed, something that tossed one or both of these planets into different orbits than the ones they were born in.
This isn't a huge surprise. Pluto may or may not be a planet by your definition, but it orbits the Sun at an angle of 17° with respect to the Earth. Sedna, an object about the same size as Pluto in the outer solar system, also has a large tilt. We know there is some mechanism that can change the orbits of big objects in the solar system, so why not in other systems, too?
In the case of Upsilon Andromedae, we have some culprits. The data hint that there may be a fourth planet orbiting the star. It's not clear if it's there or not, but if it has an elliptical orbit it could gravitationally affect the inner planets. There's also the red dwarf star orbiting farther out. Far more massive than a planet, its gravity may have some effect on the system as well. It's also certainly possible that there are other influences we haven't seen or thought of yet. [Update: I just got off the phone with the team who did this research, and Rory Barnes told me that a strong possibility as well is that there were more planets in the system initially. They would have interacted via gravity, and affected each others' orbits. A likely scenario is that a planet with about ten times the mass of Jupiter could have messed up the orbits of the other two, then been ejected out of the system. This is a common outcome when you have lots of massive objects in one system.]
The point here is that in general, our theories of how planets form is pretty good. As we study more of these systems, we'll get more and more data under our belts that will help us catalog and understand where these systems follow our theories, and where they seem to diverge. That's all good news! Theories only go so far in explaining everything, and as we observe more we modify those ideas, add to them, so they better represent the Universe around us. That's how science works, and that's how we learn.
* Unfortunately, Ups And b orbits too close to its parent star to get an accurate mass for it. That'll have to wait for the future, with new techniques and better instruments.
Image credit: NASA, ESA, and A. Feild (STScI)