Pictures of Extrasolar Planets

It’s fairly common nowadays to read about the discovery of extra-solar planets. See here for the latest. However, all of the planets detected so far have been done so by indirect means, typically by examining the “wobble” of the parent star.

My question is, what kind of technology is it going to take to ever get a picture of an extrasolar planet?

The problem is that a planet is a faint object very close to a very bright object (the star). A lot of the extrasolar planets we’ve found are very close to their stars- closer than Mercury is to the Sun. Those are the easiest ones to find with spectrographic techniques (like the technique used to find 51 Peg, the first known extrasolar planet), but are hard to image directly because they are so close.

Some people are looking into adaptive optics for direct imaging of extrasolar planets. Adaptive optics involves using a ground-based telescope (orders of magnitude cheaper than a space-based telescope, and many ground-based telescopes are a lot bigger than Hubble) and making various corrections to the image to remove atmospheric effects (twinkling stars and the like).

Of course, adaptive optics is only going to show an extrasolar planet as a dot next to a star. If you’re looking for a picture showing the level of detail of the Voyager images of Jupiter or Saturn, it’s probably going to require an interstellar space probe of some sort.

What Anne said. The star is so bright, and the planet so comparatively dim, that it’s like trying to see a candle in front of a floodlight.

There are a few projects underway that should help observers cancel out the light of the star through (IIRC) interferometry. As I understand it, these devices use multiple telescopes and combine the two images using computers that can use complicated comparative algorithms to cancel out parts of the image. Presuming I’m remembering my Science Channel documentaries right, there are two designs currently nearing completion, one with two mirrors, and one (in Argentina?) with four, plus one on the drawing board that uses six imagers in orbit (out at a Lagrange point, far beyond the reach of regular maintenance).

I’m a layman, though, and my knowledge ends there. :slight_smile:

[QUOTE=Cervaise]
What Anne said. The star is so bright, and the planet so comparatively dim, that it’s like trying to see a candle in front of a floodlight./QUOTE]
More like trying to see the hairs on an ant crawling across a floodlight.

When you start asking about pictures of extrasolar planets, it’s important to realize that we’re able to resolve very few (IIRC three) stars – and those are relatively nearby giants or supergiants. Everything else that’s a coherent object (as opposed to a nebula or galaxy) shows as a point of light.

That said, I thought I remembered seeing that visual identification had been made of one extrasolar planet – a superjovian orbiting an M dwarf. This would of course have been point-of-light-on-the-image, not a photo of a planetary surface with a giant sign saying “Stick Your Head in a Pig” or some such. I don’t remember any details of the story, though, so count it as extrasolar legend until someone knowledgeable comes along to confirm or refute it.

Resolving power of a telescope is directly proportional to its diameter. If we want to resolve a Jupiter-sized planet 10 light years away, we need a resolution of 0.5 milli-arcseconds, which requires a telescope 170 meters in diameter. Fortunately we don’t need a single mirror that big; an interferometer with a 170-meter baseline (i.e. several telescopes placed along this length, with their light combined into a single beam) will be almost as good. By the way, by “resolve” I mean recognize it as something larger than a point source. If you want to see patterns on the surface, you probably need to up the resolution (and size) by a factor of 5 at least. That’s almost a kilometer.

I don’t think adaptive optics works at that level. So what we need is a structure in space, 1 km long, with an accuracy and stability of around 50 nanometers (1/10 of wavelength of visible light). Alternatively it could be a group of satellites flying in formation, controlling the relative position to that accuracy. There is a lot of demand for precision formation flight technology*, so I’m optimistic that it will become a reality in maybe 30 years.

I have to disagree with Ann et al. about the glare from the main star. It would be a problem if the telescope has just enough resolution to separate the planet from the main star. But here we’re talking about a telescope that can resolve features on the planet. The main star will be well outside the field of view of the detector, and its glare will not be a problem.

*At least from the scientific community. One example would be a giant telescope which consists of one spacecraft carrying the telescope mirror and a separate satellite carrying the detectors floating at the focal point of the mirror.

Well, the Terrestrial Planet Finder is an actual space-based interferometry array that will fly perhaps within the next ten years. It will be able to find earth-type planets and measure their atmospheres accurately.

The TPF will use ‘nulling’ to eliminate the glare of nearby stars to detect the light reflected from the planets themselves. It will have a resolution about 100 times greater than Hubble, although I don’t think it will be enough to actually resolve the disks of planets around other stars, it will be able to make good measurements of their characteristics such as size, mass, atmosphere, temperature, rotation, etc.

After the TPF program, there are larger telescope arrays in the early planning stages. The NASA Origins program has laid out early-stage proposals for several really exciting missions:

The Life Finder is an even larger telescope array optimized for studing the atmospheres of other planets for signs of life. LF will be able to detect changing seasons, methane, temperatures, and all sorts of markers that will tell us an amazing amount about other earth-type planets around nearby stars. Life Finder should actually be able to detect large biomasses like forests and algae-rich oceans. Life Finder could be flying within 30 years.

Finally, the largest array currently on the drawing boards is the Planet Imager, a telescope array 360 kilometers in diameter, which would be theoretcially be able to resolve images of other planets about 25 x 25 pixels. This is enough to make out continents, oceans, mountain ranges, large craters, etc.

Once you’re out in space, the limits of interferometry are pretty huge. Given a large enough array, I think we should be able to image planets around other stars with the kind of resolution we have for imaging the moon from Earth.