Terrestrial Planet Finder

solarlux writes "The Terrestrial Planet Finder has taken one step closer to reality as two architectures have been approved by NASA. The first, TPF-c, will be a single optical telescope which employs a coronograph to block starlight for planet detection. TPF-i will be a flotilla of infrared telescopes flying in formation to form a interferometer. TPF-i will analyze the planets identified by TPF-c for life signatures. The telescopes are to be launched within the next 10-15 years."

Comments

Challenges of finding extrasolar planets

[klipsch_gmx]

The question becomes even more convolved once we move outside the solar system, since we now know of a wide diversity of systems, of which our own solar system is only one particular instance. (And perhaps not even typical at that.) We know that there are objects extending all the way down from massive stars (around 100 Msun) to hydrogen-burning stars like our sun to brown dwarfs to planets. Clearly any definition of a planet must apply not only to our solar system, but also to these extrasolar systems. Some of these systems are much like our own (for instance, they may contain a brown dwarf orbiting a star, or a planet orbiting a star), and some (including a few systems of low enough mass to qualify as a planet) are "free-floaters" -- just sitting out there by themselves in space.

I think ultimately the question is whether there is a single continuous "initial mass function" of isolated objects or not. The best idea as to how stars acquire their initial mass is that turbulence in the interstellar medium, which exists on all scales, establishes a power-law distribution of initial masses. Every once in a while, you get a very strong shock which passes by inside a giant molecular cloud and forces the collapse of a large region which then goes on to form a massive star. But more typically, you form stars more like our sun. And just as rare as massive collapses are very small mass ones which go on to form isolated brown dwarfs and free-floating planets. If this model holds up to be true, then we are all mincing words in our definitions of isolated systems, since they are all manifestations of the same universal formation process.

However, to avoid the difficult question of formation mechanisms, an IAU working group of some of the most respected people in the field established a working definition to define by fiat what it means to be a brown dwarf, and a planet. Extrasolar "planets" are those objects orbiting a star which are beneath the deteurium-burning limit -- regardless of how they are formed. "Brown dwarfs" are defined to be those which burn deuterium but not lithium, and "sub-brown dwarfs" (NOT free-floating planets!) are defined to be those isolated objects which do not burn deuterium. Even the working group itself admitted that this definition was not satisfying to a single member of the group, and so it is likely it will be replaced at a later time with something more physically-motivated. The "planet/planetismal/KBO" distinction was pushed back to our own solar system, since it will be some time before anyone sees anything that small in another system.

Also of interest is the following link, which gives a history of previous claims for additional planetary members of our solar system : SEDS.

Re:10 to 15 years

[kevlar]

The chances of finding an Earth-like planet "in our neighborhood" is far greater than finding one 100+ Ly away. This is due to two reasons:

1) Closer planets are easier to detect (for obvious reasons)
2) The heavy metal content in and around our "neighborhood" is greater than that which exists generally through out the Milky Way. This is because before the Solar System was formed, a massive star exploded seeding the area with heavier metals (iron+ on the periodic scale). These heavier atoms are obviously what makes up the Earth. Without this initial seeding, the solar system would only contain hydrogen based planets like Jupiter. Therefore, our local area is the best place to find heavy-metal planets.

Some details

[photonic]

I am somewhat involved with the European version of these missions (the Darwin mission, to be launched around 2014), so I might clear some things up.

Goal: to detect earth-like planets around other starts. Extra-solar planets detected thus far are usually 'hot Jupiters': big planets that orbit the star in a few days. These are relatively easy to detect. Detecting an earth-like planet (which have not been found yet) is far more difficult. It is usually compared to detecting the light of a firefly (reflection of the planet) flying very close to a lighthouse (the star). Measurements need to be done in the far infrared because there the ratio between the planet and the starlight is the highest (but still only 1:10^6 !!). With some luck they might find traces of ozone and CO2 in the spectrum that might be an indication for life.

Methods:
-Coronography: Simply put it is just a conventional big (~10 meter) telescope with a shadow mask that blocks the light of the star. The light of the planet should get past the mask on the detector.

-Interferometry: Somewhat similar to the techniques used in radio astronomy. The resolution of a telescope improves by increasing its size. The trick is to combine several small telescopes. The resolution should then be comparable to the resolution of one big telescope that is as wide as the separation between the small ones. With radio interferometry you can do the 'beam combination' by computer. In optics however you have to physically combine the beams of the different telescopes. This requires flying satellites in formation with stabilities on the order of nanometers!! Current schemes are limited to several hundred meters. There are also some attemps to do this on earth.

There is quite a lot of politics going on between NASA and ESA at the moment about how they should cooperate. First ideas where to do an interferometry mission together, but now NASA has decided to go for coronography and postpone interferometry to 2020. ESA is sticking to interferometry.