It’s the biggest thing in extrasolar planets right now so I figured it deserves an obligatory post. I would like to give more attention to instrumentation, techniques, and technology in these entries, and a post on the Kepler spacecraft seems a wonderful way to start.
The spacecraft was launched into a Heliocentric orbit on March 7, 2009 on top of a Delta II rocket. Compared to some other spacecrafts, Kepler is rather simple in design and purpose. It’s essentially a dedicated photometer attached to a 0.95 metre telescope operating in visible light (more specifically, from 400 – 865 nm wavelengths). It observes nearly 150,000 main sequence stars continuously, using the transit method to discover extrasolar planet candidates (see here for a description of how planets are found this way). It has uncovered thousands of planet candidates using this method. Light enters into the front of the telescope, bounces off the primary mirror at the back, and is focoused onto the CCDs in the middle of the telescope body. These CCDs measure the brightness of each star every 29.4 minutes for most of the stars, but some special target stars get high-cadence observations, with measurements being taken once every minute.
The CCD (imaged above) is what does all the magick. One of the squares malfunctioned and no longer works, but beside that and some trouble with the spacecraft going into safe mode and resetting early on in the mission, everything continues to go well as of this writing.
What the spacecraft ends up seeing is the following:
Kepler sees this, all day, every day. Except for the seasonal 90° roll to keep the solar arrays aimed at the sun, this field of view does not change. The image appears to be mostly hazy, but upon closer inspection, it’s actually comprised of an obscene quantity of stars (see the full image here, but careful for it is a large image).
Despite the simplicity in its design and purpose, Kepler is revolutionising the field of astrophysics and extrasolar planets. Contributions of Kepler to astronomy include studying active galactic nuclei, finding additional Hot Jupiters, performed asteroseismology on red giant stars, studied the central stars of planetary nebulae, discovered more eclipsing binary stars, finding the first transiting giant planet around a limb-darkened star and constraining its spin-orbit alignment, performed asteroseismology on white dwarfs, performed asteroseismology on sun-like stars, studying red giant granulation, finding a class of bloated white dwarfs, measuring the frequency of terrestrial planets around sun-like stars, permitting the public to discover their own planet candidates, discovering a non-transiting planet through transit timing variations for the first time, improving our knowledge of RR Lyr stars, studying stars that tidally affect each other, studying stars in open clusters, measuring giant planet reflectivity, studying sdB star pulsation behaviour, studying B-type stars and roAp stars, and work is progressing toward other fields, including discovering extrasolar moons.
The mission was designed to last 3.5 years. The need for at least three years is due to the requirement to detect three transits of a planet to confirm its candidacy. One transit reveals the planet’s existence. The second transit constrains its orbital period. The third transit confirms the orbital period. While this need for a third transit might seem redundant, consider the case of two similarly-sized planets transiting a single star. One can see how one transit of both planets may be confused for two transits of one planet. A planet at 1 AU from a solar-like star will have a period of about 1 year (like, for example, Earth). Therefore, we expect all extrasolar Earth clones to transit roughly once a year. And therefore three years of observations are required to confirm the planet as a candidate.
However Kepler found that sun-like stars are more photometrically variable than expected. It turns out our star is a bit quieter for its type. What this means is that there is more noise in the data, and the transits do not stand out as much. multiple transits must now be stacked to get more data to confirm the planet. The punch line is that for Kepler to get a complete measurement for the frequency of Earth-sized planets in the habitable zones of solar-type stars (so calledor “eta Earth”), Kepler‘s mission must get extended to six years. Kepler has the fuel onboard to do this, but the funding has not been secured. The topic is a discussion for later…