Detecting Extrasolar Planets

The first detection of extrasolar planets was announced in 1995 by two competing groups. One was headed by Geoff Marcy, then at San Francisco State University, the other by Michel Mayor at the Geneva Observatory. The same technique was used by both: measuring the back and forth motion of a star due to the teeter-totter like reaction induced by the orbiting of a massive planet. In the case of our own solar system, this motion would be predominantly due to Jupiter. Since the Sun outweighs Jupiter by more than a factor of 1000, the Suns back and forth motion is tiny: about the speed of a good runner. It is amazing that this can be precisely measured. To date well over 200 planets have been discovered using this technique, and in a few recent cases, as precision keeps improving, even multiple planets around a star can be extracted from the data.

Scale diagram of planet/star ratio for the WASP Planets. (Credit: SuperWASP project)

As powerful as this technique is, there are unavoidable limitations. We cannot measure the sizes and masses of planets accurately (because we do not know how tilted to our line-of-sight the orbits are). The technique is intrinsically biased toward the most massive planets in close orbits. Thats why so many of the planets are much more massive than Jupiter yet orbit their stars at distances closer than Mercury in our solar system.

A complementary technique is now showing its worth: planet transits. If we are lucky enough to be almost directly in the plane of the orbit of a planet, we can spot the transit of that planet against its sun. This happens here too. Venus transits the Sun as viewed from the Earth every now and then. On other stars we cannot image the planet as it crosses a stellar disk, but it is possible to measure changes in brightness with such precision that dips in the brightness of a star can be used to infer that a planet is crossing. This allows the actual size of the planet to be inferred from the dip in the stars light (assuming we can infer the radius of the star from its spectral type).

Given that we need to rely on chance alignments of the orbital planes and transits that briefly happen every few months or years for those lucky alignments, you would not think that this looks too promising. It requires monitoring hundreds of thousands of stars all at once to stand a chance of catching one of those rare events.

Amazingly this can be done, and this technique has now resulted in 46 planet detections from ground-based observatories. The latest announcement comes from astronomers at the University of California, Santa Barbara, who have found ten new planets in a project called SuperWASP, for Wide Area Search for Planets. Planets found range from half the size of Jupiter to eight times as large.

This technique is being taken into space. The NASA Kepler Mission, scheduled for launch in February 2009, is designed to find planets 30 to 600 times less massive than Jupiter. As NASA states: To detect an Earth-size planet, the photometer must be able to sense a drop in brightness of only 1/100 of a percent. This is akin to sensing the drop in brightness of a cars headlight when a fruitfly moves in front of it!

The Kepler Mission, a NASA Discovery mission, is specifically designed to find Earth-like planets in the habitable zone, which encompasses the distances from a star where liquid water can exist on a planets surface. It is amazing how quickly we are obtaining actual observations zeroing in on that most profound of questions: Are there other civilizations out there in the Universe?

Press Release from University of California, Santa Barbara.

NASA Kepler Mission home page.