Posted on Thursday 29 April 2004

Most techniques for detecting planets outside of our solar system can only spot very massive worlds which are comparable in size to the planet Jupiter but they are not much good at detecting small and close-in planets similar to our own.

There are, however, some detection techniques that are considerably more sensitive, one proposed method I mentioned here a few months ago is called nulling interferometry (here as well). It works by the cancellation of the lightwaves being emitted by a star and this allows us to directly observe any planets that may be orbiting it. The need to blank out this starlight becomes readily apparent when we consider that the light coming from a Sun-like star is billions of times brighter than any reflected light that might be coming from Earth-like planets orbiting it. This direct imaging technique is very promising and it offers a number of benefits which includes the ability to measure the chemical composition of the planet's atmosphere and to search for evidence of life. It does, however, have some practical problems, a space-based telescope using this technique would be extraordinarily difficult to build and it is likely that any implementation if it will be a few decades away at the earliest.

By contrast, a new relatively straightforward (though indirect) technique which uses conventional telescopes has proven itself to be an extremely sensitive way to detect extrasolar planets. It's called gravitational microlensing and it has just been used to discover the most distant extrasolar planet ever found. The new planet is three times further away than any previously discovered and the microlensing technique is so sensitive that it will be able to, in principle, detect planets over very large distances that are as small and as light as our own planet.

The gravitational microlensing technique is able to detect a planet through the effect of the planet's gravitational field on the light we see from a more distant background star. The simple effect of gravitational lensing of one star by another has now been detected more than thousand times. When two stars are nearly perfectly aligned as seen from Earth, the gravitational field of the foreground star acts as a lens to magnify the background star. The magnified image is very small, and can only be resolved with a telescope that produces images 1000 times sharper than the Hubble Space Telescope. Hence the term "microlensing". Even so, the effect can be observed by the increased brightness of the magnified star. It is only a temporary effect. The orbital motion of the stars in the Galaxy will occasionally bring pairs of stars into alignment as seen from Earth. Their continued motion will then bring them out of alignment again and the microlensing effect ceases.

Most of the one thousand microlensing events that have been observed have followed a brightening pattern or "light curve" characteristic of a lens composed of a single star. However, some have followed a very different light curve caused by a lens composed of a double star. The planetary microlensing event observed by the OGLE and MOA Collaborations resembles the single lens light curve for most of its four-month duration. But for a period of about a week, the gravitational field of the planet causes the light curve to resemble that of a double star lens.

The precise shape of the light curve reveals that the lighter mass of the double lens has only 0.4% of the mass of the heavier component, which implies that the lighter component must be a planet. Analysis of the light curve revealed that it was most likely a red dwarf star and a planet of about 1.5 times the mass of Jupiter at a separation of about 3 AU. (An Astronomical Unit, or AU, is the mean distance between the Earth and Sun.) It was located about 17,000 light years away toward the central part of the Galactic disk, in the constellation Sagittarius....

The gravitational microlensing technique is, in one sense, the easiest way to detect extrasolar planets because the signal of the planet can be quite large. For the present event, the maximum effect due to the planet was more than a factor of two increase in the apparent brightness of the background source star. In contrast, the radial velocity technique requires the detection of Doppler shifts of one part in ten million. Moreover, the critical data for the present event were obtained over a period of only a couple of months. By contrast, many years of data by the radial velocity technique would be required to detect a similar planet.

--- Distant Planet Found by Gravitational Microlensing, April 2004

As you might expect, microlensing has some disadvantages as well.

Being indirect, it is not possible to collect any information about the planet other than inferences about its mass and orbital characteristics. It does not make it possible to directly image the planet or to study its atmospheric composition. Also, because it relies upon the chance alignment of two stars - and this is a rare occurence - it is more likely to work in regions of space that are densely populated with stars, for example, towards the centre of our galaxy.

But the greatest disadvantage of all has to do with the unpredictable and short-lived nature of these alignments. An alignment may only last for a week or even just a few hours and once it has passed the planet returns to being undetectable again for an indeterminate period (possibly forever). The unrepeatable nature of these observations therefore makes it impossible for later scientists to confirm them and this leaves the validity of these observations open to dispute. So, in order to avoid controversy, teams of independent (and normally competing) planet finders have found the need to work together and to corroborate each others findings. In this case, data was shared between the MOA and OGLE teams as well as with several other experts.

Gravitational microlensing, then, is mainly a useful technique for randomly sampling planetary systems and for determining the relative density of these systems in different regions of space. Once the planets have been found, however, other techniques must be brought to bear in order to study them further.

The second panel shows the configuration for a planet detection. One of the light rays that is bent by the lens star's gravity comes close enough to the planet that it feels the gravity of the planet, too. This causes additional distortion of the images, and in some cases, additional images can be created which result in dramatic changes in brightness (the spikes seen in the magnification curve).

See also:


New Planet, Magnified
FIRST DETECTION of an EXTRASOLAR PLANET with MICROLENSING
MOA Page
OGLE Page
Planet Detection with Microlensing Animation (MPEG 0.5 MB)