The Economist
December 23, 2006
U.S. Edition

SECTION: SCIENCE & TECHNOLOGY
HIGHLIGHT: Building bigger telescopes

Ever larger telescopes are planned to study the heavens in ever more detail, but with a twist

SIZE matters, at least in astronomy. Large telescopes are able to detect fainter objects than their smaller counterparts can because they gather more light. They also produce crisper images because they can resolve smaller details of the objects they are pointed at. Although space-based telescopes avoid distortion caused by the atmosphere and can thus discern finer details than their Earthbound equivalents, they are doomed to have small mirrors because rockets can carry only so much weight into space. So, if the true nature of the universe--including the composition of dark matter and dark energy, the two great known unknowns of cosmology--is to be elucidated, bigger land-based equipment is needed.

Nearer to home, big telescopes would also be able to look for Earth-sized planets around stars other than the sun. Today's machines have identified more than 200 "extrasolar" planets, but these are all rather bigger than Earth. Not only might a suitable large telescope locate Earth's cousins, but it should also be able to study their atmospheres. That, in turn, would give clues as to whether they harboured life, since a chemically unstable atmosphere (such as one rich in a reactive gas like oxygen) is evidence suggesting biochemical activity.

The problem is that big telescopes are hard to make. The crucial component, the mirror that gathers the light, is more liable to distortion, the bigger it gets. The modern fashion, therefore, is to make telescope mirrors smaller, and then get them to collaborate.

One way of doing this is to build a series of independent telescopes and point them all in the same direction. Such an arrangement provides a resolution (though not a light-gathering power) equivalent to a single mirror with a diameter equal to the distance between the two telescopes that are farthest apart in the array.

This technique has been applied for a while to radio telescopes. The problem with extending it to shorter wavelengths, such as visible light, is that the accuracy with which the instruments have to be pointed is related to the wavelength they are looking in. Radio astronomers, whose wavelengths are measured in metres, can afford to be sloppy. Optical astronomers, whose wavelengths are measured in billionths of a metre, cannot.

The compromise today is to look at microwaves--the part of the spectrum that lies between radio waves and light. Astronomers in America, Europe and Japan are collaborating on the biggest microwave array to be built so far, the Atacama Large Millimetre Array now under construction in Chile. It will have up to 64 "mirrors" (actually dish-like antennae that are large versions of the sort of thing used to receive satellite television). The advantage of microwaves, other than the ease of building large telescopic arrays to look at them, is that they can pass through the interstellar dust that obscures much of the universe, and thus illuminate processes invisible to optical-wavelength astronomers.

Though optical astronomers cannot yet manage this array-building trick, they, too, have worked out how to benefit from making small mirrors collaborate. The difference is that they put the small mirrors together to form a big one inside a single instrument.

Doing that allows them to build very big mirrors indeed. The largest single-element optical telescope at the moment--confusingly called the Large Binocular Telescope, but each of its two mirrors is made as a single piece--has mirrors 8.4 metres across. The two Keck telescopes on Hawaii and the South African Large Telescope, which use mosaics of hexagonal sub-mirrors, have systems up to 11 metres across.

Even these, however, are minnows compared with what is being planned. America's National Science Foundation is now evaluating two competing designs: the Giant Magellan Telescope, some 24 metres across, and the self-explanatory Thirty Metre Telescope. A European Extremely Large Telescope is also on the drawing board. Late last month a draft design for this, with a mirror size of between 30 metres and 60 metres, was unveiled. It is based in part on what was known as the Overwhelmingly Large Telescope, an earlier venture that was abandoned after it turned out to have overwhelmingly large costs, as well.

All these telescopes will be built atop mountains in dry areas, to get as close to outer space as possible, and thus minimise the layer of air and water vapour between the sensors and the stars. Even then, achieving high resolution requires "adaptive optics"--electronic wizardry designed to undo the distorting effects of the remaining atmosphere above the instrument in question. The idea is to monitor a reference star and, by subtly adjusting the shape of the mirror, to keep this star in focus no matter what the weather. One innovation to be tested by the European Extremely Large Telescope would be to create an artificial reference star by firing a laser into the night sky. That would be a boon to those wishing to study parts of the sky that are normally void of such objects.

The new generation of part-work telescopes would operate in collaboration with space-based astronomy, of course. Just a few weeks ago NASA, America's space agency, announced that it would upgrade the Hubble space telescope. It also has plans for a new device, dubbed the James Webb space telescope and due to be launched in 2013. But modern ground-based telescopes can complement such observatories, often achieving more and costing less. Mountaintop astronomy is entering a new golden era.