Space missions designed to detect habitable Earth-like planets of other solar systems, and costing about £320 million, may not work, a scientific meeting in London has been told.
Suzanne Aigran, of the Institute of Astronomy at the University of Cambridge, told this month's specialist meeting of the Royal Astronomical Society that the problem could affect Nasa's Kepler mission and the European Space Agency's proposed Eddington spacecraft, which has a E181 million (£126 million) budget. They are both designed to spot much smaller planets than the Jupiter-sized ones already detected in other solar systems.
Dr Aigran said that the ability of these spacecraft to detect planets would be limited by the natural variation of stars' light output, which would make it hard to tell whether the satellite was observing a planet passing in front of the star or a natural fluctuation.
She said that data-filtering methods might bring the limit down to allow planets twice the size of the Earth to be detected, but not Earth-like ones. "Most of the data we have comes from the Sun, which is not a very active star compared with others. When we look in detail at how it varies, it may be possible to filter out some of the variation."
Keith Horne, of the University of St Andrews, who is involved in Eddington, said: "When a planet the size of Earth passes in front of a star like the Sun, it reduces its brightness by one part in 10,000 for 13 hours. We do not know enough yet to tell what proportion of stars are too variable on this timescale for Earths to be detected."
Professor Horne said that one planet had already been detected by this method, using the Hubble space telescope. A French mission called Corot, to be launched in 2004, will carry out a search of 1,000 stars that should reveal more about the variability problem. It will look mainly for "hot Earths", near to stars, not ones at Earth-like distances.
Professor Horne said: "All the groups are working on the problem. We are gambling that there are enough stars for which it will not be a difficulty."
But William Borucki of Nasa, which has allocated $299 million (£208 million) for Kepler, denied at the London meeting that variability was a significant problem.