Astronomers have just realised that news of a planet
orbiting a distant star came from ESA’s Hipparcos satellite eight
years ago, although no one noticed it until now. The first
observation, on 17 April 1991, was made long before Michel Mayor
and Didier Queloz of the Observatoire de Genève astounded the
world in 1995 with their discovery of a planet around the star 51
Pegasi. Since then the search for alien planets has become a highly
competitive theme in astronomy, and the present tally of stars known
to possess planets is 28.

No&oumll Robichon and Frédéric
Arenou of the Observatoire
Paris-Meudon have
re-examined sightings of the
star HD 209548, by the
Hipparcos star-mapping
mission during its life in orbit,
1989 to 1993. They find,
among 89 observations, that
the star’s light was slightly dimmed on five occasions by the shadow
of a large planet passing in front of it. The French astronomers
looked closely at the Hipparcos data when they heard that
colleagues in the USA, monitoring this star from the ground, saw it
diminish in intensity for two and a half hours, on two occasions in
September 1999.

“To be honest, we would never have found the planet in the data if
we had not known where and when to look for it,” Robichon
comments. “But the prospect of using satellite observations in this
way, for smarter detections of alien planets, is now very exciting.”

HD 209548 is a sun-like star in the Pegasus constellation. It is also
known as HIP 108859 in the Hipparcos Catalogue, and Hipparcos
fixed its distance at 153 light-years. Hints that it might possess
planets came from observations made from the ground over the past
few years, by the usual method of looking for slight shifts in the
wavelength of light. These can indicate wobbles in a star due to the
gravity of the planets circling around it.

Fresh observations by the Geneva group, using the Swiss 1.2-metre
telescope at the European Southern Observatory’s site at La Silla,
Chile, confirmed and refined the detection of wobbles in HD 209548.
They correspond with a large Jupiter-like planet so close to the star
that it orbits around it twice a week.

Only in about 10 per cent of such cases would astronomers expect
the planet to pass directly in front of the star, as seen from the
Earth. But when Michel Mayor gave out the details about HD
209548, David Charbonneau of Harvard University and Timothy
Brown of Boulder, Colorado, decided to check it with a 10-centimetre
telescope built by Brown especially for looking for transits of planets.
Their good luck prompted Robichon and Arenou in Paris to check
the Hipparcos data. Staffan Söderhjelm of Lund Observatory,
Sweden, had the same idea, and obtained similar results.

“Adding the early Hipparcos data for the transits gives us
observations spanning eight years,” Arenou notes. “We now know
the orbital period of the big planet of this star with amazing accuracy.
Our figure of 3.52474 days is twenty times more precise than one
could deduce from the star’s wobble alone in a few weeks.”

The chance now exists to check the Hipparcos data on other stars
with planets, looking for dimming by transits. More dramatic results
on alien planets should come from GAIA, a project under study by
ESA as a successor to Hipparcos. It will detect the wobbles of stars
directly, by very small changes in their motions, and should identify
about 30,000 stars with large planets. In a few thousand cases,
GAIA should also see the dimming effect of planets passing in front
of the stars.

The January 2000 issue of the journal Astronomy and Astrophysics
will carry a full report by No&oumll Robichon and Frédéric Arenou on the
Hipparcos results on HD 209548. Their results, and also those
obtained independently by Staffan Söderhjelm of Lund Observatory,
were summarized in an International Astronomical Union Circular (No.
7323, 1/12/99).

To facilitate the discovery of further planetary transits in the
Hipparcos data, ESA has provided a special-purpose web page for

The Hipparcos Epoch Photometry Search facility