Modeler's view of our Milky Way Galaxy -- at the center -- in a lumpy halo with many satellites. Photo: NASA. Simulation: Stefan Gottlober/AIP

Clues to how galaxies formed in the early Universe lie right under our nose -- in our own Galaxy. The Galaxy formed by the accretion of infalling satellite galaxies, many astronomers think. Theoretical models of the formation of galaxies predict such a scenario.

But not all astronomers are convinced yet and the topic is still controversial.

Now researchers from eleven countries have launched an ambitious project to reconstruct our Galaxy's history by gathering key components of motion and chemical compositions for its apparently brightest 50 million stars.

RAVE (RAdial Velocity Experiment) is an all-sky stellar spectroscopy survey just started on the 1.2-m UK Schmidt telescope in eastern Australia.

Projects such as Hipparcos and Tycho have accurately measured the positions and proper motions - movement across the sky - of more than 2.5 million stars.

But to get a complete picture of stellar motions, and thus to enable astronomers to reconstruct the structure and formation history of our Galaxy, they also need radial velocities - the movement of stars towards or away from the observer. And before RAVE began only about 20,000 stellar radial velocities were in the archives.

RAVE will be able to achieve velocities accurate to within 2 kms-1 - about 1% of the speed at which stars typically move in the Galaxy.

"With this accuracy and this number of radial velocities we will be able to identify dozens, perhaps hundreds, of streams of stars in the solar vicinity. The streams represent debris from disrupted old satellite galaxies now engulfed by our Galaxy," said Professor Matthias Steinmetz, Director at the Astrophysical Institute Potsdam and leader of the RAVE science team.

Even after plunging into our Galaxy, the stars of a satellite galaxy continue to move as a coherent group, and can be identified by their common velocity even after billions of years. However, only a very few of those disrupted satellites have been identified to date.

RAVE will also gather the chemical compositions of stars. These should help show which widely separated stars were formed at a common site. They should also determine whether these stars have been formed before or after the satellite galaxy of which they were a part broke up.

"RAVE will help us decide between competing models for the formation of the various structures of the Galaxy, such as the central bulge of stars and the so-called 'thick disk'," said Steinmetz.

"For a survey such as this, field of view is more important than aperture. The UK Schmidt telescope is a perfect tool for this work," said Professor Brian Boyle, Director of the Anglo-Australian Observatory, which operates the telescope. The field of view of the UK Schmidt telescope covers an area more than 100 times larger than that of the Moon.

RAVE's initial pilot phase is being carried out with the 6dF (six-degree field) instrument on the UK Schmidt. Designed and built by the Anglo-Australian Observatory, the 6dF instrument is a 'pick and place' robot that positions 150 fibres on the telescope's focal plane.

Using 6dF, astronomers can collect up to 600 stellar spectra per night. And by 2005 they plan to have 100,000 - five times as many as have been measured since Hermann Carl Vogel started such work at the Astrophysical Observatory Potsdam in 1888.

In 2006 the pace of data collection will pick up even further, when 6dF instrument is replaced by a radical new instrument from the AAO - UKidna, with 2250 fibres mounted on independently movable spines.

"With UKidna we'll be taking up to 22,000 spectra on a clear night," said Boyle.

"Then we'll be able to push beyond our local Galactic neighbourhood, out into the furthest corners of the Milky Way," said Professor Rosie Wyse of Johns-Hopkins University in Baltimore.

As well as uncovering the history of our Galaxy, RAVE will establish a huge database of stellar spectra - by far the largest to date.

"This will be a vast resource for studies of the properties and evolution of stars," said Professor Ulisse Munari of the Padova Observatory in Asagio.

With its large database of stellar spectra RAVE will also provide an ideal training set for the design of future space missions such as the European Space Agency's cornerstone mission GAIA, which will attempt to measure positions and velocities of up to a billion stars in the Milky Way.

The Anglo-Australian Observatory operates the Anglo-Australian and UK Schmidt telescopes on behalf of the astronomical communities of the UK and Australia. The Observatory is funded by the UK Government, through the Particle Physics and Astronomy Research Council, and the Australian Government, through the Department of Education, Science and Training.