Prospects for space stellar astrometry

Astrometry is the major astronomical technique to measure distances, masses and motions of stars. Dividing astrometric techniques into five types according to the size of the field in which a single instrument can produce measurements, the present achievements of the Earth-based astrometry are described. The astrometric activities such as measurements of star diameters, double star relative positioning or stellar parallaxes, search for invisible companions, photographic plate reduction, visual and photoelectric meridian and astrolable astrometry are reviewed. Then, the methods used to construct a quasi-inertial celestial reference frame and to materialize it by a fundamental catalogue are presented and discussed. A much better definition of an absolute reference frame is made possible by VLBI, but the problem of extending it to stellar positions is not yet satisfactorily resolved.The limitations of the ground based astrometry are: the atmospheric turbulence and refraction, Earth's motions and the impossibility to view the entire sky with a single instrument. These limitations are discussed and it is shown how astrometry from space can overcome them. A priori, a gain of two orders of magnitudes in accuracy for all types of astrometry is expected, but at this new level of precision, new effects and limitations will appear, as already shown in the studies of the approved programs.Then, the ESA astrometric satellite HIPPARCOS presently under development is presented. The satellite and the payload are described as well as the observing procedures. Several limitations, specific to space borne instrumentation and to the milliarc second accuracy expected have been identified. However the main limitation in precision remains the photon noise. The data reduction methods are sketched. The data downlinked at a rate of 20 kilobits per second have to be used with an equal weight all over the 21/2 years of observation. They are expected to yield a mean accuracy of 2 milliarc seconds in position and parallax and 2 m.a.s. per year in proper motion for most of the 100000 stars of the program (M _b < 9). Stars to be observed by HIPPARCOS have to be carefully selected. The main fields in which the results of HIPPARCOS will be used are listed from the proposals made by the scientific community. The task of constructing the HIPPARCOS input catalogue from these proposals is presented.Another feature of the ESA astrometric satellite is the use of the HIPPARCOS star-mapper as a photometric and position survey of the sky. This experiment, called TYCHO, should give at least 400000 star positions with accuracies of the order of 0.03 to 0.15 depending upon the magnitudes. Two colour instantaneous magnitudes should also be obtained to 0.1–0.4 mag. precision.Several Space-Telescope on-board instruments are also capable to make small field astrometric observations. Accurate imaging is possible with the Wide Field and the Faint Object cameras. Lunar occultations will be performed with the High Speed photometer. But the main astrometric mode of the Space Telescope will be the use of the Fine Guidance Sensors to measure the relative positions of stars to ±0.002. It is described together with its main scientific applications.The establishment of an absolute reference frame is subsequently discussed. Plans using simultaneously VLBI, HIPPARCOS, and Space Telescope observations are described. They consist in linking the HIPPARCOS stellar system to quasars via radio-stars or stars in the vicinity of optical quasars.Finally, several space astrometry proposals are described: long focus space astrometry and two versions of space interferometry.