The ESA Gaia mission uses two telescopes to create the most ambitious survey of the Galaxy. The angle between them must be known with exquisite precision and accuracy. An interferometer: the Basic Angle Monitoring system measures its variations. High quality data have been retrieved and analysed for more than a year. A summary of the in-orbit performance and some early results are presented.

ELSA (European Leadership in Space Astrometry) is a four-year, EU-funded ResearchTraining Network ending in September 2010. It has employed 10 postgraduate students and 5postdocs for 2–3 years and financed a number of workshops, training events and topicalmeetings, culminating in the present symposium. The primary goal of ELSA is to train youngscientists in the context of the Gaia project while contributing to the scientificpreparations for the mission. The organization, aims and history of ELSA are outlined.

We outline the basic principles of scanning space astrometry, such as represented byHipparcos, Gaia, and some other astrometric satellites planned or proposed. We explain theneed for large-angle measurements, why these are essentially one-dimensional, how it ispossible to determine absolute parallaxes, and why a Hipparcos-type scanning law isfavourable. We discuss the choice of the basic angle between the two viewing directions,the principle of self-calibration, and why the resulting numerical problem must bedifficult to solve.

Accurate characterization of the errors in the global astrometric solution for Gaia isessential for making optimal use of the catalogue data. We investigate the structure ofthe covariance between the estimated astrometric parameters by studying the properties ofthe astrometric least squares solution. We find that astrometric errors can be separatedin a star and an attitude part, due to the estimation of the star and attitude parametersrespectively. Hence the covariances can be separated in a star, an attitude and a crossterm. This is demonstrated using our scalable simulation tool AGISLab, where thecovariances are estimated statistically using Monte Carlo techniques.

We report on recent new developments around the Astrometric Global Iterative Solutionsystem. This includes the availability of an efficient Conjugate Gradient solver and theGeneric Astrometric Calibration scheme that had been proposed a while ago. The number ofprimary stars to be included in the core solution is now believed to be significantlyhigher than the 100 Million that served as baseline until now. Cloud computing servicesare being studied as a possible cost-effective alternative to running AGIS on dedicatedcomputing hardware at ESAC during the operational phase.

Gaia development is in full speed aiming to the launch in December 2011. While the entities formally responsible to make Gaia happen are very focused to be ready in time, it is necessary to explore the readiness of the wider scientific community to exploit Gaia data in the future. The GST sees a role for itself in building and enabling the community to use the Gaia catalogues when they become available. This paper gives the background for the current activities the GST is taking to promote the community to start preparations for the exploitation of Gaia data.

ELSA (European Leadership in Space Astrometry) is an EU-funded research project 2006–2010, contributing to the scientific preparations for the Gaia mission while training young researchers in space astrometry and related subjects. Nine postgraduate (PhD) students and five postdocs have been recruited to the network. Their research focuses on the principles of global astrometric, photometric, and spectroscopic measurements from space, instrument modelling and calibration, and numerical analysis tools and data processing methods relevant for Gaia.

Gaia is an ESA space astrometry mission due for launch in 2011–12. We describe part of the work carried out in the Gaia Data Processing and Analysis Consortium, namely the Astrometric Global Iterative Solution (AGIS) currently being implemented at the European Space Astronomy Center (ESAC) in Spain and largely based on algorithms developed at Lund Observatory. Some provisional results based on simulated observations of one million stars are presented, demonstrating convergence at microarcsec level independent of initial conditions.

The ESA space astrometry mission Gaia will measure the positions, parallaxes and proper motions of the 1 billion brightest stars on the sky. Expected accuracies are in the 7–25 μas range down to 15 mag and sub-mas accuracies at the faint limit (20 mag). The astrometric data are complemented by low-resolution spectrophotometric data in the 330–1000 nm wavelength range and, for the brighter stars, radial velocity measurements. The scientific case covers an extremely wide range of topics in galactic and stellar astrophysics, solar system and exoplanet science, as well as the establishment of a very accurate, dense and faint optical reference frame. With a planned launch around 2012 and an (extended) operational lifetime of 6 years, final results are expected around 2021. We give a brief overview of the science goals of Gaia, the overall project organisation, expected performance, and some key technical features and challenges.

Late-type giants (i.e., stars on the red and asymptotic giant branches, RGB/AGB, respectively) are dominant contributors to the overall spectral appearance of intermediate age and old stellar populations, especially in the red/near-infrared part of the spectrum. Being intrinsically bright, they are well suited for probing distant/obscured populations, especially those that can not be studied with their fainter members, like main sequence turn-off stars or subgiants. Late-type giants and supergiants will be the only stellar types accessible in intermediate age and old populations beyond the distances of several Mpc with the future 30-50 m class extremely large telescopes (Olsen et al. 2003). Indeed, proper understanding of their observable properties by means of theoretical models is of key importance for studying the evolution of stellar populations and their host galaxies.

From the combination of VLBI phase-referenced observations and Hipparcos satellite data, we have found evidence of a low-mass object orbiting the late-type star AB Doradus. The mass of the new object is near the hydrogen burning limit and will constitute a precise point for calibrating the low end of the main sequence. This represents the first detection of a low-mass stellar companion using the VLBI technique, which could become an important tool in future searches for planets and brown dwarfs orbiting other stars.

The Hipparcos reference frame is defined by the positions and proper motions of ~105 stars given in the Hipparcos Catalogue (ESA, 1997). It is the optical counterpart of the International Celestial Reference Frame defined by the radio positions of extragalactic sources. The linking of the Hipparcos frame to the extragalactic frame is described by Kovalevsky et al. (1997) and Kovalevsky (1997). In the present review we consider only the intrinsic properties of the Hipparcos reference frame.

A complete description of the contents of the Hipparcos Catalogue is contained in Volume 1 of ESA (1997). Recent overviews have been given by Perryman et al. (1997) and Perryman (1997). The principal astrometric data provided are the barycentric coordinates (α, δ) of the stars, the trigonometric parallaxes (π) and the proper motion components (μα* = μα cos δ, μδ). These parameters are given in the main catalogue for 117955 entries along with estimated standard errors and correlation coefficients, all data being referred to the common reference epoch To = J1991.25 on the Terrestrial Time (TT) scale. Typical ranges of the astrometric standard errors and some other characteristics of the catalogue are shown in Table 1.

The ESA Hipparcos astrometry satellite did not contain an imaging detector (such as a CCD) but used a modulating grid of alternating transparent and opaque bands to modulate the light from a given object. The Fourier coefficients of the modulated detector signal were the basis for the astrometric and photometric processing leading to the Hipparcos Catalogue. These coefficients are included (for a subset of the stars) in the CD-ROM version of the Hipparcos Catalogue as the ‘Hipparcos Transit Data’. In this work, the Transit Data are used in a completely different and novel application: the modulation components from a single scan are considered to be equivalent to the interference fringes from an interferometer oriented along the scan direction and having baselines of 0, ≃ 10 and ≃ 20 cm and a wavelength of 550 nm. Using all the scans collected over the mission lifetime, it is possible to use the complex visibilities (containing bothamplitude and phase information) calculated from the Transit Data to createaperture synthesis images of various type of objects such as visual or orbiting binary systems. More information, including sample images, can be found in: Quist C.F., Lindegren L., Söderhjelm S., 1997, Using HipparcosTransit Data for Aperture Synthesis Imaging, Hipparcos Venice ‘97 Symposium, ESA SP-402 (in press).

Space astrometry now permits accurate determinations of stellar radial motion, without using spectroscopy. Although the feasibility of deducing astrometric radial velocities from geometric projection effects was realized already by Schlesinger (1917), only with Hipparcos has it become practical. Such a program has now been carried out for the moving clusters of Ursa Major, Hyades, and Coma Berenices. Realized inaccuracies reach about 300 m/s (Dravins et al. 1997).

Discrepancies between astrometric and spectroscopic radial velocities reveal effects (other than stellar motion) that affect wavelength positions of spectral lines. Such are caused by stellar surface convection, and by gravitational redshifts. A parallel program (Gullberg & Dravins 1997) is analyzing high-precision spectroscopic radial velocities for different spectral lines in these stars, using the ELODIE radial-velocity instrument atHaute-Provence.

The Hipparcos mission demonstrated the efficiency of space astrometry (in terms of number of objects, accuracy, and uniformity of results) and the fact that a relatively small instrument can have a very large scientific potential in the area of astrometry. However, Hipparcos could probe less than 0.1 per cent of the volume of the Galaxy by direct distance measurements. Using a larger instrument and more efficient detectors, it is now technically feasible to increase the efficiency of a space astrometry mission by several orders of magnitude, thus encompassing a large part of the Galaxy within its horizon for accurate determination of parallaxes and transverse velocities. Such a mission will have immediate and profound impact in the areas of the physics and evolution of individual stars and of the Galaxy as a whole.

This contribution is intended as a ‘rough guide’ to the Hipparcos Catalogue for the non-expert user. We shall focus here on the interpretation of the measured parallaxes in terms of distances and luminosities of stars, and on the use of the covariance matrix of the errors on the astrometric parameters. Issues not addressed here are the level at which one may expect systematic errors to occur in the Hipparcos Catalogue, and the correlation of astrometric parameters over a given region of the sky. For more details on the latter two subjects we refer to the documentation of the Hipparcos Catalogue (ESA, 1997) and to Brown et al. (1997). We note that at present the Hipparcos Catalogue contains no identified systematic errors.

We present a concept for a scanning interferometer for global (wide-angle) astrometry from space. The GAIA concept has been proposed for the European Space Agency's long-term scientific programme. It consists of three Fizeau-type interferometers with 2.5 m baselines, set at large and fixed angles to each other. Complete utilization of the instrument's resolution and sensitivity requires a new type of photon-counting detector, combining very high spatial and temporal resolution. An array of superconducting tunnel junctions (STJ) may ultimately provide this capability. Pending this development we describe a focal-plane configuration for GAIA using existing technology in the form of a modulating grid and CCD detectors. We estimate that 50 million stars brighter than V = 15.5 could be observed on the 10 to 20 microarcsec accuracy level. In addition, high-precision multi-colour, multi-epoch photometry is obtained for all objects.

Instrumentation and observing methods are developed for optical high-speed astrophysics, aiming at exploring milli-, micro-, and nanosecond variability. Such rapid fluctuations can be expected from instabilities in accretion flows, and in the fine structure of photon emission. For the optical, we have constructed a dedicated instrument, whose first version was tested on La Palma to study atmospheric scintillation on very short timescales. A second version is now under development, using photon-counting avalanche photodiodes as detectors.

The foreseen complexity of the the ESA Hipparcos mission led to the establishment of two independent scientific consortia (FAST and NDAC) for the parallel processing of all main mission data. The validation of Hipparcos data through internal and external checks and by interconsortia comparisons is outlined. Examples are given based on preliminary solutions. The various checks indicate that the overall accuracy and reliability of the Hipparcos Catalogue will be extremely high.

A CCD modulation detector for a possible Hipparcos-2 mission is described. It is at least 1000 times more light efficient than the image dissector tube system used in the current Hipparcos mission. A beam combiner telescope with 29 cm aperture diameter could measure all 15 million stars brighter than 14th magnitude, providing accurate multi-colour photometry in addition to the astrometric parameters. Using two telescopes with different basic angles will further improve the rigidity of the resulting system of positions and proper motions. An accuracy of 0.5 milli-arcsec up to 10th magnitude is expected for a 2.5 year mission. The modulating grid is imaged on a CCD performing a ‘cophased integration’ of nine separate images of the whole field of view, each image representing a different part of the modulation period.