In the course of its three-year lifetime the European Space Agency's Hipparcos satellite performed some 13 million scans across the stellar objects on its pre-defined observing list. For the vast majority of targets the subsequent data reductions succeeded in determining, accurately and without ambiguity, a small set of astrometric and photometric parameters which contain practically all the useful information on the objects. In the simplest case of a non-variable single star these data would be the position of the star at a certain epoch, the trigonometric parallax, the two proper motion components, a mean magnitude, and various statistics such as the mean errors of all the parameters. For thousands of variable, double and multiple stars, the necessary additional parameters were determined without too much problem. All of these results are readily available in the Hipparcos and Tycho Catalogues (ESA 1997). For most objects and for most applications of the results, there is no need to probe deeper into the intricacies of the Hipparcos data acquisition and processing.
It was inevitable, however, that the observations of some (very small) fraction of the objects could not easily be interpreted in terms of standard models, and that inadequate, or even erroneous, object models were applied in some cases. Potentially very valuable information could have been lost for these objects if only summary results, based on inadequate models, were published. This situation has been avoided by including a good selection of intermediate results in the published catalogue. Being only partially reduced these results are less dependent on specific model assumptions, but still relatively simple to use thanks to a careful calibration into the photometric and astrometric systems of the final catalogue. The published version of the Hipparcos and Tycho Catalogues includes six CD-ROMs, five of which contain mainly such calibrated, partially reduced data. The data fall in three distinct categories:
While the TD thus provide information that is more detailed and on a more fundamental level (in the sense of being closer to the satellite "raw" data) than either the Epoch Photometry or the Intermediate Astrometric Data, two important restrictions should be noted: (1) TD are only available for about a third of the targets, or some 38000 objects; (2) TD are based solely on the results from one of the data reduction consortia, the Northern Data Analysis Consortium (NDAC; Lindegren et al. 1992). By contrast, the Epoch Photometry and the Intermediate Astrometric Data are available for virtually all the 118204 catalogue entries, and results from both NDAC and the Fundamental Astrometry by Space Techniques Consortium (FAST; Kovalevsky et al. 1992) were normally used.
Detailed satellite operations and data reductions are described elsewhere (e.g., Perryman & Hassan 1989; Perryman et al. 1989; Perryman et al. 1992; Kovalevsky et al. 1992; Lindegren et al. 1992; Vols. 2-4 of ESA 1997; van Leeuwen 1997). It will however prove useful to review how the TD were measured in order to understand their precise meaning (and limitations), aiding in the full exploitation of the information. Thus, Sect. 2 describes the convolution of the stellar diffraction image with the modulation grid, and the resulting detector signal modeled by the TD. Section 3 is a detailed description of the contents and format of the published TD files. Two applications of the TD using publicly available software are demonstrated, viz. aperture synthesis imaging (Sect. 4) and model fitting (Sect. 5).
Copyright The European Southern Observatory (ESO)