Next: The LASCO Data Archive
Previous: The XTE Data Finder (XDF)
Up: Data Archives
Table of Contents - Index - PS reprint
A. J. Wicenec
Institute for Astronomy and Astrophysics, University
of Tübingen, FRG, e-mail: wicenec@astro.uni-tuebingen.de
The number of
observations per object used to produce the TYC is 130 in the mean
over the whole sky in the two bands BT
and VT, making the
TYC/TEPA one of the largest photometric databases in the world. For a
total of about 500,000 objects, the brightest and some selected
fainter objects, the single observations will be published in the so
called Tycho Epoch Photometry Annex B (TEPA/B). This
Annex will be available through the CDS in Strasbourg, France. A
subset thereof, the TEPA/A, containing the observations of about
36,000 selected objects will be published on a CD-ROM. The
measurements of the fainter 500,000 TYC stars are not considered to be
of a quality to be published, because the errors and the censoring of
the individual observations of the faint stars are too large. This
paper presents an access and research tool for the TYC and the TEPA to
be used locally with widget based GUIs or remotely by HTML-form based
WWW-pages.
Especially the TEPA/B data base needs an easy to use, powerful access tool providing data for selected objects. The Tycho Data Analysis Consortium (TDAC) is using an access tool, where the core is a collection of IDL routines controlled by an IDL-widget based GUI. The size of the TEPA/C (all observations for all TYC objects) sums up to about 50GB binary data and it thus resides on a single host within TDAC. To provide access for all the TDAC groups there is a WWW-layer above the IDL access tool which copies some of the functionalities to a small number of HTML-form pages. The interface is currently being updated with client side image maps, JavaScript verification and Java tools to limit the network traffic and the server load. If support by the host institute is approved this interface will be opened to the WWW without restrictions at the time when the Hipparcos and Tycho catalogues are published, i.e., in June 1997.
Since the Tycho instrument on-board the Hipparcos satellite was operated at a fixed sampling rate (integration time), observations for faint stars are censored, i.e., the magnitude distribution of the observations has a clear cut-off caused by the used SNR limit. For the de-censoring and variability studies the number, time, and background of all unsuccessful observations for a given star contains very important information. The two Annexes, TEPA/A and TEPA/B will contain all successful and unsuccessful observations for about 36,442 selected stars (TEPA/A) and the 481,553 brightest stars (TEPA/B) of the TYC. The mean number of observation per star in these catalogues is 191, giving a total of more than 6 million and 90 million observations for TEPA/A and TEPA/B respectively. Access to this photometric data base should be easy to use, yet powerful and flexible.
The TYC is a quite normal star catalogue very similar to the GSC in numbering and sorting. Stars common to TYC and GSC version 1.2 do have the same primary (region) and secondary (running) numbers in both catalogues. Yet the TYC contains a third identification number in order to keep double stars resolved by Tycho under the same GSC region and running numbers. The identification numbers in TYC are called TYC1, TYC2 and TYC3, where TYC3 is just 1 for most of the stars. Because of the similar numbering and sorting scheme any catalogue browser capable of browsing through GSC should be able to access the TYC without major changes. On the other hand the TEPA catalogues are much different and they need a special tool to provide access and working capabilities such as time series analysis (variables) or selection of single observations. The TEPA/A will be published on a CD-ROM together with basic access software, i.e., it will be possible to retrieve all observations for one of the TYC-stars contained on the CD-ROM. This software might also be used to retrieve data from the TEPA/B which will only be published through astronomical data centres (CDS, Strasbourg).
The TEPA catalogues contain two different kinds of records: star
header records and observation records. Each star header record is
followed by a number of observation records. The star header records
contain some fundamental data extracted from the TYC, such as the star
numbers, and the magnitudes, but no coordinates. The contents of the
star headers is merely a compromise between necessary and useful
contents and the need to keep the record length at the same length as
the observation records are. The most important field of a star header
is the number of observation records following the star header. The
length of the observation records is adjusted to contain all relevant
data for the single observations. Since the TEPA catalogues are
intended to be used together with the TYC the user will have access to all published data
of a particular star. The TYC and TEPA/A CD-ROMs and the TEPA/B when
ordered from CDS will be accompanied by some index files giving a two
level indexing of the complete catalogue and its Annex. These index
files might be used by a quite simple program to get very fast access
to the data of a particular star in TYC and TEPA if one knows the
TYC-number (where TYC1 and TYC2 are identical to the GSC number for
that star). A more user friendly access should also give the
possibility to retrieve the catalogues via coordinates and to
produce something like maps from TYC and light curves from TEPA
data. Such possibilities are provided by
The core routines of 
are written in IDL; they
use common structures for data exchange and keywords for customising
their behaviour. There are catalogue mapping routines which are
capable of producing maps, catalogue overlay maps, and image
overlay maps. There are a number of selection and coordinate
conversion routines as well as an interface to DSS-fits files as
produced by the ESO on-line DSS. The routines are virtually
independent of the underlying catalogue due to the usage of
interface routines between the catalogue input and the internal data
structures of 
. All the routines are also usable
as stand-alone routines and any new routine might use a common
initiallisation routine yielding access to the common structures. The
access to catalogues is not part of the core routines, but belongs to
the catalogue interface. Thus 
is able to use every
catalogue if the interface routine fills up the common data
structure. Due to the flexibility of IDL, interface routines may be
written in C, Perl and/or FORTRAN. The catalogue interfaces are
responsible for the correct contents of the common structures. Since
every catalogue has its special contents some of the core routines
are also able to provide access to the complete catalogue information
for every object on a map produced from the selected catalogue. The
main advantage of using IDL as the programming language of
is the very good portability of IDL-code. This is
even enhanced by some special environment files used by
describing the path- and file-names of the
available catalogues. The core routines only use string variables for
accessing files on the local file-system and are thus independent from
file name conventions of the operating system. New catalogues may be easily
added to 
through some preparation tools providing
the interface routines.
The main user interfaces to the core routines are IDL-widget-based GUIs. They build an easy-to-use layer above the core routines and a offer most of their capabilities by mouse interaction. There are two main GUIs: one for the TYC and other star catalogues and another for the TEPA catalogues. Both are usable as stand-alone interfaces and as an integrated tool. Thus it is possible, for instance, to create a Bright Star Catalogue (BSC) map with an overlay from TYC. By clicking on a star on the map, the TEPA observations for this star will be loaded and the TEPA window will gain control. A click on a button on the latter will pop up an additional window which allows some interactive period investigation by means of a periodogram routine, a minimum entropy routine, and, of course, some plotting routines to visualize the results.
The other user-interface to the core routines is WWW-based and runs on every browser that has form and client-side image map capability. There are some simple HTML-forms which ask for coordinate or TYC number input plus some auxiliary input, such as the diameter of the map to be produced. The <SUBMIT> button will send this query to the HTTP-server. The underlying CGI mechanism produces an IDL routine from the query string and runs IDL with this routine as a startup file. The output to the HTTP-server and thus the user is of NPH-HTML type importing a GIF-image of the requested part of the sky. The HTML code contains data to describe the image as a client-side image map, merely the TYC-numbers of the stars on this image. A click onto the map will result in a query to another CGI-interface producing a list of all transits of the star closest to the click position. This interface is currently being updated to a functionality comparable to the IDL widget GUI.
This work was supported by the DARA under grants No. 01-OO-85029. I am grateful to the Tycho Data Analysis Consortium and the Tycho Data Analysis Working Group in Tübingen.
Perryman, M. A. C. P., et al. 1989, ESA-SP 1111 Vol. I-III
Next: The LASCO Data Archive
Previous: The XTE Data Finder (XDF)
Up: Data Archives
Table of Contents - Index - PS reprint