Starry Night will automatically read Hubble star data off of the CD-ROM (if it is mounted when Starry Night is opened). For better performance you can move as many Hubble data files as you wish from the CD-ROM into the Starry Night data folder on your hard disk. Any files not moved to your hard disk will still be read from the CD-ROM.
Hubble (7-9) is automatically installed on your hard disk. If you want to remove it from your hard disk (to save a small amount of disk space) then you will need to replace it with an alias to the Hubble (7-9) file on the CD-ROM. (Hubble (7-9) is NOT automatically used on the CD-ROM when Starry Night is opened because it is assumed that it will be located on your hard disk.)
The Guide Star Catalog (GSC), version 1.1, is copyrighted by the Association of Universities for Research in Astronomy, Inc. and is reproduced here by permission. Here follows the original "README.TXT" file from the HGSC CD-ROM:
THE GUIDE STAR CATALOG
VERSION 1.1
An all-sky astrometric and photometric
catalog to support the operation of the
Hubble Space Telescope
Copyright (c) 1989, 1992, The Association of Universities
for Research in Astronomy, Inc.
The Guide Star Catalog (GSC) was prepared by the Space Telescope Science
Institute (ST ScI), 3700 San Martin Drive, Baltimore, MD 21218, USA.
ST ScI is operated by the Association of Universities for Research in
Astronomy, Inc. (AURA), under contract with the National Aeronautics and
Space Administration (NASA).
1. INTRODUCTION
The Guide Star Catalog (GSC), which has been constructed to support the
operational need of the Hubble Space Telescope (HST) for off-axis guide
stars, contains nearly 19 million objects brighter than sixteenth magni-
tude, of which more than 15 million are classified as stars.
The original version of this catalog, GSC 1.0, is described in a
series of papers: Lasker et al. (1990); Russell et al. (1990); and
Jenkner et al. (1990); hereafter referred to as Papers I, II, and III.
Additions and corrections made in GSC 1.1 address
- incompleteness, misnomers, artifacts, and other errors due to the
overexposure of the brighter stars on the Schmidt plates,
- the identification of blends likely to have been incorrectly
resolved,
- the incorporation of errata reported by the user-community or iden-
tified by the analysis of HST operational problems.
Among the primary authors of the GSC 1.0 and the associated
systems, the scientific responsibilities were divided as follows: Helmut
Jenkner, system coordination and overall design; Barry M. Lasker, astro-
physics and photometry; Brian J. McLean, algorithmic analysis and
systems development; Jane L. Russell, astrometry; Michael M. Shara,
system management; and Conrad R. Sturch, production management and
quality control. GSC 1.1 analysis and production were performed
primarily by Jesse B. Doggett, Daniel Egret, Brian J. McLean, and Conrad
R. Sturch.
Helmut Jenkner is on assignment from the European Space Agency;
Jane L. Russell is currently affiliated with the Applied Research
Corporation, Landover, MD; and Conrad R. Sturch is with the Astronomy
Programs, Computer Sciences Corporation at Space Telescope Science
Institute. Daniel Egret is affiliated with Observatoire de Strasbourg,
France.
2. DISCUSSION OF THE GSC 1.0 PROJECT
Astronomical and Algorithmic Foundation
As described in Paper I, the GSC is primarily based on an all-sky,
single epoch, single passband collection of Schmidt plates. For centers
at +6 degrees and north, a 1982 epoch "Quick V" survey was obtained by
the Palomar Observatory, while for southern fields, materials from the
UK SERC J survey (epoch approximately 1975) and its equatorial extension
(epoch approximately 1982) were used. In addition, over 100 short-
exposure plates were taken with the Palomar Oschin and UK Schmidt tele-
scopes to cover complex regions including the southern Milky Way, the
Magellanic Clouds, and M31. These northern, southern, and supplemental
plates hereafter are referred to as N, S, and X plates, respectively.
The plates were digitized into 14000-square rasters at 25 um sample
intervals using modified PDS microdensitometers.
The sky-background was modeled with a bi-dimensional cubic spline
approximation to the modal level. Then an object finder, based on
locating connected pixels at a certain threshold above the background,
was used to obtain, for each plate, a list of positions, sizes, inten-
sities, and related descriptive parameters. Images with multiple peaks
were deblended by an algorithm based on correlations against a library
of stellar images.
The identified objects were classified as stars or non-stars by an
interactive multivariate Bayesian classifier that used image features
from the object-detection steps and was started from a small set of
objects visually identified on each plate. Comparison of classifica-
tions from multiply cataloged objects in the plate overlap areas shows
that the purity of objects classified as stars is typically 97 percent.
Photometric and Astrometric Calibrations
The GSC calibrations were obtained on a plate-by-plate basis by polynom-
ial modeling against the photometric and astrometric reference catalogs.
Photometry is available in the natural systems defined by the
individual plates in the GSC collection (generally J or V), and the
calibrations are done using B, V standards from the Guide Star Photo-
metric Catalog (Lasker, Sturch, et al. 1988).
In Paper II the overall quality of the photometry near the standard
stars was estimated from the fits and other tests to be 0.15 mag (one
sigma, averaged over all plates), while the quality far from the sequen-
ces was estimated from the all-sky plate-to-plate agreement and from
comparisons with independent photometric surveys to be about 0.3 mag
(one sigma), with about 10% of the errors being greater than 0.5 mag.
Additionally, Ratnatunga's (1990) comparison of the GSC against totally
independent J-band photographic photometry for three southern fields (20
sq deg area) for 12.5 < J < 15.5 shows agreement at the 0.1-0.2 mag
level.
Astrometry, at equinox J2000, is available at the epochs of the
individual plates used in the GSC; and the reductions to the reference
catalogs (AGK3, SAOC, or CPC, depending on the declination zone) use
third order expansions of the modeled plate and telescope effects. The
fits to the reference catalogs lie in the range 0.5" to 0.9", and most
of this is attributable to errors in the reference catalogs, to cen-
troiding errors on the relatively large images of the reference stars,
and to unmodeled astrometric effects.
Paper II reported estimates of the overall external astrometric
error, produced by comparisons of independently measured positions, in
the range 0.2" to 0.8" (per coordinate), depending on the areas of the
plate and the sky. Then from a more extensive analysis against the
Carlsberg Automatic Meridian Circle data, Taff et al. (1990) found that
GSC absolute positional errors from plate center to edge vary from 0.5"
to 1.1" in the north and from 1.0" to 1.6" in the south, and that rela-
tive errors at half-degree separations range from 0.33" to 0.76" depend-
ing upon hemisphere and magnitude.
Production, Database Organization, and Population Statistics
Paper III describes the software system used to produce the GSC. It
consisted of a set of (primarily non-interactive) image-processing and
calibration programs interconnected by a set of pipeline files and
supported by databases organized on a plate-by-plate basis. A set of
utility programs was also provided to support quality control and to
correct operational problems.
Object names are of the form GSC rrrrr nnnnn, where the first field
specifies an internal region number and the second is an ordinal within
it. For objects cataloged from more than one photographic plate, an
entry was made from each image; and all entries for the same object were
given the same unique name.
Paper III also reviews the database for compiling statistics of
objects with multiple entries and the details of the organization and
structure of the GSC, including the provisions for assigning unique
names, for cataloging objects lying in the plate overlap regions, for
rapidly indexing positions against regions, and for recovering the
original plate measurements. The separate count statistics for stellar
and non-stellar objects on a plate-by-plate basis are provided in the
supporting tables.
User Interfaces, Utilities, and Astronomical Applications
The all-sky collection of Schmidt plates that were digitized, archived
to optical disc, and processed to generate the Guide Star Catalog (GSC)
constitute a general image resource for astronomical research.
This data set, combined with the computing environment provided by
the Guide Star Astrometric Support Package (GASP), major elements of
which are exported within the Space Telescope Science Data Analysis
System, provides random access to a digital image in any part of the
sky. The GASP environment also supports access to the GSC and to other
major astronomical catalogs.
3. REVISIONS IN GSC 1.1
The GSC 1.1 activities performed to address a number of known problems
in GSC 1.0 are summarized here and described in detail in the text file
for this revision, REV_1_1.TBL;1.
Two concerns related to the brighter stars arise from the heavily
overexposed images on the Schmidt plates used in the GSC, namely an
incompleteness and a reduced precision. Both are addressed in the
domain V < 7.5 by the use of data from the INCA Data Base (Turon et al.
1992; Jahreiss et al. 1992; Grenon et al. 1992) in the Tycho Input Cata-
log (TIC; Egret et al. 1992). Such entries are designated by the plate
identifier +056 in GSC 1.1. The limit of V < 7.5 preserves the original
GSC data for objects that were used in the GSC 1.0 astrometric calibra-
tion.
Naming errors occur when objects catalogued from more than one
photographic plate have positional errors sufficiently large that cross-
matching of the overlapping plate areas is done incorrectly. The most
significant known instances of this in GSC 1.0 were associated with
overexposed (and therefore badly centroided) images of the brighter
stars. A search around the positions of the INCA stars facilitated the
identification of these naming errors, which were then removed in
GSC 1.1.
GSC 1.0 contains many pairs of objects (from single plates) with
separations significantly smaller than the expected resolution of the
catalog, which Garnavich (1991), based on a study of four northern
plates, estimates at ~ 10" for 8.0 < V < 14.0. Visual inspection shows
that these are generally blends that have been properly resolved, but
then affected by a centroider defect that made the separations artifici-
ally small. Such components of blends with incorrect separations are
given a classification of 2 (blend; cf. the text file for a full listing
of the codes).
For stars with V < 8, image-processing artifacts near the diffrac-
tion spikes exist in GSC 1.0. In GSC 1.1, potential artifacts were
identified by use of a purely geometrical criterion (proximity of the
object to the spike), and were assigned a classification of 5.
Small areas around southern stars brighter than V ~ 3 are not
processed from the Schmidt plates and were left blank in GSC 1.0. For
these, GSC 1.1 contains entries from supplemental astrograph plates
taken with the GPO (Gran Prisma Objectif) telescope on La Silla, and the
astrograph at the Black Birch Observatory (BBO) in Blenheim, New Zea-
land. Because of their smaller fields, the photometric and astrometric
calibrations of data from most GPO and a few BBO plates were performed
against nearby GSC entries based on the Schmidt plates.
A number of specific errors in GSC 1.0 have been identified by the
user-community and by analyses of HST operational problems. These
generally involve naming errors, plate flaws, misclassifications, and
multiple stars; most are individually corrected in GSC 1.1. Also, the
photometric error parameter in GSC 1.1 is now correctly described by
equation (3) in Paper II; i.e., the erratum of footnote 5 therein is no
longer pertinent.
Plans for GSC maintenance beyond version 1.1 include an astrometric
recalibration (cf. Taff, Lattanzi, and Bucciarelli 1990), and investiga-
tion of erroneous double entries that may exist for V > 7.5.
4. ORGANIZATION OF THE DATA FILES
The Guide Star Catalog is subdivided into regions that are bounded by
small circles of right ascension and great circles of declination, and
that are numbered consecutively from 0001 to 9537. Data for each region
are stored as separate files; these files are contained in directories,
each of which subtends a 7.5 degree zone of declination.
The Guide Star Catalog is distributed as a two CD-ROM set, divided
at a declination of -7.5 degrees. An introductory file (README.TXT;1)
and the supporting tables are duplicated on both discs.
All data files (i.e., with the exception of the file README.TXT;1
and the directory files) are in FITS (Flexible Image Transport System;
Greisen et al. 1981; Wells et al. 1981; Grosbol et al. 1988; and
