The Sloan Digital Sky Survey: Technical Summary
Donald G. York, J. Adelman, John E. Anderson, Jr., Scott
F. Anderson, James Annis, Neta A. Bahcall, J. A. Bakken, Robert Barkhouser,
Steven Bastian, Eileen Berman, William N. Boroski, Steve Bracker, Charlie
Briegel, John W. Briggs, J. Brinkmann, Robert Brunner, Scott Burles, Larry
Carey, Michael A. Carr, Francisco J. Castander, Bing Chen, Patrick L.
Colestock, A. J. Connolly, J. H. Crocker, István Csabai, Paul C. Czarapata,
John Eric Davis, Mamoru Doi, Tom Dombeck, Daniel Eisenstein, Nancy Ellman,
Brian R. Elms, Michael L. Evans, Xiaohui Fan, Glenn R. Federwitz, Larry
Fiscelli, Scott Friedman, Joshua A. Frieman, Masataka Fukugita, Bruce
Gillespie, James E. Gunn, Vijay K. Gurbani, Ernst de Haas, Merle Haldeman,
Frederick H. Harris, J. Hayes, Timothy M. Heckman, G. S. Hennessy, Robert B.
Hindsley, Scott Holm, Donald J. Holmgren, Chi-hao Huang, Charles Hull, Don
Husby, Shin-Ichi Ichikawa, Takashi Ichikawa, Zeljko Ivezi\'c, Stephen Kent,
Rita S.J. Kim, E. Kinney, Mark Klaene, A. N. Kleinman, S. Kleinman, G. R.
Knapp, John Korienek, Richard G. Kron, Peter Z. Kunszt, D.Q. Lamb, B. Lee, R.
French Leger, Siriluk Limmongkol, Carl Lindenmeyer, Daniel C. Long, Craig
Loomis, Jon Loveday, Rich Lucinio, Robert H. Lupton, Bryan MacKinnon, Edward
J. Mannery, P. M. Mantsch, Bruce Margon, Peregrine McGehee, Timothy A. McKay,
Avery Meiksin, Aronne Merelli, David G. Monet, Jeffrey A. Munn, Vijay K.
Narayanan, Thomas Nash, Eric Neilsen, Rich Neswold, Heidi Jo Newberg, R. C.
Nichol, Tom Nicinski, Mario Nonino, Norio Okada, Sadanori Okamura, Jeremiah
P. Ostriker, Russell Owen, A. George Pauls, John Peoples, R. L. Peterson,
Donald Petravick, Jeffrey R. Pier, Adrian Pope, Ruth Pordes, Angela Prosapio,
Ron Rechenmacher, Thomas R. Quinn, Gordon T. Richards, Michael W. Richmond,
Claudio H. Rivetta, Constance M. Rockosi, Kurt Ruthmansdorfer, Dale Sandford,
David J. Schlegel, Donald P. Schneider, Maki Sekiguchi, Gary Sergey, Kazuhiro
Shimasaku, Walter A. Siegmund, Stephen Smee, J. Allyn Smith, S. Snedden, R.
Stone, Chris Stoughton, Michael A. Strauss, Christopher Stubbs, Mark
SubbaRao, Alexander S. Szalay, Istvan Szapudi, Gyula P. Szokoly, Anirudda R.
Thakar, Christy Tremonti, Douglas L. Tucker, Alan Uomoto, Dan VandenBerk,
Michael S. Vogeley, Patrick Waddell, Shu-i Wang, Masaru Watanabe, David H.
Weinberg, Brian Yanny, and Naoki Yasuda (The SDSS Collaboration)
The Sloan Digital Sky Survey (SDSS) will provide the data to support detailed
investigations of the distribution of luminous and non-luminous matter in the
Universe: a photometrically and astrometrically calibrated digital imaging
survey of p steradians above about Galactic
latitude 30? in five broad optical
bands to a depth of g' ~
23m, and a spectroscopic survey of the approximately
106 brightest galaxies and 105 brightest quasars found
in the photometric object catalog produced by the imaging survey. This paper
summarizes the observational parameters and data products of the SDSS, and
serves as an introduction to extensive technical on-line documentation.
At this writing (May 2000) the Sloan Digital Sky Survey (SDSS) is ending its
commissioning phase and beginning operations. The purpose of this paper is to
provide a concise summary of the vital statistics of the project, a
definition of some of the terms used in the survey and, via links to
documentation in electronic form, access to detailed descriptions of the
project's design, hardware, and software, to serve as technical background
for the project's science papers. The electronic material is extracted from
the text (the "Project Book") written to support major funding proposals, and
is available at the Astronomical Journal web site via the on-line
version of this paper. The official SDSS web site
(http://www.sdss.org) also provides links to the on-line Project
Book, and it can be accessed directly at
http://www.astro.princeton.edu/PBOOK/welcome.htm. In the discussion
below we reference the chapters in the Project Book by the last part of the
URL, i.e. that following PBOOK/. The versions accessible at the SDSS
web sites also contain extensive discussions and summaries of the scientific
goals of the survey, which are not included here.
The text of the on-line Project Book was last updated in August 1997. While
there have been a number of changes in the hardware and software described
therein, the material accurately describes the design goals and the
implementation of the major observing subsystems. As the project becomes
operational, we will provide a series of formal technical papers (most still
in preparation), which will describe in detail the project hardware and
software in its actual operational state.
Section 2 describes the Survey's objectives: the imaging depth, sky coverage,
and instrumentation. Section 3 summarizes the software and data reduction
components of the SDSS and its data products. Section 4 reviews some recent
scientific results from the project's initial commissioning data runs, which
demonstrate the ability of the project to reach its technical goals. All
Celestial coordinates are in epoch J2000.
2 Survey Characteristics
The Sloan Digital Sky Survey will produce both imaging and spectroscopic
surveys over a large area of the sky. The survey uses a dedicated 2.5 m
telescope equipped with a large format mosaic CCD camera to image the sky in
five optical bands, and two digital spectrographs to obtain the spectra of
about one million galaxies and 100,000 quasars selected from the imaging data.
The SDSS calibrates its photometry using observations of a network of
standard stars established by the United States Naval Observatory (USNO) 1 m
telescope, and its astrometry using observations by an array of astrometric
CCDs in the imaging camera.
The SDSS telescope is a 2.5m f/5 modified Ritchey-Chrétien wide-field
altitude-azimuth telescope (see telescop/telescop.htm) located at
the Apache Point Observatory (APO), Sunspot, New Mexico
(site/site.htm). The telescope achieves a very wide (3?) distortion-free field by the use of a large
secondary mirror and two corrector lenses. It is equipped with the
photometric/astrometric mosaic camera (camera/camera.htm, Gunn et
al. 1998) and images the sky by scanning along great circles at the sidereal
rate. The imaging camera mounts at the Cassegrain focus. The telescope is
also equipped with two double fiber-fed spectrographs, permanently mounted on
the image rotator, since the spectrographs are fiber fed. This ensures that
the fibers do not flex during an exposure. The telescope is changed from
imaging mode to spectroscopic mode by removing the imaging camera and
mounting at the Cassegrain focus a fiber plug plate, individually drilled for
each field, which feeds the spectrographs. In survey operations, it is
expected that up to nine spectroscopic plates per night will be observed,
with the necessary plates being plugged with fibers during the day. The
telescope mounting and enclosure allow easy access for rapid changes between
fiber plug plates and between spectroscopic and imaging modes. This strategy
allows imaging to be done in pristine observing conditions (photometric sky,
image size ? 1.5" FWHM) and spectroscopy to be
done during less ideal conditions. All observing will be done in moonless sky.
Besides the 2.5m telescope, the SDSS makes use of three subsidiary
instruments at the site. The Photometric Telescope (PT) is a 0.5m
telescope equipped with a CCD camera and the SDSS filter set. Its task is to
calibrate the photometry. Two instruments, a seeing monitor and a
10mm cloud scanner (Hull et al. 1995;
site/site.htm) monitor the astronomical weather.
2.2 Imaging Camera
The SDSS imaging camera contains two sets of CCD arrays: the imaging array
and the astrometric arrays (camera/camera.htm, Gunn et al. 1998).
The imaging array consists of 30 2048 × 2048 Tektronix CCDs, placed
in an array of six columns and five rows. The telescope scanning is aligned
with the columns. Each row observes the sky through a different filter, in
temporal sequence r?, i?, u?, z?, and g?. The pixel size is
24mm (0.396" on the sky). The imaging survey is
taken in drift-scan (time-delay-and-integrate, or TDI) mode, i.e. the camera
continually sweeps the sky in great circles, and a given point on the sky
passes through the five filters in succession. The effective integration time
per filter is 54.1 seconds, and the time for passage over the entire
photometric array is about 5.7 minutes (strategy/strategy.htm; Gunn
et al. 1998). Since the camera contains six columns of CCDs, the result is a
long strip of six scanlines, containing almost simultaneously
observed five-color data for each of the six CCD columns. Each CCD observes
a swath of sky 13.52? wide. The CCDs are
separated in the row direction (i.e. perpendicular to the scan
direction) by 91.0 mm (25.2? on the sky)
center-to-center. The observations are filled in by a second strip, offset
from the first by 93% of the CCD width, to produce a filled stripe,
2.54? wide, with 8% (1?) lateral overlap on each side. Because of the camera's
large field of view, the TDI tracking must be done along great circles. The
Northern Galactic Cap is covered by 45 great-circle arcs (shown projected on
the sky in Figures 1 and 2).
2.3 Photometry and Photometric
The five filters in the imaging array of the camera, [u?, g?, r?, i? and z?] have effective wavelengths of [3590 Å, 4810 Å, 6230
Å, 7640 Å and 9060 Å] (Fukugita et al. 1996; Gunn et al. 1998). An a
priori model estimate of the telescope and camera throughputs and of the sky
brightness predicted that we would reach the 5s
detection limit for point sources in 1" seeing at [22.3, 23.3, 23.1, 22.3,
20.8] in the (u?,g?,r?,i?,z?) filters, respectively,
at an airmass of 1.4. We have put formal requirements on throughput at 75% of
the values used for the above estimation, and have demonstrated that we meet
this requirement in all bands with the possible exception of z?. The sensitivity limit can be tested by finding the
magnitude at which repeat observations of a given area of sky yield 50%
reproducibility of the objects detected. This has been tested most
thoroughly with data taken in less than optimal seeing (1.3"-1.6"); nevertheless, the 50% reproducibility level lies
within a few tenths of a magnitude of the above-quoted 5s detection limit in all five bands (see Ivezi\'c et al.
2000). The SDSS science requirements demand that photometric calibration
uncertainties for point sources be 0.02 in r?,
0.02 in r?-i? and g?-r?, and 0.03 in u?-g? and i?-z?. To meet these stringent
requirements in both signal-to-noise ratio and photometricity, imaging data
are declared to be survey quality only if the PT determines that the night is
photometric, with a zero-point uncertainty below 1%, and if the seeing is
better than 1.5". The imaging data saturate at about [13, 14, 14, 14, 12]
magnitudes for point sources.
The magnitude scale is on the ABn
system (Oke 1969, unpublished), which was updated to the AB79
system by Oke & Gunn (1983) and to AB95 by Fukugita et al.
(1996). The magnitudes m are related to flux density f by m ~ sinh-1(f)
rather than logarithmically (see Lupton, Gunn & Szalay 1999 and Fan et
al. 1999). This definition is essentially identical to the logarithmic
magnitude at signal-to-noise ratios greater than about 5 and is well behaved
for low and even zero and negative flux densities.
The calibration and definition of the magnitude system is carried out by the
USNO 1 m telescope and the 0.5m PT. The SDSS photometry is placed on the
ABn system using three fundamental
standards (BD+17?4708, BD+26?2606, and BD+21?609), whose magnitude scale is as defined by Fukugita et
al. (1996); a set of 157 primary standards, which are calibrated by
the above fundamental standards using the USNO 1m telescope, and which cover
the whole range of right ascension and enable the calibration system to be
made self-consistent; and a set of secondary calibration patches lying
across the imaging stripes, containing stars fainter than 14m
whose magnitudes are calibrated by the PT with respect to those of the
primary standards and which transfer that calibration to the imaging survey.
The locations of these patches on the survey stripes are shown in Figure 1.
On nights when the 2.5 m is observing, the PT observes primary standard stars
to provide the atmospheric extinction coefficients over the night and to
confirm that the night is photometric. The standard star network is described
in photcal/photcal.htm - note that the telescope described there has
now been replaced by the 0.5m PT.
2.4 Astrometric Calibration
The camera also contains leading and trailing astrometric arrays -
narrow (128×2048), neutral-density-filtered, r?-filtered CCDs covering the entire width of the camera.
These arrays can measure objects in the magnitude range r? ~ 8.5 - 16.8, i.e. they
cover the dynamic range between the standard astrometric catalog stars and
the brightest unsaturated stars in the photometric array. The astrometric
calibration is thereby referenced to the fundamental astrometric catalogues
(see astrom/astrom.htm), using the Hipparcos and Tycho Catalogues
(ESA 1997) and specially observed equatorial fields (Stone et al. 1999).
Comparison with positions from the FIRST (Becker et al. 1995) and 2MASS
(Skrutskie 1999) catalogues shows that the rms astrometric accuracy is
currently better than 150 milliarcseconds (mas) in each coordinate.
2.5 Imaging Survey: North Galactic
The imaging survey covers about 10,000 contiguous square degrees in the
Northern Galactic Cap. This area lies basically above Galactic latitude
30?, but its footprint is adjusted
slightly to lie within the minimum of the Galactic extinction contours
(Schlegel, Finkbeiner & Davis 1998), resulting in an elliptical region.
The region is centered at a =
12h 20m, d =
+32.5?. The minor axis is at an
angle 20? East of North with extent
major axis is a great circle perpendicular to the minor axis with extent
survey footprint with the location of the stripes is shown in Figure 2 - see
strategy/strategy.htm for details.
2.6 Imaging Survey: The South
In the South Galactic Cap, three stripes will be observed, one along the
Celestial Equator and the other two north and south of the equator (see
Figure 2). The equatorial stripe (a =
20.7h to 4h, d = 0?) will be observed repeatedly, both to find
variable objects and, when co-added, to reach magnitude limits about
2m deeper than the Northern imaging survey.
The other two stripes will cover great circles lying between a, d of
(20.7h, -5.8? ? 4.0h, -5.8?) and (22.4h, 8.7? ? 2.3h,
2.7 The Spectroscopic Survey
Objects are detected in the imaging survey, classified as point source or
extended, and measured, by the image analysis software (see below). These
imaging data are used to select in a uniform way different classes of objects
whose spectra will be taken. The final details of this target
selection will be described once the survey is well underway; the
criteria discussed here are likely to be very close to those finally used.
Two samples of galaxies are selected from the objects classified as
"extended". About 9 ×105 galaxies will be selected to have
Petrosian (1976) magnitudes r?P
? 17.7. Galaxies with a mean r? band surface brightness within the half light radius
fainter than 24 magnitudes/arc second2 will be
removed, since spectroscopic observations are unlikely to produce a redshift.
For illustrative purposes, a simulation of a slice of the SDSS redshift
survey is shown in Figure 3 (from Colley et al. 2000). Galaxies in this CDM
simulation are `selected' by the SDSS selection criteria. As Figure 3
demonstrates, the SDSS volume is large enough to contain a statistically
significant sample of the largest structures predicted.
The second sample, of approximately 105 galaxies, exploits the
characteristic very red color and high metallicity (producing strong
absorption lines) of the most luminous galaxies: the "Brightest Cluster
Galaxies" or "Bright Red Galaxies" (BRGs); redshifts can be well measured
with the SDSS spectra for these galaxies to about r? = 19.5. Galaxies located at the dynamical centers of
nearby dense clusters often have these properties. Reasonably accurate
photometric redshifts (Connolly et al. 1995) can be determined for these
galaxies, allowing the selection by magnitude and g?r?i? color of an essentially distance limited sample
of the highest-density regions of the Universe to a redshift of about 0.45
(see Figure 4 for a simulation).
With their power-law continua and the influence of Lyman-a emission and the Lyman-a
forest, quasars have u?g?r?i?z? colors quite distinct
from those of the vastly more numerous stars over most of their redshift
range (Fan 1999). Thus about 1.5 ×105 quasar candidates
are selected for spectroscopic observations as outliers from the stellar
locus (cf., Krisciunas et al. 1998; Lenz et al. 1998; Newberg et al. 1999;
Figure 5 below) in color-color space. At the cost of some loss of
efficiency, selection is allowed closer to the stellar locus around z = 2.8,
where quasar colors approach those of early F and late A stars (Newberg &
Yanny 1997; Fan 1999). Some further regions of color-color space outside the
main part of the stellar locus where quasars are very rarely found are also
excluded, including the regions containing M dwarf-white dwarf pairs, early A
stars, and white dwarfs (see Figure 5). The SDSS will compile a sample of
quasars brighter than i? ? 19 at z < 3.0; at redshifts between 3.0 and about
5.2, the limiting magnitude will be about i? = 20.
Objects are also required to be point sources, except in the region of
color-color space where low-redshift quasars are expected to be found.
Stellar objects brighter than i? =20 which are
FIRST sources (Becker, White and Helfand 1995) are also selected. Based on
early spectroscopy, we estimate that roughly 65% of our quasar candidates are
genuine quasars; comparison with samples of known quasars indicates that our
completeness is of order 90%.
In all cases, the magnitudes of the objects are corrected for Galactic
extinction before selection, using extinction in the SDSS bands calculated
from the reddening map of Schlegel, Finkbeiner & Davis (1998). Objects
are then selected to have a magnitude limit outside the Galaxy. If
this correction were not made, the systematic effects of Galactic extinction
over the survey area would overwhelm the statistical uncertainties in the
SDSS data set. After the imaging and spectroscopic survey is completed in a
given part of the sky, the reddening and extinction will be recalculated
using internal standards extracted from the imaging data. The SDSS plans to
use a variety of extinction probes, including very hot halo subdwarfs, halo
turnoff stars, and elliptical galaxies whose intrinsic colors can be
estimated from their line indices.
Together with various classes of calibration stars and fibers which observe
blank sky to measure the sky spectrum, the selected galaxies and quasars are
mapped onto the sky, and `tiled', i.e. their location on a 3? diameter plug plate determined (
tiling/tiling.htm). The centers of the tiles are adjusted to provide
closer coverage of regions of high galactic surface density, to make the
spectroscopic coverage optimally uniform. Excess fibers are allocated to
several classes of rare or peculiar objects (for example objects which are
positionally matched with ROSAT sources, or those whose parameters lie
outside any known range - these are serendipitous objects) and to
samples of stars. The spectra are observed, 640 at a time (with a total
integration time of 45 - 60 minutes depending on observing conditions) using
a pair of fiber-fed double spectrographs (spectro/spectro.htm). The
wavelength coverage of the spectrographs is continuous from about 3800 Å to
9200 Å, and the wavelength resolution, l/dl, is 1800 (Uomoto et al.
1999). The fibers are located at the focal plane via plug plates constructed
for each area of sky. The fiber diameter is 0.2 mm (3" on the sky), and
adjacent fibers cannot be located more closely than 55" on the sky. Both
members of a pair of objects closer than this separation can be observed
spectroscopically if they are located in the overlapping regions of adjacent
Tests of the redshift accuracy using observations of stars in M67 whose
radial velocities are accurately known (Mathieu et al. 1986) show that the
SDSS radial velocity measurements for stars have a scatter of about 3.5
3 Software and Data Products
The operational software is described in datasys/datasys.htm. The
data are obtained using the Data Acquisition (DA) system at APO (Petravick et
al. 1994) and recorded on DLT tape. The imaging data consist of full images
from all CCDs of the imaging array, cut-outs of detected objects from the
astrometric array, and bookkeeping information. These tapes are shipped to
Fermilab by express courier and the data are automatically reduced through an
interoperating set of software pipelines operating in a common computing
The photometric pipeline reduces the imaging data; it corrects the
data for data defects (interpolation over bad columns and bleed trails,
finding and interpolating over `cosmic rays', etc), calculates overscan
(bias), sky and flat field values, calculates the point spread functions
(psf) as a function of time and location on the CCD array, finds objects,
combines the data from the five bands, carries out simple model fits to the
images of each object, deblends overlapping objects, and measures positions,
magnitudes (including psf and Petrosian magnitudes) and shape parameters.
The photometric pipeline uses position calibration information from the
astrometric array reduced through the astrometric pipeline and
photometric calibration data from the photometric telescope (reduced through
the photometric telescope pipeline). Final calibrations are applied
by the final calibration pipeline, which allows refinements in the
positional and photometric calibration as the survey progresses. The
photometric pipeline is extensively tested using repeat observations,
examination of the outputs, observations of regions of the sky previously
observed by other telescopes (HST fields, for example) and a set of
simulations, described in detail in simul/simul.htm. For an example
of the repeatability of SDSS photometry over several timescales, see Ivezi\'c
et al. (2000). These repeat observations show that the mean errors (for point
sources) are about 0.03m to 20m, increasing to about
0.05m at 21m and to 0.12m at 22m.
These observed errors are in good agreement with those quoted by the
photometric pipeline. They apply only to the g?,
r? and i? bands - in
the less sensitive u? and z? bands, the errors at the bright end are about the same
as those in g?r?i?, but increase to 0.05m at 20m
and 0.12m at 21m.
The outputs, together with all the observing and processing information, are
loaded into the operational data base which is the central collection
of scientific and bookkeeping data used to run the survey. To select the
spectroscopic targets, objects are run through the target selection
pipeline and flagged if they meet the spectroscopic selection criteria
for a particular type of object. The criteria for the primary objects
(quasars, galaxies and BRGs) will not be changed once the survey is underway.
Those for serendipitous objects and samples of interesting stars can be
changed throughout the survey. A given object can in principle receive
several target flags. The selected objects are tiled as described above, plug
plates are drilled, and the spectroscopic observations are made. The
spectroscopic data are automatically reduced by the spectroscopic
pipeline, which extracts, corrects and calibrates the spectra, determines
the spectral types, and measures the redshifts. The reduced spectra are then
stored in the operational data base. The contents of the operational data
base are copied at regular intervals into the science data base for
retrieval and scientific analysis (see appsoft/appsoft.htm). The
science data base is indexed in a hierarchical manner: the data and other
information are linked into `containers' that can be divided and subdivided
as necessary, to define easily searchable regions with approximately the same
data content. This hierarchical scheme is consistent with those being adopted
by other large surveys, to allow cross referencing of multiple surveys. The
science data base also incorporates a set of query tools and is designed for
The photometric data products of the SDSS include: a catalog of all
detected objects, with measured positions, magnitudes, shape parameters,
model fits and processing flags; atlas images (i.e. cutouts from the
imaging data in all five bands) of all detected objects and of objects from
the FIRST and ROSAT catalogs; a 4 ×4 binned image of the corrected images
with the objects removed: and a mask of the areas of sky not processed
(because of saturated stars, for example) and of corrected pixels (e.g. those
from which cosmic rays were removed). The atlas images are sized to enclose
the area occupied by each object plus the PSF width, or the object size given
in the ROSAT or FIRST catalogues. The photometric outputs are described in
http://www.astro.princeton.edu/SDSS/photo.html. The data base will
also contain the calibrated 1D spectra, the derived redshift
and spectral type, and the bookkeeping information related to the
spectroscopic observations. In addition, the positions of astrometric
calibration stars measured by the astrometric pipeline and the magnitudes of
the faint photometric standards measured by the photometric telescope
pipeline will be published at regular intervals.
4 Early Science from the SDSS
The goal of the SDSS is to provide the data necessary for studies of the
large scale structure of the Universe on a wide range of scales. The imaging
survey should detect ~ 5 ×107
galaxies, ~ 106 quasars and ~ 8 ×107 stars to the survey limits. These
photometric data, via photometric redshifts and various statistical
techniques such as the angular correlation function, support studies of large
scale structure well past the limit of the spectroscopic survey. On even
larger scales, information on structure will come from quasars.
The science justification for the SDSS is discussed in several conference
papers (e.g. Gunn & Weinberg 1995; Fukugita 1998; Margon 1999). The
Project Book science sections can be accessed at
http://www.astro.princeton.edu/PBOOK/science/science.htm. Much of
the science for which the SDSS was built, the study of large scale structure,
will come when the survey is complete, but the initial test data have already
led to significant scientific discoveries in many fields. In this section, we
show examples of the first test data and some initial results. To date (May
2000), the SDSS has obtained test imaging data for some 2000 square degrees
of sky and about 20,000 spectra. Examples of these data are shown in Figures
5 (sample color-color and color-magnitude diagrams of point-source objects),
6 (sample spectra) and 7 (a composite color image of a piece of the sky which
contains the cluster Abell 267),
Fischer et al. (2000) have detected the signature of the weak lensing of
background galaxies by foreground galaxies, allowing the halos and total
masses of the foreground galaxies to be measured.
The searches by Fan et al. (1999a,b; 2000a,c), Schneider et al. (2000) and
Zheng et al. (2000) have greatly increased the number of known high redshift
(z > 3.6) quasars and include several quasars with z > 5. Fan et al.
(1999b) have found the first example of a new kind of quasar: a high redshift
object with a featureless spectrum and without the radio emission and
polarization characteristics of BL Lac objects. The redshift for this object
(z = 4.6) is found from the Lyman-a forest
absorption in the spectrum.
Some 150 distant probable RR Lyrae stars have been found in the Galactic
halo, enabling the halo stellar density to be mapped; the distribution may
have located the edge of the halo at approximately 60 kpc (Ivezi\'c et al.
2000). The distribution of RR Lyrae stars and other horizontal branch stars
is very clumped, showing the presence of possible tidal streamers in the halo
(Ivezi\'c et al. 2000; Yanny et al. 2000). Margon et al. (1999) describe the
discovery of faint high latitude carbon stars in the SDSS data.
Strauss et al. (1999), Schneider et al. (2000), Fan et al. (2000b), Tsvetanov
et al. (2000), Pier et al. (2000) and Leggett et al. (2000) report the
discovery of a number of very low mass stars or substellar objects, those of
type `L' or `T', including the first field methane (`T') dwarfs and the first
stars of spectral type intermediate between `L' and `T'. The detection rate
to date shows that the SDSS is likely to identify several thousand L and T
dwarfs. These objects are found to occupy very distinct regions of
color-color and color-magnitude space, which will enable the completeness of
the samples to be well characterized.
Measurements of the psf diameter variations and the image wander allow
variations in the turbulence in the Earth's atmosphere to be tracked. These
data demonstrate the presence of anomalous refraction on scales at least as
large as the 2.3? field of view
of the camera (Pier et al. 1999).
Of course the most exciting possibility for any large survey which probes new
regions of sensitivity or wavelength is the discovery of exceedingly rare or
entirely new classes of objects. The SDSS has already found a number of very
unusual objects; the nature of some of these remains unknown (Fan et al.
1999c). These and other investigations in progress show the promise of SDSS
for greatly advancing astronomical work in fields ranging from the behavior
of the Earth's atmosphere to structure on the scale of the horizon of the
The Sloan Digital Sky Survey (SDSS) is a joint project of The University
of Chicago, Fermilab, the Institute for Advanced Study, the Japan
Participation Group, The Johns Hopkins University, the Max-Planck-Institute
for Astronomy, Princeton University, the United States Naval Observatory, and
the University of Washington. Apache Point Observatory, site of the SDSS, is
operated by the Astrophysical Research Consortium. Funding for the project
has been provided by the Alfred P. Sloan Foundation, the SDSS member
institutions, the National Aeronautics and Space Administration, the National
Science Foundation, the U.S. Department of Energy, and Monbusho. The
official SDSS web site is www.sdss.org.
Bade, N., Engels, D., Voges, W., et al. 1998, A&AS, 127, 145
Becker, R.H., White, R.L., & Helfand, D.J. 1995, ApJ, 450, 559
Berger, J., & Fringant, A.-M., 1985, A&AS, 61, 191
Colley, W. N., Gott, J. R., Weinberg, D. H., Park, C., & Berlind, A. A.
2000, ApJ, 529, 795
Connolly, A.J., Csabai, I., Szalay, A.S., Koo, D.C., Kron, R.G., & Munn,
J.A. 1995, AJ, 110, 2655
Crawford, C.S., Allen, S.W., Ebeling, H., A.C., & Fabian, A.C. 1999,
MNRAS, 306, 857
ESA 1997, The Hipparcos and Tycho Catalogues, ESA SP-1020
Fan, X. 1999, AJ, 117, 2528
Fan, X., Strauss, M. A., Schneider, D.P., et al. 1999a, AJ, 118, 1
Fan, X., Strauss, M.A., Gunn, J.E. et al. 1999b, ApJ 526, L57
Fan, X., Strauss, M.A., Schneider, D.P., Gunn, J.E., Lupton, R.L., Knapp,
G.R., & Yanny, B. 1999c, AAS, 73.15
Fan, X., Strauss, M. A., Schneider, D.P., et al. 2000a, AJ, 119, 1
Fan, X., Knapp, G.R., Strauss, M.A., et al. 2000b, AJ, 119, 928
Fan, X., White, R.L., Davis, M., et al. 2000c, submitted to AJ
Fischer, P., McKay, T., Sheldon, E. et al. 2000, submitted to AJ
Fukugita, M. 1998, in Highlights of Astronomy, 11A, ed. J Andersen, 449.
Fukugita, M., Ichikawa, T., Gunn, J.E., Doi, M., Shimasaku, K., &
Schneider, D.P. 1996, AJ, 111, 1748
Gunn, J.E., Carr, M.A., Rockosi, C.M., Sekiguchi, M., et al. 1998, AJ, 116,
Gunn, J.E., & Weinberg, D.H. 1995, in "Wide Field Spectroscopy and the
Distant Universe", ed. S. Maddox & A. Aragòn-Salamanca, World Scientific
Hull, C., Limmongkol, S., & Siegmund, W. 1994, Proc. S.P.I.E., 2199
Ivezi\'c, Z., Goldston, J., Finlator, K., et al. 2000, AJ (in press:
Krisciunas, K., Margon, B., & Szkody, P. 1998, PASP, 110, 1342
Lenz, D.D., Newberg, H.J., Rosner, R., Richards, G.T., & Stoughton, C.
1998, ApJS, 119, 121
Lupton, R.H., Gunn, J.E., & Szalay, A. 1999, AJ, 118, 1406
Mathieu, R.D., Latham, D.W., Griffin, R.F., & Gunn, J.E. 1986, AJ, 92,
Margon, B. 1999, Philosophical Transactions of the Royal Society of London A,
Margon, B., Anderson, S.F., Deutsch, E., & Harris, H. 1999, BAAS 195.8006
Morris, S.L., Weymann, R.J., Anderson, S.F., Hewett, P.C., Foltz, C.B.,
Chaffee, F.H., Francis, P.J., & MacAlpine, G.M. 1991, AJ, 102, 1627
Newberg, H.J., & Yanny, B. 1997, ApJS, 113, 89
Newberg, H.H., Richards, G.T., Richmond, M.W., & Fan, X. 1999, ApJS, 123,
Oke, J.B., & Gunn, J.E. 1983, ApJ, 266, 713
Petravick, D., et al. 1994, S.P.I.E., 2198, 935
Petrosian, V. 1976, ApJ, 209, L1
Pier, J.R., Leggett, S.K., Strauss, M.A., et al. 2000, in "From Giant Planets
to Cool Stars", ed.C. Griffith & M. Marley, ASP Conf. Ser., in press
Pier, J.R., Munn, J.A., Hennessy, G.S., Hindsley, R.H., & Kent, S.M.
1999, AAS, 195, 81.04
Schlegel, D.J., Finkbeiner, D.P., & Davis, M. 1998, ApJ, 500, 525
Schneider, D.P., Hill, G.J., Fan, X., et al. 2000, PASP, 112, 6
Skrutskie, M.F. 1999, AAS, 195, 34.01
Strauss, M.A., Fan, X., Gunn, J.E., et al., 1999, ApJ, 522, L61
Stone, R.C., Pier, J.R., & Monet, D.G. 1999, AJ, 118, 2488
Tsvetanov, Z.I., Golimowski, D., Zheng, W., et al. 2000, ApJ, 513, L61
Uomoto, A., Smee, S., Rockosi, C., Burles, S., Pope, A., Friedman, S.,
Brinkmann, J., Gunn, J.E., & Nichol, R, 1999, AAS, 195.87.01
Yanny, B., Newberg, H., Kent, S., et al. 2000, AJ (in press:
Zheng, W., Tsvetanov, Z.I., Schneider, D.P., et al. 2000, submitted to AJ
Figure 1. Projection on the sky of the northern SDSS survey area. The
positions of the Yale Bright Star Catalogue stars are shown. The largest
symbols are stars of 0m and the smallest stars of 5m -
7m. The secondary calibration patches are shown by squares.
Figure 2. Projection on the sky (Galactic coordinates) of the Northern and
Southern SDSS surveys. The lines show the individual stripes to be scanned
by the imaging camera. These are overlaid on the extinction contours of
Schlegel, Finkbeiner and Davis (1998). The Survey pole is marked by the `X'.
Figure 3. A six-degree wide slice of the Simulated Sloan Digital Sky Survey
(from Colley et al. 2000), showing about 1/20 of the survey.
Figure 4. Simulated redshift distribution in a 6? slice of the SDSS. Small dots: main galaxy
sample (cf. Figure 3). Large dots: the BRG sample, showing about 1/30 of the
Figure 5. Color-color and color-magnitude plots of about 117,000 point
sources brighter than 21m at i* and detected at greater
than 5s in each band from 25 square degrees of
SDSS imaging data, reduced by the photometric pipeline (the i*
designation is used for preliminary SDSS photometry). The contours are drawn
at intervals of 10% of the peak density of points. The redder stars extend
to fainter magnitudes than do the bluer stars, due to the i* limit.
Figure 6. Representative SDSS spectra taken from a single spectroscopic plate
observed on 4 October 1999 for a total of one hour of integration time,
processed by the SDSS spectroscopic pipeline. For display purposes, all
spectra have been smoothed with a 3-pixel boxcar function. All spectra show
significant residuals due to the strong sky line at 5577Å. The objects
depicted are: a. An r*P = 18.00 galaxy; z = 0.1913.
This object is slightly fainter than the main galaxy target selection limit.
Note the Ha/[N II] emission at ~ 7800Å, [OII] emission at ~
4450Å, and Ca II H and K absorption at ~
4700Å. b. An r*P = 19.41 galaxy, z = 0.3735.
This object is close to the photometric limit of the Bright Red Galaxy
sample. The H and K lines are particularly strong. c. A star-forming galaxy
with r*P = 16.88, at z = 0.1582. d. A z = 0.3162
quasar, with r*psf = 16.67. Note the unusual profile
shape of the Balmer lines. This quasar is LBQS 0004+0036 (Morris et al.
1991). e. A z = 2.575 quasar with r*psf = 19.04; note
the resolution of the Lyman-a forest. This quasar
was discovered by Berger & Fringant (1985). f. A hot white dwarf, with
rpsf* = 18.09.
Figure 7. A sample frame (13?×9?) from the SDSS imaging commissioning data. The image,
a color composite made from the g?, r?, and i? data, shows a field
containing the distant cluster Abell 267 (a
= 01h 52m 41.0s,
d = +01? 00? 24.7",
redshift z = 0.23, Crawford et al. 1999); this is the cluster of galaxies
with yellow colors in the lower center of the frame. The frame also
contains, in the upper center, the nearby cluster RX J0153.2+0102, estimated
redshift ~ 0.07 (Bade et al. 1998) (a =
01h 53m 15.15s, d = +01? 02? 18.8").
The psf (optics plus seeing) was about 1.6". Right ascension
increases from bottom to top of the frame, declination from left to right.
Last modified: Mon Mar 3 15:22:59 CST 2003