The Sloan Digital Sky Survey: Overview and Operations

Figure 1. The Sloan Digital Sky Survey (hereafter SDSS) is a digital imaging and spectroscopic data bank of 25 percent of the celestial sphere, covering most of the sky above galactic latitude of 30 degrees and visible from New Mexico. Recent technical overviews of the project may be found in Margon, Phil. Trans. Roy. Soc. Lond. A, 357, 93, 1999, and Loveday and Pier, Wide Field Surveys in Cosmology, ed. S. Colombi, Y. Mellier, and B. Raban, (Paris: Editions Frontieres), 317, 1999.

Figure 2. The instrumentation for the project includes a special purpose telescope and unique mosaic camera (Gunn et al., AJ, 116, 3040, 1998). The collaboration spans a large number of institutions on three continents.

Figure 3. The project is managed under the umbrella of the Astrophysical Research Consortium, a non-profit corporation of 7 educational institutions, based in Seattle, Washington. This group also manages the 3.5-meter general purpose telescope on the same site in New Mexico, although not all members participate in both projects.

Figure 4. The Survey includes both imaging and spectroscopy, and both of these large data bases are put to multiple uses.

Figure 5. The telescope, mosaic camera, and multi-object spectrograph all have unique properties optimised specifically for the 5-year survey.

Figure 6. The SDSS is conducted at the Apache Point Observatory, located in southern New Mexico, on a peak of elevation 2788 meters. The site was chosen as a compromise between local conditions and minimisation of construction and operation costs via utilisation of existing infrastructure. With our 0.4 arcsec pixels, the superb seeing of Mauna Kea or Chile is not essential.

Figure 7. This view of the observatory shows the 3.5 meter ARC telescope enclosure in the center, with the SDSS 2.5 meter telescope housed under the roll on/roll off shed on the extreme left. The small dome in the center contains the 0.5 meter telescope used for nightly photometric calibration of the imaging data base. The Richard B. Dunn Solar Telescope of AURA's National Solar Observatory at Sunspot is visible in the distance (far right).

Figure 8. The SDSS 2.5 meter alt-az telescope is a scaled down version of the ARC 3.5 meter (which also provided the design heritage for the WIYN telescope on Kitt Peak).

Figure 9. The 2.5 meter telescope enclosure rolls back at night to expose the telescope, which then operates unencumbered by the building and possible seeing degradation due to poor airflow. The instrument is on an elevated platform at the immediate edge of a steep cliff, where the prevailing winds from White Sands are not yet turbulent.

Figure 10. The SDSS telescope is enclosed by a wind baffle which inhibits vibration despite the absence of an enclosure building; the baffle corotates with, but does not touch, the telescope.

Figure 11. The CCDs used in the mosaic imaging camera are 2048 x 2048 devices with 24 micron pixels, made by Silicon Imaging Technologies, Inc. (SITe) . The same chips are used in the spectrograph cameras as well. The quantum efficiency of these devices is very high, exceeding 80 percent at some wavelengths.

Figure 12. Although the individual CCD chips are not unusual, the ensemble detector configuration most certainly is: the camera contains 30 chips in 6 dewars, with 5 separate filters.

Figure 13. The SDSS filter system (Fukugita et al., AJ, 111, 1748, 1996) is unique to this project. The bands are largely disjoint, so the net result is the equivalent of a very low resolution spectrum, particularly well-suited, for example, for estimation of photometric redshifts of galaxies or discovery of very high redshift QSOs. The bright night-sky lines are also well-isolated in this system.

Figure 14. The 30 CCDs and filters of the mosaic camera, as viewed through the corrector plate.

Figure 15. The assembled camera, upside-down from the normal mounting position on the telescope.

Figure 16. An example of imaging data from the mosaic camera, consisting of a 2.5 by 2.5 degree region of sky, with data from three of the five filters superposed. Two successive, slightly offset scans are required to fill the small gaps created by spaces between the dewars.

Figure 17. One of the two multiobject spectrographs prior to mounting on the telescope. Each spectrograph has a red and blue camera, and receives half (320) of the fibers from the focal plane.

Figure 18. The 640 fibers are positioned at the focal plane on a punched plate, which is drilled from astrometric data from the mosaic camera, using autonomous software to select the targets according to preset criteria for a variety of different scientific projects.

Figure 19. The camera produces a considerable volume of data; both imaging and spectroscopic data are reduced offline at Fermilab.

Figure 20. The large volume of data is remarkable, but perhaps the least important of the scientific contributions of the project. Here we list the author's (strictly personal) view of the strengths of the project, in ascending order of importance.

Figure 21. Work led by Fan, Strauss et al. has led to the detection of a large number of very high redshift QSOs. These color-color plots illustrate how easily such objects separate from the normal stellar locus in the SDSS color system.

Figure 22. At very high redshift, SDSS selects apparently faint but intrinsically very luminous QSOs. Preliminary descriptions of this work have been given by Fan et al., AJ, 118, 1, 1999 and further papers (Fan et al., in preparation). Thus far spectra of the color-selected candidates have been obtained with the ARC 3.5 meter telescope, pending completion of commissioning of the SDSS spectroscopic system.

Figure 23. As of the date of this meeting, SDSS is already responsible for the majority of all known very high redshift QSOs. Figure courtesy of X. Fan.

Figure 24. The very high redshift QSOs have similar spectra, with prominent Ly-alpha emission and the Ly break. Figure courtesy of X. Fan.

Figure 25. Examples of some of the highest redshift QSOs found to date. Figure courtesy of X. Fan.

Figure 26. Unusual high redshift, luminous objects have also been found in the early survey data, including this remarkably high-z BAL QSO at z=4.91, and a featureless object with an absorption line system at z=4.58 (Fan et al., in preparation).

Figure 27. The very reddest end of the SDSS color-color diagrams yields not only the most distant known objects, but also the very nearest. The first of the field T dwarfs, sometimes termed "methane dwarfs", has been found by Strauss et al., ApJ, 522, L61, 1999 . This object, and several others later found by SDSS and 2MASS, have surface temperatures less than 1,000 K and are unequivocally brown dwarfs.

Figure 28. SDSS has made substantial progress on discovery of extremely distant halo stars. Faint halo carbon stars have colors only barely different from the normal stellar locus, but due to the high internal precision of SDSS photometry can still be reliably identified. This object, found by Margon and collaborators, is, at R=19, one of the faintest C stars yet identified, and (presuming giant luminosity) lies at or beyond 100 kpc distance.

Figure 29. The spectrum of this first SDSS C star is, apart from its faintness, perfectly similar to known very bright C giants, as well as the rare dwarf C stars.

Figure 30. We expect SDSS to optically identify huge numbers of ROSAT All Sky Survey X-ray sources, and provide a homogeneous set of spectra for each subclass of identification that is an order of magnitude larger than any previous work. In this early case study by Voges, Margon, and collaborators, a previously unidentified ROSAT X-ray source counterpart (upper left) is easily identified from SDSS imaging data (upper right), and found via a 3.5 meter spectrum (lower panel) to be a previously uncatalogued AGN.

Figure 31, Figure 32. In summary, at its completion SDSS will constitute a survey that fills an unprecedented volume of the Universe, and creates huge, homogeneous catalogs of objects suitable for further study, ranging from the nearest Earth-crossing asteroids to the most distant QSOs.

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. The 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 the Ministry of Education of Japan.