SDSS Preburst Observations of Recent Gamma‐Ray Burst Fields

Prompt long‐wavelength follow‐up observations of gamma‐ray bursts (GRBs) have revolutionized the study of these energetic and enigmatic objects. With the successful launch of recent high‐energy missions such as the High Energy Transient Explorer 2 (HETE‐2; Lamb et al. 2004), the International Gamma‐Ray Astrophysics Laboratory (INTEGRAL; Winkler et al. 2003), and Swift (Gehrels et al. 2004), rapid follow‐up of gamma‐ray bursts has come to maturity. Swift and HETE‐2 not only detect new GRBs, but also provide rapid X‐ray localization that allows for very precise positions (with uncertainties as small as a few arcseconds), far superior than was previously possible. Modern GRB samples have become large enough to study the statistical properties of GRBs (e.g., Berger et al. 2005c), and rapid follow‐up observations have allowed for the first optical, radio, and X‐ray localization of afterglows from short‐hard gamma‐ray bursts (Gehrels et al. 2005; Hjorth et al. 2005; Villasenor et al. 2005; Fox et al. 2005; Pedersen et al. 2005; Prochaska et al. 2006; Covino et al. 2006; Berger et al. 2005b; Bloom et al. 2006).

There has been large‐scale success in identifying afterglows and conducting prompt spectroscopic follow‐up; afterglows extending out to z∼6.3 (Price et al. 2005; Kawai et al. 2005) have been spectroscopically confirmed, providing luminous probes of the universe as far back as the era of reionization. Recently, high‐resolution spectroscopy of GRBs has allowed for detailed studies of absorbing systems arising in the local environment of the GRB progenitor (Berger et al. 2006; Chen et al. 2005). Absorption‐line analyses of GRB afterglow spectra show structure similar to that of quasar damped Lyα absorbers (DLAs) but extending to higher hydrogen column densities and including several metal lines not found in quasar DLA systems (Watson et al. 2006; Chen et al. 2005).

It has been suggested that GRBs can provide a new class of standard candle for future cosmological experiments (Ghirlanda et al. 2004a, 2004b; Friedman & Bloom 2005; Bloom et al. 2003). As the luminosities of these objects should allow detection to very high redshifts (z>10; Lamb & Reichart 2000), studies of GRBs could extend current work using Type Ia supernovae to high redshifts, thus providing considerable constraints on our understanding of cosmology. Before gamma‐ray bursts can be used as a cosmological tool, however, a large number of bursts must be observed spectroscopically in order to calibrate the method.

Two keys to studying GRB afterglows are the identification of a transient afterglow coincident with the burst detected at high energies, and high‐quality information for in‐field photometric calibration stars so that all observations can be placed onto a common photometric system. Currently, many afterglow searches compare new data to digitized Palomar Observatory Sky Survey photometric plates (DSS [Digitized Sky Survey]) and use US Naval Observatory (USNO) stars (Monet et al. 1998) to calibrate both astrometry and photometry. As larger telescopes join the search for afterglows, the DSS imaging quickly becomes insufficient, because these images are too shallow to make detailed comparisons to deep optical imaging; therefore, afterglow candidates can only be identified through their temporal properties. The USNO catalog is an astrometric catalog, but it was not intended to be a source of photometric calibration across the sky; USNO‐calibrated photometry may thus be strongly affected by systematic photometric residuals in the USNO catalog itself.

The Sloan Digital Sky Survey (SDSS; York et al. 2000) provides accurate photometry and astrometry for objects over one‐quarter of the sky, making it a viable alternative to the DSS images and USNO catalogs. The imaging from the SDSS extends over a magnitude deeper than that of the DSS, making it a valuable resource for identifying transient objects, while the stable photometric and astrometric calibration of the survey makes it an ideal source of calibration data in GRB fields. In order to aid the community, we are releasing SDSS imaging and spectroscopy for the fields of 27 Swift‐observed GRBs detected after 2004 December, including bursts originally localized by Swift, HETE‐2, and INTEGRAL observations. Here we offer a bulk release of GRB fields observed in the SDSS, but in the future, we will release similar data through the GRB Coordinates Network (GCN) shortly after the detection of a new burst within the SDSS survey area.

In this paper, we provide some basic documentation relating to the Sloan Digital Sky Survey and describe the data products included in each GRB release. We hope that the information provided here will be a useful primer for those who are unfamiliar with SDSS data; however, this document is by no means intended to be a full description of the survey or survey data products. The layout of the paper is as follows. In § 2, we describe the Sloan Digital Sky Survey itself. A brief description of photometric quantities measured in the SDSS is given in § 3. We describe the data products provided with each GRB release in § 4 and offer some concluding remarks in § 5.

The Sloan Digital Sky Survey (York et al. 2000; Stoughton et al. 2002a; Abazajian et al. 2003, 2004, 2005; Adelman‐McCarthy et al. 2006) is imaging π steradians of the sky through five filters, ugriz (Fukugita et al. 1996; see Fig. 1 and Table 1 for transmission curves and effective wavelengths). The imaging is conducted with a CCD mosaic in drift‐scanning mode (Gunn et al. 1998) on a dedicated 2.5 m telescope (Gunn et al. 2006) located at Apache Point Observatory in New Mexico. Images are processed (Lupton et al. 2001; Stoughton et al. 2002b; Pier et al. 2003; Lupton 2006) and calibrated (Hogg et al. 2001; Smith et al. 2002; Ivezić et al. 2004; Tucker et al. 2006), after which targets are selected for spectroscopy (Eisenstein et al. 2001; Strauss et al. 2002; Richards et al. 2002) with two double spectrographs that are mounted on the same telescope and use a fiber allocation algorithm that ensures highly complete spectroscopic samples (Blanton et al. 2003). The reduced spectra are classified, and redshifts are determined using the idlspec2d automated pipeline (D. Schlegel et al. 2006, in preparation).

Zoom In Zoom Out Reset image size

The photometric calibration of SDSS imaging results in photometry that is nearly on an AB system (Oke 1974). The details of SDSS photometric calibration are beyond the scope of this document (the interested reader can find the calibration procedure in the data‐release papers and technical papers cited above); the bottom line is that the photometric calibration in the SDSS is quite robust, with small random errors. When considering the position of the bright‐star stellar locus in color‐color space, the rms errors are found to be 0.01 mag in the g, r, and i bandpasses, 0.02 mag in the z band, and 0.03 in the u band (Ivezić et al. 2004). Similarly small photometric errors have been demonstrated by measuring the small scatter in the colors of early‐type galaxies on the red sequence (Cool et al. 2006). It should be noted that the data provided in these preburst GRB releases have been processed using a slightly different pipeline than that used for the SDSS public data releases. We cannot guarantee that the values presented here will exactly match those from the SDSS public data releases; in particular, we expect the photometric calibrations to differ on the order of 0.01 mag.

Astrometric calibration within the SDSS is also quite robust. Relative astrometry is generally better than 50 mas per coordinate, while the absolute astrometry is generally better than 0 1 per coordinate (Pier et al. 2003). It should be noted that the SDSS astrometric system can have systematic offsets with respect to other astrometric catalogs. Users requiring high‐accuracy astrometry should check for any such offsets when making comparisons between SDSS astrometry and data calibrated to another astrometric system.

Users of UBVR_CI_C imaging may also find the transformations (both empirically measured and theoretically derived) between the UBVR_CI_C and ugriz photometric systems described in Fukugita et al. (1996) and Smith et al. (2002) to be useful when using the photometry included in our GRB releases to calibrate new observations.

In this section, we outline the definitions of several of the photometric measurements that are used in the SDSS and are included in this data release. This description is by no means an exhaustive source of information. Much more detailed descriptions of the SDSS data products can be found in the data‐release papers and the technical papers referenced in the previous section.

3.1. SDSS Magnitudes and Fluxes

All SDSS magnitudes, including those presented in this and future GRB releases, are expressed in terms of asinh magnitudes (Lupton et al. 1999). At high signal‐to‐noise ratios (S/Ns), asinh magnitudes are identical to the standard logarithmic magnitude (Pogson 1856). However, asinh magnitudes are well behaved even at very low or even negative fluxes, allowing for magnitude calculations even without a formal object detection. As described in Stoughton et al. (2002a), the asinh magnitude for a measured flux f is given by

Here f₀ defines the zero point of the magnitude scale, and the softening parameter b denotes the typical 1 σ noise of the sky in a point‐spread function (PSF) aperture in 1^'' seeing. Table 2, which is a reproduction of Table 21 in Stoughton et al. (2002a), lists the values of b and the magnitude corresponding to a zero‐flux measurement. The table further lists m_c, corresponding to a flux of 10f₀b; asinh and logarithmic magnitudes differ by less than 1% in flux for objects brighter than this limit.

In this and future preburst data releases, we report photometry in flux units and in magnitudes. All flux measurements presented in these releases have units of nanomaggies. A nanomaggie is a flux‐density unit equal to 10⁻⁹ of a magnitude zero source. As SDSS is nearly an AB system, 1 nanomaggie corresponds to 3.631 μJy, or 3.631 × 10^-29 ergs s⁻¹ cm⁻² Hz⁻¹.

3.2. PSF Magnitudes

3.3. Model Magnitudes

Two galaxy models are fit to the two‐dimensional image of each object detected in the SDSS, a pure de Vaucouleurs profile,

(de Vaucouleurs 1948) and a pure exponential profile,

Each of the models is convolved with the local PSF before fitting to the image. The best‐fit model is chosen in the r band; this is the model used to calculate the model quantities for the object. In order to provide meaningful colors, the photometric pipeline fits the full profile in the r‐band image, and the images from the other bands are fitted such that only the amplitude of the profile is allowed to vary (Stoughton et al. 2002a). In the absence of color gradients, model colors provide an unbiased measurement of galaxy colors.

3.4. Petrosian Magnitudes

The Petrosian ratio _P (Petrosian 1976), the ratio of the local surface brightness at radius θ to the average surface brightness at that radius, is given by

where I(θ) is the azimuthally averaged surface brightness profile of a galaxy and α_lo and α_hi are chosen to be 0.85 and 1.25 for SDSS. The Petrosian flux is given by the flux within a circular aperture of 2 θ_P, where θ_P is the radius at which _P falls below 0.2. In the SDSS, θ_P is determined in the r band and is subsequently used in each of the other bands. In the absence of seeing effects, this flux measurement contains a constant fraction of a galaxy's light, independent of its size or distance, and thus provides a fair flux measurement when comparing galaxies of different sizes or at different redshifts. More details on SDSS Petrosian magnitudes can be found in Blanton et al. (2001), Strauss et al. (2002), and Stoughton et al. (2002a).

3.5. Flags

In this section, we describe each of the data products provided in this data release. We have removed saturated stars from all of the data tables and have not corrected the photometry for Galactic extinction. Again, magnitudes are asinh magnitudes, as is standard in the SDSS.

4.1. sdss.calstar

For each burst, we report the photometry and astrometry of bright stars (r<20.5) within 15^' of the burst location, in files labeled sdss.calstar. These stars provide a reliable basis for both the astrometric and photometric calibration of GRB follow‐up imaging. The area covered by these calibration stars is well matched to the Swift X‐Ray Telescope field of view (Gehrels et al. 2004). Table 3 lists the information provided for each of the calibrations stars in the sdss.calstar file for each GRB data set. It is important that the user consult the object flags and photometric errors when utilizing stars from this file, as some of the stars are poorly detected in the u band.

4.2. sdss.objects

In the sdss.objects_flux and sdss.objects_magnitudes files, we provide photometry and astrometry of all unsaturated objects with r<23.0 within 6^' of the GRB location. As suggested by the file names, photometric quantities in the sdss.objects_flux file use flux units (nanomaggies), while those in the sdss.objects_magnitudes file use magnitudes (asinh magnitudes). Table 4 lists the quantities reported in these files.

We recommend using model fluxes when calculating colors of galaxies with r>19.0. We generally prefer Petrosian fluxes over model fluxes for the overall flux of a galaxy, but Petrosian fluxes are noisier than model quantities at faint flux levels. We warn the user to exercise caution when using photometry with quoted flux errors larger than 20%. These objects are generally only low‐S/N detections, but sometimes the large errors indicate complications in the reductions.

4.3. sdss.spectro

If SDSS spectroscopy has been completed in the field of the GRB, the sdss.spectro file will include redshifts of spectroscopically observed objects within 6^' of the GRB location. Table 5 documents the parameters reported in the sdss.spectro files. For stars and galaxies, redshifts are accurate to approximately 30 km s⁻¹, while quasar redshifts, which are of less importance here, are measured to Δz = 0.001 (Stoughton et al. 2002a). All redshifts have been corrected to a heliocentric frame but are not corrected for Galactic rotation.

Each spectroscopically observed object is classified by a spectral class (STAR, GALAXY, or QSO) and a spectral subclass if appropriate. Galaxies can be subclassified as STARFORMING, AGN, or STARBURST, depending on the strength of the emission lines observed in the spectrum. Both galaxies and quasars can be subclassified as BROADLINE, based on the width of the emission‐line features; stars are subclassified by spectral type.

4.4. Images

For an 8^' × 8^' region around each burst, we include gzipped FITS images in each of the SDSS passbands, as well as three‐color composite (gri) images of the field in JPEG format. The FITS images are in units of nanomaggies per pixel, where a pixel is 0 396 on a side. Each of the images is oriented with north up and east to the left, and the FITS headers include the relevant World Coordinate System information. The SDSS images provide an important comparison epoch for follow‐up imaging in order to locate possible gamma‐ray burst afterglows. To illustrate the power of using SDSS images compared to those from the Digitized Sky Survey, Figure 2 shows a 1^' × 1^' region around the field of GRB 041224 as imaged in SDSS and the DSS‐2 red plates. Both images are centered on the position of the GRB as reported by the Swift Burst Alert Telescope (BAT) detection (Barthelmy et al. 2004). The imaging from the SDSS reaches over a magnitude deeper than the DSS imaging, allowing for the identification of transient objects to fainter limits than are possible using the DSS.

Zoom In Zoom Out Reset image size

In this paper, we present SDSS observations of all Swift‐observed gamma‐ray bursts occurring after 2004 December that lie in the SDSS imaging footprint. For convenience, we include not only data for burst fields with photometry included in previous SDSS data releases, but also bursts for which the data have not yet been released by the SDSS. Table 6 lists the bursts included in this release, in addition to the current status of the data and the number of calibration stars, surrounding objects, and spectroscopically observed objects located in each GRB field.

All of the data included in this release, as well as data we will release in the future, can be found on the online SDSS GRB release page.⁷ In addition to all of the data products described in the previous section, we also include a short text file for each burst. This short description provides a brief introduction to the data products and reports the local mean galactic extinction for each GRB field. In the future, we will release SDSS imaging and spectroscopy of individual GRB fields shortly after the bursts are localized. These future data releases will be announced via a GCN Circular and, unless otherwise noted, follow the same data release model described in this document. All data from this release and future preburst data releases may be used freely, but we request that both the most recent SDSS data release paper (currently Adelman‐McCarthy et al. 2006) and this paper (for bursts included in the current release) or the GCN Circular (for bursts released in the future) be cited.

R. J. C. is funded through a National Science Foundation Graduate Student Fellowship. This research has made use of NASA's Astrophysics Data System bibliographic services. The Second Palomar Observatory Sky Survey (POSS‐II) was made by the California Institute of Technology with funds from the National Science Foundation, the National Aeronautics and Space Administration, the National Geographic Society, the Sloan Foundation, the Samuel Oschin Foundation, and the Eastman Kodak Corporation.

Funding for the SDSS and SDSS‐II has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, the US Department of Energy, the NASA, the Japanese Monbukagakusho, the Max Planck Society, and the Higher Education Funding Council for England. The SDSS Web Site is http://www.sdss.org.

The SDSS is managed by the Astrophysical Research Consortium for the Participating Institutions. The Participating Institutions are the American Museum of Natural History, Astrophysical Institute Potsdam, University of Basel, Cambridge University, Case Western Reserve University, University of Chicago, Drexel University, Fermilab, the Institute for Advanced Study, the Japan Participation Group, Johns Hopkins University, the Joint Institute for Nuclear Astrophysics, the Kavli Institute for Particle Astrophysics and Cosmology, the Korean Scientist Group, the Chinese Academy of Sciences (LAMOST), Los Alamos National Laboratory, the Max‐Planck‐Institute for Astronomy (MPA), the Max‐Planck‐Institute for Astrophysics (MPIA), New Mexico State University, Ohio State University, University of Pittsburgh, University of Portsmouth, Princeton University, the United States Naval Observatory, and the University of Washington.