Browse
this table...

SDSSNBCKDE - SDSS NBCKDE Catalog of Photometrically Selected Quasar Candidates

HEASARC
Archive

Overview

This table contains a catalog of 1,015,082 quasar candidates selected from the photometric imaging data of the Sloan Digital Sky Survey (SDSS) using a non-parametric Bayesian classification kernel density estimator (NBC-KDE). It excludes 157,075 initial candidates that were culled as known or likely contaminants. The objects are all point sources to a limiting magnitude of i = 21.3 from 8417 deg2 of imaging from SDSS Data Release 6 (DR6). This sample extends the previous catalog (Paper I: Richards et al. 2004, ApJS, 155, 257) by using the latest SDSS public release data and probing both ultraviolet (UV)-excess and high-redshift quasars. While the addition of high-redshift candidates reduces the overall efficiency (quasars:quasar candidates) of the catalog to ~80%, it is expected to contain no fewer than 850,000 bona fide quasars, which is ~8 times the number of the previous sample and ~10 times the size of the largest spectroscopic quasar catalog. Cross-matching between this photometric catalog and spectroscopic quasar catalogs from both the SDSS and 2dF survey yields 88,879 spectroscopically confirmed quasars. For judicious selection of the most robust UV-excess sources (~500,000 objects in all), the efficiency is nearly 97 - more than sufficient for detailed statistical analyses. The catalog's completeness to type 1 (broad-line) quasars is expected to be no worse than 70%, with most missing objects occurring at z < 0.7 and 2.5 < z < 3.0. In addition to classification information, the authors provide photometric redshift estimates (typically good to Delta(z) +/- 0.3 [2-sigma]) and cross-matching with radio, X-ray, and proper-motion catalogs. Finally, the authors have considered the catalog's utility for determining the optical luminosity function of quasars and are able to confirm the flattening of the bright-end slope of the quasar luminosity function at z ~ 4 as compared to z ~ 2.

Much more information on the SDSS is available at the project's web site at http://www.sdss.org/.


Catalog Bibcode

2009ApJS..180...67R

References

Efficient photometric selection of quasars from the Sloan Digital Sky Survey.
II. ~1,000,000 quasars from data release six.
    Richards G.T., Myers A.D., Gray A.G., Riegel R.N., Nichol R.C.,
    Brunner R.J., Szalay A.S., Schneider D.P., Anderson S.F.,
   <Astrophys. J. Suppl. Ser., 180, 67-83 (2009)>
   =2009ApJS..180...67R

Efficient photometric selection of quasars from the Sloan Digital Sky Survey:
100000 z<3 quasars from data release one. [Paper I]
    Richards G.T., Nichol R.C., Gray A.G., Brunner R.J., Lupton R.H.,
    Vanden Berk D.E., Chong S.S., Weinstein M.A., Schneider D.P.,
    Anderson S.F., Munn J.A., Harris H.C., Strauss M.A., Fan X., Gunn J.E.,
    Ivezic Z., York D.G., Brinkmann J., Moore A.W.
   <Astrophys. J. Suppl. Ser., 155, 257-269 (2004)>
   =2004ApJS..155..257R

Provenance

This table was created by the HEASARC based on an electronic version of Table 1 in the reference paper which was obtained from the ApJ web site.

Parameters

Source_Number
A unique catalog number for each source in order of increasing J2000.0 Right Ascension.

Name
The standard SDSS designation of the object, in the format 'SDSS Jhhmmss.ss+ddmmss.s' using the sexagesimal J2000.0 coordinates.

RA
The Right Ascension of the quasar candidate in the selected equinox. This was given in J2000.0 decimal degrees to a precision of 0.1 microdegrees (0.36 milliarcseconds) in the original table

Dec
The Declination of the quasar candidate in the selected equinox. This was given in J2000.0 decimal degrees to a precision of 0.1 microdegrees (0.36 milliarcseconds) in the original table

LII
The Galactic Longitude of the quasar candidate.

BII
The Galactic Latitude of the quasar candidate.

SDSS_Object_ID
The SDSS object identification of the object.

Redshift
The photometric redshift of the quasar candidate: see Weinstein et al. (2004, ApJS, 155, 243) for more details.

Redshift_Lower_Limit
The lower limit to the photometric redshift range.

Redshift_Upper_Limit
The upper limit to the photometric redshift range.

Redshift_Probability
The photometric redshift range probability, zphotprob. See Weinstein et al. (2004, ApJS, 155, 243) for more details.

Umag
The u-band PSF ubercalibrated asinh magnitude (corrected for Galactic extinction).

Gmag
The g-band PSF ubercalibrated asinh magnitude (corrected for Galactic extinction).

Rmag
The r-band PSF ubercalibrated asinh magnitude (corrected for Galactic extinction).

Imag
The i-band PSF ubercalibrated asinh magnitude (corrected for Galactic extinction).

Zmag
The z-band PSF ubercalibrated asinh magnitude (corrected for Galactic extinction).

Umag_Error
The error in the u-band PSF ubercalibrated asinh magnitude.

Gmag_Error
The error in the g-band PSF ubercalibrated asinh magnitude.

Rmag_Error
The error in the r-band PSF ubercalibrated asinh magnitude.

Imag_Error
The error in the i-band PSF ubercalibrated asinh magnitude.

Zmag_Error
The error in the z-band PSF ubercalibrated asinh magnitude.

Reddening
The (B-V) differential extinction, E(B-V), in magnitudes. Note that the extinctions in the SDSS bands are related to this quantity by the relations A_u/A_g/A_r/A_i/A_z = 5.155/3.793/2.751/2.086/1.479 x E(B-V), respectively.

Concentration
The concentration parameter c for star/galaxy separation, being equal to (PSFMag_i - modelMag_i).

Flux_20_cm
The 20-cm flux density, in mJy, of the radio counterpart to the catalog object. Objects wich were not detected or not covered in the radio have been given null values.

RASS_Count_Rate
The ROSAT All-Sky Survey (RASS) broadband (0.1 - 2.4 keV) vignetting-corrected count rate, in ct/s, of the X-ray counterpart to the catalog object. Objects wich were not detected or not covered in the radio have been given null values.

Proper_Motion
The proper motion of the catalog object, in mas/yr, based on USNO-B+SDSS proper-motion information as tabulated in the SDSS database.

Moving_Flag
This parameter is a flag which is set to 1 to indicate a likely moving object.

Qsots_Flag
The qsots flag for the catalog object. The initial classification used a stellar prior of 0.95 (i.e., ~95% of objects in the test set are expected to be stars). These objects are flagged in the catalog with qsots = 1 (see Section 4).

Lowzts_Flag
The lowzts flag for the catalog object. The authors have also classified all of the objects in the test set after restricting the quasar training set to three narrower redshift ranges (moving the quasars outside of these ranges to the "stars" training set). They classified objects as low redshift (z <= 2.2), mid redshift (2.2 < z < 3.5), and high redshift (z >= 3.5). The rationale for this process is that the distribution of quasar colors considerably changes with redshift, sometimes being more consistent with the stellar locus than others. Thus, subclassification by redshift can improve the robustness of the sample. The priors for these subsamples were set to a somewhat more conservative value of 0.98 rather than 0.95. The bandwidth optimizing algorithm was also rerun for these subclassifications, and the paired (star, quasar) bandwidth values were (0.16, 0.13), (0.12, 0.12), and (0.185, 0.195) for low-z, mid-z, and high-z, respectively, as compared to (0.11, 0.12) for the full sample. Small changes (of order of the range quoted here) in these values would have relatively little impact on the authors' results. The redshift-dependent selected entries in the catalog are flagged as lowzts = 1, midzts = 1, and highzts = 1, respectively.

Midzts_Flag
The midzts flag for the catalog object. The authors have also classified all of the objects in the test set after restricting the quasar training set to three narrower redshift ranges (moving the quasars outside of these ranges to the "stars" training set). They classified objects as low redshift (z <= 2.2), mid redshift (2.2 < z < 3.5), and high redshift (z >= 3.5). The rationale for this process is that the distribution of quasar colors considerably changes with redshift, sometimes being more consistent with the stellar locus than others. Thus, subclassification by redshift can improve the robustness of the sample. The priors for these subsamples were set to a somewhat more conservative value of 0.98 rather than 0.95. The bandwidth optimizing algorithm was also rerun for these subclassifications, and the paired (star, quasar) bandwidth values were (0.16, 0.13), (0.12, 0.12), and (0.185, 0.195) for low-z, mid-z, and high-z, respectively, as compared to (0.11, 0.12) for the full sample. Small changes (of order of the range quoted here) in these values would have relatively little impact on the authors' results. The redshift-dependent selected entries in the catalog are flagged as lowzts = 1, midzts = 1, and highzts = 1, respectively.

Highzts_Flag
The highzts flag for the catalog object. The authors have also classified all of the objects in the test set after restricting the quasar training set to three narrower redshift ranges (moving the quasars outside of these ranges to the "stars" training set). They classified objects as low redshift (z <= 2.2), mid redshift (2.2 < z < 3.5), and high redshift (z >= 3.5). The rationale for this process is that the distribution of quasar colors considerably changes with redshift, sometimes being more consistent with the stellar locus than others. Thus, subclassification by redshift can improve the robustness of the sample. The priors for these subsamples were set to a somewhat more conservative value of 0.98 rather than 0.95. The bandwidth optimizing algorithm was also rerun for these subclassifications, and the paired (star, quasar) bandwidth values were (0.16, 0.13), (0.12, 0.12), and (0.185, 0.195) for low-z, mid-z, and high-z, respectively, as compared to (0.11, 0.12) for the full sample. Small changes (of order of the range quoted here) in these values would have relatively little impact on the authors' results. The redshift-dependent selected entries in the catalog are flagged as lowzts = 1, midzts = 1, and highzts = 1, respectively.

Uvxts_Flag
The uvxts flag for the catalog object. For backwards compatibility with the catalog from Paper I (and their unpublished DR3 and DR4 catalogs), the authors have also provided a flag that indicates whether each object would be selected by that UV-excess algorithm as well. See Paper I for more details on this selection procedure. These entries in the catalog are flagged as uvxts = 1.

Log_Quasar_Density
The qsodens parameter for the catalog object, being the log of the KDE (kernel density estimate) quasar density, i.e., a measure of the probability that the object is a quasar.

Log_Star_Density
The stardens parameter for the catalog object, being the log of the KDE (kernel density estimate) star density, i.e., a measure of the probability that the object is a star.

Quality_Flag
The 'good' quality flag for the catalog object. This is meant to be indicative of how likely the authors feel that the object is truly a quasar. It has integer values that span the range from -6 to +6 in the full (pre-culled) set of quasar candidates. More positive values indicate greater confidence in the quasar classification, and the authors generally recommend using objects with values >= 0 for statistical analysis (with the possible exception of mid- and high-z candidates). As such, objects with values less than zero and/or that are known contaminants have been 'culled' and are not included in this table. The rejected candidates are listed in Table 3 of the reference paper.

Previous_Classification
A previous type classification for the object (or 'U' if there is none), including the reference source from wich it was obtained. Each object in the catalog was cross-matched to the DR5 quasar catalog (Schneider et al. 2007, AJ, 134, 102), the 2QZ quasar catalog (Croom et al. 2004, MNRAS, 349, 1397), the SDSS-2dF LRG and QSO Survey (2SLAQ) Early Data Release quasar catalog (Croom et al. 2008, MNRAS, in press), and the SDSS-DR6 spectroscopic database (Adelman-McCarthy et al. 2008, ApJS, 175, 297). The matching was done in the above order. Once a match was found, no further matches were allowed for that object as this hierarchy represents the most effective path to robust identifications. Objects from the DR6 spectroscopic database were required to have a high confidence zStatus flag. See Section 4.1 of the reference paper for more details.

Previous_Redshift
A previous redshift for the catalog object from the reference source named in the previous_classification parameter.


Contact Person

Questions regarding the SDSSNBCKDE database table can be addressed to the HEASARC User Hotline.

Page Author: Browse Software Development Team
Last Modified: 27-Apr-2009