Browse
this table...

CHANDFS4MS - Chandra Deep Field South 4-Megasecond Catalog

HEASARC
Archive

Overview

This table contains the main Chandra source catalog for the 4 megasecond (Ms) Chandra Deep Field-South (CDF-S), which is the deepest Chandra survey to date and covers an area of 464.5 arcmin2. It contains 740 X-ray sources that are detected with wavdetect at a false-positive probability threshold of 10-5 in at least one of three X-ray bands (0.5-8 keV, full band; 0.5-2 keV, soft band; and 2-8 keV, hard band) and also satisfy a binomial-probability source-selection criterion of P < 0.004 (i.e., the probability of sources not being real is less than 0.004); this approach is designed to maximize the number of reliable sources detected. A total of 300 main-catalog sources are new compared to the previous 2 Ms CDF-S main-catalog (the HEASARC CHANDFS2MS table) sources. The authors determined X-ray source positions using centroid and matched-filter techniques and obtained a median positional uncertainty of ~0.42 arcseconds. In their paper, they also provided a supplementary catalog (not included in this HEASARC table), which consists of 36 sources that are detected with wavdetect at a false-positive probability threshold of 10-5, satisfy the condition of 0.004 < P < 0.1, and have an optical counterpart with R < 24. Multiwavelength identifications, basic optical/infrared/radio photometry, and spectroscopic/photometric redshifts are provided for the X-ray sources in the main and supplementary catalogs. Seven hundred sixteen (~97%) of the 740 main-catalog sources have multiwavelength counterparts, with 673 (~94% of 716) having either spectroscopic or photometric redshifts. The 740 main-catalog sources span broad ranges of full-band flux and 0.5-8 keV luminosity; the 300 new main-catalog sources span similar ranges although they tend to be systematically lower.

Basic analyses of the X-ray and multiwavelength properties of the sources indicate that >75% of the main-catalog sources are active galactic nuclei (AGNs); of the 300 new main-catalog sources, about 35% are likely normal and starburst galaxies, reflecting the rise of normal and starburst galaxies at the very faint flux levels uniquely accessible to the 4 Ms CDF-S. Near the center of the 4 Ms CDF-S (i.e., within an off-axis angle of 3'), the observed AGN and galaxy source densities have reached 9800 (+1300,-1100) deg-2 and 6900 (+1100,-900) deg-2, respectively. Simulations show that the main catalog is highly reliable and is reasonably complete. The mean backgrounds (corrected for vignetting and exposure-time variations) are 0.063 and 0.178 counts Ms-1 pixel-1 (for a pixel size of 0.492 arcseconds) for the soft and hard bands, respectively; the majority of the pixels have zero background counts. The 4 Ms CDF-S reaches on-axis flux limits of ~3.2 x 10-17, 9.1 x 10-18, and 5.5 x 10-17 erg cm-2 s-1 for the full, soft, and hard bands, respectively. An increase in the CDF-S exposure time by a factor of ~2-2.5 would provide further significant gains and probe key unexplored discovery space.

This HEASARC table comprises Table 3 from the reference paper, the Main Chandra Source Catalog of 740 X-ray sources. The 36 optically bright Chandra sources that were listed in Table 6 of the reference paper are thus not included herein.


Catalog Bibcode

2011ApJS..195...10X

References

The Chandra Deep Field-South Survey: 4 Ms Source Catalogs
    Y.Q. Xue, B. Luo, W.N. Brandt, (22 more co-authors)
    <Astrophys. J. Suppl. 195, 10 (2011)>
    =2011ApJS..195...10X

Provenance

This table was created by the HEASARC in July 2011 based on an electronic version of Table 3 from the reference paper which was obtained from the ApJS web site.

HEASARC Implementation

In this representation, the HEASARC has replaced values of -1.00 which were used in the machine-readable version of table 3 obtained from the ApJS web site to indicate the absence of valid data values for the following parameters by nulls: all magnitudes and redshifts, the errors in the band counts (fb_counts_neg_err, fb_counts_pos_err, hb_counts_neg_err, hb_counts_pos_err, sb_counts_neg_err, sb_counts_pos_err), the band ratio, the offset, and the rest-frame X-ray luminsoity (rf_lx).

Parameters

XID_Source_Number
A unique identification source number, i.e., the XID number; sources are listed in order of increasing Right Ascension.

Alt_Name
An alternative source designation in the style recommended by the Dictionary of Nomenclature of Celestial Objects, which was created by the HEASARC using the prefix '[XLB2011]' for Xue, Luo, Brandt (2011) and the XID source number.

Name
The primary source designation in the style recommended by the Dictionary of Nomenclature of Celestial Objects, which was created by the HEASARC using the prefix 'CXOCDFS' for Chandra X-ray Observatory Chandra Deep Field South and the J2000.0 Right Ascension and Declination truncated to 0.1 seconds of time in RA and 1 arcsecond in Declination.

RA
The Right Ascension of the Chandra X-ray source in the selected equinox. This was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors determined the X-ray source positions following the procedure detailed in Section 3.2 of the reference paper.

Dec
The Declination of the Chandra X-ray source in the selected equinox. This was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors determined the X-ray source positions following the procedure detailed in Section 3.2 of the reference paper.

LII
The Galactic Longitude of the Chandra X-ray source.

BII
THe Galactic Latitude of the Chandra X-ray source.

Log_Min_Ns_Prob
The minimum value of log P for the Chandra X-ray source, where P is the AE-computed binomial no-source probability, among the three standard bands (full-band from 0.5 to 8 keV, soft-band from 0.5 to 2 keV and hard-band from 2 to 8 keV), More negative values of log P indicate a more significant source detection. The authors set log P = -99.0 for sources with P = 0. For the main-catalog sources, the median value of log P is -8.9 (note that P < 0.004, corresponding to log P < -2.4, is the condition for a source to be included in the main catalog). There are 493, 57, 69, and 121 sources with minimum wavdetect probabilities of 10-8, 10-7, 10-6, and 10-5, respectively (see footnote 38 and Figure 5 of the reference paper).

Log_Min_FP_Prob
The logarithm of the minimum wavdetect false-positive probability detection threshold for the Chandra X-ray source. More negative values of this parameter indicate a more significant source detection.

Error_Radius
The ~68% confidence level X-ray positional uncertainty of the Chandra X-ray source, in arcseconds, computed using Equation (2) from the reference paper. This is dependent on both the off-axis angle and the aperture-corrected net source counts. The ~68% confidence-level X-ray positional uncertainty was used in the likelihood-ratio matching procedure (see Section 4.3 of the reference paper). The positional uncertainty for the main-catalog sources ranges from 0.10 to 1.51 arcseconds, with a median value of 0.42 arcseconds.

Off_Axis
The off-axis angle of the X-ray source, in arcminutes, i.e., the angular separation between the X-ray source and the CDF-S average aim point (given in Table 1 of the reference paper as [RA, Dec (J2000.0)] = [03 32 28.06, -27 48 26.4]). The off-axis angle for the main-catalog sources ranges from 0.33 to 12.36 arcminutes, with a median value of 5.82 arcminutes. The maximum off-axis angle of 12.36 arcminutesis slightly larger than a half of the diagonal size of the ACIS-I field of view (11.95 arcminutes), due to the fact that the CDF-S observations have varying aim points and roll angles, as shown in Table 1 of the reference paper.

FB_Counts
The aperture-corrected net (i.e., background-subtracted) source counts in the full 0.5 - 8 keV band. The photometry was calculated by AE using the position given in the ra and dec parameters for all bands and following the procedure described in Section 3.2 of the reference paper, and was not corrected for vignetting or exposure-time variations. To be consistent with their source-detection criterion (i.e., P < 0.004), the authors considered a source to be "detected" for photometry purposes in a given band only if the AE-computed binomial no-source probability for that band was less than 0.004. For sources not detected in a given band, they calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

FB_Counts_Pos_Err
The upper uncertainty in the full-band counts in the X-ray source. For sources not detected in a given band, the authors calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

FB_Counts_Neg_Err
The lower uncertainty in the full-band counts in the X-ray source. For sources not detected in a given band, the authors calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

SB_Counts
The aperture-corrected net (i.e., background-subtracted) source counts in the soft 0.5 - 2 keV band. The photometry was calculated by AE using the position given in the ra and dec parameters for all bands and following the procedure described in Section 3.2 of the reference paper, and was not corrected for vignetting or exposure-time variations. To be consistent with their source-detection criterion (i.e., P < 0.004), the authors considered a source to be "detected" for photometry purposes in a given band only if the AE-computed binomial no-source probability for that band was less than 0.004. For sources not detected in a given band, they calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

SB_Counts_Pos_Err
The upper uncertainty in the soft-band counts in the X-ray source. For sources not detected in a given band, the authors calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

SB_Counts_Neg_Err
The lower uncertainty in the soft-band counts in the X-ray source. For sources not detected in a given band, the authors calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

HB_Counts
The aperture-corrected net (i.e., background-subtracted) source counts in the hard 2-8 keV band. The photometry was calculated by AE using the position given in the ra and dec parameters for all bands and following the procedure described in Section 3.2 of the reference paper, and was not corrected for vignetting or exposure-time variations. To be consistent with their source-detection criterion (i.e., P < 0.004), the authors considered a source to be "detected" for photometry purposes in a given band only if the AE-computed binomial no-source probability for that band was less than 0.004. For sources not detected in a given band, they calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

HB_Counts_Pos_Err
The upper uncertainty in the hard-band counts in the X-ray source. For sources not detected in a given band, the authors calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

HB_Counts_Neg_Err
The lower uncertainty in the hard-band counts in the X-ray source. For sources not detected in a given band, the authors calculated upper limits and placed -1.00 values in the corresponding error parameters, which the HEASARC has represented by null values in this table. When the total number of counts within the polygonal extraction region of an undetected source was <= 10, the authors computed the upper limit using the Bayesian method of Kraft et al. (1991, ApJ, 364, 344) for a 99% confidence level; otherwise, they computed the upper limit at the 3-sigma level for Poisson statistics (Gehrels 1986, ApJ, 306, 336).

Extent_Flag
This parameter contains a flag indicating whether a source shows any evidence for spatial extent in basic testing. In Section 3.2 of the reference paper, the authors ran wavdetect using nine wavelet scales up to 16 pixels, which potentially allows detection of sources that are extended on such scales. They utilized the following procedure to assess extent. They first derived a set of cumulative encircled energy fractions (EEFs) by extracting the PSF power within a series of circular apertures (centered at the source position) up to a 90% EEF radius from the merged PSF image. They then derived another set of cumulative EEFs by extracting source counts within a series of circular apertures (also centered at the source position) up to the same 90% EEF radius from the merged source image. Finally, they used a Kolmogorov-Smirnov (K-S) test suitable for two distributions to compute the probability (phoKS) that the two sets of cumulative EEFs were consistent with each other. Of the 740 main-catalog sources, 7 have rhoKS <= 0.01, i.e., the merged PSF and source images are inconsistent with each other at or above a 99% confidence level, and have the value of this parameter set to 2; 24 have 0.01 < rhoKS <= 0.05 and have the value of this parameter set to 1; all the remaining sources have the value of this column set to 0. See page 12 of the reference paper for more details.

Onir_RA
The Right Ascension of the optical/near-infrared/infrared/radio (ONIR) counterpart in the selected equinox (see Section 4.3 of the reference paper for the details of the multiwavelength identifications): this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. Sources without multiwavelength identifications have had their Right Ascension and Declination values set to null values.

Onir_Dec
The Declination of the optical/near-infrared/infrared/radio (ONIR) counterpart in the selected equinox (see Section 4.3 of the reference paper for the details of the multiwavelength identifications): this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. Sources without multiwavelength identifications have had their Right Ascension and Declination values set to null values.

Offset
The measured offset between the X-ray source and its ONIR counterpart, in arcseconds. Sources without multiwavelength identifications have a value set to null.

Crtpart_AB_Mag
The AB magnitude m_AB_of the ONIR counterpart, measured in the counterpart-detection band. The AB magnitudes for the radio counterparts wereconverted from the radio flux densities fnu using mAB = -2.5 * log(fnu) - 48.60. Sources without counterparts have a value set to null.

Ctrpart_Catalog
The name of the ONIR catalog (i.e., VLA, GOODS-S, GEMS, MUSIC, WFI, MUSYC, or SIMPLE) from which the primary counterpart has been taken. Sources without counterparts have this parameter set to blank values.

Wfi_RA
The Right Ascension of the WFI R-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Wfi_Dec
The Declination of the WFI R-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Rmag
The R-band AB magnitude of the counterpart to the X-ray source in the WFI catalog. Sources without counterparts have their magnitudes set to nulls.

Goodss_RA
The Right Ascension of the GOODS-S z-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Goodss_Dec
The Declination of the GOODS-S z-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Goodss_Zmag
The z-band AB magnitude of the counterpart to the X-ray source in the GOODS-S catalog. Sources without counterparts have their magnitudes set to nulls.

Gems_RA
The Right Ascension of the GEMS z-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Gems_Dec
The Declination of the GEMS z-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Gems_Zmag
The z-band AB magnitude of the counterpart to the X-ray source in the GEMS catalog. Sources without counterparts have their magnitudes set to nulls.

Music_RA
The Right Ascension of the MUSIC K-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Music_Dec
The Declination of the MUSIC K-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Music_Kmag
The K-band AB magnitude of the counterpart to the X-ray source in the MUSIC catalog. Sources without counterparts have their magnitudes set to nulls.

Musyc_RA
The Right Ascension of the MUSYC K-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Musyc_Dec
The Declination of the MUSYC K-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Musyc_Kmag
The K-band AB magnitude of the counterpart to the X-ray source in the MUSYC catalog. Sources without counterparts have their magnitudes set to nulls.

Simple_RA
The Right Ascension of the SIMPLE 3.6-um-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Simple_Dec
The Declination of the SIMPLE 3.6-um-band counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Simple_3p6_um_Mag
The 3.6-micron-band AB magnitude of the counterpart to the X-ray source in the SIMPLE catalog. Sources without counterparts have their magnitudes set to nulls.

Vla_RA
The Right Ascension of the VLA 1.4-GHz radio counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Vla_Dec
The Declination of the VLA 1.4-GHz radio counterpart to the X-ray source in the selected equinox: this was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. The authors cross-matched the positions of their primary ONIR counterparts (i.e., onir_ra and onir_dec) with the seven ONIR catalogs using likelihood-ratio matching. Sources without counterparts have their corresponding right ascension and declination values set to null values and their AB magnitudes also set to nulls. The authors find ~ 75%, 61%, 72%, 55%, 70%, 88%, and 18% of the main-catalog X-ray sources have WFI, GOODS-S, GEMS, MUSIC, MUSYC, SIMPLE, and VLA counterparts, respectively, with a false-match probability of <2% for each ONIR catalog (see footnote 43 and Section 4.3 of the reference paper for more details).

Vla_20_cm_Mag
The 1.4 GHz AB magnitude of the counterpart to the X-ray obtained by the VLA. The AB magnitudes for the radio counterparts wereconverted from the radio flux densities fnu using mAB = -2.5 * log(fnu) - 48.60. Sources without counterparts have their magnitudes set to nulls.

Spect_Redshift
The spectroscopic redshift. Spectroscopic redshifts were collected from Le Fevre et al. (2004, A&A, 428, 1043), Szokoly et al. (2004, ApJS, 155, 271), Zheng et al. (2004, ApJS, 155, 73), Mignoli et al. (2005, A&A, 437, 883), Ravikumar et al. (2007, A&A, 465, 1099), Vanzella et al. (2008, A&A, 478, 83), Popesso et al. (2009, A&A, 494, 443), Treister et al. (2009, ApJ, 693, 1713: these redshifts were flagged as 'Insecure' since the authors did not provide redshift quality flags), Balestra et al. (2010, A&A, 512, 12), and Silverman et al. (2010, ApJS, 191, 124) with the reference numbers of 1-10 in the ref_spect_redshift parameter value, respectively. The authors cross-matched the positions of primary ONIR counterparts with the above catalogs of spectroscopic redshifts using a matching radius of 0.5 arcsecs. Of the 716 main-catalog sources that have multiwavelength identifications, 419 (58.5%) have spectroscopic redshift measurements. Three hundred forty-three (81.9%) of these 419 spectroscopic redshifts are secure, i.e., they are measured at >=95% confidence levels with multiple secure spectral features (flagged as 'Secure' in the value of their spect_redshift_flag parameter value); 76 (18.1%) of these 419 spectroscopic redshifts are insecure (flagged as 'Insecure'). The authors estimated the false-match probability to be <~1% in all cases. Sources without spectroscopic redshifts have been set to null values.

Spect_Redshift_Flag
This parameter contains a quality flag for the spectroscopic redshift. Of the 716 main-catalog sources that have multiwavelength identifications, 419 (58.5%) have spectroscopic redshift measurements. Three hundred forty-three (81.9%) of these 419 spectroscopic redshifts are secure, i.e., they are measured at >=95% confidence levels with multiple secure spectral features (flagged as 'Secure' in the value of their spect_redshift_flag parameter value); 76 (18.1%) of these 419 spectroscopic redshifts are insecure (flagged as 'Insecure'). Sources without spectroscopic redshifts have this parameter set to 'None'.

Ref_Spect_Redshift
The reference for the spectroscopic redshift, coded as follows:

  Value       Reference

  -1          No available spectroscopic redshift,
   1          Le Fevre et al. (2004, A&A, 428, 1043),
   2          Szokoly et al. (2004, ApJS, 155, 271),
   3          Zheng et al. (2004, ApJS, 155, 73),
   4          Mignoli et al. (2005, A&A, 437, 883),
   5          Ravikumar et al. (2007, A&A, 465, 1099),
   6          Vanzella et al. (2008, A&A, 478, 83),
   7          Popesso et al. (2009, A&A, 494, 443)
   8          Treister et al. (2009, ApJ, 693, 1713),
   9          Balestra et al. (2010, A&A, 512, 12),
  10         Silverman et al. (2010, ApJS, 191, 124).
  

Phot_Redshift_1
The photometric redshift for the source taken from Luo et al. (2010, ApJS, 187, 560). Sources with no such information have null values for this parameter. The photometric redshift catalogs were chosen because they utilized extensive multiwavelength photometric data and produced accurate photometric redshifts. Luo et al. (2010) derived high quality photometric redshifts for the 462 Luo et al. (2008, ApJS, 179, 19) main catalog X-ray sources with a treatment of photometry that included utilizing likelihood matching, manual source deblending, and appropriate upper limits.

Phot_Redshift_1_Min
The 1-sigma lower bound on the Luo et al. (2010) photometric redshift for the source. Sources with no such information have null values for this parameter. See Footnote 45 of the reference paper for a caveat about the reliability of this parameter.

Phot_Redshift_1_Max
The 1-sigma upper bound on the Luo et al. (2010) photometric redshift for the source. Sources with no such information have null values for this parameter. See Footnote 45 of the reference paper for a caveat about the reliability of this parameter.

Phot_Redshift_2
The alternative photometric redshift for the source given in Luo et al. (2010, ApJS, 187, 560). Sources with no such information have null values for this parameter.

Phot_Redshift_3
The photometric redshift for the source taken from Cardamone et al. (2010, ApJS, 189, 270). Sources with no such information have null values for this parameter. Cardamone et al. (2010) employed new medium-band Subaru photometry and a PSF-matching technique to create a uniform photometric catalog and derived photometric redshifts for over 80,000 sources in the E-CDF-S; their photometric redshifts are of high quality, in particular for bright sources.

Phot_Redshift_3_Min
The 1-sigma lower bound on the Cardamone et al. (2010) photometric redshift for the source. Sources with no such information have null values for this parameter.

Phot_Redshift_3_Max
The 1-sigma upper bound on the Cardamone et al. (2010) photometric redshift for the source. Sources with no such information have null values for this parameter.

Phot_Redshift_3_Flag
The quality flag Q_z_for the Cardamone et al. (2010) photometric redshift, where smaller values of Qz indicate better quality; 0 < Qz <~ 1-3 indicates a reliable photometric redshift estimate.

Phot_Redshift_4
The photometric redshift for the source taken from Rafferty et al. (2011, ApJ, submitted). Sources with no such information have null values for this parameter. Rafferty et al. (2011) derived photometric redshifts for over 100,000 sources in the E-CDF-S, using a compiled photometric catalog that probes fainter magnitudes than the Cardamone et al. (2010, ApJS, 189, 270) catalog by including sources in the GOODS-S MUSIC catalog (Grazian et al. 2006, A&A, 449, 951; Santini et al. 2009, A&A, 504, 751); their photometric redshifts are accurate down to faint fluxes.

Phot_Redshift_4_Min
The 1-sigma lower bound on the Rafferty et al. (2011) photometric redshift for the source. Sources with no such information have null values for this parameter. See Footnote 45 of the reference paper for a caveat about the reliability of this parameter.

Phot_Redshift_4_Max
The 1-sigma upper bound on the Rafferty et al. (2011) photometric redshift for the source. Sources with no such information have null values for this parameter. See Footnote 45 of the reference paper for a caveat about the reliability of this parameter.

Redshift
The preferred redshift adopted by the authors in the reference paper. They chose redshifts, in order of preference, as follows:

  (1) secure spectroscopic redshifts;

  (2) insecure spectroscopic redshifts that are in agreement with at least one
  of the Luo et al. (2010), Cardamone et al. (2010), or Rafferty et al. (2011)
  photometric-redshift estimates (i.e., |(zspec - zphot)/(1 + zspec)| <= 0.15,
  where zspec/zphot is the spectroscopic/photometric redshift);

  (3) the Luo et al. (2010) photometric redshifts;

  (4) the Cardamone et al. (2010) photometric redshifts; and

  (5) the Rafferty et al. (2011) photometric redshifts.
  
Of the 716 main catalog sources that have multiwavelength identifications, 673 (94.0%) have spectroscopic or photometric redshifts.

L08_Source_Number
The corresponding 2 Ms CDF-S source number from the main and supplementary Chandra catalogs presented in Luo et al. (2008, ApJS, 179, 19)= L08. The authors matched their X-ray source positions (i.e., the ra and dec parameters in this table) to the L08 source positions (corrected for the systematic positional shifts described in Section 3.1 of the reference paper) using a 2.5 arcsec matching radius for sources with off-axis angle theta < 6' and a 4.0 arcsec matching radius for sources with theta >= 6'. The mismatch probability is ~1% using this approach. For the 740 main-catalog sources, the authors find:

(a) Four hundred forty have matches to the 462 L08 main-catalog sources (the value of l08_source_number is that from Column 1 of Table 2 in L08; see Section 4.5 of the reference paper for more details);

(b) Forty-one have matches to the 86 L08 supplementary CDF-S plus E-CDF-S Chandra catalog sources (the value of l08_source_number is that from Column 1 of Table 5 in L08 with a prefix of "SP1_," e.g., SP1_1);

(c) Twenty-two have matches to the 30 L08 supplementary optically bright Chandra catalog sources (the value of l08_source_number is that from Column 1 of Table 6 in L08 with a prefix of "SP2_," e.g., SP2_1);

(d) Six were outside of the 2 Ms CDF-S footprint of L08 (the value of l08_source_number is set to -1); the detection of these sources is simply due to the new sky coverage (rather than the improved sensitivity) of the 4 Ms CDF-S;

(e) Two hundred thirty-one have no match in any of the L08 main and supplementary Chandra catalogs; these sources were inside the 2 Ms CDF-S footprint but are only detected now due to the improved sensitivity of the 4 Ms observations (the value of l08_source_number is set to 0).

In summary, of the 740 main-catalog sources, 503 were detected previously in the 2 Ms CDF-S observations (the value of l08_source_number is greater than 0) and 237 were detected only in the 4 Ms observations (the value of l08_source_number is either -1 or 0). Compared to the L08 main catalog, there are 300 (i.e., 740 - 440 = 300) new main-catalog sources (see Section 4.7 of the reference paper for more details of these 300 sources).

L08_RA
The Right Ascension of the corresponding L08 source (corrected for the systematic positional shifts described in Section 3.1 of the reference paper) in the selected equinox. This was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. Sources without an L08 match have their L08 Right Ascension values set to null.

L08_Dec
The Declination of the corresponding L08 source (corrected for the systematic positional shifts described in Section 3.1 of the reference paper) in the selected equinox. This was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. Sources without an L08 match have their L08 Declination values set to null.

L05_Source_Number
The corresponding 250 ks E-CDF-S source number from the main and supplementary Chandra catalogs presented in Lehmer et al. (2005, ApJS, 161, 21) = L05. The authors adopted the same matching approach between X-ray catalogs as used for the matching with the Luo et al. (2008) sources, again with the E-CDF-S source positions corrected for the systematic positional shifts described in Section 3.1 of the reference paper. For the 740 main-catalog sources, the authors find that:

(a) 239 have matches in the E-CDF-S main Chandra catalog (the value of l05_source_number is that from Column 1 of Table 2 in L05);

(b) 5 have matches in the E-CDF-S supplementary optically bright Chandra catalog (the value of l05_source_number is that from Column 1 of Table 6 in L05 with a prefix of "SP_," e.g., SP_1); and

(c) 496 have no match in either of the E-CDF-S main or supplementary Chandra catalogs (the value of l05_source_number is set to 0).

L05_RA
The Right Ascension of the corresponding L05 source (corrected for the systematic positional shifts described in Section 3.1 of the reference paper) in the selected equinox. This was given in J2000.0 equatorial coordinates to a precision of 0.01 seconds of time in the original table. Sources without an L05 match have their L08 Right Ascension values set to null.

L05_Dec
The Declination of the corresponding L05 source (corrected for the systematic positional shifts described in Section 3.1 of the reference paper) in the selected equinox. This was given in J2000.0 equatorial coordinates to a precision of 0.1 arcseconds in the original table. Sources without an L08 match have their L05 Declination values set to null.

FB_Exposure
The full-band effective exposure time for the source, in seconds. This was derived from the exposure maps (detailed in Section 3.1 of the reference paper) for the full band. Dividing the corresponding counts in this band for the source by this effective exposure time will provide an effective count rate that has been corrected for vignetting, quantum-efficiency degradation, and exposure-time variations.

SB_Exposure
The soft-band effective exposure time for the source, in seconds. This was derived from the exposure maps (detailed in Section 3.1 of the reference paper) for the soft band. Dividing the corresponding counts in this band for the source by this effective exposure time will provide an effective count rate that has been corrected for vignetting, quantum-efficiency degradation, and exposure-time variations.

HB_Exposure
The hard-band effective exposure time for the source, in seconds. This was derived from the exposure maps (detailed in Section 3.1 of the reference paper) for the hard band. Dividing the corresponding counts in this band for the source by this effective exposure time will provide an effective count rate that has been corrected for vignetting, quantum-efficiency degradation, and exposure-time variations.

Band_Ratio
The band ratio for the source. The authors defined the band ratio as the ratio of the counts between the hard and soft bands, correcting for differential vignetting between the hard and soft bands using the appropriate exposure maps. The band ratios and their corresponding errors have been set to null values for those sources which were detected only in the full band.

Band_Ratio_Pos_Err
The upper error in the band ratio for the source. The authors followed the numerical error-propagation method described in Section 1.7.3 of Lyons (1991, Data Analysis for Physical Science Students) to compute the band ratio errors. This method avoids the failure of the standard approximate variance formula when the number of counts is small and the error distribution is non-Gaussian (e.g., see Section 2.4.5 of Eadie et al. 1971, Statistical Methods in Experimental Physics). The authors calculated upper limits for sources detected in the soft band but not the hard band and lower limits for sources detected in the hard band but not the soft band. For these sources, they set the upper and lower errors to the computed band ratio value. The band ratios and their corresponding errors were set to null values for those sources which were detected only in the full band.

Band_Ratio_Neg_Err
The lower error in the band ratio for the source. The authors followed the numerical error-propagation method described in Section 1.7.3 of Lyons (1991, Data Analysis for Physical Science Students) to compute the band ratio errors. This method avoids the failure of the standard approximate variance formula when the number of counts is small and the error distribution is non-Gaussian (e.g., see Section 2.4.5 of Eadie et al. 1971, Statistical Methods in Experimental Physics). The authors calculated upper limits for sources detected in the soft band but not the hard band and lower limits for sources detected in the hard band but not the soft band. For these sources, they set the upper and lower errors to the computed band ratio value. The band ratios and their corresponding errors were set to null values for those sources which were detected only in the full band.

Photon_Index
The effective photon index Gamma of the source for a power-law model with the Galactic column density value of 8.8 x 1019 cm-2 which was given in Section 1 of the reference paper. The authors calculated the effective photon index based on the band ratio value, using a conversion between the effective photon index and the band ratio. They derived this conversion using the band ratios and photon indices calculated by the AE-automated XSPEC-fitting procedure for relatively bright X-ray sources (with full-band counts greater than 200; this ensures reliable XSPEC-fitting results). This approach takes into account the multi-epoch Chandra calibration information and thus has an advantage over methods using only single-epoch calibration information such as the Portable, Interactive, Multi-Mission Simulator method used by L08. For low-count sources, the authors were unable to determine the effective photon index reliably; they therefore assumed Gamma = 1.4, which is a representative value for faint sources that should yield reasonable fluxes, and set the corresponding upper and lower bounds to values of 0.00. The authors defined sources with a low number of counts as those which were (1) detected in the soft band with <30 counts and not detected in the hard band, (2) detected in the hard band with <15 counts and not detected in the soft band, (3) detected in both the soft and hard bands, but with <15 counts in each, or (4) detected only in the full band. #

Photon_Index_Max
The upper bound for the effective photon index Gamma of the source for a power-law model with the Galactic column density value of 8.8 x 1019 cm-2 which was given in Section 1 of the reference paper. The authors calculated upper limits for sources detected in the hard band but not the soft band and lower limits for sources detected in the soft band but not the hard band. For these sources, they set the upper and lower bounds to be the same as the computed effective photon index. For low-count sources, the authors were unable to determine the effective photon index reliably; they therefore assumed Gamma = 1.4, which is a representative value for faint sources that should yield reasonable fluxes, and set the corresponding upper and lower bounds to values of 0.00.

Photon_Index_Min
The lower bound for the effective photon index Gamma of the source for a power-law model with the Galactic column density value of 8.8 x 1019 cm-2 which was given in Section 1 of the reference paper. The authors calculated upper limits for sources detected in the hard band but not the soft band and lower limits for sources detected in the soft band but not the hard band. For these sources, they set the upper and lower bounds to be the same as the computed effective photon index. For low-count sources, the authors were unable to determine the effective photon index reliably; they therefore assumed Gamma = 1.4, which is a representative value for faint sources that should yield reasonable fluxes, and set the corresponding upper and lower bounds to values of 0.00.

FB_Flux
The full-band observed-frame flux for the source, in units of erg cm-2 s-1. The authors computed fluxes using the corresponding counts in the same band, the appropriate effective exposure in this band, and the effective power-law photon indices. They did not correct fluxes for absorptions by Galactic material or material intrinsic to the source. Negative flux values indicate upper limits. The authors note that, due to the Eddington bias, sources with low net counts could have true fluxes lower than those computed here (see, e.g., Vikhlinin et al. 1995, ApJ, 451, 553; Georgakakis et al. 2008, MNRAS, 388, 1205). They did not attempt to correct for the Eddington bias, since they aim to provide only observed fluxes herein. Determining more accurate fluxes for these sources would require (1) using a number-count distribution prior to estimating the flux probabilities for sources near the sensitivity limit and/or (2) directly fitting the X-ray spectra for each observation; these analyses were beyond the scope of their paper.

SB_Flux
The soft-band observed-frame flux for the source, in units of erg cm-2 s-1. The authors computed fluxes using the corresponding counts in the same band, the appropriate effective exposure in this band, and the effective power-law photon indices. They did not correct fluxes for absorptions by Galactic material or material intrinsic to the source. Negative flux values indicate upper limits. The authors note that, due to the Eddington bias, sources with low net counts could have true fluxes lower than those computed here (see, e.g., Vikhlinin et al. 1995, ApJ, 451, 553; Georgakakis et al. 2008, MNRAS, 388, 1205). They did not attempt to correct for the Eddington bias, since they aim to provide only observed fluxes herein. Determining more accurate fluxes for these sources would require (1) using a number-count distribution prior to estimating the flux probabilities for sources near the sensitivity limit and/or (2) directly fitting the X-ray spectra for each observation; these analyses were beyond the scope of their paper.

HB_Flux
The hard-band observed-frame flux for the source, in units of erg cm-2 s-1. The authors computed fluxes using the corresponding counts in the same band, the appropriate effective exposure in this band, and the effective power-law photon indices. They did not correct fluxes for absorptions by Galactic material or material intrinsic to the source. Negative flux values indicate upper limits. The authors note that, due to the Eddington bias, sources with low net counts could have true fluxes lower than those computed here (see, e.g., Vikhlinin et al. 1995, ApJ, 451, 553; Georgakakis et al. 2008, MNRAS, 388, 1205). They did not attempt to correct for the Eddington bias, since they aim to provide only observed fluxes herein. Determining more accurate fluxes for these sources would require (1) using a number-count distribution prior to estimating the flux probabilities for sources near the sensitivity limit and/or (2) directly fitting the X-ray spectra for each observation; these analyses were beyond the scope of their paper.

Rf_Lx
A basic estimate of the absorption-corrected, rest-frame 0.5-8 keV luminosity L0.5-8keV of the source, in units of erg s-1. The authors calculated L0.5-8keV using the procedure detailed in Section 3.4 of Xue et al. (2010, ApJ, 720, 368). Briefly, this procedure models the X-ray emission using a power law with both intrinsic and Galactic absorption (i.e., zpow x wabs x zwabs in XSPEC) to find the intrinsic column density that reproduces the observed band ratio, assuming a typical power-law photon index of Gammaint = 1.8 for intrinsic AGN spectra. It then corrects for both Galactic and intrinsic absorption to obtain the absorption-corrected intrinsic 0.5-8.0 flux fint, as opposed to the observed flux given in the fb_flux parameter), and follows the equation L0.5-8 = 4 * pi * (dL)2 * fint * (1 + z)^(Gammaint - 2) to derive L0.5-8keV (where dL is the luminosity distance and z is the adopted redshift given in the redshift parameter). In this procedure, the authors set the observed band ratio to a value that corresponds to Gamma = 1.4 for sources detected only in the full band; for sources having upper or lower limits on the band ratio, they adopted their upper or lower limits for this calculation. Basic luminosity estimates derived in this manner are generally found to agree with those from direct spectral fitting to within a factor of ~ 30% (but see Footnote 47 in the reference paper for a caveat on this); the direct spectral-fitting approach should produce more reliable estimates, but is beyond the scope of their paper. Sources without redshift estimates have this parameter set to null; negative luminosity values indicate upper limits.

Broad_Type
A basic estimate of the likely source type. The authors categorized the X-ray sources into three basic types: "AGN," "Galaxy," and "Star." They utilized four criteria that are based on distinct AGN physical properties and one criterion that is based on optical spectroscopic information to identify AGN candidates, which must satisfy at least one of these five criteria. They briefly describe these criteria and their limitation on pages 15 - 16 of the reference paper, as well as the procedures which they followed in order to identify likely stars. They inspected each of the sources identified as stars in HST images and retrieved sources that appear to be galaxies (i.e., set their classification to "galaxy"). The sources that were not identified as AGNs or stars are classified as "galaxies". Of the 740 main-catalog sources, 568 (76.8%), 162 (21.9%), and 10 (1.3%) are identified as AGNs, galaxies, and stars, respectively.

Notes
This parameter contains notes on the sources. The authors annotated sources at the field edge that lie partially outside of the survey area with "E" (one source only) and sources in close doubles or triples with "C" (a total of 35 sources; these 35 sources have overlapping polygonal extraction regions that correspond to ~40%-75% EEFs; see Section 3.2 of the reference paper). Sources not annotated have this parameter set to blank.

Class
The Browse object classification based on the value of the broad_type parameter.


Contact Person

Questions regarding the CHANDFS4MS database table can be addressed to the HEASARC User Hotline.
Page Author: Browse Software Development Team
Last Modified: Friday, 29-Jul-2011 11:03:36 EDT