GOODS Source Catalogs release r2.0z based on v2.0 data

Document version: 23 May 2008

TABLE OF CONTENTS


1.0 General information

The GOODS Team is releasing version r2.0z of the ACS multi-band source catalogs.

These catalogs, prepared using the SExtractor package (Bertin & Arnouts 1996, A&A 117, 393), are based on the version v2.0 of the reduced, calibrated, stacked and mosaiced images acquired with HST and ACS as part of the GOODS ACS Treasury program and of the PANS program of search for Type Ia supernovae at high redshift.

The catalogs are z-band based, that is, source detection has been made using the z-band images. A variety of photometric apertures defined during the detection process have then been used as "fixed apertures" in the i, v and b-band images to derive the multi-band photometry.

2.0 What is being released

The r2.0z release of the multi-band source catalogs comprises: eight (8) primary ASCII files (the updated catalogs); one (1) updated file required to configure SExtractor; and this "README" file.

The eight primary files are named:

Groups of letters, numbers or punctuation symbols separated by underscore ("_") are used to specify the main parameter of the catalogs.

The prefix "h" is used in the STScI multimission archive to indicate HST data products.

The second group the survey identifier.

The first letter of the following two-letter group indicates which GOODS field is cataloged (values are either "s" for GOODS-S, or "n" for GOODS-N), while the second letter indicates the photometric passband (values are "b", "v", "i" and "z").

The group "r2.0z" specifies that this is the release 2.0 of the catalogs, and that detections are based on the z band.

The following group "_cat" specifies the nature of the data product (source catalog).

The extension ".txt" specifies that the catalogs consist of text ASCII files.

The updated SExtractor configuration files are:

These file specify the list of columns output by SExtractor in its photometry runs, the size of the convolution kernel, and the setting of the neural network algorithm for image deblending, respectively. They are the same as the files released as part of the previous r1.1 GOODS ACS catalog version.

All other SExtractor configuration files are also the same as the previous r1.1 release (listed in the r1.0z release notes), except that they contain updated photometric zero points and exposure time (see also the documentation of the v2.0 image release). These files configure SExtractor to perform source detection in the F850LP (z) band (the _detect_ files) and to carry out photometry of the detected sources in each of the four passbands (the _phot_ files). They are:

3.0 Changes from previous GOODS ACS data releases

The r2.0z catalog release is based on the v2.0 images, which, as described above, have significantly longer total exposure times in the z850 bandpass, and somewhat longer exposure times in the i775 and V606 bands as well.

The only significant difference in the catalogs, other than that of being based on deeper data, is that we have applied a small astrometric offset to the GOODS-North images (only), as explained in the v2.0 documentation. This offset is a simple translational change to the world coordinates for the GOODS-N data, based on comparisons to catalogs from the Sloan Digital Sky Survey, 2MASS, and to new, deep VLA 20cm data (Morrison et al. 2010, ApJ, 188, 178). This correction is:

           RA(v2.0)  - RA(v1.0)  =  0.000 arcsec =  0.0000000 degrees
           Dec(v2.0) - Dec(v1.0) = -0.320 arcsec = -0.0000889 degrees

This correction was applied to the images by updating the CRVAL2 header keyword in each image:

	   CRVAL2:    62.238661 ->  62.238572   (62:14:19.18 -> 62:14:18.86)

Coordinates in the v2.0 source catalogs were also updated accordingly.

We note that as of this writing (14 May 2008), this change in GOODS-N astrometry has not yet been applied to other GOODS-North public data sets, such as the Spitzer IRAC and MIPS data. This will be corrected in future releases of data from those other facilities.

4.0 How the catalogs were prepared

With the scale of the images set at 0.03 arcsec per pixel (to improve the sampling of the PSF), each GOODS field would have resulted in images too large in size for practical purposes (40,000x40,000 pixels for the HDF-N and 32,000x40,000 pixels for the CDF-S). As a result, each field has been divided into contiguous sections, each 8,192 x 8,192 pixels in size. A total of 17 sections cover the HDF-N, and a total of 18 sections cover the CDF-S.

For source detection, we prepared special overlapping sections, still 8192x8192 pixels in size, but with a 600 pixel overlap at the borders to ensure that no objects were split across section borders. After each section was SExtracted independently, objects in the overlapping areas were matched by RA and Dec, and duplicates were eliminated by selecting the objects furthest from the section edge. The resulting catalog was sorted by RA and Dec, and unique sequential ID_MOSAIC identifiers were assigned. Some duplicated objects were removed from the catalog by hand, and as a result the ID_MOSAIC values do not include every ordinal number. Object detection and merging were performed on the z band images only, and the b, v and i photometry was obtained processing these images in dual-image mode. The merged b, v and i catalogs were obtained by selecting the appropriate object record based on the z-band object ID_MOSAIC numbers.

Because the images are derived from overlapping ACS frames, taken at different orientations, the exposure time is not uniform across the images, and there is a complex pattern to the exposure-time variations. This is recorded in the weight maps released with the v2.0 images. We also use this information to create a flag map, with various bits set to represent the exposure time of the image at each pixel position. The algorithm for creating the flag maps does two levels of smoothing prior to setting a flag. (Even so, there is still some blotchiness in the flags). The algorithm for setting a flag is as follows:

  1. median filter the weight map with a 7x7 kernel
  2. flag pixels below a certain threshold with value = 1.0
  3. boxcar smooth this 1,0 map with a 7x7 kernel
  4. set pixels with values > 0 to the desired flag value

The thresholds are set roughly in to catch the typical weight (w) of pixels, calculated as median in a 7x7 pixel box, with a given fraction of the median weight value of the stack (m_w). The bit values have been assigned as follows:

Flag values for GOODS ACS images
Number epochs Bit set
0 weight (off the edge) 32
within 33 pixels of edge (zero weight regions) 16
w <= (1.2/5)*m_w 8
w <= (2.2/5)*m_w 4
w <= (3.2/5)*m_w 2
w <= (4.2/5)*m_w 1
w > (4.2/5)*m_w 0

The median weight ("m_w") in each band for both GOODS fields are:

Median weight map values for GOODS ACS images
FIELDF435WF606WF775WL850LP
GOODS-N438511119113158074423859
GOODS-5477099121406146677407598

The flag maps were was passed to SExtractor using the FLAG_IMAGE configuration parameter. The catalog output parameter IMAFLAGS_ISO gives the logical 'OR' of the flag map bit values for pixels within the isophotal "segmentation" area of the object. This information can be used to identify objects that have less than the typical exposure time. As a rule-of-thumb: Pixels with flag values > 8 are generally suspect due to poor cosmic-ray rejection.

  • 5.0 Source detection and photometry

    Source detection was carried out on the z-band images, and photometry was then measured in the other bandpasses using SExtractor "dual image mode", i.e., through fixed apertures defined from the z-band detection catalog.

    Because the exposure times and local noise levels vary over the images, the weight map images provided with the v2.0 data release were fed to SExtractor to ensure that source detection occurred at a constant S/N level based on the local sky noise. Specifically, the weight maps were constructed according to a noise model, as described in the GOODS ACS v2.0 image product documentation, and were normalized so that their values are approximately equal to the inverse variance of the background noise. As described in the image data release documentation, the RMS computed from the weight maps takes into account account the small scale correlation between pixels that is introduced by the drizzling process, so that noise over larger areas (such as photometry apertures) should be approximately correct. Before cataloging, we verified the noise scaling by measuring the amplitude and autocorrelation of the background noise in the drizzled science images, and using this information to rescale the weight maps ("w") by a weight scale factor ("s") before converting to RMS maps ("rms"), rms = 1/sqrt(s*w). The scaling factors that were used used are:

    GOODS ACS weight map rescaling factors
    Bandpass s (North) s (South)
    F435W 0.9052 0.8487
    F606W 0.9386 0.8987
    F775W 0.9094 0.9199
    F850LP 0.9307 0.9295

    In most cases, these factors change the scaling of the RMS maps by less than 5%.

    When SExtractor is provided with RMS maps (WEIGHT_TYPE = MAP_RMS), it assumes that they are absolutely normalized and does not attempt to rescale them further. Flux and magnitude errors in the catalogs are computed as described in the SExtractor documentation, using the normalized RMS maps to estimate the contribution from the background shot noise.

    The primary parameters that affect the detection process, namely the minimum connected area, the isophotal threshold (in unit of the background fluctuation RMS) and the convolution kernel have been set with the help of numerical simulations to maximize sensitivity to faint and relatively compact sources while reducing the number of spurious sources to a minimum. In the simulations we inserted a number of artificial galaxies in the images and retrieved them with the same procedures adopted for real sources. The morphology of these galaxies was, with equal probability, either an exponential disk or a de Vaucouleur spheroid. Apparent magnitudes and half-light radii were extracted from uniform distribution functions, covering the range 20 <= z_850 <= 28 (AB) and PSF <= r_1/2 <= 2.0 arcsec, respectively. The simulated disk sample (viewed from arbitrary directions) is drawn from a population of oblate optically-thin spheroids with a Gaussian distribution of intrinsic axial ratios with mean b/a = 0.05 and sigma = 0.01. The spheroids were drawn from a population of oblate spheroids with intrinsic axial ratios uniformly populating the range 0.3 < b/a < 0.9. Position angles were randomly distributed.

    We fine-tuned the SExtractor parameters to maximize the number of detected sources, while keeping the number of spurious sources, estimated with the method of the "negative images", essentially negligible.

    One should keep in mind, however, that there is no one single catalog that is perfect for all scientific problems and applications. While we believe that the released catalogs are good for most investigations on faint galaxies, users with specific problems (e.g. low-surface brightness galaxies) are strongly urged to experiment with different settings of the SExtraction detection parameters.

    SExtractor measures photometry through a variety of apertures, which are explained in more detail the SExtractor documentation. The ones of primary interest in the GOODS catalogs are the isophotal aperture (ISO), which the "AUTO" aperture which uses an elliptical aperture whose size and orientation are determined by the galaxy light profile in the 2D images, and a series of circular apertures (APER) with fixed radii. The radii of the 11 circular apertures used for photometry in the GOODS catalogs are spaced by factors of sqrt(2), and are:

    or equivalently:

    6.0 Modifications to SExtractor

    In the process of building the catalogs, we realized that two small modifications to the SExtractor software (version 2.2.2) resulted in somewhat better catalogs. These modifications affect the splitting/merging behavior and the computation of the local sky background, and are described below.

    7.0 Column description for the catalogs

    A listing of the individual columns of the catalogs, reproduced from the catalog file headers, is provided below. We have added brief descriptions of the output quantities here, but in many cases the user should refer to the SExtractor documentation for more detailed explanations.

    Most of the output parameters are as defined by SExtractor, but we have added several other parameters. Most of these additional parameters end with _MOSAIC, to indicate global pixel positions pertaining to a virtual mosaic of the image sections. This virtual mosaic has coordinate value (0,0) at the central reference point of the world coordinate system (CRVAL1,2), and spans coordinate ranges from -19000 to +19000.

    The ID_MOSAIC column contains the unique sequential id assigned to the objects. There are a few small gaps in this sequence as a result of objects being removed at a later quality assessment stage of the process. This should not be a cause of concerns.

    The X_MOSAIC, Y_MOSAIC, XMIN_MOSAIC, YMIN_MOSAIC, XMAX_MOSAIC, YMAX_MOSAIC are the various x and y pixel positions in the reference frame of the full (virtual) mosaic of the GOODS field.

    The SECT_REFNUM indicates the _SECT.FITS file from the v2.0 image release which contains the barycenter of the object, at pixel coordinates (X_SECT, Y_SECT).

    Finally, we have also assigned object names based on their position in the sky, in accordance with the IAU naming conventions. This is recorded in first column of each catalog, named ID_IAU.

    A few parameters occur in groups, particularly, the FLUX_RADIUS values (3), and the APER magnitudes, fluxes, and errors (11 of each).

    GOODS ACS catalog entries:

    # 1    ID_IAU              IAU-format coordinate-based name for object
    # 2    ALPHA_J2000         J2000 Right Ascension (degrees)
    # 3    DELTA_J2000         J2000 Declination (degrees)
    # 4    SECT_REFNUM         section number from GOODS v2.0 imaging data
    # 5    X_SECT              centroid pixel x coordinate within section
    # 6    Y_SECT              centroid pixel y coordinate within section
    # 7    X_MOSAIC            centroid pixel x coordinate within a virtual mosaic of all sections
    # 8    Y_MOSAIC            centroid pixel y coordinate within a virtual mosaic of all sections
    # 9    XPEAK_MOSAIC        peak pixel x coordinate within virtual mosaic
    # 10   YPEAK_MOSAIC        peak pixel y coordinate within virtual mosaic
    # 11   XPEAK_WORLD         peak RA coordinate within virtual mosaic
    # 12   YPEAK_WORLD         peak Dec coordinate within virtual mosaic
    # 13   XMIN_MOSAIC         minimum x coordinate of object segmentation map within virtual mosaic
    # 14   YMIN_MOSAIC         minimum y coordinate of object segmentation map within virtual mosaic
    # 15   XMAX_MOSAIC         maximum x coordinate of object segmentation map within virtual mosaic
    # 16   YMAX_MOSAIC         maximum y coordinate of object segmentation map within virtual mosaic
    # 17   ISOAREAF_IMAGE      filtered isophotal area (pixels)
    # 18   ISOAREA_IMAGE       isophotal area (pixels)
    # 19   THETA_IMAGE         position angle (degrees)
    # 20   ELLIPTICITY         ellipticity
    # 21   ELONGATION          elongation
    # 22   ERRTHETA_IMAGE      positional angle uncertainty (degrees)
    # 23   KRON_RADIUS         "Kron radius" from light profile (pixels)
    # 24   FLUX_RADIUS(1)      20% enclosed flux radius (pixels)
    # 25   FLUX_RADIUS(2)      50% enclosed flux radius (pixels)
    # 26   FLUX_RADIUS(3)      80% enclosed flux radius (pixels)
    # 27   FWHM_IMAGE          full width at half maximum (pixels)
    # 28   CLASS_STAR          star-galaxy classification parameter
    # 29   FLAGS               SExtractor internal flags
    # 30   IMAFLAGS_ISO        external flags from flag map, logical OR combination over segmentation area
    # 31   NIMAFLAGS_ISO       number of pixels with non-zero IMAFLAGS_ISO 
    # 32   BACKGROUND          local sky background value (DN/s)
    # 33   FLUX_MAX            peak counts (DN/s)
    # 34   MAG_ISO             isophotal AB magnitude
    # 35   MAGERR_ISO          isophotal AB magnitude uncertainty
    # 36   FLUX_ISO            isophotal flux (DN/s)
    # 37   FLUXERR_ISO         isophotal flux uncertainty (DN/s)
    # 38   MAG_ISOCOR	   corrected isophotal AB magnitude
    # 39   MAGERR_ISOCOR	   corrected isophotal AB magnitude uncertainty
    # 40   FLUX_ISOCOR	   corrected isophotal flux (DN/s)
    # 41   FLUXERR_ISOCOR	   corrected isophotal flux uncertainty (DN/s)
    # 42   MAG_AUTO            AB magnitude in SExtractor "AUTO" aperture
    # 43   MAGERR_AUTO         AB magnitude uncertainty in SExtractor "AUTO" aperture
    # 44   FLUX_AUTO           flux in SExtractor "AUTO" aperture (DN/S)
    # 45   FLUXERR_AUTO        flux uncertainty in SExtractor "AUTO" aperture (DN/s)
    # 46   MAG_BEST            SExtractor "BEST" AB magnitude (not recommended!)
    # 47   MAGERR_BEST         SExtractor "BEST" AB magnitude uncertainty (not recommended!)
    # 48   FLUX_BEST           SExtractor "BEST" flux (DN/S) (not recommended!)
    # 49   FLUXERR_BEST        SExtractor "BEST" flux uncertainty (DN/S) (not recommended!)
    # 50-60   MAG_APER         circular aperture AB magnitudes (see documentation for aperture sizes)
    # 61-71   MAGERR_APER      circular aperture AB magnitude uncertainties (see documentation for aperture sizes)
    # 72-82   FLUX_APER        circular aperture fluxes (DN/s) (see documentation for aperture sizes)
    # 83-93   FLUXERR_APER     circular aperture flux uncertainties (DN/s) (see documentation for aperture sizes)
    # 94   X2_IMAGE            second order x moment of the light distribution (pixels)
    # 95   Y2_IMAGE            second order x moment of the light distribution (pixels)
    # 96   XY_IMAGE            second order xy moment of the light distribution (pixels)
    # 97   ERRX2_IMAGE         second order x moment uncertainty (pixels)
    # 98   ERRY2_IMAGE         second order x moment uncertainty (pixels)
    # 99   ERRXY_IMAGE         second order x moment uncertainty (pixels)
    # 100  A_IMAGE             semi-major axis length (pixels)
    # 101  B_IMAGE             semi-minor axis length (pixels)
    # 102  ERRA_IMAGE          semi-major axis uncertainty (pixels)
    # 103  ERRB_IMAGE          semi-minor axis uncertainty (pixels)
    # 104  ID_MOSAIC           object identifier number from virtual mosaic
    
    8.0 Citations

    A description of the GOODS observations and data products is given in "The Great Observatories Origins Deep Survey: Initial Results from Optical and Near-Infrared Imaging" by M. Giavalisco and the GOODS Team [2004, ApJ, 600, L93].  Another paper describing in more details the GOODS HST/ACS data and source catalogs is in preparation and, for the time being, should be referenced as "M. Giavalisco and the GOODS Team, 2008, in preparation".


    Back to ftp site.

    Back to top of page.